Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
A percutaneously implantable wireless neurostimulator for treatment of stress urinary incontinence
(USC Thesis Other)
A percutaneously implantable wireless neurostimulator for treatment of stress urinary incontinence
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
A Percutaneously Implantable Wireless Neurostimulator
for Treatment of Stress Urinary Incontinence
By
Xuechen Huang
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(BIOMEDICAL ENGINEERING)
December 2017
1
To my family
To my daughter
Mandi Huang
2
ACKNOWLEDGEMENT
The thesis presented here is a record of my research and study work in University of Southern California.
The five years’ experience here in USC gave me an opportunity to meet many different of remarkable
people, work with them, and learn from them. I would like to express my sincere gratitude to those people
whose encouragement and support have enabled me to dedicate to my work. Thanks for helping this
dissertation a reality.
First and foremost, I would like to thank my thesis supervisor, Gerald Eli Loeb, who walked me throughout
the journey. Thanks for him that I have this unique opportunity to gain a mix experience from academic
and industry research. With his support, encouragement, and most importantly, his patience, I could
concentrate on my goal, focusing on academic research that could be commercialized and help patients. I
really appreciate the thousands of hours and tremendous energy he has invested on me.
I would like to thank my committee members, Dr. Francisco Valero-Cuevas, Dr. Ellis Meng, Dr. Daniel
Kirages, and Dr. Michael Khoo, for their valuable feedbacks for keeping me focused on right questions and
doing right works. In addition to mentorship, they are good role models as a team leader and a scientific
researcher.
I would like to thank my colleagues and collaborators within General Stim, Medical Device Development
Facility, The USC International Center of Regulatory Science, and Syntouch. I appreciate for all of the
input that has been given to make this project possible. I am grateful to Sam Kohan, Petcharat May
Denprasert, Katie Zheng, Ronald Yu, Longjiao Peng, Sisi Shi, Thomas Yeh, Ray Peck and Gary Lin for
contribution in technical development. I am grateful to my lab mates and friends, Li Zhou and Adriana N,
Vest, your valuable inputs and advice have given me new strategies to tackle difficult problems. I am
grateful to Hao Li, Jinjin Yu, Zeyu Jia, Haifeng Lao and people in manufacture team, all of whom
continuously work hard to improve this technology. Thank you to Dr. Frances Richmond for not only your
expertise in regulatory science and physiology, but your patience, encouragement, and support.
3
I would like to express my appreciation to the group of research clinicians, including Dr. Limin Liao, Dr.
Fan Zhang, Dr. Werner Schaefer, and Dr. Med Helmut Madersbacher. Thank you for your dedication to
this project and your advice to me. Thanks to Xing Li, Tianji Lu, Zhaoxia Wang and Han Deng, and Linxiao
Yan for assistance in animal study.
I would like to express my appreciation for the generous funding from USC Provost Fellowship and General
Stim that have kept me able to dedicate my time to research.
Lastly, as a personal note, I want to thank to all my friends and family who have supported me throughout
the years. To my sworn brother, Yongdong Cai, thank you for encouraging me as one of your own. Thanks
for my best friends, Yuanjie Zhang, Hanxiao Xu, and Anqi Shi, all of whom inspired me with their own
accomplishments and supported me as always without hesitation. Thank you to my parents and my
grandparents, who have been my primary role models, providing my first examples of hardworking,
perseverance, integrity and discipline. It is the freedom they provided that I am able to follow my heart and
pursue my own career. Wish my colleagues, friend and family all the best.
4
TABLE OF CONTENTS
ACKNOWLEDGEMENT ............................................................................................................................ 2
TABLE OF CONTENTS .............................................................................................................................. 4
LIST OF TABLES ........................................................................................................................................ 8
LIST OF FIGURES ...................................................................................................................................... 9
ABSTRACT ................................................................................................................................................ 13
CHAPTER 1: INTRODUCTION ............................................................................................................... 15
MOTIVATION ....................................................................................................................................... 15
OUTLINE ............................................................................................................................................... 15
CHAPTER 2: REVIEW OF STRESS URINARY INCONTINENCE CLINICAL BACKGROUND ...... 18
INTRODUCTION .................................................................................................................................. 18
MECHANISM OF CONTINENCE AND INCONTINENCE ............................................................... 19
INCIDENCE ........................................................................................................................................... 21
ETIOLOGY ............................................................................................................................................ 21
CLINICAL DIAGNOSIS AND EVALUATION ................................................................................... 22
CHOICE OF INTERVENTION ............................................................................................................. 23
CONSERVATIVE TREATMENT ..................................................................................................... 23
PHARMACOLOGICAL THERAPY ................................................................................................. 29
SURGICAL INTERVENTIONS ........................................................................................................ 31
SUI AND TREATMENT OPTIONS FOR FEMALE ............................................................................ 32
SUI AND TREATMENT OPTIONS FOR MALE ................................................................................ 33
SUMMARY ............................................................................................................................................ 35
CHAPTER 3: NEUROMUSCULAR STIMULATION ............................................................................. 36
NEUROMUSCULAR PHYSIOLOGY .................................................................................................. 36
ELECTRODE TISSUE INTERFACE .................................................................................................... 37
STRENGTH DURATION RELATIONSHIP ........................................................................................ 38
PULSE POLARITY ................................................................................................................................ 40
HISTORICAL REVIEW OF NEUROMUSCULAR STIMULATION ................................................. 41
CHAPTER 4: MICROSTIMULATION STRATEGY ............................................................................... 45
PREFACE ............................................................................................................................................... 45
5
PRECLINICAL STUDY ........................................................................................................................ 46
TECHNICAL AND SAFETY REQUIREMENTS ................................................................................ 50
THERAPEUTIC FUNCTION ............................................................................................................ 50
PHYSICAL FORM AND PACKAGING ........................................................................................... 52
RELIABILITY AND SAFETY .......................................................................................................... 53
CHAPTER 5: NEUROSTIMULATION STRATEGY FOR STRESS URINARY INCONTINENCE ..... 57
PREFACE ............................................................................................................................................... 57
PERSONAL ROLE ................................................................................................................................ 57
ABSTRACT ............................................................................................................................................ 57
INTRODUCTION .................................................................................................................................. 58
DESIGN .................................................................................................................................................. 60
SYSTEM OPERATION AND REQUIREMENTS ............................................................................ 60
IMPLANT ........................................................................................................................................... 62
INSERTION TOOL ............................................................................................................................ 65
CUSHION ........................................................................................................................................... 68
SOFTWARE APP ............................................................................................................................... 71
METHODS ............................................................................................................................................. 74
VERIFICATION IN VITRO .............................................................................................................. 74
PRECLINICAL VALIDATION IN VIVO ......................................................................................... 74
RESULTS ............................................................................................................................................... 75
SYSTEM BENCH TESTING ............................................................................................................. 75
SYSTEM VALIDATION IN VIVO ................................................................................................... 79
DISCUSSION ......................................................................................................................................... 83
ACKNOWLEDGMENTS ...................................................................................................................... 86
CHAPTER 6: ACCELERATED LIFE-TEST METHODS AND RESULTS FOR IMPLANTABLE
ELECTRONIC DEVICES WITH ADHESIVE ENCAPSULATION........................................................ 88
PREFACE ............................................................................................................................................... 88
PERSONAL ROLE ................................................................................................................................ 88
ABSTRACT ............................................................................................................................................ 88
INTRODUCTION .................................................................................................................................. 89
6
DESIGN AND METHODS .................................................................................................................... 91
COMB PATTERN .............................................................................................................................. 91
FUNCTIONAL DEVICES ................................................................................................................. 95
WIRELESS DETECTION .................................................................................................................. 96
WIRELESS POWERING ................................................................................................................... 98
ACCELERATED LIFE-TESTING .................................................................................................... 99
RESULTS ............................................................................................................................................. 101
DISCUSSION ....................................................................................................................................... 106
INTERPRETATION OF TEST DATA ............................................................................................ 106
EXTRAPOLATION OF EXPECTED LIFETIME ........................................................................... 109
CONCLUSION ..................................................................................................................................... 112
ACKNOWLEDGMENTS .................................................................................................................... 113
CHAPTER 7: PROPOSAL FOR CLINICAL VALIDATION ................................................................. 114
PREFACE ............................................................................................................................................. 114
OBJECTIVES ....................................................................................................................................... 116
HYPOTHESIS ...................................................................................................................................... 116
PILOT CLINICAL TRIAL PROTOCOL ............................................................................................. 116
RECRUITMENT .............................................................................................................................. 116
INFORMED CONSENT .................................................................................................................. 118
BASELINE MEASUREMENT ........................................................................................................ 118
DEVICE IMPLANTATION ............................................................................................................. 119
DEVICE ACTIVATION AND EXERCISE PRESCRIPTION ........................................................ 123
IN-HOME EXERCISE SESSIONS .................................................................................................. 123
OUTCOME MEASUREMENT ....................................................................................................... 123
ADJUSTMENT OF EXERCISE PRESCRIPTION ......................................................................... 124
IMPLANTATION OF SECOND DEVICE ...................................................................................... 125
X-RAY OF DEVICE LOCATIONS ................................................................................................. 125
DISCONTINUATION OF EXERCISE ............................................................................................ 125
REMOVAL OF NUSTIM IMPLANTS ........................................................................................... 126
ANTICIPATED DISCOMFORT AND RISKS OF ADVERSE EVENTS ...................................... 126
7
CHAPTER 8: TRANSLATIONAL AND REGULATORY PATHS ....................................................... 128
CFDA PATHWAY ............................................................................................................................... 128
STANDARDS TESTING ................................................................................................................. 129
ANIMAL TESTING ......................................................................................................................... 131
CLINICAL TRIAL ........................................................................................................................... 131
INNOVATIVE MEDICAL DEVICE PATH.................................................................................... 132
CHAPTER 9. PILOT CLINICAL STUDY .............................................................................................. 134
OBJECTIVE ......................................................................................................................................... 134
RESULTS ............................................................................................................................................. 134
MEDICAL HISTORY ...................................................................................................................... 134
DIAGNOSIS ..................................................................................................................................... 134
IMPLANTATION ............................................................................................................................ 135
ACTIVATION .................................................................................................................................. 136
1
ST
VISIT .......................................................................................................................................... 137
PAD TESTING ................................................................................................................................. 137
CESSATION OF EXERCISE .......................................................................................................... 137
INTERPRETATION ............................................................................................................................. 138
DISCUSSION ....................................................................................................................................... 138
CHAPTER 10: CONCLUSION AND FUTURE DIRECTIONS ............................................................. 140
APPENDIX-1: INFORMED CONSENT FORM ..................................................................................... 141
APPENDIX-2: CASE REPORT FORM ................................................................................................... 152
APPENDIX-3: ANIMAL STUDY REPORT ........................................................................................... 170
APPENDIX-4: HISTOPATHOLOGY REPORT ..................................................................................... 176
REFERENCES ......................................................................................................................................... 187
8
LIST OF TABLES
Table 2-1. Possible pathophysiological mechanisms for weakness of pelvic floor muscle and urethral
sphincter. ..................................................................................................................................................... 20
Table 2-2. Treatment options comparison among side effects and effectiveness. ...................................... 33
Table 6-1. Comparisons of different cleaning procedure for comb pattern devices in Deion. Clean
group, Dist. Clean group and functional microstimulators. ........................................................................ 95
Table 9-1. NuStim delivered charge at each level. ................................................................................... 136
Table 9-2. 1-hour pad testing data from the 1st patient when the study is in progress. ............................ 137
9
LIST OF FIGURES
Figure 2-1. Diagram of SUI. The external urethral sphincter as part of pelvic floor muscle is atrophied.
.................................................................................................................................................................... 18
Figure 2-2. Prevalence of urinary incontinence (and leakage) in women 20 years+ (Reprint from 2005
Hunskaar research)[4]. ................................................................................................................................ 21
Figure 3-1. a. The electrode/electrolyte interface. Equivalent electrical circuit model of the interface
(reprinted from Merrill paper [62]. ............................................................................................................. 38
Figure 3-2. Example of strength duration curve and charge duration curve[62]. ....................................... 39
Figure 3-3. Experimentally determined strength duration relationship for motor nerve stimulation and
direction stimulation of the muscle [64]. .................................................................................................... 40
Figure 3-4. Photograph depicting three neuromuscular microstimulators (BIONs) in different packages.
(Reprinted from Jianguang Qiang paper with permission of the American Association of Neurological
Surgeons[92]) .............................................................................................................................................. 42
Figure 4-1. a. An X-ray showing dislodgement of the leads in an early implant. b. Fracture of the
electrode (Reprint from 1966 Caldwell research) ....................................................................................... 45
Figure 4-2. The experiment set up of the acute human study. The catheter with a balloon is placed
inside the bladder. Constant flow is provided in the catheter with a small opening on the part of urethra
where it goes through the urethral sphincter. if the EUS has responded to electrical stimulation, this
opening is obstructed, which increases the pressure measurement. It should also be possible to record
an M wave, the electromyographical signal associated with synchronous action potentials in a large
population of muscle fibers, but this was obscured by the relatively large stimulus artefact in these
experiments. ................................................................................................................................................ 48
Figure 4-3. Electrodes placement and stimulation location for female and male subjects. ........................ 48
Figure 4-4. Details information about patients’ condition and testing results. It is observed to produce
EUS response as shown with increased pressure measurement, requiring only low stimulation level.
For patient with spinal cord injury, it may not be possible to produce sufficient response due to
denervation of the EUS muscle. .................................................................................................................. 49
Figure 4-5. a. Muscle recruitment curve. Distances (mm) from the nerve to electrode are given in the
legend. b. Effects of distance on the current for a half- maximal response. (Reprinted from Stein[98] )
.................................................................................................................................................................... 52
Figure 5-1. Concept of NuStim operation. The patient receives passive exercise while sitting on the
RF-Cushion and performing daily activities. Exercise is adjusted on a tablet and transmitted wirelessly
to the RF-Cushion, which wirelessly powers and controls the implanted microstimulator. ....................... 61
10
Figure 5-2. a. Design of NuStim implant. b-j. Construction steps. b. Lead free solder paste is applied
on the PCB. c. The surface-mount components are placed and soldered on hot plate (240°C for 30s).
d. Ferrite is wound with insulated copper wire and cleaned in ultrasound. e. PCB is mounted on the
ferrite with epoxy. f. coil terminals and electrodes are soldered. Wirebonds are placed to connect the
PUT. g. Device is inserted and tacked inside the glass capillary. h. Device is loaded inside silicone
tubes and filled with epoxy. i. Epoxy is cured. j. Extra epoxy is trimmed off and the device is ready
for functional testing. .................................................................................................................................. 63
Figure 5-3. Schematic circuit and conceptual waveforms of the NuStim implant for two output pulses
with low and high charge, respectively. L2=11.5µH. C2=47pF. C3=3.3nF. R1=1MΩ. R=39kΩ.
Cout=0.33µF. .............................................................................................................................................. 65
Figure 5-4. Assembled NuStim insertion tool. The dilator (3.26 mm o.d. x 117.6 mm length) passes
through the sheath (4.27 mm o.d.). ............................................................................................................. 66
Figure 5-5. a. A Teflon-coated needle electrode with sharp beveled tip is connected to the disposable
stimulator. The needle is used to find the low-threshold target stimulation site. b. The assembled
insertion tool is used to locate the low-threshold target identified with the needle. c. Stimulation charge
is passed through NuStim to confirm the low-threshold target site. d. The NuStim is released at the
targeted stimulation site. ............................................................................................................................. 67
Figure 5-6. RF-Cushion system architecture. ............................................................................................. 69
Figure 5-7. a. NuStim system test configuration with saline tube to simulate dissipative loading by
conductive tissues of the body. b. The electromagnetic field strength measured from the center to the
edge of the primary coil (radius) as the distance is increased. The blue plane indicates the minimum
field strength needed for regulated stimulation. The intersection with the measured field strength
indicates the maximum operating range of the device. ............................................................................... 70
Figure 5-8. a. Prescription mode. Various stimulation parameters are determined by the physician on
tablet. b. Prescription testing mode. Exercise is first tested in the prescription testing mode before being
sent to the patient. c. Patient mode. Patient performs the prescribed exercise at home. ............................. 73
Figure 5-9. a. Train of stimulation pulses delivered from the NuStim implant. (parameter settings:
Threshold = 1, Target = 20, ramp = 1s, hold = 2s, off = 1s, frequency = 10 pps). b. NuStim output as
measured on oscilloscope for stimulus level = 13 (yellow trace). Electromagnetic field generated from
RF-Cushion in cycle of charging period (blue trace). ................................................................................. 76
Figure 5-10. a. NuStim output measured at different distances from the plane of primary coil as a
function of stimulus strength steps in the app. b. NuStim output measured at different tilted angle,
height and intensity level. Dotted red trace is calculated output decreased as the tilted angle without
presence of ferrite at level 20. c. Stimulus charge comparison between measured values and ideal
calculation at the RF burst durations programmed to produce the 20 stimulus step values. ...................... 78
11
Figure 5-11. a. Threshold measurements as a function of post-implantation day (normalized to implant
activation day). b. Muscle block with a capsule after active device removal. c-e. Photomicrographs of
a tissue section stained with H&E from the block in d showing typical features of cellular response
and capsule formation 1 month after implantation. e. Typical histological appearance around active
device after 3 months. f and g. tissue section from control side with typical appearance around non-
active device after 3 months. ...................................................................................................................... 82
Figure 6-1. Design of IDE device compared to clinical NuStim microstimulator. a. IDE device before
epoxy filling. b. Microstimulator after encapsulation. ................................................................................ 92
Figure 6-2. a. Comb pattern ceramic PCB board with IDE. b. Schematic of comb pattern board. C1
and C2 represent the capacitors circled by red and black dotted lines in the PCB board. R1 and R2
represent virtual resistance in parallel with capacitors. L1 is the winding coil on ferrite. Z1 and Z2 are
Zener diodes. ............................................................................................................................................... 94
Figure 6-3. a. Two coil wireless query system for resonant frequency and bandwidth detection. The
powering coil (2 turns of 18 AWG insulated copper wire, 37.5 mm in diameter) and the detecting coil
(30 turns 22 AWG insulated copper wire, 35 mm in diameter) were placed orthogonally to minimize
direct electromagnetic coupling between them. The test article was placed at 45° angle between them
so that it coupled to both coils. b. Oscillogram of self-resonant frequency and bandwidth (green trace)
as measured in wireless detection, sweeping from 16-24MHz. Generated bias DC voltage (yellow trace)
on the diodes as measured by needle probes is correlated to device’s resonant properties. ....................... 97
Figure 6-4. a. Measured bias DC voltage as self-resonant frequency changes. B Measures bias DC
voltage as Q factor changes. ....................................................................................................................... 99
Figure 6-5. a. Accelerated life test system for both comb pattern devices and the microstimulator. b.
Accelerated life-testing vial for the microstimulator to convert stimulation pulses into visible light
flashes. The rubber tube acts as an O-ring to force the output stimulation current through the
surrounding saline and into the Pt/Ir wire electrodes, which do not touch the output electrodes of the
microstimulator. ........................................................................................................................................ 101
Figure 6-6. Self-resonant frequency measurements of all comb pattern devices as a function of number
of days in accelerated life-testing.............................................................................................................. 102
Figure 6-7. a. Self-resosnant frequency measurment of comb pattern devices with defects as function
of days in testing. Marker A and B represent day of observation with blistering (Fig. 8d and e) and
delamination on gold trace in one device with huge bubble (Fig. 8f). b. Calculated Q factor as function
of days in testing. ...................................................................................................................................... 104
Figure 6-8. a. Typical corrosion on IDE with finger print oil contamination. b. Typical corrosion on
IDE with adhesive contamination. c. Increased corrosion area over time. d and e. The giant bubble
filled with water and blisters forming on the gold traces. Markers A: optical microscopy correlated
12
with measured resonant frequency in Fig. 7a. f. Further delamination formed along the gold trace.
Marker B: optical microscopy was correlated with measured resonant frequency in Fig. 7a. ................. 106
Figure 6-9. Number of survived devices in deionized clean group (Deion. clean), distilled clean group
(Dist. Clean) and neurostimulator group as a function of accelerated testing days (normalized to
number of devices starting soaking). The x-axis in red represents expected life in months at 37°C,
assuming only Arrhenius temperature acceleration. ................................................................................. 108
Figure 6-10. a, b and c. typical corrosion of 100, 220 and 350 days after soaking respectively, visible
only on anodally-polarized gold traces. .................................................................................................... 108
Figure 7-1. Assembled insertion tool. Patient position and implantation angle. ..................................... 120
Figure 7-2. Implantation procedure. A. Identify target location using EMG needle. B. Confirm target
location using assembled insertion tool. C. Re-confirm location before release of NuStim. D. Release
the NuStim. ............................................................................................................................................... 122
Figure 8-1. NuStim regulatory pathway and timeline. .............................................................................. 129
13
ABSTRACT
Stress urinary incontinence is a common problem in which there is involuntary urine leakage during
activities that increase abdominal pressure. Although it is not a life-threatening problem, it has high
prevalence, affecting millions of elderly people. Most patients with this problem have weakness or damage
of the external urethral sphincter and pelvic floor muscle as a result of childbirth or prostate surgery.
Invasive surgical treatment such as vaginal tape and hydraulic sphincters have significant long term-
complications. Voluntary muscle exercise program have been documented to produce excellent results but
most patients find it difficult to perform such exercises correctly or consistently enough to obtain such
results. Previous attempts to induce muscle exercise by electrical stimulation failed due either to unpleasant
sensation from external electrodes, or to connection failure of long leads to conventional, fully implantable
stimulators. Intramuscular electrical stimulation via a minimally invasively implanted microstimulator
should overcome these challenges.
We have developed a single channel, monolithic microstimulator (3.5mm diameter * 10 mm long) that can
be implanted close to targeted motor axons via minimally invasive procedure to generate charge-regulated
pulses for strong muscle contraction. The implanted device receives stimulus power and timing by inductive
coupling from a radio frequency transmitter in a seat cushion. The physician and patient can use an app in
a tablet computer or smartphone to prescribe exercise patterns and control and record daily use via
Bluetooth communication to the seat cushion.
Preclinical testing demonstrated that the implant produces the desired output pulse (0.05-2.8µC) up to 12cm
from the face of the seat cushion. Accelerated life testing in vitro of the non-hermetic epoxy encapsulation
demonstrated expected lifetime in vivo > 1 year. Chronically stimulated implants in an animal study reliably
activated skeletal muscle without apparent discomfort and were well anchored by minimally reactive
connective tissue encapsulation after 1-3 months. The validated system is now being used in a clinical trial
to demonstrate safety and efficacy. The primary outcome measure is a conventional pad weight test for
14
urine leakage during controlled exercise. The inert, passive implants are intended to be left implanted after
3 months of daily, electrically induced exercise.
15
CHAPTER 1: INTRODUCTION
MOTIVATION
It is difficult to develop a successful medical device for a novel medical application. The design,
development, and testing of such a device requires wide domain of knowledge, including electrical
engineering, mechanical engineering, software engineering, material science, regulatory science,
physiology, neuroscience, anatomy and surgical technique. The body of work presented in this thesis
provides an example of innovative medical device development. It starts from the background review of
the clinical problem, to technology design, prototyping, and testing in vitro and in vivo, concluding with
the plan for and first results of clinical validation.
OUTLINE
Chapter 2. Review of Stress Urinary Incontinence - Clinical Background
This chapter is a background review of stress urinary incontinence (SUI) for both female and male. The
prevalence of the disease, mechanism of continence and incontinence, etiology and the state of art of various
treatment options are provided in the chapter. The clinical need for improved treatment of SUI is identified.
Chapter 3. Neuromuscular Stimulation
This chapter reviews the underlying neuromuscular physiology, which enables an alternative method for
muscle training by using neurostimulation. It addresses concerns in efficacy and safety about how to
produce a practical electrical stimulation of motor axon for a strong muscle contraction without damage.
The review of electrical stimulation also presents the historical design implementation of electrical
stimulation for stress urinary incontinence.
Chapter 4. Microstimulation Strategy
This chapter presents a feasibility test performed on human subjects. The objective of this acute experiment
was to understand the pelvic floor neurophysiology, location for stimulation, and required charge for
16
stimulation. Technological design requirements were then extracted for the design of a miniaturized and
percutaneously implantable neurostimulator for treatment of stress urinary incontinence
Chapter 5. Neurostimulation Strategy for Stress Urinary Incontinence
This chapter provides the details of engineering work to bring the device into realization. Preliminary data
from in vitro and in vivo study are also presented to set the stage for future experimental direction.
Chapter 6. Accelerated Life-Test Methods and Results for Implantable Electronic Devices with
Adhesive Encapsulation
This chapter delves deeper into a major technical question related to the reliability of the non-hermetic
epoxy encapsulation. It presents implant packaging lifetime assessment and evaluation, including a design
and validation of a highly sensitive wireless detection system for accelerated life testing. Quantitative and
qualitative projection of reliability of the device are provided to address the chief concern about the lifetime
of non-hermetic epoxy encapsulation.
Chapter 7. Proposal for Clinical Validation
This chapter describes ongoing demonstration of the efficacy and safety of the NuStim system in a clinical
trial. It is extracted from the protocol that was submitted for IRB review. The questions that should be
answered during the clinical trial are listed in the proposal. It provides the overall design of the clinical trial.
Chapter 8. Translational and Regulatory Paths
This chapter presents the required regulatory strategy and timeline for the CFDA approval of the NuStim
system. The goal of this is to provide guidance for technological translation from United States, where the
NuStim technology was developed, to China, where the NuStim system will be manufactured and marketed.
Chapter 9. Pilot Clinical Study
This chapter presents the collected data and results of the ongoing pilot clinical study of the NuStim system.
The discussion presents the limitation of the pilot study and plan for future validation.
17
Chapter 10. Conclusion and Future Directions
This chapter summarizes the results from this body of work and proposed adaptation of current technology
to address additional clinical problems.
18
CHAPTER 2: REVIEW OF STRESS URINARY
INCONTINENCE CLINICAL BACKGROUND
INTRODUCTION
Unexpected urine leakage is a common problem known as urinary incontinence (UI) among adults. Stress
urinary incontinence (SUI) is one common type of incontinence with an increase of abdominal pressure
during physical activities. It is the result of the sphincter pelvic muscles, which support the bladder and
urethra, are weakened. SUI occurs when urethral sphincter is no longer able to resist the flow of urine from
the bladder during increased abdominal pressure during activities such as coughing, sneezing or laughing
(Figure 1). The weak muscle in the pelvic floor is unable to maintain its normal function, so these patients
must wear absorptive undergarments or adult diapers in their daily life. Although there are various available
treatment options for SUI in females, limitations or side effects from those treatments leave an opportunity
for further improvement. The SUI is also a common problem for male patient after the prostate removal
surgery. Currently, there are only limited treatment options for the urine leakage due to physiological
limitations.
Figure 2-1. Diagram of SUI. The external urethral sphincter as part of pelvic floor muscle is atrophied.
19
MECHANISM OF CONTINENCE AND INCONTINENCE
The mechanism of micturition control is determined by bladder and urethra coordination[1]. In the bladder
filling phase, the detrusor smooth muscle on the bladder is suppressed to allow bladder expansion with an
increase of urine. In order to maintain continence, the bladder outlet and urethra must be closed at a pressure
that exceeds the bladder pressure. In a normal healthy subject, the urethral pressure can be increased
promptly in an involuntary dynamic reflex to overcome the sudden increase of bladder pressure during
activities such as coughing and sneezing. During voluntary micturition, the contraction of the bladder and
relaxation of the pelvic floor musculature and urinary sphincter permit urine flow through the urethra. The
urethra is kept opened until the completion of bladder emptying. A simple analogy for the mechanism is a
balloon for fluid storage and pipe with a switch to open for fluid drainage or closed to hold the capacity in
the filling or emptying phases, respectively.
The dominant element for the urethral closure function is the urethral sphincter, which contains layers of
submucosal vasculature, smooth muscles and striated muscles. Steady muscle tone is provided by the slow
twitch muscle for urethral closure at rest; this is the internal urethral sphincter. Fast twitch muscle provides
additional closure force in a condition when there is a rise in abdominal pressure; this is the external urethral
sphincter. The pelvic floor muscles (PFM) around the urethra are comprised of three-layer muscular plates
that provide structural support for the pelvic organs. The muscles described have different muscle fiber
directions, and if each region of the muscles could contract separately, they would all have different
functions. However, there is only one known voluntary contraction pattern of these muscles as a mass
inward lifting contraction that simultaneously squeezes the urethra, vagina, and anus [2].
SUI occurs when the external urethral sphincter does not generate sufficient pressure on the urethra to
prevent urine flow when bladder pressure is high, either from the volume of urine or from increases in intra-
abdominal pressure generated by physical activity. The decrease in muscle function, strength and structural
support in SUI patients compared to normal subjects were well demonstrated in various studies that
measured vaginal squeeze pressure, EMG, ultrasound and MRI[2]. As described previously, the neural
pathways for controls of continence and incontinence are organized as a simple on-off switching circuit for
20
fluid filling and emptying. Any damage in the circuit that affects neural signal transmission or muscle
control could lead to weakness of muscle function. All possible pathophysiological mechanisms that could
result in such weakness are shown in the table 2-2. Damage of pelvic floor muscle and motor axons from
trauma directly lead to muscle weakness. Signal transmission through motor axon could be interrupted by
diseases like diabetes and multiple sclerosis which attack nerve fibers and neurons. Such diseases affect the
entire neural system non-selectively which also affect afferent and efferent nerve fibers for bladder control,
therefore usually developing other type of bladder problems, mixed with SUI. Loss of inputs from upper
brain control is seen in complete spinal cord injury which results in muscle atrophy. It also causes detrusor
over activity and detrusor-sphincter-dyssynergia. Depending on the level of spinal cord injury, the effects
on neural pathways and developed bladder problems will vary.
The above interruptions in neural pathways sometimes may develop another type of common urine leakage
problem, known as urge incontinence. It is an involuntary urine leakage with a sudden unstoppable urgency
to urine, which mostly relates to overactive detrusor and strong bladder contraction. Patients with urge
incontinence and stress incontinence have combined symptoms from both known as mixed incontinence.
Depend on the diagnosis of type and cause of incontinence, treatment plan will be adjusted for each
patient[3]. In the thesis, we mainly focus on pure SUI caused by damage or aging on pelvic floor muscle
and motor axons, in which most of neural innervations and circuits are intact.
Pathophysiology Possible reasons
Damage to muscle Trauma
Damage to motor axons Trauma, diabetes
Failure of neuromuscular transmission Myasthenia gravis, medication
Loss of motoneurons in spinal cord Multiple sclerosis
Loss of inputs to motoneurons in spinal cord Spinal cord injury, multiple sclerosis
Table 2-1. Possible pathophysiological mechanisms for weakness of pelvic floor muscle and urethral
sphincter.
21
INCIDENCE
The estimated prevalence of urinary incontinence in women varies between 25%-45% in several studies
[4]. Isolated stress urinary incontinence accounts for about half of all urinary incontinence[5, 6]. Prevalence
increases in middle age as shown (Figure 2-2). Differences in prevalence can be attributed to different
population samples, definitions and measurements of UI. Most studies consider patient survey reports of
any leakage in the past 12 months. However, these estimations can be under-reported due to the social
embarrassment associated with the condition. Only a small number of patients have searched for medical
help, usually when the problem has become unbearable. Most patients have a tendency to accept
incontinence as a normal part of aging that is annoying but not life threating.
Figure 2-2. Prevalence of urinary incontinence (and leakage) in women 20 years+ (Reprint from 2005
Hunskaar research)[4].
ETIOLOGY
The etiology of poor urethral function in female SUI is not completely understood. SUI is often seen in
women who have had more than one pregnancy and vaginal delivery. It is also common in women whose
bladder, urethra, or rectal wall protrude into the vagina (pelvic prolapse)[7]. Identifiable risk factors for this
condition include pregnancy, childbirth, menopause, cognitive impairment, obesity and advanced age.
22
The vaginal delivery has been found the most prominent risk factors for the SUI. It causes laxity because
of muscle and connective tissue stretching, acute damage to the surrounding muscles, which is then replaced
by scar tissue, and nerve damage.
Aging is a normal process that causes density loss in muscle, at least in part because of greatly reduced
estrogen levels after menopause. The striated muscle fibers within the ventral wall of vagina were found to
have a 70% muscle fibers loss as women progress from 15 to 80 years of age[7]. The decrease of cross
section area of the muscle and loss of motor neurons contribute to the loss of muscle power and bulk.
Although some denervated muscle fibers may be innervated again by the adjacent motor neuron, the general
decrease is inevitable.
Urinary incontinence can lead to severe medical problems including perineal rash, pressure ulcers and
urinary tract infections. It is an also undeniable social problem, creating embarrassment and negative self-
perception. It was found to reduce both social interactions and physical activities, is associated with poor
self-rated health, impaired emotional and psychological well-being and impaired sexual relationships.
Compared to general patients, an adding urinary incontinence to any other medical problem increases the
need for nursing care and the risk of developing a complication.
CLINICAL DIAGNOSIS AND EVALUATION
The purpose of clinical diagnosis for SUI is to understand the cause of incontinence and exclude subjects
with urge incontinence, overactive bladder, and mixed incontinence. The general procedure for diagnosis
includes reviewing medical history, a physical examination, post-void residual urine volume, and urinalysis.
Patients with history of SUI usually have urine leakage during physical activities. The onset of leakage, the
history of liquid intake prior to leakage, frequency, volume of leakage, and numbers of pads used can be
usually acquired in the patient interview to give the physician some basic ideas of the cause of UI. A 24- or
72-hours urinary diary describing leakage can help to quantify the symptoms. The physical examination
includes a pelvic examination with assessment of pelvic organ supports and prolapse severity. A
23
neurological examination detects the integrity of neural functions of the bladder and pelvic floor muscle.
The cough test or cystometry can be performed to confirm the diagnosis of SUI. Urine leakage pad testing
is the most commonly used objective assessment for severity and response to treatment. The test asks
patients to take a known quantity of liquid and perform a series of standardized activities while wearing a
pre-weighed absorptive pad. The total urine leakage is determined from the increased weight immediately
after the test.
CHOICE OF INTERVENTION
Various treatment options are available for SUI patients. The choice of the intervention depends on the
severity of the problem and the cost-benefit justification. Most patients begin with a conservative treatment.
Pharmacological treatments and invasive surgeries are only considered for when the condition becomes
unbearable. There are pros and cons for different types of options. The purpose of listing all available
choices of interventions is to compare the efficacy and safety.
CONSERVATIVE TREATMENT
BEHAVIOR CHANGES
For most patients, conservative treatment should always be considered and offered as the primary
noninvasive interventions for SUI symptoms management. The purpose of the conservative treatment is to
increase awareness of bladder control and to alter physical activities. During the consultation, patients
usually have the chance to change their attitude about incontinence, learn about different types of
incontinence, risk factors and treatment options. It helps to alter erroneous beliefs in most patients about
accepting incontinence as a normal part of aging. The patients in conservative treatment are usually
instructed to use a urinary diary to self-manage the symptoms. It allows patients to realize if they are
consuming excessive fluids and to modify their behavior accordingly. Also, it aims to increase the interval
between voids and increase void volume gradually. The change in behavior includes having the patients
24
empty the bladder before physical activities, decrease fluid and caffeinated beverage intake, and wear
absorptive undergarments. Because behavior change is mostly instructed as a combined intervention with
other options such as pelvic floor muscle training, it is not clear that how much behavior change itself
contributes to incontinence improvement. In most case studies, the self-management of behavior change is
usually used as a control group in the study that achieves limited improvement compared to other treatment
groups[8]. Smoking and obesity are considered as two risk factors for SUI and so the cessation of smoking
and weight loss are recommended for SUI alleviation. Weight loss has been found to have significant
contributions for UI improvement. A 5% reduction in body weight resulted in a decrease of 50% leakage
frequency[9]. It is obvious that weight reduction decreases stress on the bladder and pelvic floor muscle,
thereby reducing incontinence. There are no studies regarding the effect of cessation of smoking on
incontinence improvement.
EXTERNAL MECHANICAL DEVICES
Although mechanical devices are not widely accepted in the market due to infections and compliance, they
are marketed as a nonsurgical correction for SUI. They can be divided into several categories that aim to
provide mechanical support for the bladder neck or block the urethral opening. The most commonly used
mechanical devices are pessaries placed in the vagina to hold the bladder neck in position. Other less
common vaginal support devices are similar in the mechanism to provide upward support. Intraurethral
devices are inserted inside the urethra and replaced each time after voiding. The last common type of
mechanical devices are urethral occlusive devices that are placed on the outside urethral opening and
temporally sealed by gel. There are only limited valuable studies available for the effects of the mechanical
devices, likely due to limited patient acceptance[10]. Compared to other conservative treatments,
mechanical devices only provide external solution similar to absorptive pads, which do not improve
intrinsic function of the muscle or solve the problem internally. Compared to surgical treatment, they
require extra effort of regular replacement, which may induce additional problems like pain or infection.
25
PELVIC FLOOR MUSCLE TRAINING
Pelvic floor muscle training (PFMT, also known as Kegel exercise) is the most recommended first-line
therapy for stress urinary incontinence. Like any striated skeletal muscle, the external urethral sphincter
(EUS) muscle and pelvic floor muscle with intact neuromuscular innervation respond to exercise by
increasing its bulk and strength. The principle of the PFMT is to directly address the problem via repetitive
contraction of the pelvic floor muscle, which builds strength and perineal support. This is the basis of PFMT,
which is highly effective with up to 70% improvement when done properly and conscientiously [2, 11, 12].
The rationale that the PFMT is effective in treatment is that:
1. The bladder neck is supported by the strong pelvic floor muscle, thereby limiting its downward
movement during effort and exertion, which prevent leakage[2, 13].
2. The strength training builds up muscle bulk to provide structural support and elevate the levator
muscle plate, thereby enhancing hypertrophy and stiffness of connective tissues[2].
3. The PFMT helps patients to build up skill to consciously contract the pelvic floor muscle prior to
a physical stress and to maintain the contraction during the stress. It is known as ‘the knack’
mechanism to re-educate patients’ self-care awareness, which has been found highly effective.
After one week of training, Miller et al. found it reduced about 98% leakage with a moderate cough
and about 73% with a deep cough[14]. Although the automatic activation of pelvic floor muscle
before and during the physical exertion seems to be a natural reflex without conscious efforts in
continence women, the ‘Knack’ may work as an enforcement learning for patients to acquire well-
timed, fast and strong voluntary contraction.
4. It directly increases the muscle tone and clamping force of the urethra[15].
The exercise is a cyclical, voluntary squeeze and relaxation of the pelvic floor for a few seconds with a
recommended repetition up to 1 hour a day by Dr. Kegel [16]. There is no gold standard for how to perform
the best PFMT. The endurance exercise program may help to build up slow-twitch muscle to improve the
26
urethral muscle tone. The strength program with high load and low repetition may increase muscle bulk to
respond to increased abdominal pressure efficiently. The number of exercises recommended across the
studies ranged from 8 to 200 contractions, and repeated for up to four times a day. The basic concept is to
receive a significant improvement of muscle function which requires proper instruction and regular and
persistent exercise for at least several months. When the training is initially successful, the benefits seem
to persist[17-20]. Nevertheless, PFMT is difficult to do correctly and few patients are trained properly or
sufficiently disciplined to do the exercise often and long enough to obtain substantial benefit. Thus, patients
with SUI are often subjected to surgical procedures or pharmacological treatments rather than this
conservative and effective treatment.
In order to acquire full benefit of PFMT, several accessories have been added into the treatment to teach
the exercise and enforce consistent daily training. These types of accessories include vaginal cones,
biofeedback, and electrical stimulation.
VAGINAL CONE
Vaginal cones come into different sizes and weights that are placed intravaginally during the PFMT. The
patients learn to retain the cone in the vagina during exercise. Once the patient can retain the cone with no
active contraction effort, a heavier cone should be used which requires a stronger voluntary pelvic floor
contraction to prevent the cone from slipping out of the vagina. Based on previous clinical trials, there is
no significant difference in efficacy, including the satisfaction of the treatment, decrease in urinary leakage
and number of pad usage, between patients who performed PFMT only or PFMT with the addition of a
vaginal cone[21, 22]. Although the pelvic floor muscle contraction pressure was found to be higher in
patients who used vaginal cone compared to the group with PFMT only, there is no clear relationship
between the contraction pressure and urine leakage. Meanwhile, patients who receive either PFMT only or
vaginal cone with PFMT seem to both maintain the improvement in long term[23].
27
BIOFEEDBACK
In the current standard of care, most patients with uncomplicated SUI are assigned to a physical therapist
after systematic physical exams and tests. In order to learn how to perform correct PFMT, physical therapy
education sessions are important to help patients to identify pelvic floor muscles and learn how to contract
and relax the correct muscles selectively. A biofeedback system is usually used during the education session
to assist with real time feedback from surface electromyography (sEMG). The sEMG biofeedback relies on
a vaginal probe with surface electrodes to measure the recruitment of activated motor units, which allows
for consistent measurement of muscle activities with the patient remaining static or dynamic. It translates
pelvic muscle activity to the patient in a readily understandable signal that is particularly helpful for
identifying and isolating the correct muscle. To achieve long-term full benefit of PFMT, the patient is
required to make multiple visits to a clinic with biofeedback instrumentation and teaching assistance and
encouraged to perform the same PFMT in home without such feedback. Most studies reported significant
improvement after several months of treatment. In a Dannecker et.al study, 60% of patients prior to study
had severe incontinence with jet-leakage in the supine position, whereas directly after the treatment only
5% had a severe condition and 50% of patients had no leakage during coughing[24]. One short term study
showed a cure rate of 50% to 69% for PFMT with and without biofeedback, respectively[25]. Nevertheless,
no statistical significance has been found to claim that the addition of biofeedback to PFMF for SUI has
additional effects, which agrees with other previous researchers.
A perineometer is another type of biofeedback device that detects the vaginal squeeze pressure. Compared
to sEMG biofeedback, pressure monitor biofeedback is not widely accepted in the clinical setting due to
the ambiguous pressure measurement associated with muscle stiffness/tightness and inability to
differentiate pelvic floor muscle contraction from increased abdominal pressure or gluteal squeezing [26].
With the help of biofeedback, patients may get better at producing the correct contraction of their pelvic
floor muscles, but the adherence to treatment and motivation are still a problem for PFMT. In one study,
128 patients out of 390 patients broke off the therapy prematurely, mostly due to lack of time and
motivation[24].
28
ELECTRICAL STIMULATION
Electrical stimulation is another alternative to treat stress urinary incontinence. Like any striated skeletal
muscle, electrical stimulation can be used to directly activate the motor units of the external sphincter,
which should provide benefits similar to voluntarily exercise. Electrical stimulation can be used externally
or internally for short term or chronic stimulation with various reported ranges of cure rates and
improvement. The details of our implanted neuromuscular stimulation strategy for stress urinary
incontinence is described in the next several chapters. The electrical stimulation described here is the current,
clinically available stimulator used externally, typically in the vagina. The efficacy of using transvaginal
electrical stimulation for stress incontinence is unclear. No significant improvement in stress incontinence
was demonstrated between the sham group and active group in Brubaker’s study, but they found surprising
improvement in detrusor overactivity[27]. Only limited improvement was reported by Bo et al, but they
received frequent complaints from patients about lack of motivation, difficulty in use, and discomfort from
the stimulation[8]. Some recent studies reported similar improvement between PFMT and electrical
stimulation[21]. The lack of consistent improvements from electrical stimulation limits the practice of
combining electrical stimulation with conventional PFMT. The discomfort also limits the acceptance by
physicians and patients. The inconsistency in outcomes of electrical stimulation can be attributed to the
following reasons:
1. The placement of the probe, the parameters of the stimulation, and subjective sensation tolerance
are different among patients.
2. The muscle nerves lie under highly innervated skin and mucosa, so transcutaneous or intravaginal
or intrarectal stimulation necessarily tends to produce unpleasant sensations.
3. The location and innervation of the target muscle differs among patients, so some patients may
require strong but discomforting stimulation to produce sufficient contraction.
4. The damage from childbirth and formation of connective tissue around the vagina increases the
impedance for electrical current conduction, which may reduce the muscle response to stimulation.
29
PHARMACOLOGICAL THERAPY
Pharmacological therapy achieves varying degrees of success with substantial side effects. Medication
produces some improvement in mild cases but rarely brings about total dryness in cases of severe or even
moderate SUI. There are several types of pharmacological treatment options available that may be used
alone or in combination.
ANTICHOLINERGIC DRUGS
Anticholinergic drugs are one type of drug (oxybutynin, tolterodine, Enablex, Sanctura, Vesicare, Oxytrol)
to control overactive bladder. It inhibits the receptors of the neurotransmitter acetylcholine, thereby
suppressing the involuntary smooth muscle contraction in the bladder. It has been recommended as a
medication option for overactive and urge incontinence patients, but the efficacy for SUI is doubtful. It may
be useful for patients with mixed incontinence who have both weakness of the urethral sphincter muscle
and overactive bladder. One previous systematic review of 32 clinical studies of using anticholinergic drugs
for patients with overactive bladder have showed only statistically significant but with only small
improvements compared to the placebo group, regardless of whether the drugs were applied orally or
intravesically[28]. Side effects of the anticholinergic drugs usually were not well tolerated by patients with
about 10 to 20% drop-out rates, since it has other interferences to the central and/or peripheral nervous
system. The reported maximal cytometric bladder capacity only increased slightly from 43 ml to 66 ml,
compared to the normal bladder capacity (usually more than about 250 to 300 ml). Patients with SUI usually
have normal bladder capacity, so may only acquire minor benefits from the small increase of relaxation of
the bladder with an anticholinergic drug.
30
ALPHA-ADRENERGIC AGONIST DRUGS
Alpha-adrenergic agonist drugs such as ephedrine, phenylpropanolamine and pseudoephedrine (common
ingredients in over-the-counter cold medications), have a high affinity to alpha-adrenergic receptors in the
bladder neck and proximal urethra that, when applied, induce urethral smooth muscle contractions[29]. It
was found to be useful to increase urethra outlet resistant around the bladder neck and urethra which
contains high concentrations of alpha-adrenergic receptors in both voiding and filling period. Several
clinical studies have reported limited improvement in patients. However, the off-label use of those drugs
are rarely prescribed because of possible side effects in stimulation of the central nerve system, including
elevated blood pressure, cardiac arrhythmias, increased risk of hemorrhagic stroke, and respiratory
difficulties[30, 31].
ANTIDEPRESSANT DRUGS
Some antidepressant medications have been found to enhance the muscle tone of the proximal urethra
muscle. Duloxetine is one of the known drugs used to treat depression, and has been recommended off-
label to treat stress urinary incontinence[32]. It inhibits serotonin (5-HT) and noradrenaline (NA) reuptake
in the presynaptic neuron which leads to serotonin and norepinephrine elevation in the synaptic cleft, acts
at Onuf’s nucleus and increases the excitability of the pudendal nerve[33]. A previous study reported to
enhancement of striated sphincter muscle activity in animal model [34]. However, human trails only
showed only about 50% improvement in incontinence episode frequency compared to the 27 to 40% of
placebo effects in the control group[34, 35]. Although one report showed the combination of PFMT with
duloxetine improved the condition significantly compared to using PFMT or duloxetine alone[36], more
large trials, further analysis, and long term outcome measurement are needed to support the claim. Besides
its common side effects, including nausea, fatigue, dry mouth and constipation, duloxetine was associated
with higher than expected suicidal attempts in those patients. Another drug is imipramine, tricyclic
antidepressant, has some properties similar to alpha-agonist and anticholinergic drugs [37]. SUI is not an
FDA-approved indication for antidepressant drugs.
31
ESTROGEN
Estrogen replacement therapy is a controversial hormonal therapy for SUI patients who have gone through
menopause. The onset and/or worsening of SUI peaks clinically during menopause. Although some studies
argued that this may be a general consequence of aging rather than hormonal [38], the decline of hormones
has a direct impact on the functional and physiological changes of genital tract and its surrounding tissues.
The pelvic floor has a high concentration of estrogen receptors [39]. It is clear that the estrogen stimulates
cellular proliferation and blood floor in the vagina, urethra and bladder, increasing maximal closing
pressure and increasing the concentration and sensitivity of several neuromuscular receptors[38]. However,
there are no consistent results showing the loss of estrogen as a contributor for SUI[40]. Meanwhile, clinical
trials using different preparations, dosage, and routes of administration do not demonstrate strong efficacy
compared to a placebo. Some studies even showed no effects on SUI or worsening of symptoms[41]. Some
hormone treatments given after menopause have been shown to be more harmful than helpful to women's
health. Women who have a history of breast or uterine cancer should not use estrogen therapy for any
purpose. Estrogen therapy has been associated with an increased incidence of breast cancer as well as
changes in blood coagulation and cardiovascular diseases. Estrogen withdrawal at menopause may
contribute to the increased severity of stress incontinence with age, but it is difficult to separate from the
general tissue atrophy associated with aging and it may not be reversible with hormone replacement therapy.
SURGICAL INTERVENTIONS
Surgery is recommended for many patients who fail to respond to an initial trial of conservative therapy.
The surgical treatment causes minimal to moderate pain and discomfort, and it is an invasive surgical
procedure with general risk of complications from anesthesia and wound infections. There are also specific
complications such as inadvertent mechanical damage to the bladder or urethra (perforation), or difficulty
in urinating after surgery [42-44]. The most commonly used surgical procedure known as tension-free
vaginal tape (TVT) suspension has an initially high success rate (75-95%), but incontinence may recur over
32
time and require repeat surgery with significantly more adverse events and less success than the first
procedure [31]. Recently, some implanted tape materials have been voluntarily recalled after increased
reports of severe long-term complications including erosion and perforation of adjacent tissues such as the
vaginal wall [45-47]. Retropubic urethropexy and suburethral slings are the most two common surgery
procedures. Recent modifications of those two original procedures seem to provide better outcomes but
long-term clinical data are limited.
SUI AND TREATMENT OPTIONS FOR FEMALE
There is no optimal treatment option for all patients with SUI. An appropriate intervention for a properly
motivated patient usually results in an adequate improvement. Based on the currently known information
about the available treatment options, a table with all listed treatment options is presented here to compare
the pros and cons, measured in a subjective analogue scale (Table 2-1). Treatment for non-complicated SUI
should usually start with conservative interventions to manage the condition and prevent further
deterioration. The behavior change with PFMT is highly effective for most patients with mild or moderate
conditions. The PFMT does not have any known side-effects. The main problem for PFMT is lack of
motivation for most patients to adhere to the repetitive muscle contraction sessions for three months.
Various tools are available to improve PFMT such as vaginal cone, biofeedback, and electrical stimulation.
Those tools seem to increase ability to activate the correct muscle and are useful for education, but do not
actually improve the efficacy significantly. The placement of those tools has potential side-effects for
uncomfortable or painful sensations or irritation.
Studies of the safety and efficacy of pharmacological treatments failed to demonstrate sufficient efficacy
to justify the high incidence of side effects. When conservative treatment for SUI fails, most patients
currently are referred for surgical treatment. This usually results in immediate change from incontinence to
complete dryness but may result in severe, long-term adverse events.
33
Intervention Effectiveness Side effects
Behavior change + +
Mechanical devices +++ +++
PFMT w/o guidance ++ +
PFMT w guidance ++++ +
PFMT plus vaginal cone ++++ ++
PFMT plus biofeedback ++++ ++
PMFT plus electrical stimulation ++++ +++
Pharmacological treatment ++ ++++
Surgery +++++ ++++
Level from + to +++++ refers to lowest level for minor or not obvious change with one +, and
most severe or effective with ++++++.
Table 2-2. Treatment options comparison among side effects and effectiveness.
SUI AND TREATMENT OPTIONS FOR MALE
Although SUI is rare in male patients, it is a major complication affecting about 5% to 30% of male patients
after radical prostatectomy surgery[48]. Patients with varying degrees of postoperative urine leakage
usually recover to normal continence in the first year after surgery. A small percentage (<5%) suffer from
persistent urine leakage beyond one year[48]. This problem is mainly caused by the decrease of urethral
closing pressure due to urethral sphincter damage and detrusor instability after the surgery[49, 50]. The
degree of damage to the relevant surrounding nerves mainly depends on details of the surgical procedures.
The nerve preservation procedure developed to reduce post-operative impotence has been found also to
decrease incontinence[51].
Similarly to the conservative treatment for female patients with SUI, PFMT is also recommended as the
first-line conservative treatment for the male patients with SUI after surgery. The exercise consists of
repetitive muscle contraction to train and strengthen the pelvic floor muscle, similar to PFMT prescribed to
female patients. It has been reported to be effective to accelerate symptomatic improvement several months
after surgery, compared to the patients without formal trained PFMT[52]. Review of most studies, however,
did not find any significant difference of incontinence improvement between group with or without PFMT
34
1 year after surgery[53]. This may be related to the very high variability of spontaneous recovery of
continence in this post-surgical population. Researches have tried various methods to improve the efficacy
from PFMT. There were studies to investigate preoperative biofeedback or brief instructed PFMT for
incontinence reduction, but the result showed no significant difference[54]. There were studies to compare
efficacy of postoperative electrical stimulation, biofeedback or PFMT for incontinence improvement, but
those aids did not enhance effects of PFMT after surgery after 3 months and 12 months[55]. There is no
study in which PFMT was performed on the selected patients with stable incontinence at least 1-year after
surgery.
Pharmacological treatment is considered in some cases, but the limited efficacy and side effects affect the
acceptance. Many efforts have been made to treat the incontinence surgically if the symptoms are
complicated or unbearable for patients. All currently available surgical interventions are invasive
procedures with operative complications, possible need of further surgical correction, and long-term
complications. The sling procedure is an available option for male patients. It has been reported to have
similar postoperative complications as the female sling, including erosion or infection. Symptomatic
improvement is highly effective if the sling is placed correctly the first time. However, a relatively high
percentage of patients require a second sling or more adjustments[56]; 40% of patients was reported in
previous studies to have additional surgery for sling position adjustment[57]. Injectable bulking agents
(collagen, stem cells) are another alternative method with varying success rates depending on the injection
volume, material degradation, and injection location. Improved efficacy has been shown in recent studies,
but may require further research into long term outcomes and safety demonstration[58]. Urinary artificial
urethral sphincter (AUS) has been recommended as a gold standard correction for male SUI but only if all
above solutions failed[59]. The expensive AUS is placed in as a major invasive surgery: a circular inflatable
cuff is placed around the urethra which is controlled by a pump connected to an air reservoir under skin.
Although satisfaction rate is high, there are severe complications of erosion, mechanical failure and
infection which require surgical revision[60].
35
SUMMARY
From the review of SUI for both female and male, PFMT appears to be a cost-effective intervention for
most patients compared to other options, at least in patients with damaged but sufficiently intact
neuromuscular anatomy to respond to the trophic effects of exercise. PFMT requires proper training and
consistent exercise to acquire maximal benefit from muscle exercise. The main limitation for PFMT is lack
of mechanism to assure treatment adherence and motivation. It may be possible to achieve the same exercise
improvement and solve the limitation by performing passive PFMT by neuromuscular electrical stimulation
(NMES). The commonly used external stimulation relies on electrodes placement in the vagina or anus,
which limits the acceptance due to inconvenience, uncomfortable sensations and limited muscle activation.
Implantable stimulator built like cardiac pacemakers with a bulky case and long lead have been effective
but often failed due to lead dislodgement, pain and infection around the implantation site[61].
36
CHAPTER 3: NEUROMUSCULAR STIMULATION
NEUROMUSCULAR PHYSIOLOGY
Skeletal muscles are composed on hundreds of motor units, each consisting of one motor neuron and a few
hundred muscle fibers that it controls. A typical motor behavior consists of many motor units working in
parallel to exert the force needed to move the body. The motor neurons activate their muscle fibers by
sending trains of action potentials at 5-30 pulses/s depending on the amount of force desired. Command
signals from the nervous system depolarize the motor neurons, which then generate action potentials that
propagate along the axon to its terminals in the neuromuscular synapses. The neurotransmitter acetylcholine
is then released, which causes an action potential to propagate along the cell membrane of the muscle fiber.
In response to the action potential, calcium sequestered within the muscle fiber is released to trigger
contractile protein binding and sliding. Because each motor axon branches and innervates hundreds of
muscle fibers, the contraction of all muscle fibers of a motor unit occurs simultaneously. The exact force
exerted by the muscle is determined by how many of its motor units are activated and the rate of the action
potentials in each.
Electrical stimulation applied inside a muscle results in generating action potentials in the local branches
of the motor neuron axons rather than direct stimulation of muscle fibers, which have a much higher
threshold to such stimulation. Stimulating any branch of an axon leads to the action potential spreading to
all muscle fibers innervated by that motor unit. Once the action potentials reach the muscle fibers, they are
indistinguishable from the activation that would occur with voluntary exercise. At the level of the whole
muscle and its response to voluntary exercise, there may be some differences, however. When muscles are
activated voluntarily, the motoneurons are recruited in a fixed order that depends on their physiological
properties and they fire asynchronously at rates of 5-30 pulses/s. When they are stimulated electrically, the
motoneurons are recruited in a different order based mostly on how close their axons are to the electrodes
and they fire synchronously at whatever stimulation rate is applied (typically 1-50 pulses/s). For equidistant
motor axons, the recruitment order of electrical stimulation is opposite of the physiological case in which
the smallest motor units are activated first. The large diameter axons that innervate the fast twitch, easily
37
fatigued muscle fibers are activated by lower strength electrical stimulation than the small, slower and
fatigue-resistant motor units.
As described below, muscle fibers can change their physiological type in response to different patterns of
both normally induced and electrically stimulated exercise.
ELECTRODE TISSUE INTERFACE
The electrode tissue interface provides for the transfer electrical current between the metal electrode and
the extracellular fluid in the body. When two electrodes with power source are in contact with body, it
forms a conductive closed loop. If a voltage is applied, the electrons are pulled away from one electrode
which becomes move positive, and aggregate on the other electrode which becomes move negative. The
electrical field created between two electrodes causes reorientation and distribution of ion and water
molecules in the extracellular fluid. There are two mechanisms for the transfer of charge in the interface,
known as non-Faradaic reactions and Faradaic reactions. The non-Faradaic reactions involves only
redistribution of charged particles, which is reversible. As shown in the electrical model of the electrode
tissue interface, this reaction can sustain only alternating current and is modeled as a capacitor. If too much
charge is applied in one phase of the AC waveform, non-Faradaic reactions will occur. These are non-
reversible chemical reactions associated with electrolysis and corrosion. The platinum iridium electrode
has relative high reversible charge-storage capacity, which is about 300-350 µC/cm
2
. Charge level below
this range is safe for electrical stimulation with only non-Faradaic reactions.
38
Figure 3-1. a. The electrode/electrolyte interface. Equivalent electrical circuit model of the interface
(reprinted from Merrill paper [62].
STRENGTH DURATION RELATIONSHIP
Strength-duration curve is the relationship between the amplitude of applied constant current pulse required
to initiate an action potential and the duration of the current (see Figure 3.2) It represents the threshold
value for stimulation. There are two important concepts in this curve that have been used to characterize
the tissue response. Firstly, as the pulse width increased, the required intensity of current decreases. The
minimum current for tissue excitation is called rheobase current Irh when the pulse width is very long. The
curve asymptotes to this value because the neurons themselves act like high pass filters that do not respond
to DC (unchanging) potential gradients induced by the stimulus current. Secondly, the required time for
action potential initiation at current amplitude twice of the rheobase is called chronaxie. It is generally most
efficient to provide stimulation at duration of chronaxie, which is described as required charge for excitation
as shown in figure of charge duration curve. The charge is simply the multiplication of duration of time and
amplitude of constant current for a square wave or the integral of current over time for an arbitrary
waveform. The minimum charge Qmin can be only achieved by pulse with very short duration, which is
about 0.1 µs in practice.
39
In general practice, the motor axons are stimulated to activate muscle instead of direct stimulation of muscle
fibers, whose electrical threshold is much higher. Stimulation of motor neurons allows action potential
propagation along the motor axons to activate all innervated muscle fibers. In contrast, direct muscle
stimulation only transmits an action potential within the muscle fiber, which produces only a local response.
There are many factors that affect the strength duration relationship, including the proximity between
electrode and nerve, tissue state, stimulation wave form, etc. For stimulation that is close to the motor axon
(less than 1 mm), the threshold for stimulation is about 0.1 milliampere current for a 100 µs long pulse,
which is about 0.1µC[63].
Figure 3-2. Example of strength duration curve and charge duration curve[62].
40
Figure 3-3. Experimentally determined strength duration relationship for motor nerve stimulation and
direction stimulation of the muscle [64].
PULSE POLARITY
Pulse polarity as another important parameter should be considered to provide effective electrical
stimulation, avoid tissue damage and electrode corrosion. Cathodic electrical stimulation is more effective
than anodic stimulation. The potential field generated from the cathodic (negative) stimulation directly
attract the cations (positive) away from the outside of cell membrane, which decreases the normally
negative potential difference between inside and outside of cell membrane. When this depolarization
reaches threshold for voltage-dependent sodium channels in the cell membrane, an action potential is
initiated. A continuously monophasic electrical stimulation will eventually build up the potential difference
between each electrode and the extracellular fluid until irreversible reactions such as electrolysis and
corrosion can occur. The most common used of biphasic pulses consists of a negative pulse for stimulation
and a positive pulse with equal but opposite charge to reverse these polarization potentials. Charge balance
and safety can be achieved with a monophasic pulse generator by adding a coupling capacitor in series with
the stimulator and the electrodes. The high-pass characteristic of a capacitor prevents direct DC current
41
injection, and guarantees charge-balance over time to avoid net charge accumulation as long as there is a
path to discharge the capacitor through the electrodes when the stimulation pulse is off.
HISTORICAL REVIEW OF NEUROMUSCULAR STIMULATION
Electrically controlled exercise has been used by researchers in experiments to understand the trophic
mechanisms that cause muscles to change their properties according to the demands of their usage[65-72].
Since 1960s, electrical stimulation has been largely adopted in both research and clinical settings as a
common rehabilitation/training method. The efficacy of electrical stimulation for functional improvement
of various disorders has been examined by several groups: for example, decreasing spasticity[73, 74],
preventing or reducing shoulder subluxation[75-77], preventing and treating shoulder pain[78], increasing
range of motion [79, 80].
Several neuromuscular stimulation technologies have been used with some inevitable shortcomings,
thereby limiting their application. External stimulators with percutaneously implanted electrodes and long
wire leads are subject to lead dislodgement, potential infection, and tissue damage. Fully implanted
multichannel stimulators require extensive surgery, which is undesirable, especially in early stages disease
with minor/moderate symptoms. Transcutaneous stimulators employ surface electrodes attached to the skin
over the muscles of interest, but this is usually not selective enough for individual muscle stimulation and
fine control. For deep and complete muscle contraction, the stimulation would need to be adjusted to a
strong intensity, which can produce unpleasant or even painful sensations related to the electrical
stimulation of sensory nerve fibers in the skin beneath the electrodes, thereby further limiting the
effectiveness and application scope. These electrodes are hard to maintain in the desired location and are
likely to cause skin irritation unless applied by an experienced therapist.
A solution to these problems was proposed 30 years ago and involved a small single-channel implantable
stimulator that can be inserted near the targeted nerve fiber. After decades of development, the BION™
provided solutions to most of these problems, and has been tested successfully in several clinical trials for
42
different applications. Detailed reports of scientific research results are available including
safety/biocompatibility testing in animal studies[63, 81, 82], device performance testing both in vitro [83]
and in vivo[84, 85], and clinical usability and efficacy testing in different clinical trials[86-91]. Sufficient
research experience of this revolutionary new technology facilitates development of similar devices and
paves the way for commercialization.
Figure 3-4. Photograph depicting three neuromuscular microstimulators (BIONs) in different packages.
(Reprinted from Jianguang Qiang paper with permission of the American Association of Neurological
Surgeons[92])
As an implantable medical device, biocompatibility and stimulation safety are always the primary
objectives to be examined. Miniature stimulator like the BION consists of a cylindrical glass package with
electrodes sealed on both ends. It is inserted via minimally invasive procedure into muscles or next to nerves.
Multiple scientific studies have been done in both the short term and long term to understand the device
performance and the body’s response. The first in vivo testing of 3 cats with 12 devices implanted around
43
the hind muscle with intermittent stimulation for 3 months found no detectable tissue changes or
damage[82]. Similar results were found in the following animal studies as well[81]. The tissue responses
were compared histologically to those provoked by active devices, failed devices and components. All
devices were encapsulated with a thin layer of connective tissue as the result of normal foreign response
without any migration in tissue. The minimal stimulation pulse strength required to elicit a muscle twitch
was very stable over time, indicating that the implant did not move within or damage the nearby tissues.
Necrosis of surrounding tissue was not observed due to the highly inert and biocompatible nature of the
materials chosen. The stability in the target after implantation provides stability of muscle fiber recruitment
over a long period of time. Implanting the device near the part of the muscle where its nerve entered made
it possible to produce a strong contraction by activating all or most of its motor units. As stimulation
parameters, intensity, frequency and/or pulse width are changed, muscle force and contraction type are also
adjusted until maximum contraction, known as tetanus, is reached.
All skeletal muscles have similar innervation and responses to implanted devices and electrical stimulation.
Most animal studies are done in hind limb muscles because it is easy to implant into a particular muscle
near its motor nerve entry site, which allows selective stimulation, contraction, observation, and post
mortem histological study. Such animal studies demonstrate that the device is safe to implant in any skeletal
muscle and that it is effective in producing strong contractions that will provide useful exercise[63]. Clinical
trials may then be conducted to determine whether such exercise will be useful to treat human patients with
specific clinical problems.
Since 1998, a total of about 150 first generation BION implants were implanted in about 100 patients in
eight small pilot studies in Italy, Canada and the US. The first clinical experience of BION implants was in
post-stroke shoulder subluxation in hemiplegic patient and knee osteoarthritis[88]. Measurements of the
shoulder study and osteoarthritis with subluxation reduction and knee pain decrease respectively showed
the outcome of treatment. Patients received treatment and were evaluated for 3 months but the implants
were left permanently in the patients without adverse events. Other potential applications include foot drop
correction of stroke patients, neuromodulation on pudendal nerve of urinary urge incontinence patients, and
44
ischial pressure ulcers prevention[87, 89, 90]. As originally described by Loeb and colleagues, the first
BION stimulators were developed with a hermetically sealed glass package and complex stimulus control.
Other versions of the BION relied on a ceramic package and incorporated a battery for usability in
ambulatory applications.
45
CHAPTER 4: MICROSTIMULATION STRATEGY
PREFACE
Electrically stimulated exercise seems to be a reasonable solution which may provide the hypertrophy effect
as voluntary exercise for stress urinary incontinence treatment[8, 93]. Because the muscle nerves lie under
highly innervated skin and mucosa, commercial stimulators that use transcutaneous, intravaginal, or
intrarectal electrodes are often unacceptable to patients due to somatosensory nerve excitation with
unpleasant sensations [8, 94]. Transcutaneous magnetic stimulation (similar to transcranial magnetic
stimulation, TMS) can induce eddy currents in the pelvis that are capable of activating muscle nerves, but
this requires high-power, expensive instrumentation in a clinic [10]. Intramuscular electrodes can excite the
terminal branches of motor axons with little or no sensation other than from the muscle contraction itself.
Conventional implantable stimulators first used by Caldwell (Figure 4-1) achieved efficacy for SUI without
adherence problems, but the bulky case and long leads required invasive surgery with a risk of infection
and high possibility of lead dislodgement [95].
Figure 4-1. a. An X-ray showing dislodgement of the leads in an early implant. b. Fracture of the electrode
(Reprint from 1966 Caldwell research)
46
Compared with previous NMES strategies, a fully implantable leadless stimulator could be a solution.
Although a similar fully implantable microstimulator (the BION®) was described 15 years ago [96, 97], it
has not been commercially available, partly because it utilized fairly expensive technologies that were not
cost-effective for applications like SUI that are problematic but not life-threatening. From the review of the
above chapters, it is possible to develop an injectable and wireless microstimulator that can be implanted
percutaneously into the pelvic floor muscles to generate strong contractions of the EUS without producing
unpleasant sensations or requiring voluntary effort.
PRECLINICAL STUDY
There are still several questions should be answered before the engineering work to build the complete
wireless microstimulator system. In order to investigate the neurophysiology, acute human studies were
performed to identify the feasibility in this study.
The study setup is shown in the Figure 4-2. It used the percutaneous needle electrodes to locate electrical
stimulation sites that recruited the EUS as measured by intra-urethral pressure catheter or by intramuscular
M-waves recording. The study was performed on total 6 subjects with 2 healthy males, 2 healthy females,
1 female spinal cord injury and 1 female with stress urinary continence. The selected site for stimulation,
details of patients’ information and testing results are shown in Figure 4-3 and Figure 4-4. The EUS could
be activated by stimulation in pelvic floor sites that were about 3-7cm deep in both male and female.
Perceptual and pressure thresholds were generally concurrent in the range of 0.2-1mA at 0.2ms cathodal
stimulation in female subjects. Unpleasant sensation was reported only in patients with urethral catheters
or with high strength stimulations above 7 mA, which may relate to perception of a strong EUS contraction.
The placement of stimulation electrode was not critical to produce strong muscle activation using pulses
within the comfortable range. Simultaneous bilateral stimulation at levels associated with increased catheter
pressure or increased sensation when stimulated unilaterally produced no obvious change in pressure or
subjective sensations.
47
The results are consistent with the original speculation about how these structures might be innervated. The
ease of eliciting EUS in pelvic floor area suggests that an action potential induced by stimulation of one
motor axon could easily propagate to EUS. The muscle nerves lying under the pelvic floor are known
originate from the pudendal nerve, which enters both sides of pelvic floor laterally with myelinated
branches covering the pelvic floor muscle and terminating laterally to the vagina and urethra. The absence
of synergistic effects from bilateral stimulation suggests that effective sphincter closure can be achieved by
unilateral stimulation, either because the motor units extend bilaterally or the sphincter is stiff enough that
unilateral contraction is effective or unilateral stimulation gives rise to contralateral reflex activation. The
effectiveness of unilateral stimulation with strong EUS activation suggests that a single-channel
percutaneous implantation can be used to produce reliably effective results. Stimulus charge will probably
be well below 1uC (5mA x 0.2ms), which should be easily achieved with low voltage circuits (<5V). The
target locations for implantation appear to be compatible with RF power transmission (shallow, oriented
approximately vertical to seating surface) and unlikely to interfere with coitus in the female.
48
Figure 4-2. The experiment set up of the acute human study. The catheter with a balloon is placed inside
the bladder. Constant flow is provided in the catheter with a small opening on the part of urethra where it
goes through the urethral sphincter. if the EUS has responded to electrical stimulation, this opening is
obstructed, which increases the pressure measurement. It should also be possible to record an M wave, the
electromyographical signal associated with synchronous action potentials in a large population of muscle
fibers, but this was obscured by the relatively large stimulus artefact in these experiments.
Figure 4-3. Electrodes placement and stimulation location for female and male subjects.
49
Figure 4-4. Details information about patients’ condition and testing results. It is observed to produce EUS
response as shown with increased pressure measurement, requiring only low stimulation level. For patient
with spinal cord injury, it may not be possible to produce sufficient response due to denervation of the EUS
muscle.
50
TECHNICAL AND SAFETY REQUIREMENTS
As described above, electrical stimulation can be used to activate directly the motor units of the EUS and
pelvic floor muscle, providing the same benefits as voluntary exercise, potentially with even higher levels
of recruitment and force than can be achieved voluntarily. The proposed microstimulator is designed to
stimulate the intramuscular motor axons directly so as to produce strong contractions without unpleasant
sensation or requiring voluntary efforts. It is designed for minimally invasive implantation, thereby greatly
reducing costs and risks against which the therapeutic benefits must be weighted. The implant can be
powered and controlled wirelessly to minimize the size of implant and need of battery for muscle exercise
in SUI application that only requires intermittent stimulation. The user interface should be friendly for
patient in-home use. The microstimulator is designed to be implanted chronically for muscle exercise over
a period of up to one year of PFMT. The requirements listed below are derived from this basic therapeutic
strategy.
THERAPEUTIC FUNCTION
STIMULUS SUFFICIENCY
The stimulation must recruit a substantial proportion of the motor units comprising the external urethral
sphincter to induce trophic effects. Stimulus trains at 20pps delivered to a single site in or near the external
urethral sphincter on one side can generate urethral pressures comparable to strong voluntary contractions.
Stimulus pulses with <1µC charge delivered over <0.2ms are generally effective[98]. It is not known
whether the therapeutic effects will require tetanic muscle forces or whether those will be well-tolerated.
Alternatively, useful trophic effects on hypotrophic muscle fibers may be achieved with low stimulus rates
that activate trophic effects mediated by intracellular calcium release[66, 68, 84].
51
STIMULUS CONTROL
Control of stimulus pulse intensity and repetition rate must be sufficient to achieve desirable therapeutic
effects for the expected range of electrode locations. The efficacy of a stimulus pulse for recruitment of a
spatially distributed group of myelinated motor axons is a function of stimulus charge rather than waveform
for pulse durations shorter than ~0.3ms = 2 time constants of the nerve fiber. In order to activate motor
axon at a 10 mm distance from the electrode, stimulation intensity at 10 mA with pulse duration about 0.2
ms is sufficient to produce half-maximal muscle response (shown in Figure 4-5). The pelvic floor muscles
have many fine, intramuscular nerve branches distributed throughout, so a range of stimulus strengths from
0.02 to 2µC and a coarse step resolution of 20% should be sufficient to control of wide range of muscle
contraction strength. In practice, stimulus intensity is best turned up to the maximal level that can be
tolerated without unpleasant sensations because that recruits the largest percentage of available muscle
fibers. Human slow-twitch muscle fibers generally fire at rates 5-40pps and force output saturates around
50pps.
52
Figure 4-5. a. Muscle recruitment curve. Distances (mm) from the nerve to electrode are given in the legend.
b. Effects of distance on the current for a half- maximal response. (Reprinted from Stein[98] )
PHYSICAL FORM AND PACKAGING
MINIMALLY INVASIVE IMPLANTATION
All implanted components must be delivered accurately to their desired location via minimally invasive
techniques. If the implant has a narrow cylindrical shape, it can be delivered through a large gauge needle
or cannula insertion tool.
WIRELESS OPERATION
The implanted components must be free of percutaneous wires or flexible leads to minimize the danger of
infection or failure through lead dislodgement or stress fatigue. Power and command signals can be
transmitted transcutaneously from an external transmitter by inductive coupling of a RF magnetic field. If
the implant has a suitable size and shape, the stimulating electrodes can be a monolithic part of the implant
body, avoiding leads and connections.
53
EASE OF USE
The graphical user interface is designed to be user-friendly for physician and patients. The first version of
the software can be installed on a one tablet that is used by both physician and patients. The physician can
prescribe exercise, test parameters and retrieve patient adherence data. The patient can use the same tablet
to control a stored exercise program for in-home exercise. The next version of the software will use a cloud
server that can be download easily and installed on any kind of tablet/cell phone. All patients’ data will be
uploaded automatically via the cellular network or internet and stored in a cloud server. The physician can
access these data to follow all patients and prescribe exercise.
RELIABILITY AND SAFETY
PROJECTED LIFE
Implant must provide electrically induced exercise for a period sufficient to realize therapeutic benefit. For
any skeletal muscle, substantial benefits from regular exercise can be realized in a matter of weeks to
months. As suggested by voluntary PFMT, substantial improvement can be achieved after 3 months of
exercise. Once strength has been built up to satisfactory levels, it should be retained by normal use in
neurologically intact patients. As suggested by voluntary PFMT, the effects appear to persist for years[2].
The ability to repeat a course of stimulation at any time in the future is always desirable, but function for
more than one year is not necessary for this clinical application. Implanted devices that are no longer needed
by the patient or are not functioning will not be surgically removed unless medically indicated. The NuStim
may be useful for other clinical applications that require longer term treatment. The relative ease of the
implantation procedure and small package size reduce the obligation to guarantee function for an indefinite
period of time, particularly if the failure of one implant does not preclude the continuing use of other
implants. In general, patients and physicians will expect at least several years of function from a chronically
implanted device. Accelerated life-testing methods are available to determine the statistical probability that
devices tested for a short period of time are likely to continue to operate under normal usage conditions for
much longer periods.
54
The main risk to a small implanted electronic device is moisture from the surrounding saline body fluids.
Water causes damage only after it condenses in voids or dissects through seams as a liquid, eventually
causing corrosion of metals, particularly accelerated by the presence of operating voltages on electrical
conductors in circuits. Liquid water has a high dielectric constant, so its physical presence can change the
electrical properties of reactive components such as inductors and capacitors. Sodium ions dissolved in
water easily “poison” semiconductor junctions by changing the charge distributions. All of these failure
mechanisms can be prevented either by hermetic enclosures that completely prevent inward diffusion of
water as liquid or vapor or by sufficient adhesion that prevent water vapor dissolved into and diffusing
through nonhermetic encapsulants from condensing on hydrophilic surfaces.
BIOCOMPATIBILITY
Passive: Chronically implanted microstimulators shall comply with all applicable regulations for
biocompatibility of Class III Medical Devices. They should provoke only a stable and benign long-term
foreign-body response. Failure modes that might affect electrical function should not expose toxic materials
or otherwise pose a threat to surrounding tissues.
Electrically Active: Chronic application of the strongest available stimulation pulses should not cause
electrochemical damage to either the microstimulator electrodes or the surrounding tissues. It is essential
for chronic use of any stimulating electrode that the stimulus pulses be charge-balanced, with no net direct
current. This is usually met by using a capacitor in series with the electrode. For the NuStim, it may be
desirable to use a capacitive charge-metering system, in which the maximal stimulus strength is given by
the charge stored by the regulated voltage on this coupling capacitor. Candidate electrode materials such
as platinum, iridium and stainless steel all have well-known charge-density limits to avoid electrochemical
damage[62]. The actual charge density is given by the stimulus charge divided by the real surface area of
the electrodes.
Mechanical: Repeated voluntary and electrically stimulated contractions of the muscle around or near the
implant must not pose a risk to its mechanical integrity or stable location. Previous experience with BION
55
implants identified that they are tightly anchored by surrounding connective tissue that is usually continuous
with the epimysial collagen. This connective tissue tends to keep the implant from migrating but it can
subject it to large mechanical forces, particularly if the implant has narrow regions that can flex between
wider regions that act as anchors for such connective tissue.
NuStim implants should withstand mechanical stresses related to occasional minor trauma over the implant
site (e.g. contusions from impact with external objects). The presence of the implant must not interfere
with normal sexual function. In the female, the desired stimulation site may be close to the vaginal wall.
The size, shape and location of the implanted microstimulator must not be detectable by either sexual
partner during coitus and must not put the implant or the surrounding tissues at risk of mechanical damage.
REPLACEMENT OPTIONS
It should be possible to remove surgically an implanted microstimulator that has failed prematurely and
needs to be replaced for therapeutic reasons or that poses an undesirable or risky foreign body in the patient.
ENVIRONMENTAL TOLERANCE
Devices should be able to maintain functionality over a normal range of environmental extremes that would
be encountered during shipment and storage and at least a limited number of cycles of re-sterilization.
Systems including devices should be able the resist breakage when dropped or handled with surgical
instruments or instruments that are provided
MRI COMPATIBILITY
The compatibility of the stimulator with magnetic resonance imaging must be determined and indicated on
the labeling. Any image artifact associated with the presence of a device should not interfere with the
interpretation of regions of clinical interest. If the presence of this device in the body is a contra-indication
to such imaging, there must be a means to inform the patient and any caregivers of any known hazards.
It should be relatively easy to design a muscle stimulator that will pose little, if any, hazard during clinical
MR imaging because the device will be small, the lead short, and there is no need for magnetic materials
56
other than a small amount of dielectric ferrite. Standard protocols for measuring such hazards over the range
of clinical MR field strengths are available:
• Devices should not generate spurious electrical output.
• Devices should not heat perceptibly.
• Magnetic forces on the devices should not exceed gravitational force.
• Devices should continue to work according to all electrical specifications following MR imaging.
• Thresholds for electrical stimulation of the muscle should be unchanged following MR imaging.
57
CHAPTER 5: NEUROSTIMULATION STRATEGY FOR
STRESS URINARY INCONTINENCE
© IEEE Transactions on Neural Systems and Rehabilitation Engineering. Reprint from:
X. Huang; K. Zheng; S. Kohan; P. M. Denprasert; L. Liao; G. E. Loeb, "Neurostimulation Strategy
for Stress Urinary Incontinence," in IEEE Transactions on Neural Systems and Rehabilitation
Engineering, vol. PP, no.99, pp.1-1
doi: 10.1109/TNSRE.2017.2679077
PREFACE
This peer reviewed publication describes the design, fabrication, and testing of the neuromuscular
stimulation system in application of stress urinary incontinence treatment in vitro and in vivo. The design
of the system is based on the previous electrical, mechanical, and clinical requirements. The complete
validation of the system on the bench test and animal study ensure that it is ready for clinical study.
PERSONAL ROLE
I was the primary researcher on all the work described in this chapter.
ABSTRACT
We have developed a percutaneously implantable and wireless microstimulator (NuStim®) to exercise the
pelvic floor muscles for treatment of stress urinary incontinence. It produces a wide range of charge-
regulated electrical stimulation pulses and trains of pulses using a simple electronic circuit that receives
power and timing information from an externally generated RF magnetic field. The complete system was
validated in vitro and in vivo in preclinical studies demonstrating that the NuStim can be successfully
58
implanted into an effective, low threshold location and the implant can be operated chronically to produce
effective and well-tolerated contractions of skeletal muscle.
Index Terms—Neuromuscular Stimulation, Implant, Wireless
INTRODUCTION
Stress urinary incontinence (SUI) is a common type of incontinence with symptoms of involuntary leakage
of urine during activities that increase intra-abdominal pressure (e.g., coughing, sneezing or laughing). The
majority of patients with mild to moderate SUI have weakness of the external urethral sphincter (EUS) and
pelvic floor muscle, usually as a result of damage during childbirth or surgery and often exacerbated by
hypotrophic changes associated with aging and declining hormone levels [40]. Chronic urinary
incontinence is a common condition associated with severe medical and social consequences [99] .
Like any striated skeletal muscle, the EUS with intact neuromuscular innervation responds to exercise by
increasing its bulk and strength. This is the basis of pelvic floor muscle training (PFMT, also known as
Kegel exercise), which is highly effective when done properly and conscientiously [11]. Significant
improvement of muscle function requires proper instruction and regular and persistent exercise for at least
several months. If PMFT is initially successful, the benefits appear to persist [2]. However, PFMT is
difficult to perform correctly, and few patients are trained properly or sufficiently disciplined to perform
the exercise often and long enough to obtain substantial benefit. Thus, patients with SUI are often subjected
to surgical procedures or pharmacological treatments rather than PMFT.
Pharmacological therapy achieves varying degrees of success, but with substantial side effects [31]. Surgery
is recommended for many patients who fail to respond to an initial trial of PMFT. While the surgical
treatment causes minimal to moderate pain and discomfort, it is an invasive surgical procedure with general
risk of complications from anesthesia and wound infections. There are also specific complications such as
inadvertent mechanical damage to the bladder or urethra (perforation), or difficulty in urinating after
59
surgery [100]. The most commonly used surgical procedure is the tension-free vaginal tape (TVT)
suspension. TVT has an initially high success rate (75-95%), but incontinence may recur over time and
requires repeat surgery with significantly more adverse events and less success than the initial procedure
[31]. Recently, some tape products have been voluntarily recalled after increased reports of severe long-
term complications, including erosion and perforation of tissues such as the vaginal wall [46].
Rather than relying on voluntary PFMT, it may be possible to achieve the same trophic effects on muscle
fibers by activating them by neuromuscular electrical stimulation (NMES) [8, 93]. Because the muscle
nerves lie under highly innervated skin and mucosa, commercial stimulators that use transcutaneous,
intravaginal, or intrarectal electrodes are often unacceptable to patients due to somatosensory nerve
excitation with unpleasant sensations [8, 94]. Transcutaneous magnetic stimulation (similar to transcranial
magnetic stimulation, TMS) can induce eddy currents in the pelvis that are capable of activating muscle
nerves, but this requires high-power, expensive instrumentation in a clinic [10]. Intramuscular electrodes
can excite the terminal branches of motor axons with little or no sensation other than from the muscle
contraction itself. Conventional implantable stimulators first used by Caldwell achieved efficacy for SUI
without adherence problems, but the bulky case and long leads required invasive surgery with a risk of
infection and high possibility of lead dislodgement [95].
Compared with previous NMES strategies, a fully implantable leadless stimulator could be a solution. We
have developed a percutaneously implantable and wireless microstimulator (NuStim®) that can be
implanted into the pelvic floor muscles to generate strong contractions without producing unpleasant
sensations or requiring voluntary effort. Although a similar fully implantable microstimulator (the BION®)
was described 15 years ago [96, 97], it has not been commercially available, partly because it utilized fairly
expensive technologies that were not cost-effective for applications like SUI that are problematic but not
life-threatening. Compared to conventional surgical treatments for stress incontinence, we hypothesize that
the NuStim treatment will be less invasive, less expensive, and have fewer post-operative and long-term
complications, while achieving significant reduction of urinary leakage in patients with moderate stress
60
incontinence from innervated but hypotrophic muscle. This paper describes the functional requirements,
technological strategies, and in vitro and in vivo testing of this device.
DESIGN
SYSTEM OPERATION AND REQUIREMENTS
The NuStim has been designed as a low cost, minimally invasive, wireless system that patients can use at
home to deliver precisely prescribed and reproducibly delivered NMES to a single site in the pelvic floor
over a period of up to one year of PFMT for the treatment of SUI. The NuStim system contains three major
subsystems (Fig. 5-1): an implanted microstimulator for chronic electrical stimulation, an external
transmitter in a seat cushion, and a remote control in an Android smart-phone/tablet app.
The design requirements as listed below were derived from the therapeutic strategy:
• The microstimulator is small enough to be implanted accurately into the desired location via
minimally invasive procedures.
• Control of various stimulus parameters is sufficient to achieve therapeutic effects for a range of
electrode locations.
• Power and commands are transmitted wirelessly to the implant from outside the body.
• Packaging method can achieve required, limited longevity without expensive or bulky technologies.
• Using low-cost off-the-shelf components reduces both non-recurring engineering development and
manufacturing costs.
• Graphical user interface enables easy clinical programming in professional setting and patient self-
treatment at home.
61
Figure 5-1. Concept of NuStim operation. The patient receives passive exercise while sitting on the
RF-Cushion and performing daily activities. Exercise is adjusted on a tablet and transmitted
wirelessly to the RF-Cushion, which wirelessly powers and controls the implanted microstimulator.
62
IMPLANT
Mechanical: The NuStim implant (3.4 mm diameter x 10 mm long) is designed for chronic implantation
with projected functional life at least 1 year (Fig. 5-2a). It avoids expensive integrated circuits and hermetic
packaging. The exterior consists of a borosilicate glass tube with a platinum/iridium (80%/20%) electrode
on each end. The electronic circuitry consists of a two-sided ceramic printed circuit board (PCB; Fig. 5-2b),
on which discrete electronic components are surface-mounted by reflow soldering (Fig. 5-2c). The
externally generated RF magnetic field is received by a coil wound on a machined ferrite core (26.5 turns
of 3-mil insulated copper; Fig. 5-2d) that serves as a substrate for the ceramic PCB (Fig. 5-2e). Wirebonds
from the top of the large central component (programmable unijunction transistor – PUT; CP622-2N6027-
CT, Central Semiconductor Corp, NY, USA) and connections to the electrodes are added to the ceramic
PCB (Fig. 5-2f). The electronic subassembly slides into a glass tube (Fig. 5-2g), which is then attached to
plastic tubes that apply vacuum to one side and inject centrifuged and de-gassed epoxy (Epotek #302-3M,
Epoxy Technology Inc.) under pressure on the other side (Figs. 5-2h-i). After curing under pressure at 40C
for 12h, the tubing and excess epoxy is removed from the ends, leaving the finished implant (Fig, 5-2j).
With proper cleaning procedures before encapsulation, a strong adhesion between the epoxy and the
surfaces of the electronic components prevents water condensation and corrosion[101].
63
Figure 5-2. a. Design of NuStim implant. b-j. Construction steps. b. Lead free solder paste is
applied on the PCB. c. The surface-mount components are placed and soldered on hot plate (240°C
for 30s). d. Ferrite is wound with insulated copper wire and cleaned in ultrasound. e. PCB is
mounted on the ferrite with epoxy. f. coil terminals and electrodes are soldered. Wirebonds are
placed to connect the PUT. g. Device is inserted and tacked inside the glass capillary. h. Device is
loaded inside silicone tubes and filled with epoxy. i. Epoxy is cured. j. Extra epoxy is trimmed off
and the device is ready for functional testing.
64
Electronic: The circuitry relies on low-cost, off-the-shelf, surface mount components in a limited space to
achieve functions of inductive power reception and pulse generation (schematic diagram and idealized
waveforms in Fig. 5-3). We elected to control the charge per pulse rather than voltage or current because
this is the determinant of stimulus strength for short pulses in which most of the charge is delivered faster
than the 150 µs time constant of the myelinated motor axons.
The inductive coupling between the primary coil in the RF-Cushion and the secondary coil in the implant
is very weak, and thus required systematic analysis to maximize the magnetic field capture by the small-
sized implant, as discussed by Vest et al. [102]. The ferrite in the implant enhances the capture of magnetic
flux, particularly if the axis of the implant is tilted with respect to the applied magnetic field. The inductive
coil L2 and tuning capacitor C2 in parallel resonate at the carrier frequency to maximize the amplitude of
the received RF signal, which is then half-wave rectified by D1 and regulated by Zener diodes and filter
capacitor C3 to provide a constant charging voltage Vs = +16 VDC. While the RF signal is being transmitted,
capacitor Cout is slowly charged through a circuit consisting of limiting resistor R1 and the electrodes
(modeled as a series Ctissue and Rtissue in Fig. 5-3). As long as the voltage on Cout is below Vs, the PUT
remains in a high impedance state. When the RF signal is removed, Vs drops quickly to zero and the PUT
goes into a low impedance state that rapidly discharges Cout through the electrodes, creating the effective
stimulation pulse. For each stimulus output pulse, the amount of charge that the output capacitor delivers
is precisely controlled by the duration of the transmitted RF burst, which ranges from 120µs to 19.9ms. A
set of 20 burst durations generate a set of pulse strengths that form an exponential series from approximately
0.05 to 4.9 µC in which each successive step represents about 27% increase over the previous step. The
repetition rate can be controlled from 1 to 50 pulses per second (pps). The output is charge-balanced,
capacitively-coupled with a cathodal stimulus phase that has a time-constant of approximately 0.33 ms
duration when the tissue impedance is approximately 1 kOhm. For comparison, a square pulse with 10 mA
amplitude and 0.33 ms duration represents 3.3 µC, similar to the maximal output of the NuStim.
65
INSERTION TOOL
The implantation and deployment strategy utilizes a sterile NuStim insertion tool consisting of a needle
electrode inside a dilator inside a sheath plus a disposable handheld stimulator (Fig. 5-4). The NuStim
implantation can be performed as an outpatient procedure under local anesthesia in a lithotomy position. A
low threshold implantation site is first located using a disposable hypodermic EMG needle (Fig. 5-5a)
connected to the handheld stimulator via a pinjack adapter. The return electrode is connected to the back of
the handheld stimulator and attached to the skin. A skin incision is made at a different location as an entry
Figure 5-3. Schematic circuit and conceptual waveforms of the NuStim implant for two output
pulses with low and high charge, respectively. L2=11.5µH. C2=47pF. C3=3.3nF. R1=1MΩ.
R=39kΩ. Cout=0.33µF.
66
site for the NuStim insertion tool to be oriented approximately perpendicular to the perineum, and aimed
so that the end of the needle is at approximately the target determined above. The insertion tool with its
needle electrode attached to the handheld stimulator is advanced in 1 cm steps toward the target while
applying stimulation pulses that are calibrated to the same clinical units as produced by the NuStim implant
(Fig, 5-5b). The threshold for a visible twitch decreases as the needle electrode approaches the motor axons,
then increases after passing them in the next step. The stimulator with needle electrode and dilator are
removed without moving the sheath, leaving the end of the sheath at the location where the threshold
minimum was obtained. The NuStim is placed into the sheath with cathode facing the tissue. The needle +
dilator is used to push the NuStim through the sheath to its tapered end, where there is a snug fit and some
resistance (Fig. 5-5c). When the needle on the stimulator makes contact with the back of the NuStim
implant, the stimulation pulses pass through the implant to its cathodal stimulating electrode, which is used
to confirm the lowest threshold that was obtained previously. The NuStim is finally released to its site as
the sheath is retracted over the dilator (Fig. 5-5d).
Figure 5-4. Assembled NuStim insertion tool. The dilator (3.26 mm o.d. x 117.6 mm length) passes
through the sheath (4.27 mm o.d.).
67
Figure 5-5. a. A Teflon-coated needle electrode with sharp beveled tip is connected to the disposable
stimulator. The needle is used to find the low-threshold target stimulation site. b. The assembled insertion
tool is used to locate the low-threshold target identified with the needle. c. Stimulation charge is passed
through NuStim to confirm the low-threshold target site. d. The NuStim is released at the targeted
stimulation site.
68
CUSHION
Electrical: The sequence of stimulus pulses for each exercise cycle are sent to the microcontroller unit
(MCU) in the external RF-Cushion from the Android tablet via BLE communication, which then generates
a sequence of RF bursts with the required durations and intervals. The RF driver is operated at 6.78 MHz,
an ISM (Industrial-Scientific-Medical) band exempt from FCC limits on emitted field strength. This RF
carrier signal is modulated and amplified at the buffer circuit to drive Q1 (IRLR024 N-channel MOSFET,
VISHAY, CA), which is operated in a high power and highly efficient Class-E configuration to feed the
antenna (Fig. 5-6). The ideal calculation and practical tuning procedure of the class E amplifier is based on
Sokal’s method to keep voltage and current out of phase by means of a stagger-tuned LC circuit [103]. The
output impedance of the class E amplifier and the input impedance of the primary coil loaded by the
dissipative tissues of the body are matched to 50-ohms impedance for efficient power transmission. The
MOSFET achieves desired 50% duty cycle and draws 18W from +12 VDC supply. The voltage driving
the 50Ω impedance is 40V which represents about 89% power efficiency. The electromagnetic field
strength of the RF-Cushion was calculated theoretically and measured by a calibrated detection coil (2 turns
on 17.5 mm radius, 18AWG insulated copper wire) [102] with the body load simulated as a saline solution
inside a toroidal inner tube (Figs. 5-7a and b).
The physical size of the primary coil is confined by the 24 cm diameter of the seat cushion. Two turns of
wide copper trace create 15 A/m field strength up to 10 cm distance from the plane of the coil, the maximal
anticipated depth of the NuStim in patients. The minimal field strength required to reach Zener-regulated
voltage in the implant is approximately 10 A/m. The primary coil is optimized to have 50Ω input impedance
when the subject is actually present, which allows maximal output efficiency. The mismatch that occurs
when the patient is not present is detected by the load detection circuit, which then turns off the exercise
session. Current and temperature detection are used to prevent circuit damage from excessive loading such
as might occur if the cushion is placed on a conductive metal surface.
69
Figure 5-6. RF-Cushion system architecture.
70
Figure 5-7. a. NuStim system test configuration with saline tube to simulate dissipative loading by
conductive tissues of the body. b. The electromagnetic field strength measured from the center to
the edge of the primary coil (radius) as the distance is increased. The blue plane indicates the
minimum field strength needed for regulated stimulation. The intersection with the measured field
strength indicates the maximum operating range of the device.
71
Mechanical: The primary coil and external electronics are populated on a 4-layer PCB board that is
embedded in the RF-Cushion. The electronic circuit has an aluminum cover on top and bottom for heat
dissipation and electromagnetic shielding. The RF-Cushion has an outer shell of waterproof polyurethane
foam with no exposed electrical components or controls, and thus should avoid an electrical shock hazard
even if the patient leaks urine. The electronic components face the bottom of the RF-Cushion to decrease
the foam thickness between the subject and primary coil. The exposed end of the aluminum cover has a
LED power indicator and a magnetic power port for connection to a medical-grade 12 VDC power supply.
SOFTWARE APP
The software application has two functions: to allow a physician to identify the appropriate range of
stimulus strength and exercise program for a given patient, and to allow the patient to adjust stimulation
within that range to obtain and track the prescribed exercise.
In physician mode, the system provides a range of 20 stimulus intensities that can be adjusted from
threshold to target level while generating single twitches at 2 pps stimulation frequency. The threshold level
is based on identification of first twitch sensation at the lowest stimulus; the target level is identified as
maximal strength twitch or maximal comfortable level, whichever comes first. After these intensities are
locked, stimulation cycle parameters are then selected to provide strong, cyclical contractions and
relaxations for the desired exercise period (typically 30-60 minutes/day). During each exercise cycle at the
selected repetition rate, the pulse intensity ramps up from threshold to the selected intensity level, holds at
that level, and then ramps down, followed by a pause between cycles (Fig. 5-8a). After the clinician tests
these exercise cycles (screenshot in Fig. 5-8b), the subject can take home their RF-Cushion paired with a
tablet computer on which the prescribed exercise parameters have been stored.
In patient mode, the subject is instructed to self-administer the prescribed exercise on a daily basis. The
subject can only adjust the stimulus intensities over the range from the determined threshold to target in a
scale that goes from 1-10 in linear steps of RF burst duration (Fig. 5-8c). The prescription allows the patient
72
to go beyond that range up to 150% by linear extrapolation (limited to the maximal RF burst duration of
20ms). The app is designed to encourage the patient to use the highest comfortable stimulus strength from
the range prescribed by the physician, which can result in accomplishing the prescribed daily exercise in
the shortest period of time according to an algorithm for tracking adherence to treatment in the tablet. The
app keeps track of whether the patient is ahead or behind the prescribed daily regimen. All adherence
information can be read out by during follow-up visits.
73
Figure 5-8. a. Prescription mode. Various stimulation parameters are determined by the physician
on tablet. b. Prescription testing mode. Exercise is first tested in the prescription testing mode
before being sent to the patient. c. Patient mode. Patient performs the prescribed exercise at home.
74
METHODS
VERIFICATION IN VITRO
Tests were performed in vitro to verify the functionality of the NuStim system. A saline-filled inner tube
was placed on the RF-Cushion to simulate human tissue load (Fig. 5-7a). The microstimulator was placed
in a fixture with spring-loaded probes to connect the electrodes to a 1 kOhm resistive load. The stimulus
intensity was adjusted on the software APP from level 1 to level 20 while the implant was positioned at
different heights and angles from the RF-Cushion. Various exercise patterns (different stimulation
parameters: ramp, hold, off, pps and exercise time) were adjusted on the software APP to validate the
complete system. Stimulus output from the microstimulator was measured on an oscilloscope screen to
verify that the stimulus generation meets requirements for sufficient muscle activation.
PRECLINICAL VALIDATION IN VIVO
Before the NuStim system can be used clinically, chronic animal experiments are needed to provide
evidence of safety and efficacy. The objectives of the animal study were i) to identify the feasibility of the
minimally invasive implantation procedure, ii) evaluate the complete system for threshold level
identification and muscle activation on a daily basis, and iii) to evaluate the short and long-term device
stability and tissue response via threshold measurement and histology. The animal experiments conformed
to the Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal
Care and Use Committees at the Capital Medical University, Beijing, China. Experiments were carried out
on three adult beagle dogs (9.4 to 11.2 kg) sedated with 2 to 3 ml Xylazine and anesthetized with sodium
pentobarbital (2.5%, 1 ml/kg intramuscular). NuStim implants and insertion tools were sterilized in ethylene
oxide and implanted using aseptic technique. A total of five active devices were implanted into quadriceps
femoris or triceps brachii and exercised daily for two weeks. Four non-activated devices were placed in the
comparable contralateral muscles. Insertions were directed perpendicularly to the skin in order to orient
the devices transversely to the muscle fibers, but there was no attempt to correct for the substantial
pennation angle of these muscles. The insertion tool was used to locate the depth at which the threshold to
evoke a muscle twitch was minimal; once identified, the NuStim was deposited at this site. Non-active
devices were implanted without regard to optimal stimulation site. The animals recovered for at least 7 days
75
after implantation before initial activation. The threshold was measured for each active device at least once
per week. Daily exercise was performed in the first two weeks for each active implant at an intensity that
produced an apparently maximal twitch, using pulse trains intended for clinical use (0.5s ramp, 4s holding,
5s off, 6 pps and 30 min/day exercise period). The prescribed exercise pattern was designed empirically to
simulate the voluntary PFMT with a cyclical, voluntary squeeze and relaxation of the muscle for a few
seconds with repetition up to 1 hour a day [11]. The stimulus pulse rate was limited to 6 pps to avoid
hyperextension from tetanic contraction.
At the end of study, Dog 1 had been implanted for 98 days, Dog 2 for 27 days and Dog 3 for 72 days. All
three were euthanized with intravenous potassium chloride and the implant sites were examined for gross
and histological pathology. Tissue was fixed in 10% formalin for 7 to 10 days before removing the device
and blocking for paraffin embedding and sectioning. Sections were obtained from tissue near the middle
of the cylindrical device (3.4 mm diameter x 10.0 mm long) and were oriented perpendicularly to its long
axis. The tissue was stained with hematoxylin and eosin (H&E) and examined under a light microscope.
RESULTS
SYSTEM BENCH TESTING
Trains of stimulus pulses prescribed by the software were generated as illustrated in Fig. 5-9a. The peak
cathodal voltage of each stimulus pulse (yellow trace in Fig. 5-9b) was compared to the theoretical value
expected according to the charge accumulated on Cout by the regulated Vs flowing through limiting resistor
R for the period of time during which the RF carrier was on (blue trace in Fig. 5-9b). Fig. 5-10a plots the
output voltages (log scale) obtained for each of the 20 clinical steps when the device was placed at various
distances above the center of the RF cushion. For distances up to 10.5 cm and tilt angles to 45° from vertical
as shown in Fig. 5-10b, the values agree closely with the theoretical value. This indicates
sufficient power was received to activate the Zener diodes that clamp Vs at +16 VDC. The dotted red trace
indicates how the output would have decreased according to the cosine of the tilt angle without the presence
76
of the ferrite core. The stimulus output charge (orange trace in Fig. 5-10c) was compared to the theoretical
value (blue trace in Fig. 5-10c) of the 20 clinical steps. Each marker represents a clinical step in the app.
The measured stimulus charge follows the same exponential curve as the ideal calculation, but an increasing
difference from the ideal calculation was observed for the six highest stimulus charge values.
Figure 5-9. a. Train of stimulation pulses delivered from the NuStim implant. (parameter settings:
Threshold = 1, Target = 20, ramp = 1s, hold = 2s, off = 1s, frequency = 10 pps). b. NuStim output
as measured on oscilloscope for stimulus level = 13 (yellow trace). Electromagnetic field generated
from RF-Cushion in cycle of charging period (blue trace).
77
78
Figure 5-10. a. NuStim output measured at different distances from the plane of primary coil as a
function of stimulus strength steps in the app. b. NuStim output measured at different tilted angle,
height and intensity level. Dotted red trace is calculated output decreased as the tilted angle without
presence of ferrite at level 20. c. Stimulus charge comparison between measured values and ideal
calculation at the RF burst durations programmed to produce the 20 stimulus step values.
79
SYSTEM VALIDATION IN VIVO
When properly positioned over the RF-Cushion, each active implanted device was activated separately to
cause muscle contractions that could be palpated on the skin and observed as cyclical limb motion. When
the repetition rate was above 20 pps, the muscle contraction was smooth and sufficient to fully extend the
limb. The animals generally ignored the stimulation in the course of exercise without sedation, but required
custody and petting to stay on the RF-cushion during the exercise. Sedation was administered for most of
the daily 30-minute exercise periods for convenience. Thresholds for implanted devices were all in the
range of 6 to 9 clinical units when initially activated 10 days after implantation. Fig. 5-10a shows trends
in the thresholds over time normalized to the value on the activation day and compared to the thresholds
obtained during implantation on day zero. The target stimulus strength needed to generate apparent
maximal twitch was different for each device. The active device in Dog 1 vastus muscle only required 3 to
4 clinical units above threshold while the device in Dog 2 produced more gradual increases in recruitment
over 11 steps. One of the 5 active devices ceased to produce stimulation pulses 5 days after implantation.
Electrical function testing of this device removed at necropsy showed that it no longer resonated at the
tuned frequency. The most likely cause would be a cold solder joint where the copper wire is attached to
the ceramic PCB. This failure mode was subsequently mitigated with a manufacturing change to pretin the
copper wire before soldering to the PCB.
Each implanted muscle was removed at necropsy with the device left in the tissue block. All active and
passive devices were well-integrated with surrounding tissues, which made them somewhat difficult to find.
There were no gross pathological signs of reaction or infection. Tissues from which active and non-active
devices were removed after fixation are shown in Figs. 5-11b to g. The capsule layer was peeled away from
the surrounding tissue in some places, presumably because of forces on the tissues during device extraction,
but remained in position in most samples. The thickness of the fibrotic capsule around the three-month
implants (Fig 5-11. e) was about half as thick as that after one month (Fig 5-11. d). Close to the capsule,
there were some small muscle fibers with central nuclei suggestive of ongoing recovery from damage
during the initial implantation. Further from the capsule, the myonuclei were spaced around the periphery
80
of the muscle fibers in a normal pattern for healthy muscle fibers. Around the long-term implants, the
unaffected muscle fibers were closer to the capsule, which suggests progressive healing of local insertion
trauma. Muscle fibers immediately adjacent to the capsule tended to be cut transversely (i.e., running
parallel to the long axis of the cylindrical implant) but further away the plane of section appeared to be
more oblique, consistent with the intent to implant the device transversely to the muscle fibers. The
mechanical presence of the device may have resulted in some local reorientation of muscle fibers,
particularly those recovering from implantation damage. Some of the non-active devices were found wholly
or partially in loose connective tissue rather than within a muscle (Fig 5-11. f). Their surrounding capsules
tended to collapse from lack of support by fixed muscle when the implants were removed prior to
embedding and sectioning. No clinically significant histological differences were observed between the
active and passive devices at the same time points.
81
82
Figure 5-11. a. Threshold measurements as a function of post-implantation day (normalized to
implant activation day). b. Muscle block with a capsule after active device removal. c-e.
Photomicrographs of a tissue section stained with H&E from the block in d showing typical
features of cellular response and capsule formation 1 month after implantation. e. Typical
histological appearance around active device after 3 months. f and g. tissue section from control
side with typical appearance around non-active device after 3 months.
83
DISCUSSION
The NuStim system described here successfully meets the design requirements. We validated that the
Android software app allowed researchers to adjust stimulation parameters to accommodate substantial
differences in orientation of the implant with respect to the local motor axons that are the target of the
neuromuscular stimulation. The RF-Cushion responded to these commands and generated a sufficiently
strong magnetic field to allow the implant to generate the requested stimulus charges over the desired range
of distances and orientations.
Some of the voltage and charge measurements in vitro were lower than calculations predicted. The output
voltage for the lowest charge pulses decreased slightly as the distance increased between the secondary and
primary coil. Generally, the RF-Cushion requires about 30 cycles for the class E amplifier to ring up to full
field strength. When the implant is further from the transmitter, it reaches the regulated voltage later in this
ramp. For the lowest charge pulses, this delay becomes a noticeable portion of the relatively brief RF burst.
Such variability might be a concern for other applications requiring accurate control of partially recruited
muscles, but it is not critical for the proposed SUI application, which aims for strong stimulation to achieve
complete muscle recruitment. The difference between the measured and calculated charge for the highest
clinical steps is mostly due to the nonlinearity of the multilayer ceramic output capacitor Cout
(C1005X5R1E334K050BB, TDK Corporation). This is the only commercially available capacitor with a
combination of acceptable capacitance, voltage rating and small package size for the application. The
capacitance of C2 decreases nonlinearly as the bias voltage across the capacitor increases, making the
capacitor is less able to store charge at the higher voltages associated with the highest clinical levels. The
physiological efficacy of these pulses is actually less affected than the reduction in delivered charge because
the missing charge would have been delivered in the exponential tail of the stimulus pulse, which is well
after the 150µs time constant of the myelinated motor axons.
The animal study confirmed that the insertion tool could be used to implant the device in a low-threshold
location where strong skeletal muscle contraction could be achieved (< clinical step level 13 = ~0.8 µC)
well before reaching the maximal measured output of the NuStim implant (about 3 µC). A significant
84
change of threshold after implantation may indicate migration through or damage to surrounding tissue.
Proper healing after implantation is necessary for implant stabilization. The stability of the threshold values
over an extended period of electrically induced muscle contractions suggests that the implants did not
migrate or damage the muscle. The connective tissue capsule that starts to form around cylindrical implants
in the first few days after implantation [104] gradually becomes less reactive and better integrated into the
endomysial connective tissue that surrounds and supports all muscle fibers [105]. The absolute value of the
threshold was expected to differ between implantations because it is quite sensitive to the distance between
the cathodal electrode and the nearest motor axons. Accordingly, the rate at which twitch force increases
with stimulus strength depends on the distribution of intramuscular nerve branches with respect to the
implant, which is likely to vary considerably depending on the neuromuscular architecture of the muscle
and the location of the stimulator [63, 98, 106]. Healing and local reorganization of tissues after
implantation (as well as uncertainties in palpating twitch threshold in an awake animal) are the likely cause
of the small shifts in electrical thresholds noted after implantation. Similar encapsulation and recruitment
patterns were described previously for BION stimulators [81]. The physician can palpate contractions of
the pelvic floor muscles, but we expect that it will be simpler and perhaps more accurate to rely on the
patient’s perception of the electrically induced muscles contractions.
NuStim implants with non-hermetic epoxy packaging were found in most cases to maintain functionality
for the limited duration of these experiments (about 3 months). The one failure appeared to be an
idiosyncratic flaw during manufacturing rather than an encapsulation failure. Similar devices have been
subjected to an accelerated life-test that involves soaking in saline at 50°C while operating at continuous
maximal output (article in preparation). This methodology was used to refine the cleaning and
encapsulation processes described in this article. The NuStim is made from inert, biocompatible materials
that may be left in the body permanently, similar to medical devices such as nonresorbable sutures, vascular
clips and orthopedic bone screws. If there is a medical indication for NuStim removal such as infection or
pain, this will require a minor surgical procedure under local anesthesia, probably with ultrasound guidance
to locate the implant easily.
85
There are no good animal models for SUI in humans. One animal model of urinary incontinence used
surgical dissection of the muscle surrounding the urethra, which is likely to damage the muscle nerve in a
manner that is unlike human SUI [107]. It is impractical to incorporate the extended periods of post partem
reorganization and post-menopausal hormone withdrawal that characterize most clinical cases of SUI.
Furthermore, terrestrial quadrupeds do not require pelvic floor muscles as strong as a human, which must
support pelvic viscera during upright posture. The canine quadriceps and triceps have sufficient thickness
for transverse implantation, similar to the implantation orientation expected in the human pelvic floor, and
they facilitate visual monitoring of muscle recruitment. The chronic animal experiment presented here
confirms that the NuStim can produce strong, well-controlled, repetitive contractions in skeletal muscle. It
remains to be demonstrated that exercise patterns similar to PFMT can be obtained in the skeletal muscle
of the human pelvic floor, which has similar neuromuscular physiology and size but somewhat different
innervation and fiber architecture.
The RF magnetic field strength required to achieve inductive power transmission over the required
distance must be considered in terms of the specific absorption rate (SAR) allowed by the
guidelines/standards [108]. Tests of a similar wireless transmission system for a fetal micropacemaker
indicated that about 50% of the applied magnetic field was absorbed by eddy currents induced in the
conductive tissues of the body [102]. This corresponds to about 8 Watts of energy from the NuStim RF-
Cushion (peak value when the RF burst is on), which will be dissipated as heat in the pelvic region overlying
the transmission coil. If we model that region as a cylinder with 0.1 m radius and 0.1 m height and specific
gravity of 1.0, we get a 3.1 kg mass. The foam insulation layer of the RF-Cushion ensures that no tissue
will be in direct contact with transmission coil. The app software was programmed to limit the RF duty
cycle to avoid overheating the circuitry in the RF-Cushion. For example, when the stimulus intensity is
maximal at level 20, the clinician cannot prescribe a stimulation frequency greater than 20 pps, which
produces a strong, smooth contraction in skeletal muscle. After allowing for the minimal off-time between
stimulus trains, the maximal RF duty cycle is 35%. This corresponds to 0.9 W/kg, well below the 1.6 W/kg
safety limit.
86
Eddy currents induced by very strong magnetic fields have been used to excite pelvic nerves and muscles
for exercise therapy of SUI. It avoids the need for vaginal or anal electrodes but it is poorly selective,
leading to unwanted sensory and motor effects, even when the patient is positioned optimally and the
magnetic field strength is adjusted carefully by a clinician [109]. The NuStim must be implanted in a simple
out-patient procedure, but it then appears likely to enable selective and consistent recruitment of just the
target muscle when used by the patient at home. It remains to be determined whether some patients will
require implants in both sides of the pelvic floor.
The NuStim provides a wide range of stimulation parameters with which to treat SUI. It remains to be
determined which stimulation patterns will be both comfortable for the patients and effective and efficient
to exercise the EUS muscle to treat SUI. For typical SUI patients with severe EUS atrophy, muscle fibers
are easy to fatigue. Slow-twitch muscle fibers are non-fatiguing, and thus are capable of continuous or
frequent contractions for urethral closure at rest. Fast-twitch muscle fibers are more easily fatigued but can
respond more rapidly and forcefully to sudden recruitment (e.g., during coughing). Low-frequency and
high-intensity electrical stimulation tends to reverse disuse muscle atrophy, building bulk and peak force
generation [110]. Longer cycles of stimulation with intermediate frequencies and longer exercise periods
tend to improve fatigue resistance [111]. High frequency bursts of stimulation produce maximal contractile
force. It is unclear how patients may tolerate the sensations produced by these different patterns of muscle
contraction. The pulse train used in the animal study is the default exercise pattern for clinical use. The
planned pilot clinical study is intended to identify stimulus and exercise parameters that achieve a useful
balance of acceptability for the subjects and clinical improvement of SUI. The usability of the clinical
system and its software APP by physicians and patients will be tested in the clinical trial.
ACKNOWLEDGMENTS
The authors would like to thank Xing Li, Tianji Lu, Zhaoxia Wang and Han Deng for assistance in animal
care, Dr. Frances J. Richmond for assistance with histological analysis, consultant Thomas Yeh for software
development and engineers Ray Peck, Gary Lin, Sisi Shi, and Longpeng Jiao for contributions to design
87
and manufacture. This research was funded by General Stim Inc.
88
CHAPTER 6: ACCELERATED LIFE-TEST METHODS AND
RESULTS FOR IMPLANTABLE ELECTRONIC DEVICES
WITH ADHESIVE ENCAPSULATION
© Biomedical Microdevices. Reprint from:
X. Huang; P. M. Denprasert; L. Zhou; A. N. Vest; S. Kohan; G. E. Loeb, "Accelerated life-test methods
and results for implantable electronic devices with adhesive encapsulation," in Biomed Microdevices, 2017,
19: 46. doi:10.1007/s10544-017-0189-9
PREFACE
This peer reviewed publication describes the major technical challenge of package reliability for the
proposed implantable neurostimulator. The non-hermetic packaging relies on epoxy adhesion to prevent
failure from water condensation on the electric circuit. In order to understand the failure modes and weak
points of the design, a novel accelerated life testing method using a highly sensitive circuit was designed
and validated, then the critical manufacture process was optimized after several iterations. Strong adhesion
from epoxy was demonstrated to be sufficient for the implanted neurostimulator to function in the human
body for more than 1 year.
PERSONAL ROLE
I was the primary researcher on all the work described in this chapter.
ABSTRACT
We have developed and applied new methods to estimate the functional life of miniature, implantable,
wireless electronic devices that rely on non-hermetic, adhesive encapsulants such as epoxy. A comb pattern
89
board with a high density of interdigitated electrodes (IDE) could be used to detect incipient failure from
water vapor condensation. Inductive coupling of an RF magnetic field was used to provide DC bias and to
detect deterioration of an encapsulated comb pattern. Diodes in the implant converted part of the received
energy into DC bias on the comb pattern. The capacitance of the comb pattern forms a resonant circuit with
the inductor by which the implant receives power. Any moisture affects both the resonant frequency and
the Q-factor of the resonance of the circuitry, which was detected wirelessly by its effects on the coupling
between two orthogonal RF coils placed around the device. Various defects were introduced into the comb
pattern devices to demonstrate sensitivity to failures and to correlate these signals with visual inspection of
failures. Optimized encapsulation procedures were validated in accelerated life tests of both comb patterns
and a functional neuromuscular stimulator under development. Strong adhesive bonding between epoxy
and electronic circuitry proved to be necessary and sufficient to predict 1-year packaging reliability of 99.97%
for the neuromuscular stimulator.
Index Terms—Accelerated life test, Non-hermetic packaging, Encapsulation, Wireless, Implantable
INTRODUCTION
A long functional lifetime is critical for clinical applications of most implantable medical devices. This is
usually evaluated by accelerated life-testing in which failure modes that might occur after years of normal
use are revealed more quickly by increasing the temperature and duty cycle. One common source of failures
is the device packaging that is designed to protect the electronic components from body fluids and protects
the body from the electronic components [112]. Hermetic packages employing exotic metal and ceramic
technologies have few long-term failure modes but tend to result in relatively expensive, bulky and rigid
devices that must be implanted surgically. For applications requiring treatment for months to years rather
than decades, such packaging is unnecessary. Polymeric encapsulation was used with considerable success
in the early days of pacemakers and spinal cord stimulators. Non-hermetic encapsulated devices can be
90
inexpensive and reliable, permitting small and elongated shapes that can be implanted via minimally
invasive procedures [112, 113]. However, this type of encapsulation may still fail within months to years
due to diffusion and condensation of water vapor and corrosion of electronic components [114, 115]. Its
integrity depends on conformal layer adhesion to component surfaces rather than a diffusion barrier [116].
Such adhesion can fail over time as a result of various stressors, so it is important to test such devices under
actual use conditions, but it may be impractical to do so for the design life of the device. The alternative is
to introduce higher than usual aging factors (such as temperature, voltage, pressure, usage level, etc., or
combinations thereof) to cause test units to fail more rapidly than under usual conditions [117, 118]. It is
also important, however, to avoid introducing failure modes that may be specific to the test methods, such
as adding connections to energizing and monitoring instrumentation.
The most commonly used acceleration factor is elevated temperature, but many components including the
polymeric encapsulant may have limited operating temperature ranges, such as its glass transition
temperature. An alternative is to detect incipient failure at an early stage by using a highly sensitive test
device. Rather than waiting for water to condense, corrode and cause circuit failure, a more sensitive “comb
pattern” consisting of many closely interdigitated elements is used detect initial water molecule diffusion
and condensation. Traditionally, the encapsulated comb pattern device is made with two wire connections
to an external power source for DC bias voltage and to an impedance spectrometer to monitor package
performance [119]. A DC bias applied across the comb pattern results in a uniformly high voltage stress
(V/cm) over a large test area. Moisture between the electrodes of the comb pattern greatly increases the
capacitance and reduces the resistance between them. The combination of elevated temperature, elevated
voltage stress, and high sensitivity to incipient failure provides an accelerated validation of encapsulants.
Connecting external leads to the comb pattern may be difficult or impossible without changing the structure
and fabrication procedure of the clinical device, which can lead to inaccurate extrapolation of lifetime and
reliability [120]. Furthermore, lead connection may not even be possible for the next generation of devices
that use bi-directional wireless transmission of power and data [121, 122]. Some implantable devices such
as RF identification (RFID) “biochips”, video cameras, photonic stimulators and drug delivery pumps may
91
not normally have any externalized electrical connections [123, 124]. We utilized non-contact methods to
detect changes in the resonance of tuned inductor-capacitor circuits, which have long been used in radio
engineering, intracranial pressure monitors [125] and other applications [126].
In this study, we have developed and tested wireless accelerated life-testing methods to evaluate the epoxy
packaging performance of a wireless neuromuscular microstimulator called NuStim™, which is designed
to exercise the pelvic floor muscle in stress urinary incontinence patients for up to one year [127]. The
ceramic circuit board in the microstimulator was replaced by the comb pattern board to increase sensitivity
to incipient failure mechanisms. The devices were stressed by increased temperature, voltage and duty cycle
in saline. We developed a non-contact method to measure the resonant frequency and quality factor of the
capacitive comb pattern connected to the same inductive coil that normally receives power. Comb pattern
devices with different known defects were tracked over time to validate sensitivity to failure mechanisms.
Different cleaning and fabrication procedures were compared during the study. Failure rates under these
accelerated conditions were extrapolated mathematically to estimate package reliability under normal use
conditions for the microstimulator.
DESIGN AND METHODS
COMB PATTERN
The comb pattern device (Fig. 6-1a) used the same construction as the microstimulator (Fig. 6-1b) described
elsewhere [127]. The package consists of a thin-wall borosilicate glass sleeve that was filled with a low
permeability epoxy (Epotek #302-3M, Epoxy Technology Inc.). The comb pattern was printed on an
aluminum oxide ceramic substrate (2.4 mm width x 3 mm length) by photolithography (deposited layers of
gold over nickel, gold, and titanium-tungsten) with the spacing between the electrodes about 25 µm, as
shown in Fig. 6-2a. The surface mount discrete components were placed on the ceramic PCB board by
hotplate soldering at 230°C (solder paste: WS488-SAC305, AIM solder, CA, USA). The external radio
frequency power was received by a machined ferrite core with wound coil (26.5 turns of 0.003” insulated
copper). The complete electronic assembly was cleaned thoroughly as described below and slipped into a
92
clean glass capillary. The epoxy was de-gassed in a vacuum chamber, drawn into a syringe, and centrifuged
to further remove bubbles. Then the epoxy was injected through a plastic tube slipped over the subassembly
while applying a vacuum to the other end. The epoxy was cured at 40°C and 230 psi to accelerate polymer
crosslinking and prevent dissolved air from forming bubbles. After overnight curing, the plastic tube and
excess epoxy was cut away and the electrode surfaces exposed cleanly by abrasive buffing.
Figure 6-1. Design of IDE device compared to clinical NuStim microstimulator. a. IDE device before epoxy
filling. b. Microstimulator after encapsulation.
The interdigitated electrodes of the comb pattern function as a capacitor C that was connected in parallel
with the winding coil L to achieve inductive power transmission and encapsulation failure detection (Fig.
6-2b). The capacitor C and the inductive coil L in parallel resonate at a frequency 𝑓 𝑆𝑅𝐹
, determined by:
𝑓 𝑆𝑅𝐹
=
1
2𝜋 √ 𝐿𝐶
Eq. 1
The value of capacitance C depends on the dielectric properties of the material between the electrodes. Any
condensed moisture present will substantially increase C because the dielectric constant of water (ɛ=80.4)
is much larger than epoxy (ɛ=3.6). This will change the self-resonant frequency of the circuit. Condensed
b
a
93
moisture will also greatly reduce the resistance between the electrodes of the comb pattern by providing a
conductive liquid path between them. This can be detected by a decrease in the quality factor (Q factor) of
the circuit based on the equation for a RLC parallel circuit,
Q = R√
𝐶 𝐿 =
𝑓 𝑆𝑅𝐹
Δ𝑓 Eq. 2
The Q factor can be computed by dividing the self-resonant frequency 𝑓 𝑆𝑅𝐹
by the bandwidth Δ𝑓 , where
Δ𝑓 is identified by a 3db decrease in the amplitude of a received RF signal.
The comb pattern device has the functionality to introduce DC bias voltage as an accelerating stressor. The
comb pattern board is separated into two capacitors C1 and C2 in parallel as shown in the schematic (Fig.
6-2b). The introduced Zener diodes (GDZT2R8.2, Rohm Semiconductor) do not affect the resonant
frequency of the circuit, which is mainly determined by the parallel C1 and C2. One side of the comb pattern
C1 is connected to the inductive coil terminals directly to receive the AC signal. The other half of the comb
pattern C2 is connected in series with the Zener diodes to introduce a DC voltage bias between the capacitor
electrodes. A high voltage sinusoidal AC signal is generated across C1 and C2 branches equally during
resonance. The Zener diodes provide half-wave rectification that is limited to the reverse-bias breakdown
voltage of the Zener diodes (8VDC each, in-series total 16VDC). C2 integrates the difference between high
forward current through the Zeners when the positive phase of the received AC waveform exceeds their
forward bias and the low reverse current during the negative phase until they reach their negative Zener
threshold. This results in a bias voltage (measurement described in Results) that provides the desired voltage
stress across the interleaved electrodes that form C2. Epoxy with dissolved and condensed water form
virtual resistors R1 and R2 between the capacitor electrodes shown in the schematic.
94
Figure 6-2. a. Comb pattern ceramic PCB board with IDE. b. Schematic of comb pattern board. C1 and C2
represent the capacitors circled by red and black dotted lines in the PCB board. R1 and R2 represent virtual
resistance in parallel with capacitors. L1 is the winding coil on ferrite. Z1 and Z2 are Zener diodes.
A total of 12 comb pattern devices with three different common defects during fabrication process were
made to demonstrate sensitivity to failures. 1) Bubble defects were induced intentionally in the epoxy
injection step with undegassed epoxy and injection without a vacuum. 2) Finger print oil was introduced
after final cleaning by swiping a cotton swab contaminated with finger oil over the surface of the comb
pattern before encapsulation. 3) Contamination by a cyanoacrylate adhesive (normally used to mount the
PCB and control the terminations of the coil) was applied by hand.
Preliminary work had suggested that epoxy adhesion was adversely affected by using deionized water
instead of distilled water in the final rinsing step of the PCB cleaning process before epoxy encapsulation.
Two groups of 20 comb pattern devices each were constructed to evaluate the differences using the
following cleaning procedures (Table 1) before epoxy encapsulation (as described above): 1) The comb
patterns designated “Deion. Clean” were flushed by isopropyl alcohol (IPA) for 30s and brushed if there
was visible debris left on the board., then rinsed with deionized water for 30s and dried in the oven for 2
b
a
95
hours, and 2) The comb patterns designated “Dist. Clean” were placed in separate clean vials for ultrasound
cleaning with IPA (10 min.) and distilled water (10 min.), then dried in the oven for 2 hours.
Table 6-1. Comparisons of different cleaning procedure for comb pattern devices in Deion. Clean group,
Dist. Clean group and functional microstimulators.
FUNCTIONAL DEVICES
After the studies of comb pattern devices, 9 functional microstimulators were fabricated and encapsulated
according to the final cleaning procedure as shown in Table 1, which included some further improvements.
Before the coil winding procedure, any loose particulates from the machined ferrites were removed by
sonicating in detergent for 10 min. (1% detergent in distilled water, Detergent 8, Alconox, NY, USA) and
then rinsed with distilled water to flush out residual detergent. The PCB was placed on the ferrite using the
same epoxy adhesive as for encapsulation instead of cyanoacrylate. The functional device was cleaned
before the wire bonding by the same cleaning procedure as for the final subassembly. The final subassembly
was rinsed after ultrasonic cleaning by distilled water to flush out any contaminants from the ultrasonic
bath.
96
WIRELESS DETECTION
The bandwidth and resonant frequency of the LC circuit of the comb pattern devices were detected
wirelessly by the concentric, orthogonal, inductive coil pair shown in Fig. 6-3a. The coils were carefully
aligned to minimize direct inductive coupling between the transmitter and receiving coils by finding the
null orientation at all test frequencies. The function generator (Tektronix AFG3102 Dual Channel Function
generator) generated a sinusoidal sweep frequency 16 – 24 MHz at 10 Vp-p to provide power to the
transmitter coil. The device under test was inserted at 45 degrees into the center of both coils. The output
voltage of the receiver coil was recorded on the oscilloscope to determine resonant frequency and
bandwidth. When the frequency of the transmitted magnetic field equals the resonant frequency of the
device under test, it induces a maximal circulating current in the LC circuit. The induced current flow in
the coil of the device under test creates its own magnetic field, which is picked up by the receiving coil of
the test system. The self-resonant frequency (green trace in Fig. 6-3b) of a comb pattern device was
measured by the receiving coil as displayed on oscilloscope, which was ~20.7 MHz with ~1 MHz
bandwidth before encapsulation. The bias DC voltage (yellow trace in Fig. 6-3b) was also measured during
wireless detection and was found to be correlated with the measured resonant properties of the device under
test. Instead of directly probing the comb pattern capacitor, which is the summation of the both the capacitor
output and the bias, the bias DC voltage was measured directly using fine needle probes between Zener
diodes. The bias DC voltage tended to drop substantially when the self-resonant frequency and/or Q factor
of the device was affected by moisture applied on the bare comb pattern board.
97
Figure 6-3. a. Two coil wireless query system for resonant frequency and bandwidth detection. The
powering coil (2 turns of 18 AWG insulated copper wire, 37.5 mm in diameter) and the detecting coil (30
turns 22 AWG insulated copper wire, 35 mm in diameter) were placed orthogonally to minimize direct
electromagnetic coupling between them. The test article was placed at 45° angle between them so that it
coupled to both coils. b. Oscillogram of self-resonant frequency and bandwidth (green trace) as measured
in wireless detection, sweeping from 16-24MHz. Generated bias DC voltage (yellow trace) on the diodes
as measured by needle probes is correlated to device’s resonant properties.
98
WIRELESS POWERING
The comb pattern devices were powered inductively during accelerated life-testing by a custom transmitter
coil driven by a class E power amplifier. The external power source produced a 20 MHz continuous
electromagnetic field at least 10 A/m strength up to 10 cm distance from the plane of the coil. The design
of this RF powering system and methods to measure and optimize power output were similar to those
described previously for the microstimulator [102]. When the devices were around 20MHz, Q factor at 15,
and placed at various distances above the RF transmitter, the measured Zener-regulated bias DC voltage as
displayed on oscilloscope was stable at ~11.5 VDC for distances up to 10cm and tilt angles to 45° from
vertical. The bias DC voltage is not regulated at the ideal voltage (16V) due to component frequency
characteristics. During the accelerated life-testing, the resonant frequency and Q factor would drop due to
comb pattern capacitance and resistance changes, which decreased the DC bias voltage. The data points in
Fig. 6-4 were acquired by adding different values of capacitors or resistors in parallel with C2 and R2 in
the bare comb pattern board, respectively, to shift the resonant frequency or Q factor independently. Fig.
6-4a plots value of the bias DC voltage as a function of self-resonant frequency when the Q factor was
maintained at 15.5±0.4 (mean ± standard deviation). The bias DC voltage dependency on Q factor is shown
in Fig. 6-4b when the self-resonant frequency was maintained at 20.0±0.4MHz (mean ± standard deviation).
By measuring the self-resonance frequency and calculating the Q factor, the bias DC voltage can be
conservatively estimated using the fitted curves. For example, if one device had self-resonant frequency at
about 19MHz, the bias voltage would be about 7V and 4.7V when Q factor was about 15 and 10,
respectively, based on the fitted curve. It is important to note that shifts of this magnitude for resonant
frequency and Q were always associated with rapid further shifts that led to device failure as defined below.
99
Figure 6-4. a. Measured bias DC voltage as self-resonant frequency changes. B Measures bias DC voltage
as Q factor changes.
ACCELERATED LIFE-TESTING
After initial baseline measurement, all comb pattern devices were positioned vertically inside a saline tank
at 50º C for accelerated life-testing (Fig. 6-5a) and were removed at regular intervals for wireless detection
(Fig. 6-3a) of their resonant properties. A saline tank was placed over the transmitter coil, and the devices
were powered continuously via inductive coupling. Its temperature was monitored by an
electromagnetically shielded thermocouple and regulated by a circulating water bath with accuracy of ±1°C.
If the measured resonant properties of comb pattern devices met the criteria for incipient failure (resonant
frequency 17MHz), the devices were examined carefully under a microscope to identify possible failure
mechanisms. As indicated by the fitted curve in Fig. 6-4, when the resonant frequency was around 17MHz
with Q factor about 7, the voltage stress on the comb pattern dropped below 0.5 V, which would be
insufficient to drive most electrolysis reactions. From the preliminary experiments, whenever the self-
resonant frequency was lower than 17MHz or there was no detectable resonant frequency, there were visible
gross failures including ionic dissolution, galvanic corrosion and dendrite formation, bridging, and short-
circuits between comb pattern traces.
100
Data was collected for one year in accelerated life-testing. The time to failure of the device was derived
from the resonant properties measurement that met failure criteria. An exponential distribution with
constant failure rate was assumed for the failure times of two groups of comb pattern devices (Deion. Clean
and Dist. Clean groups)[128]. This was used to extrapolate the reliability of epoxy encapsulation of the
microstimulators for 1 year at 37ºC, based on differences in the magnitude and distribution of voltage stress
levels between the two types of device. One year of data collection in accelerated life-testing predicts a
much longer lifetime than the one year of intermittent use required for the intended clinical application of
the neurostimulator.
The acceleration model used in the paper had three combined accelerating factors of temperature, voltage
stress and duty cycle. Each individual accelerating factor was assumed to be independent and to contribute
to the acceleration multiplicatively [118]. The accelerated life test was performed at 50ºC, below the glass
transition temperature of the epoxy to avoid excessive thermal expansion and glass breakage. The aging
factor was determined by an empirical estimation of Arrhenius’ Law, for which every 10ºC elevation results
in twice acceleration of all chemical reactions [129]. The calculated acceleration factor was thus about 2.4
times faster in 50ºC compared to human body environment at 37ºC. The inverse power relation was used
for non-thermal acceleration due to voltage stress:
𝐴𝐹 = (
𝐸 𝑎 𝐸 𝑢 )
β
Eq. 3
where the acceleration factor AF is the ratio between the accelerated voltage stress level and the normal
voltage stress level. β =1 assumes that increased voltage stress magnitude has positive, linear acceleration
for the device failure. The effects of duty cycle were also estimated from the continuous stress in the
accelerated life tester (24h/d) divided by the maximal daily treatment prescription (2h/d).
The functional microstimulators were powered in the accelerated life-testing by a separate RF transmitter
tuned to the operating frequency of the microstimulator at 6.78MHz. Rather than placing the devices into
saline tank directly as done with the comb pattern devices, each functional microstimulator was loaded into
a test vial of saline equipped with Pt-Ir electrodes to capture its output pulses and a light emitting diode
101
(LED) to provide a visual monitor of its functionality (Fig. 6-5b). The vials were then placed into heated
water bath at 50º C above the RF transmitter. The microstimulator was powered continuously with 20ms
long bursts of RF every 33ms, generating the maximal stimulus output through the LED of ~3 µC @ 30
pps with peak voltage of 15V. If and when the LED ceased blinking, the device was removed from the test
vial and examined for failure analysis.
Figure 6-5. a. Accelerated life test system for both comb pattern devices and the microstimulator. b.
Accelerated life-testing vial for the microstimulator to convert stimulation pulses into visible light flashes.
The rubber tube acts as an O-ring to force the output stimulation current through the surrounding saline and
into the Pt/Ir wire electrodes, which do not touch the output electrodes of the microstimulator.
RESULTS
Measured self-resonant frequency of all comb pattern devices as a function of accelerated life-testing in
days is shown in Fig. 6-6. Details regarding the failure rates for each type of deliberately introduced defect
in the comb patterns are shown in Fig. 6-7. Each trace corresponds to the measurements over time for one
device, with color codes to indicate classification according to details of fabrication that were intended to
affect reliability. Generally, all devices experienced the same three phases but differed in the rate at which
they progressed between phases depending on fabrication classification. In phase 1, self-resonant frequency
102
around 20 MHz dropped within the first couple days of soaking, representing water vapor diffusion into the
epoxy encapsulant with dielectric change. In phase 2, if there were no voids or unbounded cavities for water
vapor condensation, the measurements were stable with only slight fluctuations that appeared to be related
to environmental temperature changes. In phase 3, self-resonant frequency first dropped below 17 MHz,
with wide bandwidth and low Q factor due to condensation of moisture, then progressed rapidly towards
undetectable resonant frequency, shown as a sharp decline in the plot, representing metal corrosion and
ultimate circuit failure from dendrite growth and shorting between the comb electrodes.
Figure 6-6. Self-resonant frequency measurements of all comb pattern devices as a function of number of
days in accelerated life-testing.
The measured resonant frequency and calculated Q factor of comb pattern devices with different defects
over duration of accelerated life testing in days are shown in Fig. 6-7.
Fingerprint oil-contaminated devices (red traces in Fig. 6-7) had slightly higher initial self-resonant
frequency than other devices. The epoxy over the fingerprints appeared to have pulled away from the comb
pattern electrodes during curing, leaving air-filled spaces between electrodes that had lower dielectric
103
constant than the epoxy. Compared with other devices, the finger print oil contaminated devices usually
failed most quickly due to direct water vapor condensation in the void and dendrite formation between
electrodes (Fig. 6-8a).
Cyanoacrylate-contaminated devices (green traces in Fig. 6-7) tended to initiate corrosion on the edge of
the PCB where the low-viscosity adhesive could seep onto the comb pattern before curing. This was
associated with unreliable adhesion, corrosion, dendritic short circuits and undetectable self-resonant
frequency (Fig. 6-8b). The damage could expand even after electrical function ceased (Fig. 6-8c).
Bubble-defect devices (blue traces in Fig. 6-7) usually lasted longer in accelerated life-testing than other
devices, except for one device with a huge bubble over the comb pattern that left only a thin layer of epoxy
on the electrodes after curing. Water vapor condensation could be found in the bubble after 10 days of
soaking (Fig. 6-8d). Optical microscopy revealed blistering on the gold electrode (Fig. 6-8e) with
progressive delamination (Fig. 6-8f) that was correlated with resonant frequency measurements (markers
A and B on Fig. 6-7a, respectively).
104
Figure 6-7. a. Self-resosnant frequency measurment of comb pattern devices with defects as function of
days in testing. Marker A and B represent day of observation with blistering (Fig. 8d and e) and
delamination on gold trace in one device with huge bubble (Fig. 8f). b. Calculated Q factor as function of
days in testing.
105
The percentages of surviving devices in the Deion. Clean and Dist. Clean groups are plotted over time in
Fig. 6-9. The Deion. Clean group had a total of 20 devices subjected to the soaking test, whereas the Dist.
Clean group had 13 devices after excluding 7 devices that failed from unrelated manufacturing problems.
In the Deion. Clean group, about 50% of devices failed after 30 days and 90% failed after 100 days. In the
Dist. Clean group, about 54% (7 of the 13 devices) were still working after more than 365 days in the
accelerated life-testing. The surviving comb pattern devices had stable self-resonant frequencies around 19
MHz with Q factor ~ 9, which was estimated to generate 4.2V voltage stress (see Fig. 6-5). Corrosion
usually started on the gold trace (Fig. 6-10a) and gradually increased over time (Fig. 6-10b). Eventually,
the epoxy barrier between comb pattern traces broke in at least one location, which led to functional failure
(Fig. 6-10c).
All 9 functional microstimulators were stressed continuously in the accelerated life-testing chamber. The
survival rate over time is shown in Fig. 6-9. Two devices failed after 119 and 155 days; the other 7 devices
were still functioning normally after >180 days at 50°C (equal to ~14 months at 37°C). The two failed
devices were subjected to failure analysis by applying various test signals to their output electrodes. Neither
failure was related to epoxy encapsulant failure. One device had no visible corrosion but had no self-
resonant frequency; this is most likely caused by a cold solder joint where the copper wire is attached to
the ceramic PCB. A procedure to pretin the copper wire before soldering to the PCB was added to alleviate
this problem. The other failed device had the characteristic appearance of “purple plague” around one end
of the gold bondwire to an aluminum pad on the programmable unijunction transistor in the stimulation
circuit [130]. This is a classic example of interfacial intermetallic compound (IMC) formation when such
bonds are poorly made and subjected to high current flow.
106
Figure 6-8. a. Typical corrosion on IDE with finger print oil contamination. b. Typical corrosion on IDE
with adhesive contamination. c. Increased corrosion area over time. d and e. The giant bubble filled with
water and blisters forming on the gold traces. Markers A: optical microscopy correlated with measured
resonant frequency in Fig. 7a. f. Further delamination formed along the gold trace. Marker B: optical
microscopy was correlated with measured resonant frequency in Fig. 7a.
DISCUSSION
INTERPRETATION OF TEST DATA
The measured self-resonant frequency and calculated Q factor of the comb pattern devices are both sensitive
indicators of encapsulant performance. They are related to each other, but Q factor provides additional
information. In the first several days of testing, capacitance increases as absorbed water vapor increases the
dielectric constant of the epoxy encapsulant, which decreases self-resonant frequency (Fig. 6-7a) and should
increase Q factor (Fig. 6-7b). Instead, the Q factor decreases as the result of significant decrease in
107
resistance based on Eq. 2, which is related to an increase of epoxy conductivity from water vapor dissolved
in it [119]. As more water vapor diffuses through the epoxy, the Q factor then increases, which represents
an increase of capacitance due to the expected dielectric change from water vapor with no further change
in resistance between traces. When water vapor reaches saturation, the measured Q factor becomes stable,
indicating stable capacitance and resistance. If the adhesive bonding is damaged, the resistance between
electrodes drops significantly, as indicated by a rapid decrease of Q factor. The comb pattern circuity is not
an ideal RLC parallel circuit as shown in Eq. 2. The circuit has additional resistance introduced by the
winding coil wire and impedance from the Zener diodes, which complicates the equivalent circuit. Rather
than providing quantitative measurements, the circuit allows qualitative evaluation of resistance and
capacitance changes from water vapor diffusion and condensation.
One objective of this study is to determine and optimize the long-term reliability of the epoxy encapsulation
for a wireless microstimulator. Instead of relying on long-term functional testing of complete devices, a
comb pattern is preferable for accelerated identification of failure modes in several ways:
• It has a large area of high density of closely spaced conductors that are vulnerable to water
condensation.
• High voltage stress can be applied to accelerate incidence of failure without introducing electrical
connections that might change failure susceptibility.
• Water vapor diffusion, condensation and corrosion are detected as changes of capacitance and
resistance to provide an early indication of incipient device failure.
• These changes can be computed from their effects on self-resonant frequency and bandwidth as
measured wirelessly by an external pair of inductive coils.
• The comb pattern devices can be made using the same materials and processes as the
microstimulator, which allows extrapolation and prediction of performance for functional devices.
The last point is especially important, as none of the 9 microstimulators undergoing accelerated life-testing
experienced a failure due to encapsulation even after 6 months of testing. Their reliability must be
108
extrapolated from the encapsulation failures of the comb pattern devices, which exhibit large and rapid
changes in their normally very high impedance and Q-factor.
Figure 6-9. Number of survived devices in deionized clean group (Deion. clean), distilled clean group (Dist.
Clean) and neurostimulator group as a function of accelerated testing days (normalized to number of devices
starting soaking). The x-axis in red represents expected life in months at 37°C, assuming only Arrhenius
temperature acceleration.
Figure 6-10. a, b and c. typical corrosion of 100, 220 and 350 days after soaking respectively, visible only
on anodally-polarized gold traces.
109
EXTRAPOLATION OF EXPECTED LIFETIME
The failure rate of the microstimulator encapsulation can be estimated quantitatively by considering the
probability of packaging failure in comb pattern devices and correcting for acceleration due to stressors
including temperature, voltage stress and duty cycle. Exponential curves were fit to the data of the Deion.
Clean and Dist. Clean devices, as shown in Fig. 6-9 (goodness of fit: R-square about 0.967 and 0.946
respectively). To extrapolate at least 1 year lifetime at 37°C, the device should last 152 days at 50°C, based
on 2.4 aging factor due to temperature. The 1 year calculated reliability is 5.3% and 78.8% for Deion. Clean
and Dist. Clean respectively. The reliability of 78.8% in one year can be interpreted as probability of success
of a random device in the Dist. Clean group to last at least for 1 year, which can be expressed as the
following equation,
𝑃 𝑠𝑢𝑐𝑠𝑒𝑠𝑠 𝑟𝑎𝑡𝑒 = 78.8% = 𝑒 −0.001569∗152
= 𝑅𝑒𝑙𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑖𝑛 𝑜𝑛𝑒 𝑦 𝑒 𝑎 𝑟 Eq. 4
This probability assumes that failures only occur on the half of the comb pattern device that was subjected
to voltage stress from the DC bias, as was observed. This half comb pattern has total length of parallel
electrodes about 𝑙 = 40𝑚𝑚 . If an occurrence of failure is linearly distributed on this 40mm electrode plate
with a constant probability, and if the occurrence of each failure point is independent, the probability of
success of the comb pattern can be assumed as the product of success probabilities of all forty units of
electrodes with length 𝑙 = 1 𝑚𝑚 , which can be described as
𝑃 𝑠𝑢𝑐𝑐𝑒𝑠𝑠 𝑟𝑎𝑡𝑒𝑠 = (𝑃 𝑠 (𝑢𝑛𝑖𝑡 =1𝑚𝑚 )
)
40
Eq. 5
The calculated success probability of a unit 𝑃 𝑠 (𝑢𝑛𝑖𝑡 =1 𝑚𝑚 )
is equal to 99.41%, which means only 0.59%
chance that a failure will occur on the 1 mm long parallel plate electrodes at the voltage stress 𝐸 𝑐𝑜𝑚𝑏 . The
traces subjected to high voltage stress 𝐸 𝑁𝑢𝑆𝑡𝑖𝑚 in the microstimulator are about 1mm in length with
measured distance between traces about 150 µm. The probability to fail on the unit length at 1mm on comb
pattern device is assumed to be same in the microstimulator after correcting for the different voltage stress
(see below Eq. 7). The voltage aging factor 𝐴𝐹
𝑣𝑜𝑙𝑡𝑎𝑔𝑒 between the microstimulators and comb pattern
110
devices is determined by the inverse power relationship. In the accelerated life-testing, the microstimulator
is powered 24/7 at its maximal rated output. In actual practice, the microstimulator will be only used by
patients for no more than 2 hours per day in the targeted treatment of stress urinary incontinence. Because
the increase of use rate is assumed to accelerate the device failure linearly [131], the aging factor of use rate
𝐴𝐹
𝑢𝑠𝑒 𝑟𝑎𝑡𝑒 is increased by a factor of 12 in continuous use as shown in Eq. 8 (This factor will be even larger
for likely use conditions at lower frequencies and/or lower pulse strengths, which reduce the duty cycle of
the RF bursts powering each output pulse of the implant). Overall, the following derived equation can be
used to provide quantitative estimation of encapsulant failure fate of the microstimulator,
𝑃 𝑓 (𝑁𝑢𝑆𝑡𝑖𝑚 )
= 𝑃 𝑓 (𝑢𝑛𝑖𝑡 =1𝑚𝑚 )
.
1
𝐴𝐹
𝑣𝑜𝑙𝑡𝑎𝑔𝑒 .
1
𝐴𝐹
𝑢 𝑠 𝑒 𝑟𝑎𝑡𝑒 Eq. 6
Where, 𝐴𝐹
𝑣𝑜𝑙𝑡𝑎𝑔𝑒 = (
𝐸 𝑐𝑜𝑚𝑏 𝐸 𝑁𝑢𝑆𝑡𝑖𝑚 )
β
= (
𝑉 𝑐 𝑑 𝑐 𝑉 𝑁 𝑑 𝑁 )
β
Eq. 7
Where, 𝐴𝐹
𝑢𝑠𝑒𝑟𝑎𝑡𝑒 =
24 ℎ𝑟𝑠
2 ℎ𝑟𝑠
Eq. 8
In Eq.6, 𝑃 𝑓 (𝑁𝑢𝑆𝑡𝑖𝑚 )
is the probability of the microstimulator epoxy encapsulant failure after one year of
maximal intended use in the human body. 𝑃 𝑓 (𝑢𝑛𝑖𝑡 =1 𝑚𝑚 )
is the probability of failure of the comb pattern,
which is 0.59%. The aging factor of voltage 𝐴𝐹
𝑣𝑜𝑙𝑡𝑎𝑔𝑒 is 1.64, where voltage stress 𝐸 𝑐𝑜𝑚𝑏 = 168 𝑉 /𝑚𝑚
with 𝑉 𝑐 = 4.2 𝑉 and 𝑑 𝑐 = 25 µm, and voltage stress 𝐸 𝑁𝑢𝑆𝑡𝑖𝑚 = 100 𝑉 /𝑚𝑚 with 𝑉 𝑐 = 15 𝑉 and 𝑑 𝑐 =
150 µm, with assumption of β = 1 indicating a linear effect of voltage stress.
The quantitative estimation of the microstimulator failure rate at one year 𝑃 𝑓 (𝑁𝑢𝑆𝑡𝑖𝑚 )
is about 0.03% based
on consideration of aging factors from temperature, voltage stress and use rate. This estimated rate is
surprisingly low but further supported by the observed result in the microstimulator accelerated life-testing,
which had no devices failed due to encapsulation after 180 days testing at elevated temperature and maximal
output. All devices (excluding the two that failed for other reasons) have functioned for the equivalent of >1
year at 37°C (the original design goal for the clinical application) despite continuous powering. This
conclusion must be tempered by the relatively small number of devices tested to date, which results in a
111
relatively large statistical uncertainty about actual reliability. The accelerated life-test methods described
herein now need to be applied systematically to larger samples of devices. According to parametric
binomial statistics, 36 comb pattern device samples with 4 failures occurring over 152 days in accelerated
life-testing would demonstrate the claimed reliability at a 90% confidence level [132]. The much higher
reliability extrapolated for the microstimulators would require a much larger sample and/or a much longer
test period to confirm empirically. Conversely, failures occurring more frequently would suggest failure
modes different from those related to the epoxy encapsulation in the comb pattern devices.
Non-hermetic encapsulation relies on adhesive bonding to the component surfaces that is maintained by
electrostatic attraction and mechanical interlocking [120]. The cleaning procedure is a critical step for
encapsulation performance because even a monolayer of a surface contaminant can prevent the necessary
bonds from forming. The major contaminant residue left on a populated circuit board is flux in the solder
paste. It is formulated from many ionic and organic compounds that include reactive species. The soldering
problem can be managed by using water soluble solder paste. The deionized water was not as effective as
distilled water to clean the subassembly. Deionization removes ions such as sodium and chloride but may
leave or even introduce non-ionic contaminants. Deionized water is typically shipped and stored in plastic
bottles, from which oligomers, catalysts, and plasticizers may leech. Water distilled and stored in glass
bottles should be used for final rinses to prevent contaminants from coating surfaces after the water
evaporates away. The fully cured epoxy Epotek 302-3M has high chemical stability, strong dielectric
properties, low water uptake and slow diffusion rate; it was found to be a reliable encapsulant [133]. Low
permeability itself cannot prevent water vapor diffusion throughout a device as small as the microstimulator,
but it tends to be associated with low water absorption; such absorption and swelling can produce
mechanical stresses that may disrupt chemical adhesion, leaving voids for water condensation. Strong
adhesion between encapsulant and active circuitry can reliably prevent water vapor condensation and
corrosion for many years [134]. Although the epoxy thermal expansion would induce extra pressure,
thereby improving adhesion and decreasing the likelihood of delamination, it is negligible in this study due
to the small cross-sectional area of the epoxy above the PCB and the low thermal expansion coefficient.
112
Although the comb pattern device has been made to mimic the fabrication process of the functional wireless
microstimulator, there are still several differences that may affect the encapsulation performance in various
ways. First, the microstimulator has more discrete components and complicated structure than the comb
pattern device, which has only two Zener diodes on the PCB located away from the sensitive interdigitated
electrodes. Compared to the unbroken flat surface of the comb pattern, the microstimulator might be
subjected to residual stress due to non-uniform shrinkage and shear forces from epoxy curing, which could
introduce potential voids for water vapor condensation. Although hybrid discrete electronic components
are generally not vulnerable to water vapor, these components and their solder joints are susceptible to
liquid water that might condense in a void between components and circuit board or a cavity within a
component. The cleaning procedure is sufficient to get rid of contamination on the board when components
are placed and soldered properly. The controlled low rate of epoxy infiltration and curing under pressure
can help epoxy penetration into cavities and compress any residual bubbles to provide a defect-free
encapsulation [135]. Second, wire-bonding in microelectronic circuits is another vulnerability with its own
failure modes, which were not present in the comb pattern devices. The thermosonic gold wire-bonding to
aluminum pads in the microstimulator inevitably forms intermetallic compounds at the junction that tend
to have poor electrical conductivity, the source of the purple-plague failure of one device. Wire-bond quality
is controlled by proper cleaning procedure before bonding and the settings of the wire-bonding machine.
Our cleaning procedure used ultrasound with solvents to clean the pad surface, resulting in mechanically
acceptable wire-bonds with a breaking force of 7 to 12 grams. The thickness and resistance of the
intermetallic compound at the gold-aluminum junction needs to be minimized by providing an appropriate
substrate temperature during bonding [136, 137].
CONCLUSION
We have evaluated the functional long-term reliability of miniature, implantable, wireless electronic
devices that rely on non-hermetic, epoxy encapsulants. The lifetime was measured by an encapsulated
comb pattern with high density of interdigitated electrodes that was used to detect incipient failures. The
113
capacitance of the comb pattern forms a resonant circuit with the inductor by which the implant receives
power. Any moisture affects both the resonant frequency and the Q-factor of the resonance, which was
detected wirelessly by its effects on the coupling between two orthogonal RF coils placed around the device.
Inductive coupling of an RF magnetic field was used in the accelerated life-testing to provide DC bias.
Strong adhesive bonding between epoxy and electronic circuity proved to be necessary for package
reliability; this can be achieved by proper cleaning and encapsulation procedures. Comb pattern devices
were estimated to have 78.8% reliability for one year, which extrapolated to 1 year reliability of 99.97%
for the neuromuscular stimulator after accounting for various stressors in the accelerated life-test. This is
consistent with the results from accelerated life-testing of a small sample of the microstimulators, in which
there were no encapsulant failures even beyond 6 months of continuous output at elevated temperature.
ACKNOWLEDGMENTS
The authors would like to thank engineers Ray Peck, Sisi Shi, and Longpeng Jiao for help in design and
manufacturing. The project is funded by General Stim Inc.
114
CHAPTER 7: PROPOSAL FOR CLINICAL VALIDATION
PREFACE
The majority of patients with mild to moderate stress incontinence have weakness of the external urethral
sphincter (EUS) muscle, usually as a result of damage during childbirth or surgery and often exacerbated
by hypotrophic changes associated with aging and declining hormone levels. Like any striated skeletal
muscle, the external sphincter responds to exercise by increasing its bulk and strength. This is the basis of
Kegel exercises, which are highly effective when done properly and conscientiously. Unfortunately, many
patients lack the ability to voluntarily contract these muscles fully and/or are unwilling to do the exercises
frequently enough to obtain the full benefit. Kegel exercises, when properly done, will effectively treat
most cases of urinary stress incontinence by strengthening the pelvic floor muscles including the EUS. As
far as the muscle fibers are concerned, there is no difference between voluntary exercise and electrically
stimulated exercise. Because the muscle nerves lie under highly innervated skin and mucosa, conventional
transcutaneous, intravaginal or intrarectal stimulation tends to produce unpleasant sensations. We have
developed an injectable and wireless microstimulator (NuStim®) that can be implanted percutaneously into
the pelvic floor muscles. Electrical stimulation of the intramuscular motor axons generates strong
contractions of the EUS without producing unpleasant sensations or requiring voluntary effort. The power
and command signal are transmitted inductively to the device without percutaneous wires or leads,
minimizing dangers of infection and failure. Compared to conventional surgical treatments for stress
incontinence, we hypothesize that this treatment will be less invasive, less expensive, and have fewer post-
operative and long-term complications, while achieving significant reduction of urinary leakage in patients
with moderate stress incontinence from innervated but hypotrophic EUS muscle.
There is an extensive literature about how muscles respond to voluntary and electrically induced exercise
programs, including prior published research with a similar injectable microstimulator called the BION
[121, 138]. A battery powered BION somewhat longer than the NuStim was approved in Europe with the
CE-Mark for treatment of urinary urge incontinence by stimulation of the pudendal nerve[139, 140].
However, there are several questions that need to be answered in a clinical trial:
115
• There is no animal model for stress incontinence. We demonstrated safety and efficacy of
intramuscular stimulation in hind limb skeletal muscle of beagle dogs. Human pelvic floor muscles
have different neuromuscular architecture.
• We do not know what stimulus patterns will be comfortable for patients. If we implant the
stimulator close to sensory fibers, the stimulation intensity will be limited due to discomfort and
stimulation may not be strong enough for effective muscle exercise. The proposed implantation
procedure should minimize this possibility by finding a low motor threshold location before release
of the NuStim.
• We do not know what specific electrical stimulation parameters will be effective and efficient for
training the EUS muscle to treat urinary stress incontinence. For typical urinary stress incontinence
patients with severe EUS atrophy, muscle fibers are easy to fatigue. Slow-twitch muscle fibers are
non-fatiguing, so capable of continuous or frequent contractions; fast-twitch muscle fibers are more
easily fatigued but can respond more rapidly and forcefully to sudden recruitment such as during
coughing. Low frequency and high intensity stimulation pulses tends to reverse disuse muscle
atrophy, building bulk and peak force generation[110]. Longer cycles of stimulation with
intermediate frequencies and longer exercise periods tends to improve fatigue resistance[111].
High frequency bursts of stimulation produce maximal contractile force. The sensations produced
by these different patterns of muscle contraction may be more or less desirable for the subjects.
• We don’t know if we need to stimulate the pelvic floor on both sides. Acute, percutaneous
stimulation via needle electrodes showed no difference in urethral closure pressure between
unilateral and bilateral stimulation.
This pilot study is intended to identify stimulus and exercise parameters that achieve a useful balance of
acceptability for the subjects and clinical improvement of their incontinence.
116
OBJECTIVES
The objective of this clinical trial is to demonstrate that a minimally invasive microstimulator can be used
to activate and strengthen the pelvic floor muscles similarly to Kegel exercises. Acute stimulation via
percutaneous needle electrodes achieved strong muscle contraction of the EUS. The main question that
must be answered in the pilot study is whether we can produce electrical stimulation patterns that are both
effective and comfortable chronically. Up to 10 subjects will participate in this pilot study, which will
provide valuable information about basic physiology of the pelvic nerve network and its responses to
electrical stimulation. This will allow us to determine and optimize the implantation site and procedure and
the stimulation parameters for reversing hypotrophy and building strength. We will evaluate and, if
necessary, refine the clinical outcome measure that quantifies urine leakage during normal daily activities.
This pilot study will test the performance characteristics and capabilities of study designs, measures,
procedures, recruitment criteria, and operational strategies that will be employed in a subsequent pivotal
clinical trial to demonstrate safety and efficacy in order to obtain premarket approval of a clinical product.
HYPOTHESIS
The NuStim system provides comfortable and safe exercise of the pelvic floor muscles that is effective in
reducing urinary stress incontinence.
PILOT CLINICAL TRIAL PROTOCOL
RECRUITMENT
Subjects are recruited from the urological practice of Dr. Liao and his colleagues, or referred by other
hospitals. The opportunity is offered to participate in this research study without cost or remuneration
beyond expenses.
117
INCLUSION CRITERIA
1. Female patient age 35-70 who is diagnosed with mild or moderate urinary stress incontinence for
more than 6 months; and male patients with stress urinary incontinence at least 1 year after
prostatectomy, and a low PSA <1.0ng/ml consistent with no active malignancy.
2. Patient qualified for NuStim treatment after diagnosis.
3. Patient wants to participate in the clinical trial and the patient or the guardian is willing to sign the
consent form.
4. Patient who is cooperative, willing to communicate with the researchers, and follow the trail
requirement.
EXCLUSION CRITERIA
1. History of spinal cord injury or/and abnormal spinal cord
2. Overactive bladder
3. Urinary tract and related infection
4. History of surgery treatments for stress incontinence and still in recovery (prostatectomy is not
included)
5. Narrow urethra after prostatectomy (Residual bladder urine >100ml)
6. Intrinsic urethra sphincter deficiency (Type 3 stress urinary incontinence)
7. History of neurodegenerative and neuromuscular diseases
8. History of malignant tumor in pelvic organ or ovary.
9. Severe perineum trauma from congenital perineum malformation
10. History of fecal incontinence surgery, bladder surgery, fistula surgery, and other similar surgery
that can be clearly related to the cause of urinary incontinence
11. Coagulation disorders
12. Immune system disorders
13. Female patients with plan of pregnancy, pregnancy and lactation in the trial period
14. Epilepsy
118
15. Heart disease
16. Acute infectious disease
17. Fever
18. Abnormal blood pressure
19. Dementia, brain atrophy, cognitive disorders, cerebral vascular disease and sequelae
20. Acute local or systemic bacterial infection without proper control
21. Ill or dying patient who fail to cooperate and evaluate the efficacy
22. Mental or/and neurological disorders who fail to cooperate with physician
23. Constipation patient
24. Female patient without vaginal delivery history
25. patients who are prone to allergies and/or scarring
26. Patients who have participated in other clinical trial in previous 3 months
27. Planned to change lifestyle of routine habit
28. Other condition that may not proper for this study after evaluation
INFORMED CONSENT
Patients who meet the above criteria and are considering participation in this study will be seen as out-
patients. The Informed Consent Form attached in the Appendix will be provided to and read by prospective
subjects and explained to them by a urologist who has no financial interest in General stim Inc. or the
NuStim technology.
BASELINE MEASUREMENT
For the duration of their participation in the study, all subjects will be asked to perform 1 hour pad testing
using absorptive undergarments similar to those commonly used by most patients with urinary stress
incontinence. All provided undergarments are pre-weighted in the clinic, thereby allowing to measure urine
leakage in the testing. The subject will be asked to fast at least 3 hours before the testing and consume a
119
500ml liquid fluid within 15 mins. The subject will then wear the absorptive undergarment and perform a
series of standard activities in one hour. The used pad will be weighed and counted to establish baseline
leakage. The patient will be also given the bladder diary and enough instruction to record their bladder
diary throughout the treatment and follow-up period. The diary will be scan at each visit and stored as
patient record for treatment evaluation. Questionnaire of incontinence symptom improvement and quality
of life will be provided to the patient and information will then be extracted and documented by the
physician. All similar collection and measurement will be done between device implantation and device
activation and at regular intervals throughout the remainder of the treatment and follow-up period.
Urodynamic testing is only performed before the implantation and the end of treatment. It is quite invasive
procedure but provide EUS performance in a quantitative way. The scheduled event in each visit is shown
in the attached table in appendix.
DEVICE IMPLANTATION
The procedure is to position the cathodal stimulating electrode of the NuStim in a location in the pelvic
floor muscles that can produce strong contractions that will exercise the external urethral sphincter. The
pelvic floor muscles are diffusely innervated by highly branched motor axons, so the exact location is not
critical as long as the NuStim implant is at the correct depth. The procedure consists of test stimulation to
find the approximate target followed by delivery of the NuStim implant to the target through a specially
designed insertion tool. After implantation, the NuStim should not be tested or used for exercise for at least
4-6 days to allow it to become fixed in the muscle.
SETTING UP IMPLANTATION EQUIPMENT
The NuStim implant comes in a wrapped presterilized tray with disposable tools for testing and implantation.
It should be opened on a small sterile field with the usual sterile technique.
The knob on the stimulator allows you to turn the stimulator on and deliver test pulses at 2 pulses per second
with an intensity that ranges from about 3 to 13 clinical units as marked on the stimulator. Pull the small
120
plastic tab hanging from the back completely out to activate the stimulator. The LED will flash when each
stimulus is delivered.
LOCATION CONFIRMATION
Patient is in lithotomy position with regular disinfection on the perineum. Before the NuStim implantation,
an off-the-shelf standard EMG needle is first used to locate the implantation site. The EMG needle is
connected with the stimulator via a custom-made adapter. The start of stimulus strength is 3 (medium
intensity). The ideal implantation side is defined as low intensity stimulus (less than 4mA x 0.2 ms, i.e.,
less than NuStim stimulation level 8) can produce strong pelvic floor muscle and external urethral sphincter
contraction. Based on previous experiment, the physician can identify the contraction based on palpation
or observation, the patient can identify the subjective contraction on this region.
NUSTIM IMPLANTATION
Pick an entry site that allows you to reach the target with the insertion tool oriented approximately
perpendicular to the perineum. Anesthetize the skin and make a 3-mm long stab wound through the skin
at this entry site.
Insert the back end of the long-insulated needle into the socket on the front of the stimulator. Assemble the
insertion tool by passing the long-insulated needle through the dilator and passing the dilator through the
sheath. When correctly assembled, the tip of the needle should protrude about 2 mm past the end of the
dilator and the taper on the dilator should end just past the end of the sheath.
Figure 7-1. Assembled insertion tool. Patient position and implantation angle.
121
Insert the insertion tool in the desired location until it is at least 1 cm under the skin and turn the stimulator
up until the patient feels a twitch. Advance 1 cm at a time, using the marks on the sheath as a guide. Do
NOT turn the stimulator to a higher strength until then to avoid stimulating the skin nerves. Turn the
stimulation up until the threshold where the patient reports feeling a muscle twitch. As you push the
electrode further in, the stimulation strength required to produce a twitch should go down. When you take
the next 1 cm step past the target, the threshold will probably go up. At this depth, the end of the sheath is
now at the target depth. The depth at which the threshold is lowest is the target location for the NuStim.
You should be able to elicit a progressively stronger muscle twitch by going 4-6 clinical units higher at this
location. Leave the skin electrode connected.
If a satisfactory twitch cannot be produced, make an additional incision at another location and repeat the
procedure.
Remove the stimulator with the needle and dilator from the sheath without moving the sheath. Insert the
NuStim into the back of the sheath so that the NuStim enters the lumen of the sheath. (Please confirm the
NuStim is placed in the correct direction. The NuStim electrode on the PCB board side should be facing
toward the tissue). Assemble the dilator and needle back to push the NuStim inside the sheath. You will
start to feel resistance. Stop when the first marking on the dilator reaches the top of the sheath. The
NuStim implant will then be at the end of the sheath.
When the needle on the stimulator contacts the back of the NuStim implant, the stimulation pulses pass
through the implant to its cathodal stimulating electrode. Use the stimulator to confirm that you can elicit
a muscle twitch at a strength within three clinical units of the lowest threshold that you obtained previously.
If a twitch cannot be reproduced within three clinical units with the NuStim inside, pull out the insertion
tool and discard the NuStim. Use a new NuStim implant and repeat the previous procedure. If this does not
work, probe again for a new location using the procedure outlined above. Hold the stimulator, probe and
dilator in place and gently pull the back on the sheath until it reaches the second marking on the dilator to
122
release the NuStim at this location. Do NOT use the dilator to push the NuStim out of the sheath because
it will wind up deeper than your target.
Remove all parts of the implantation and test equipment from the patient. Apply a small, sterile dressing
over the implantation site. Do NOT test the implant or use it for exercise for at least 4-6 days to allow it to
become fixed in the muscle.
If you decide to implant a second NuStim in a patient, be aware that BOTH implants will generate the same
stimulation pulses at the same time when the patient sits on the cushion.
Figure 7-2. Implantation procedure. A. Identify target location using EMG needle. B. Confirm target
location using assembled insertion tool. C. Re-confirm location before release of NuStim. D. Release the
NuStim.
123
DEVICE ACTIVATION AND EXERCISE PRESCRIPTION
Approximately one week after implantation, the physician or designated caregiver will use the external
hardware (RF-cushion and tablet computer) and software (graphical application on the tablet computer) to
determine the range of stimulus intensities from threshold to maximal comfortable level, based on the
subject’s report of the sensations experienced. The physician will set up an exercise program that provides
strong, cyclical contractions and relaxations with parameters that are comfortable for the patient to use for
~30-60 minutes/day. As experience is gained with parameters that appear to produce comfortable and
effective exercise, the default parameters in the physician’s software may be modified by the study sponsor
to indicate these recommended values.
IN-HOME EXERCISE SESSIONS
The subject will take home their RF-cushion paired with a tablet computer or smartphone on which the
prescribed exercise parameters have been stored. The subject will be instructed to self-administer the
prescribed exercise on a daily basis. The subject will be encouraged to use the highest comfortable stimulus
strength from the range prescribed by the physician, which will result in accomplishing the prescribed daily
exercise in the shortest period of time according to an algorithm for tracking subject compliance in her
personal tablet computer or smartphone. The subject will be asked to keep a hand-written diary describing
their usage of and experiences with the exercise system and their subjective assessment of their incontinence
symptoms. It is expected that subjects will require 3-6 months of regular exercise to show significant
improvements in stress incontinence.
OUTCOME MEASUREMENT
The subjects will be followed in the outpatient clinic 3 weeks, 8 weeks, and 13 weeks after implant
activation, at which time their compliance with the prescribed exercise will be read from their tablet
computer or smartphone and their incontinence will be measured from their collected used undergarments.
At each visit, 1-hour pad testing will be performed to measure the change of urine leakage. Subjects will
124
be asked to complete a questionnaire regarding their use of and satisfaction with the treatment system and
to quantify their symptoms and quality of life. Key information will be extracted from their personal diaries
and incorporated onto Case Report Forms with coded subject identifiers as provided or amended during the
course of these pilot studies. The primary and secondary outcome measurements are provided as follows.
PRIMARY OUTCOME MEASUREMENT
• 1h pad testing after 13 weeks of treatment compared to the baseline changes
SECONDARY OUTCOME MEASUREMENT
• Subjective satisfaction and quality of life change after 3 weeks, 8 weeks, and 13 weeks of treatment
to the baseline.
• Incontinence diary (includes 24 hours of total urination, urine volume/time, urgency score, mean
improvement urine leakage) change after 3 weeks, 8 weeks, and 13 weeks of treatment to the
baseline.
• Urodynamic measurement of abdominal leak point pressure (ALPP, includes VLPP and CLPP)
change after 13 weeks of treatment to the baseline.
• Urethral pressure profile (UPP includes MUP, MUCP) change after 13 weeks of treatment to the
baseline.
• Patient perception of bladder condition (PPBC) change after 13 weeks of treatment to the baseline.
• Pelvic floor muscle contraction force change after 13 weeks of treatment to the baseline.
ADJUSTMENT OF EXERCISE PRESCRIPTION
At each follow-up visit, the physician will repeat the determination of threshold and maximal comfortable
stimulation intensity. This provides an indication of the stability and integrity of the implant and
surrounding tissues. It is also likely that these values will change slowly as the exercised muscle increases
its bulk and the subjects become accustomed to the sensations produced by the muscle contractions. The
exercise programs will be adjusted to continue to encourage the subjects to use the strongest tolerable
125
stimulation, thereby exercising the largest number of muscle fibers. If the physician feels that the subject’s
incontinence is not improving as desired, the other parameters of the prescribed exercise such as stimulation
rate, contraction duration and exercise time per day may be changed within the ranges provided by the
prescribing software.
IMPLANTATION OF SECOND DEVICE
If the physician feels that the subject’s incontinence is not improving as desired, a second NuStim may be
implanted in the contralateral pelvic floor muscles using the same procedure as for the first implant. After
at least five days without stimulation for the implant to stabilize, a new exercise program that activates both
devices will be prescribed.
X-RAY OF DEVICE LOCATIONS
If the physician feels that the location of the implanted NuStim(s) is unusually favorable or unfavorable or
if there is some question about the ability to transmit sufficient power from the RF-cushion to the implant(s),
the physician may request an x-ray in the seated position (AP, lateral or both views) to determine the actual
location and orientation of the implanted NuStim(s).
DISCONTINUATION OF EXERCISE
The subjects are expected to discontinue exercise at some point within one year of device implantation,
which may occur when any of the following situations arises:
• The subject has achieved essentially complete relief from urinary stress incontinence.
• The subject’s condition has shown no signs of further improvement.
• The subject’s implanted NuStim(s) are no longer capable of generating EUS muscle contractions.
• The subject decides to withdraw from the study for any reason.
126
At this point, the subject may be asked to return the RF-cushion and/or tablet computer or smartphone,
which may be cleaned and refurbished to be used by other subjects in the study. The subjects will be asked
to continue to maintain their diary to record their subjective evaluation of their incontinence symptoms.
These will be collected and evaluated in the out-patient clinic at occasional intervals for as long as the study
is underway.
REMOVAL OF NUSTIM IMPLANTS
The NuStim is made from inert, biocompatible materials. It is designed to be left in the body permanently,
whether it is functional or not, similar to medical devices such as nonresorbable sutures, vascular clips and
orthopedic bone screws. Removing a NuStim from the pelvic floor muscle would require a minor surgery
under local anesthesia. If there is a medical indication for its removal such as infection or pain, the NuStim
will be removed surgically; otherwise it will be left implanted indefinitely.
ANTICIPATED DISCOMFORT AND RISKS OF ADVERSE EVENTS
IMPLANTATION PROCEDURE
This is a minimally invasive procedure using sterile, biocompatible instruments and devices. There is a
minimal risk of subcutaneous or intramuscular bleeding, damage to intramuscular nerve fibers, or post-
operative infection. These may cause temporary discomfort but should resolve with time and/or
conventional medical treatment.
If the implanted NuStim becomes infected, it will probably need to be removed via a minor surgical
procedure to resolve the infection using conventional antibiotic therapy.
The probe electrode is a small caliber needle that should cause minimal discomfort when inserted
percutaneously into the pelvic floor muscles. The injection of local anesthetic will cause some transient
discomfort but will minimize sensation during an incision to enlarge the entry site and as the dilator is
pushed through the skin and underlying fascia. Deep anesthesia of the target muscle must be avoided
127
because it would block the muscle contractions that the patient must perceive and report to guide the
implantation process.
STIMULATION IN THE CLINIC
The subjects will experience unusual sensations from the electrically induced contraction of the EUS
muscles, but we do not expect these sensations to be painful because the stimulation is not near skin or
cutaneous nerves. The software requires the clinician to increase the stimulation gradually from the lowest
levels so that he/she can stop before painful levels are reached.
EXERCISE AT HOME
The subject has full control of the stimulus intensity via the application software, which always starts at the
minimal threshold level and must be turned up by the subject to the desired intensity. The NuStim implant
has no internal power source, so all stimulation can be stopped immediately by getting up from the RF-
cushion on which the subject normally sits for the exercise treatment. The RF-cushion detects whether the
subject is actually present on it and turns itself off when the subject is not present.
The RF-cushion is powered by a medical-grade 12 VDC power supply that plugs into the AC power outlet
and is connected to the RF-cushion by a magnetically coupled connector that detaches easily if the cord is
pulled. The RF-cushion has an outer shell of waterproof polyurethane skin and foam with no exposed
electrical components or controls, so should not pose an electrical shock hazard even if the patient leaks
urine onto it.
The maximal electromagnetic field strength emitted by the RF-cushion is below the standard safety limit
of 39V/m, thereby giving specific absorptive rate (SAR) about 0.75W/kg, which is much lower than cellular
phone SAR from Federal Communications Commission.
128
CHAPTER 8: TRANSLATIONAL AND REGULATORY
PATHS
CFDA PATHWAY
In order to advance the NuStim neurostimulator from an initial prototype to a commercial product available
in China, several steps must be taken along a regulatory pathway that the Chinese Food and Drug
Administration (CFDA) has, to some extent, adapted from medical device regulations of the United States
Food and Drug Administration (FDA). Based on Rules for Classification of Medical Devices from CFDA,
all types of implantable medical devices are classified as Class III medical devices, which require a pathway
similar to FDA requirements for Class III medical devices pursuing Premarket Approval (PMA). Since the
NuStim is an innovative Class III product with no similar product on the market, more effort will be needed
to assure its commercial registration.
The NuStim regulatory pathway and timeline can be divided into three phases. In phase 1, the system is
tested by a third party agency that assesses compliance with relevant standards. At the same time, the safety
and efficacy of the system is tested by performing a series of animal studies. After these two sets of activities
are complete, the project then goes into phase 2, a step requiring clinical validation. Unlike the situation in
the US, documents need not be submitted to CFDA prior to starting these clinical trials. CFDA only needs
to be notified through a registration procedure through the state CFDA office. The study must be conducted
at one of a limited set of government hospitals authorized to run clinical trials, however, and all relevant
test reports and clinical documents must be submitted, reviewed and approved by the hospital IRB prior to
the start of the clinical trial. During these two phases, the sponsor should follow the Chinese requirements
for Good Manufacturing Practice (GMP) and the ISO13485 quality system requirements to finalize its own
quality system, which will be audited by a third party to obtain certification. After clinical validation, all
reports are ready for CFDA submission to obtain a PMA. At this phase, the team should be prepared to
answer questions and host a preapproval quality audit by CFDA.
129
Figure 8-1. NuStim regulatory pathway and timeline.
STANDARDS TESTING
The system of standards testing was instituted when medical devices in China were either imports or copies
of pre-existing medical devices whose specifications and operation were already well-understood. For a
novel device such as the NuStim, the manufacturer must develop and justify its specific standards within a
general regulatory framework. The purpose of testing the device to these standards is to have a certified
third party ensure the compliance of product to design specifications and general regulations. The testing
of the NuStim system was directed at its two principal components:
• The internal system of the implanted microstimulator
• The external transmitter in a seat cushion
The product as a whole has two additional parts, the insertion tool and the software. These parts were
validated as part of the testing of the major components. The implantation tool, customized for implantation
of the NuStim, was validated simultaneously with the microstimulator. The Android smart-phone/tablet
app is designed to work exclusively with the NuStim SUI treatment, and cannot be sold as stand-alone
130
software for other purposes. Therefore, the software can bypass most standards testing since it was validated
as an integral part of the transmitting cushion that it controls.
IMPLANT STANDARD TESTING
The NuStim neurostimulator is regulated under ISO14708-3: Implant for Surgery–Active Implantable
Medical Device–Part 3: Implantable Neurostimulators, which is specific for medical devices intended to
electrically stimulate the central or peripheral nervous system. Guidance, requirements, and test methods
are listed in the standard, and these should be considered at the beginning of design. Some examples of its
requirements include protection of the medical device from electrostatic discharge, radiation, and external
defibrillators. Biocompatibility testing is another major set of tests called out by ISO14708-3. It is important
that implantable medical devices do not cause any undesirable local or systemic effects in human body.
The general requirements describe tests to assess toxicity, carcinogenicity, hemocompatibility, for example.
Some of these tests are simple in vitro tests, while others require chronic animal experiments.
The proposed NuStim device was designed to be implanted in human body chronically with direct contact
to soft tissues. Therefore, it was tested extensively in vitro and in vivo tests in a designated test lab for the
biocompatibility. The most onerous of the tests involved implantation for periods of up to 6 months.
Because the biocompatibility testing usually takes a longer time than other tests, we scheduled the
biocompatibility testing in advance. Some details of test methods are not provided in the ISO 14708-3, but
they can be found in the references listed in the standard. Because NuStim is an innovative product with
new features other than listed in the standard, we determined the requirements, test methods, and expected
test results for those new features. The testing agency then used the standards that we provided to perform
the testing.
EXTERNAL RF-CUSHION STANDARD TESTING
The external RF-Cushion is regulated by GB9706.1, which is the general electrical safety regulation for
medical devices used in the home. Because the RF-Cushion is designed to generate electromagnetic fields,
131
the electromagnetic compatibility (EMC) is the most difficult set of specifications to meet. This requirement
in GB9706.1 refers to YY0505/IEC60601 which is a general EMC standard for medical devices.
ANIMAL TESTING
The objective of animal testing is to validate the safety and efficacy of the system before human use.
General rules are in place to govern the conduct of all animal studies in an ethical way without unnecessary
harm to animals. The design of the animal testing is determined by the sponsor to gather the best information
possible regarding safety and mechanism of action in order to justify readiness for subsequent human
studies. The animal study reports (attached in Appendix-3: Animal Study Report) that were submitted for
IRB review describes the details of this study.
CLINICAL TRIAL
Unlike the United States where clinical trial rules have been in place for more than 50 years, China has
limited experience in performing clinical trials for innovative medical devices whose efficacy has not been
established in other jurisdictions. Facilities are typically only accustomed to carrying out clinical trials on
medical devices that resemble imported products and whose minor modifications are likely to produce
similar, predictable clinical results. For innovative medical device like NuStim, the rules are different and
more similar to the Investigational Device Exemption process of the US FDA. The device must be subject
to two clinical trials; the pilot clinical trial studies the new treatment in a small group of patients whereas
the pivotal clinical trial validates the efficacy in a large group of patients located in multiple clinical research
centers. The protocol (Chapter 7: Proposal for Clinical Validation) that was submitted for IRB review
described the design of the pilot clinical trial.
132
INNOVATIVE MEDICAL DEVICE PATH
The usual approach for the review of a medical device is to submit relevant information to CMDE. CMDE
performs a technical review of the device application and provides recommendation to CFDA for approval.
Under standard review procedures, medical devices are placed on a waiting list for panel review. In 2014,
CFDA released a new policy for a “priority review” pathway designed for certain medical devices with
breakthrough technologies. The goal of this policy is to facilitate direct communication with the Center for
Medical Device Evaluation (CMDE) to accelerate the novel applications. The proposed pathway puts the
project on a priority list for more expedited review. Also, the proposed pathway allows expedited review
in the complementary standards testing phase. Therefore, the priority review pathway provides a significant
advantage to applicants. The specific timeline for review process is not provided.
Under the new policy, the CFDA considers the medical device appropriate for the priority pathway if the
device meets all conditions as follows:
The applicant owns the core technology or intellectual property after innovation activities; or has been
authorized to use the innovative technologies; or has patent in application at the time of review process
The device has an innovative working principle/application, significant improvement in efficacy or safety
compared to similar products, leading technology in international level, and significant clinical impact.
The applicant has finished the early research and prototypes. The research is complete and follow the quality
regulation. All research data are well documented and traceable.
After CMDE determination of the eligibility for innovative devices, a list of medical devices considered
for priority review will be published.
Unfortunately, The NuStim system was not approved for priority review. The major reason is there is no
precedent for this product and application. Although researchers have shown efficacy of using electrical
stimulation for SUI, no available device exists in the market for this type of application. Ironically, the
priority review is granted for a product which is simply a copy of existing product with minor modification.
133
Evaluation of efficacy and safety on an innovative product requires strong expertise and takes
responsibilities of possible unexpected adverse effects. It is a much riskier decision for the reviewer than
granting the priority to products with expected efficacy and safety. It is possible to have a higher chance of
acquiring priority review for NuStim after the pilot clinical study with data of the treatment. Overall, the
regulatory experience with NuStim should provide useful guidance for other innovative medical devices
that intend to enter the Chinese market.
134
CHAPTER 9. PILOT CLINICAL STUDY
OBJECTIVE
This chapter presents the collected data and results of the ongoing pilot clinical study of the NuStim system.
As of the date of the thesis submission, there is one male patient who met inclusion/exclusion criteria to be
included in the study. His data are presented below as a case study in progress.
Timeline: The male patient agreed to participate in the study and signed the Informed Consent Form on
May 25, 2017. NuStim implantation was on June 1, 2017. The 1
st
stimulus parameters testing was on June
8, 2017. The 1
st
follow up of this patient was on June 26, 2017. The 2
nd
follow up was on August 3, 2017.
RESULTS
MEDICAL HISTORY
The patient underwent radical prostatectomy for cancer on June 4, 2013, and was diagnosed with stress
urinary incontinence after the surgery. The patient was taught to perform the PFMT for treatment of urine
leakage, but he could not persist on this exercise, therefore no sufficient improvement was obtained. He
then tried prescribed Chinese herb medication, but there was no improvement. Patient then stopped the
medication in April 2017. The symptom of leakage was severe initially after the surgery, then had slight
improvement after recovery of the surgery and has been stable for a long period of time. The patient has
frequent urine leakage during coughing, physical activities and on the way to restroom. He regularly uses
absorptive undergarments.
DIAGNOSIS
The prostate-specific antigen (PSA) level is less than 0.003 ng/ml, which is within a normal range,
indicating absence of prostate cancer. The patient only experiences urine leakage during activities that
increase abdominal pressure, like standing from sitting position. From the 1-hour pad testing results a few
days before NuStim implantation, he had average of 41.5g leakage, which is classified as severe leakage.
135
IMPLANTATION
The implantation site was determined during the implantation procedure as specified in the protocol. The
actual delivered charges at each level were shown in the table 9-1. When using the needle electrode with 2
pps pulses, a subjective tapping feeling was reported at level 6 at depth about 4cm which was estimated
visually. Pelvic floor area contraction could be seen at level 12-13. The assembled insertion tool was then
inserted parallel to the needle electrode, aiming the same target as determined previously. The threshold
level was reconfirmed at level 5 at the depth about 4.5 cm which was measured by markers on the sheath.
The insertion tool was then pushed about 0.5cm further past the target and the threshold value increased to
level 10. The insertion tool needle and dilator were withdrawn and NuStim was loaded for implantation.
The threshold level was measured at level 10 before release of NuStim. The determined target site was
about 3cm lateral away from the anus and 4cm in depth.
136
Level Charge (µC)
1 0.05
2 0.065
3 0.078
4 0.096
5 0.117
6 0.155
7 0.181
8 0.25
9 0.315
10 0.385
11 0.48
12 0.606
13 0.84
14 1.077
15 1.17
16 1.5
17 1.8
18 2.107
19 2.37
20 2.46
Table 9-1. NuStim delivered charge at each level.
ACTIVATION
Device was activated 7 days after the implantation. The low threshold was measured at level 10. The target
level was first determined at level 15, however, patient felt no difference in sensation from level 16 to level
20. The target level then was set at level 20 to maximize muscle contraction. Stimulation at 20 pps was
associated with a sensation of rapid tapping, while 30 pps at the same maximal intensity was reported to be
a smooth “numb” sensation. The other parameters were set at default level (ramp 0.5 sec, hold 4 secs, off
5 secs, frequency 20 pps). The patient volunteered that he would like to perform 60 minutes’ exercise per
day, probably in two 30-minute sessions.
137
1
ST
VISIT
The 1
st
visit was 2 weeks after the treatment. The average value of the 1-hour pad testing was 49.5g. The
patient reported increase stool frequency from once a day to 2-3 times a day. Also, the patient reported
uncomfortable needle pain sensation at site of stimulation if he moved during exercise. The frequency was
increased from 20pps to 50 pps, then patient reported ‘much better’ with decrease of pain sensation. The
patient had history of back pain, but it worsened after he began daily NuStim treatment.
PAD TESTING
As of the date of the thesis submission, the patient has received total 4 times of total 16 sessions of 1-hour
pad testing. All recorded data are shown in the table 9-2.
Visit Test 1
(g)
Test 2 (g) Test 3 (g) Test 4 (g) Average (g)
Baseline 1
st
24.5 38.7 49.8 53.3 41.6
2
nd
20.5 33.7 18.6 28.6 25.4
Follow-ups 3
rd
53.4 95.5 26.7 21.6 49.3
4
th
53.5 59.2 11.4 29.1 38.2
5
th
...
Table 9-2. 1-hour pad testing data from the 1st patient when the study is in progress.
CESSATION OF EXERCISE
The patient reported severe back pain and was referred to surgery to address the pain. Patient has history of
back pain prior to participation of the study. The diagnosis for back pain was lumbar disc herniation and
sacroiliac arthritis. The treatment was nucleoplasty using RF ablation in lumbar level 4 and 5. The NuStim
treatment was stopped temporarily for 3 weeks due to the arrangement of surgery and recovery.
138
INTERPRETATION
The insertion tool proved useful to identify a suitable implantation site for the NuStim. The functionality
of the NuStim system was confirmed from the 1
st
patient. A wide range of stimuli could be delivered to
generate various stimulation patterns, which the patient found generally comfortable. It is premature to
evaluate efficacy due to number of patient and lack of follow up data. We found out the huge variance of
1-hour pad testing in this patient was because of consumption of blood pressure medication, which is a
strong diuretic. Due to irregular diuretic consumption, it is difficult to interpret the collected data including
baseline. This patient would be treated continuously for effectiveness observation, but would be excluded
from efficacy evaluation due to protocol deviation. The back pain probably was not related to the NuStim
stimulation itself, but may have been exacerbated by sitting on the cushion for one hour per day.
DISCUSSION
The primary objective of the pilot study is to validate whether the leakage from SUI could be improved
from passive electrical stimulation. If the system is validated to improve symptom in the pilot study, it
could be very helpful then to compare the efficacy among different treatment strategies, including NuStim
treatment only, NuStim with additional PFMT instruction, and NuStim plus voluntary contraction.
Compared to voluntary exercise, the electrical stimulation itself does not require patient mentally effort for
voluntary contraction. Considering about recruitment order and muscle activation pattern, the electrical
stimulation is different from voluntary exercise. With some basic physiology and PFMT instruction before
NuStim treatment, it might improve patient engagement in the study, build knowledge of SUI, identify
proper muscle control, and improve understanding of the NuStim treatment. Combining NuStim with
voluntary contraction would be beneficial for certain type of patients. For a patient who has extremely weak
pelvic floor muscle, the patient could initiate the voluntary muscle contraction then be reinforced by NuStim
electrical stimulation. It would be helpful to build brain muscle neural connection which could be important
for patient to identify pelvic floor muscle and perform voluntary contraction in activities with sudden
abdominal pressure increase (known as “the knack”). Or the NuStim could be used vice versa as a cue for
139
strong voluntary contraction. The pilot study paves the way as a fundamental first step for future evaluation
of different combined treatment strategies.
140
CHAPTER 10: CONCLUSION AND FUTURE DIRECTIONS
The NuStim system has been developed to meet all specified technology requirements of the SUI clinical
application. Significant progress has been made toward a clinical product, including design of prototype,
testing of the system in vivo and in vitro, transferring manufacture process, and ongoing clinical validation.
The primary goal of this new treatment is to find stimulation parameters that provide sufficient
improvement of the symptom. If the treatment is proven to provide similar or better outcome compared to
voluntary PFMT, this cost-effective method will be widely accepted by patients due to low invasiveness,
low cost and low post-implantation complications. The success of NuStim system provides a model of
neuromodulation systems that can be implanted via a minimally invasive approach, a feature that could be
useful to a wider application that may not be feasible for a conventional neurostimulator with long lead and
bulky case. Therefore, future development will involve expanding technology to be used in other
application that will require:
1. Multiple implants that can be controlled separately.
2. Packaging that will sustain decades of active use.
3. Continuous stimulation with a small battery and wireless charging.
There are several applications that require at least one of the above technological enhancements to address
complications from using conventional stimulator. Obstructive sleep apnea has been successfully treated
by neuromuscular stimulation of the tongue to open the airway. However, the implantation procedure is
quite invasive which requires several incisions for a stimulation lead, a sensor lead, and a pocket for pulse
generator. Similar challenges also occur in other applications using electrical stimulation in the upper
gastrointestinal tract to increase stomach mobility and control of obesity. These and other applications
require one or more stimulating electrodes to be implanted in a location subjected to body motion with high
possibility of lead dislodgement and failure.
141
APPENDIX-1: INFORMED CONSENT FORM
INFORMED CONSENT FORM
This Informed Consent form is for men and women who attend clinic Chinese rehabilitation research center
in Beijing Boai Hospital., and who we are inviting to participate in the research will lead to a treatment for
urinary stress incontinence. The title of our research project is “Pilot Clinical Trial of NuStim Therapy for
Urinary Stress Incontinence”.
Name of Principal Investigator: Limin Liao
Name of Organization: Chinese rehabilitation research center in Beijing Boai Hospital.
Name of Sponsor: Hangzhou General Stim Medical Technology Co., Ltd.
Name of Proposal and version: “Pilot Clinical Trial of NuStim Therapy for Urinary Stress Incontinence”,
This modified version is based on the current ICF submitted to IRB.
This Informed Consent Form has two parts:
• Information Sheet (to share information about the research with you)
• Certificate of Consent (for signatures if you agree to take part)
You will be given a copy of the full Informed Consent Form
Chinese Rehabilitation Research Center in Beijing Boai Hospital.
142
PART I: INFORMATION SHEET
INTRODUCTION
I am Limin Liao, working for Chinese rehabilitation research center in Beijing Boai Hospital. We are doing
research on urinary stress incontinence, which is very common in this country. I am going to give you
information and invite you to be part of this research. You do not have to decide today whether you will
participate in the research. Before you decide, you can talk to anyone you feel comfortable with about the
research.
There may be some words that you do not understand. Please ask me to stop as we go through the
information and I will take time to explain. If you have any questions later, you can ask me or the study
doctor or the staff.
PURPOSE OF THE RESEARCH
Urinary stress incontinence occurs when an activity such as coughing, sneezing or laughing causes leakage
of urine. This is a common problem as women and men get older. The treatments currently available for
this problem require surgery. We are working on a less invasive treatment that involves exercising the
muscles by electrical stimulation.
TYPE OF RESEARCH INTERVENTION
The electrical stimulation is delivered directly to the muscles by a small device (NuStim®) implanted into
the muscle through a small tube. You will control when and how strongly the muscle is exercised each day
by sitting on a special cushion that communicates with the implanted device.
PARTICIPANT SELECTION
We are expecting to enroll total 5-10 patients to attend this study. If you are a 35-80 years old female patient
who have been diagnosed with moderate urinary stress incontinence, or male patient who have urine
incontinence at least 1 year after prostatectomy with PSA < 1.ong/ml, we would like to invite you to
participate in this research. This research is not recommended to pregnant women or patients on
anticoagulant therapy or if you have taken aspirin within the past three days. After you agree this study, the
143
doctor will determine whether you are an appropriate subject for this study based on clinical test results and
inclusion/exclusion criteria.
VOLUNTARY PARTICIPATION
Your participation in this research is entirely voluntary. It is your choice whether to participate or not.
Whether you choose to participate or not, all the services you receive at this clinic will continue and nothing
will change. You may change your mind later and stop participating even if you agreed earlier.
DESCRIPTION OF THE PROCEDURES AND PROTOCOL
During the study, you will make several visits to the clinic.
After agreement to participate in this research, you will be provided several tests to ensure you are an
appropriate participant of this study. All enrolled participants must satisfy the following inclusion/exclusion
criteria:
INCLUSION CRITERIA
1. Female patient age 35-70 who is diagnosed with mild or moderate urinary stress incontinence for
more than 6 months; and male patients with stress urinary incontinence at least 1 year after
prostatectomy, PSA <1.0ng/ml.
2. Patient qualified for NuStim treatment after diagnosis.
3. Patient want to participate in the clinical trial and the patient or the guardian is willing to sign the
consent form.
4. Patient who is cooperative, willing to communicate with the researchers, and follow the trail
requirement.
EXCLUSION CRITERIA (SUBJECTS WHO MEET ANY FOLLOWING CONDITIONS WILL BE
EXCLUDE FROM THE STUDY)
1. History of spinal cord injury or/and abnormal spinal cord
2. Overactive bladder
3. Urinary tract and related infection
144
4. History of surgery treatments for stress incontinence and still in recovery (prostatectomy is not
included)
5. Narrow urethra after prostatectomy (Residual bladder urine >100ml)
6. Intrinsic urethra sphincter deficiency (Type 3 stress urinary incontinence)
7. History of neurodegenerative and neuromuscular diseases
8. History of malignant tumor in pelvic organ or ovary.
9. Severe perineum trauma from congenital perineum malformation
10. History of fecal incontinence surgery, bladder surgery, fistula surgery, and other similar surgery
that can be clearly related to the cause of urinary incontinence
11. Coagulation disorders
12. Immune system disorders
13. Female patients with plan of pregnancy, pregnancy and lactation in the trial period
14. Epilepsy
15. Heart disease
16. Acute infectious disease
17. Fever
18. Abnormal blood pressure
19. Dementia, brain atrophy, cognitive disorders, cerebral vascular disease and sequelae
20. Acute local or systemic bacterial infection without proper control
21. Ill or dying patient who fail to cooperate and evaluate the efficacy
22. Mental or/and neurological disorders who fail to cooperate with physician
23. Constipation patient
24. Female patient without vaginal delivery history
25. patients who are prone to allergies and/or scarring
26. Patients who have participated in other clinical trial in previous 3 months
27. Planned to change lifestyle of routine habit
145
28. Other condition that may not proper for this study after evaluation
If you meet all above requirements, you will be arranged for a minimally invasive implantation. During the
implantation surgery, you will lie on an exam table with your legs elevated. Small amount of local
anesthesia will be injected. One small tube (The NuStim device) will be inserted through skin into the pelvic
muscle layer. The tip of the tube will be used to apply a weak electrical stimulus that will cause the muscle
to contract. You will feel the muscle twitch but you should not feel any electrical shock. Your sensation
feedback will be used to determine the implantation site. The device will be implanted into the muscle
through the small tube (wound less than 4mm).
After at least 5 days’ recovery, we will ask you to sit on a special cushion that activates the device. The
exercise will be prescribed from tablet computer. Based on your sensation feedback, we can set up your
personal exercise program. You will be provided an operation training session about the equipment and
procedures.
You will take the cushion and a tablet computer or smartphone home your customized exercise program so
that you can use it to exercise your muscles every day.
You will record urinary diary at least one week before implantation, and perform several times of 1. Similar
collection and measurement continues throughout the treatment and follow-up period.
You will be followed in the clinic at 1-2 weeks’ intervals during the period of treatment. At each visit, we
review your exercise and your urinary diary. You will be asked several simple questions to evaluate
satisfaction and describe your symptoms. We may adjust your exercise program based on your progress.
You will be asked to drink 500ml of water, wear a diaper, and perform a series of exercise for 1-hour. We
will ask you to perform this same exercise twice in each visit.
If your improvement is not as desired, a second NuStim may be implanted in the pelvic floor muscles using
the same procedure as for the first implant.
We may take a single x-ray picture of your pelvis while you are sitting down during the visit.
146
If this treatment is going to work, you should see substantial improvement in urine leakage within a few
months. The implant is not designed to work for more than one year. If it is working and your condition
is still improving when the study period ends, you will be allowed to continue the exercise treatment. At
some point, you will stop using the implant, either because you are completely cured or you are no longer
obtaining any benefit or the implant is no longer working. The implant is designed to be left in your body
indefinitely even when not in use, but it is easily removed in a minor surgery if there is any medical reason
to do so.
DURATION
The treatment will finish within one year when any of the following situation arises:
• You have no more leakage
• You do not have any more improvement
• You do not feel any muscle contraction within exercise
• You want to quit the treatment
SIDE EFFECTS
You may feel temporary discomfort during and after implantation but this will resolve quickly. During
stimulation to exercise your muscles, you will feel contractions in the muscles around your urethral and
anal area but these should not be uncomfortable. If you feel any electrical shocks from the stimulation, we
will reduce the strength of the stimulation to prevent this. If unanticipated side effects occur, we will
prescribe necessary treatment.
There are no lasting side effects from X-ray. The x-ray exposure is similar to a single chest x-ray.
RISKS
The needle and tube used to implant the stimulator may cause bleeding, nerve damage or infection. This is
unlikely and can be treated. The implanted stimulator may cause discomfort or infection. If this occurs, it
will be removed in a minor operation.
147
BENEFITS
We hope that the muscle exercise produced by this system will strengthen the muscles and reduce the
amount of urine leakage that you are now having, but there is no guarantee. This is a research study to
determine what exercise will be most effective. What we learn from your participation is likely to help
other patients with this common problem.
ALTERNATIVE TREATMENTS
There are several alternative treatments for urine leakage. You may choose not to participate this study
which will have no side effects for you to access other options.
Non-surgical Treatments/Options:
1. PELVIC FLOOR MUSCLE TRAINING, PFMT
This method is convenient, effective, suitable for all types of stress urinary incontinence. In long term
studies, patients who achieved benefits from complete PFMT may pertain the muscle function and
symptoms improvement for a long period. There are no uniform training methods. The idea is to perform
the exercise consistently to acquire strong pelvic floor muscle.
2. LOSE WEIGHT
Obesity is a clear and relevant factor for female stress urinary incontinence. Weight loss can help prevent
the occurrence of stress urinary incontinence. For obese SUI women, if weight loss 5% to 10%, the
frequency of urine leakage will reduce more than 50%.
3. QUIT SMOKING
The association between smoke and urine leakage is unclear. There is an evidence to show smoking
increases the risk of urine leakage, but there is no evidence to show cessation in smoking relieves symptoms
of urine leakage.
4. CHANGE DIET
There is no clear evidence between development of urine leakage and consumption of water, caffeine, or
alcohol, but changing diet help to relieve the symptoms.
148
5. VAGINAL CONE TRAINING
It requires a cone shape weight (20g or 40g) into female vagina. In order to avoid the cone slippery, pelvic
floor muscle contracts to hold the cone. Efficacy of this method is unclear. Some scholars believe that it
has similar outcome as the pelvic floor muscle training. This treatment generally has poor adherence and
no efficacy for severe condition. Side effects include: abdominal pain, vaginal infection and vaginal
bleeding.
6. ELECTRICAL STIMULATION
This electrical stimulation relies on surface electrodes temporary placement in vagina for stimulation. It
uses similar theory to repeatedly stimulate the pelvic floor muscles for increase of muscle contraction. It
has side effects like, vaginal infections, bleeding, perineal discomfort and rash, which are not well accepted
by patients.
7. MAGNETIC STIMULATION
The principle of this treatment is similar to electrical stimulation. The difference is this treatment relies on
external magnetic field stimulation. It is found to be effective for a short period of time, but further study
is needed to prove the efficacy and safety. It has unknown side effects.
8. MEDICINE
The main idea of using medication is to increase the urethral closure pressure and improve urethral closure.
Currently common used medications include, selective α1-adrenergic receptor agonists, imipramine, β-
adrenergic receptor antagonist, β-adrenal receptor agonists, and estrogen. They all have different side
effects and cannot treat the urine leakage completely.
SURGICAL TREATMENTS
Surgical treatments are only recommended for the following patients:
1. Patients who have poor improvement from conservative treatments.
2. Patients with severe condition which affects the quality of life.
3. Patients who require high quality of life.
149
4. Patients who have pelvic organ prolapse and require pelvic floor reconstruction.
AVAILABLE OPTIONS:
1. Tension-free vaginal tape, TVT
2. Burch Colposuspension
3. Vaginal Sling Procedure
4. Marshall-Marchetti-Krantz, MMK
5. Needle sling
6. Filling injection
7. Artificial urethral sphincter
8. Anterior &posterior vaginal wall repair
REIMBURSEMENTS
All equipment needed for your treatment will be provided free of charge. You will be expected to return
the cushion and tablet when your treatment is complete or you decide to stop participating in the study.
You will not be charged for any loss or damage that may occur. You will not be charged for any
hospitalization, clinical testing and surgeries that relate to the study. You will be reimbursed for all
reasonable travel expenses to the clinic. You will not be given any other money or gifts to take part in this
research. Withdrawal from the study is allowed at any time.
RIGHTS
Your participation in this study is voluntarily. You have plenty of time to decide whether you want to join
this study. If you decide to participate this study, you will receive the copy of relevant documents. You
have the right to stop participating in the research at any time that you wish without losing any of your
rights as a patient here. Your treatment at this clinic will not be affected in any way. If you have any question
about this study, you can always ask for relevant information and ask questions to your doctors. If there is
an occurrence of any adverse events related to the treatment, you will receive free treatment to the adverse
events, an alternative treatment for your urine leakage, and compensation. If there are any updates about
the study, we will inform you and ask for your agreement.
150
CONFIDENTIALITY
Your participation in this research project will be kept confidential. Personal information about you that
will be collected during the research will be put away and no-one but the researchers will be able to see it.
Information about your experiences and results will be identified by a code number instead of your name.
Only the researchers will know what your number is and we will lock that information up. Your identifying
information will not be shared with or given to anyone except the doctor and other medical personnel who
examine and treat you. After the study, we may disclose the study without your personal information in the
following conditions: research, publication, teaching, and government inspection.
RIGHT TO REFUSE OR WITHDRAW
You do not have to take part in this research if you do not wish to do so. Refusing to participate will not
affect your treatment at this clinic in any way. You will still have all the benefits that you would otherwise
have at this clinic. You have the right to stop participating in the research at any time that you wish without
losing any of your rights as a patient here. Your treatment at this clinic will not be affected in any way.
WHO TO CONTACT
If you have any questions you may ask them now or later, even after the study has started. If you wish to
ask questions later, you may contact any of the following: 010- 87589667.
This proposal has been reviewed and approved by IRB in Chinese rehabilitation research center in Beijing
Boai Hospital, which is a committee whose task it is to make sure that research participants are protected
from harm. If you wish to find about more about the IRB, contact (010-87589667).
You can ask me any more questions about any part of the research study, if you wish to. Do you have
any questions?
151
PART II: CERTIFICATE OF CONSENT
I have read the foregoing information, or it has been read to me. I have had the opportunity to ask questions
about it and any questions that I have asked has been addressed properly. I consent voluntarily to participate
in this research.
Print Name of Participant__________________
Signature of Participant ___________________
Date ___________________________
Day/month/year
Statement by the researcher/person taking consent
I have accurately read out the information sheet to the potential participant, and to the best of my ability
made sure that the participant understands the experimental procedures and his/her right to withdraw at any
time.
I confirm that the participant was given an opportunity to ask questions about the study, and all
the questions asked by the participant have been answered correctly and to the best of my ability. I confirm
that the individual has not been coerced into giving consent, and the consent has been given freely and
voluntarily.
A copy of this ICF has been provided to the participant.
Print Name of Researcher/person taking the consent________________________
Signature of Researcher /person taking the consent__________________________
Date ___________________________
Day/month/year
152
APPENDIX-2: CASE REPORT FORM
The instructions for CRF
1. All recorded data have the original data record can be inquired (including medical records, the message
form doctors or nurses, laboratory inspection reports, screening and selected records, etc.)
2. Fill in with a pen; fill in the CRF correctly in block letters. When answering the check box, use "×" to
indicate that it is selected.
3. The subject number consists of 2 digits. For example, the second participant has patient number 02.
4. For the same answers of different questions, please do not use "ibid".
5. If the answer is "0", please do not leave a blank, write "0".
6. If the answer is "Unknown", please write "NK".
7. If the required inspection is not done, please write "ND" and indicate the reason.
8. If the question is not suitable, please use "NA".
9. If you fill in incorrectly, please follow the following methods to correct:
② cross the wrong answer with a single line
② Write the correct answer above or next to the crossed part
③ indicate the initials of who modify and the date of modification
Do not use rubber or correction fluid / tape to correct.
10. The CRF for each selected patient should be completed on time, including those who have withdrawn
(complete the End of Study section).
11. Each CRF should be signed by the principal researcher and the CRF should be checked by a clinical
research inspector before signing.
12. For abbreviation of patients’ name, take the first three Chinese phonetic alphabet of the name, such as
abbreviation of
Wang Li : W L I; Wang Dali: W D L; Ouyang Meli: O M L.
13. Accompanied treatment and adverse events must be record in the corresponding sites.
153
14. Any serious adverse event must be reported to the Ethics Committee, the CFDA, the Principal
Researcher and the Sponsor, and the Food and Drug Administration at the location of the Research
Center within 24 hours.
154
Visit date :201|__|year|__|__|month|__|__|day
Date when inform consent signed :201|__|year|__|__|month|__|__|day
Demographic Information
Date of Birth :19|__|__| year|__|__| month|__|__|
day
Gender : Male Female
Height :|__|__|__| cm
Weight :|__|__|__|. |__| kg
Medical History of stress urinary incontinence
Date of diagnosis of stress urinary incontinence :|__|__|__|__|year|__|__|month|__|__|day
For male patients, please continue to fill in the following :
The date of prostatectomy : |__|__|__|__|year|__|__|month|__|__|day
PSA (Prostate Specific Antigen) Examination Date :201|__| year |__|__| month |__|__| day Not Checked
Test Result : ng/ml
Note: If patient is continuing in treatment, please fill in "NA" for end date
Previous treatment for stress urinary incontinence
Name of
medicine
Start date
Is patient still taking the
medicine
End date
|__|__|__|__|y|__|__|m|__|__|d Y N |__|__|__|__|y|__|__|m|__|__|d
|__|__|__|__|y|__|__|m|__|__|d Y N |__|__|__|__|y|__|__|m|__|__|d
|__|__|__|__|y|__|__|m|__|__|d Y N |__|__|__|__|y|__|__|m|__|__|d
|__|__|__|__|y|__|__|m|__|__|d Y N |__|__|__|__|y|__|__|m|__|__|d
|__|__|__|__|y|__|__|m|__|__|d Y N |__|__|__|__|y|__|__|m|__|__|d
|__|__|__|__|y|__|__|m|__|__|d Y N |__|__|__|__|y|__|__|m|__|__|d
155
Note: If the patient is taking anticholinergic, cholinergic drugs, patient can join the group of the trial 2 weeks after stop taking
medicine. Patient cannot take the medicine during the trial.
Other previous medical history / concomitant diseases
Whether the patient has other diseases beside the stress urinary incontinence :
No Yes Please fill in the following form
Name of disease /
surgery
Start date If still exist End date
|__|__|__|__|y|__|__|m|__|__|d Y N |__|__|__|__|y|__|__|m|__|__|d
|__|__|__|__|y|__|__|m|__|__|d Y N |__|__|__|__|y|__|__|m|__|__|d
|__|__|__|__|y|__|__|m|__|__|d Y N |__|__|__|__|y|__|__|m|__|__|d
|__|__|__|__|y|__|__|m|__|__|d Y N |__|__|__|__|y|__|__|m|__|__|d
|__|__|__|__|y|__|__|m|__|__|d Y N |__|__|__|__|y|__|__|m|__|__|d
|__|__|__|__|y|__|__|m|__|__|d Y N |__|__|__|__|y|__|__|m|__|__|d
Note:
1, if the above-mentioned diseases are still drug treatment, please fill in the medication on the 75-78 page
'combined medication form'.
2, Please check whether the disease is in the exclusion criteria listed above. If any, the subject cannot be
selected.
3, If the disease still exists, please fill in "NA" for the end date.
156
Physical Examination
System Normal Abnormal
Not
Checked
Description
1.Skin
2.Face
3.Lymph node
4.Cardiovascular
5.Lung
6.abdomen
7.Spine / limbs
8.Genitourinary
9.Neuron
10.Others
Vital Signs
Pulse :|__|__|__| times/min Breath :|__|__|__|times/min
Blood Pressure :|__|__|__|/|__|__|__| mmHg Body Temperature :|__|__|. |__|℃
Activate NuStim Neuron Stimulator
Whether NuStim neuron stimulator is activated or not?
□ Yes ,Date of activete :201|__|year|__|__|month|__|__|day
□ No ,Please note the reason :
157
Patient NuStim System Operation Training
Patient receives training : □Yes □ No ,please explain :
NuStim New Parameters Setup
Please record compliance as shown on the Software
Achieved Exercise Days
Days
Total Exercise duration
Minutes
Average intensity
CU
How does the patient feel about the stimulation? (Tick one box)
If the stimulation is too weak or too strong, the patient may need
parameters adjustment in NEW PRESCRIBTION section.
Otherwise, please put down you signature below and keep the record.
New Prescription (Please circle the new prescription value below)
• Too strong
• Just right
• Too weak
158
1-hour Urine Pad Test
Date :201|__|year|__|__|month|__|__|day Starting time :|__|__|:|__|__| Am/Pm
Clean pad weight before starting: ________ g Is the patient’s bladder empty? : Yes No
Last time eating before pad test start time is over 3 hours : Yes No
Steps :
1 )Drink 500ml of sodium-free liquid in 15 minutes ; Yes No
2) Walking for half an hour, including climbing up and down a staircase ; Yes No
3) Do the following experiment in the remaining time :
① Stand up 10 times from a seat Yes No
② Hard cough 10 times Yes No
③ In situ running 1 minute Yes No
④ Bend down and pick up the small objects on the ground 5 times Yes No
⑤ Wash hands with running water for 1 minute Yes No
Whether the urine pad wet saturated need to remove the urine pad, the replacement of another new pad?
Yes No
Does the patient urinate during the examination? Yes No
If you select "Yes", please record the following, and arrange the patient to do another test if necessary. The
relevant information is recorded on the _______ page of this study.
The 1-hour urine pad test duration :|__|__| min ;Amount of urine :________ ml
Stop time of test :|__|__|:|__|__| am/ pm
Weigh the urine pad after use :________ g ; Record the amount of patient’s urine : ________ ml
Measure urine leakage by weighing method :________g 。
Note: leakage of urine equal to the weight of the pad at the end of test minus the weight of a clean pad, 1 gram is
equivalent to 1ml of urine
159
Life Satisfaction Questionnaire
International Consultation on Incontinence Modular Questionnaire Short Form (ICI-Q-SF )
1 .Date of Birth : |__|__|__|__| year |__|__| month |__|__| day
2 .Gender : Female Male
3 .Score of the number of leakage of urine : _ _ points
4. Urine leakage score :_ _points
5. The influence of urine leakage on daily life :_ _points
ICI-Q-SF score (Sum scores 3+4+5):_ _points
6. When does urine leak?
never – urine does not leak
leaks before you can get to the toilet
leaks when you cough or sneeze
leaks when you are asleep
leaks when you are physically active/exercising
leaks when you have finished urinating and are dressed
leaks for no obvious reason
leaks all the time
Abdominal leak point pressure measurement
Date :201|__| year |__|__| month |__|__| day Not checked
Result VLPP : cmH 2O ; CLPP : cmH 2O ;
Urethral Pressure Profile (UPP )
Date :201|__| year |__|__| month |__|__| day Not Checked
Result MUP : cmH 2O ; MUCP : cmH 2O ;
160
The Patient Perception of Bladder Condition (PPBC Test)
Which of the following statements describes your bladder condition best now? Please mark ”X” in one
box only.
1. My bladder condition does not cause me any problems at all.
2. My bladder condition causes me some very minor problems.
3. My bladder condition causes me some minor problems.
4. My bladder condition causes me (some) moderate problems.
5. My bladder condition causes me severe problems.
6. My bladder condition causes me many severe problems.
Urine Pregnancy Test
Date :201|__| year |__|__| month |__|__| day Not Checked Not applicable reason :
;
Result : Negative ; Positive (If its positive, then subject is not qualified for this trial )
Laboratory Test
Please fill in every laboratory test result in following form and check if the measurement is in the normal
scope. If its abnormal, please find out if it has clinical relevance to treatment.
Date :201|__| year |__|__| month |__|__| day Not Checked
Blood test
Item Result Unit Normal
Abnormal
Not
Checked
No
clinica
l
propos
e
Clinic
al
propos
e
Descriptio
n
白细胞WBC ×10
9
/L
161
红细胞RBC ×10
12
/L
血红蛋白HGB g/L
血小板PLT ×10
9
/L
Date :201|__| year |__|__| month |__|__| day Not Checked
Urine
Test
红细胞BLD
--
白细胞LEU
--
尿蛋白PRO
--
尿糖GLU
--
Date :201|__| 年 |__|__| 月 |__|__| 日 Not Checked
Comprehen-
sive
Metabolic
Panel
谷丙转氨 酶 ALT U/L
谷草转氨 酶 AST U/L
肌酐 Cr umol/L
尿素氮 BUN
mmol/
L
总胆红素 TBIL umol/L
碱性磷酸 酶 ALP U/L
γ- 谷氨 酰转移 酶
r-GT
U/L
Date :201|__| year |__|__| month |__|__| day Not Checked
Electrolyte
钾 K
-
mmol/
L
钠 Na
+
mmol/
L
162
氯 C1
-
mmol/
L
钙 Ca
2+
mmol/
L
Date :201|__| year |__|__| month |__|__| day Not Checked
Coagulation
Function
凝血酶原 时间
(PT)
sec 秒
活化部分 凝血活
酶时间
(APTT )
sec 秒
纤维蛋白 原
(FIB )
g/L
凝血酶时 间
(TT )
sec 秒
Note: If the result is ‘abnormal’ and has ‘relevant to treatment’, please check whether there is a relevant
medical history.
If yes, please check if the medical history form on page 3 is filled.
12 Leads Electrocardiogram
Date :201|__| year |__|__| month |__|__| day Not Checked
Result : Normal Abnormal, not related Normal ,related to treatment
If its abnormal ,please briefly describe :
163
Enrolment Criteria
1. Male patients with stress urinary incontinence who had a PSA < 1.0 ng / ml
for at least 1 year after prostatectomy; Female patients who is older than 35
years old and had mild and moderate stress urinary incontinence clinical
conditions over than 6 months.
Yes No
2. Patients diagnosed by physician as suitable for NuStim training Yes No
3. Willing to participate in the trial, patient himself/herself or legal guardian
signed the informed consent form;
Yes No
4. Good communication with researchers and compliance with the test
requirements.
Yes No
If any of the answer criteria is "No", the patient cannot participate in this study.
Exclusion criteria
1. History of spinal cord injury or/and abnormal spinal cord Yes No
2. Overactive bladder Yes No
3. Urinary tract and related infection Yes No
4. History of surgery treatments for stress incontinence and still in recovery
(prostatectomy is not included) Yes No
5. Narrow urethra after prostatectomy (Residual bladder urine >100ml) Yes No
6. Intrinsic urethra sphincter deficiency (Type 3 stress urinary
incontinence) Yes No
7. History of neurodegenerative and neuromuscular diseases Yes No
8. History of malignant tumor in pelvic organ or ovary. Yes No
9. Severe perineum trauma from congenital perineum malformation Yes No
10. History of fecal incontinence surgery, bladder surgery, fistula surgery,
and other similar surgery that can be clearly related to the cause of urinary
incontinence Yes No
11. Coagulation disorders Yes No
12. Immune system disorders Yes No
13. Female patients with plan of pregnancy, pregnancy and lactation in the
trial period Yes No
14. Epilepsy Yes No
15. Heart disease Yes No
16. Acute infectious disease Yes No
164
17. Fever Yes No
18. Abnormal blood pressure Yes No
19. Dementia, brain atrophy, cognitive disorders, cerebral vascular disease
and sequelae Yes No
20. Acute local or systemic bacterial infection without proper control Yes No
21. Ill or dying patient who fail to cooperate and evaluate the efficacy Yes No
22. Mental or/and neurological disorders who fail to cooperate with
physician Yes No
23. Constipation patient Yes No
24. Female patient without vaginal delivery history Yes No
25. patients who are prone to allergies and/or scarring Yes No
26. Patients who have participated in other clinical trial in previous 3 months Yes No
27. Planned to change lifestyle of routine habit Yes No
28. Other condition that may not proper for this study after evaluation Yes No
Exclusion Criteria Any of the questions answered Yes, the patient was unable to participate in the study
Signature :____________________ Date :____________________
165
Issue and Collect Diary card
In this visit, whether the completed diary card has been collected ?
□ Yes □ No ,Please explain :___________________________
Average urine times three days before visit :________ times ;
Average amount of urine three days before visit :________ ml ;
Average amount of urine pad used three days before visit :________ pads ;
Average amount of urine leakage three days before visit : __________ ml ;
Average score of urgency of urine three days before visit :_________ points ;
In the visit, whether to issue a new diary card to the subjects ?
□ Yes □ No ,Please explain :___________________________
Combined Medication
Since the last visit to this visit, whether the subjects have a new combination of medication occurred
or the original merger drug is change?
Yes ,please fill in “combine medication form” in page 75-78 No
Note: Please check for all indications or medication purposes of the new combination and if adverse
events should be recorded. If yes, please check if the Adverse Events Form has been filled
Adverse events
Has the subject ever experienced any adverse events since the last visit?
□ Yes Please fill in the “adverse events form” in page 79-83. □ No
Note: Please check if there is any combine medicine using when treating the adverse events. If there is, pleae check
the “combine medicine form” on page 75-78 is filled in or not.
166
Summary of research
Trial complete/ending date 201|__| year |__|__| month |__|__| day
Does the patient complete the trial? Yes No ,Please write done the reason for termination.
Reason for termination of the trial :(choose only one )
1 Lack of effective
2 adverse events (Please complete the adverse event form )
3 patient ask to quit (non- adverse event )
4 Violation of rules of trial ,please describe
5 Lose visit
6 Pregnant
7 Others ,Please describe:
167
Principal Researcher Statement
I declare that this case report has been inspected and that all records in the case report form are complete and
true.
Signature of principle researcher :_____________________
Date :|__|__|__|__| year |__|__| month |__|__| day
168
Combined Medicine Form
If there is combined medicine on this page : None Yes ,Please fill in the following form.
Note :” Continue using” means the patient is still taking the medicine after the end of research.
Signature :______________________ Date :_____________________
169
Adverse Event Form
Are there any adverse events? :
No Yes ,Please follow the instruction of adverse event form to fill in the following form.
Instruction :
1. If there are any serious adverse events, please also fill out the Severe Adverse Event Report Form ;
2. Severity : 1 Mild, 2 Medium ,3 Severe ;
3. Correlation with the medicine :1 related, 2 probably related, 3 probably related, 4 probably unrelated, 5 unrelated ;
4. Take action: 0 None, 1 Exit Study, 2 Set new usage parameters, 3 Implant second implant stimulator, 4 Treatment, 5 Others ;
5. Result :1 Disappear, 2 Relief, 3 Continue, 4 Sequelae, 5 Death, 6 Unknown
Signature :______________________ Date :___________________
170
APPENDIX-3: ANIMAL STUDY REPORT
NUSTIM
®
IMPLANT
ANIMAL STUDY
Xuechen Huang
Experiment performed in Beijing Capital University, from 12.9.2016 to 3.16.2017
Tissue received in Beijing Capital University, Date: 3.16.2016
Tissue processed at USC Histology Laboratory, Date: 4.18.2016
Tissue examined USC Liver Laboratory, Date: 5.5.2016
Tissue reported date: 5.6.2016
Date: 7.20.2016
Reformatted: 7.72017
Experiment participants:
Limin Liao. PhD, MD, Gerald E. Loeb. MD, Frances J. Richmond. PhD. Xuechen Huang, Xing Li,
Deng Han, Tianji Lu, Zhaoxia Wang, Hao Li, Linxiao Yan
171
The animal experiments conformed to the Guide for the Care and Use of Laboratory Animals and
were approved by the Institutional Animal Care and Use Committees at the Capital Medical
University, Beijing, China.
OBJECTIVES
The objective of the NuStim system is to provide comfortable and safe exercise of the pelvic floor
muscles which is effective in reducing urinary stress incontinence. An animal study is one of the
steps necessary to accomplish the objective. The primary objective of an animal study is to examine
the safety and device performance in the environment of a body before the device can be used
clinically. The purpose of the animal study is
• To identify the feasibility of the NuStim implantation procedure
• To evaluate the performance of the NuStim system via daily stimulation.
• To compare the short and long-term stability of the NuStim in tissue via threshold measurement
and histology.
• To compare the short and long-term tissue response between an active and non-active NuStim
METHODS
SURGERY
Experiments were carried out on three dogs (9.4 to 11.2 kg) sedated with 2-3 ml Xylazine and
anesthetized with sodium pentobarbital (2.5%, 1 ml/kg intramuscular). One dog was used for a one-
month study, and two dogs were used for three-month studies. A total of five active devices were
implanted into quadriceps femoris or triceps brachii and activated for up to two weeks. Four non-
activated devices were placed in the comparable contralateral muscles. Insertions were directed
perpendicularly to the skin in order to orient the devices transversely to the muscle fibers, but there
was no attempt to correct for the substantial pennation angle of these muscles. Thresholds were
measured during and after active device implants to assure proper intramuscular location. Non-
active devices were implanted blindly.
Two implantation strategies were tested in the experiment:
• In the first strategy, the conventional off-the-shelf disposable hypodermic EMG needle
electrode was inserted at an angle on the skin starting with lowest stimulation strength. As the
needle pushed in, the lowest threshold contraction was determined by detecting the first muscle
172
twitch. A skin incision was made at a different location as an entry site for the NuStim insertion
tool. The NuStim insertion tool was oriented approximately perpendicular to the muscle, passed
through the skin, and aimed so that the end of the sheath was at approximately the target
determined above. The insertion tool was advanced in 1cm steps until the depth of the target
had a similar stimulation threshold.
• The second strategy used the needle electrode from the NuStim insertion tool to locate the ideal
target. The needle was held in position once the stimulation threshold was acquired. The skin
incision was made directly at the needle entry site. The dilator and and sheath were passed over
the needle and assembled to complete the insertion tool so that the final assembled NuStim
insertion tool was positioned at approximately the same location and depth as the first strategy.
Both strategies rely on the same NuStim release strategy. During the implantation, the needles were
connected to the disposable customized stimulator which delivered test pulses at 2 pulses per
second with an intensity that ranges from about 3 to 13 arbitrary clinical units as marked on the
stimulator. An alligator clip was used as return electrode on the skin. After location of the ideal
implantation position, the NuStim insertion tool was pushed 1 cm further past the target. At this
depth, the end of the sheath was at the target depth. The needle and the dilator were retracted
without moving the sheath. The NuStim holder was attached to the back end of the sheath with the
cathode of NuStim oriented toward the tissue. The NuStim was pushed by the needle through the
lumen of the sheath and stopped at the end of sheath when the operator felt resistance. When the
needle on the stimulator contacted the back of the NuStim implant, the stimulation pulse passed
through the implant to its cathodal stimulating electrode. The elicited muscle twitch confirmed the
location at a strength within three clinical units of the lowest threshold obtained previously. The
sheath was then pulled out to release the NuStim at this location.
THRESHOLD DETERMINATION
The animals recovered for at least 5 days after surgery. The threshold measurement was determined
for each active device at least once a week. Animals were sedated with Xylazine to eliminate
voluntary muscle contraction, and positioned on the external cushion to produce threshold twitches
of muscle. The threshold was obtained by setting the level to the lowest setting that still produced
a palpable muscle twitch. The setting of the stimulus pulse that was delivered 2 at pulses per second
was adjusted on the app with an up and down arrow in a single set of up to 20 clinical units.
STIMULATION
The active devices were activated to simulate daily exercise of the muscles. Before stimulation,
threshold measurement was performed to determine the threshold level and the target stimulation
173
strength based on different recruitment pattern in each animal. The target level is generally about
3 to 6 levels higher than the threshold level. The stimulation settings (ramp up from threshold, hold
at selected target intensity, ramp down and off period between contraction) were set at default at
0.5 second ramp, 4 seconds holding, and 5 seconds off for a complete stimulation cycle. The
stimulation frequency was set at 6 pulses per second to eliminate over-contraction. Stimulus trains
were applied for 30 mins a day, five days a week for a period up to two weeks.
TERMINATION
At the end of study, Dog 1 had been implanted for 98 days, Dog 2 for 27 days and Dog 3 for 72
days. Lateral and anterior x-rays were given to confirm the orientation and location of the implants.
External cushion was used to locate the active device. All three were euthanized with intravenous
potassium chloride and the implant sites were examined for gross and histological pathology.
Tissue was fixed in 10% formalin for 7 to 10 days before removing the device and blocking for
paraffin embedding and sectioning. Sections were obtained from tissue near the middle of the
cylindrical device (3.4 mm diameter x 10.0 mm long) and were oriented perpendicularly to its long
axis. The tissue was stained with hematoxylin and eosin (H&E) and examined under a light
microscope. All implants were removed from the animals and examined under microscope for any
visible damage or changes in surface characteristics.
RESULTS
SYSTEM OPERATION
After simple training, the operators could use the surgery tool to identify low threshold location.
The insertion tool allowed confirmation of implantation location before release of NuStim. If the
location was not ideal, the operator could use the surgery tool to retract the NuStim and found the
low threshold location using the same implantation procedure. In the exercise training session, the
operator could use the software to set parameters, test exercise and prescribe training. The external
RF-Cushion could generate sufficient energy to power the implant from different angles and
distance at determined output level. When properly positioned over the RF-Cushion, each active
implanted device was activated separately to cause muscle contractions that could be palpated on
the skin and observed as cyclical limb motion. When the repetition rate was above 20 pps, the
muscle contraction was smooth and sufficient to fully extend the limb. Animal generally ignored
the stimulation pattern bur required petting to stay on the RF-Cushion during the exercise.
174
STIMULATION
Thresholds for implanted devices were all in the range of 6 to 9 clinical units when initially
activated 10 days after implantation. Fig. 10a shows trends in the thresholds over time normalized
to the value on the activation day and compared to the thresholds obtained during implantation on
day zero. The target stimulus strength needed to generate apparent maximal twitch was different
for each device. The active device in Dog 1 vastus muscle only required 3 to 4 clinical units above
threshold while the device in Dog 2 produced more gradual increases in recruitment over 11 steps.
One of the 5 active devices ceased to produce stimulation pulses 5 days after implantation.
Electrical function testing of this device removed at necropsy showed that it no longer resonated at
the tuned frequency. The most likely cause would be a cold solder joint where the copper wire is
attached to the ceramic PCB. This failure mode was subsequently mitigated with a manufacturing
change to pretin the copper wire before soldering to the PCB.
A significant change of threshold after implantation may indicate migration through or damage to
surrounding tissue. Proper healing after implantation is necessary for implant stabilization. The
stability of the threshold values over an extended period of electrically induced muscle contractions
suggests that the implants did not migrate or damage the muscle. The connective tissue capsule that
starts to form around cylindrical implants in the first few days after implantation gradually becomes
less reactive and better integrated into the endomysial connective tissue that surrounds and supports
all muscle fibers. The absolute value of the threshold was expected to differ between implantations
because it is quite sensitive to the distance between the cathodal electrode and the nearest motor
axons. Accordingly, the rate at which twitch force increases with stimulus strength depends on the
distribution of intramuscular nerve branches with respect to the implant, which is likely to vary
considerably depending on the neuromuscular architecture of the muscle and the location of the
stimulator. Healing and local reorganization of tissues after implantation (as well as uncertainties
in palpating twitch threshold in an awake animal) are the likely cause of the small shifts in electrical
thresholds noted after implantation. Similar encapsulation and recruitment patterns were described
previously for BION stimulators. The physician can palpate contractions of the pelvic floor muscles,
but we expect that it will be simpler and perhaps more accurate to rely on the patient’s perception
of the electrically induced muscles contractions.
HISTOLOGY
Each implanted muscle was removed at necropsy with the device left in the tissue block. All active
and passive devices were well-integrated with surrounding tissues, which made them somewhat
difficult to find. There were no gross pathological signs of reaction or infection. Tissues from which
175
active and non-active devices were removed after fixation are shown in Figs. 11b to g. The capsule
layer was peeled away from the surrounding tissue in some places, presumably because of forces
on the tissues during device extraction, but remained in position in most samples. The thickness of
the fibrotic capsule around the three-month implants (Fig 11.e) was about half as thick as that after
one month (Fig 11.d). Close to the capsule, there were some small muscle fibers with central nuclei
suggestive of ongoing recovery from damage during the initial implantation. Further from the
capsule, the myonuclei were spaced around the periphery of the muscle fibers in a normal pattern
for healthy muscle fibers. Around the longer-term implants, the unaffected muscle fibers were
closer to the capsule, which suggests progressive healing of local insertion trauma. Muscle fibers
immediately adjacent to the capsule tended to be cut transversely (i.e., running parallel to the long
axis of the cylindrical implant) but further away the plane of section appeared to be more oblique,
consistent with the intent to implant the device transversely to the muscle fibers. The mechanical
presence of the device may have resulted in some local reorientation of muscle fibers, particularly
those recovering from implantation damage. Some of the non-active devices were found wholly or
partially in loose connective tissue rather than within a muscle (Fig 11.f). Their surrounding
capsules tended to collapse from lack of support by fixed muscle when the implants were removed
prior to embedding and sectioning. No clinically significant histological differences were observed
between the active and passive devices at the same time points. Details report can be found in the
NuStim
®
Implant Histopathology Report.
CONCLUSION
NuStim system can operate functionally to power and control the implantable neurostimulator. The
implant can generate wide range of adjustable parameters for muscle activation without sign of
surrounding tissue damage. The threshold measurement and histopathology study prove device
compatibility, safety and stability. The insertion tool is easy to manipulate to find low threshold
implantation site. No Significant difference was found between active and non-active device.
176
APPENDIX-4: HISTOPATHOLOGY REPORT
NUSTIM
®
IMPLANT
HISTOPATHOLOGY
REPORT
Xuechen Huang
Tissue received in Beijing Capital University, Date: 3.16.2016
Tissue processed at USC Histology Laboratory, Date: 4.18.2016
Tissue examined USC Liver Laboratory, Date: 5.5.2016
Tissue reported date: 5.6.2016
Study and report supervised by: Gerald E. Loeb. MD, Frances J. Richmond. PhD, Date: 5.9.2016
Researcher: Xuechen Huang
Date: 5/12/2016
Supervisor: Gerald E. Loeb. MD
Date :5/12/2016
177
SUMMARY
Preclinical studies of NuStim implants were conducted to provide data in support of safety and
efficacy of chronic intramuscular implantation and stimulation. This report describes the histology
of the tissues surrounding the implants. Data regarding the stability of stimulus thresholds for
eliciting muscle contractions are reported elsewhere. All results are consistent with the presence
of a benign, stable foreign body, similar to the responses reported previously for similar active and
passive BION implants.
MATERIALS AND METHODS
Tissue was removed at necropsy with the device in place. All devices were well-integrated with
surrounding tissues, which made them somewhat difficult to find. There were no gross pathological
signs of reaction or infection. Tissue was fixed in 10% formalin for 7-10 days before removing the
device and blocking for paraffin embedment and sectioning. Sections were obtained from tissue
near the middle of the cylindrical device (3.5 mm diameter x 10.0 mm long) and were oriented
perpendicularly to its long axis. Three sets of sections stained with hematoxylin and eosin (H&E)
were obtained from each specimen. Two sets of sections and one set of half-blocks are being sent
to Beijing; one set will be retained in Los Angeles.
Experiments were carried out on three dogs (9.4 to 11.2 kg) sedated with 2-3 ml Xylazine and
anesthetized with sodium pentobarbital (2.5%, 1 ml/kg intramuscular). One dog was used for a one-
month study, and two dogs were used for three-month studies. A total of five active devices were
implanted into quadriceps femoris or triceps brachii and activated for up to two weeks. Four non-
activated devices were placed in the comparable contralateral muscles. Insertions were directed
perpendicularly to the skin in order to orient the devices transversely to the muscle fibers, but there
was no attempt to correct for the substantial pennation angle of these muscles. Thresholds were
measured during and after active device implants to assure proper intramuscular location. Non-
active devices were implanted blindly.
178
The code for the tissues is listed in the table:
Tissue code
Implant location
(Dog #: muscle)
Device
(Active/Non-active: ID #)
Implant duration
D2NRH 2: R vastus N: 2015112401 3 mo.
D2ALH 2: L vastus A: 2015102701 3 mo.
D2ALF 2: L triceps A: 2015092506 3 mo.
D3NRH 3: R vastus N: 2015112402 1 mo.
D3ALH 3: L vastus A: 2015092516 1 mo.
D4NRH 4: R vastus N: 2015121602 3 mo.
D4ALH 4: L vastus A: 2015122905 3 mo.
D4NRF 4: R triceps N: 2015121601 3 mo.
D4ALF 4: L triceps A: 2015122802 3 mo.
179
INTRAMUSCULAR IMPLANT ASSESSMENT OF LOCAL EFFECTS
BACKGROUND:
The typical body reaction from a biocompatible material is described and illustrated from Alves,
Metz and Render, in "Microscopic and ultrastructural pathology in medical
devices." Biocompatibility and Performance of Medical Devices (2012): 457. They note that the
implantation or injection of a biomaterial or medical device results in a physical injury to local
tissues, with attendant cellular and tissue responses. Durable smooth-surfaced materials will typical
elicit a limited macrophage and giant cell response, and will be surrounded by a well-formed fibrous
capsule.
Figure 1: Fibrosis illustrated by Alves et al. showing minimal cellular reaction surrounding metal
device implanted into skeletal muscle. Plastic tissue section stained with H&E after grinding. Alves
et al.
180
COMPARISON OF ACTIVE DEVICE IN SHORT/LONG TERM
Tissues from which active devices were removed after fixation are shown in Figure 2. The implant
space is surrounded by a thin connective tissue capsule with minimal cellular reaction. The capsule
layer is peeled away from the surrounding tissue in some places presumably because of forces on
the tissues during device extraction, but remains in position in most samples. In the short-term
implantation, the thickness of the fibrotic capsule is about twice as thick as that after longer term
implantation. Close to the capsule there are some small muscle fibers with central nuclei suggestive
of ongoing recovery from damage during the initial implantation. Further from the capsule the
myonuclei are spaced around the periphery of the muscle fiber in a normal pattern for healthy
muscle fibers. In the long-term specimens, the unaffected muscle fibers are closer to the capsule,
which suggests healing from damaged state back to healthy muscle. Muscle fibers immediately
adjacent to the capsule tended to be cut transversely (i.e. running parallel to the long axis of the
cylindrical implant) but further away the plane of section appears to be more oblique and consistent
with the intent to implant the device transversely to the muscle fibers. The mechanical presence of
the device may result in some local reorientation of muscle fibers, particularly those recovering
from implantation damage.
181
Figure 2: Tissue sections from dogs implanted with active microstimulators. Tissues after short
term implantation (1 month) are on the left whereas those after longer term implantation (3
months) are on the left. Magnification = 5X and 10X
182
COMPARISON OF NON-ACTIVE DEVICE IN SHORT/LONG TERM
Tissues with short term implantation are shown on the left and long-term implantation are on the
right of Figure 3. The implant is surrounded by a capsule with minimal cellular reaction. The
capsule layer is peeled away from the surrounding tissue due to the extraction force of the implant.
Some of the non-active devices were found wholly or partially in loose connective tissue rather
than within a muscle. Their surrounding capsules tended to collapse from lack of support by fixed
muscle when the implants were removed prior to embedding and sectioning. No significant
differences were observed between the active and passive devices at the same time points.
Figure 3: Tissue sections from dogs implanted with passive microstimulators. Tissues after short
term implantation (1 month) are on the left whereas those after longer term implantation (3
months) are on the left. Magnification = 5X and 10X
183
IMPLANTATION OF ACTIVE DEVICE IN DIFFERENT REGION
All active devices were implanted in the muscle. All capsules were well formed in thin layer. It
indicated that the implantation procedure allowed correct implantation into right position.
Figure 4: Tissue sections from dogs implanted with active microstimulators. Tissues after long
term implantation (3 month) into the quadriceps femoris is on the left whereas this implantation
into the triceps brachii is on the right. Magnification = 5X top and 10X bottom.
184
IMPLANTATION OF NON-ACTIVE DEVICE IN DIFFERENT REGION
All Non-Active devices were implanted blindly. Some devices were found only partially implanted
into the muscle. One device was found surrounded by fat tissue. Forming of fibrosis was determined
by the device location. The device located in the place with different movement was encapsulated
inside the fibrosis. The extended fibrosis penetrated surrounding tissue helped to fix the foreign
body in a position.
Figure 5: Tissue sections from dogs implanted with passive microstimulators. Magnification = 5X
185
Figure 6: Tissue sections from dogs implanted with passive microstimulators.
186
CONCLUSION
The long-term devices look even better than the short-term devices, consistent with healing of the
initial damage from the implantation. In the short-term implantation, the thickness of the fibrotic
capsule is thicker than the long-term implantation. Several of the inactive devices look somewhat
different from the active devices because they were completely or partially in non-muscle
tissue. This is because we did not do complete stimulation tests during the inactive
implantations. The stimulation testing allowed us to accurately place all of the active devices in
muscle, where their stimulation thresholds were steady over time, indicating that they stayed where
we initially implanted them. All devices elicit mild cellular response. The biocompatible durable
smoothed surface material finally encapsulated by well-formed capsule.
187
REFERENCES
[1] A. L. Baert, Imaging pelvic floor disorders: Springer Science & Business Media,
2010.
[2] K. Bø, "Pelvic floor muscle training is effective in treatment of female stress
urinary incontinence, but how does it work?," International Urogynecology
Journal, vol. 15, pp. 76-84, 2004.
[3] P. Abrams, K.-E. Andersson, L. Birder, L. Brubaker, L. Cardozo, C. Chapple, et
al., "Fourth International Consultation on Incontinence Recommendations of the
International Scientific Committee: Evaluation and treatment of urinary
incontinence, pelvic organ prolapse, and fecal incontinence," Neurourology and
urodynamics, vol. 29, pp. 213-240, 2010.
[4] S. Hunskaar, K. Burgio, A. Clark, M. Lapitan, R. Nelson, U. Sillen, et al.,
"Epidemiology of urinary (UI) and faecal (FI) incontinence and pelvic organ
prolapse (POP)," Incontinence, vol. 1, pp. 255-312, 2005.
[5] L. Zhu, J. Lang, C. Liu, S. Han, J. Huang, and X. Li, "The epidemiological study
of women with urinary incontinence and risk factors for stress urinary incontinence
in China," Menopause, vol. 16, pp. 831-836, 2009.
[6] K. S. Coyne, C. C. Sexton, D. E. Irwin, Z. S. Kopp, C. J. Kelleher, and I. Milsom,
"The impact of overactive bladder, incontinence and other lower urinary tract
symptoms on quality of life, work productivity, sexuality and emotional well‐being
in men and women: Results from the EPIC study," BJU international, vol. 101, pp.
1388-1395, 2008.
[7] J. A. ASHTON‐MILLER and J. DeLANCEY, "Functional anatomy of the female
pelvic floor," Annals of the New York Academy of Sciences, vol. 1101, pp. 266-296,
2007.
[8] K. Bø, T. Talseth, and I. Holme, "Single blind, randomised controlled trial of pelvic
floor exercises, electrical stimulation, vaginal cones, and no treatment in
management of genuine stress incontinence in women," Bmj, vol. 318, pp. 487-493,
1999.
[9] L. L. Subak, R. Wing, D. S. West, F. Franklin, E. Vittinghoff, J. M. Creasman, et
al., "Weight loss to treat urinary incontinence in overweight and obese women,"
New England Journal of Medicine, vol. 360, pp. 481-490, 2009.
[10] A. Lipp, C. Shaw, and K. Glavind, "Mechanical devices for urinary incontinence
in women," The Cochrane Library, 2011.
[11] C. Dumoulin and J. Hay-Smith, "Pelvic floor muscle training versus no treatment,
or inactive control treatments, for urinary incontinence in women," Cochrane
Database Syst Rev, vol. 1, 2010.
[12] K. Bø, "Pelvic floor muscle exercise for the treatment of stress urinary incontinence:
an exercise physiology perspective," International Urogynecology Journal and
Pelvic Floor Dysfunction, vol. 6, pp. 282-291, 1995.
[13] U. M. Peschers, D. B. Vodus ̆ ek, G. Fanger, G. N. Schaer, J. O. DeLancey, and B.
Schuessler, "Pelvic muscle activity in nulliparous volunteers," Neurourology and
urodynamics, vol. 20, pp. 269-275, 2001.
188
[14] J. M. Miller, J. A. Ashton‐Miller, and J. O. DeLancey, "A pelvic muscle
precontraction can reduce cough‐related urine loss in selected women with mild
SUI," Journal of the American Geriatrics Society, vol. 46, pp. 870-874, 1998.
[15] J. M. Miller, D. Perucchini, L. T. Carchidi, J. O. DeLancey, and J. Ashton-Miller,
"Pelvic floor muscle contraction during a cough and decreased vesical neck
mobility," Obstetrics and gynecology, vol. 97, p. 255, 2001.
[16] A. H. Kegel, "Progressive resistance exercise in the functional restoration of the
perineal muscles," American journal of obstetrics and gynecology, vol. 56, pp. 238-
248, 1948.
[17] S. Henalla, C. Hutchins, P. Robinson, and J. MacVicar, "Non-operative methods in
the treatment of female genuine stress incontinence of urine," Journal of obstetrics
and gynaecology, vol. 9, pp. 222-225, 1989.
[18] L. Mouritsen, C. Frimodt‐Møller, and M. Møller, "Long‐term effect of pelvic floor
exercises on female urinary incontinence," British journal of urology, vol. 68, pp.
32-37, 1991.
[19] K. Bø and T. Talseth, "Long-term effect of pelvic floor muscle exercise 5 years
after cessation of organized training," Obstetrics & Gynecology, vol. 87, pp. 261-
265, 1996.
[20] H. Cammu, M. Van Nylen, and J. Amy, "A 10‐year follow‐up after Kegel pelvic
floor muscle exercises for genuine stress incontinence," BJU international, vol. 85,
pp. 655-658, 2000.
[21] R. A. Castro, R. M. Arruda, M. R. Zanetti, P. D. Santos, M. G. Sartori, and M. J.
Girão, "Single-blind, randomized, controlled trial of pelvic floor muscle training,
electrical stimulation, vaginal cones, and no active treatment in the management of
stress urinary incontinence," Clinics, vol. 63, pp. 465-472, 2008.
[22] V. Pereira, M. De Melo, G. Correia, and P. Driusso, "Vaginal cone for
postmenopausal women with stress urinary incontinence: randomized, controlled
trial," Climacteric, vol. 15, pp. 45-51, 2012.
[23] V. S. Pereira, M. V. de Melo, G. N. Correia, and P. Driusso, "Long‐term effects of
pelvic floor muscle training with vaginal cone in post‐menopausal women with
urinary incontinence: A randomized controlled trial," Neurourology and
urodynamics, vol. 32, pp. 48-52, 2013.
[24] C. Dannecker, V. Wolf, R. Raab, H. Hepp, and C. Anthuber, "EMG-biofeedback
assisted pelvic floor muscle training is an effective therapy of stress urinary or
mixed incontinence: a 7-year experience with 390 patients," Archives of
Gynecology and Obstetrics, vol. 273, pp. 93-97, 2005.
[25] S. Mørkved, K. Bø, and T. Fjørtoft, "Effect of adding biofeedback to pelvic floor
muscle training to treat urodynamic stress incontinence," Obstetrics & Gynecology,
vol. 100, pp. 730-739, 2002.
[26] K. Bø, B. Kvarstein, R. R. Hagen, and S. Larsen, "Pelvic floor muscle exercise for
the treatment of female stress urinary incontinence: II. Validity of vaginal pressure
measurements of pelvic floor muscle strength and the necessity of supplementary
methods for control of correct contraction," Neurourology and Urodynamics, vol.
9, pp. 479-487, 1990.
189
[27] L. Brubaker, J. T. Benson, A. Bent, A. Clark, and S. Shott, "Transvaginal electrical
stimulation for female urinary incontinence," American journal of obstetrics and
gynecology, vol. 177, pp. 536-540, 1997.
[28] P. Herbison, J. Hay-Smith, G. Ellis, and K. Moore, "Effectiveness of
anticholinergic drugs compared with placebo in the treatment of overactive bladder:
systematic review," Bmj, vol. 326, p. 841, 2003.
[29] L. Viktrup and R. C. Bump, "Pharmacological agents used for the treatment of
stress urinary incontinence in women," Current medical research and opinion, vol.
19, pp. 485-490, 2003.
[30] W. N. Kernan, C. M. Viscoli, L. M. Brass, J. P. Broderick, T. Brott, E. Feldmann,
et al., "Phenylpropanolamine and the risk of hemorrhagic stroke," New England
Journal of Medicine, vol. 343, pp. 1826-1832, 2000.
[31] E. S. Rovner and A. J. Wein, "Treatment options for stress urinary incontinence,"
Reviews in urology, vol. 6, p. S29, 2004.
[32] P. Mariappan, A. Alhasso, Z. Ballantyne, A. Grant, and J. N’Dow, "Duloxetine, a
serotonin and noradrenaline reuptake inhibitor (SNRI) for the treatment of stress
urinary incontinence: a systematic review," European urology, vol. 51, pp. 67-74,
2007.
[33] S. Boy, A. Reitz, B. Wirth, P. A. Knapp, P. M. Braun, A. Haferkamp, et al.,
"Facilitatory neuromodulative effect of duloxetine on pudendal motor neurons
controlling the urethral pressure: a functional urodynamic study in healthy women,"
European urology, vol. 50, pp. 119-125, 2006.
[34] D. D. Sweeney and M. B. Chancellor, "Treatment of stress urinary incontinence
with duloxetine hydrochloride," Reviews in urology, vol. 7, p. 81, 2005.
[35] R. Millard, K. Moore, R. Rencken, I. Yalcin, and R. Bump, "Duloxetine vs placebo
in the treatment of stress urinary incontinence: a four‐continent randomized clinical
trial," BJU international, vol. 93, pp. 311-318, 2004.
[36] G. M. Ghoniem, J. S. Van Leeuwen, D. M. Elser, R. M. Freeman, Y. D. Zhao, I.
Yalcin, et al., "A randomized controlled trial of duloxetine alone, pelvic floor
muscle training alone, combined treatment and no active treatment in women with
stress urinary incontinence," The Journal of urology, vol. 173, pp. 1647-1653, 2005.
[37] H. H. Lin, B. C. Sheu, M. C. Lo, and S. C. Huang, "Comparison of treatment
outcomes of imipramine for female genuine stress incontinence," BJOG: An
International Journal of Obstetrics & Gynaecology, vol. 106, pp. 1089-1092, 1999.
[38] G. Legendre, V. Ringa, A. Fauconnier, and X. Fritel, "Menopause, hormone
treatment and urinary incontinence at midlife," Maturitas, vol. 74, pp. 26-30, 2013.
[39] P. Smith, G. Heimer, A. Norgren, and U. Ulmsten, "Localization of steroid
hormone receptors in the pelvic muscles," European Journal of Obstetrics &
Gynecology and Reproductive Biology, vol. 50, pp. 83-85, 1993.
[40] S. Hunskaar, E. Arnold, K. Burgio, A. Diokno, A. Herzog, and V. Mallett,
"Epidemiology and natural history of urinary incontinence," International
Urogynecology Journal, vol. 11, pp. 301-319, 2000.
[41] M. A. Malallah and T. F. Al-Shaiji, "Pharmacological treatment of pure stress
urinary incontinence: a narrative review," International urogynecology journal, vol.
26, pp. 477-485, 2015.
190
[42] G. Novara, A. Galfano, R. Boscolo-Berto, S. Secco, S. Cavalleri, V. Ficarra, et al.,
"Complication rates of tension-free midurethral slings in the treatment of female
stress urinary incontinence: a systematic review and meta-analysis of randomized
controlled trials comparing tension-free midurethral tapes to other surgical
procedures and different devices," european urology, vol. 53, pp. 288-309, 2008.
[43] J. S. Dunn Jr, A. E. Bent, R. M. Ellerkman, M. A. Nihira, and C. F. Melick,
"Voiding dysfunction after surgery for stress incontinence: literature review and
survey results," International Urogynecology Journal, vol. 15, pp. 25-31, 2004.
[44] R. Abouassaly, J. R. Steinberg, M. Lemieux, C. Marois, L. I. Gilchrist, J. L.
Bourque, et al., "Complications of tension‐free vaginal tape surgery: a multi‐
institutional review," BJU international, vol. 94, pp. 110-113, 2004.
[45] P. A. Nosti and C. B. Iglesia, "Medicolegal issues surrounding devices and mesh
for surgical treatment of prolapse and incontinence," Clinical obstetrics and
gynecology, vol. 56, pp. 221-228, 2013.
[46] D. Y. Deng, M. Rutman, S. Raz, and L. V. Rodriguez, "Presentation and
management of major complications of midurethral slings: Are complications
under‐reported?," Neurourology and urodynamics, vol. 26, pp. 46-52, 2007.
[47] U. Food and D. Administration, "UPDATE on serious complications associated
with transvaginal placement of surgical mesh for pelvic organ prolapse: FDA safety
communication," Silver Spring: FDA, 2011.
[48] M. Peyromaure, V. Ravery, and L. Boccon‐Gibod, "The management of stress
urinary incontinence after radical prostatectomy," BJU international, vol. 90, pp.
155-161, 2002.
[49] P. Hammerer and H. Huland, "Urodynamic evaluation of changes in urinary control
after radical retropubic prostatectomy," The Journal of urology, vol. 157, pp. 233-
236, 1997.
[50] R. Chao and M. E. Mayo, "Incontinence after radical prostatectomy: detrusor or
sphincter causes," The Journal of urology, vol. 154, pp. 16-18, 1995.
[51] R. S. Hollabaugh, R. R. Dmochowski, T. G. Kneib, and M. S. Steiner, "Preservation
of putative continence nerves during radical retropubic prostatectomy leads to more
rapid return of urinary continence," Urology, vol. 51, pp. 960-967, 1998.
[52] M. T. Filocamo, V. L. Marzi, G. Del Popolo, F. Cecconi, M. Marzocco, A. Tosto,
et al., "Effectiveness of early pelvic floor rehabilitation treatment for post-
prostatectomy incontinence," European urology, vol. 48, pp. 734-738, 2005.
[53] R. MacDonald, H. A. Fink, C. Huckabay, M. Monga, and T. J. Wilt, "Pelvic floor
muscle training to improve urinary incontinence after radical prostatectomy: a
systematic review of effectiveness," BJU international, vol. 100, pp. 76-81, 2007.
[54] G. T. Bales, G. S. Gerber, T. X. Minor, D. A. Mhoon, J. M. McFarland, H. L. Kim,
et al., "Effect of preoperative biofeedback/pelvic floor training on continence in
men undergoing radical prostatectomy," Urology, vol. 56, pp. 627-630, 2000.
[55] S. Wille, A. Sobottka, A. Heidenreich, and R. Hofmann, "Pelvic floor exercises,
electrical stimulation and biofeedback after radical prostatectomy: results of a
prospective randomized trial," The Journal of urology, vol. 170, pp. 490-493, 2003.
[56] A. Sousa-Escandon, J. Cabrera, F. Mantovani, M. Moretti, E. Ioanidis, N.
Kondelidis, et al., "Adjustable suburethral sling (Male Remeex System®) in the
191
treatment of male stress urinary incontinence: a multicentric European study,"
european urology, vol. 52, pp. 1473-1480, 2007.
[57] W. A. Hübner, H. Gallistl, M. Rutkowski, and E. R. Huber, "Adjustable
bulbourethral male sling: experience after 101 cases of moderate‐to‐severe male
stress urinary incontinence," BJU international, vol. 107, pp. 777-782, 2011.
[58] T. Yamamoto, M. Gotoh, M. Kato, T. Majima, K. Toriyama, Y. Kamei, et al.,
"Periurethral injection of autologous adipose‐derived regenerative cells for the
treatment of male stress urinary incontinence: Report of three initial cases,"
International Journal of Urology, vol. 19, pp. 652-659, 2012.
[59] A. Kumar, E. R. Litt, K. N. Ballert, and V. W. Nitti, "Artificial urinary sphincter
versus male sling for post-prostatectomy incontinence—what do patients choose?,"
The Journal of urology, vol. 181, pp. 1231-1235, 2009.
[60] A. E. Gousse, S. Madjar, M.-M. Lambert, and I. J. Fishman, "Artificial urinary
sphincter for post-radical prostatectomy urinary incontinence: long-term subjective
results," The Journal of urology, vol. 166, pp. 1755-1758, 2001.
[61] K. Caldwell, P. Cook, F. Flack, and D. James, "TREATMENT OF POST‐
PROSTATECTOMY INCONTINENCE BY ELECTRONIC IMPLANT," British
journal of urology, vol. 40, pp. 183-186, 1968.
[62] D. R. Merrill, M. Bikson, and J. G. Jefferys, "Electrical stimulation of excitable
tissue: design of efficacious and safe protocols," Journal of neuroscience methods,
vol. 141, pp. 171-198, 2005.
[63] T. Cameron, F. J. Richmond, and G. E. Loeb, "Effects of regional stimulation using
a miniature stimulator implanted in feline posterior biceps femoris," IEEE
transactions on biomedical engineering, vol. 45, pp. 1036-1043, 1998.
[64] J. T. Mortimer, "Motor prostheses," Comprehensive Physiology, 1981.
[65] G. A. Cox, Y. Sunada, K. P. Campbell, and J. S. Chamberlain, "Dp71 can restore
the dystrophin-associated glycoprotein complex in muscle but fails to prevent
dystrophy," Nature genetics, vol. 8, pp. 333-339, 1994.
[66] J. C. Jarvis, H. Sutherland, C. N. Mayne, S. J. Gilroy, and S. Salmons, "Induction
of a fast-oxidative phenotype by chronic muscle stimulation: mechanical and
biochemical studies," American Journal of Physiology-Cell Physiology, vol. 270,
pp. C306-C312, 1996.
[67] J. C. Jarvis, T. Mokrusch, M. Kwende, H. Sutherland, and S. Salmons, "Fast‐to‐
slow transformation in stimulated rat muscle," Muscle & nerve, vol. 19, pp. 1469-
1475, 1996.
[68] C. N. Mayne, H. Sutherland, J. C. Jarvis, S. J. Gilroy, A. J. Craven, and S. Salmons,
"Induction of a fast-oxidative phenotype by chronic muscle stimulation:
histochemical and metabolic studies," American Journal of Physiology-Cell
Physiology, vol. 270, pp. C313-C320, 1996.
[69] S. J. Gilroy, S. Salmons, and S. R. Pennington, "Changes in nuclear protein
composition in response to chronic electrical stimulation of skeletal muscle,"
Electrophoresis, vol. 18, pp. 809-813, 1997.
[70] P. Mela, P. H. Veltink, P. A. Huijing, S. Salmons, and J. C. Jarvis, "The optimal
stimulation pattern for skeletal muscle is dependent on muscle length," IEEE
Transactions on neural systems and rehabilitation engineering, vol. 10, pp. 85-93,
2002.
192
[71] H. Sutherland, J. C. Jarvis, and S. Salmons, "Pattern dependence in the stimulation‐
induced type transformation of rabbit fast skeletal muscle," Neuromodulation:
Technology at the Neural Interface, vol. 6, pp. 176-189, 2003.
[72] J. Henriksson, M. Chi, C. S. Hintz, D. A. Young, K. K. Kaiser, S. Salmons, et al.,
"Chronic stimulation of mammalian muscle: changes in enzymes of six metabolic
pathways," American Journal of Physiology-Cell Physiology, vol. 251, pp. C614-
C632, 1986.
[73] T. Bajd, M. Gregoric, L. Vodovnik, and H. Benko, "Electrical stimulation in
treating spasticity resulting from spinal cord injury," Archives of physical medicine
and rehabilitation, vol. 66, pp. 515-517, 1985.
[74] J. Chae, F. Bethoux, T. Bohinc, L. Dobos, T. Davis, and A. Friedl, "Neuromuscular
stimulation for upper extremity motor and functional recovery in acute
hemiplegia," Stroke, vol. 29, pp. 975-979, 1998.
[75] P. D. Faghri, M. M. Rodgers, R. M. Glaser, J. G. Bors, C. Ho, and P. Akuthota,
"The effects of functional electrical stimulation on shoulder subluxation, arm
function recovery, and shoulder pain in hemiplegic stroke patients," Archives of
physical medicine and rehabilitation, vol. 75, pp. 73-79, 1994.
[76] S. L. Linn, M. H. Granat, and K. R. Lees, "Prevention of shoulder subluxation after
stroke with electrical stimulation," Stroke, vol. 30, pp. 963-968, 1999.
[77] L. Ada and A. Foongchomcheay, "Efficacy of electrical stimulation in preventing
or reducing subluxation of the shoulder after stroke: a meta-analysis," Australian
Journal of Physiotherapy, vol. 48, pp. 257-267, 2002.
[78] G. J. Renzenbrink and M. J. Ijzerman, "Percutaneous neuromuscular electrical
stimulation (P-NMES) for treating shoulder pain in chronic hemiplegia. Effects on
shoulder pain and quality of life," Clinical rehabilitation, vol. 18, pp. 359-365,
2004.
[79] C. Kerr and B. McDowell, "Electrical stimulation in cerebral palsy: a review of
effects on strength and motor function," Developmental Medicine & Child
Neurology, vol. 46, pp. 205-213, 2004.
[80] A. I. Kottink, L. J. Oostendorp, J. H. Buurke, A. V. Nene, H. J. Hermens, and M. J.
IJzerman, "The orthotic effect of functional electrical stimulation on the
improvement of walking in stroke patients with a dropped foot: a systematic
review," Artificial organs, vol. 28, pp. 577-586, 2004.
[81] T. Cameron, T. L. Liinamaa, G. E. Loeb, and F. J. Richmond, "Long-term
biocompatibility of a miniature stimulator implanted in feline hind limb muscles,"
IEEE transactions on biomedical engineering, vol. 45, pp. 1024-1035, 1998.
[82] T. Cameron, G. E. Loeb, R. A. Peck, J. H. Schulman, P. Strojnik, and P. R. Troyk,
"Micromodular implants to provide electrical stimulation of paralyzed muscles and
limbs," IEEE Transactions on Biomedical Engineering, vol. 44, pp. 781-790, 1997.
[83] T. Cameron, G. E. Loeb, R. Peck, J. H. Schulman, P. Strojnik, and P. R. Troyk,
"Micromodular implants to provide electrical stimulation of paralyzed muscles and
limbs," Biomedical Engineering, IEEE Transactions on, vol. 44, pp. 781-790, 1997.
[84] A. D. Salter, F. J. Richmond, and G. E. Loeb, "Prevention of muscle disuse atrophy
by low-frequency electrical stimulation in rats," IEEE Transactions on neural
systems and rehabilitation engineering, vol. 11, pp. 218-226, 2003.
193
[85] K. Singh, F. J. Richmond, and G. E. Loeb, "Recruitment properties of intramuscular
and nerve-trunk stimulating electrodes," IEEE Transactions on Rehabilitation
Engineering, vol. 8, pp. 276-285, 2000.
[86] B. Burns, L. Watkins, and P. J. Goadsby, "Treatment of hemicrania continua by
occipital nerve stimulation with a bion device: long-term follow-up of a crossover
study," The Lancet Neurology, vol. 7, pp. 1001-1012, 2008.
[87] J. Bosch, "The bion device: a minimally invasive implantable ministimulator for
pudendal nerve neuromodulation in patients with detrusor overactivity
incontinence," Urologic Clinics, vol. 32, pp. 109-112, 2005.
[88] A. C. D. Salter, S. D. Bagg, J. L. Creasy, C. Romano, D. Romano, F. J. Richmond,
et al., "First clinical experience with BION implants for therapeutic electrical
stimulation," Neuromodulation: Technology at the Neural Interface, vol. 7, pp. 38-
47, 2004.
[89] D. J. Weber, R. B. Stein, K. M. Chan, G. Loeb, F. Richmond, R. Rolf, et al.,
"BIONic WalkAide for correcting foot drop," IEEE Transactions on Neural
Systems and Rehabilitation Engineering, vol. 13, pp. 242-246, 2005.
[90] H. M. Kaplan, L. L. Baker, S. Rubayi, and G. E. Loeb, "Preventing ischial pressure
ulcers: III. Clinical pilot study of chronic neuromuscular electrical stimulation,"
Applied Bionics and Biomechanics, vol. 8, pp. 345-359, 2011.
[91] J. Pain and N. Pain, "Occipital nerve stimulation with the Bion® microstimulator
for the treatment of medically refractory chronic cluster headache," 2011.
[92] G. Jiang and D. D. Zhou, "Technology advances and challenges in hermetic
packaging for implantable medical devices," in Implantable Neural Prostheses 2,
ed: Springer, 2009, pp. 27-61.
[93] T. Yamanishi, T. Kamai, and K. I. Yoshida, "Neuromodulation for the treatment of
urinary incontinence," International journal of urology, vol. 15, pp. 665-672, 2008.
[94] N. T. Galloway, R. E. El-Galley, P. K. Sand, R. A. Appell, H. W. Russell, and S. J.
Carlan, "Extracorporeal magnetic innervation therapy for stress urinary
incontinence," Urology, vol. 53, pp. 1108-1111, 1999.
[95] K. Caldwell, "The treatment of incontinence by electronic implants. Hunterian
Lecture delivered at the Royal College of Surgeons of England on 8th December
1966," Annals of the Royal College of Surgeons of England, vol. 41, p. 447, 1967.
[96] G. E. Loeb, R. A. Peck, W. H. Moore, and K. Hood, "BION system for distributed
neural prosthetic interfaces," Medical Engineering & Physics, vol. 23, pp. 9-18,
2001.
[97] G. E. Loeb, F. J. Richmond, and L. L. Baker, "The BION devices: injectable
interfaces with peripheral nerves and muscles," Neurosurgical focus, vol. 20, pp.
1-9, 2006.
[98] R. Stein, D. Weber, K. Chan, G. Loeb, R. Rolf, and S. Chong, "Stimulation of
peripheral nerves with a microstimulator: experimental results and clinical
application to correct foot drop," in Proceedings of the 9th annual conference of
the International FES Society, 2004.
[99] A. Grimby, I. Milsom, U. Molander, I. Wiklund, and P. Ekelund, "The influence of
urinary incontinence on the quality of life of elderly women," Age and ageing, vol.
22, pp. 82-89, 1993.
194
[100] E. Petri and K. Ashok, "Comparison of late complications of retropubic and
transobturator slings in stress urinary incontinence," International urogynecology
journal, vol. 23, pp. 321-325, 2012.
[101] P. Donaldson, "The essential role played by adhesion in the technology of
neurological prostheses," International journal of adhesion and adhesives, vol. 16,
pp. 105-107, 1996.
[102] A. N. Vest., L. Zhou, X. Huang, V. Norekyan, Y. Bar-Cohen, R. H. Chmait, et al.,
"Design and Testing of a Transcutaneous RF Recharging System for a Fetal
Micropacemaker," IEEE Transactions on Biomedical Circuits and Systems,
Accepted. 2016.
[103] N. O. Sokal, "Class-E RF power amplifiers," QEX Commun. Quart, vol. 204, pp.
9-20, 2001.
[104] T. L. Fitzpatrick, T. L. Liinamaa, I. E. Brown, T. Cameron, and F. J. R. Richmond,
"A novel method to identify migration of small implantable devices," Journal of
Long-Term Effects of Medical Implants, vol. 6, pp. 157-168, 1997.
[105] J. A. Trotter, F. J. R. Richmond, and P. P. Purslow, "Functional morphology and
motor control of series-fibered muscles," Exercise & Sport Sciences Reviews, pp.
167-213, 1995.
[106] H. Mino, J. T. Rubinstein, C. A. Miller, and P. J. Abbas, "Effects of electrode-to-
fiber distance on temporal neural response with electrical stimulation," IEEE
transactions on biomedical engineering, vol. 51, pp. 13-20, 2004.
[107] A. Hijaz, F. Daneshgari, K.-D. Sievert, and M. S. Damaser, "Animal models of
female stress urinary incontinence," The Journal of urology, vol. 179, pp. 2103-
2110, 2008.
[108] T. Shimamoto, M. Iwahashi, Y. Sugiyama, I. Laakso, A. Hirata, and T. Onishi,
"SAR evaluation in models of an adult and a child for magnetic field from wireless
power transfer systems at 6.78 MHz," Biomedical Physics & Engineering Express,
vol. 2, p. 027001, 2016.
[109] T. Yamanishi, Y. Homma, O. Nishizawa, K. Yasuda, and O. Yokoyama,
"Multicenter, randomized, sham‐controlled study on the efficacy of magnetic
stimulation for women with urgency urinary incontinence," International Journal
of Urology, vol. 21, pp. 395-400, 2014.
[110] A. C. D. Salter, F. J. R. Richmond, and G. E. Loeb, "Prevention of muscle disuse
atrophy by low-frequency electrical stimulation in rats," IEEE Transactions on
Neural Systems and Rehabilitation Engineering, vol. 11, pp. 218-226, 2003.
[111] A.-C. D. Salter, S. D. Bagg, J. L. Creasy, C. Romano, D. Romano, F. J. R.
Richmond, et al., "First Clinical Experience with BION Implants for Therapeutic
Electrical Stimulation," Neuromodulation: Technology at the Neural Interface, vol.
7, pp. 38-47, 2004.
[112] T. Stieglitz, "Manufacturing, assembling and packaging of miniaturized neural
implants," Microsystem technologies, vol. 16, pp. 723-734, 2010.
[113] S. Kim, R. Bhandari, M. Klein, S. Negi, L. Rieth, P. Tathireddy, et al., "Integrated
wireless neural interface based on the Utah electrode array," Biomedical
microdevices, vol. 11, pp. 453-466, 2009.
[114] C. Hassler, T. Boretius, and T. Stieglitz, "Polymers for neural implants," Journal
of Polymer Science Part B: Polymer Physics, vol. 49, pp. 18-33, 2011.
195
[115] P. Wang, S. Lachhman, D. Sun, S. Majerus, M. S. Damaser, C. A. Zorman, et al.,
"Non-hermetic micropackage for chronic implantable systems," in International
Symposium on Microelectronics, 2013, pp. 000166-000170.
[116] P. E. K. Donaldson, "The Encapsulation of Microelectronic Devices for Long-Term
Surgical Implantation," IEEE Trans.Biomed.Engng, vol. 23, pp. 281-285, 1976.
[117] W. Nelson, "Accelerated life testing-step-stress models and data analyses," IEEE
transactions on reliability, vol. 29, pp. 103-108, 1980.
[118] L. A. Escobar and W. Q. Meeker, "A review of accelerated test models," Statistical
science, pp. 552-577, 2006.
[119] G. Bierwagen, D. Tallman, J. Li, L. He, and C. Jeffcoate, "EIS studies of coated
metals in accelerated exposure," Progress in Organic Coatings, vol. 46, pp. 149-
158, 2003.
[120] B. Yacobi, S. Martin, K. Davis, A. Hudson, and M. Hubert, "Adhesive bonding in
microelectronics and photonics," Journal of applied physics, vol. 91, pp. 6227-6262,
2002.
[121] G. E. Loeb, R. A. Peck, W. H. Moore, and K. Hood, "BION™ system for
distributed neural prosthetic interfaces," Medical Engineering & Physics, vol. 23,
pp. 9-18, 1// 2001.
[122] L. Zhou, A. N. Vest, R. A. Peck, J. P. Sredl, X. Huang, Y. Bar-Cohen, et al.,
"Minimally invasive implantable fetal micropacemaker: mechanical testing and
technical refinements," Medical & biological engineering & computing, pp. 1-12,
2016.
[123] H. Gensler, R. Sheybani, P.-Y. Li, R. L. Mann, and E. Meng, "An implantable
MEMS micropump system for drug delivery in small animals," Biomedical
microdevices, vol. 14, pp. 483-496, 2012.
[124] A. Koulaouzidis, D. K. Iakovidis, A. Karargyris, and E. Rondonotti, "Wireless
endoscopy in 2020: Will it still be a capsule?," World journal of gastroenterology,
vol. 21, pp. 5119-5130, 2015.
[125] J. G. Chubbuck, "Intracranial pressure monitor," ed: Google Patents, 1977.
[126] K. G. Ong and C. A. Grimes, "A resonant printed-circuit sensor for remote query
monitoring of environmental parameters," Smart materials and structures, vol. 9,
p. 421, 2000.
[127] K. Z. Xuechen Huang, Sam Kohan, May Denprasert, Limin Liao, Gerald E. Loeb,
"Neurostimulation Strategy for Stress Urinary Incontinence," IEEE Transactions
on Neural Systems and Rehabilitation Engineering, 2016.
[128] S. Minnikanti, G. Diao, J. J. Pancrazio, X. Xie, L. Rieth, F. Solzbacher, et al.,
"Lifetime assessment of atomic-layer-deposited Al2O3–Parylene C bilayer coating
for neural interfaces using accelerated age testing and electrochemical
characterization," Acta Biomaterialia, vol. 10, pp. 960-967, 2// 2014.
[129] D. Hukins, A. Mahomed, and S. Kukureka, "Accelerated aging for testing
polymeric biomaterials and medical devices," Medical engineering & physics, vol.
30, pp. 1270-1274, 2008.
[130] R. C. Blish, S. Li, H. Kinoshita, S. Morgan, and A. F. Myers, "Gold–aluminum
intermetallic formation kinetics," IEEE Transactions on Device and Materials
Reliability, vol. 7, pp. 51-63, 2007.
196
[131] X. Xie, L. Rieth, R. Caldwell, S. Negi, R. Bhandari, R. Sharma, et al., "Effect of
bias voltage and temperature on lifetime of wireless neural interfaces with Al2O3
and parylene bilayer encapsulation," Biomedical microdevices, vol. 17, pp. 1-8,
2015.
[132] H. Guo, E. Pohl, and A. Gerokostopoulos, "Determining the right sample size for
your test: theory and application," in 2013 Annual Reliability and Maintainability
Symposium. Available from http://www. reliasoft.
com/pubs/2013_RAMS_determining _right_sample_size. p df [Accessed on 5 June
2014], 2013.
[133] K. Birkelund, L. Nørgaard, and E. V. Thomsen, "Enhanced polymeric
encapsulation for MEMS based multi sensors for fisheries research," Sensors and
Actuators A: Physical, vol. 170, pp. 196-201, 2011.
[134] P. Donaldson, "Aspects of silicone rubber as an encapsulant for neurological
prostheses," Medical and Biological Engineering and Computing, vol. 29, pp. 34-
39, 1991.
[135] A. C. Loos and G. S. Springer, "Curing of epoxy matrix composites," Journal of
composite materials, vol. 17, pp. 135-169, 1983.
[136] H. Xu, C. Liu, V. V. Silberschmidt, S. Pramana, T. J. White, Z. Chen, et al., "A
micromechanism study of thermosonic gold wire bonding on aluminum pad,"
Journal of applied physics, vol. 108, p. 113517, 2010.
[137] C. Breach and F. Wulff, "New observations on intermetallic compound formation
in gold ball bonds: general growth patterns and identification of two forms of Au 4
Al," Microelectronics Reliability, vol. 44, pp. 973-981, 2004.
[138] Gerald E. Loeb, Frances J. R. Richmond, and Lucinda L. Baker, "The BION
devices: injectable interfaces with peripheral nerves and muscles," Neurosurgical
Focus, vol. 20, pp. 1-9, 2006.
[139] J. L. Bosch, "The bion device: a minimally invasive implantable ministimulator for
pudendal nerve neuromodulation in patients with detrusor overactivity
incontinence," Urol Clin North Am, vol. 32, pp. 109-12, Feb 2005.
[140] J. Groen, C. Amiel, and J. L. H. R. Bosch, "Chronic pudendal nerve
neuromodulation in women with idiopathic refractory detrusor overactivity
incontinence: Results of a pilot study with a novel minimally invasive implantable
mini-stimulator," Neurourology and Urodynamics, vol. 24, pp. 226-230, 2005.
Abstract (if available)
Abstract
Stress urinary incontinence is a common problem in which there is involuntary urine leakage during activities that increase abdominal pressure. Although it is not a life-threatening problem, it has high prevalence, affecting millions of elderly people. Most patients with this problem have weakness or damage of the external urethral sphincter and pelvic floor muscle as a result of childbirth or prostate surgery. Invasive surgical treatment such as vaginal tape and hydraulic sphincters have significant long term-complications. Voluntary muscle exercise program have been documented to produce excellent results but most patients find it difficult to perform such exercises correctly or consistently enough to obtain such results. Previous attempts to induce muscle exercise by electrical stimulation failed due either to unpleasant sensation from external electrodes, or to connection failure of long leads to conventional, fully implantable stimulators. Intramuscular electrical stimulation via a minimally invasively implanted microstimulator should overcome these challenges. ❧ We have developed a single channel, monolithic microstimulator (3.5mm diameter * 10 mm long) that can be implanted close to targeted motor axons via minimally invasive procedure to generate charge-regulated pulses for strong muscle contraction. The implanted device receives stimulus power and timing by inductive coupling from a radio frequency transmitter in a seat cushion. The physician and patient can use an app in a tablet computer or smartphone to prescribe exercise patterns and control and record daily use via Bluetooth communication to the seat cushion. ❧ Preclinical testing demonstrated that the implant produces the desired output pulse (0.05-2.8µC) up to 12cm from the face of the seat cushion. Accelerated life testing in vitro of the non-hermetic epoxy encapsulation demonstrated expected lifetime in vivo > 1 year. Chronically stimulated implants in an animal study reliably activated skeletal muscle without apparent discomfort and were well anchored by minimally reactive connective tissue encapsulation after 1-3 months. The validated system is now being used in a clinical trial to demonstrate safety and efficacy. The primary outcome measure is a conventional pad weight test for urine leakage during controlled exercise. The inert, passive implants are intended to be left implanted after 3 months of daily, electrically induced exercise.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Mechanical design and preclinical testing of a percutaneously implantable fetal micropacemaker
PDF
Investigating the role of muscle physiology and spinal circuitry in sensorimotor control
PDF
Neuromuscular electrical stimulation for pressure ulcer prevention
PDF
Electronics design and in vivo evaluation of a wirelessly rechargeable fetal micropacemaker
PDF
Electromyography of spinal cord injured rodents trained by neuromuscular electrical stimulation timed to robotic treadmill training
PDF
On the electrophysiology of multielectrode recordings of the basal ganglia and thalamus to improve DBS therapy for children with secondary dystonia
PDF
Towards a high resolution retinal implant
PDF
Sencil project: design and fabrication of a disposable, percutaneous sensor platform
PDF
Next generation neural interfaces for neural modulation and prosthesis
PDF
A virtual reality exergaming system to enhance brain health in older adults at risk for Alzheimer’s disease
Asset Metadata
Creator
Huang, Xuechen
(author)
Core Title
A percutaneously implantable wireless neurostimulator for treatment of stress urinary incontinence
School
Viterbi School of Engineering
Degree
Doctor of Philosophy
Degree Program
Biomedical Engineering
Publication Date
09/25/2017
Defense Date
08/15/2017
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
implant,minimal invasiveness,motor axon,neurostimulation,non-hermetic packaging,OAI-PMH Harvest,pelvic floor muscle training,stress urinary incontinence,wireless
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Loeb, Gerald (
committee chair
), Valero-Cuevas, Francisco (
committee chair
), Kirages, Daniel (
committee member
), Meng, Ellis (
committee member
)
Creator Email
xcrhuang@gmail.com,xuechenh@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c40-436304
Unique identifier
UC11264439
Identifier
etd-HuangXuech-5769.pdf (filename),usctheses-c40-436304 (legacy record id)
Legacy Identifier
etd-HuangXuech-5769.pdf
Dmrecord
436304
Document Type
Dissertation
Rights
Huang, Xuechen
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
implant
minimal invasiveness
motor axon
neurostimulation
non-hermetic packaging
pelvic floor muscle training
stress urinary incontinence
wireless