BACKGROUND: The purpose was to assess the current variation in complication rates and evaluate the association between specific types of complications and in-hospital mortality and total hospital charges for patients having abdominal aortic surgery. PATIENTS/ METHODS: We studied 2987 patients for abdominal aortic surgery in Maryland from 1994 to 1996 and used discharge diagnoses and procedure codes to identify diagnoses that most likely represent major surgery complications. We evaluated how in-hospital mortality and total hospital charges related to specific complications, adjusting for patient demographics, severity of illness, comorbidity, and hospital and surgeon volumes. Discharge data was obtained from the hospital marketing departments. RESULTS: Complication rates varied widely among hospitals. Complications independently associated with increased risk of in-hospital death include cardiac arrest with an odds ratio (OR) of 90 and a 95% confidence interval (CI) of 32-251, septicemia (OR 6.1, CI 3.3-11.3), acute myocardial infarction (OR 5.7, CI 2.3-14.3), acute renal failure (OR 5.0, CI 2.3-11.0), surgical complications after a procedure (OR 3.1, CI 2.0-4.9), and reoperation for bleeding (OR 2.2, CI 1.1-4.8). The population-attributable risk for in-hospital mortality was 47% for cardiac arrest and 27% for acute renal failure. CONCLUSIONS: In abdominal aortic surgery on patients in Maryland, the rates of some complications vary widely and are independently associated with increased in-hospital mortality and hospital charges (charges differ from costs). Efforts to reduce these complications should help to decrease both levels.
BACKGROUND: The purpose was to assess the current variation in complication rates and evaluate the association between specific types of complications and in-hospital mortality and total hospital charges for patients having abdominal aortic surgery. PATIENTS/ METHODS: We studied 2987 patients for abdominal aortic surgery in Maryland from 1994 to 1996 and used discharge diagnoses and procedure codes to identify diagnoses that most likely represent major surgery complications. We evaluated how in-hospital mortality and total hospital charges related to specific complications, adjusting for patient demographics, severity of illness, comorbidity, and hospital and surgeon volumes. Discharge data was obtained from the hospital marketing departments. RESULTS: Complication rates varied widely among hospitals. Complications independently associated with increased risk of in-hospital death include cardiac arrest with an odds ratio (OR) of 90 and a 95% confidence interval (CI) of 32-251, septicemia (OR 6.1, CI 3.3-11.3), acute myocardial infarction (OR 5.7, CI 2.3-14.3), acute renal failure (OR 5.0, CI 2.3-11.0), surgical complications after a procedure (OR 3.1, CI 2.0-4.9), and reoperation for bleeding (OR 2.2, CI 1.1-4.8). The population-attributable risk for in-hospital mortality was 47% for cardiac arrest and 27% for acute renal failure. CONCLUSIONS: In abdominal aortic surgery on patients in Maryland, the rates of some complications vary widely and are independently associated with increased in-hospital mortality and hospital charges (charges differ from costs). Efforts to reduce these complications should help to decrease both levels.