• Vascular Annual Meeting
• General Information
• Call for Abstracts
• Exhibitor Information
• Future Meeting Dates
• Past Meeting Information
• Contact Us

Hospital Characteristics, Volume Thresholds and In-hospital Mortality After Endovascular Aortic Aneurysm Repair in the United States

Carlos H Timaran¹, Frank J Veith², Eric B Rosero¹, John G. Modrall¹, Frank Arko¹, Rawson J Valentine¹, George P Clagett.^1
1Univ of Texas Southwestern Med Ctr, Dallas, TX; ²Cleveland Clinic Foundation, Cleveland, OH.

OBJECTIVES: A recent evidence report prepared for the Agency for Healthcare Research and Quality (AHRQ) on surgical repairs for nonruptured abdominal aortic aneurysm (AAA) revealed that there are no data to estimate the effect of hospital volume on endovascular repair (EVAR) outcomes. Our purpose was to define volume thresholds and the influence of other hospital factors on risk-adjusted in-hospital mortality after EVAR in the United States (US).

METHODS: The Nationwide Inpatient Sample (NIS) identified EVAR procedures for nonruptured abdominal aortic aneurysms (AAA) during the years 2000-2004. Risk stratification was based on the Charlson comorbidity index. Weighted polynomial and logistic regression analyses were used to determine the association between hospital factors and risk adjusted in-hospital mortality taking into account the NIS sample design.

RESULTS: 67,923 EVARs were performed during the 5-year period with an in-hospital mortality of 2.2% (1487 of 67,923). Of these, 85% were elective EVAR procedures. Elective EVAR had significantly lower in-hospital mortality than nonelective EVARs (1.1% vs. 9.2%, respectively; P <.001). Increasing elective hospital volume had a significant nonlinear correlation with decreasing in-hospital mortality for EVARs (Spearman r, -.32; P=.003). Logistic regression models identified 10 elective EVARs per year as the optimal cutoff value below which decreasing hospital volume resulted in significantly higher in-hospital mortality. Of interest, the decrease in mortality was rather small above this cutoff point. The odds of in-hospital mortality were 2.1 times greater (95% confidence interval, 1.8-2.5) for patients undergoing procedures in hospitals with low volumes (<10 EVARs) compared to high-volume hospitals ((≥10 EVARs) after adjusting for age, sex and comorbidity index. Other hospital characteristics (location, teaching status and type of ownership) did not influence in-hospital mortality.

CONCLUSIONS: NIS data analysis suggests lower hospital volume correlates with higher in-hospital mortality for elective EVAR procedures. Such correlation, however, is nonlinear and less significant reduction in elective in-hospital mortality may be demonstrated after exceeding a threshold hospital volume of 10 EVAR procedures.