A nationwide study of conversion from laparoscopic to open cholecystectomy

Purpose To determine the national incidence and risk factors for conversion from laparoscopic to open cholecystectomy. Background Most series reporting the rates at which laparoscopic cholecystectomies are performed, relative to the open procedure, have come from centers specializing in laparoscopic surgery. The rates at which conversions occur from these centers may not reflect those in community practice. We sought to determine the actual, and therefore acceptable, conversion rate by examining nationally representative discharge data. Methods The National Hospital Discharge database for 1998 to 2001 was acquired from the Centers for Disease Control. All gallbladder disease related admissions were extracted, and the cholecystectomies (ICD-9-CM codes 51.2X) were analyzed using the SAS package. Stepwise logistic regression was used to determine what factors were associated with the risk of conversion from laparoscopic to open cholecystectomy. Results Approximately 25% of all cholecystectomies are performed by the open technique. Of the remaining 75%, there is an approximately 5% to 10% conversion rate. The major risk factors for conversion included male sex, obesity, and cholecystitis. Concurrent choledocholithiasis, cholelithiasis, and cholecystitis were associated with a conversion rate of 25%. Length of stay (LOS) was reduced for laparoscopic operations and although conversion added 2 to 3 days to the LOS, for most cases the LOS was still less than for primary open operations. Conclusions Three quarters of all cholecystectomies are performed laparoscopically, and the national conversion rate is 5% to 10%. Cholecystitis, choledocholithiasis, male sex, and obesity are major predictors for conversion. The data presented in terms of conversion rates and LOS were derived from population-adjusted hospital discharge data and represent the current U.S. experience for cholecystectomy. From these data the community experience for conversion rates, risk factors, and LOS can be derived.