The impact of peripheral arterial disease on left ventricular assist device implantation: A propensity-matched analysis of the nationwide inpatient sample database.

Abstract

Left ventricular assist device (LVAD) candidacy screening includes evaluation for peripheral arterial disease (PAD). However, given current evidence, the impact of PAD on post-LVAD complications remains unknown. The National Inpatient Sample (NIS) database (2002-2017) was utilized to identify all LVAD cases. The in-hospital safety endpoints included major cardiovascular adverse events and its components. A propensity-matched analysis was used to obtain adjusted odds ratios (aOR). A subgroup analysis of patients with diabetes mellitus (DM) with PAD was also performed. A total of 27 424 patients with LVAD implantation (PAD: 516 [1.8%] and no-PAD 26908 [98.2%]) were included. There were significant intergroup differences in the demographics and baseline comorbidities. A weighted sample of 1053 (no-PAD 537, PAD 516) propensity-matched population was selected. The adjusted odds for in-hospital mortality (aOR 1.7; 95% CI, 1.2-2.44, P=.004) were found to be significantly higher for LVAD-patients with PAD. There was no significant difference in the adjusted odds of MACE (aOR 1.16, 95% CI 0.87-1.5), postprocedure bleeding (aOR 0.88, 95% CI 0.62-1.26, P=.54) and risk of pneumonia (aOR 0.67, 95% CI 0.44-1.15, P=.63) between the two groups. A selected cohort of DM-only population (7339) consistently showed a higher adjusted mortality rate in PAD patients with LVAD implantation (aOR 2.3, 95% CI 1.2-4.47, P=.01). The rate of MACE (P=.17), myocardial infarction (P=.12), stroke (P=.60), postprocedural (0.10), and major bleeding (P=.51) remained identical between patients with PAD and those with no-PAD. PAD confers an increased risk of in-hospital all-cause mortality in patients undergoing LVAD implantation. This risk increases further in patients with a concomitant diagnosis of DM.