Left Ventricular Assist Device Implantation Can Be Performed in Patients With Severe Renal Insufficiency Without Increasing Morbidity or Mortality
Chetan Pasrija1, Hannah Voorhees1, Praveen George1, Erik Sorensen1, Mariem Sawan1, Mehrdad Ghoreishi1, Erica Feller1, Van-Khue Ton1, David J Kaczorowski1, Bartley P Griffith1, Si M Pham2, Zachary N Kon3
1University of Maryland, Baltimore, MD;2Mayo Clinic, Jacksonville, FL;3New York University Langone Health, New York, NY
OBJECTIVES: Renal insufficiency has commonly been considered a risk factor for mortality after left ventricular assist device (LVAD) implantation. Severe renal insufficiency has, therefore, been considered a relative contraindication to LVAD placement. After establishing an algorithmic approach to preoperative and postoperative management of these patients, we hypothesized that LVAD implantation in the setting of severe renal insufficiency could be performed without increasing morbidity or mortality.
METHODS: All consecutive patients (7/2015-12/2017) who underwent centrifugal continuous-flow LVAD implantation (HeartWare, Framingham, MA) at a single center were retrospectively reviewed. Patients were stratified by preoperative glomerular filtration rate (GFR): ≤30 (low-GFR) and >30 (high-GFR) mL/min/1.73m2. Subset analysis was performed comparing patients with a GFR ≤15 and ≥60 mL/min/1.73m2. Preoperative optimization included aggressive inotropic support and temporary mechanical circulatory support prior to LVAD implantation in any patient with an acute component of renal insufficiency. However, the presence of persistent renal insufficiency did not preclude implantation. All patients with a preoperative GFR ≤15 were prophylactically initiated on temporary continuous renal replacement therapy postoperatively. The primary outcome was Kaplan-Meier one-year survival.
RESULTS: 58 patients (low-GFR: 21, high-GFR: 37) were identified, with a median age of 56 years. Patients in the low-GFR cohort had a higher median INTERMACS score, Kormos score, and HeartMate II mortality score. Furthermore, preoperative end-organ function was significantly worse in the low-GFR group (Table 1). Preoperative optimization resulted an increase in GFR to >30 mL/min/1.73m2 in 52% (11/21) patients from the low-GFR cohort. Both cardiopulmonary bypass and operative time were similar between the 2 groups (Table 2). The incidence of postoperative moderate or severe RV failure was also similar between the 2 groups. Moreover, inotrope days, ventilator time, intensive care unit (ICU) LOS, and hospital LOS were not significantly different between the low-GFR and high-GFR group. Kaplan-Meier one-year survival was 81% in the low-GFR cohort and 86% in the high-GFR cohort (p=0.98). Subset analysis of GFR≤15 (n=11) and ≥60 (n=19) mL/min/1.73m2 demonstrated an increase in inotrope days (GFR≤15: 10 vs. GFR≥60: 6 days, p=0.002) and hospital LOS (GFR≤15: 21 vs. GFR≥60: 13 days, p=0.04) in the GFR≤15 subset, but no statistically significant difference in ventilator time, ICU LOS, or RV failure. One-year survival was identical between the two subsets (89% vs. 89%, p=0.91).
CONCLUSIONS: Despite an overall sicker cohort, patients undergoing LVAD implantation with severe renal insufficiency did not exhibit an increased morbidity or mid-term mortality. Based on these data, using a comprehensive management strategy, including pre-, intra-, and post-operative considerations, durable LVAD implantation in patients with severe renal insufficiency can be safely performed. This strategy may serve as a successful option to bridge patients to heart-kidney transplantation, or destination therapy with renal recovery or outpatient dialysis.
Back to 2018 Program