A new study of outcomes for elderly patients with heart failure who received implanted mechanical heart pumps (ventricular assist devices, or VADs) to support their circulation finds that many patients still die during or shortly after surgery. Many of the survivors are readmitted to the hospital within 6 months, and treatment costs remain high. The researchers conclude that improving patient selection, to reduce deaths during and shortly after surgery, is important to improve overall outcomes.
Researchers examined claims from Medicare patients in fee-for-service plans over a 6-year period. Overall, 2,943 patients at 570 hospitals received VADs. Half of the patients had not had prior open-heart surgery (the primary device group), while half had undergone open-heart surgery within 30 days before VAD implantation (the postcardiotomy group). Few of the hospitals performed more than 5 device implants annually (ranging from 19 hospitals in the year 2000 to 28 in the year 2005).
For all years, 815 patients (55.2 percent) in the primary device group were discharged alive from the hospital with a VAD, while 140 (9.5 percent) received heart transplants and 455 (30.8 percent) died during their hospital stay. At 1 year after implantation of the VAD, 417 primary device patients (32.2 percent) were still alive with the device, 299 (20.7 percent) had received a heart transplant, and 617 (42.2 percent) had died. Among the patients with prior open-heart surgery, 493 patients (33.6 percent) were discharged live with a VAD, 21 (1.4 percent) underwent a heart transplant, and 824 patients (56.2 percent) died during their hospital stay. At 1 year of followup, 333 postcardiotomy patients (24.1 percent) were alive with a VAD implanted, 51 patients (3.5 percent) had received heart transplants, and 929 (63.4 percent) had died.
The researchers used retrospective data on all Medicare fee-for-service patients receiving VADs, so the study provided a complete picture of the use of mechanical heart assist devices outside of clinical trials. They did not observe improved outcomes for prior cardiotomy patients from the beginning to the end of the study period, but saw a decline in 1-year survival from 41 percent in the year 2000 to 22 percent by 2006. The researchers suggest that this decline in survival may represent an increase over time in patients who would not have previously been considered for the procedure receiving VAD implants. The study was funded in part by the Agency for Healthcare Research and Quality (HS16964).
More details are in "Long-term outcomes and costs of ventricular assist devices among Medicare beneficiaries," by Adrian F. Hernandez, M.D., M.H.S., Alisa M. Shea, M.P.H., Carmelo A. Milano, M.D., and others, in the November 26, 2008, Journal of the American Medical Association 300(20), pp. 2398-2406.