Increased hemoglobin A(1c) in obese pregnant women after exclusion of gestational diabetes.

Preconception obesity is a strong predictor of glucose intolerance during pregnancy that has the potential to lead to adverse maternal and offspring effects. An increased prevalence of overweight offspring of obese mothers has been reported, despite normal values in an oral glucose tolerance test (OGTT) during pregnancy (1 ). Such findings might reflect effects of the intrauterine milieu, genetic background, or lifestyle factors related to maternal obesity or, alternatively, failure to diagnose glucose intolerance in obese pregnant women. We investigated whether increases in hemoglobin A1c (Hb A1c) at delivery as a surrogate marker of glycometabolic disturbances are prevalent in obese mothers, despite the exclusion of gestational diabetes (GDM) even after applying the stringent new criteria as recommended by the International Association of Diabetes and Pregnancy Study Groups (IADPSG). We performed a retrospective analysis of 137 obese pregnant Caucasian women. Recruitment was through an ongoing cohort study on the long-term effects of preconception maternal obesity (PEACHES, Programming of Enhanced Adiposity Risk in CHildhood – Early Screening), which had been initiated in 15 recruiting maternity clinics in Bavaria, Germany. Inclusion criteria before delivery were an age 18 years, singleton pregnancy, full-term birth, and a pregestation body mass index (BMI) 30 kg/m. Women with a preexisting diagnosis of type 1 or type 2 diabetes mellitus or other chronic diseases were excluded. Ethics approval was obtained from the Ludwig-Maximilians-Universität München, and informed consent was provided by each participant. The main exposure was the diagnosis or exclusion of GDM according to the results of a 75-g OGTT and the new IADPSG criteria or the former American Diabetes Association (ADA) criteria (Table 1). The primary outcome was an Hb A1c value 5.7% at delivery, a threshold that has been reported to be above the reference interval for late pregnancy (2, 3). Maternal EDTA-containing plasma samples were analyzed for Hb A1c via cationexchange chromatography with a Tosoh G8 HPLC Analyzer (Tosoh Bioscience) in a central laboratory (interassay CV, 2.0%; analytical bias, 0.03% Hb A1c at a target of 5.44%). A complete blood count was performed at delivery. Within subgroups, we compared the prevalence of increases in Hb A1c by calculating the proportions with binomial 95% CIs (Table 1) and SAS software (version 9.2; SAS Institute). According to the new IADPSG criteria, 38.9% (95% CI, 30.7%– 47.5%) of the women had a GDM diagnosis, whereas the former ADA criteria yielded a percentage of only 8.0% (95% CI, 4.1%– 13.9%). Subgroups did not differ significantly with respect to prepregnancy BMI, smoking during pregnancy, week of gestation of the OGTT, and parity. Iron deficiency anemia at delivery (4 ) was excluded by a complete blood count [mean corpuscular hemoglobin, 28.1 pg (median); mean corpuscular volume, 83.8 fL (median)]. The main finding was an increased prevalence of an increased Hb A1c of 5.7% at delivery in obese women— even considering the lower limit of the 95% CI— despite the exclusion of GDM even by the sensitive new IADPSG criteria (Table 1). Studies of Hb A1c levels during pregnancy for 500 pregnancies (2, 3 ) have yielded late-pregnancy estimates of the upper Hb A1c reference limit of 5.6%, compared with 6.2% in nonpregnant women. From these crosssectional data, we used an Hb A1c value of 5.7% at delivery as a cutoff for potentially undiagnosed glycometabolic disturbances. An Hb A1c value above this threshold at delivery was associated with increased risks of newborns being large for their gestational age (odds ratio, 3.1; 95% CI, 1.3–7.6) and having hypoglycemia (odds ratio, 6.2; 95% CI, 1.3–29.0) (2 ), suggesting a potential role for Hb A1c as a surrogate risk marker. Indeed, recent research points toward a role for Hb A1c as a risk marker of cardiovascular disease and a predictor of mortality in nondiabetic adult populations, results that suggest that abnormal glycation, even at a subdiabetic level, may already reflect a considerable health risk (5 ). Besides indicating glycemic control, Hb A1c variation may be due to nonglycemic factors that lead to states of high or low glycation, such as ethnic disparities and disease conditions, including those that affect erythrocyte life span (4 ). In addition to biological variation, the potential influence of analytical variation in Hb A1c measurements on decision thresholds needs cautious interpretation. In our study, both analytical and preanalytical variation was low, and we included only Caucasian mothers and excluded relevant microcytic hypochromic anemia. Our finding that the prevalence of increased Hb A1c values at delivery is still increased in seemingly nondiabetic obese preg1 Nonstandard abbreviations: OGTT, oral glucose tolerance test; Hb A1c, hemoglobin A1c; GDM, gestational diabetes; IADPSG, International Association of Diabetes and Pregnancy Study Groups; PEACHES, Programming of Enhanced Adiposity Risk in CHildhood – Early Screening (study), BMI, body mass index; ADA, American Diabetes Association. Clinical Chemistry 58:7 1152–1160 (2012) Letters to the Editor