Improving mortality trends for hospitalization of aortic dissection in the National Inpatient Sample.

OBJECTIVE: Population-based assessment of aortic dissection (AD) hospitalizations in the general United States population is limited. We assessed the current trends in AD admissions and in-hospital mortality for surgical and medical AD treatment.
METHODS: Patients admitted for primary diagnosis of AD were identified from the National Inpatient Sample database (2003-2012). Patients were identified by International Classification of Diseases-Ninth Revision diagnosis codes and categorized by treatment type: type A open surgical repair (TASR), type B open surgical repair (TBSR), thoracic endovascular aortic repair (TEVAR), and medical management (MM). Our primary outcomes were to evaluate admission trends and in-hospital mortality of AD. Secondary outcomes included postoperative complications. We used weighted national estimates of admissions to assess trends over time using linear regression. We also identified factors associated with mortality via a hierarchical multivariable logistic regression model.
RESULTS: We identified 15,641 patients (60.7% male; mean age, 63.5 years) admitted with a primary diagnosis of AD between 2003 and 2012. Intervention types included TASR in 3253 (20.8%), TBSR in 3007 (19.2%), TEVAR in 1417 (9.1%), and MM in 7964 (50.9%). Overall weighted admissions for AD increased significantly, from 6980 in 2003 to 8875 in 2012 (P < .01, test of trend), with increases in admission for TASR, from 1143 in 2003 to 2130 in 2012 (P < .01, test of trend), and TEVAR from 96 in 2005 to 1130 in 2012 (P < .01, test of trend). TBSR and MM admissions were stable, with TBSR admissions at 1519 in 2003 and 1540 in 2012 (P = .9, test of trend) and MM admissions at 4319 in 2003 and 4075 in 2012 (P = .8, test of trend). During the same interval, overall in-hospital mortality rates for AD decreased from 18.1% to 13.0% (P < .01, test of trend). When stratified by intervention type, mortality rates decreased for TASR, from 20.5% to 14.8% (P < .01, test of trend), for TBSR, from 18.0% to 14.3% (P = .03, test of trend), and for MM, from 17.5% to 13.9% (P < .01, test of trend). Mortality rates for TEVAR were stable, with an average mortality of 7.9% (P = .8, test of trend) during the study period. Factors associated with increased mortality included older age, Caucasian race, nonelective admission, pre-existing peripheral vascular disease, and acute postoperative complication of myocardial infarction, stroke, or kidney failure. Admissions at a center with high surgical volume were associated with a decreased mortality for TBSR admissions only (odds ratio, 0.55; 95% confidence interval, 0.4-0.7).
CONCLUSIONS: Overall and surgical admission rates for AD appear to be increasing, and in-hospital mortality rates are decreasing. TEVAR mortality remains mostly unchanged, however, suggesting targets for further improvement in mortality for AD treatment. Decreased mortality for TBSR at centers with a high surgical volume may suggest a need for regionalization of AD care.