Hypoglycaemia with Oral Antidiabetic Drugs Results from Prescription-Event Monitoring Cohorts of Rosiglitazone, Pioglitazone, Nateglinide and Repaglinide

Background: Hypoglycaemia is an acute complication associated with intensive treatment of patients with diabetes mellitus. This complication poses a major challenge in diabetes management. Furthermore, severe hypoglycaemia may be life threatening. Although hypoglycaemia is more often associated with insulin treatment, oral hypoglycaemic agents have the potential to trigger hypoglycaemia. Aim: The aim of this study was to quantify the incidence of hypoglycaemic events and to describe the pattern of these incident events during the first 9 months of treatment with four oral antidiabetic drugs, rosiglitazone, pioglitazone, nateglinide and repaglinide, prescribed in general practice in England. Methods: We used data collected for prescription-event monitoring (PEM) studies of rosiglitazone, pioglitazone, nateglinide and repaglinide. PEM is an observational, non-interventional, incept cohort study. Observation time for each patient and incidence rate (IR) per 1000 patient-years of treatment for hypoglycaemia was calculated for each drug cohort. Smoothed hazard estimates were plotted over time. Case/non-case analysis was performed to describe and compare patients who had at least one hypo,glycaemic event in the first 9 months of treatment with those who did not. Results: The total number of patients included in the analysis was 14373, 12768, 4549 and 5727 in rosiglitazone, pioglitazone, nateglinide and repaglinide cohorts, respectively. From these, 276 patients experienced at least one episode of hypoglycaemia. The IR was between 50% and 100% higher in patients receiving treatment with meglitinides compared with those treated with the thiazolidinediones(TZDs) [IR=9.94, 9.64, 15.71 and 20.32 per 1000 patient-years for rosiglitazone, pioglitazone, nateglinide and repaglinide, respectively]. The plot of the hazard function and the estimated shape parameter from the Weibull regression model showed that pioglitazone, nateglinide and repaglinide had non-constant (decreasing) hazards over time, whereas the hazard for rosiglitazone-treated patients was approximately constant over time. Nateglinide and repaglinide had similar shape hazard function, indicating a significantly higher number of hypoglycaemic episodes shortly after starting treatment. For women treated with TZDs, hypoglycaemia was reported more frequently than for men. Conclusion: This analysis shows that the frequency of reported hypoglycaemia within the study cohorts was relatively low. The rates of hypoglycaemia were not equal between drug classes. Treatment with nateglinide or repaglinide was characterized by a higher incidence of hypoglycaemia at the beginning of treatment. Further investigation is necessary to assess whether women treated with TZDs are more prone to hypoglycaemia than men. Findings from this study should be taken into account with other clinical and pharmacoepidemiological studies.