Table 1.
Reported associations of 25(OH)D with insulin secretion and sensitivity/resistance. To convert nmol/L to ng/mL, divide by 2.496. 25(OH)D: 25-hydroxyvitamin D; OGTT: oral glucose tolerance test; T2DM: type 2 diabetes mellitus; NHANES III: National Health and Nutrition Examination Survey 1988–1994; CA: Caucasian American; AA: African American; MA: Mexican American; HOMA-β: homeostasis model assessment of β-cell function; HOMA-IR: homeostasis model assessment of insulin resistance; QUICKI: quantitative insulin sensitivity check index; HbA1c: glycosylated hemoglobin; WHR: waist-to-hip ratio; BIA: bioelectrical impedance.
| Author, year (ref) | Subjects' characteristics | Mean serum 25(OH)D | Method to determine insulin outcomes | Association of 25(OH)D with insulin outcome | Covariates | Major limitations |
|---|---|---|---|---|---|---|
| (a) Insulin Secretion | ||||||
|
| ||||||
| Boucher 1995 [41] | 44 glucose-intolerant East London Asians, mean age 44.9 years, mean BMI: 25.9 kg/m2 | <27.5 nmol/L | OGTT | Positive association with postchallenge C-peptide and insulin | Age, sex, BMI | No measure of dietary intake or physical activity; indirect measures of adiposity and insulin secretion |
| Orwoll et al. 1994 [42] | 35 adults with T2DM, mean age: 61 years, mean BMI: 29.8 kg/m2 (ethnicity not reported) | 35 ± 7 nmol/L | Meal challenge | No association with fasting or postchallenge glucose, insulin, C-peptide, or glucagon | — | Indirect measure of insulin secretion; no adjustment for confounders |
| Scragg et al. 2004 [5] | 6228 NHANES III participants (Caucasian, African, and Mexican American), ages ≥ 20 years | 79.6 ± 36.8 nmol/L (CA) 49.1 ± 37.5 nmol/L (AA) 66.0 ± 41.5 nmol/L (MA) | Fasting glucose and insulin (HOMA-β) | No association with HOMA-β | Age, sex, BMI, physical activity, season, HOMA-IR | No measure of dietary intake; indirect measures of adiposity and insulin secretion |
| Baynes et al. 1997 [4] | 142 Dutch men, mean age: 75.7 years | 42 ± 29.1 nmol/L | OGTT | Inverse association with postchallenge glucose and insulin | Season, physical activity, BMI, skinfolds, smoking, alcohol, and fish, fat, and oil consumption | No adjustment for calcium consumption; indirect measure of adiposity |
| Chiu 2004 et al. [43] | 126 glucose-tolerant Asian, African, Caucasian, and Mexican American, mean age 26 ± 6 years, mean BMI: 24.7 kg/m2 | 46.9 nmol/L (Asian American), 47.3 nmol/L (AA), 69.4 nmol/L (CA), 50.2 nmol/L (MA) | Hyperglycemic clamp | Inverse association with first- and second-phase insulin response (not significant after adjustment for insulin sensitivity index) | Age, sex, ethnicity, BMI, WHR, blood pressure, season, insulin sensitivity index | No measure of dietary intake or physical activity; indirect measure of adiposity |
| Kamycheva et al. 2007 [40] | 15 Norwegian subjects with secondary hyperparathyroidism, mean age: 62.9 years, mean BMI: 27.1 kg/m2; 15 sex-, age-, and BMI-matched controls | 58.4 nmol/L (patients) 61.9 nmol/L (controls) | Hyperglycemic clamp | Inverse association with 2nd phase insulin secretion | Sex, age, BMI | No measure of dietary intake or physical activity; indirect measure of adiposity |
|
| ||||||
| (b) Insulin sensitivity/resistance | ||||||
|
| ||||||
| Scragg et al. 2004 [5] | 6228 NHANES IIIparticipants (Caucasian, African, and Mexican American), ages ≥ 20 years | 79.6 ± 36.8 nmol/L (CA) 49.1 ± 37.5 nmol/L (AA) 66.0 ± 41.5 mol/L (MA) | Fasting glucose and insulin (HOMA-IR) | Inverse association with HOMA-IR in Caucasian ( P = .058) and Mexican Americans (P = .002) | Age, sex, BMI, physical activity, season | No measure of dietary intake; indirect measures of adiposity and insulin sensitivity |
| Chonchol and Scragg 2007 [45] | 14697 NHANES III participants (Caucasian, African, and Mexican American), ages ≥ 20 years | 78 ± 58.4 nmol/L (men) 71.6 ± 61.5 nmol/L (women) | Fasting glucose and insulin (HOMA-IR) | Inverse association with fasting insulin and HOMA-IR | Age, sex, ethnicity, BMI | No measure of dietary intake or physical activity; indirect measures of adiposity and insulin sensitivity |
| Lu et al. 2009 [48] | 3262 Chinese adults (Nutrition and Health of Aging Population in China study), ages 50–70 years | 40.4 nmol/L | Fasting insulin and glucose (HOMA-IR) | Inverse association with fasting glucose, HbA1c, fasting insulin, and HOMA-IR. | Age, sex, residence, season, education, physical activity, smoking, alcohol, family history, CRP, IL-6, BMI | No measure of dietary intake; indirect measures of adiposity and insulin sensitivity |
| Gannagé-Yared et al. 2009 [47] | 381 Lebanese university students; ages 18–30 years; mean BMI: 23.9 ± 4.1 kg/m2 | 77.4 ± 31.2 nmol/L | Fasting insulin and glucose (HOMA-IR) | Inverse association with fasting glucose, fasting insulin, and HOMA-IR. Only glucose held after adjustment for all confounders. | Sex, BMI, exercise | No measure of dietary intake; indirect measures of adiposity and insulin sensitivity |
| Clifton-Bligh et al. 2008 [49] | 307 pregnant women of various ethnicities, mean age: 32.6 years | 53.8 ± 23.9 nmol/L | Fasting insulin and glucose (HOMA-IR) | Inverse association with fasting insulin, glucose, and HOMA-IR but not after adjustment for confounders | Age, BMI, ethnicity | No measure of dietary intake or physical activity; indirect measures of adiposity and insulin sensitivity |
| Alemzadeh et al. 2008 [49] | 127 children (Caucasian, Mexican, and African American), mean age: 13 years, BMI > 95th percentile for age | 59.9 ± 23.2 nmol/L | Fasting insulin and glucose (QUICKI), HbA1c | Inverse association with HbA1c, but not after adjustment for confounders | Fat mass (BIA), age, sex, ethnicity, season | No measure of dietary intake or physical activity; indirect measures of adiposity and insulin sensitivity |
| McGill et al. 2008 [46] | 243 adults of various ethnicities (New Zealand), mean age: 47.6 years, mean BMI: 35.4 kg/m2 | 62.2 ± 22.7 nmol/L | Fasting glucose and HbA1c | Inverse association with HbA1c and glucose (negated after removal of 3 outliers) | Sex, age, ethnicity, season, BMI | No measure of dietary intake or physical activity; indirect measures of adiposity and insulin sensitivity |
| Ford et al. 2005 [53] | 8421 NHANESIII participants, ages ≥ 20 years, multiple ethnicities | 74 nmol/L (range: 8.7–227.9 nmol/L) | Fasting glucose | Inverse association with fasting glucose | Age, sex, ethnicity, education, smoking, cotinine, cholesterol, abdominal obesity, triglyceridemia, low HDL, high blood pressure | No measure of dietary intake or physical activity; indirect measures of adiposity and insulin sensitivity |
| Need et al. 2005 [63] | 753 postmenopausal Caucasian (Australian) women, mean age: 63 years, mean BMI: 26.5 kg/m2 | 62 ± 24.3 nmol/L | Fasting glucose | Inverse association with fasting glucose | Age, BMI | No measure of dietary intake or physical activity; indirect measures of adiposity and insulin sensitivity |
| Hyppönen et al. 2008 [64] | 6810 British Caucasian adults, age: 45 years, BMI variable | 53.8 nmol/L (men) 51.5 nmol/L (women) | HbA1c | Inverse association with HbA1c | Sex, month, IGF-1, physical activity, smoking, alcohol, social class, abdominal obesity, BMI | No measure of dietary intake; indirect measures of adiposity and insulin sensitivity |
| Baynes et al. 1997 [4] | 142 Dutch men, mean age: 75.7 years | 42 ± 29.1 nmol/L | OGTT | Inverse association with fasting insulin | — | — |
| Chiu et al. 2004 [43] | 126 glucose-tolerant Asian, African, Caucasian, and Mexican American, mean age 26 ± 6 years, mean BMI: 24.7 kg/m2 | 46.9 nmol/L (Asian American), 47.3 nmol/L (AA), 69.4 nmol/L (CA), 50.2 nmol/L (MA) | OGTT Hyperglycemic clamp | Inverse association with postchallenge glucose; Positive association with insulin sensitivity index | Age, sex, ethnicity, BMI, WHR, blood pressure, season | No measure of dietary intake or physical activity; indirect measure of adiposity |
| Kamycheva et al. 2007 [40] | 15 Norwegian subjects with secondary hyperparathyroidism, mean age: 62.9 years, mean BMI: 27.1 kg/m2 and 15 sex-, age-, and BMI-matched controls | 58.4 ± 15.3 nmol/L (patients) 61.9 ± 18.5 nmol/L (controls) | Hyperglycemic clamp | Positive association with insulin sensitivity index among all subjects | Sex, age, BMI | No measure of dietary intake or physical activity; indirect measure of adiposity |
| Lind et al. 1995 [51] | 34 Caucasian men, mean age: 63 years | 90 ± 19 nmol/L | Euglycemic clamp | Positive association with insulin sensitivity, inverse association with fasting insulin | Age, BMI, WHR, serum creatinine | No measure of dietary intake or physical activity; indirect measure of adiposity |
| Lind et al. 1989 [52] | 10 Danish men with impaired glucose tolerance, ages 60–63 years | 107 ± 62 nmol/L | Euglycemic clamp | Positive association with insulin sensitivity | Serum calcium, phosphate, magnesium, and other minerals | No measure of dietary intake or physical activity; no adjustment for adiposity |
| Manco et al. 2005 [59] | 116 Caucasian, morbidly obese (mean BMI: 48.8 kg/m2) women before and after bariatric surgery, ages 20–35 years | Pre-surgery: 39.2 ± 22.3 nmol/L, 5 years postsurgery: 27.4 ± 16.4 nmol/L, 10 years postsurgery: 25.1 ± 13.9 nmol/L | Euglycemic clamp | No association with insulin sensitivity before or after bariatric surgery | Triglycerides, fat mass, PTH, serum calcium, phosphorus, HDL, LDL, total cholesterol | No measure of dietary intake or physical activity |