Association Between Percutaneous Ventricular Assist Device Hospital Volumes and In-hospital Outcomes.

Background:
The use of percutaneous ventricular assist devices (pVAD) has increased considerably in the past 10 years for the treatment of cardiogenic shock. We aimed to identify if hospitals with higher volume of pVAD placement have better outcomes than low-volume hospitals.

Methods:
Using the National Readmission Database from 2010 to 2015, we identify adult patients that underwent pVAD placement using ICD-9 DM codes and divided them into low-moderate volume centers (≤ 12 pVAD/year) and high-volume centers (>12 pVAD/year). We used multivariate logistic regression to compare mortality, readmission rates and in-hospital outcomes between both groups.

Results:
We included 1,671 hospitals (121 categorized as high-volume and 1,450 as low-moderate volume) which comprised 17,218 patients (5,073 admitted to high-volume and 12,145 admitted to low-moderate volume hospitals). In-hospital mortality was significantly lower in high-volume centers when compared to low-moderate volume centers (32.1 % vs 35.9%, OR 0.77, 95%CI 0.65-0.92, p=0.003), as was vascular complications (25.5% vs 28.7%, OR 0.80, 95%CI 0.67-0.94, p=0.008) and cardiac complications (7.5% vs 9.1%, OR 0.73, 95%CI 0.56-0.96, p=0.025). High volume centers also had lower hospital costs ($63,205 vs $67,329, p=0.016). There was no significant difference in 30-day readmissions, 30-day mortality, stroke rates, major bleeding, acute kidney injury and length of stay.

Conclusions:
Hospitals with higher pVAD volume had lower in-hospital mortality, vascular complications, cardiac complications and lower hospital costs when compared with low-moderate volume hospitals.