| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
Department of Obstetrics and Gynecology, University of Medicine and Dentistry of New Jersey, School of Osteopathic Medicine, Stratford, New Jersey 08084
Address all correspondence and requests for reprints to: Dr. Theresa O. Scholl, Department of Obstetrics and Gynecology, University of Medicine and Dentistry of New Jersey, School of Medicine, 2 Medical Drive, Science Center, Suite 185, Stratford, New Jersey 08084. E-mail: scholl{at}umdnj.edu.
Abstract
There are ethnic differences in insulin secretion and resistance in healthy nondiabetic adults, children, and adolescents. It is not known whether these ethnic differences are also detectable during normal pregnancy. The objective of this study was to examine whether ethnic differences in glucose homeostasis (C-peptide/insulin/glucose dynamics) are present in nondiabetic pregnant women. Fasting serum C-peptide, insulin, and plasma glucose were measured in the second and third trimesters in 773 pregnant women (343 African-Americans, 312 Hispanics, and 118 Caucasians), and a 50-g oral glucose challenge test was performed in the third trimester. Significantly reduced C-peptide levels and C-peptide to insulin ratio and elevated fasting insulin to glucose ratios were observed in African-American women compared with Caucasians and/or Hispanics. Similar results were found after a 50-g glucose load. In addition, African-Americans had greater insulin and lower glucose levels at glucose challenge test. There were ethnic differences in insulin production and resistance in both fasting and glucose-stimulated conditions in normal young nondiabetic pregnant women.
PREVIOUS STUDIES IN healthy nondiabetic adults have reported that there are ethnic differences in carbohydrate metabolism that have the potential to be linked to complications during pregnancy, decreased infant birth weight, as well as increased risks of obesity and type 2 diabetes in later life (1, 2, 3, 4). Nondiabetic African-Americans have higher levels of fasting insulin, elevated insulin, and C-peptide in response to a glucose load and lower insulin sensitivity compared with Caucasians (5, 6, 7).
The differences in insulin secretion and sensitivity are detectable during childhood and adolescence (8, 9, 10). The Bogalusa Study (11, 12) reported that African-American children, girls in particular, had changes consistent with increased insulin resistance: elevated levels of insulin, an increased insulin/glucose ratio, a decreased ratio of C-peptide to insulin, and lower postprandial glucose levels. Although most studies compare African-Americans to Caucasians, some suggest that nondiabetic Hispanics have a higher insulin response to a glucose load than Caucasians, which also implies greater insulin resistance in Hispanics (6, 13).
It is not known whether these ethnic differences are also detectable during pregnancy. Late pregnancy is characterized by accelerated fetal growth, rising concentrations of several diabetogenic hormones, progressive increases in the nutrient-stimulated insulin response, and insulin resistance (14, 15). Ethnic comparisons have been limited to a single measurement, mostly the 1 h glucose concentration after the 50-g glucose screening test, or by small sample size. These studies have shown that during pregnancy African-American women had lower glucose concentrations determined by the glucose screening test compared with other ethnic groups (3, 16, 17). The purpose of this study was to examine the ethnic differences in C-peptide/insulin/glucose dynamics and metabolism (insulin secretion and resistance) using simple indexes derived from fasting measures and the glucose challenge test (GCT) in healthy nondiabetic pregnant women of three ethnic groups (African-American, Hispanic, and Caucasian) from Camden.
Subjects and Methods
Study design
Data were collected as part of the Camden Study, an observational prospective cohort study on the relationship of maternal nutrition during pregnancy in young, generally healthy women residing in one of the poorest cities in the continental United States (18, 19, 20). The institutional review board at the University of Medicine and Dentistry of New Jersey approved the study protocol. Informed written consent was obtained from each participant after explanation of the nature and purpose of the study.
Subjects
The 773 women (343 African-Americans, 312 Hispanics, and 118 Caucasians) enrolled between 1997 and 2000 were included in this analysis. Less than 5% of women screened for eligibility were excluded from participation because of serious nonobstetric problem (e.g. lupus, type 1 or 2 diabetes, seizure disorders, malignancies, drug or alcohol abuse, and psychiatric problems). Some 80.5% of the patients who were eligible agreed to participate in this study. Of these, an additional 26 women who developed gestational diabetes mellitus in the course of pregnancy and six Asian women (inadequate statistical power) were excluded from this analysis.
Socioeconomic, demographic, and lifestyle data were obtained by interview at entry to care and updated during pregnancy and postpartum (4–6 wk). Ethnicity was assessed by self-report. Maternal weight was measured at each visit; height was measured at entry to prenatal care, and pregravid weight was obtained by recall. Body mass index (BMI) was computed as pregravid weight for height (kilograms per meter squared). Information on current and past pregnancy outcomes, complications, and infant abnormalities was abstracted from the prenatal records, the delivery record, delivery logbooks, and the infant’s chart. Gestation duration was based upon gestation from the gravidas’ last normal menstrual period and was confirmed or modified by ultrasound.
Samples collection and analytic procedures
Fasting (>8 h) blood samples were collected (75% in the morning, the remainder in the afternoon) in the second (18 ± 0.5 wk) and third trimesters (30 ± 0.11 wk) from each subject. A 50-g oral glucose load (GCT) as an initial screen to detect gestational diabetes was performed at 24–28 wk gestation, followed by plasma glucose determination 1 h later. Eighty-three percent of the GCT samples were fasting and collected in the morning. According to clinical practice recommendations from American Diabetes Association, the patient need not be fasting, and no baseline sample is required (21).
Blood samples collected at each visit were immediately refrigerated and centrifuged at 4 C within a maximum of 5 h. Plasma and serum samples were stored at -70 C until assayed. Fasting plasma glucose was measured with the glucose oxidase method (glucose reagent from Sigma, St. Louis, MO) on a spectrophotometer at a wavelength of 505 nm. Serum insulin was determined by RIA using a kit with a specific antibody that cross-reacts only minimally (<0.2%) with proinsulin and has a high sensitivity (2 µU/ml or 12 pmol/liter). Plasma C-peptide was determined by RIA with a high sensitivity (0.1 ng/ml or 0.033 nmol/liter) and low cross-reactivity to proinsulin (<4%; Linco Research, Inc., St. Charles, MO). The coefficients of variation within and between assays were 3.2% and 6.3%, respectively, for C-peptide, 3.5% and 6.5% for insulin, and 1.5% and 3.0% for glucose.
Indexes of insulin resistance, secretion, and clearance
The deconvolution method using C-peptide concentrations under steady state conditions and euglycemic-hyperinsulinemic clamp are considered the gold standard to assess insulin secretion rate and insulin sensitivity (22, 23, 24). However, it has been suggested that indexes such as fasting or stimulated insulin and glucose concentrations are reasonable surrogates for more sophisticated measures to evaluate insulin secretion and sensitivity in epidemiological studies (25, 26, 27). The C-peptide concentration has been interpreted as an indicator of insulin secretion because it is cosecreted on an equimolar basis from the ß-cell with insulin and is not extracted or metabolized by the liver (28). The metabolic clearance rate of C-peptide is constant under physiological conditions and over a wide range of plasma concentrations. Therefore, peripheral C-peptide and the C-peptide to insulin molar ratio reflect insulin production, changes in hepatic insulin extraction, and insulin clearance at both steady and nonsteady conditions (7, 28). The ratio of insulin to glucose concentration has been used as a measure of insulin resistance in nonpregnant adults and children (9, 27).
Statistical analyses
Metabolic parameters [insulin, glucose, C-peptide, C-peptide to insulin ratio (C/I ratio), and insulin to glucose ratios (I/G ratio)] were compared using analysis of covariance or ordinary least squares regression with control for potential confounding variables (version 8.0; SAS Institute, Inc., Cary, NC). Confounding was assessed by comparing crude and adjusted regression coefficients. Gestation at entry was always included as a control variable for the entry measures. The least squares means from the model were tested among ethnic groups using Bonferroni’s correction for multiple comparisons. Pearson correlation coefficients were computed to assess the relationship between C-peptide and glucose levels at fasting and during the GCT.
Results
Maternal characteristics by ethnic group are shown in Table 1
. There were no significant differences among the three groups in characteristics such as parity, prior history of preterm delivery, prior infant low birth weight, pregravid BMI, or the proportion of women who were overweight or obese (BMI, >26) before the current pregnancy. However, Caucasians were older than African-Americans and Hispanics; more of them smoked while pregnant, and more of them had private insurance.
|
Mean levels of fasting glucose, insulin, C-peptide, the I/G ratio and the C/I ratio, controlled for potential confounding variables, are presented in Table 2.
|
). However, there were marked ethnic differences in glucose after the GCT. Results of GCT
In response to a 50-g oral glucose load, African-Americans had a significantly lower glucose concentrations compared with Hispanics (P < 0.01) and Caucasians (P < 0.01; Table 3). Insulin concentrations and the I/G ratio were markedly higher in African-Americans and Hispanics than in Caucasians. Consistent with the fasting data, African-Americans had significantly lower C-peptide levels than Hispanics and a lower C/I ratio than Hispanics and Caucasians.
|
To determine the relations between C-peptide and glucose levels during pregnancy, we performed a correlation analysis (Pearson’s r) between the parameters. We found a positive relationship between C-peptide and glucose in both the fasting state and after the GCT (Table 4). These correlations were statistically significant for and present in each ethnic group as well (data not shown).
|
It was the objective of this study to determine whether ethnic differences in metabolic parameters were present in nondiabetic pregnant women. Previous studies showed ethnic differences in glucose metabolism among normal nonpregnant adults and adolescent even including children (6, 10, 11, 12, 29). To our knowledge, this is the first report to demonstrate the existence of ethnic differences in insulin production and resistance during normal pregnancy.
Our data suggested that African-American women may have significantly reduced insulin production, as indicated by C-peptide concentrations at fasting and after GCT, compared with Hispanics and/or Caucasians. Jiang et al. (11) presented similar findings in children. In contrast, Osei et al. (7) reported that there were no differences in fasting and poststimulation C-peptide levels between African-American and Caucasian nonpregnant adults. Data also showed that the mean plasma C-peptide response after a 75-g OGTT was greater in nonpregnant African-Americans than in Caucasians (30). The mechanisms of the ß-cell response and sensitivity to the effect of glucose and/or other nutrients ingested in African-Americans remain unknown.
The C/I ratios were markedly decreased in African-Americans (vs. Hispanics and Caucasians). The differences were present and became highly significant by the third trimester as well as after GCT, and this may suggest decreased insulin clearance in African-Americans. Thus, it is possible that higher insulin levels after a glucose load in pregnant African-Americans could be the result of decreased insulin clearance instead of hypersecretion of insulin.
The hyperinsulinemia of insulin resistance stems from an increase in insulin secretion and a decrease in insulin clearance (28). At the late stage of pregnancy, our results showed that African-Americans and Hispanics were more insulin resistant than Caucasians, suggested by I/G ratios and the somewhat higher insulin concentrations that were not statistically significant. Previous studies showed African American adolescents and adults had higher fasting insulin levels, a greater 2-h insulin response to an oral glucose challenge, an increased acute insulin response to an insulin-modified frequently sampled iv glucose tolerance test, and a lower insulin sensitivity index than Caucasians (6, 11). Current data are generally consistent with data reported for nonpregnant subjects.
After a 50-g glucose challenge, African-Americans and Hispanics had greater I/G ratios and insulin levels than Caucasians, suggesting that they are more insulin resistant than their Caucasian counterparts, which is in agreement with previous findings in nonpregnant subjects (6, 7). In contrast to the fasting state, African-Americans had significantly lower glucose concentrations than Caucasians and Hispanics (P < 0.01), although the absolute difference was small. This finding is in agreement with previous reports, although the mechanism remains unknown (7, 16, 17). The maternal glucose concentration has an important influence on fetal growth (31). In prior studies in nondiabetic Camden gravidas, we illustrated that lower maternal glucose after the screening test for gestational diabetes mellitus was associated with reduced infant birth weight, a reduced risk of large for gestation births, and an increased risk of fetal growth restriction (32). Lower postprandial glucose in African-Americans may diminish the amount of glucose transferred to the fetus, and fetal insulin secretion may be lower. These factors may play a role in increasing the risk of lower birth weight and fetal growth restriction among infants born to African-American women (33, 34). Thus, a reduced glucose flux between mother and fetus may be one of the mechanisms that underlie this ethnic difference.
African-Americans are at increased risk of obesity, type 2 diabetes mellitus, and hypertension (2, 35). Our data suggest that ethnic differences in metabolic parameters present in the early to middle stages of pregnancy and become highly statistically significant by the late stage of pregnancy in both the fasting state and after the GCT. Ethnic differences in insulin secretion and action thus may play an important role in increasing insulin resistance and in the development of glucose intolerance, obesity, and type 2 diabetes.
There are several limitations to this study. Firstly, insulin production and resistance were not measured directly by C-peptide decay kinetics or hyperinsulinemic clamp, because it was not feasible under the conditions of the study. However, data have shown that the fasting I/G ratio and homeostasis model assessment-insulin resistance are strongly correlated with estimates from the hyperinsulinemic clamp (27). Secondly, the I/G ratio has not been validated in pregnant women. During pregnancy, plasma volume is expanded, and the degree of insulin resistance is increased. Thus, further study may be needed to compare the estimates from the different methods.
In summary, our data showed ethnic differences in C-peptide concentration, C/I ratio, and I/G ratio in both fasting and glucose-stimulated conditions in normal young nondiabetic pregnant women, suggesting changes in insulin production, resistance, and/or clearance among the three ethnic groups. Compared with their Caucasian and Hispanic counterparts, African-American women had indexes suggestive of lower insulin production and greater insulin resistance, that is, a lower C-peptide concentration, a lower C/I ratio, and elevations in insulin and the I/G ratio.
Acknowledgments
We thank the staff of the Osborn Family Health Center, Our Lady of Lourdes Hospital, and St. John the Baptist Prenatal Clinic for providing access to patients; Janet Mead for her outstanding technical assistance; and Deborah Cruz for typing the manuscript.
Footnotes
This work was supported by NICHD Grant HD-18269.
Abbreviations: BMI, Body mass index; C/I ratio, C-peptide to insulin ratio; GCT, glucose challenge test; I/G ratio, insulin to glucose ratio.
Received December 10, 2001.
Accepted July 2, 2002.
References
This article has been cited by other articles:
 |
|
 |
|
  R. Retnakaran, A. J. G. Hanley, P. W. Connelly, M. Sermer, and B. Zinman Ethnicity Modifies the Effect of Obesity on Insulin Resistance in Pregnancy: A Comparison of Asian, South Asian, and Caucasian Women J. Clin. Endocrinol. Metab., January 1, 2006; 91(1): 93 - 97. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. O. Scholl, X. Chen, C. S. Khoo, and C. Lenders The Dietary Glycemic Index during Pregnancy: Influence on Infant Birth Weight, Fetal Growth, and Biomarkers of Carbohydrate Metabolism Am. J. Epidemiol., March 1, 2004; 159(5): 467 - 474. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Chen, T. O. Scholl, M. J. Leskiw, M. R. Donaldson, and T. P. Stein Association of Glutathione Peroxidase Activity with Insulin Resistance and Dietary Fat Intake during Normal Pregnancy J. Clin. Endocrinol. Metab., December 1, 2003; 88(12): 5963 - 5968. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
