The total artificial heart (TAH) is an effective treatment for biventricular heart failure and bridge to heart transplantation. The original hospital device driver weighed several hundred pounds and did not allow hospital discharge.1 The portable driver for the TAH was approved in 2014 by the Food and Drug Administration. Data regarding hospital admissions with left ventricular assist devices are increasing2,3 but there are limited data regarding admissions for patients using the Freedom Driver.4 We examined hospital admissions in patients that were implanted with the TAH and discharged with a portable driver at our institution to understand the frequency and reasons for admission.
This study was approved by the institutional review board and ethics committee at Virginia Commonwealth University Health System, Richmond, VA. We included patients implanted with the TAH from December 2010 to August 2018. The medical records were reviewed and data pertaining to patient demographics, hospitalizations, and clinical outcomes were extracted. Ambulatory patients were transitioned to the portable driver when they were hemodynamically stable, did not require intravenous medications, without surgical drains, and without dialysis dependence. We collected data for each patient until the time of heart transplant, death, or data censoring date of August 1, 2018. Data were analyzed using either a t test or χ2 analysis with SPSS (Armonk, NY).
A total of 62 patients were implanted with the TAH during the study period and most patients were male (82%) and INTERMACS profile 1 (69%). Twenty-six patients (42%) were transitioned to the portable driver and 20 patients (32%) were successfully discharged from the hospital. Six of the patients transitioned to the portable driver remained in the hospital because of acute renal failure requiring dialysis (2), nausea and vomiting (1), psychologic instability (1), hemodynamic instability (1), and patient choice (1). Clinical comparisons of the patients discharged from the hospital to those were not are presented in Table 1.
Patients were supported on the portable driver in hospital for an average of 16.9 ± 15.9 days, then discharged and supported for a median [25th,75th] of 257 [129, 742] days (mean 418, range: 49–1193 days). During follow-up, patients experienced 148 hospital admissions (Table 2) and each patient experienced a median [25th, 75th] of 6 [3, 10.3] admissions (mean 7.4, range: 2–20 admissions, average 8.2 ± 4.0 admissions per year). Eight patients (40%) were admitted within 30 days of hospital discharge, 16 patients (80%) were admitted within 90 days and 19 patients (95%) were admitted within 180 days. Average time to first admission was 70.0 ± 63.9 days (range: 6–302 days). Patients spent 17.0% of their days alive on the portable driver admitted to the hospital and each hospital stay averaged 9.3 ± 8.4 days in duration.
Thirty-two admissions (21.6%) were planned and included four unique admission diagnoses: scheduled portable driver device exchange (n = 13, 8.8%), planned heart transplant (n = 11, 7.4%), desensitization in preparation for transplant (n = 7, 4.7%), and admission to inpatient rehabilitation (n = 1, 0.7%). Portable driver exchanges are now typically done during a clinic visit, however early in our study patients were intermittently admitted for freedom driver exchange. They would participate in cardiopulmonary rehab and medications were optimized. Length of stay was typically one or two days. One hundred sixteen admissions (78.4%) were unplanned and were most commonly due to driveline crack/fracture (n = 21, 14.2%), portable driver alarm of unknown etiology (n = 19, 12.8%), and abnormal blood pressure (n = 9, 6.1%). Neurologic complications were limited to three patients (15%, two ischemic cerebrovascular accident (CVA) and one hemorrhagic CVA).
Patients reported driver alarms in 21.0% (31/148) of all admissions. A cause of alarm was not identified in the majority (n = 19, 61.3%) of these instances. When this occurred, the patient was typically observed for 24 hours and discharged after a portable driver exchange. In 12 cases (38.7%), the cause of the alarm was identified: hypertension >160 systolic (n = 5, 41.7%), nausea/vomiting (n = 2, 16.7%), driver overheating (n = 2, 16.7%), syncope (n = 2, 16.7%), and pulmonary edema (n = 1, 8.3%). The driver did not alarm every time one of these events occurred however.
Patients successfully discharged on the portable driver were likely to have a longer wait to transplantation (319 ± 232 vs. 139 ± 87 days; p = 0.013) or death (934 ± 521 vs. 28 ± 26 days; p = 0.04) and trended toward a higher proportion of blood type O patients (75% vs. 50%; p = 0.06). Other factors, including baseline clinical characteristics and survival to transplant, did not show a significant difference between the two groups (Table 1).
The TAH is an option as a bridge to transplant for patients with biventricular heart failure. The portable driver allows these patients to be discharged from the hospital, but data after their index hospitalization are sparse. We observed that patients can safely live at home on a portable driver; however, they have frequent hospital admissions and experienced longer wait times to transplant or death. We partly attribute the difference in wait time to the prior United Network for Organ Sharing (UNOS) allocation system, where hospitalized patients with a TAH were listed as 1A and patients with a TAH discharged from the hospital were listed as 1B. We expect this discrepancy to fade with the updated UNOS allocation system which went into effect October 18, 2018.
These data suggest that device improvements with driveline durability and implementation of clinical protocols to closely follow fluid volume status, blood pressure, and INR may offer opportunities to reduce hospital readmissions and time spent in the hospital for patient with the TAH.
1. El Banayosy A, Arusoglu L, Morshuis M, et al. Home discharge and out-of-hospital follow-up of total artificial heart patients supported by a portable driver. ASAIO J 2014.60: 148–153.
2. Forest SJ, Bello R, Friedmann P, et al. Readmissions after ventricular assist device: Etiologies, patterns, and days out of hospital. Ann Thorac Surg 2013.95: 1276–1281.
3. Tsiouris A, Paone G, Nemeh HW, et al. Factors determining post-operative readmissions after left ventricular assist device implantation. J Heart Lung Transplant 2014.33:1041–1047.
4. Quader MA, Green AJ, Shah KB, et al. Hospital readmissions after discharge to home with the total artificial heart freedom driver: Readmission reasons, clinical outcomes, and health care costs. J Heart Lung Transplant 2016.35:251–252.