Warfarin with sulfonylureas and hypoglycemic events: consider dose, length of treatment, and all sulfonylureasBMJ 2016; 354 doi: https://doi.org/10.1136/bmj.i4328 (Published 05 August 2016) Cite this as: BMJ 2016;354:i4328
- 1Université de Bordeaux, UMR1219, CHU de Bordeaux, 146 Rue Léo Saignat, 33076 Bordeaux Cedex, France
Romley and colleagues studied the association between the use of warfarin with common sulfonylureas and the development of serious hypoglycemic events.1 Data on potential interactions are essential because both drugs have a narrow therapeutic range and can also interact with many other drugs.2 3
The pharmacological effect of these drugs is strongly related to the dose, and a higher dose inevitably increases the risk of hypoglycemic events. However, Romley and colleagues did not adjust for the dose of glipizide/glimepiride and the length of exposure, which may result in residual confounding of the risk estimate. Moreover, a hypoglycemic event is more likely to occur and be detected early during treatment because of non-optimal control of diabetes or more frequent monitoring. Indeed, recent users of sulfonylureas may be more at risk of hypoglycemic events than those who have taken the drugs for some time, but the analyses were not stratified by recent and current users.
More importantly, glyburide (glibenclamide) was not considered to be a commonly prescribed sulfonylurea, but, even in the US data, it is not prescribed much less than glipizide or glimepiride.4 5 Glyburide is a second generation sulfonylurea also known to cause hypoglycemia: Romley and colleagues do not give enough information to justify its exclusion from the principal analysis. However, their separate analysis in fig 4 shows a non-significant association between severe hypoglycemic events and the concomitant use of warfarin and glyburide.
The authors should provide the global results for all commonly used sulfonylureas and the risk of severe hypoglycemia when prescribed with warfarin.
Competing interests: None declared.