Seasonal variation associated with gestational diabetes diagnosis

Among Swedish women, gestational diabetes diagnoses peak during the summer months, likely due to an association with ambient temperature, according to study findings presented at 52nd European Association for the Study of Diabetes Annual Meeting.

“While seasonality in the onset of type 1 diabetes is well documented, less is known about the seasonality in the diagnosis of type 2 and gestational diabetes,” Anastasia Katsarou, of Lund University and Skåne University Hospital in Malm, Sweden, and colleagues wrote. “Temperature-induced increase in post-load glucose levels has been described, possibly related to increased arterialization of venous blood at higher ambient temperature. The aim of the present study was to examine seasonal pattern in glucose tolerance and in the diagnosis of [gestational diabetes].”

Katsarou and colleagues analyzed data from 11,538 women participating in the Mamma study who underwent a standard 75-g oral glucose tolerance test (OGTT) at 28 weeks’ gestation between 2003 and 2005 in southern Sweden. Researchers stratified OGTT results by month and season and used Chi-squared tests to determine differences in frequencies of gestational diabetes and ANOVA models to test for differences in mean glucose concentrations. Researchers used logistic regression analysis to determine whether month or season were associated with the diagnosis of gestational diabetes; linear regression analysis was used to examine the corresponding associations with 2-hour glucose levels. Mean monthly temperature information was obtained from the Swedish Meteorological and Hydrological Institute.

Within the cohort, 487 women (4.2%) were diagnosed with gestational diabetes. Monthly frequency ranged from 2.9% in March to 5.8% in June; seasonal frequency ranged from 3.3% in spring to 5.5% in summer. Differences were significant for both month (P < .01) and season (P < .0001), according to researchers.

Researchers observed a similar seasonal trend for mean 2-hour glucose concentrations, with a mean difference of 0.15 mmol/L between winter and summer (P < .0001). Mean monthly temperature ranged from -0.6 C in the winter to 17.7 C in the summer.

“In a simple linear regression [analysis] with 2-hour plasma glucose as the dependent variable and mean monthly temperature as the predictor variable, the coefficient in the equation was 0.009, suggesting that the 2-hour glucose level increased by 0.009 mmol/L for every degree increase in temperature,” the researchers wrote.

Researchers found that summer months were associated with an increased 2-hour glucose level in regression analysis after adjustment for age (P < .001) as well as increased frequency for gestational diabetes vs. other seasons (OR = 1.507; 95% CI, 1.241-1.829). However, the associations did not persist after adjustment for mean monthly temperature, according to the researchers.

“The prevalence of [gestational diabetes] and the 2-hour glucose values showed statistically significant variations during the months and the seasons, with a peak in the summer,” Katsarou said during her presentation. “A 0.15-mmol/L difference in glucose concentration between summer and winter would in fact increase the [gestational diabetes] frequency in summer.”

Katsarou noted during her presentation that the researchers did not have access to individual data other than age, and only had access to mean monthly, and not daily temperatures. – by Regina Schaffer

Reference: Katsarou A, et al. Abstract#86. Presented at: 52nd EASD Annual Meeting; Sept. 12-16, 2016; Munich.