Left ventricular assist devices complicated by gastrointestinal bleeding and outcomes on transplant
Aurora L, Sadiq O, Nemeh H, and Williams C. Left ventricular assist devices complicated by gastrointestinal bleeding and outcomes on transplant. Am J Transplant 2018; 18:651.
Am J Transplant
Left Ventricular Assist Devices (LVAD) are becoming the standard of care in advanced heart failure patients. However, despite ongoing advances in this technology, there remains a signifi cant early postoperative morbidity and mortality. LVAD implantation is associated with signifi cant complications including infection, arrhythmias, post-operative, and non-surgical bleeding. Gastrointestinal bleeding (GIB) is one of the most common long-term complications with an incidence approaching 30%. There is literature pertaining to global outcomes after LVAD implantation, but few have studied outcomes after transplant in the presence of GIB. In this retrospective study, 234 patients who received FDA approved LVAD devices (HeartMate II or HeartWare) in a tertiary care center between 2005 and 2015 were studied. Demographics such as age, race, sex, and socioeconomic status were recorded along with comorbidities such as hypertension, diabetes, coronary artery disease, and etiology of heart failure. Incidence of GIB was investigated along with post-transplant outcomes. 68 patients (29.06%) had GIB whereas 166 (70.94%) did not have any occurence of GIB. Out of the GIB cohort, 28 (41.18%) underwent transplant whereas 40 (58.82%) did not undergo transplant or died. 63 (37.95%) of the non-GIB cohort underwent transplant. There was no signifi cant difference between the groups in the likelihood of receiving a transplant (p=0.755). 1 year post-transplant outcomes did not differ signifi cantly between the cohorts. Specifically, length of stay for the transplant hospitalization was 30.4 days in the GIB transplant group compared to 27.83 in non-GIB transplant patients (p=0.77). There were 2 incidences of rejection (defi ned as grade 2R/3A) in the GIB transplant group compared to 4 in the non-GIB transplant group (p=1.00). Patients with GIB had an average 1.33 readmissions post-transplant compared to 2.03 in the non-GIB group (p=0.148). Only 2 patients died (7.14%) in the GIB group compared to 12 deaths (19.05%) in the non-GIB group (p=0.255). Our results of a single center study reveal that GIB in patients who have LVAD is not predictive of transplant outcomes. Nonetheless, health care providers need to appropriately manage GIB as this will continue to occur when LVADs are implanted as destination therapy or bridge to transplant. Understanding prognostic factors in this high risk population can help appropriately and effectively target patient needs.