University of Southern California Dissertations and Theses

Metabolic effects of magnesium supplementation in women with a history of gestational diabetes mellitus

(USC Thesis Other)

Metabolic effects of magnesium supplementation in women with a history of gestational diabetes mellitus

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content METABOLIC EFFECTS OF MAGNESIUM SUPPLEMENTATION IN WOMEN WITH A HISTORY OF GESTATIONAL DIABETES MELLITUS by Christopher David Hass A Thesis Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree MASTER OF SCIENCE (APPLIED BIOSTATISTICS AND EPIDEMIOLOGY) August 2003 Copyright 2003 Christopher David Hass Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Num ber: 1417922 Copyright 2 0 0 3 by H ass, C hristopher David All rights reserved. INFORMATION TO U S E R S T he quality of this reproduction is d e p e n d e n t upon th e quality of th e copy subm itted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, s u b sta n d a rd m argins, and im proper alignm ent can adversely affect reproduction. In th e unlikely ev e n t that the au th o r did not se n d a com plete m anuscript and th ere a re missing p ag es, th e s e will b e noted. Also, if unauthorized copyright material had to be rem oved, a note will indicate th e deletion. UMI UMI Microform 1417922 Copyright 2004 by P ro Q u est Information and Learning C om pany. All rights reserved. This microform edition is protected ag ain st unauthorized copying u n d er Title 17, United S ta te s Code. P ro Q u est Information a n d Learning C o m p an y 300 North Z e e b R oad P.O. Box 1346 Ann Arbor, Ml 48 1 0 6 -1 3 4 6 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UNIVERSITY O F SO U T H E R N CALIFORNIA THE GRADUATE S C H O O L UNIVERSITY P A R K LOS ANG ELES. CALIFORNIA 9 0 0 0 7 This thesis, written by under the direction of h..i& Thesis Committee, and approved by all its members, has been pre sented to and accepted by the Dean of The Graduate School, in partial fulfillment of the requirements for the degree of £ i ......... Dem THESIS' COMMITTEE ChawmaM Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ii TABLE OF CONTENTS List of Tables iii Abstract iv CHAPTER I: INTRODUCTION 1 CHAPTER IT . RESEARCH METHODS Study Participants 3 Study Design 3 Clinical Testing 4 Statistical Analysis 4 CHAPTER III: RESULTS Recruitment and Randomization 7 Baseline Comparison of Treatment Groups 7 Compliance with Treatment 10 Changes in Insulin Sensitivity 10 Changes in Fasting Glucose 11 Changes in Fasting Insulin 11 Changes in Glucose Effectiveness 18 Changes in Other Outcomes 19 Comparison of Treatment Groups at End of Study 19 CHAPTER TV: DISCUSSION 21 WORKS CITED 24 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES Table 1: Means, Medians, and Inter-Quartile-Ranges for Baseline Demographic and Outcome Variables by Treatment Group (All Randomized Subjects) Table 2: Means, Medians, and Inter-Quartile-Ranges for Baseline Demographic and Outcome Variables by Treatment Group (Subjects Completing Trial) Table 3: Treatment Compliance Rates by Treatment Group for Different Periods of Follow-up Table 4: Means, Medians, and Inter-Quartile Ranges for Absolute Changes in Outcome Variables Between Baseline and 3-month End of Study Visit (Subjects Completing Trial) Table 5: Means, Medians, and Inter-Quartile Ranges for Percent Changes in Outcome Variables Between Baseline and 3-month End of Study Visit (Subjects Completing Trial) Table 6: P-values for Differences by Treatment Group for Absolute Changes in Medical and Outcome Variables, Unadjusted and Adjusted for Specified Unbalanced Baseline and/or On-trial Variables (Subjects Completing Trial) Table 7: P-values for Differences by Treatment Group for Percent Changes in Medical and Outcome Variables, Unadjusted and Adjusted for Specified Unbalanced Baseline and/or On-trial Variables (Subjects Completing Trial) Table 8: Predicted Adjusted Means for Selected Outcome Variables by Treatment Group (Subjects Completing Trial) Table 9: Means, Medians, and Inter-Quartile-Ranges for End-of-Study Demographic and Outcome Variables by Treatment Group (Subjects Completing Trial) Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT In this randomized trial of 13 subjects in each of two treatment groups, magnesium lactate (504 mg) or placebo was administered over 3 months to study the effects of magnesium supplementation on insulin sensitivity and glucose tolerance in non-pregnant Hispanic women with a history of gestational diabetes mellitus. After comparing baseline and 3-month oral and intravenous glucose tolerance test results between treatment groups, no significant improvement in insulin sensitivity (p>0.5) or glucose tolerance (p>0.5) was detected due to magnesium supplementation. However, glucose effectiveness did show significant improvement after adjustment for age, BMI, and 6-week compliance rates (p=0.02) and was significantly higher in an end of study comparison between treatment groups (p=0.048). Additionally, the glucose disappearance rate was significantly higher (p=0.04) in the magnesium group at the end of the study. These results suggest magnesium supplementation may have effects over a longer intervention period. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 CHAPTER I: INTRODUCTION The prevalence of type 2 diabetes mellitus (T2DM) is increasing in the United States, and it is one of the 10 leading causes of death (11, p. 3; 4, p. B43). Women with a history of gestational diabetes mellitus (GDM) are at increased risk for developing T2DM, with a 30-50% chance of developing the disease within 5 years of their index pregnancy (4, p. B44; 7, p. 782; 3, p. 1302). With 3-5% of all pregnancies affected (7, p. 782), the relatively large number of women with a history of GDM are ideal candidates for diabetes prevention. It has been shown that hypomagnesemia is a common characteristic of diabetes, occurring in 25-38% of all T2DM patients (14, p. 3783; 8, p. 2151; 12, p. 265; 9, p. 682). Although the metabolic pathways linking magnesium (Mg) to diabetes are not fully understood, Mg is known to be involved in the metabolism of both glucose and insulin (14, p. 3783; 12, p. 265; 13, p. 767). Furthermore, a number of studies have suggested an association between Mg depletion and insulin resistance (14, p. 3783; 1, p. E745; 5, p. 188); and magnesium supplementation has been shown to improve metabolic control in patients with T2DM. Finally, in several studies using rodent models of diabetes. Mg supplementation appeared to prevent the onset of diabetes (1, p. E745). Traditional methods of diabetes prevention, such as diet, weight loss, and exercise, although very effective, may require significant behavioral changes by the patient. Patients often show little benefit from these traditional prevention methods, as they historically tend to have poor long-term compliance rates (10, p. 786; 2, p. 1573). In contrast, magnesium supplementation with Mg lactate has been well tolerated by patients with few reported side effect, including abdominal pain and diarrhea (9, p. 683; 5, p. 191). Therefore, it has good Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. potential for high treatment compliance. If magnesium supplementation can prevent or delay the onset of diabetes, it would provide an ideal method of intervention. The purpose of the pilot study described in this report was to test the effect of magnesium supplementation over a 3-month period on insulin sensitivity and glucose tolerance in non-pregnant Hispanic women with a prior history of GDM. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3 CHAPTER II: RESEARCH METHODS Study Participants Women’s Hospital at the USC/LAC Medical Center maintains a program of glucose tolerance testing for women with a history of gestational diabetes (GDM) (2, p. 1573). OGTT tests are performed 4-16 weeks postpartum with annual follow-up testing recommended for non-diabetic patients. Potential subjects had to be free of medical conditions that could affect glucose tolerance or magnesium metabolism, and have no plans to become pregnant within 6 months. Subjects >18 years of age meeting all other eligibility criteria were invited to enroll. The pilot study enrolled a total of 26 subjects. After subject recruitment was completed, the sample size of the study population was sufficient to detect a >45% treatment effect in insulin sensitivity between treatment groups with 80% power at the a=0.05 level of significance. For glucose effectiveness (Sg ) and the glucose disappearance rate (Kg ), the sample size of the study population was sufficient to detect a >15% treatment effect and >11% treatment effect respectively with 80% power at the a=0.05 level of significance. Study Design A baseline OGTT was conducted for all subjects. The same baseline visit included measurements of height, weight, blood pressure, and blood chemistries. Study participants returned to the General Clinical Research Center (GCRC) for a baseline intravenous glucose tolerance test (IVGTT) within 6 weeks of the initial OGTT screening. Following the IVGTT, subjects were randomly assigned to treatment with placebo or 504 mg/day magnesium lactate tablets for a 12-week period. Placebo and Mg medications were supplied Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4 by Niche Pharmaceuticals as identical tablets in numbered vials, masking the randomization assignments to both subjects and investigators. Six weeks after randomization, subjects returned to the GCRC for a physical examination, fasting glucose measurement, and a pill count. At 12 weeks, subjects returned for an OGTT, IVGTT, and final pill count. Clinical Testing Weight was measured with a standard balance beam; height was measured with a stadiometer. OGTTs were administered in the morning after a 12-hour fast. Thirty minutes after an indwelling antecubital venous catheter was inserted, a 75g dextrose solution was given orally. Blood samples were taken at -10, 30, 60, 90, and 120 minutes relative to the ingestion of the dextrose solution. IVGTTs were conducted after a 12-hour fast. Two intravenous catheters were inserted: one in an antecubital vein for injections and one in a hand vein for blood sampling. The hand was warmed to ~65° C for the duration of the procedure. After 30 minutes, baseline blood samples were drawn and a dextrose solution (300 mg/Kg body weight) was given intravenously over 1 minute. Twenty minutes later an injection of tolbutamide (125 mg/m2 body surface area) was given. At 2, 3, 4, 5, 6, 8, 10, 12, 14,16,19, 22, 24, 25, 27, 30, 40,50, 60, 70, 80, 90,100, 120,140,160, 180, and 240 minutes after the dextrose injection, blood samples were drawn to measure glucose and insulin levels. Statistical Analysis Body mass index (BMI) was calculated as weight divided by height squared (Kg/m2). The MINMOD program (2, p. 1573) developed by Dr. Richard Bergman at Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5 University of Southern California was used to calculate the insulin sensitivity (Sj) and glucose effectiveness (Sg ) by minimal model analysis of IVGTT insulin and glucose results (2, p. 1573). The trapezoid approximation was used to calculate the area under the curve (AUC) for the glucose and insulin levels measured during the IVGTT and OGTT. LDL cholesterol levels were calculated as (total cholesterol) - (HDL cholesterol + [total triglycerides/5]) (13, p. 1573). Treatment compliance rates were defined as (total pills dispensed - total pills returned) / (days on medication * 6 pills/day), and calculated separately for both 6-week intervals of the 12-week study period. An overall compliance rate for the full 12-week study period was calculated for subjects with complete pill counts, using the same definition. Due to the skewed distribution of the data and small sample sizes, the Wilcoxon Rank Sum test was used to compare baseline parameters between treatment groups and was repeated for the same parameters at the end of study. Two analyses, Wilcoxon Rank Sum and analysis of variance, were performed on both the absolute and percent change in outcome measures. The Wilcoxon Rank Sum test was used to test for unadjusted differences between the Mg and placebo groups. Analysis of variance was used to adjust comparisons for age, BMI, and 6-week treatment compliance. All tests were performed at the a=0.05 level of statistical significance. The following variables were examined for between group changes: age, BMI, systolic blood pressure, diastolic blood pressure, total cholesterol, LDL cholesterol, HDL cholesterol, total triglycerides, Si, Sg, fasting insulin, AIRG, S; * AIRG, and IVGTT Kg l0- 4o , insulin IVGTT AUC 19-40, insulin IVGTT A U C 0 -2 4 0 , insulin OGTT A U C o - n o , and fasting Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6 glucose, glucose OGTT AUCq-uo- All variables were also compared for between group differences at baseline and the end of the study. Three binary compliance flags were created to indicate whether each subject had complete pill counts: one for the first 6 weeks, one for the second 6 weeks, and one for the total 12 weeks. Missing compliance rates were set to 0%. To adjust for overall compliance, the 12-week flag was included in the analysis of variance prediction model as well as the 12- week compliance rate. Similarly, to adjust for 6-week compliance, both the compliance flag and the corresponding 6-week compliance rate for both 6-week periods were included in the multivariate analysis. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7 CHAPTER III: RESULTS Recruitment and Randomization By design, 26 participants were recruited and randomly assigned to treatment, 13 to Mg group and 13 to placebo group. Twenty-four of the study participants completed the study, 13 from the Mg group and 11 from the placebo group. Baseline Comparison of Treatment Groups At baseline (Table 1), the women subsequently randomized to placebo were slightly older and had a higher BMI than women subsequently randomized to Mg, but the differences were of borderline statistical significance. The two groups were similar in their levels of blood pressure, lipids, glucose tolerance, insulin sensitivity, and all variables examined later as outcomes (Table 1). When the analysis was restricted to data from the 24 subjects who completed the study (Table 2), the placebo group had a significantly higher BMI than the Mg group (p=0.03). Additionally, the subgroup of women randomized to placebo who completed the study had a slightly lower glucose disappearance rate (Kg ) on baseline IVGTTs compared to Mg-treated women (Table 2). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission of the copyright owner. Further reproduction prohibited without permission. Table 1: Means, Medians, and Inter-Quartile-Ranges for Baseline Demographic and Outcome Variables by Treatment Group (All Randomized Subjects) P a r a m e te r Magnesium (n ■ = 13) Placebo (n^ 13) P-value Mean± S.D. Median Interquartile Range* Mean ± S.D. Median Interquartile Range* Age (years) 32.5 ± 4.6 31 2 9 ,3 6 37.3 ± 5.9 36 34,41 0.06 BMI (k g/m 2 ) 30.2 ±4.1 30.1 28.8,31.7 35.9 ±7.1 33.1 30.3,41.8 0.051 Systolic Blood Pressure (mmHg) 111 a b 11 108 105, 321 118± 14 120 111,125 0.24 Diastolic Blood Pressure (mmHg) 70 ± 6 69 6 7 ,7 3 68 ± 9 67 6 6 ,7 6 0.49 Total Cholesterol (mg / dl) 170 ± 40 173 147,195 177 ±35 175 151,190 0.77 LDL Cholesterol (m g/di) 106 ± 3 6 118 72,130 103 ±33 97 84,118 0.65 HDL Cholesterol (mg / dl) 37 ± 8 35 3 1 ,4 6 40 ± 9 41 3 3 ,4 8 0.32 Total Triglycerides (mg / dl) 136± 112 91 76,137 167 ±97 127 105,216 0.23 Si (min4 * pU4 • m l"1 • lO") 2.01 ± 1.27 1.63 1.33,2.12 1.72 ±0.93 1.52 0.89,2.41 0.57 SG(min4 -102 ) 1.34 ±0.34 1.30 1.17,1.41 1.15 ±0.41 1.24 1.04,1.34 0.48 Fasting Insulin (pU • ml'1 ) 26 ± 22 18 15,26 26 ± 19 18 14,37 0.98 AIRG (pU ■ min • ml'1 ) 419 ±318 357 192,375 331 ±190 243 197,436 0.76 Si* AIRG (ml'2 ) 677 ±355 611 424,813 504 ±272 438 339,601 0.23 IVGTT Kg 10-40 (min'1 • 102 ) -1.21 ±0.24 -1.19 -1.27,-1.09 -1.06 ±0.27 -1.08 -1.11,-0.90 0.14 Insulin IVGTT AUC 19-40 (pU • min • ml'1 ) 2129 ±856 2112 1542,2388 2061 ±924 1870 1403,2904 0.80 Insulin IVGTT AUC 0-240 (pU • min • ml4) 10929 ±5135 9642 7268,12757 11084 ± 6077 8126 7812,17765 0.88 Insulin OGTT AUC 0-120 (pU • min • ml"1 ) 10229 ±6226 8470 7635, 10950 9560 ± 5482 7590 5603,13959 0.73 Fasting Glucose (mg / dl) 99.0 ±17.0 102 89,105 106.4 ±12.0 109 102,118 0.24 Glucose OGTT AUC 0-120 (mg ■ min • dl'1 ) 24162 ±4219 24643 21073 , 26035 23393 ±2365 23079 21700,25251 0.61 AUC represents the total m m under the plasma concentration curve between the indicated time indexes. P wines determined using Wilcoxon Rank Sum test # 25% - 75% percentile 0 0 Reproduced w ith permission of the copyright owner. Further reproduction prohibited without permission. Table 2: Means, Medians, and Inter-Quartile-Ranges for Baseline Demographic and Outcome Variables by Treatment Group (Subjetfs Completing Trial) Parameter Magnesium (n = 13) Placebo (n = 11) P-value Mean± S.D. Median Interquartile Range* Mean ± S.D. Median Interquartile Range* Age (years) 32.5 ±4.6 31 2 9 ,3 6 36.3 ±5.8 35 3 4 ,3 9 0.16 BMI (kg /m 2 ) 30.2 ±4.1 30.1 28.8,31.7 37.0 ±7.2 34.7 30.3,42.0 0.03* Systolic Blood Pressure (mmHg) 111*11 108 105,121 115* 12 113 105,125 0.44 Diastolic Blood Pressure (mmHg) 70 ± 6 69 67 ,7 3 66± 9 67 5 9 ,6 9 0.17 Total Cholesterol (mg / dl) 170 ± 40 173 147,195 174 ± 38 171 149,186 0.93 LDL Cholesterol (mg / dl) 106 ± 36 118 72,130 103 ± 3 6 96 80,125 0.69 HDL Cholesterol (mg / dl) 37 ± 8 35 3 1 ,4 6 41 ± 9 41 3 5 ,4 9 0.24 Total Triglycerides (mg / dl) 136 ± 112 91 76,137 151 ± 87 116 103,208 0.45 Si (min'1 • ulT1 • ml"1 • iO4 ) 2.01 ± 1.27 1.63 1.33,2.12 1.59 ±0.96 1.14 0.85,2.27 0.26 SG (m in1 ■ 102 ) 1.34 ±0.34 1.30 1.17,1.41 1.08 ±0.41 1.19 1.01,1,28 0.18 Fasting Insulin (pU • ml'1 ) 26 ± 22 18 15,26 27 ± 19 18 14,40 0.82 AIRG (pU • min ■ ml'1 ) 419± 318 357 192,375 330 ±199 243 194 , 436 0.69 Si* AIRG (ml'2 ) 677 ±355 611 424,813 441± 175 420 3 0 8 ,557 0.13 IVGTT Kg 10-40 (min'1 • 103 ) -1.21 ±0.24 -1.19 -1.27,-1.09 -1.02 ±0.28 -1.01 -1.11,-0.88 0.06 Insulin IVGTT AUC 19-40 (jiU • min ■ ml'1 ) 2129 ±856 2112 1542 , 2388 2171 ±951 1883 1403,3026 0.95 Insulin IVGTT AUC 0-240 (pU • min • ml'1 ) 10929 ±5135 9642 7268,12757 12101 ±6076 8725 7818,18679 0.69 Insulin OGTT AUC 0-120 (pU • min • ml'1 ) 10229 ± 6226 8470 7635,10950 9885 ±5627 8513 5330,14615 0.91 Fasting Glucose (mg / dl) 99.0 ±17.0 •102 89,105 107.2 ±12.8 109 102,119 0.20 Glucose OGTT AUC 0-120 (mg • min • dl'1 ) 24162 ±4219 24643 2 1 0 7 3 ,26035 23.2 ±2.4 22.8 21.6,25.5 0.57 AUC represents the total area under the plasma concentration curve between the indicated time indexes. P values determined using Wilcoxon Rank Sum test. #25% -75% percentile * Significantly different using Wilcoxon Rank Sum test. 10 Compliance with Treatment Pill counts were not recorded for 2 subjects in each treatment group at 6 weeks and for 5 subjects in each treatment group at 12 weeks. Of the 24 subjects who completed the study, complete 12-week pill counts were available for only 7 subjects in the Mg group and 4 in the placebo group. For those with complete data, compliance with medication schedules averaged 97%±4% and 74%±9% for the Mg and placebo groups, respectively, showing a significant difference (p=G.03 between groups, Table 3) between treatment groups. This difference in compliance rates did not reach the criteria for statistical significance when calculated separately for the two 6-week periods between follow-up visits. During the first 6-week period, compliance rates averaged 98%±6% and 84%±35% for the Mg and placebo treatment groups, respectively (p=0. 11, Table 3). During the second 6-week period, compliance rates averaged 94%±6% and 79%±12% for the Mg and placebo groups, respectively (p=0.08, Table 3). Table 3: Treatment Compliance Rates by Treatment Group for Different Periods of Follow-up Study Period 504 mg/day Placebo P-value n = 11 n = 10 1 - 6 Weeks 98 ±6% 84 ±35% 0.11 n = 8 n = 6 7 - 1 2 Weeks 94 ±6% 79 ± 12% 0.08 n = 7 n = 4 1 - 1 2 Weeks * 97 ±4% 74 ±9% 0.03* Table lists means ± SD * Statistically Significant using Wilcoxon Rank Sum Test Changes in Insulin Sensitivity After 12 weeks, insulin sensitivity (Si) in the Mg treatment group decreased by 0.21±0.65(min'i-p.ir1 -mr1 -l(rt) compared to an average decrease of Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 11 0.1 l±0.58(min'1 -|xU'1 -ml'1 -KT4 ) in the placebo group (Table 4), but this difference was easily compatible with chance (p>0.5G). The percent changes in insulin sensitivity were likewise similar in the two groups (-2%±38% vs. +2%±33%, respectively; Table 5, p>0.50). Differences between treatment groups in both the absolute change and percentage change in Si remained non-significant after adjustment for age, BMI, and treatment compliance (Tables 6 and 7). Changes in Fasting Glucose Fasting glucose levels increased in both the Mg and placebo treatment groups; absolute increases were 12.3±35.2 and 3.7±13.6 (mg/dl) respectively, in the Mg and placebo groups (Table 4). However, the difference between these increases was compatible with chance (p>0.50, Table 4). Likewise, percent increases in fasting glucose did not differ significantly between groups (18%±45% and 3%±12%, respectively; p>0.50, Table 5). After adjustment for BMI, age, and treatment compliance, the absolute changes and percent changes remained non-significant (Tables 6 and 7). Changes in Fasting Insulin Fasting insulin levels fell by 4±19 and 3±6(pU',-nir1 ) in the Mg and placebo treatment groups, respectively (Table 4). However, fasting insulin levels showed a mean percent increase of 4%±49% in the Mg treatment group compared to a mean percent decrease of 8%±30% in the placebo group (Table 5). Neither the difference in absolute or percent changes between groups was statistically significant, before or after adjustments (Tables 6 and 7). Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f the copyright owner. Further reproduction prohibited without permission. Table 4; Means, Medians, and Inter-Quartile Ranges for Absolute Changes in Outcome Variables Between Baseline and 3-month End o f Study Visit (Subjects Completing Trial) Magnesium (n = >13) . Placebo* (n = 13) P-valuet Mean ± S.D. Median Interquartile Range* Mean±S.D. Median Interquartile Range* BMI (kg / m2 ). -0.8 ± 1.3 -0.4 -1.5,0.2 -0.2 ±1.7 0.2 -1.5,0.9 0.30 Systolic Blood Pressure (mmHg) -3 ± 6 -3 - 7 ,0 -2 ± 8 -4 - 9 ,2 1.00 Diastolic Blood Pressure (mmHg) -1 ± 8 -4 - 6 ,6 3 ± 7 1 - 1 ,8 0.14 Total Cholesterol (mg / dl) 9 ± 2 8 10 -1 1 ,2 9 -7 ±21 -11 -1 5 ,0 0.16 LDL Cholesterol (mg / dl) 3 ±27 1 -2 1 ,3 0 -10 ±28 -12 -2 1 ,4 0.59 HDL Cholesterol (mg / dl) 4 ± 8 5 - 4 ,9 -1 ± 7 0 4 , 3 0.22 Total Triglycerides (mg / dl) 11 ±63 14 - 1 7 ,26 16 ± 66 0 -2 3 ,4 7 0.98 Si (min'1 ■ pU'1 • ml'1 • 104 ) -0.21 ±0.65 -0.23 -0.59,0.09 -0.11 ±0.58 0.08 -0.26,0.12 0.65 S0 (min'1 • 102 ) 0.11 ±0.29 0.03 -0.07,0,26 0.08 ±0.42 0.09 -0.24,0.29 0.73 Fasting Insulin (pU • ml'1 ) -4 ± 19 0.0 -8 ,4 -3 ± 6 -5 -9 ,2 0.48 AIRG (pU • min • ml'1 ) 7 ±121 -5 -74,53 -5 ±83 4 -50,72 1.00 Si* AIRG (ml'2 ) 70 ±403 -114 -142,26 -16± 114 32 -129,51 0.55 IVGTT Kg 10-40 (min'1 ■ 102 ) 0.010±0.12 0.0 -0.013,0.112 0.024 ±0.21 0.0 0.026,0.059 0.89 Insulin IVGTT AUC 19-40 (|iU • min • ml'1 ) -33 ±423 -41 -181,226 30 ±464 -119 -196,197 0.73 Insulin IVGTT AUC 0-240 (JJ.U • min • ml'1 ) 126 ±3082 -129 -2074,1168 701 ±2608 408 -1300,2364 0.49 Insulin OGTT AUC 0-120 (pU ■ min • ml'1 ) -710 ± 3996 273 -4860,1540 74 ± 2729 -233 -2583,1598 0.73 Fasting Glucose (mg / dl) 12.3 ±35.2 -1 -8 ,1 5 3.7 ±13.6 4 - 1 ,8 0.83 Glucose OGTT AUC 0-120 (mg • min • dl'1 ) -732 ± 5628 -558 -4638,2073 -282 ±3329 -1370 -2220,2978 0.82 Table lists means ± SD [calculated as (3 month - baseline)], f Unadjusted p-value (using Wilcoxon Rank Sum test) # 25% -75% percentile N J Reproduced w ith permission o f the copyright owner. Further reproduction prohibited without permission. Tab!e5: Means, Medians, and Inter-Quartile Ranges for Percent Changes in Outcome Variables Between Baseline and 3-month End of Study Visit (Subjects Completing Trial) Parameter Magnesium (n °1 3 ) Placebo* (n m 11) P-valuet Mean ± S.D. Median Interquartile Range* Mean ± S.D. Median Interquartile Range* BMI (k g/m 2 ) -2 ± 5 -2 -5 ,1 -1 ± 5 1 - 4 ,3 0.32 Systolic Blood Pressure (mmHg) -3 ± 6 -2 - 6 ,0 -1 ± 7 -3 -7 ,1 0.98 Diastolic Blood Pressure (mmHg) -1± 11 -5 - 7 ,9 5 * 1 1 I -1 ,1 2 0.14 Total Cholesterol (mg / dl) 8 ± 2 0 5 -8 ,2 3 -3 ±11 -7 - 9 ,0 0.16 LDL Cholesterol (mg / dl) 10± 31 0 -18,43 -8 ±23 -14 -2 3 ,4 0.32 HDL Cholesterol (mg / dl) 12± 21 14 -1 0 ,2 6 -1 ± 15 0 - 8 ,8 0.24 Total Triglycerides (mg / dl) 23 ±51 19 -1 7 ,3 2 15 ± 49 0 -15,41 0.83 Si (m in1' pU'11 ml'1 x IQ4 ) -2.10 ±37.7 -17.9 -26.74,6.8 1.55 ±33.2 -7.5 -20.70,15.2 0.65 SG (min'1 x 103 ) 10.21 ± 27.5 2.8 -4.37,16.3 -0.97 ±22.7 4.2 -1 8 .0 3 ,13.6 0.52 Fasting Insulin (p U ' ml'1 ) 4 ± 4 9 0 -3 3 ,3 5 -8 ± 3 0 -18 -2 8 ,8 0.61 AIRG (pU 1 m in' ml'1 ) 13 ±48 -1 -1 3 ,2 8 8 ±25 2 - 1 6 ,34 0.91 Si * AIRG (ml'2 ) 9 ±61 -14 -2 8 ,8 5 ± 3 6 11 -20,11 0.65 IVGTT Kg 10-40 (n u n 1 x 10s) -0.94 ±9.4 0.0 -10.1,1.4 4.08 ±35.7 0.0 -6.7,3.0 0.93 Insulin IVGTT AUC 19-40 (pU ■ m in ' m l'1 ) 0 ± 2 2 -4 -8 ,1 3 11 ± 38 -6 -8 ,1 4 1.00 Insulin IVGTT AUC 0-240 (pU ■ m in' ml'1 ) 3 ± 2 6 -3 -18,11 10 ±24 5 -1 0 ,2 9 0.53 Insulin OGTT AUC 0-120 (pU ■ m in' ml'1 ) 0 ± 3 7 5 -2 6 ,1 9 0 ± 2 4 -10 -1 9 ,2 4 1.00 Fasting Glucose (mg / dl) 18 ± 45 -1 -8 ,1 7 3 ± 12 3 0 , 7 0.77 Glucose OGTT AUC 0-120 (mg • min ■ dl'1 ) -1 ± 22 -2 -1 8 ,1 0 -1 ± 15 -5 -12,13 0.86 Table lists means ± SD [calculated as 100 * (3 month - baseline) / baseline], t Unadjusted p-vatue (using Wilcoxon Rank Sum test) #25% -75% percentile Reproduced w ith permission o f the copyright owner. Further reproduction prohibited without permission. Table 6; P-values for Differences by Treatment Group for Absolute Changes in Medical and Outcome Variables, Unadjusted and Adjusted for Specified Unbalanced Baseline and/or On-trial Variables (Subjects Completing Trial) Parameter P-value4 P-valueb P-valuec P-value4 P-value* P-valns BMI (k g/m 2 ) 0.30 N/A 0.41 N/A 0.69 N/A Systolic Blood Pressure (mmHg) 1,00 0.23 0.99 0.39 0.92 0.50 Diastolic Blood Pressure (mmHg) 0.14 0.25 0.25 0.44 0.11 0.21 Total Cholesterol (mg / dl) 0.16 0.35 0.28 0.48 0.38 0.55 LDL Cholesterol (mg / dl) 0.59 0.87 0.67 0.85 0.60 0,77 HDL Cholesterol (mg / dl) 0.22 0,41 0.21 0.36 0.63 0.71 Total Triglycerides (mg / dl) 0.98 0.81 0.42 0.45 0.99 0.97 Si (min*1 • pU'1 • ml"1 • 104 ) 0.65 0.37 0.77 0.59 0.64 0.40 S0 (min4 -103 ) 0.73 0.74 0.84 0.78 0.48 0.57 Fasting Insulin (pU • ml'1 ) 0.48 0.72 0.72 0.86 0.22 0.32 AIRG (pU • min • ml'1 ) 1.00 0.88 0.65 0.68 0.052 0.10 Si* AIRG (ml'2 ) 0.55 0.34 0.66 0.30 0.27 0.23 IVGTT Kg 10-40 (min'1-102 ) 0.89 0.87 0.35 0.56 0.61 0.71 Insulin IVGTT AUC 19-40 (pU • min ■ ml'1 ) 0.73 0.65 0.55 0.69 0.91 0.86 Insulin IVGTT AUC 0-240 (pU • min • ml'1 ) 0.49 0.50 0.78 0.69 0.82 0.76 Insulin OGTT AUC 0-120 (pU ■ min • ml'1 ) 0.73 0.14 0.89 0.19 0.88 0.40 Fasting Glucose (mg / dl) 0.83 0.20 0.90 0.37 0.64 0.22 Glucose OGTT AUC 0-120 (mg ■ min • dl'1 ) 0.82 0.42 0.65 0.31 0.62 0.44 8 Unadjusted p-value (using Wilcoxon Rank Sum test) b P-value adjusted for BMI (using analysis o f variance) c P-value adjusted for age (using analysis of variance) d P-value adjusted for BMI, age (using analysis o f variance) e P-value adjusted for overall compliance (using analysis of variance) f P-value adjusted for BMI & overall compliance (using analysis of variance) * Significantly different by indicated test. Reproduced w ith permission o f the copyright owner. Further reproduction prohibited without permission. Table 6: Continued P-values for Differences by Treatment Group for Absolute Changes in Medical and Outcome Variables, Unadjusted and Adjusted for Specified Unbalanced Baseline and/or On-trial Variables (Subjects Completing Trial) Parameter P-value8 P-value1 1 P-value1 P-value1 P-valuek P-value BMI (kg/m 2 ) 0.73 N/A 0.93 N/A 0.92 N/A Systolic Blood Pressure (mmHg) 0.98 0.64 0.96 0.71 0.99 0.81 Diastolic Blood Pressure (mmHg) 0.20 0.30 0.12 0.21 0.20 0.31 Total Cholesterol (mg / dl) 0.52 0.68 0.48 0.63 0.61 0.74 LDL Cholesterol (mg / dl) 0.71 0.84 0.71 0.83 0.83 0.90 HDL Cholesterol (mg / dl) 0.42 0.57 0.32 0.33 0.29 0.36 Total Triglycerides (mg / dl) 0.73 0.71 0.99 0.99 0.67 0.53 Si (min'1 • pU4 • ml'1 • 104 ) 0.79 0.53 0.66 0.57 0.66 0.68 SQ(min4 -102 ) 0.66 0.69 0.003* 0.005* 0.007* 0.01* Fasting Insulin (pU • mi'1 ) 0.36 0.46 0.07 0.11 0.12 0.16 AIRG (pU * min * ml4) 0.11 0.18 0.051 0.09 0.07 0.09 Si* AIRG (ml'2 ) 0.41 0.28 0.18 0.20 0.29 0.28 IVGTT Kg 10-40 (min4 • 102 ) 0.54 0.69 0.10 0.15 0.12 0.19 Insulin IVGTT AUC 19-40 (pU • min • ml4) 0.87 0.90 0.75 0.76 0.53 0.64 Insulin IVGTT AUC 0-240 (pU • min • ml4) 0.92 0.82 0.51 0.51 0.63 0.65 Insulin OGTT AUC 0-120 (pU • min • ml4) 0.95 0.46 0.75 0.35 0.85 0.46 Fasting Glucose (mg / dl) 0.79 0.34 0.78 0.46 0.88 0.59 Glucose OGTT AUC 0-120 (mg • min • dl4) 0.64 0.31 0.78 0.68 0.82 0.61 g P-value adjusted for age & overall compliance (using analysis o f variance) h P-value adjusted for age, BMI, & overall compliance (using analysis of variance) i P-value adjusted for 6 wk compliance (using analysis of variance) j P-value adjusted for BMI & 6 wk compliance (using analysis of variance) k P-value adjusted for age & 6 wk compliance (using analysis o f variance) 1 P-value adjusted for BMI, age, & 6 wk compliance (using analysis o f variance) * Significantly different by indicated test. T able 7 16 £ 3 ! >1 ^ 000000^0000000000 0 * 1 n h - ' ^ ^ 8 o c 8 r t \ 0 s n « 8 e ’t f n \ o M ' © o r ^ © © w c ^ r > i « o o \< 3 0 f n c ^ o c ^ ® n © ^ e * - ' 0 5 !^ ^ i oooooooooooooooooo < £ r* ~ o «n t* - 2 W n f- * © ^ O O O © © C S .......... _. m « n ^ s © s s -‘ t-*^'*3"V © r'-«-*frneo ooooooooooo 5 g s s? s a sss sssK g sss oooooooooooooooooo ^ o o o ^ n a s m v D O s o ^ c ^ a s O ^ D ^ r ^ e o r ^ jcr^«^icnv>-?t-«i©€^'©i^os€,<iQK?*-«no*-^^ Oooooooooooooooooo ^ ^ m O s t* © © © c * * ^ S * m m s n i M ^ ^ s n e n o c n o s ^ v © K>\om%D@s,ooso«oor^e© o o o o o o o ® » - o ^ o o 1 1 i b i t - 6 £ 'js s i l l pq w 5 c ; Pw a. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f the copyright owner. Further reproduction prohibited without permission. Table 7: Continued P-values for Differences by Treatment Group for Percent Changes in Medical and Outcome Variables, Unadjusted and Adjusted for Specified Unbalanced Baseline and/or On-trial Variables (Subjects Completing Trial) Parameter P-value® P-value1 1 P-value1 P-value1 P-vaIuek P-value BMI (k g /m 2 ) 0.75 N/A 0.94 N/A 0.92 N/A Systolic Blood Pressure (mmHg) 0.95 0.68 0.96 0.76 0.99 0.85 Diastolic Blood Pressure (mmHg) 0.19 0.30 0.19 0.30 0.30 0.42 Total Cholesterol (mg / dl) 0.43 0.59 0.51 0.66 0.62 0.75 LDL Cholesterol (mg / dl) 0.37 0.51 0.38 0.53 0.53 0.66 HDL Cholesterol (m g/dl) 0.43 0.59 0.33 0.39 0.31 0.41 Total Triglycerides (mg / dl) 0.67 0.68 0.99 0.99 0.63 0.54 Si (miff1 • gU*1 • ml*1 x 104 ) 0.92 0.56 0.56 0.44 0.63 0.58 S0 (miff1 x 10*) 0.53 0.63 0.005* 0.009* 0.01* 0.02* Fasting Insulin (p U ' ml'1 ) 0.38 0.47 0.12 0,16 0.20 0.25 AIRG (p U ' min ■ ml*1 ) 0.16 0.26 0.06 0.10 0.07 0.10 Si* AIRG (ml*2 ) 0.44 0.29 0.26 0.24 0.38 0.32 IVGTT Kg 10-40 (miff1 x 102 ) 0.45 0.60 0.08 0.14 0.09 0.15 Insulin IVGTT AUC 19-40 fttU' min • ml*1 ) 0,84 0.90 0.90 0.91 0.72 0.82 Insulin IVGTT AUC 0-240 (jiU ■ m in' ml*1 ) 0.89 0.82 0.58 0.60 0.67 0.71 Insulin OGTT AUC 0-120 (p U ' m in • ml*1 ) 0,67 0.27 0.89 0.31 0.87 0.43 Fasting Glucose (mg / dl) 0.73 0.27 0.80 0.46 0.89 0.58 Glucose OGTT AUC 0-120 (m g' min ■ dl*1 ) 0.55 0.29 0.70 0.63 0.76 0.61 g P-value adjusted for age & overall compliance (using analysis of variance) h P-value adjusted for age, BMI, & overall compliance (using analysis of variance) i P-value adjusted for 6 wk compliance (using analysis of variance) j P-value adjusted for BMI & 6 wk compliance (using analysis o f variance) k P-value adjusted for age & 6 wk compliance (using analysis of variance) 1 P-value adjusted for BMI, age, & 6 wk compliance (using analysis o f variance) * Significantly different by indicated test. 18 Changes in Glucose Effectiveness Absolute values of glucose effectiveness (Sg ) increased 0.11±0.29 and 0.08±0.42 (min *T02 ) in the Mg and placebo groups, respectively (p>0.50, Table 4). These values corresponded to percent changes from baseline of 10%±28% and -1%±22%, respectively (p>0.50, Table 5). Adjustment for differences in treatment compliance (in 6-week intervals), revealed first, that the absolute increase observed in the Mg group was significantly higher than the increase observed in the placebo group (p=0.0Q3, Table 6), and second, that the percent increase in Sg in the Mg group was significantly different from the small decrease in the placebo group (p=0.005, Table 7). This adjustment yielded predicted mean absolute changes in glucose effectiveness levels of 0.21 and 0.11 (min'Mo2 ) in the Mg and placebo groups, respectively (Table 8). Table 8 Predicted Adjusted Means for Selected Outcome Variables by Treatment Group (Subjects Completing Trial) Parameter _____ Magnesium (n = 13)____ Placebo (n = 11) Mean* M ean** Mean3 Meanb Sofm in’x 102 ) 0.206 0.189 0.105 0.116 AIRG (pU min ml'1 ) 73 67 39 41 IVGTT Kg 10-40 (min4 x lO 2 ) 0.013 0.0035 0.023 0.022 a Predicted mean adjusted for 6 wk compliance b Predicted mean adjusted for BMI, age, & 6 wk compliance Additional adjustments for BMI and age tended to attenuate the p-values associated with these between group differences, but both adjusted measures of change in glucose effectiveness remained statistically significant, p=0.005 and p=0.009 for absolute and percent changes respectively. This attenuation is illustrated by the predicted mean absolute changes in glucose effectiveness levels of 0.19 and 0.12 (min1 TO2 ) for the Mg and placebo groups respectively (Table 8). When overall 12-week compliance was substituted for compliance in 6-week intervals, none of these differences reached statistical Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 19 significance. However, at the end of this 12-week pilot study, both the average glucose effectiveness and the average rate of glucose disappearance (IVGTT Kg) were significantly higher in the Mg treatment group than in the placebo group (Table 9). Changes in Other Outcomes Differences between the changes in AIRG, insulin IVGTT AUCo-240, insulin OGTT AUCo-120, and all other outcome variables were compatible with chance, whether based on absolute or percentage changes and whether adjusted for age, BMI, and treatment compliance or not (Table 4, Table 5, Table 6, Table 7). However, AIRG did show marginally significant absolute (p=0.051, Table 6) and percent changes (p=0.06, Table 7) between treatment groups after adjustment for treatment compliance in 6-week intervals. Comparison of Treatment Groups at End of Study At the end of this pilot study, both treatment groups were similar in AIRG, insulin IVGTT AUCo-24o , insulin OGTT AUCo-120, and most outcome variables. However, the placebo group maintained a significantly higher BMI than the Mg group (31±42 vs. 29±4kg/m2 respectively; p=0.01). In addition, the Mg treatment group had a higher Sg (1.44±0.41 vs. 1.15±0.27, p=0.048) and larger Kg i0-40 (-1.20±0.29 vs. -1.00±0.2Q, respectively, p=0.04; Table 9). Adjustment for age, baseline BMI, and treatment compliance (in 6-week intervals) attenuated the inter-group p-values to 0.20 and 0.10 for Sg and Kgl0j} o respectively. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced w ith permission o f the copyright owner. Further reproduction prohibited without permission. Table 9 Means, Medians, and Isiter-Quartile-Ranges for End-of-Study Demographic and Outcome Variables by Treatment Group (Subjects Completing Trial) __ ....... ........ ........ _ _ Parameter Magnesium (n 3513) Placebo (n ■ » 11) P-value Mean ± S.D. Median Interquartile Range* Mean± S.D. Median Interquartile Reap* Age (years) 32.5 ±4.6 31 2 9 ,3 6 36.3 ± 6 35 3 4 ,3 9 0.16 BMI (kg / m2 ) 29.4 ± 3 .9 29.1 27.3,30.9 36.9 ±7.1 35.0 30.7,41.5 0.01* Systolic Blood Pressure (rnmHg) 108 ± 8 106 103,114 114± 10 114 106,123 0.16 Diastolic Blood Pressure (mmHg) 69 ± 8 69 6 4 ,7 0 70 ± 10 68 6 3 ,7 8 0.86 Total Cholesterol (mg / dl) 111 ±38 175 157,212 170 ±32 165 141,207 0.52 LDL Cholesterol (mg / dl) 110 ± 27 112 91,126 96 ± 3 0 104 76,115 0.31 BDL Cholesterol (mg / dl) 41 ± 12 37 3 3 ,4 7 40 ± 10 38 3 5 ,4 6 0.95 Total Triglycerides (mg / dl) 147 ±91 107 92,204 167 ±94 155 97,185 0.40 Si(min‘1 • pU4 ' m l'1 x Iff*) 1.80 ±0.97 1.57 1.30,1.94 1.48 ±0.68 1.44 0.86,2.20 0.52 So (min4 x 102 ) 1.44 ±0.41 1.32 1.25,1.46 1.15 ±0.27 1.10 1.03,1.37 0.048* Fasting Irauiin (pU ■ ml’1 ) 22 ± 15 16 13,29 24 ± 18 14 12,33 0.86 AIRG (p U ' m in' ml'1 ) 426 ±274 318 240,485 325 ±137 271 229,369 0.61 Si * AIRG (ml4) 746 ± 640 517 391,623 429 ±122 472 329,516 0.18 IVGTT Kg 10-40 (min4 x 102 ) -1.20 ±0.29 -1.09 -1.24,-1.03 -1.00 ±0.20 -0.91 -1.11,-0.86 0.04* Insulin IVGTT AUC 19-40 (p U ' min ■ ml'1 ) 2096 ±892 2000 1515,2383 2201 ±773 2012 1529 , 2945 0.73 Insulin IVGTT AUC 0-240 (pU ■ m in• ml'1 ) 11055 ±6007 9627 8085,11634 12803 ± 6056 10480 8219,17643 0.57 Insulin OGTT AUC 0-120 (g U ' min ■ ml4) 9519 ± 6021 7073 5010,10793 9958 ± 6849 6938 5698, 12033 0.86 Fasting Glucose (mg / dl) 111.3 ±26.9 101 93,134 111.1 ±19.6 108 99,123 0.69 Glucose OGTT AUC 0-120 (mg ■ m in' dl4) 23430 ± 5240 22970 21123,24085 22971 ±4304 22975 21380,25575 0.82 AUC represents the total area under the plasma concentration curve between the indicated time index®. P values determined using Wilcoxon Rank Sum test. # 25% - 75% percentile * Significantly different using Wilcoxon Rank Sum test. to o 21 CHAPTER IV: DISCUSSION This study was conducted to examine potential effects of magnesium supplementation on insulin sensitivity and glucose response in women with a history of GDM and an increased risk of developing T2DM. For the primary outcome measures of interest, insulin sensitivity and glucose tolerance, no significant changes were observed among treatment groups. The only outcome for which significant differences were observed between the treatment groups was glucose effectiveness, with the Mg group showing greater improvement than the placebo group. Although there was not a significant change between treatment groups in the rate of glucose disappearance (Kg) until the differences were adjusted for differences in treatment compliance, the groups were significantly different on this variable at the conclusion of the study. The findings of this pilot study suggest that Mg supplementation may not be an effective method of improving insulin sensitivity or glucose tolerance in individuals at risk for developing T2DM (neither outcome showed statistical improvement), but several limitations of this study should be noted. First, this pilot study involved only 26 patients, which means the power to detect differences in the outcome measures was diminished. In fact, a treatment effect >45% needed to be observed to detect a statistically significant difference in insulin sensitivity with reasonable power. Second, since data on magnesium levels were not available at the time of this analysis, either before or after treatment, it was not possible to determine the extent to which Mg supplementation altered serum Mg levels. Third, patients for this pilot trial were not selected on the basis of low baseline magnesium levels, which may be necessary in order for supplementation to be useful. Much of the previous diabetes research with magnesium supplementation has focused on diabetic patients Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 22 known to have hypomagnesemia, which could have been a consequence of their diabetes rather than a risk factor for developing it. Thus, patients with normal to high levels of serum Mg may not benefit from magnesium supplementation. Finally, 12 weeks of magnesium supplementation may not have been long enough for a benefit to be observed. Previous research has demonstrated that at least 3 months of magnesium supplementation were required to achieve normal blood serum levels of magnesium in patients with hypomagnesemia (5, p. 189; 6, p. 2031), and 6 months was necessary for these treatment effects to wear off when magnesium supplementation was stopped after normal levels were attained (5. p. 190). Even though compliance rates were not consistently monitored in all subjects, and despite the fact that there were significant differences between treatment groups for those compliance rates that were recorded, higher compliance rates, when recorded, occurred in the magnesium treatment group rather than the placebo group. This imbalance provides reasonable confidence, in the absence of measurements of circulating Mg levels, that subjects in the Mg group actually received the intended treatment. Moreover, the fact that our finding on changes in glucose effectiveness was not apparent until it was adjusted for the most specific compliance data available (i.e. the 6-week data) lends further credibility to our measures of compliance as reflection of true magnesium supplementation. Additional evidence that magnesium supplementation potentially has an observable treatment effect, although not for the primary outcomes of this study, may be found by examining the percent changes in glucose effectiveness stratified by treatment group and treatment compliance. Increases in glucose effectiveness were observed for subjects with complete compliance data in both the Mg and placebo treatment groups; however, the Mg Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 23 group showed a larger overall increase and included the subject with the largest increase in glucose effectiveness. In contrast, the placebo group’s increase was smaller and included the subject with the largest decrease in glucose effectiveness. This separation of outlying data contributed to the fact that for the placebo group, the percent change in glucose effectiveness showed a mean decrease, while the absolute change showed a mean increase. Since subjects were not asked at the end of this pilot study to guess whether they had been assigned to the magnesium or placebo groups, we do not know if their blinding was maintained throughout the trial. However, it is difficult to explain how measures of glucose effectiveness could have been influenced by patient knowledge or beliefs about their treatment group. There was no question about the blinding of the laboratory analyzing the blood specimens, since no group identifier was included with the samples. Even though no significant changes were observed in the primary outcomes of insulin sensitivity and glucose tolerance, the results of this study provide guidance for future studies that might examine the usefulness of magnesium supplementation in delaying or preventing the onset of T2DM. The significant improvement in glucose effectiveness and rate of glucose disappearance in the Mg treatment group at the end of the study suggest that Mg supplementation may have observable effects. Whether Mg supplementation is effective in preventing T2DM remains to be established, and may depend on whether hypomagnesemia is a precursor or a symptom of T2DM. Future studies should begin with a larger sample of participants known to have low baseline levels of Mg, and continue Mg supplementation over a six to twelve month period. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 24 WORKS CITED 10. Balon, Thomas W., et al. “Magnesium Supplementation Reduces Development of Diabetes in a Rat Model of Spontaneous NIDDM,” American Journal of Physiology 269 (1995): E745-E752. 13. Berkowitz, Kathleen, et al. “Effect of Troglitazone on Insulin Sensitivity and Pancreatic P-Cell Function in Women at High Risk for NIDDM.” Diabetes 45 (1996): 1572-1579. 4. Buchanan, Thomas A., et al. “Gestational Diabetes: Antepartum Characteristics That Predict Postpartum Glucose Intolerance and Type 2 Diabetes in Latino Women.” Diabetes 47 (1998): 1302-1309. 2. Domhorst, Anne and Michela Rossi. “Risk and Prevention of Type 2 Diabetes in Women With Gestational Diabetes.” Diabetes Care 21 (1998): B43-B47. 11. Eibl, Nicole L., et al. “Hypomagnesemia in Type II Diabetes: Effect of a 3-Month Replacement Therapy.” Diabetes Care 18 (1995): 188-191. 14. Eibl, Nicole L., Christoph Schnack, and Guntram Schernthaner. “Magnesium Supplementation in Type 2 Diabetes.” Diabetes Care 21 (1998): 2031-2032. 3. Gregory, Kimberly D., Siri L. Kjos, and Ruth K. Peters. “Cost ofNon-Insulin- Dependenl Diabetes in Women With a History of Gestational Diabetes: Implications for Prevention.” Obstetrics & Gynecology 81 (1993): 782-786. 6. Kao, W. H. Linda, et al. “Serum and Dietary Magnesium and the Risk for Type 2 Diabetes Mellitus.” Archives of Internal Medicine 159 (1999): 2151-2159. 8. Lima, Maria De Lourdes, et al. “The Effect of Magnesium Supplementation in Increasing Doses on the Control of Type 2 Diabetes.” Diabetes Care 21 (1998): 682-685. 12. McCarty, M. F. “Toward Practical Prevention of Type 2 Diabetes.” Medical Hypotheses 54 (2000): 786-793. 1. Minino, Arialdi M., and Betty L. Smith. “Deaths: Preliminary Data for 2000.” National Vital Statistics Reports 49 (2001): 1-5. 7. Paolisso, Giuseppe, et al. “Improved Insulin Response and Action by Chronic Magnesium Administration in Aged NIDDM Subjects.” Diabetes Care 12 (1989): 265-268. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 25 9. Resnick, L. M., et al. “Intracellular and Extracellular Magnesium Depletion in Type 2 (Non-Insulin-Dependent) Diabetes Mellitus.” Diabetologia 36 (1993): 767-770. 5. Rosolova, Hana, Otto Mayer, Jr., and Gerald Reaven. “Effect of Variations in Plasma Magnesium Concentration on Resistance to Insulin-Mediated Glucose Disposal in Non-Diabetic Subjects.” Journal of Clinical Endocrinology and Metabolism 82 (1997): 3783-3785. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Comparisons of metabolic factors among gestational diabetes mellitus probands, siblings and cousins

PDF

Predictive value of CT coronary artery calcium scanning for coronary heart disease in asymptomatic subjects with diabetes mellitus

PDF

Polymorphisms in genes involved in steroid hormone metabolism and mammographic density changes in women randomized to menopausal estrogen and progesterone therapy

PDF

Validation of serum cotinine as a biomarker of environmental tobacco smoke exposure: Validation with self-report and association with subclinical atherosclerosis in non-smokers

PDF

Predictors of mammography use among California teachers

PDF

The influence of family structures on adolescent smoking among multicultural adolescents in Hawaii

PDF

Recreational physical activity and risk of breast cancer: The California Teachers Study

PDF

The effects of tobacco smoke on respiratory symptoms among young adults in the Children's Health Study II

PDF

Risk factors for diabetic retinopathy in Latinos: Los Angeles Latino eye study

PDF

Cigarettes and alcohol in relation to colorectal cancer within the Singapore Chinese Health Study

PDF

The association between recreational physical activity and mammographic density

PDF

Progression of carotid intima-media thickness and plasma antioxidants: The Los Angeles Atherosclerosis Study

PDF

Selective laser trabeculoplasty for the treatment of glaucoma

PDF

Effect of hormone therapy on the progression of carotid-artery atherosclerosis in postmenopausal women with and without established coronary artery disease

PDF

P53 and bladder cancer outcome: A combined analysis from the Keck School of Medicine

PDF

Post-intensive care unit mechanical ventilation: Relationship of infections to outcomes of weaning from prolonged mechanical ventilation

PDF

Extent, prevalence and progression of coronary calcium in four ethnic groups

PDF

Association between body mass and benign prostatic hyperplasia in Hispanics: Role of steroid 5-alpha reductase type 2 (SRD5A2) gene

PDF

Development and evaluation of standardized stroke outcome measures in a population of stroke patients in rural China

PDF

A case-control study of passive smoking and bladder cancer risk in Los Angeles

Asset Metadata

Creator Hass, Christopher David (author)

Core Title Metabolic effects of magnesium supplementation in women with a history of gestational diabetes mellitus

Degree Master of Science

Degree Program Applied Biostatistics and Epidemiology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag biology, biostatistics,health sciences, nutrition,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-307950

Unique identifier UC11336902

Identifier 1417922.pdf (filename),usctheses-c16-307950 (legacy record id)

Legacy Identifier 1417922.pdf

Dmrecord 307950

Document Type Thesis

Rights Hass, Christopher David

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 au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA