Association between reduced kidney function and incident hypoglycaemia in people with diabetes: The Stockholm...

To evaluate possible associations between estimated glomerular filtration rate (eGFR) and hypoglycaemia in adults with diabetes.

We conducted an observational study in adults with diabetes from the Stockholm Creatinine Measurement (SCREAM) project, a Swedish healthcare utilization cohort during 2007 to 2011. We evaluated diagnoses and outpatient glucose tests for incidence rate ratios (IRRs) of hypoglycaemia (overall and by severity) in outpatient care by eGFR strata using zero-inflated negative binomial regression. We identified clinical predictors through ordinal logistic regression and assessed 7-day and 30-day mortality from hypoglycaemia in relation to eGFR with Cox proportional hazard models.

We identified 29 434 people with diabetes (13% with type 1 diabetes). Their mean age was 66 years, 43% were women and the median eGFR was 80 mL/min/1.73 m² . During 2 years of follow-up, 1812 patients (6.2%) had hypoglycaemia registered at least once. The risk of hypoglycaemia increased linearly with lower eGFR, with an IRR of 1.2 (95% confidence interval [CI] 1.0-1.4) for eGFR 60-89 mL/min/1.73 m² and 5.8 (95% CI 3.8-9.0) for eGFR <15 mL/min/1.73 m² compared to eGFR 90 to 104 mL/min/1.73 m² . This trend was observed for both mild and severe hypoglycaemia. Both 7-day and 30-day post-hypoglycaemia mortality increased with lower eGFR, peaking in those with eGFR <15 mL/min/1.73 m² (hazard ratio 21.2, 95% CI 5.1-87.9) as compared to those with eGFR 90 to 104 mL/min/1.73 m² . Lower eGFR categories, type 1 diabetes, previous hypoglycaemia, liver disease, presence of diabetic complications and use of insulin and sulphonylureas increased the odds of hypoglycaemia.

In this large, observational study, low eGFR was strongly associated with the occurrence, severity and fatality of hypoglycaemia in people with diabetes.