Impella Use in Cardiogenic Shock: National Trends Using NIS 2016

Background Cardiogenic shock (CS) remains a significant burden amongst the critically ill. The percutaneous micro-axial flow mechanical circulatory support (pMCS) device Impella provides significantly better hemodynamic support in such patients. The aim of our study was to describe national trends among patients admitted with CS who had Impella placement. Methods We conducted a retrospective analysis using the 2016 Nationwide Inpatient Sample (HCUP-NIS) database. Patients with an admitting diagnosis of CS as identified by the corresponding International Classification of Disease (ICD-10 CM) were analyzed. The subgroup of patients who had an IMPELLA placed in the index admission was identified by the ICD-10 procedural codes. A weighted descriptive analysis was performed to obtain In-hospital mortality, length of stay (LOS) and total charges. Statistical analyses was performed using STATA statistical software. The following confounders were adjusted for using multivariate regression analysis: age, sex, income, Charlson score, Hospital bed size, teaching status and urban location. Results During 2016, there were a total of 139205 CS admissions in the US of which 5805 underwent Impella placement. 28.3% were women. Majority of the patients were Caucasian (72.8%) followed by African Americans (10.46%). Majority of the patients had Medicare (54.1%) while only 5.2% had private insurance. Mean age was 63.8 years (95% CI 63.03-64.59). Notably, 28.3% of admissions came in as transfers from other acute care hospitals and 80.8% of admissions were at teaching hospitals. Among the 47.1% of patients who died, uninsured patients had a higher mortality when compared to medicare patients (adjusted odds ratio (aOR)1.89; p=0.048). Mean total charge per admission was $444,620. Transfers from other acute care centers resulted in an increase in hospital charges (adjusted mean difference(aMD) $166726, p=0.08). Large hospitals charged significantly more compared to smaller ones (aMD $94210, p=0.006) as did teaching hospitals (aMD $120196, p=0.00). Large hospitals had a significantly longer LOS compared to small hospitals (aMD 3.84 days, p=0.004) as did teaching hospitals (aMD 3.93 days, p=0.00). Conclusions 47.1% of the patients who got Impella placement for CS died during the hospitalization. There are significantly higher mortality rates for uninsured patients. Increasing age also increases mortality among patients who get an Impella for CS. The financial burden is compounded in larger hospitals and teaching hospitals (who receive the vast majority of admissions) as reflected in the increased LOS and hospitalisation charges.