Impaired Fasting Glucose and Impaired Glucose Tolerance in Women With Prior Gestational Diabetes Are Associated With a Different Cardiovascular Profile

This research was conducted in the Diabetes and Pregnancy Unit at the University Hospital La Paz in Madrid between 1992 and 2000. We studied 838 Caucasian women (mean age 32.4 ± 4.6 years; mean parity 1.8 ± 0.9) with singleton gestations complicated by GDM and who attended the initial postpartum assessment (3–6 months after delivery) when lactation was concluded. The total number of women with singleton gestations who had a diagnosis of GDM during this period was 1,350. Women who did not participate in the postpartum examination (37.9%) were similar to those who attended with regard to age, parity, prepregnancy BMI, and degree of glucose intolerance during pregnancy. Ethical permission for the study was received from the hospital ethical committee, and informed consent was obtained from all subjects.

The diagnosis of GDM was made using the criteria of the National Diabetes Data Group (11) after performing a fasting 3-h, 100-g oral glucose tolerance test (OGTT) in all pregnant women with a screening test (50-g oral glucose challenge) showing a 1-h glucose value ≥140 mg/dl (7.8 mmol/l). The screening test was performed at the first prenatal visit if clinical characteristics were consistent with a high risk of GDM (obesity, family history of diabetes, personal history of GDM, or poor obstetric outcome) or at 24–28 weeks of gestation when these characteristics were not present.

All women with a diagnosis of GDM received nutrition counseling with individualized caloric intake and providing 50% of calories from carbohydrates, 30% from fat, and 20% from protein. Self-glucose monitoring was performed by the patients on alternate days. Insulin therapy, given on the basis of a fasting blood glucose ≥105 mg/dl (5.8 mmol/l) or a 2-h postprandial blood glucose ≥120 mg/dl (6.7 mmol/l) on two or more occasions, was needed in 386 women (46.1%).

At postpartum assessment, a standard 75-g OGTT was given to 834 women (4 women had fasting glucose values that were diagnostic of diabetes). Reclassification of glycemic status was done according to the criteria of the World Health Organization (12) with the following modification: diabetes, fasting glucose ≥126 mg/dl (7.0 mmol/l) or 2-h glucose ≥200 mg/dl (11.1 mmol/l); IFG (isolated IFG), fasting glucose ≥110 mg/dl (6.1 mmol/l) and <126 mg/dl (7.0 mmol/l) and 2-h glucose <140 mg/dl (7.8 mmol/l); IGT (isolated IGT), fasting glucose <110 mg/dl (6.1 mmol/l) and 2-h glucose ≥140 mg/dl (7.8 mmol/l) and <200 mg/dl (11.1 mmol/l); IFG plus IGT, fasting glucose ≥110 mg/dl (6.1 mmol/l) and <126 mg/dl (7.0 mmol/l) and 2-h glucose ≥140 mg/dl (7.8 mmol/l) and <200 mg/dl (11.1 mmol/l); and normal, fasting glucose <110 mg/dl (6.1 mmol/l) and 2-h glucose <140 mg/dl (7.8 mmol/l). At the same time, a fasting blood sample was drawn for the measurement of triglyceride, cholesterol, and HDL cholesterol levels. Systolic (sBP) and diastolic (dBP) blood pressures were recorded with the patient in a sitting position after at least 5 min of rest, using a sphyngomanometer. The average of two blood pressure measurements was used for the analysis. Anthropometric measurements included weight, height, and waist and hip circumferences (13). BMI (kg/ m²) was computed from current weight and height, and waist-to-hip ratio (WHR) was computed from waist and hip circumferences.

Postpartum area under the glucose curve was calculated by the trapezoidal method. A woman was classified as having abnormal lipids if she had any of the following characteristics: triglyceride levels ≥150 mg/dl (1.69 mmol/l), HDL cholesterol levels ≤39 mg/dl (1.01 mmol/l), or total cholesterol levels ≥200 mg/dl (5.17 mmol/l). Obesity was defined as the presence of BMI ≥30 kg/m² and/or WHR ≥0.85. Hypertension was defined as the presence of sBP ≥140 mmHg and/or dBP ≥90 mmHg.

Plasma glucose was enzymatically measured on an automated analyzer (Hitachi 704; Boehringer-Mannheim, Indianapolis, IN). Cholesterol and triglyceride levels were determined using enzymatic assays (Kit CHOL-PAP and Kit GPO-PAP; Boehringer-Mannheim, Indianapolis, IN). HDL cholesterol levels were measured using an enzymatic assay after precipitation of LDLs and VLDLs with dextran sulfate (Kit CHOL-HDL; Sclavo Diagnostici, Sienna, Italy).

Statistical analyses were conducted using SPSS version 8.0 statistical software (SPSS, Chicago, IL). Values are reported as means ± SD or as percentages. Data were tested for normal distribution using the Kolmogorov-Smirnov test. To compare continuous variables among groups, one-way ANOVA was used (F test) and post hoc tests were performed using the Bonferroni method. Categorical variables were compared using χ² and Fisher’s exact tests of significance. Multiple-variable logistic regression models were used to estimate adjusted ORs. Association between variables was evaluated using linear correlation coefficients and stepwise multiple linear regression analysis. A P value <0.05 was considered significant.

At postpartum assessment of the 838 women with GDM, 681 women (81.3%) had a normal glycemic status, 30 (3.5%) had diabetes, 40 (4.8%) had IFG, 62 (7.4%) had IGT, and 25 (3%) had IFG plus IGT. Mean age was 32.5 ± 4.2 years for women with normal glycemic status, 32.6 ± 4.8 years for women with diabetes, 32.8 ± 5.1 years for women with IFG, 32.7 ± 4.7 years for women with IGT, and 32.8 ± 4.8 years for women with IFG plus IGT.

Potential cardiovascular risk factors by postpartum categories of glucose tolerance are presented in Table 1. Mean BMI, sBP, and dBP were significantly higher in women with IFG than in women with normal glucose status; these differences were not observed for women with IGT. Women with IFG plus IGT had significantly higher mean BMI and waist circumference than women with normal glucose status or with IGT. Mean sBP and dBP in women with IFG plus IGT were not significantly different from women with normal glucose status, women with IFG, or women with IGT. Mean BMI, waist circumference, dBP, and triglyceride were significantly higher in women with diabetes than in women with normal glucose status. Mean cholesterol and HDL cholesterol did not differ by postpartum glucose tolerance categories.

Table 2 shows linear correlation coefficients between potential cardiovascular risk factors and postpartum fasting glucose and area under the glucose curve. BMI, waist circumference, trygliceride, HDL cholesterol, and sBP and dBP were significantly related to both fasting glucose and area under the glucose curve. To test the independent associations, multiple linear regression analysis was used. Variables independently associated with the area under the glucose curve were waist circumference (β = 0.236; P = 0.0001) and triglycerides (β = 0.172; P = 0.0001). Variables independently associated with fasting glucose were BMI (β = 0.227; P = 0.0001), dBP (β = 0.121; P = 0.007), and HDL cholesterol (β = −0.084; P = 0.049).

The prevalence of cardiovascular risk factors among the different postpartum categories of glucose tolerance is provided in Table 3. The odds ratio for hypertension (3.7; P < 0.0001) and obesity (2.4; P = 0.009) was significantly increased in IFG compared with normal glucose status, whereas IGT showed no increased prevalence of cardiovascular risk factors. The IFG plus IGT category showed a significantly higher odds ratio for abnormal lipids (3; P = 0.036) and obesity (2.9; P = 0.024) compared with normal glucose status. The prevalence of obesity (OR 3.1; P = 0.003) and abnormal lipids (OR 3.3; P = 0.002) was significantly increased in diabetes compared with normal glucose status. Finally, we estimated the prevalence of two or more of the cardiovascular risk factors assessed among the different postpartum categories of glucose tolerance (Fig. 1). Only the IFG and diabetes categories showed a significantly increased prevalence; women with IFG had an OR of 3.7 (95% CI 1.6–8.9) and women with diabetes had an OR of 5.7 (95% CI 2.3–14) for two or more cardiovascular risk factors.

At postpartum glycemic assessment using an OGTT, ∼20% of the women with prior GDM had impaired glucose metabolism, and ∼4% had diabetes, in agreement with previous data (14).

According to the 1997 American Diabetes Association criteria (10), impaired glucose metabolism comprises two categories: IFG and IGT. Since the establishment of these two intermediate metabolic stages, their identity with future diabetes and with cardiovascular risk has been put forward. A modification in the criteria used to ascertain these two intermediate categories is required for an accurate evaluation of the consequences of fasting hyperglycemia and postprandial hyperglycemia. Hence, it is necessary to take into account isolated IFG with normal glucose tolerance, isolated IGT with normal fasting glucose, and the occurrence of both IFG and IGT (IFG plus IGT). Finally, it must be remembered that the limited repeat test reproducibility of fasting and 2-h glucose levels can lead to different classifications of an individual when tested more than once (15).

With the scheme proposed, women with IFG and IGT at the postpartum evaluation showed similar lipid levels, blood pressure values, BMI, and waist circumference, whereas women with IFG plus IGT had higher mean BMI and waist circumference than women with IGT. Comparison of women with impaired glucose metabolism with women with normal glycemic status disclosed that only women with IFG and women with IFG plus IGT were different in relation to potential cardiovascular risk factors. Regarding the prevalence of each cardiovascular risk factor, obesity and abnormal lipids were more common in women with diabetes than in women with normal glucose status. As for the intermediate metabolic stages, IGT showed no increased prevalence of cardiovascular risk factors, whereas obesity and hypertension were more frequent in IFG, and obesity and abnormal lipids occurred more often in IFG plus IGT. In a similar way, IFG but not IGT had a higher prevalence of two or more cardiovascular risk factors.

Recent studies have established an epidemiological relation between postprandial hyperglycemia and mortality from coronary heart disease in type 2 diabetes (16). Moreover, long-term morbidity and mortality from coronary heart disease appeared to be higher in IGT than in IFG (17). Finally, the DECODE study (18) showed that post-challenge glycemia, compared with fasting glucose, had a higher predictive value for the occurrence of cardiovascular complications.

When cardiovascular risk factors have been analyzed across the categories of impaired glucose metabolism, the results have not been so conclusive. De Pablos-Velasco et al. (19) showed that in a Caucasian population, both IFG and IGT had higher prevalence of certain cardiovascular risk factors compared with normal glucose tolerance; and at the same time, IFG was associated with higher total-to-HDL cholesterol ratio than IGT. Among a Swedish population, Larsson et al. (20) did not find differences between IFG and IGT, whereas IFG plus IGT showed higher BMI, cholesterol, and triglyceride values than IGT and higher triglyceride values than IFG. To date, combined IFG plus IGT seems to be associated with the highest cardiovascular risk (21). Our data showed that BMI and waist circumference in the IFG plus IGT category were higher than in the normal glucose tolerance and in the IGT categories. Given the small number of women (n = 25) in the IFG plus IGT group in our study, it is possible that differences in the prevalence of two or more cardiovascular factors compared with the normal tolerance group were not detected.

Apart from the above discussion on the cardiovascular consequences of fasting or postprandial hyperglycemia, the overall degree of impaired glucose metabolism (assessed as the area under the glucose curve) was significantly associated with all the cardiovascular risk factors analyzed in this study (BMI, waist circumference, triglyceride, HDL-cholesterol, sBP, and dBP), except cholesterol.

In conclusion, our data indicate that IFG has a stronger association than IGT with traditional cardiovascular risk factors in women with prior GDM, although the underlying determinant of the relation may well be the overall degree of impaired glucose metabolism. Further data comparing cardiovascular risk factors across the categories of impaired glucose metabolism are desirable.