Introduction: In patients with cardiogenic shock due to acute myocardial infarction (AMI), percutaneous ventricular assist devices (PVADs) improve hemodynamic parameters, but do not offer mortality benefit when compared to intra-aortic balloon pumps (IABP). There is limited data on the outcomes of the use of these devices in patients with cardiogenic shock due to non-ischemic etiology. This study compared outcomes of use of these two devices among patients who developed cardiogenic shock not due to acute myocardial infarction or coronary revascularization.
Methods: We extracted patients with a diagnosis of cardiogenic shock who underwent either PVAD or IABP implantation from the national inpatient sample database from 2010 to 2014. We excluded patients with AMI or any revascularization procedure during their hospital. We compared clinical characteristics and inpatient outcomes in both groups. Our outcomes of interest were mortality and length of stay.
Results: Of the 238,091 patients with cardiogenic shock of non-ischemic origin, 7.6% (n=18,046) underwent implantation of either device (IABP 92.1% and PAD 7.9%). After exclusions, 18,032 patients were included in our analysis. The mean age of all patients was 59.2±15.3 and 33.2% were female. The PVAD group was younger (55.8±17.2 vs 19.5±15.1, p<.01) and less likely to be female (27.7% vs 33.6%, p=.01). Other than a higher proportion of patients with LVAD in the PVAD arm (1.8% vs 0.4%, p=.<01), comorbidities were similar in each arm. There was a higher proportion of patients with acute renal failure (72.4% vs 61.4%, p<.001), dialysis for acute renal failure (15.4% vs 10.4%, p=.01), ECMO use (11.4% vs 5.5%, p<.01), and mechanical ventilation (68.2% vs 59.7%, p=.002). Mortality was significantly higher in the PVAD arm (51.9% vs 30.2%, p<.01). In multivariate analysis, PVAD use in these patients was associated with higher mortality (OR 2.12 95% CI 1.54-2.90; p<.01). Length of hospital stay in patients that survived to discharge was similar in both groups (15, IQR 25 vs 16, IQR 21, p= .68).
Conclusions: In a retrospective analysis of a large national database, the use of PVAD was associated with higher mortality compared to IABP use in patients with cardiogenic shock of non-ischemic etiology.