Implantable left ventricular assist device. Approaching an alternative for end-stage heart failure. Implantable LVAD Study Group

Abstract

Background: The implantable left ventricular assist device (LVAD) was designed to provide circulatory support as an alternative to heart transplantation or to continued medical therapy of end-stage heart failure. Initial experience with the implantable LVAD used as a bridge to heart transplantation provides a clinical opportunity to study the function of the device and adaptation by the patient.

Methods and results: Nineteen heart transplant candidates (mean age, 50 years; 17 males) underwent insertion of the HeartMate LVAD as a bridge to heart transplantation from December 1991 to November 1993. All patients were in cardiogenic shock on inotropes, and 16 (84%) were on an intra-aortic balloon pump. Three patients died because of multiple organ failure; all had right ventricular (RV) dysfunction (2 required RV assist devices). Sixteen patients (84%) improved markedly and were rehabilitated to New York Heart Association functional class I-II. Three patients are still on support. Significant improvements in hemodynamic function (based on analysis of the percent change from pre-LVAD condition to pretransplantation) were observed: cardiac index rose from 1.6 +/- 0.2 to 3.2 +/- 0.9 L/min per m2 (P = .0002), left atrial pressure fell from 22.9 +/- 9.5 to 8.0 +/- 5.5 mm Hg (P = .003), RV ejection fraction increased from 19.8 +/- 11.3% to 40.8 +/- 8.9% (P = .0004), pulmonary vascular resistance decreased from 5.2 +/- 2.6 to 2.0 +/- 0.8 Wood units (P = .004). Thirteen patients had successful transplants after a mean duration of 66 days on the LVAD (range, 22 to 101 days). There were no thromboembolic events while the patients were on the LVAD. Only aspirin with dipyridamole was given for anticoagulation during a total of > 1100 patient days of support.

Conclusions: Bridge to transplant implantable LVAD experience indicates that hemodynamic improvement should be significant after insertion of the devices and that the risk of thromboembolic events with the HeartMate LVAD should be extremely low. Rehabilitation and quality of life should be markedly improved. Limitations of extrapolating this clinical experience to the permanent implantable LVAD include that these patients were hospitalized (permanent implants will be outpatients); the "vented-electric" HeartMate LVAD was not tested (it is a portable, battery-powered device), and true "chronic" LVAD support (> 1 year) was not tested, so questions regarding long-term device reliability and the chronic risk of infection are unknown.