Background: Phosphodiesterase type-5 inhibitors (PDE5is) exert cardioprotective effects in small mammal models of myocardial ischaemia. There is currently little data on whether a similar effect exists in humans. We determined whether PDE5i use in males with type 2 diabetes (T2DM) was associated with reduced mortality.

Methods: We retrospectively analysed the pseudoanonymised records of 48 GP practices in Cheshire, UK and identified all 7029 men (mean age 72.8 years) diagnosed with T2DM before 1 January 2007. Baseline clinical characteristics and PDE5i treatment data were obtained. Mean follow-up was 6.4 years (January 31, 2014) and all deaths were ascertained from GP records.

Findings: Of the 1,663 (23.7%) men prescribed a PDE5i, the proportion of deaths was significantly lower than those never prescribed (16.9% versus 29.4%). All-cause mortality rates (per 1000 person-years) were similarly lower (21.1 (19.1–24.5) versus 34.4 (32.5–36.5); P<0.0001). There was a 38% reduction in all-cause mortality (univariate Cox proportional hazards HR: 0.62 (0,54–0.71); P<0.0001) in men on a PDE5i over the period. This reduction remained but was attenuated (HR: 0.80 (0.65, 0.98); P<0.05) after multivariate regression adjusting for age (1.11 (1.09–1.12); P<0.0001 per year), smoking history (1.31 (1.16–1.47); P<0.0001), HbA1c, systolic BP, creatinine levels, prescribed statins, aspirin and beta-blocker use.

Interpretation: Around 70% of deaths in T2DM are attributable to cardiovascular disease. Our data demonstrates that PDE5i use is associated with significantly reduced mortality in men with T2DM at high risk of CVD. Further evidence is required to elucidate the role of PDE5is in cardioprotection.