Short stature and gestational diabetes in Brazil

Aims/hypothesis. To examine the association between maternal stature and ge stational diabetes mellitus. Methods. We studied a sample of 5564 consecutive Brazilian women 20 or more years old, who were pregnant for approximately 21-28 weeks, had no history of diabetes outside pregnancy and were attending general prenatal care uni ts in six state capitals in Brazil from 1991 to 1995. We did a 2-h, 75-g or al glucose tolerance test, defining gestational diabetes by World Health Or ganisation criteria. Results. Those in the shortest quartile of height (less than or equal to 15 1 cm) had a 60 % increase in the odds of having gestational diabetes, indep endently of prenatal clinic, age, global obesity, family history of diabete s, skin colour, referral pattern, waist circumference, parity, previous ges tational diabetes, education, ambient temperature and gestational age compa red with the tallest quartile [odds ratio (OR)= 1.60, p = 0.005]. This asso ciation was observed for those with above median values of skinfold thickne ss (OR = 1.74, p = 0.006) but not for those with below median values (OR = 1.22, p = 0.51). Associations of short stature with high 2-h glycaemia (gre ater than or equal to 7.8 mmol/l) (OR = 1.61, p = 0.005) were essentially t he same as those for gestational diabetes. There was, however, no associati on between short stature and gestational hyperglycaemia when the latter was defined exclusively by fasting values (OR = 0.97, p = 0.90). Conclusion/interpretation. In Brazil short stature associates with gestatio nal diabetes, principally in women with greater fat mass. This difference i n glycaemic levels is present postprandially but not in the fasting state.