Abstract
Background and aims: Ethnicity influences the prevalence of gestationaldiabetes (GDM) and its progression to manifest diabetes postpartum, beinghigher in non-European populations. This may partly be explained by differences in insulin secretion and action. Aims of the present study were toevaluate glucose homeostasis after GDM, the impact of ethnicity and otherdeterminants of glucose tolerance postpartum.Material and methods: Women in southern Sweden undergoing a 75 g oralglucose tolerance test (OGTT) during pregnancy in 2003-2005 were invited to follow-up postpartum. Diagnostic criteria were those defined by theWHO in 1999. At 1-2 years after delivery 470 women with GDM and 166women with normal glucose tolerance (NGT) during pregnancy performedan OGTT with measurements of plasma glucose and insulin concentrationsat fasting, 30 min and 120 min. Homeostasis model assessment (HOMA-IR)was used to estimate insulin resistance. Beta cell function was quantified asthe ratio of the incremental insulin to glucose during the first 30 min of theOGTT (I/G30). The disposition index was used to adjust insulin secretion forthe degree of insulin resistance ([I/G30)]/HOMA-IR). Women were groupedaccording to ethnicity based on stated country of origin in at least three oftheir grandparents. Indices were log transformed and differences in meanswere tested by ANCOVA, adjusting for age, parity and interval to follow-up(results given as geometric mean [95% confidence interval (CI)]). Frequencydifferences were tested by the Chi-square test. Multivariate logistic regressionanalysis was used to assess the association of known predictor variables (age,BMI, parity, first degree relative(s) with diabetes, non-European origin) withdiabetes postpartum, adjusting for time to follow-up.Results: Comparing women with previous GDM (n=470) to controls (NGTduring pregnancy and follow-up, n=150), the former had higher HOMA-IR Diabetologia (2012) 55:[Suppl1]S1–S538 S 4411 C(1.5 [1.4-1.7] vs. 1.3 [1.2-1.5], p=0.020) and lower disposition index (8.4 [7.7-9.2] vs. 12.8 [10.8-15.2], p<0.001). These differences were more pronouncedin women with GDM who had diabetes postpartum (HOMA-IR 3.1 [2.2-4.4],disposition index 2.6 [1.9-3.7]) compared to controls (p<0.001), while thosewho stayed normoglycaemic had similar HOMA-IR as controls but lower disposition index (9.6 [8.7-10.6], p<0.001). Among women with GDM, estimatesof beta cell function did not differ between non-European (n=94) and European women (n=362), whereas non-European women were more insulin resistant (HOMA-IR 2.0 [1.7-2.3] vs. 1.5 [1.3-1.6], p=0.002, after adjustment forBMI p=0.015). Similarly, Arabic women (n=41) had higher HOMA-IR (2.1[1.6-2.7]) than European women (p=0.006), but insignificant after adjustment for BMI. Non-European origin was associated with higher frequency ofdiabetes at follow-up (16%) than was European origin (4%, p<0.001). Of thepredictor variables tested for an association with diabetes after GDM, BMIand non-European origin showed the highest associations; odds ratio (95%CI), 1.1 (1.1-1.2), p<0.001, and 5.3 (1.9-14.9), p=0.002, respectively.Conclusions: Women with a history of GDM display abnormalities in glucose homeostasis, also in the presence of NGT postpartum, including betacell dysfunction and insulin resistance. These derangements may be influenced by ethnicity and BMI.
Background and aims: Ethnicity influences the prevalence of gestational diabetes (GDM) and its progression to manifest diabetes postpartum, being higher in non-European populations. This may partly be explained by differences in insulin secretion and action. Aims of the present study were to evaluate glucose homeostasis after GDM, the impact of ethnicity and other determinants of glucose tolerance postpartum. Material and methods: Women in southern Sweden undergoing a 75 g oral glucose tolerance test (OGTT) during pregnancy in 2003-2005 were invited to follow-up postpartum. Diagnostic criteria were those defined by the WHO in 1999. At 12 years after delivery 470 women with GDM and 166 women with normal glucose tolerance (NGT) during pregnancy performed an OGTT with measurements of plasma glucose and insulin concentrations at fasting, 30 min and 120 min. Homeostasis model assessment (HOMA-IR) was used to estimate insulin resistance. Beta cell function was quantified as the ratio of the incremental insulin to glucose during the first 30 min of the OGTT (I/G30). The disposition index was used to adjust insulin secretion for the degree of insulin resistance ([I/G30)]/HOMA-IR). Women were grouped according to ethnicity based on stated country of origin in at least three of their grandparents. Indices were log transformed and differences in means were tested by ANCOVA, adjusting for age, parity and interval to follow-up (results given as geometric mean [95% confidence interval (CI)]). Frequency differences were tested by the Chi-square test. Multivariate logistic regression analysis was used to assess the association of known predictor variables (age, BMI, parity, first degree relative(s) with diabetes, non-European origin) with diabetes postpartum, adjusting for time to follow-up. Results: Comparing women with previous GDM (n=470) to controls (NGT during pregnancy and follow-up, n=150), the former had higher HOMA-IR (1.5 [1.41.7] vs. 1.3 [1.21.5], p=0.020) and lower disposition index (8.4 [7.79.2] vs. 12.8 [10.815.2], p<0.001). These differences were more pronounced in women with GDM who had diabetes postpartum (HOMA-IR 3.1 [2.24.4], disposition index 2.6 [1.93.7]) compared to controls (p<0.001), while those who stayed normoglycaemic had similar HOMA-IR as controls but lower disposition index (9.6 [8.710.6], p<0.001). Among women with GDM, estimates of beta cell function did not differ between non-European (n=94) and European women (n=362), whereas non-European women were more insulin resistant (HOMA-IR 2.0 [1.72.3] vs. 1.5 [1.31.6], p=0.002, after adjustment for BMI p=0.015). Similarly, Arabic women (n=41) had higher HOMA-IR (2.1 [1.62.7]) than European women (p=0.006), but insignificant after adjustment for BMI. Non-European origin was associated with higher frequency of diabetes at follow-up (16%) than was European origin (4%, p<0.001). Of the predictor variables tested for an association with diabetes after GDM, BMI and non-European origin showed the highest associations; odds ratio (95% CI), 1.1 (1.11.2), p<0.001, and 5.3 (1.914.9), p=0.002, respectively. Conclusions: Women with a history of GDM display abnormalities in glucose homeostasis, also in the presence of NGT postpartum, including beta cell dysfunction and insulin resistance. These derangements may be influenced by ethnicity and BMI.