A National Analysis of the Relationship Between Hospital Volume, Academic Center Status, and Surgical Outcomes for Abdominal Hysterectomy Done for Leiomyoma

Abstract

Volume-to-outcomes relationships have been established for high-risk surgical procedures. To determine whether hospital volume and academic center status affect surgical outcomes in a lower-risk procedure, morbidity and mortality in patients undergoing abdominal hysterectomy for leiomyoma were evaluated. Administrative data from the National Inpatient Sample were used to conduct a retrospective analysis of 172,344 individuals who had primary diagnoses of leiomyomata (ICD-9 diagnosis codes of 218.x in the first 2 positions) and who underwent abdominal hysterectomy (ICD-9 procedure codes 68.4 in the first 2 positions) from 1999 to 2003. Comparison was made between teaching hospitals versus nonteaching hospitals and annual case volume in quintiles. Morbidity was considered to be any postoperative condition that is not an expected outcome of hysterectomy and defined as instances in which a patient suffered hemorrhage, ureteral injury, bladder injury, intestinal injury, wound dehiscence, wound infection, deep vein thrombosis, pulmonary embolism, or required blood transfusion. A total of 37 deaths were observed. Mortality was not significantly related to hospital volume or academic medical center status. In contrast, morbidity was found to have a positive association with academic medical center status (odds ratio = 1.34; 95% CI, 1.23 to 1.45), although an inverse relationship between volume and morbidity was observed for extended length of stay (> 3 days) and blood transfusion outcomes in the first 3 (lowest) volume quintiles and for pulmonary embolism in the highest-volume quintile. No important association with volume was found for hemorrhage, ureteral injury, bladder injury, or intestinal injury. Unlike high-risk procedures, such as esophagectomy, pediatric cardiac surgery, and pancreaticoduodenectomy, mortality for abdominal hysterectomy done for benign indication does not improve with hospital volume or academic center status. The statistically significant positive association between academic medical center status and morbidity merits additional characterization to target areas for improvement.