Impact of surgeon and hospital experience on outcomes of abdominal aortic aneurysm repair in New York State.

OBJECTIVE: This study aimed to assess the impact of the surgeon's and hospital's experience on the outcomes of open surgical repair (OSR) and endovascular aneurysm repair (EVAR) of intact and ruptured abdominal aortic aneurysms (AAAs) in New York State.

METHODS: New York Statewide Planning and Research Cooperative System data were used to identify patients undergoing AAA repair from 2000 to 2011. Characteristics of the provider and hospital were determined by linkage to the New York Office of Professions and National Provider Identification databases. Distinct hierarchical logistic regression models for EVAR and OSR for intact and ruptured AAAs were created to adjust for the patient's comorbidities and to evaluate the impact of the surgeon's and hospital's experience on outcomes. The provider's years since medical school graduation as well as annual volume of the facility and provider are examined in tertiles. Adjusted odds ratios and 95% confidence intervals are presented.

RESULTS: A total of 18,842 patients underwent AAA repair by a vascular surgeon. For intact AAAs (n = 17,118), 26.2% of patients underwent OSR and 73.8% underwent EVAR. For ruptured AAAs (n = 1724), 63.9% underwent OSR and 36.1% underwent EVAR. After intact AAA repair, OSR adjusted outcomes were significantly influenced by the surgeon's annual volume but not by the facility's volume or the surgeon's age. The lowest volume providers (1-4 OSRs) had higher in-hospital mortality rates than high-volume (>11 OSRs) surgeons (adjusted odds ratio, 1.87 [95% confidence interval, 1.1-3.17]). Low-volume providers also had higher odds of major complications (1.23 [1-1.51]). For patients with intact AAA undergoing EVAR, mortality was higher at low-volume facilities (2.6 [1.3-5.3] and 2.7 [1.5-4.8] for <33 EVARs and 34-81 EVARs, respectively). After OSR for ruptured AAA, treatment at a low-volume facility (<9 OSRs for ruptured AAA) was associated with greater mortality than at high-volume (>27 OSRs for ruptured AAA) centers (1.56 [1.02-2.39]), whereas low-volume physicians (<4 OSRs for ruptured AAA) had higher odds of major complications (1.58 [1.04-2.41]). In the case of EVAR for rupture, there were no characteristics of the hospital or surgeon significantly associated with poorer outcomes.

CONCLUSIONS: For intact AAA, the surgeon's volume was an important factor for OSR outcomes, whereas low facility volume was associated with worse outcomes after EVAR. For ruptured AAA, low-volume surgeons and low-volume facilities had worse outcomes after OSR but not after EVAR. The interaction between the surgeon's volume and the hospital's volume is complex and varies on the basis of the acuity of presentation and treatment modality.