Empagliflozin therapy for type 2 diabetes: viewing possible benefits in relative and absolute terms

Abstract

Background The impact of therapeutic interventions on outcomes needs assessing in both relative and absolute terms. We therefore undertook such assessments for empagliflozin treatment in type 2 diabetes, using data published in the EMPA-REG clinical trial,1 in which all patients also received standard treatment for cardiovascular risk factors including diabetes. Methods As described previously,2 for several critical outcomes we computed unadjusted RR and NNT)/year values and corresponding 95% CIs for empagliflozin treatment, using pooled data for 10 and 25 mg/day recipients (n=4687). Placebo recipients numbered 2333. The median patient observation time was 3.1 years. Results Our findings are summarized below. The only significant adverse effect was genital infection; respective RR and ‘number-needed-to-harm’/year values were 3.57 (2.57-4.96) and 67 (56-83). Conclusions Statistically significant benefits were mainly driven by mortality reductions, whereas there were no significant non-fatal event rate changes. This implies conversion of fatal to non-fatal outcomes coupled with prevention of some non-fatal events. The NNT/year for preventing any death was 120. This absolute benefit was about half that for preventing fatal and non-fatal coronary events enjoyed by high risk patents in the 4S trial (in whom the NNT/year was 63).2 Moreover, such benefit appears additional to that derived from standard anti-lipidaemic, anti-hypertensive, and other interventions offered to these patients. References 1.Zinman B et al., NEJM 2015; 373:2117-2128. http://dx.doi.org/10.1056/NEJMoa1504720.

2.Kumana CR et al., JAMA 1999; 282:1899-1907.