Background: The population level frequency of adverse events after antireflux procedures and its relationship to surgical experience has not been well studied.
Study design: Two parallel retrospective, population-based cohort studies were conducted using the Washington State discharge database and the United States Health Care Utilization Project (HCUP) database. All adult patients assigned ICD-9 procedure codes for antireflux surgery from 1992 to 1997 were included. The frequency of case fatality, splenectomy, and esophageal injury was measured. In Washington State, the relationship of adverse outcomes to the cumulative number of procedures performed by a given surgeon (case-order) was determined.
Results: Nationwide, an estimated 86,411 patients underwent antireflux surgery between 1992 and 1997. Splenectomy was performed in 2.3%, suture of esophageal laceration in 1.1%, and in-hospital death occurred in 0.8%. Adverse events were significantly more likely when procedures at case-order less than or equal to 15 (median) were compared with those at case-order greater than 15. As case-order increased by 1, the risk of death decreased by 1.7% (p = 0.001), and the risk of splenectomy and injury repair decreased by 1.6% (p = 0.001). If performed at case-order less than 15, the odds ofsplenectomy were 2.7 times, esophageal laceration repair 2.3 times, and death 5.6 times greater than the odds of adverse outcomes for procedures performed at later case-orders.
Conclusions: On a national level, morbidity and mortality associated with antireflux surgery performed in the 1 990s was quite low, but was somewhat higher than suggested by case series. Surgical experience with the procedure was linked to better outcomes.