International Journal of Diabetes Mellitus 2 (2010) 3–9

Contents lists available at ScienceDirect

International Journal of Diabetes Mellitus

journal homepage: ees.elsevier.com/locate/ijdm

Original Article

Clinical assessment of insulin action during late pregnancy in women at risk

for gestational diabetes: Association of maternal glycemia with perinatal outcome
Karen E. Elkind-Hirsch a,*, Beverly W. Ogden b, Carmen J. Darensbourg a, Brett L. Schelin b
a
Woman’s Health Research Department, Woman’s Hospital, Baton Rouge, LA, United States
b
Woman’s Pathology Laboratory, Woman’s Hospital, Baton Rouge, LA, United States

a r t i c l e i n f o a b s t r a c t

Article history: Objective: We prospectively evaluated differences in fasting- and oral glucose tolerance test (OGTT)-
Received 1 September 2009 derived indices of insulin action in Caucasian (Cau) and African-American (AA) pregnant women and
Accepted 16 December 2009 compared them with obstetric outcomes.
Study design: IRB-approved prospective study in 171 pregnant women undergoing a 3-h OGTT. Mathe-
matical modeling was used to evaluate insulin response, insulin activity and glucose tolerance in fasting
Keywords: and postglucose ingestion state. Insulin sensitivity indices derived from fasting (HOMA-IR) and glucose-
Gestational diabetes
stimulated values (SIOGTT) were compared. An insulin sensitivity-secretion index (IS-SI) was calculated
Insulin sensitivity
Insulin secretion
from the product of the SIOGTT and early-phase insulin secretion.
Racial disparity Results: Forty-nine patients had gestational diabetes (GDM), 28 had gestational impaired glucose toler-
Perinatal outcome ance (GIGT) and 94 had normal glucose tolerance after an abnormal glucose challenge test (NGT-
abnGCT). Insulin sensitivity was lowest in women with GDM. In all groups, pregnant AA women were sig-
nificantly more insulin resistant than Cau women, based on both HOMA-IR and SIOGTT, but had
enhanced insulin secretion compared to their Cau counterparts. The mean IS-SI progressively improved
for all women from GDM to GIGT to NGT-abnGCT. Women with NGT-abnGCT had a higher prevalence of
large-for-gestational age (LGA) newborns and significantly higher cesarean section rate.
Discussion: Insulin measures along with glucose determinations during OGTT testing in pregnant women
at risk for diabetes provide valuable information that varies according to race. We observed that pregnant
women with a lesser degree of glucose tolerance abnormality during pregnancy who receive no interven-
tion have a higher risk for LGA infants and significantly increased C-section rate (ClinicalTrials.gov num-
ber, NCT006874791).
Ó 2009 International Journal of Diabetes Mellitus. Published by Elsevier Ltd.
Open access under CC BY-NC-ND license.

1. Introduction Women with GDM are at an increased risk of cesarean delivery,

Pregnancy is a diabetogenic condition characterized by insulin
resistance with a compensatory increase in b-cell response and
hyperinsulinemia [1]. Normal women compensate for insulin resis-
tance by increasing insulin secretion to maintain normal glucose
tolerance [2]. Gestational diabetes mellitus (GDM), defined as car-
bohydrate intolerance with onset or first recognition during preg-
nancy, occurs when insulin secretion is insufficient to compensate
for the insulin resistance of pregnancy. Women with GDM have a
limited ability to increase their insulin secretion [3–6]. As a result,
they have very blunt glucose-stimulated insulin responses com-
pared with the augmented insulin responses of normal pregnant
women [1,7,8].
* Corresponding author. Address: Woman’s Health Research Department,
Woman’s Hospital, 9050 Airline Highway, Baton Rouge, LA 70815, United States.
Tel.: +1 225 231 5278; fax: +1 225 924 8225.
E-mail address: Karen.Elkind-Hirsch@womans.org (K.E. Elkind-Hirsch).
while their infants tend to experience higher rates of macrosomia
and shoulder dystocia [9]. Abnormal fetal growth in diabetic preg-
nancy seems to occur with any elevation in the maternal glucose
level [10]. Pregnant women with elevated glucose levels have a
higher risk of delivering increased birth weight infants, even when
their glucose levels are below those diagnostic of GDM [10,11].
Pregnant women with impaired glucose tolerance exhibit insulin
resistance comparable to women with GDM, and have an increased
risk of macrosomic infants and other morbidities [11]. It has been
suggested that even minor degrees of increased glucose intoler-
ance during pregnancy in women without GDM are related in a
continuous and graded pattern with a significantly increased inci-
dence of macrosomia, cesarean section, pre-eclampsia and an in-
creased need for neonatal intensive care unit admission, as well
as greater length of maternal and neonatal hospital stay [12–16].
Women of ethnic minority populations are at a greater risk for
developing GDM [17,18]. Solomon et al. [17] found that the risk of
GDM increased among non-Caucasian women in the Nurses’

1877-5934 Ó 2009 International Journal of Diabetes Mellitus. Published by Elsevier Ltd. Open access under CC BY-NC-ND license.
doi:10.1016/j.ijdm.2009.12.006
4 K.E. Elkind-Hirsch et al. / International Journal of Diabetes Mellitus 2 (2010) 3–9
Health Study Cohort II. A significant interaction between glucose
status and race was identified by Saldana et al. [19], so their anal-
yses were stratified by race looking at African-American (AA) and
Caucasian (Cau) mothers separately. Obesity-related risks during
pregnancy were also found to vary by race, with obese AA women
more likely to have adverse outcomes than obese Cau women [20].
Other researchers report the racially disparate effects of impaired
glucose tolerance and glucose levels on birth outcomes, with these
conditions leading to higher levels of macrosomic babies among
AA women, but not among Cau women [19,21].
Gravidas with GDM generally demonstrate higher degrees of
post-pregnancy insulin resistance, b-cell dysfunction, higher
BMI, central obesity, and exaggerated hyperlipidemia, which sug-
gest that GDM is a transient manifestation of longstanding meta-
bolic dysfunction [7,8]. The oral glucose tolerance test (OGTT) in
pregnancy can provide valuable insights into the underlying met-
abolic phenotype and risk potential of young, otherwise healthy
women. Notably, the diagnosis of GDM, based on glucose values
from an antepartum OGTT, identifies a population of young
women at elevated risk of developing diabetes later in life [22].
Investigators have often used fasting glucose and insulin levels
or levels after glucose administration as indicators of insulin
sensitivity. Kirwan et al. [4] reported that insulin sensitivity esti-
mated from glucose and insulin levels during an OGTT was signif-
icantly improved compared with fasting values in pregnant
women with normal glucose tolerance and GDM. This study
examined the use of fasting- and OGTT-derived indices to mea-
sure insulin sensitivity and secretion in pregnant women in
southern Louisiana with varying degrees of glucose tolerance.
We further explored the potential use of these measures to define
racially diverse risk profiles for these pregnant women and com-
pare them with obstetric and perinatal outcomes.
2. Materials and methods
2.1. Treatment protocol
The Institutional Review Board of the Woman’s Hospital Foun-
dation approved the protocol, and all participants gave written in-
formed consent. Pregnant women were asked to participate if they
met all of the following criteria: (1) gestational age between 20–
30 weeks, (2) at least 18 years of age, (3) were either Cau or AA
and (4) had an otherwise uncomplicated pregnancy. Women with
known pregestational type 1 or type 2 diabetes or women of a dif-
ferent ethnicity were excluded from this study. The women were
screened for carbohydrate intolerance by performing a standard
1-h, 50-g oral glucose challenge test (GCT) between the 20th and
28th week of gestation. If the plasma glucose level was greater
than 135 mg/dL (GCT positive), they then underwent a 3-h, 100-g
oral glucose tolerance test (OGTT).
Study participants consisted of 176 pregnant women (135 Cau,
41 AA) attending outpatient obstetrics clinics, who had been re-
ferred for OGTT testing. All OGTTs were performed in the early
morning (7:00–9:00 AM) after a 10- to 12-h overnight fast at the
outpatient Woman’s Hospital Pathology Laboratory. On the morn-
ing of the test, demographic, anthropometric and clinical data
(maternal age, race/ethnicity, family history of diabetes, obstetric
history, and prepregnancy BMI) were collected by an inter-
viewer-administered questionnaire. Venous blood samples were
drawn for measurement of glucose and insulin at fasting and 30,
60, 120, and 180 min after oral ingestion of 100 g glucose load. Five
women (4 Cau, 1 AA) were unable to complete the test, because of
vomiting after glucose intake. Based on the GCT and OGTT, partic-
ipants were stratified into the following three glucose tolerance
groups:
 Gestational diabetes mellitus (GDM), defined by at least ‘‘two
glucose levels at or higher than the following OGTT cutoffs: fast-
ing, 95 mg/dL; 1 h, 180 mg/dL; 2 h, 155 mg/dL; 3 h, 140 mg/dL
[23].
 Gestational impaired glucose tolerance (GIGT), defined by meet-
ing only one abnormal glucose of above criteria during OGTT.
 Normal OGTT, GCT positive (NGT-abnGCT), defined as having an
abnormal GCT followed by NGT on the OGTT (defined by meet-
ing none of the above criteria).
2.2. Laboratory measurements and physiologic indexes
Determinations of glucose and insulin levels in the fasting state
and during an oral glucose load were performed, as previously de-
scribed [24]. Insulin secretion and sensitivity were expressed using
glucose and insulin, measured in conventional units (milligrams
per deciliter and microunits per milliliter, respectively). The
homeostasis model of assessment for insulin resistance (HOMA-
IR) was calculated from fasting values as described by Matthews
et al. [25]. The HOMA-IR generally provides a partial estimate of
body insulin sensitivity because it mainly correlates with basal he-
patic insulin resistance [26]. This is why we also evaluated dy-
namic insulin sensitivity using the OGTT insulin sensitivity
(ISOGTT) model of Matsuda and DeFronzo, which correlates with to-
tal glucose disposal, as extensively validated vs. the glucose clamp
in various pathophysiological conditions [26]. In pregnant women,
ISOGTT exhibits better correlation with insulin sensitivity, derived
using the glucose clamp, than did the HOMA-IR model [3]. Insulin
secretion was estimated after oral glucose loading by two meth-
ods; (1) the corrected insulin response at glucose peak (CIRgp)
[27] and (2) the insulinogenic index divided by HOMA-IR (IGI/
HOMA-IR) [28,29] which have been applied previously in pregnant
women with and without GDM [30]. The insulinogenic index (IGI)
was calculated as the ratio of change in insulin concentration to
change in glucose during the first 30 min of the OGTT [31]. The
early-phase insulin release calculated by the IGI is used as a surro-
gate marker of first-phase insulin secretion measured during the
glucose clamp. b-Cell compensatory capacity was calculated using
the insulin sensitivity-secretion index (IS-SI) defined as the prod-
uct of SIOGTT and first-phase insulin release index (IGI). The IS-SI
expresses the overall ability of the b-cell to increase its release rate
relative to insulin resistance in response to a glucose stimulus and
reveals the progressive loss of b-cell function in individuals with
IGT and GDM that was originally demonstrated using the disposi-
tion index calculated from the glucose clamp [29]. An analogous
mathematically derived measure; the insulin sensitivity secretion
index (ISSI) has been utilized previously in both pregnant diabetic
and nondiabetic women [32].
2.3. Obstetrical and perinatal outcomes
Obstetrical outcome information was obtained from a database
that tracks labor and delivery data for all deliveries at the Woman’s
Hospital. Each woman’s demographic information such as age and
race was obtained from the computerized hospitalization record
and confirmed with self-reported information. Neonatal data were
abstracted by the review of maternal and newborn medical re-
cords. We recorded maternal prepregnancy weight, parity, age,
race, maternal drug or tobacco use, delivery mode, and obstetric
history (previous GDM, delivery mode), and infant’s weight and
height, gestational age at delivery, and birth weight for gestational
age. According to the gestational age-specific weight distribution
of the study population, infants were considered large-for-gesta-
tional age (LGA) if their sex-specific birth weight for gestational
age exceeded the 90th percentile of the US population fetal growth
 K.E. Elkind-Hirsch et al. / International Journal of Diabetes Mellitus 2 (2010) 3–9 5

Table 1
Clinical features of pregnant study subjects and neonatal outcomes by glucose tolerance group.

NGT/abGCT GDM GIGT P value

Maternal attributes n = 94 n = 49 n = 28
[AA = 22; Cau = 72] [AA = 13; Cau = 36} [AA = 5; Cau = 23}
Age (year) 29 ± 0.6 29 ± 0.8 29.5 ± 0.9 0.84a
AA 27.2 ± 1.1 28.2 ± 1.5 27.6 ± 2.4 0.14b
Cau 29.3 ± 0.6 29 ± 0.9 29.9 ± 1.1
Parity (n) 0.9 ± 0.1 0.78 ± 0.14 1.00 ± 0.2 0.51a
AA 1.4 ± 0.2 1.4 ± 0.3 1.2 ± 0.4 0.001b
Cau 0.83 ± 0.1 0.56 ± 0.17 0.96 ± 0.2
Pregravid BMI (kg/m2) 27.6 ± 0.7 29.7 ± 1.3 27.6 ± 1.3 0.25a
AA 28.7 ± 1.6 34.5 ± 2.1 32.5 ± 3.3 0.003b
Cau 27.2 ± 0.9 28 ± 1.2 26.5 ± 1.5
Obstetric outcomes (n = 91) (n = 46) (n = 28)
[AA = 20; Cau = 71] [AA = 13; Cau = 33} [AA = 5; Cau = 23}
Infant birthweight (gm) 3391 ± 50 3129 ± 106 3155 ± 128 0.025a; 0.0141
AA 3160 ± 130 3150 ± 161 2913 ± 260 0.14b
Cau 3457 ± 69 3121 ± 101 3207 ± 121
Cesarean section (%) 48 37 43 0.049c
AA 50 31 40 0.59b
Cau 48 42 43

Gestational age-specific weight (rate)

SGA 1 (1%) 3 (6%) 3 (11%) 0.41d
AGA 63 (69%) 38 (78%) 18 (64%)
LGA 27 (30%) 8 (16%) 7 (25%)

Values are expressed as means ± SEMs. P < 0.05 was considered statistically significant.
Race: African-American – AA; Caucasian – Cau.
a
Overall differences between three glucose tolerance groups: 1NGT vs. GDM.
b
Overall main effect of race – AA vs. Cau.
c
NGT vs. impaired (combined GDM and GIGT).
d
Overall differences across diagnosis groups for birth weight rates for gestational age.

curves, or small for gestational age (SGA) if their birth weight was
below the 10th percentile.
2.4. Statistical analyses
All analyses were conducted using SPSS statistical software
(version 15.1, SPSS, Chicago, IL). Continuous variables were tested
for normality of distribution using the Kolmogorov–Smirov test.
Comparisons of clinical measures and insulin action indices be-
tween glucose tolerance groups were performed using one-way
ANOVA for continuous parameters with means following normal
distribution. Nonparametric Kruskal–Wallis test was carried out
if appropriate. The Bonferroni–Dunn post hoc test was applied to
analyze the variation among the three groups if the ANOVA
showed overall differences that were significant (P < 0.05). To eval-
uate differences in the impact of maternal race on glucose toler-
ance status category, data were further analyzed using a two-
way ANOVA with race and glucose tolerance diagnosis as indepen-
dent variables, with both main and interaction effects calculated.
Regression analysis between fasting and OGTT-derived measures
of insulin action was performed using Pearson product-moment
correlation test. We further explored categorical differences in
the distributions of family history of GDM, gestational age-specific
weight, cesarean delivery (C-section) rates and Apgar scores by
glucose tolerance group using v2 tests for comparison. All P values
were two-tailed; P < 0.05 was considered statistically significant.
The results are presented as means ± standard error of the mean
(SEM) unless otherwise indicated.
remaining 171 patients completing the OGTT, 131 (76.6%) were
Cau and 40 (23.4%) were AA (Table 1). Table 1 reveals that 49
(29%) and 28 (16%) women were diagnosed with GDM and GIGT,
respectively. Of the remaining 94 participants, 72 were Cau (55%)
and 22 were AA (56.4%). There were no significant differences
among glucose tolerance groups with respect to age, weeks’ gesta-
tion tested for the condition, prepregnancy BMI, or parity. As seen
in Table 1, AA women had a higher pregravid BMI (P < 0.003) and
increased parity (P < 0.001) compared with their Cau counterparts.
(Table 1).
3.2. Comparison of insulin sensitivity and insulin secretion measures
in pregnant subjects
Table 2 shows the comparative data for fasting to glucose-stim-
ulated insulin sensitivity. Overall agreement between the two
measures of insulin sensitivity was modest (r = 0.61, P < 0.001)
for all pregnant women. Prevalence of insulin resistance was high-
er using estimates of insulin sensitivity from the ISOGTT than using
fasting values to classify women as insulin resistant. The HOMA-IR
index did not reveal decreased insulin sensitivity in 23% (11 Cau
and 5 AA) of the women tested compared with the ISOGTT. The
HOMA-IR failed to determine 5 Cau women with GDM and 1AA
and 2 Cau women with GIGT as having insulin resistance. The
two measures of insulin secretion, IGI/HOMA-IR and CIRgp, were
highly correlated (r = 0.74; P < 0.0001; Table 2). Poor insulin
responsiveness to a glucose challenge was evident in both AA
and Cau women that had diabetes.

3. Results 3.3. Metabolic measures

3.1. Participants and prevalence of gestational diabetes Metabolic characteristics pertaining to insulin action for the
three glycemic tolerance groups are summarized in Table 2.
Of the 176 consented participants, five pregnant women were Evaluation of fasting insulin resistance demonstrated the greatest
excluded because of vomiting after the glucose load. Of the sensitivity in the NGT-abnGCT group wherein GIGT had HOMA-IR
6 K.E. Elkind-Hirsch et al. / International Journal of Diabetes Mellitus 2 (2010) 3–9

Table 2
Subject glucose tolerance during late pregnancy.

NGT/abGCT GDM GIGT P value

(n = 94) (n = 49) (n = 28)
[AA = 22; Cau = 72] [AA = 13; Cau = 36] [AA = 5; Cau = 23]
HOMA-IR 2.5 ± 0.15 3.7 ± 0.35 3.6 ± 0.82 0.001a;
0.0041; 0.03
AA 2.7 ± 0.5 5.5 ± 0.6 7.2 ± 1.1 0.001b
Cau 2.4 ± 0.3 3.2 ± 0.4 3 ± 0.5 0.02c
SIOGTT 4.3 ± 0.24 2.8 ± 0.3 4.0 ± 0.5 0.001a
0.0031; 0.0353
AA 3.9 ± 0.5 1.7 ± 0.7 1.9 ± 1.0 0.003b
Cau 4.4 ± 0.3 3.2 ± 0.4 4.4 ± 0.5 0.21c
CIRgp 0.97 ± 0.07 0.74 ± 0.06 0.91 ± 0.14 0.002a
0.0251
AA 1.5 ± 0.1 0.87 ± 0.16 2.1 ± 0.3 0.0001b
Cau 0.8 ± 0.06 0.7 ± 0.09 0.7 ± 0.1 0.007c
IGI/HOMA-IR 1.0 ± 0.06 0.32 ± 0.09 0.43 ± 0.13 0.0001a
0.00011; 0.0062
AA 1.4 ± 0.11 0.3 ± 0.16 0.44 ± 0.2 0.03b
Cau 0.6 ± 0.06 0.3 ± 0.09 0.43 ± 0.1 0.001c
ISSI 333.7 ± 35 115 ± 8.5 186 ± 14.3 0.0001a;
0.00011; 0.0142
AA 377 ± 51 115 ± 69 183 ± 79 P = 0.06b
Cau 260 ± 28 115 ± 40 187 ± 49 P = 0.02c

Values are expressed as means ± SEMs. P < 0.05 was considered statistically significant.
African-American – AA; Caucasian – Cau.
a
Overall differences between three glucose tolerance groups: 1 NGT vs. GDM; 2NGT vs. GIGT; 3GDM vs. GIGT.
b
Overall main effect of race – AA (n = 40) vs. Cau (n = 131).
c
Interaction differences of race by diagnosis.

values similar to GDM (P < 0.001; Table 2). In AA women, both
GDM and GIGT were significantly less sensitive than NGT-
abnGCT, whereas in Cau women, the sensitivity progressively de-
clined from NGT-abnGCT to GIGT to the GDM group (P < 0.02). As
shown in Fig. 1A, glucose-stimulated measures revealed a differ-
ent overall pattern with the highest SIOGTT index in the NGT-
abnGCT group, followed in turn by the GIGT and GDM groups,
respectively (P < 0.01). Consistent with fasting measures, in AA
women, both GDM and GIGT SIOGTT values differ from NGT-
abnGCT but not each other, whereas in Cau women, both NGT-
abnGCT and GIGT subjects were significantly more sensitive than
the GDM subjects (P < 0.003; Table 2). Pregnant AA women were
less sensitive than their Cau counterparts overall as illustrated in
Fig. 1A.
Evaluation of overall insulin secretion supported the observed
glycemic trends across the three study groups. For all pregnant wo-
men, CIRgp was highest in NGT-abnGCT, followed by GIGT and then
GDM (P < 0.002; Fig. 1B). Fig. 1B illustrates that in pregnant Cau
women CIRgp was highest in NGT-abnGCT compared to GIGT and
GDM that are not different from each other. In pregnant AA wo-
men, CIRgp was highest in GIGT and NGT-abnGCT groups whereas
in CIRgp was significantly lower in GDM (P < 0.007). Nondiabetic
AA women have significantly greater CIRgp compared to Cau
whereas there is no racial difference in diabetics which is shown
in Fig. 1B. Table 2 shows the differences in b-cell function in preg-
nant women assessed by IGI/HOMA-IR; insulin secretion was high-
est in the NGT-abnGCT group compared with GDM (P < 0.0001)
and GIGT groups (P < 0.006). The IGI/HOMA-IR was significantly
higher in AA pregnant women compared to the corresponding
Cau subjects (P < 0.03).
The relationship between insulin sensitivity and insulin secre-
tion demonstrated a racial dissimilarity as is evident in Fig. 1C.
The insulin sensitivity-secretion index (IS-SI) of NGT-abnGCT AA
women was significantly higher than Cau counterparts (P < 0.02)
whereas mean IS-SI in GDM or GIGT groups showed no racial dis-
parity (Fig. 1C).
3.4. Maternal and perinatal outcomes
Shown in Table 1 and 165 patients delivered at Woman’s Hos-
pital of which 127 (77%) were Caucasian and 38 (23%) were Afri-
can-American. Six patients (4 GDM, 2 NGT-abnGCT) delivered
elsewhere. Birth weights of infants from NGT-abnGCT mothers
were significantly higher than from GDM mothers (P < 0.014; Table
2). Eight (17%) infants from GDM mothers, seven (23%) infants
from GIGT mother and 27 (30%) babies of mother with NGT-
abnGCT were LGA (Table 1). No consistent differences in length
of gestation, infant gender or frequency of Apgar scores <7 at 1
and 5 min between glucose tolerance groups were found. No vari-
ations due to race or race by diagnosis were observed to impact
gestational age-specific weight, gender or length of gestation.
Women with NGT-abnGCT had a significantly higher C-section
rate compared with mothers having any impaired glucose metab-
olism on the 3-h OGTT (P < 0.049; Table 1). Overall, 48% of women
with NGT-abnGCT had a C-section compared with 39% of women
with either GDM or GIGT. No racial disparity was found.
4. Discussion
In current clinical practice, women with GDM are identified on
the basis of hyperglycemia on routine glucose tolerance testing in
pregnancy. We report our institutional experience with the 100-g
OGTT in which glucose and insulin levels were evaluated in the
fasting state and after an oral glucose load in a cohort of pregnant
women across the glycemic spectrum. Adding insulin levels to the
OGTT provided a clearer picture of the subtle metabolic abnormal-
ities in both insulin sensitivity and b-cell function in this at-risk
pregnant population. Other investigators suggested the use of fast-
ing measure, such as HOMA-IR as an alternative but sensitive
screening test for GDM, which avoids oral administration of glu-
cose-containing solutions [33]. While the HOMA-IR was correlated
with the OGTT-derived insulin sensitivity assessment, we found
 K.E. Elkind-Hirsch et al. / International Journal of Diabetes Mellitus 2 (2010) 3–9 7

Fig. 1. Insulin sensitivity index [SIOGTT] values derived from the OGTT (A); corrected insulin response at glucose peak [CIRgp] (B); and insulin secretion-sensitivity index [IS-SI]
(C) stratified by race and glucose tolerance group. The mean SIOGTT values progressively decreased from GDM to GIGT to NGT-abGCT (A). Findings further illustrate that GDM
is characterized by impairment of pancreatic b-cell function (CIRgp) with a progressive reduction going from women with NGT to those with GIGT and GDM status (B). In all
glucose tolerance groups, AA women had greater insulin secretory responses than Cau. The worsening glycemic profile is reflected in the trend of decreasing mean IS-SI score
from NGT-abnGCT to IGT to GDM (C). The IS-SI was higher in AA women with NGT compared with Cau, no impact of race on IS-SI was seen in women with GIGT and GDM.
8 K.E. Elkind-Hirsch et al. / International Journal of Diabetes Mellitus 2 (2010) 3–9
the HOMA-IR provided a weaker predictive index compared to the
glucose-stimulated ISOGTT measure. Furthermore, based on OGTT-
derived indices of insulin secretion, it was quite clear that b-cell
function progressively deteriorates with worsening of glucose tol-
erance, consistent with results obtained in other studies [30,32].
Similar to the ISSI index first reported by Retnakaran et al. [32],
we calculated an IS-SI for each pregnant patient as a novel inte-
grated measure of insulin sensitivity relative to insulin secretion.
Specifically, we found that compared with NGT-abnGCT, women
with GDM and GIGT had lower insulin sensitivity, poorer insulin
secretion, and greater glycemia. Poor b-cell compensation for insu-
lin resistance was evident in all pregnant women with GDM. A lim-
itation of this study is that only women who failed the GCT
underwent OGTT assessment and therefore all the pregnant wo-
men studied had some subtle impairment in insulin action. A fur-
ther limitation is that the estimates of insulin action have been
made on calculations based on the OGTT, and not by a ‘‘gold stan-
dard” test, euglycemic clamp study. The indices are being used in
population studies as the clamp studies are not feasible in large
numbers.
Racial differences in basal and post-stimulation glucose homeo-
static regulation were present over the spectrum of glucose toler-
ance. Nondiabetic pregnant AA women were more insulin resistant
but had higher baseline and glucose-stimulated insulin levels com-
pared to Cau counterparts, findings that are consistent with earlier
observations [34]. One limitation to this finding of ethnic differ-
ences is the modest size (n = 40) of the AA pregnant group com-
pared to the Cau group (n = 131) studied. However, these
differences are not unique to pregnancy; AA children have higher
fasting insulin, greater glucose-stimulated insulin levels and lower
insulin sensitivity as assessed by a variety of methods [35]. As was
observed by other investigators [18], we found that pregnant AA
women were more obese than pregnant Cau women in all glucose
tolerance groups that might account for the increased insulin resis-
tance. Even so, with further adjustments for obesity and body fat
distribution, nondiabetic African-Americans continued to have
lower insulin sensitivity but higher fasting and 2-h insulin levels
and acute insulin response to glucose than whites [18].
Although maternal GDM has long been associated with fetal
macrosomia, our data support a growing number of studies that re-
port a continuous effect of maternal glucose levels, even in the ab-
sence of GDM, on offspring birth weight and pregnancy outcomes
[10–16,36–39]. Hillier et al. [36] observed among women with
both normal and abnormal GDM screenings, increasing level of
maternal glucose was linearly related to macrosomia risk. Unde-
tectable glucose intolerance and a resistance to insulin are sup-
posed mechanisms for this subgroup of patients. Retnakaran et
al. [37] demonstrated that women with NGT-abnGCT are clearly
distinct from those with NGT-normal GCT on the basis of lower
insulin sensitivity and greater glycemia. We found that pregnant
women who failed only the GCT showed a higher prevalence of
delivering high birthweight infants and were significantly more of-
ten delivered by cesarean section than GDM and GIGT mothers.
Similarly, Lapolla et al. [38] found that the newborns of false pos-
itives screening test (NGT-abnGCT) mothers were heavier than
those with NGT or GDM. Others have also reported that increasing
maternal glucose levels within the normal nondiabetic range were
consistently related to larger offspring birth size (large-for-gesta-
tional age, LGA) and increased risk of interventional deliveries
[10,30]. In a prospective study of greater than 6000 women, Yogev
et al. [39] found a gradual increase in the rate of macrosomia, LGA
and cesarean section in relation to increasing GCT severity catego-
ries in women without GDM. Moreover, they demonstrated that
increased fetal weight and cesarean section rates in both obese
and non-obese women are associated with higher degrees of car-
bohydrate intolerance [39]. In this study, NGT-abnGCT patients
had a complication rate that was worse than patients diagnosed
and/or treated as GDM. However, all pregnant women with GDM
and GIGT at our institution receive interventional glucose-lowering
therapy (diet, oral medication, insulin) to improve obstetrical out-
come. Standard obstetrical practice at our institution generally
does not precipitate any specific intervention or treatment recom-
mendations for women with normal glucose tolerance results after
an abnormal GCT.
5. Conclusions
In summary, antepartum OGTT screening identifies carbohy-
drate intolerance that occurs when insulin secretion is insufficient
to compensate for the insulin resistance of pregnancy. The work
presented here argues that insulin measures along with glucose
determinations during oral glucose tolerance testing provide valu-
able screening test information which is racially disparate and fu-
ture work should examine the predictive value of derived insulin
action indices for the diagnosis of GDM risk in a large prospective
ethnically-diverse cohort. Hyperglycemia during pregnancy, less
severe than overt DM, is associated with increased risk of adverse
maternal and fetal outcome that is independently related to the
degree of metabolic disturbance. Pregnancy glycemia, insulin sen-
sitivity, and insulin secretion all contribute to offspring adiposity
and macrosomia, and may be separate targets for intervention to
optimize birth outcomes and later offspring health.
Acknowledgements
We wish to thank the Pathology Laboratory technicians for their
unrelenting efforts in patient recruitment.
This research was an investigator-initiated study supported in
part by a grant from the Baton Rouge Area Foundation awarded
to KEH. This study was presented in part at the 91st annual Meet-
ing of the Endocrine Society, Washington, DC, June 10–13, 2009.
References
[1] Stanley K, Fraser R, Bruce C. Physiological changes in insulin resistance in
human pregnancy: longitudinal study with the hyperinsulinaemic
euglycaemic clamp technique. Br J Obstet Gynaecol 1998;105:756–9.
[2] Catalano PM, Tyzbir ED, Wolfe RR, Calles J, Roman NM, Amini SB, et al.
Carbohydrate metabolism during pregnancy in control subjects and women
with gestational diabetes. Am J Physiol 1993;264:E60–7.
[3] Catalano PM, Huston L, Amini SB, Kalhan SC. Longitudinal changes in glucose
metabolism during pregnancy in obese women with normal glucose tolerance
and gestational diabetes mellitus. Am J Obstet Gynecol 1999;180:903–16.
[4] Kirwan JP, Huston-Presley H, Kalhan SC, Catalano PM. Clinically useful
estimates of insulin sensitivity during pregnancy. Diabetes Care
2001;24:1602–7.
[5] Ryan EA, O’Sullivan MJ, Skyler JS. Insulin action during pregnancy. Studies with
the euglycemic clamp technique. Diabetes 1985;34:380–9.
[6] Homko C, Sivan E, Chen X, Reece EA, Boden G. Insulin secretion during and
after pregnancy in patients with gestational diabetes mellitus. J Clin
Endocrinol Metab 2001;86:568–73.
[7] Xiang AH, Peters RK, Trigo E, Kjos SL, Lee WP, Buchanan TA. Multiple metabolic
defects during late pregnancy in women at high risk for diabetes. Diabetes
1999;48:848–54.
[8] Buchanan TA. Pancreatic b-cell defects in gestational diabetes: implications for
the pathogenesis and prevention of type 2 diabetes. J Clin Endocrinol Metab
2001;86:989–93.
[9] Boyd ME, Usher RH, McLean FH. Fetal macrosomia: prediction, risks, proposed
management. Obstet Gynecol 1983;61(6):715–22.
[10] Metzger BE, Lowe LP, Dyer AR, Trimble ER, Chaovarindr U, Coustan DR, et al.
For HAPO Study Cooperative Research Group. Hyperglycemia and adverse
pregnancy outcomes. N Engl J Med 2008;358:1991–2002.
[11] Schafer-Graf UM, Dupak J, Vogel M, Dudenhausen JW, Kjos SL, Buchanan TA.
Hyperinsulinism, neonatal obesity and placental immaturity in infants born to
women with one abnormal glucose tolerance test value. J Perinat Med
1998;26:27–36.
[12] Sermer M, Naylor CD, Gare DJ, et al. Impact of increasing carbohydrate
intolerance on maternal–fetal outcomes in 3637 women without gestational
diabetes: the Toronto Tri-Hospital gestational diabetes project. Am J Obstet
Gynecol 1995;173:146–56.
 K.E. Elkind-Hirsch et al. / International Journal of Diabetes Mellitus 2 (2010) 3–9
[13] Jensen DM, Damm P, Sorensen B, Molsted-Pedersen L, Westergaard JG, Klebe J.
Clinical impact of mild carbohydrate intolerance in pregnancy: a study of 2904
nondiabetic Danish women with risk factors for gestational diabetes mellitus.
Am J Obstet Gynecol 2001;185:413–9.
[14] Leikin EL, Jenkins JH, Pomerantz GA, Klein L. Abnormal glucose screening tests
in pregnancy: a risk factor for fetal macrosomia. Obstet Gynecol
1987;69:570–3.
[15] Mello G, Parretti E, Mecacci F, et al. Risk factors for fetal macrosomia: the
importance of a positive oral glucose challenge test. Eur J Endocrinol
1997;137:27–33.
[16] Tallarigo L, Giampietro O, Penno G, Miccoli R, Gregori G, Navalesi R. Relation of
glucose tolerance to complications of pregnancy in nondiabetic women. N Engl
J Med 1986;315:989–92.
[17] Solomon CG, Willet WC, Carey VJ, Rich-Edwards J, Hunter DJ, Colditz GA, et al.
A prospective study of pregravid determinants of gestational diabetes
mellitus. JAMA 1997;278:1078–83.
[18] Haffner SM, D’Agostino Jr R, Saad MF, Bergman RN, Savage PJ, Rewers M, et al.
Increased insulin resistance and insulin secretion in nondiabetic African-
Americans and Hispanics compared with non-Hispanic whites. Diabetes
1996;45:742–8.
[19] Saldana TM, Siega-Riz AM, Adair L, Savitz DA, Thorp JM. The association
between impaired glucose tolerance and birth weight among black and white
women in central North Carolina. Diabetes Care 2003;26:656–61.
[20] Steinfeld JD, Valentine S, Lerer T, Ingardia CJ, Wax JR, Curry SL. Obesity-related
complications of pregnancy varies by race. J Matern Fetal Med 2000;9:238–41.
[21] Scholl TO, Chen X, Gaughan C, Smith WK. Influence of maternal glucose level
on ethnic differences in birth weight and pregnancy outcome. Am J Epidemiol
2002;156:498–506.
[22] Kjos S, Buchanan T, Greenspoon J, Montoro M, Bernstein G, Mestman J.
Gestational diabetes mellitus: the prevalence of glucose intolerance and
diabetes mellitus in the first two months postpartum. Am J Obstet Gynecol
1990;163:93–8.
[23] National Diabetes Data Group. Classification and diagnosis of diabetes mellitus
and other categories of glucose intolerance. Diabetes 1979;28:1039–57.
[24] Elkind-Hirsch K, Darensbourg C, Ogden B, Ogden L, Hindelang P. Contraceptive
vaginal ring use for women has less adverse metabolic effects than an oral
contraceptive. Contraception 2007;76:348–56.
[25] Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC.
Homeostasis model assessment: insulin resistance and b-cell function from
fasting plasma glucose and insulin concentrations in man. Diabetologia
1985;28:412–9.
9
[26] Matsuda M, DeFronzo R. Insulin sensitivity indices obtained from oral glucose
tolerance testing. Diabetes Care 1999;22:1462–70.
[27] Sluiter WJ, Erkelens DW, Terpstra P, Reitsma WD, Doorendos H. Glucose
intolerance and insulin release, a mathematical approach, I: assay of the beta
cell response after glucose loading. Diabetes 1976;25:241–4.
[28] Kitabchi AE, Temprosa M, Knowler WC, Kahn SE, Fowler SE, Haffner SM, et al.
The Diabetes Prevention Program Research Group: role of insulin secretion
and sensitivity in the evolution of type 2 diabetes in the diabetes prevention
program. Diabetes 2005;54:2404–14.
[29] Kahn SE. The relative contributions of insulin resistance and b-cell dysfunction
to the pathophysiology of type 2 diabetes. Diabetologia 2003;46:3–19.
[30] Ergin T, Lembet A, Duran H, Kuscu E, Bagis T, Saygili E, et al. Does insulin
secretion in patients with one abnormal glucose tolerance test value mimic
gestational diabetes mellitus? Am J Obstet Gynecol 2002;186:204–9.
[31] Wareham NJ, Phillips DI, Byrne CD, Hales CN. The 30 min insulin incremental
response in an oral glucose tolerance test as a measure of insulin secretion.
Diabetes Med 1995;12:931.
[32] Retnakaran R, Hanley AJ, Raif N, Hirning CR, Connely PW, Sermer M, et al.
Adiponectin and b-cell dysfunction in gestational diabetes: pathophysiological
implications. Diabetologia 2005;48:993–1001.
[33] Kauffman RP, Castracane VD, Peghee P, Baker TE, Van Hook JW. Detection of
gestational diabetes mellitus by homeostatic indices of insulin sensitivity: a
preliminary study. Am J Obstet Gynecol 2006;194:1576–82.
[34] Osei K, Cottrell DA, Harris B. Differences in basal and poststimulation glucose
homeostasis in nondiabetic first-degree relatives of black and white patients
with type 2 diabetes mellitus. J Clin Endocrinol Metab 1992;75:82–6.
[35] Preeyasombat C, Bacchetti P, Lazar AA, Lustig RH. Racial and etiopathologic
dichotomies in insulin hypersecretion and resistance in obese children. J
Pediatr 2005;146:474–81.
[36] Hillier TA, Pedula KL, Vesco KK, Schmidt MM, Mullen JA, LeBlanc ES, et al.
Excess gestational weight gain: modifying fetal macrosomia risk associated
with maternal glucose. Obstet Gynecol 2008;112(5):1007–14.
[37] Retnakaran R, Qi Y, Sermer M, Connelly PW, Hanley AJ, Zinman B. Glucose
intolerance in pregnancy and future risk of pre-diabetes or diabetes. Diabetes
Care 2008;31:2026–31.
[38] Lapolla A, Dalfrà MG, Bonomo M, Castiglioni MT, Di Cianni G, Masin M, et al.
Can plasma glucose and HbA1c predict fetal growth in mothers with different
glucose tolerance levels? Diabetes Res Clin Pract 2007;77:465–70.
[39] Yogev Y, Langer O, Xenakis EM, Rosenn B. The association between glucose
challenge test, obesity and pregnancy outcome in 6390 non-diabetic women. J
Matern Fetal Neonatal Med 2005;17:29–34.