PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of capmcAbout usÀ propos de nous
 
J Clin Endocrinol Metab. Author manuscript; available in PMC 2010 May 26.
Published in final edited form as:
PMCID: PMC2877113
CAMSID: CAMS819

The Antepartum Glucose Values that Predict Neonatal Macrosomia Differ from Those that Predict Postpartum Prediabetes or Diabetes: Implications for the Diagnostic Criteria for Gestational Diabetes

Abstract

Background/Objective

The diagnosis of gestational diabetes mellitus on oral glucose tolerance test (OGTT) is used to identify risk of both neonatal large-for-gestational-age (LGA) and maternal postpartum prediabetes/diabetes. An assumption inherent in this practice, however, is that the glucose values that define gestational diabetes mellitus on the OGTT relate to both of these outcomes in the same way. Thus, to test this assumption, we sought to evaluate the predictive capacity of each glucose value on antepartum OGTT in relation to LGA and postpartum prediabetes/diabetes.

Design/Setting/Participants

A total of 412 women representing the full spectrum of antepartum glucose tolerance underwent 3-h OGTT in pregnancy, assessment of obstetrical outcome at delivery, and 2-h OGTT at 3 months postpartum.

Results

Of the four glucose values (fasting, 1h, 2 h, 3 h) on antepartum OGTT, only the fasting measure was a significant predictor of LGA [odds ratio (OR) 2.00 per mmol/liter, 95% confidence interval (CI) 1.20–3.34] (P = 0.0076). In contrast, all three postload glucose values were significant predictors of postpartum prediabetes/diabetes (1 h glucose: OR 1.37, 95% CI 1.17–1.61, P < 0.0001; 2 h glucose: OR 1.55, 95% CI 1.32–1.83, P < 0.0001; 3 h glucose: OR 1.30, 95% CI 1.10–1.53, P = 0.002), whereas fasting glucose was not. Furthermore, whereas fasting glucose had the highest area under the receiver operating characteristic curve for predicting LGA (0.62), the 1- and 2-h glucose measures had the highest area under the receiver operating characteristic curve values for postpartum prediabetes/diabetes (0.68 and 0.72, respectively).

Conclusions

On antepartum OGTT, the fasting glucose value best predicts LGA risk, whereas postload glucose values predict postpartum prediabetes/diabetes. These relationships may have implications for the glycemic thresholds that define obstetrical and metabolic risk.

The clinical significance of diagnosing gestational diabetes mellitus (GDM) lies in the recognition that affected women are at risk of both adverse obstetrical outcomes, particularly related to fetal overgrowth, and the future development of type 2 diabetes mellitus (T2DM) (1). In 1964 when O’Sullivan and Mahan (2) proposed the initial glycemic thresholds for diagnosing GDM on oral glucose tolerance test (OGTT), their criteria were defined on the basis of identifying those women at risk of ultimately developing T2DM. Since that time, however, the ongoing debate regarding the appropriate diagnostic criteria for GDM has largely focused on the detection of fetal overgrowth and its associated obstetrical complications, leading to various sets of diagnostic criteria proposed by different organizations (including the National Diabetes Data Group, the American Diabetes Association, and the World Health Organization) (35). Importantly, in current clinical practice, regardless of the criteria used, the diagnosis of GDM is generally interpreted as an indicator of risk of both fetal overgrowth and future T2DM (leading to treatment with glucose lowering therapy in pregnancy and the recommendation for postpartum glucose tolerance screening, respectively). Thus, with GDM, the situation exists wherein a single set of diagnostic criteria on OGTT in pregnancy is used to identify women at risk of two very different adverse outcomes.

An assumption inherent in this practice is that the glucose values that define GDM on the OGTT relate to both of the outcomes of interest in the same way (i.e. such that a single set of diagnostic criteria can optimally capture both obstetrical and postpartum metabolic risk). Indeed, because there are many non-glycemic determinants of fetal overgrowth [particularly maternal obesity (6)], we hypothesized that the individual glucose measures on antepartum OGTT may not relate to both fetal overgrowth and postpartum glycemia in the same way. Thus, to test this hypothesis, our objective in this study was to evaluate the predictive capacity of each glucose value on antepartum OGTT in relation to neonatal large for gestational age (LGA) and post-partum prediabetes/diabetes in a cohort of women representing the full spectrum of glucose tolerance in pregnancy (from normal to impaired to GDM).

Patients and Methods

This analysis was conducted in the context of an ongoing observational study of early events in the natural history of T2DM, in which a cohort of women recruited at the time of antepartum screening for GDM is undergoing longitudinal metabolic characterization in pregnancy and the postpartum period (7, 8). Standard clinical care at our institution involves universal screening for GDM in all pregnant women at 24–28 wk gestation by 50 g glucose challenge test (GCT), followed by referral for a diagnostic OGTT if the GCT is abnormal (plasma glucose ≥7.8 mmol/liter at 1 h). In the current study, regardless of the GCT result, all participants underwent a 3-h, 100-g OGTT for assessment of glucose tolerance status in pregnancy. Recruitment was performed either before or after the GCT but before the OGTT. It should be noted that the recruitment of women after an abnormal GCT enriched the study population for women with varying degrees of antepartum glucose intolerance. At 3 months postpartum, participants returned for reassessment by 2-h, 75-g OGTT. The study protocol was approved by the Mount Sinai Hospital Research Ethics Board, and all participants have given written informed consent. As previously described (7), 487 women had completed both the pregnancy OGTT and the 3-month postpartum OGTT by September 2007. For the current analysis, the study population was restricted to women of Caucasian, Asian, or South Asian ethnicity because these are the three groups for which Canadian-based ethnicity-specific birth weight centiles are available (9, 10). Because multiple gestation pregnancy (i.e. twins) can affect fetal growth, the analysis was further restricted to those women with singleton pregnancies, yielding a study population of 412 women.

Participant assessments

On the morning of the OGTT in pregnancy, data regarding medical, obstetrical, and family history were collected by interviewer-administered questionnaire. As described previously (7), National Diabetes Data Group (NDDG) criteria (11) applied to the OGTT stratified subjects into the following four glucose tolerance groups in pregnancy: 1) GDM (defined by exceeding two or more NDDG glycemic thresholds); 2) gestational-impaired glucose tolerance [GIGT; a designation that was not originally described by NDDG but that we have previously applied (7, 8) to describe those women exceeding only one NDDG glycemic threshold on the OGTT]; 3) normal glucose tolerance (NGT) with an abnormal preceding GCT; and 4) NGT with a normal preceding GCT. Women with GDM received glucose-lowering treatment in pregnancy, consisting of dietary counseling ±insulin therapy. As per standard clinical practice at our institution, women in the other three glucose tolerance groups did not receive this treatment.

At delivery, data on obstetrical outcome were entered into a database that tracks labor and delivery data at Mount Sinai Hospital. LGA was defined as birth weight for gestational age above the 90th percentile of Canadian fetal growth curves for the ethnic group under study (Caucasian, Asian, or South Asian) (9, 10).

At 3 months postpartum, participants returned for a 2-h, 75-g OGTT, on which glucose tolerance status was defined according to Canadian Diabetes Association guidelines (12). Prediabetes refers to impaired glucose tolerance (IGT), impaired fasting glucose (IFG), or combined IFG and IGT. Postpartum glucose intolerance collectively refers to prediabetes and diabetes, as described earlier (7).

Statistical analyses

All analyses were conducted using the Statistical Analysis System (SAS, version 9.1; SAS Institute, Cary NC). Continuous variables were tested for normality of distribution, and natural log transformations of skewed variables were used, where necessary, in subsequent analyses. In Table 1, continuous variables are presented as mean followed by SD if normally distributed or median followed by interquartile range if skewed. Categorical variables are presented as percentages. Logistic regression analysis (Table 2) was performed to evaluate the effect of a 1 mmol/liter increase in any glucose value (fasting, 1 h, 2 h, or 3 h) from the antepartum OGTT on the following outcomes: neonatal LGA and postpartum glucose intolerance (i.e. prediabetes or diabetes). Receiver-operating-characteristic (ROC) analysis was performed to assess the discriminative capacity of each of the antepartum OGTT glucose values for predicting these two outcomes (LGA in Fig. 1A and postpartum glucose intolerance in Fig. 1B). Comparisons of area under the ROC curve (AROC) were performed as described by DeLong et al. (13). As part of a sensitivity analysis, the ROC analysis for LGA was repeated with adjustment for glucose-lowering treatment. In addition, for each antepartum OGTT glucose value, the AROC from the adjusted LGA analysis was compared with that from the unadjusted LGA analysis. Finally, as a further sensitivity analysis, the ROC analyses were repeated with exclusion of the women with GDM (supplementary figure, published as supplemental data on The Endocrine Society’s Journals Online Web site at http://jcem.endojournals.org) to eliminate any potential effect of glucose-lowering treatment in pregnancy.

FIG. 1
ROC curves for each glucose value on antepartum OGTT for prediction of neonatal LGA (A) and postpartum glucose intolerance (i.e. prediabetes or diabetes) (B). In A, AROC values for prediction of LGA are 0.62 for fasting glucose (P < 0.05 vs. 1 ...
TABLE 1
Antepartum characteristics, obstetrical outcomes, and 3-month postpartum metabolic outcomes of the study population
TABLE 2
Comparison of glucose values on antepartum OGTT in the prediction of neonatal LGA and postpartum glucose intolerance (i.e. prediabetes or diabetes)

Results

Baseline characteristics, obstetrical outcomes, and postpartum metabolic outcomes

Table 1 shows the antepartum characteristics, obstetrical outcomes, and 3-month postpartum metabolic outcomes of the 412 study participants. The study population represented the full spectrum of glucose tolerance in pregnancy, ranging from normal GCT NGT (n = 82) to abnormal GCT NGT (n = 137) to GIGT (n = 81) to GDM (n = 112). Mean age was 34.1 yr and median prepregnancy body mass index (BMI) was 23.5 kg/m2. Of the women with GDM, 29 (25.9%) required insulin therapy. Mean infant birth weight in the study population was 3376 ±540 g and the prevalence of LGA was 10.9%. At 3 months post-partum, the prevalence of glucose intolerance was 15% (IGT 12.6%; IFG 0.5%, combined IFG and IGT 0.2%; diabetes 1.7%).

Relationships between antepartum glucose values and clinical outcomes

We next studied the relationships between the glucose values on the OGTT in pregnancy and the following clinical outcomes: delivery of an LGA neonate and postpartum glucose intolerance (i.e. prediabetes or diabetes). On logistic regression analysis (Table 2), the fasting glucose value emerged as a significant predictor of neonatal LGA [odds ratio (OR) 2.00 per 1 mmol/liter, 95% confidence interval (CI) 1.20–3.34] (P < 0.0076), whereas none of the postload glucose values (1, 2, 3 h) were significantly associated with this obstetrical outcome. In contrast, however, logistic regression analysis of the metabolic outcome of postpartum glucose intolerance showed the opposite pattern: all three postload glucose values were significant predictors of postpartum prediabetes/diabetes (1 h glucose: OR 1.37, 95% CI 1.17–1.61, P < 0.0001; 2 h glucose: OR 1.55, 95% CI 1.32–1.83, P < 0.0001; 3 h glucose: OR 1.30, 95% CI 1.10–1.53, P = 0.002), whereas fasting glucose was not a significant predictor.

Although the primary interest in the current analysis is the unadjusted relationship between glucose values on the antepartum OGTT and each of the two outcomes under study, we also queried the potential effects of covariates. In Table 2B, the logistic regression analyses were adjusted for covariates at the time of the OGTT, including age, weeks gestation, prepregnancy BMI, weight gain in pregnancy preceding the OGTT, parity, and previous GDM/macrosomia in an earlier pregnancy. On these adjusted analyses, all three postload glucose values remained significant predictors of postpartum prediabetes/diabetes (1 h glucose: OR 1.31, 95% CI 1.09–1.57, P = 0.0036; 2 h glucose: OR 1.54, 95% CI 1.28–1.84, P < 0.0001; 3 h glucose: OR 1.30, 95% CI 1.09–1.55, P = 0.0030), whereas fasting glucose was not a significant predictor (OR 1.17, 95% CI 0.68–2.00, P = 0.5754). For the outcome of neonatal LGA, the previously significant association of fasting glucose was now attenuated (OR 1.68, 95% CI 0.91–3.10, P = 0.099), reflecting the dominant impact of prepregnancy BMI, which was the only significant independent predictor of LGA on the adjusted analyses (data not shown), consistent with earlier reports (6, 14, 15).

Having demonstrated differences in the predictive relationships between the glucose values from antepartum OGTT and neonatal LGA and postpartum glucose intolerance, respectively, we next conducted ROC analyses to assess the discriminative capacity of these antepartum glucose values for the prediction of these two clinical outcomes. As shown in Fig. 1, the predictive characteristics of the antepartum glucose values were strikingly different for these two outcomes. For the prediction of LGA (Fig. 1A), the AROC for fasting glucose (0.62) exceeded that of the postload glucose values [1 h glucose: 0.52 (pairwise P = 0.033), 2 h glucose: 0.53 (pairwise P = 0.067), 3 h glucose: 0.57], consistent with the superiority of the fasting value as a predictor of LGA. In contrast, however, when predicting postpartum prediabetes/diabetes (Fig. 1B), the fasting value was the antepartum glucose measure with the smallest AROC, easily surpassed by the postload values. Indeed, the AROC of the 2-h glucose value (0.72) and the AROC of the 1-h glucose value (0.68) were both significantly higher than that of fasting glucose (0.57) (pairwise comparisons: P = 0.003 and P = 0.013, respectively), reflecting the markedly superior discriminative capacity of the postload values in the prediction of postpartum prediabetes/diabetes.

Sensitivity analysis

Recognizing that glucose-lowering therapy for GDM can decrease neonatal birth weight and hence the risk of LGA, we conducted a sensitivity analysis to determine whether the treatment of women with GDM may have differentially affected the relationships between individual glucose values on the antepartum OGTT (i.e. fasting, 1 h, 2 h, 3 h) and the neonatal LGA outcome. On ROC analysis adjusted for glucose-lowering treatment in pregnancy (dietary therapy or insulin), the AROC of fasting glucose (0.65) continued to exceed that of the postload values in the prediction of LGA (1 h glucose 0.60, 2 h glucose 0.58, 3 h glucose 0.60) (data not shown). Furthermore, for each glucose value, when its AROC from this adjusted analysis was compared with its AROC on the analogous unadjusted analysis (i.e. Fig. 1A), there were no significant differences (fasting glucose: P = 0.345; 1 h glucose: P = 0.160; 2 h glucose: P = 0.343; 3 h glucose: P = 0.330). Thus, whereas it can reduce birth weight and the risk of LGA, it appears that glucose-lowering therapy in pregnancy did not differentially affect the relationships between the individual glucose values on the antepartum OGTT and the outcome of neonatal LGA.

Finally, to address the potential confounding effect of glucose-lowering therapy in a different way, we repeated the ROC analyses after restricting the data set to only the women without GDM (n = 300) (i.e. none of these women received glucose lowering therapy). For the prediction of LGA (supplementary figure A), the AROC for fasting glucose (0.63) again exceeded that of the postload glucose values. Nevertheless, as before, when predicting postpartum glucose intolerance (supplementary figure B), the AROC for fasting glucose (0.52) was lower than that of the postload glucose values. Thus, as observed in the full data set, fasting glucose was associated with LGA, whereas the post-load glucose values related to postpartum prediabetes/diabetes.

Discussion

In this report, we demonstrate that, on antepartum OGTT, the fasting glucose value best predicts risk of neonatal LGA, whereas the postload glucose values (particularly at 1 and 2 h) predict postpartum prediabetes/diabetes. These differences between the antepartum glycemic determinants of fetal overgrowth and post-partum glucose intolerance are readily apparent, statistically on both logistic regression and AROC analyses and graphically, with the ROC curves in Fig. 1. Importantly, these findings may have implications for the ongoing debate regarding diagnostic criteria for GDM, in that, without considering this issue, a single set of glycemic thresholds on antepartum OGTT may not optimally capture both obstetrical and metabolic risk.

The current analysis has three key strengths compared with preceding studies in this area. First, to our knowledge, this analysis is unique in its objective of evaluating the predictive capacity of each individual glucose measure on antepartum OGTT in relation to both fetal macrosomia and postpartum glucose intolerance (i.e. previous studies have focused on either obstetrical or metabolic risk). Second, as shown in Table 1, the current study population provided relatively balanced representation across the full spectrum of glucose tolerance in pregnancy (from normal to impaired to GDM), a feature not typically present in other studies that have generally focused on GDM or abnormal glucose tolerance (1418). This broad range is required for defining the relationship between glucose values and the outcomes of interest across the full glycemic spectrum that may be observed clinically on the OGTT. Third, in the current study, all women, regardless of their glucose tolerance status in pregnancy, underwent glucose tolerance testing at 3 months postpartum, a departure from previous studies in which such testing was performed only in women with GDM (1618). This feature enables characterization of the antepartum glycemic predictors of postpartum prediabetes/diabetes across the full range of glucose tolerance status in pregnancy (rather than just in women with GDM), which is again relevant to the utility of the OGTT for this purpose in the general population.

Design differences notwithstanding, our study nevertheless yields findings that are consistent with earlier observations. Specifically, previous studies have reported that the postload glucose values on antepartum OGTT are superior to the fasting value in predicting postpartum diabetes, although these analyses were conducted solely in women with GDM (1618). Similarly, earlier reports noted that the fasting glucose value was better than the postchallenge glucose levels in identifying macrosomic risk (14, 15, 19, 20), although other factors (most notably maternal obesity) were generally more important than any of the glycemic measures in this regard (14, 15). Indeed, when one considers that maternal obesity is an important nonglycemic determinant of fetal overgrowth (6, 14, 15) whereas the OGTT in pregnancy is directly testing the very same physiologic axis that determines postpartum glucose tolerance, the current demonstration of significant differences in the antepartum glycemic predictors of macrosomia and postpartum prediabetes/diabetes is perhaps not surprising.

Because the objective of the current analysis was to evaluate the predictive capacity of each glucose value on antepartum OGTT in relation to LGA and postpartum prediabetes/diabetes, the primary analyses of interest are the unadjusted relationships between the glucose measures and these outcomes. Nevertheless, we also conducted adjusted analyses to evaluate the potential effect of covariates and to test our hypothesis that differences in the underlying determinants of the outcomes may explain the observed differences in glycemic predictors. These adjusted analyses showed that the postload glucose values were robustly associated with postpartum prediabetes/diabetes after adjustment, consistent with the concept that the OGTT in pregnancy is directly testing the very same physiologic axis that determines postpartum glucose tolerance. On the other hand, as expected, the association between fasting glucose and LGA was attenuated on adjustment for prepregnancy BMI, again consistent with the previously described significance of maternal obesity as a key determinant of fetal overgrowth (6, 14, 15). Thus, taken together, the adjusted analyses further support our explanation (i.e. that of different underlying determinants) for why significant differences in the antepartum glycemic predictors of macrosomia and postpartum prediabetes/diabetes are to be expected.

The current findings are also biologically plausible. Specifically, because IGT represents the majority of postpartum prediabetes (7), it is fully reasonable that the postload glucose values on antepartum OGTT would be most predictive of this outcome (because postload glycemia on both antepartum and postpartum testing may be reflecting the same defect in glucose homeostasis). In addition, in the Hyperglycemia and Adverse Pregnancy Outcomes Study, although fasting glucose was not a stronger predictor of macrosomia than the postload values on adjusted analyses (i.e. after adjustment for covariates including obesity), it was the strongest predictor of cord-blood serum C-peptide levels, which reflect the fetal insulinemia through which maternal hyperglycemia is believed to cause fetal overgrowth (21). Finally, it should be noted that the current demonstration that the risks associated with fasting glycemia in pregnancy differ from those related to postload glycemia may be analogous to the emerging concept that, in the nonpregnant state, IFG and IGT represent metabolically distinct conditions (22).

The clinical implications of the findings reported herein may relate to the longstanding debate regarding diagnostic criteria for GDM on antepartum OGTT. Whereas it is anticipated that the Hyperglycemia and Adverse Pregnancy Outcomes Study will soon provide recommendations for new diagnostic criteria based on fetal macrosomic risk, the current report raises the possibility that different thresholds may be needed for detecting postpartum risk of diabetes. Indeed, it may be that, on antepartum OGTT, one set of glycemic criteria will be needed for determining macrosomic risk (and hence the need for glucose lowering therapy in pregnancy), whereas a second set of criteria will be required for identifying those women who are at greatest risk of ultimately developing T2DM (and hence warrant postpartum glucose testing). Because definitive clinical recommendations cannot be provided at this time based on the current initial report, further study is needed to address the possibility that such a dual approach to the antepartum OGTT may be required to optimally capture both obstetrical and metabolic risk.

A limitation of this analysis is that we cannot determine optimal glycemic thresholds for detecting macrosomic risk from this study, owing to the fact that women with GDM by NDDG criteria needed to be treated with diet or insulin to lower glucose levels in pregnancy, as per our institutional standard of care. Although this treatment can lower the risk of LGA, our sensitivity analysis indicates that it does not differentially affect the relationships between the individual glucose values on antepartum OGTT and neonatal LGA. Thus, this treatment has not precluded the current demonstration of significant differences in the glycemic predictors of obstetrical and postpartum metabolic risk. It should also be recognized that distinction of the respective risks associated with fasting and postload glycemia may be complicated by the fact that women with fasting hyperglycemia are more likely to also have postload hyperglycemia. In this regard, it is encouraging that, of the 112 women with GDM in our study, only seven had abnormal fasting glucose values.

In summary, on antepartum OGTT, the fasting glucose value best predicts risk of neonatal LGA, whereas the postload glucose values predict postpartum prediabetes/diabetes. It thus emerges that, without considering these relationships, diagnostic criteria for GDM may not optimally capture both obstetrical and metabolic risk. Indeed, ultimately, it may be that two sets of separate criteria need to be applied to the antepartum OGTT to optimally detect the respective risks of these two distinct outcomes.

Acknowledgments

We thank the Mount Sinai Hospital Department of Pathology and Laboratory Medicine and Patient Care Services.

This work was supported by Operating Grants MOP 67063 and 84206 from the Canadian Institutes of Health Research (CIHR). R.R is supported by a CIHR Clinical Research Initiative New Investigator Award, Canadian Diabetes Association (CDA) Clinician-Scientist incentive funding, and a University of Toronto Banting and Best Diabetes Centre New Investigator Award. A.J.G.H. holds a Tier II Canada Research Chair in Diabetes Epidemiology and is supported through a CDA Scholarship. B.Z. holds the Sam and Judy Pencer Family Chair in Diabetes Research at Mount Sinai Hospital and University of Toronto.

Abbreviations

AROC
Area under the ROC curve
BMI
body mass index
CI
confidence interval
GCT
glucose challenge test
GDM
gestational diabetes mellitus
GIGT
gestational-impaired glucose tolerance
IFG
impaired fasting glucose
IGT
impaired glucose tolerance
LGA
large for gestational age
NDDG
National Diabetes Data Group
NGT
normal glucose tolerance
OGTT
oral glucose tolerance test
OR
odds ratio
ROC
receiver-operating-characteristic
T2DM
type 2 diabetes mellitus

Footnotes

Disclosure Statement: The authors have nothing to disclose.

References

1. Buchanan TA, Xiang AH. Gestational diabetes mellitus. J Clin Invest. 2005;115:485–491. [PMC free article] [PubMed]
2. O’Sullivan JB, Mahan CM. Criteria for the oral glucose tolerance test in pregnancy. Diabetes. 1964;13:278–285. [PubMed]
3. Hanna FW, Peters JR. Screening for gestational diabetes: past, present and future. Diabet Med. 2002;19:351–358. [PubMed]
4. Ferrara A, Weiss NS, Hedderson MM, Quesenberry CP, Jr , Selby JV, Ergas IJ, Peng T, Escobar GJ, Pettitt DJ, Sacks DA. Pregnancy plasma glucose levels exceeding the American Diabetes Association thresholds, but below the National Diabetes Data Group thresholds for gestational diabetes mellitus, are related to the risk of neonatal macrosomia, hypoglycaemia and hyperbilirubinaemia. Diabetologia. 2007;50:298–306. [PubMed]
5. Agarwal MM, Dhatt GS, Punnose J, Koster G. Gestational diabetes: dilemma caused by multiple international diagnostic criteria. Diabet Med. 2005;22:1731–1736. [PubMed]
6. Ricart W, López J, Mozas J, Pericot A, Sancho MA, González N, Balsells M, Luna R, Cortázar A, Navarro P, Ramírez O, Flández B, Pallardo LF, Hernández-Mijas A, Ampudia J, Fernández-Real JM, Corcoy R. Spanish Group for the Study of the Impact of Carpenter and Coustan GDM Thresholds. Body mass index has a greater impact on pregnancy outcomes than gestational hyperglycemia. Diabetologia. 2005;48:1736–1742. [PubMed]
7. Retnakaran R, Qi Y, Sermer M, Connelly PW, Hanley AJ, Zinman B. Glucose intolerance in pregnancy and future risk of prediabetes or diabetes. Diabetes Care. 2008;31:2026–2031. [PMC free article] [PubMed]
8. Retnakaran R, Qi Y, Sermer M, Connelly PW, Zinman B, Hanley AJ. Isolated hyperglycemia at 1 hour on oral glucose tolerance test in pregnancy resembles gestational diabetes in predicting postpartum metabolic dysfunction. Diabetes Care. 2008;31:1275–1281. [PMC free article] [PubMed]
9. Kramer MS, Platt RW, Wen SW, Joseph KS, Allen A, Abrahamowicz M, Blondel B, Breart G. (Fetal/Infant Health Study Group of the Canadian Perinatal Surveillance System) A new and improved population-based Canadian reference for birthweight for gestational age. Pediatrics. 2001;108:e35–e41. [PubMed]
10. Kierans WJ, Kramer MS, Wilkins R, Liston R, Foster L, Uh S-H. Charting birth outcome in British Columbia: determinants of optimal health and ultimate risk–an expansion and update. Victoria, British Columbia: British Columbia Vital Statistics Agency; 2003. ( http://www.vs.gov.bc.ca/stats/features/index.html)
11. National Diabetes Data Group. Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes. 1979;28:1039–1057. [PubMed]
12. Canadian Diabetes Association Clinical Practice Guidelines Expert Committee. Canadian Diabetes Association. 2008 clinical practice guidelines for the prevention and management of diabetes in Canada. Can J Diabetes. 2008;32(Suppl 1):S10–S13.
13. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a non-parametric approach. Biometrics. 1988;44:837–845. [PubMed]
14. Schaefer-Graf UM, Heuer R, Kilavuz O, Pandura A, Henrich W, Vetter K. Maternal obesity not maternal glucose values correlates best with high rates of fetal macrosomia in pregnancies complicated by gestational diabetes. J Perinat Med. 2002;30:313–321. [PubMed]
15. Ouzilleau C, Roy MA, Leblanc L, Carpentier A, Maheux P. An observational study comparing 2-h 75-g oral glucose tolerance with fasting plasma glucose in pregnant women: both poorly predictive of birth weight. CMAJ. 2003;168:403–409. [PMC free article] [PubMed]
16. Buchanan TA, Xiang AH, Kjos SL, Trigo E, Lee WP, Peters RK. Antepartum predictors of the development of type 2 diabetes in Latino women 11–26 months after pregnancies complicated by gestational diabetes. Diabetes. 1999;48:2430–2436. [PubMed]
17. Dalfrà MG, Lapolla A, Masin M, Giglia G, Dalla Barba B, Toniato R, Fedele D. Antepartum and early postpartum predictors of type 2 diabetes development in women with gestational diabetes mellitus. Diabetes Metab. 2001;27:675–680. [PubMed]
18. Weijers RN, Bekedam DJ, Goldschmidt HM, Smulders YM. The clinical usefulness of glucose tolerance testing in gestational diabetes to predict early postpartum diabetes mellitus. Clin Chem Lab Med. 2006;44:99–104. [PubMed]
19. Voldner N, Frøslie KF, Bo K, Haakstad L, Hoff C, Godang K, Bollerslev J, Henriksen T. Modifiable determinants of fetal macrosomia: role of lifestyle-related factors. Acta Obstet Gynecol Scand. 2008;87:423–429. [PubMed]
20. Kautzky-Willer A, Bancher-Todesca D, Weitgasser R, Prikoszovich T, Steiner H, Shnawa N, Schernthaner G, Birnbacher R, Schneider B, Marth Ch, Roden M, Lechleitner M. The impact of risk factors and more stringent diagnostic criteria of gestational diabetes on outcomes in central European women. J Clin Endocrinol Metab. 2008;93:1689–1695. [PubMed]
21. HAPO Study Cooperative Research Group. Metzger BE, Lowe LP, Dyer AR, Trimble ER, Chaovarindr U, Coustan DR, Hadden DR, McCance DR, Hod M, McIntyre HD, Oats JJ, Persson B, Rogers MS, Sacks DA. Hyperglycemia and adverse pregnancy outcomes. N Engl J Med. 2008;358:1991–2002. [PubMed]
22. Nathan DM, Davidson MB, DeFronzo RA, Heine RJ, Henry RR, Pratley R, Zinman B. American Diabetes Association. Impaired fasting glucose and impaired glucose tolerance: implications for care. Diabetes Care. 2007;30:753–759. [PubMed]