Association between pulmonary artery pulsatility and mortality after implantation of left ventricular assist device

Abstract

Aims: Right ventricular failure after left ventricular assist device (LVAD) implantation is a major concern that remains challenging to predict. We sought to investigate the relationship between preoperative pulmonary artery pulsatility index (PAPi) and mortality after LVAD implantation.

Methods and results: A retrospective analysis of the ASSIST-ICD multicentre registry allowed the assessment of PAPi before LVAD according to the formula [(systolic pulmonary artery pressure - diastolic pulmonary artery pressure)/central venous pressure]. The primary endpoint was survival at 3 months, according to the threshold value of PAPi determined by the receiver operating characteristic (ROC) curve. A multivariate analysis including demographic, echographic, haemodynamic, and biological variables was performed to identify predictive factors for 2 year mortality. One hundred seventeen patients were included from 2007 to 2021. The mean age was 58.45 years (±13.16), with 15.4% of women (sex ratio 5.5). A total of 53.4% were implanted as bridge to transplant and 43.1% as destination therapy. Post-operative right ventricular failure was observed in 57 patients (48.7%), with no significant difference between survivors and non-survivors at 1 month (odds ratio 1.59, P = 0.30). The median PAPi for the whole study population was 2.83 [interquartile range 1.63-4.69]. The threshold value of PAPi determined by the ROC curve was 2.84. Patients with PAPi ≥ 2.84 had a higher survival rate at 3 months [PAPi < 2.84: 58.1% [46.3-72.8%] vs. PAPi ≥ 2.84: 89.1% [81.1-97.7%], hazard ratio (HR) 0.08 [0.02-0.28], P < 0.01], with no significant difference after 3 months (HR 0.67 [0.17-2.67], P = 0.57). Other predictors of 2 year mortality were systemic hypertension (HR 4.22 [1.49-11.97], P < 0.01) and diabetes mellitus (HR 4.90 [1.83-13.14], P < 0.01). LVAD implantation as bridge to transplant (HR 0.18 [0.04-0.74], P = 0.02) and heart transplantation (HR 0.02 [0.00-0.18], P < 0.01) were associated with a higher survival rate at 2 years.

Conclusions: Preoperative PAPi < 2.84 was associated with a higher risk of early mortality after LVAD implantation without impacting 2 year outcomes among survivors.

Keywords: Circulatory assist devices; Database; Haemodynamics; Heart failure; Left ventricular assist device; Ventricle, right.

Figure 1

Flow chart of the study. ECMO, extracorporeal membrane oxygenation; HD, haemodynamic; LVAD, left ventricular assist device.

Figure 2

Definition and severity of post‐left ventricular assist device right ventricular failure (RVF) according to the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) registry. CVP, central venous pressure; IVC, inferior vena cava; RAP, right atrial pressure; RV‐MCS, right ventricular mechanical circulatory support.

Figure 3

Receiver operating characteristic curve of pulmonary artery pulsatility index as a predictor of 3 month mortality. AUC, area under the curve [95% confidence interval].

Figure 4

Comparative survival according to the pulmonary artery pulsatility index (PAPi) threshold value of 2.84.

Figure 5

Univariate analysis including preoperative risk factors of mortality (univariate analysis). ACE, angiotensin‐converting enzyme; ARBs, angiotensin receptor blockers; BMI, body mass index; CI, confidence interval; CRT, cardiac resynchronization therapy; ECMO, extracorporeal membrane oxygenation; IABP, intra‐aortic balloon pump; MRA, mineralocorticoid receptor antagonist; OR, odds ratio; VA < 30 days, early ventricular arrhythmia at 1 month. **Statistical test was considered statistically significant if P < 0.05.

Figure 6

Univariate analysis including post‐operative risk factors of mortality (univariate analysis). CI, confidence interval; LVAD, left ventricular assist device; NO, nitrous oxide; OR, odds ratio; RRT, renal replacement therapy; RV‐MCS, right ventricular mechanical circulatory support; VA‐ECMO, veno‐arterial extracorporeal membrane oxygenation; VV‐ECMO, veno‐venous extracorporeal membrane oxygenation. **Statistical test was considered statistically significant if P < 0.05.