Growing Doubt, Uncertainty Over PDE5i Use in Left Heart Disease

Patrice Wendling

June 12, 2019

A new INTERMACS study calls into question the off-label use of preoperative phosphodiesterase-5 inhibitors (PDE5i) to reduce the risk for right heart failure (RHF) in patients receiving a left ventricular assist device (LVAD).

LVAD recipients treated with preoperative PDE5i therapy were more likely than control subjects to experience severe early RHF in unadjusted (29.4% vs 23.1%) and propensity-matched (odds ratio, 1.31; 95% CI, 1.09 - 1.57) analyses.

Major bleeding events were also significantly higher in those on a PDEF5i, according to the study, published June 11 in Circulation: Heart Failure.

"Our research didn't demonstrate a signal of benefit and, in fact, suggested even a signal of potential harm, and thus really indicates the need for well-done prospective research looking at the role of the medication in this population," senior author Michael Kiernan, MD, MS, Tufts Medical Center, Boston, told | Medscape Cardiology.

Although PDE5 inhibitors are known to be an effective therapy for primary pulmonary arterial hypertension, evidence is thin to support off-label use to reduce right ventricular (RV) afterload before LVAD implantation.

"It's something that I am increasingly looking at askance," Brian A. Houston, MD, director of mechanical circulatory support, Medical University of South Carolina, Charleston, told | Medscape Cardiology. "Previously, we've leaned on small trials with surrogate outcomes or, even more vexingly, on biological plausibility to say it makes sense that if you give someone a pulmonary vasodilator, the RV afterload is reduced and they'll do better."

"But this study, even though it's retrospective and all the caveats of that, certainly calls into doubt the strength of that argument," he said. "And then, when you couple that with the SIOVAC trial," in which patients had worse outcomes when treated with sildenafil after valve repair or replacement, "it certainly calls into question whether we should be using these therapies in patients with left heart disease, which includes our LVAD patients."

In the absence of good alternative therapies, however, the temptation is strong. One in 10 patients (1199; 10.3%) were on preoperative PDE5i therapy in the study involving 11,544 participants with a first-time LVAD surgery in the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS), a representative cross section of LVADs implanted in North America.

Patients on PDE5i therapy had more severe INTERMACS profiles, higher use of renal replacement therapy, and higher mean pulmonary artery pressures (35.2 vs 33.0 mm Hg) and pulmonary vascular resistance values (2.6 vs 2.3 WU).

The primary end point of severe early RHF — defined as death from RHF within 30 days, need for right ventricular assist device (RVAD) within 30 days, or use of inotropes beyond 14 days — occurred in 23.7% of patients overall.

The association between severe early RHF and PDE5i use after propensity matching for 22 variables was driven by a higher incidence of prolonged inotrope support (24.0% vs 19.9%; P = .009). There were no significant differences between groups in RVAD use, death from RHF, or in 3- and 6-month quality of life and 6-minute walk distances, the authors report.

The PDE5i group had more major bleeding events than control subjects during the first month after LVAD implantation (24.5% vs 17.9%; hazard ratio [HR], 1.20; 95% CI, 1.06 - 1.36) that concentrated in the immediate postoperative period. A landmark analysis confirmed a significantly higher major bleeding event rate at 7 days with preoperative PDE5i use (HR, 1.52; 95% CI, 1.15 - 2.00), but similar rates thereafter.

Houston noted that PDE5 inhibitors can have a strong effect on platelet function that is often underappreciated going into surgery. He suggested that the 52% increased risk of bleeding may explain the signal for longer inotrope support because patients who bleed more get more volume resuscitation, the RV's going to be volume loaded, and inotropes won't be pulled off as quickly. After all that settles down, however, the patient will probably do okay.

"So it may be that explains the signal for longer inotrope use but not more mortality or morbidity, and that's really interesting because that would change practice for sure," he said. "You'd make sure to get patients off their PDE5 inhibitors before their operation. And I think that's an interesting hypothesis to draw from this study."

"It's certainly a reasonable explanation for what may have occurred in these patients," observed Kiernan. "But the converse is that it could be purely a reflection of the patients underlying substrate."

Subgroup analyses failed to identify a signal of benefit for preoperative PDE5i among patients stratified by markers of pulmonary hypertension or right heart dysfunction. The sole exception was among patients in the middle tertile of pulmonary artery systolic pressure, who had a higher frequency of early severe RHF.

"The call to action as presented by this study is increasingly pressing — if pulmonary arterial hypertension-specific therapy is indeed shown to increase the risk of harm to our patients with LVAD or LHD, guidelines and practice patterns should expeditiously abandon the arguments of biological plausibility and surrogate end points just as they did with antiarrhythmic agents for premature ventricular contractions and inotropic medications for LV failure," Houston writes in an accompanying editorial.

Limitations of the retrospective analysis are the potential for residual unmeasured confounders, missing pulmonary arterial wedge pressure values in 30% of cases, and a lack of information on the indication for preoperative PDE5i therapy, timing of its initiation in relation to LVAD implant, and whether it was continued in the 30 days after surgery.

Both Kiernan and Houston noted that the ongoing SOPRANO trial of macitentan may shed light on the utility of endothelin receptor antagonist therapy after LVAD. In the recent MELODY-1 study, however, macitentan failed to improve pulmonary vascular resistance and was associated with numerically more adverse events in patients with pulmonary hypertension due to left heart disease.

Kiernan reports consulting fees from Medtronic and travel support from Abbott. Coauthor disclosures are listed in the paper. Houston reports no relevant disclosures.

Circ Heart Fail. Published online June 11, 2019. Full text, Editorial

Follow Patrice Wendling on Twitter: @pwendl. For more from | Medscape Cardiology, join us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: