Understanding Value and Patient Complexity Among Common Inpatient Vascular Surgery Procedures

OBJECTIVE Vascular surgery patients are highly complex, second only to patients undergoing cardiac procedures. However, unlike cardiac surgery, work relative value units (wRVU) for vascular surgery were undervalued based on an overall patient complexity score. This study assesses the correlation of patient complexity with wRVUs for the most commonly performed inpatient vascular surgery procedures. METHODS The 2014-17 National Surgical Quality Improvement Program Participant Use Data Files (NSQIP PUF) were queried for inpatient cases performed by vascular surgeons. A previously developed patient complexity score utilizing perioperative domains was calculated based on patient age, ASA class ≥4, major comorbidities, emergent status, concurrent procedures, additional procedures, hospital length of stay (LOS), non-home discharge, and 30-day major complications, readmissions, and mortality. Procedures were assigned points based on their relative rank and then an overall score was created by summing the total points. An observed to expected ratio (O/E) was calculated using open ruptured abdominal aortic aneurysm repair (rOAAA) as the referent and then applied to an adjusted median wRVU/operative minute. RESULTS Among 164,370 cases, patient complexity was greatest for rOAAA (complexity score =128) and the least for carotid endarterectomy (CEA) (complexity score=29). Patients undergoing rOAAA repair had the greatest proportion of ASA class ≥4 (84.8%; 95% CI=82.6-86.8%), highest mortality (35.5%; 95% CI=32.8-38.3%), and major complication rate (87.1%; 95% CI=85.1-89.0%). Patients undergoing CEA had the lowest mortality (0.7%; 95% CI=0.7-0.8%), major complication rate (8.2%; 95% 95% CI=8.0-8.5%), and shortest LOS (2.7 days; 95% CI=2.7-2.7). Median wRVU ranged from 10-42.1 and only weakly correlated with overall complexity (Spearman's ρ=0.11; p<0.01). Median wRVU/operative minute was greatest for thoracic endovascular aortic repair (0.25) and lowest for both axillary-femoral artery bypass (0.12) and open femoral endarterectomy, thromboembolectomy, or reconstruction (0.12). After adjusting for patient complexity, CEA (O/E=3.8) and transcarotid artery revascularization (O/E=2.8) had greater than expected O/E. In contrast, lower extremity bypass (O/E=0.77), lower extremity embolectomy (O/E=0.79), and open abdominal aortic repair (O/E=0.80) had lower than expected O/E. CONCLUSIONS Patient complexity varies substantially across vascular procedures and is not effectively captured by wRVUs. Increased operative time for open procedures is not adequately accounted for by wRVUs, which may unfairly penalize surgeons who perform complex open operations.