Abstract WP398: In-Hospital Mortality: Is the Rate Associated with Hemicraniectomy Changing Over Time?

Abstract

Introduction: Although there are recent data which suggest that surgical decompression can reduce mortality in patients with acute ischemic stroke (AIS), hemicraniectomy has remained a hot topic in stroke care, with candidates for this procedure at high risk for excessive morbidity and mortality. We determined in-hospital mortality rates for patients with AIS undergoing hemicraniectomy, and compared these rates to those reported previously. Methods: Data from the National Inpatient Sample (NIS), a nationwide administrative database for the year 2009 (>7,800,000 US hospital admissions), and the associated ICD-9-CM diagnosis and procedure codes were used to identify patients. We examined in-hospital mortality associated with hemicraniectomy, with and without thrombolysis in patients admitted emergently, having a discharge diagnosis of acute ischemic stroke and/or precerebral arterial occlusion (multi-level clinical classification software codes 109 and 110). Results: From 114982 adult patients (53% Female) with AIS, 118 (0.1%) underwent hemicraniectomy, and 3115 (2.7%) were treated with thrombolysis. Compared with the nonsurgical group, patients treated with hemicraniectomy were younger (mean age 55 vs. 72 years), but had a higher inpatient mortality rate (25% vs. 7%). In patients treated with thrombolysis, mortality was also higher in the hemicraniectomy group compared to the nonsurgical group (27% vs. 10%). Stroke associated complication rates for systemic sepsis (9% vs. 2%) and aspiration pneumonia (9% vs. 4%) were higher in the hemicraniectomy-treated patients. Median cost per stay was $170,600 in the hemicraniectomy treated patients vs. $25,100 nonsurgical. The 2009 mortality rate in hemicraniectomy-treated patients was lower than previously reported in NIS cohorts from 2000-2006 (25% vs. 32%). Conclusion: In-hospital mortality remains high for patients treated with hemicraniectomy, but appears to have declined from previous years. Despite the fact that candidates for hemicraniectomy are by definition at very high risk for excessive morbidity and mortality, use of databases such as the NIS can supply valuable information on demographics and trends for inpatient outcomes, and the associated costs.