Use of percutaneous mechanical circulatory support for right ventricular failure.

Abstract

Utilization of right ventricular mechanical circulatory support (RV-MCS) devices has been limited by a lack of recognition of RV failure as well as a lack of availability and experience with RV-MCS. We report a single-center experience with the use of percutaneous RV-MCS and report predictors of adverse outcomes. This was a single-center retrospective cohort study. Data from consecutive patients who received RV-MCS for any indication between June 2015 and January 2022 were included. Data on baseline comorbidities, hemodynamics, and laboratory values were collected. The primary outcome was in-hospital mortality analyzed as a logistic outcome in a multivariable model. These variables were further ranked by their predictive value. Among 58 consecutive patients enrolled, the median age was 66 years, 31% were female and 53% were white. The majority of the patients (48%) were hospitalized for acute on chronic heart failure. The majority of the patients were SCAI SHOCK Stage D (67%) and 34 (64%) patients had MCS placed within 24 h of the onset of shock. Before placement of RV-MCS, median central venous pressure (CVP) and RV stroke work index were 20 mmHg and 8.9 g m/m2, respectively. Median serum lactate was 3.5 (1.6, 6.2) mmol/L. Impella RP was implanted in 50% and ProtekDuo in the remaining 50%. Left ventricular MCS was concomitantly used in 66% of patients. Twenty-eight patients (48.3%) died. In these patients, median serum lactate was significantly higher (4.1 [2.3, 13.0] vs. 2.2 [1.4, 4.0] mmol/L, p = 0.007) and a trend toward higher median CVP (24 [18, 31] vs. 19 [14, 24] mmHg, p = 0.052). In the multivariable logistic model, both serum lactate and CVP before RV-MCS placement were independent predictors of in-hospital mortality. Serum lactate had the highest predictive value. In our real-world cohort, 52% of patients treated with RV-MCS survived their index hospitalization. Serum lactate at presentation and CVP were the strongest predictors of in-hospital mortality.