Not the hardest data, but raises some interesting questions.

Alright, let’s get something out of the way. This week, we’re talking about medications for erectile dysfunction and long-term mortality, thanks to this study, appearing in the Journal of the American College of Cardiology.

And, I know, you think I’m going to be immature about this. Crack jokes. Bad puns, that sort of thing. Well, I’m sorry to disappoint you, but this is a serious commentary about a serious study.

Briefly, the researchers found a vas deferens between men treated with phosphodiesterase five inhibitors, like Viagra, and those treated with alprostadil.

OK sorry. I’m done now.

The background here is that it is long-established that ED is associated with bad cardiovascular outcomes – likely because it is a proxy for vascular disease. Prior studies of men with ED found that those receiving PDE5-inhibitors had lower rates of those bad outcomes than men who didn’t get those medications – but comparing men who get ED prescriptions to those who don’t might introduce bias – men getting those prescriptions are healthy enough for sexual activity, for example.

A better control group might be men with ED receiving another type of treatment – like alprostadil, which is applied locally (either via injection, a cream, or a urethral suppository).

The study comes out of Sweden, home of great pharmacoepidemiology thanks to their nationwide healthcare database which includes a huge amount of data on healthcare status, medications, and outcomes on everyone in the country.

The researchers identified around 240,000 Swedish men who had a prior MI or revascularization. Of that group, around 20,000 were receiving medication for ED, mostly PDE5-inhibitors but enough on alprostadil to do the analysis.

Top-line results, men taking PDE5 inhibitors for ED were way less likely to have MIs, coronary revascularization, or heart failure than those taking alprostadil. In fact, over the course of up to 15 years of follow-up, 14% of men died of any cause in the PDE5 group, compared to 26% in the alprostadil group.

Of course, when you see results like that, you immediately think about confounding. Who are these men using injections when there is a pill on the market that achieves the same effect? You can see here that the populations were dramatically different – men on alprostadil were more likely to have diabetes, COPD, stroke, and active cancer and were basically sicker in any way the researchers could measure.

Adjustment for all of these factors dramatically attenuated the observed benefit of PDE5 inhibitors, but didn’t eliminate it. The class was still associated with lower rates of cardiovascular and all-cause mortality.

My gut is to interpret studies like this conservatively. The dramatic observed differences in baseline characteristics in the two study groups suggests there are pretty dramatic unobserved differences that couldn’t be accounted for by statistics. The authors didn’t have data on smoking status or body mass index for example.
They did their best with what they had, showing that there was something of a dose-response effect here, with men who filled MORE PDE5-inhibitor prescriptions having lower risk than those who filled less.  And of course, there really is biological plausibility for this effect – remember PDE5s were initially developed to be anti-hypertensives and anti-anginal medications. The effect on ED was either a happy accident or perhaps the result of a pharmaceutical executive finding a magic lamp.

And there are other, ahem, mechanisms by which these drugs might improve long-term outcomes. Without data on sexual activity, we can’t untie the pharmacologic effects of these drugs on blood vessels from their effect on… well… lifestyle. Maybe these men had more to live for. Several studies – yes, observational – have suggested that more sexual activity is associated with longer life.

You’d need a randomized trial to tease all this apart – one that I’m sure will have no trouble with recruiting. In the meantime, the data we have seem to suggest that, with PDE5-inhibitors, you might be getting more than you bargained for. Just remember – if your life is extended by more than 4 hours – phone your doctor.

A version of this commentary first appeared on medscape.com.