Dialysis After Left Ventricular Assist Device Implantation

Abstract

Purpose Renal replacement therapy (RRT) post-LVAD implantation is often needed due to post-operative complications from AKI and pre-implant cardio-renal syndrome. RRT can assist with post-operative renal function recovery, but long-term use is associated with complications and poor outcomes. Identifying risk factors that increase the likelihood of RRT initiation and withdrawal post-LVAD can help multi-disciplinary teams risk-stratify patients to improve outcomes. Methods A retrospective chart review from 2018 to 2019 was conducted on 26 LVAD recipients (53.9% INTERMACS Class I) in our program. Patients on RRT prior to LVAD implantation were excluded. Clinical, laboratory, and outcomes data were compared between RRT (n=11) and no RRT (n=15) patients and between RRT recipients who continued (n=4) and withdrew (n=7). RRT withdrawal was defined as stopping RRT before hospital discharge. Bivariate analyses performed with T-test, Wilcoxon Rank-Sum Test, Chi-square, and Fisher's Exact test. Logistic regression with backward elimination used to predict RRT initiation and withdrawal. Results Patients on RRT were less healthy compared to patients not on RRT. More patients on RRT had hypertension (p=0.01) and history of myocardial infarction (p=0.05). Patients requiring RRT had longer mean hospitalizations (50.5 vs 30.3 days, p =0.01). Pre-operative renal function characterized by BUN, Cr, and eGFR were worse among RRT recipients (p=0.02, 0.04, and 0.04), respectively. Adjusting for age and sex, our reduced model identified a 12% increase in RRT initiation for each 1mg/dL increase in BUN prior to LVAD implantation (p=0.03). Patients withdrawing or continuing RRT post-LVAD were similar in terms of co-morbidities and hemodynamics. INTERMACS Class I was not associated with RRT initiation or withdrawal. Patients withdrawing from RRT had better renal function than those who continued. A serum biomarker, LDH (U/L), was elevated after (p=0.01) LVAD implantation among patients who withdrew from RRT. Conclusion Pre-operative renal function was significantly worse among LVAD recipients requiring post-implantation RRT. LVAD recipients may benefit from optimization of renal function and identification of serum biomarkers prior and after surgery, which may reduce the number of patients requiring RRT after LVAD implantation.