Volume-Outcome Relationships for Endovascular Recanalization Therapy of Acute Ischemic Stroke in Korea and Identifying Volume Thresholds

Abstract

Introduction:<br/> Endovascular recanalization therapy (ERT) is becoming increasingly important in the management of acute ischemic stroke (AIS), but the hospital threshold volume for optimal ERT remains unknown. We investigated whether there was a correlation between hospital volume of ERT and risk-adjusted patient outcomes.<br/> <br/> Materials and Methods: <br/> From the National Health Insurance (NHI) claims data, we selected 11,745 AIS patients who underwent ERT from 2011 to 2015, in 111 hospitals. The outcome measures of the study were 30-day mortality, readmission rate, and symptomatic intracranial hemorrhage (ICH) rate. For each outcome measure, risk-adjusted mortality prediction models incorporating demographic variables, modified Charlson comorbidity index, and stroke severity index (SSI) were built, and validated through a comparison with two independent hospital registries. Hospitals were divided into quartiles and risk-adjusted outcomes of AIS cases were compared across the hospital quartile to confirm volume-outcome relationship (VOR) in ERT. Spline regression was performed to determine the volume threshold.<br/> <br/> Results:<br/> The mean AIS volume was 14.8 cases per hospital per year and the unadjusted means of outcome variables were mortality 11.6%, readmission 4.6%, and ICH rate 8.6%. The VOR was observed in the risk-adjusted 30-day mortality rate across all quartile groups, and in the ICH rate between the first and fourth quartiles (P value <0.05). Volume threshold was 24 cases per year, which was estimated by examining the relative effect on the adjusted odds of outcome for an increase in hospital volume by 10 cases/year for a hospital of a given size.<br/> <br/> Discussion: <br/> In this study, the trends of most ERT procedures performed in Korea for the past 5 years, demographic information of patients, and patient histories were obtained. Hyperlipidemia showed abnormally high, and it seems to be the matter of claims of health insurance. Comparing administrative and clinical database showed relatively good result, but there was no comparative studies. For derivation and validation of SSI, risk-adjustment modelling methods were successful, and these methods would be helpful in the future studies. In the VOR analysis, there were many low-volume good-outcome hospitals, so various dichotomous comparisons from quartile groups were made to calculate the volume thresholds. Volume threshold may be used for hospital management of ERT. However, as technology and knowledge of ERT continues to evolve, and other clinical factors critical to the ERT outcomes were not included into the category of severity correction, the above results should be used with caution.<br/> We found an association between hospital volume and outcomes, and we identified the volume threshold in ERT. Policies to ensure the implementation of the AIS volume threshold for hospitals performing ERT is needed.