Mortality Higher in Real-World LVAD Recipients Ineligible for Clinical Trials

By Megan Brooks

January 13, 2021

NEW YORK (Reuters Health) - Roughly half of patients with advanced heart failure who receive a left ventricular assist device (LVAD) in clinical practice do not meet eligibility criteria for the clinical trials that led to expanded LVAD use, a new study suggests.

The study also found a higher death rate in trial-ineligible LVAD recipients and identified several clinical factors tied to increased risk.

"Clinical trialists, regulators and payers should consider the generalizability and prognostic value of individual eligibility criteria to further refine LVAD trial populations," Dr. Alexander Brescia of Michigan Medicine and the University of Michigan, Ann Arbor, told Reuters Health by email.

"Likewise, clinicians should use trial criteria to inform clinical decision-making when evaluating LVAD candidates and prognosticating in real-world practice," said Dr. Brescia, first author of the study, which appears in JAMA Network Open.

Wide-scale adoption of durable LVADs for end-stage heart failure followed favorable outcomes of randomized controlled trials. However, restrictive trial selection criteria may limit generalizability to real-world patients, the authors note in their article.

"As with other treatments and types of surgery, the clinical outcomes observed in a real-world setting are not necessarily the same as what is found in the setting of a clinical trial. This analysis was important to determine whether patients who would not have qualified to be in the clinical trial do as well as those who would qualify, within a group of real-world patients undergoing heart surgery to receive this device," Dr. Brescia explained.

By applying inclusion and exclusion criteria to a real-world population of LVAD recipients, "we were able not only to characterize how many of real-world recipients would have even been eligible to be enrolled in the trial, but whether 'trial-ineligible' patients do worse and furthermore describe what other factors are associated with that group doing worse," he said.

Among a total of 14,679 LVAD recipients, 6,429 (43.8%) were trial ineligible in the limited analysis and 7,888 (53.7%) were trial-ineligible in the comprehensive analysis, the authors report.

In the group of trial-ineligible LVAD recipients (primary analysis), 4,226 (65.7%) met one ineligibility criterion, 1,442 (22.4%) met two criteria, and 761 (11.8%) met three or more ineligibility criteria.

Mortality at one and three years was significantly higher for trial-ineligible LVAD recipients (25.3% and 42.8%, respectively) than their trial-eligible peers (16.2% and 36.4%, respectively).

Among patients with one ineligibility criterion, prior or ongoing mechanical circulatory support, elevated bilirubin or creatinine levels, and low albumin or prealbumin levels were independently associated with mortality.

Each additional criterion of ineligibility was associated with increasing risk of mortality. The hazard ratios for one, two and three or more ineligibility criteria were 1.16, 1.51 and 2.09, respectively.

"This analysis is important not only because it informs clinicians who are advising on the decision of whether or not to offer LVAD therapy and communicating individual prognosis to patients, but also trialists designing the next clinical trial and regulators who may determine strategies for approving the use of these devices and managing reimbursement," Dr. Brescia told Reuters Health.

He noted, however, that even trial-ineligible patients at a higher risk of mortality "should be considered by comparing prognosis with medical management alone, which was not investigated in this study."

"Importantly, this study period included LVADs implanted between 2012 and 2017. Future analyses will include assessing the generalizability of trial findings for the HeartMate 3 device, which is the most commonly used device in practice currently," Dr. Brescia said.

The study had no commercial funding and the authors have no relevant conflicts of interest.

SOURCE: http://bit.ly/3q9q4sV JAMA Network Open, online January 8, 2021.

processing....