Multivariate analysis for predicting the presence of bacteria in bile in patients with acute cholecystitis.
OBJECTIVE: to find out whether the presence of bacteria in bile could be predicted accurately from preoperative data in patients with acute cholecystitis. DESIGN: Prospective open study. SETTING: University hospital. SUBJECTS: 42 patients undergoing cholecystectomy for acute gallstone cholecystitis. MAIN OUTCOME MEASURES: Correlations between 24 preoperative clinical and laboratory variables, and the incidence of pathogenic organism in bile. RESULTS: 4 of the 24 variables tested were of predictive significance. These were external body temperature on admission, percentage of neutrophils, preoperative white blood cell count, and total serum concentration of bilirubin. When these predictive variables were evaluated in the discriminant analysis equation they had a sensitivity of 92% and a specificity of 100% in predicting positive bile culture. CONCLUSION: Multivariate discriminant analysis permits accurate preoperative prediction of bile cultures growing pathogens in patient undergoing cholecystectomy for acute cholecystitis.
Duke Scholars
Published In
ISSN
Publication Date
Volume
Issue
Start / End Page
Location
Related Subject Headings
- Surgery
- Sensitivity and Specificity
- Risk Factors
- Prospective Studies
- Multivariate Analysis
- Middle Aged
- Male
- Humans
- Female
- Discriminant Analysis
Citation
Published In
ISSN
Publication Date
Volume
Issue
Start / End Page
Location
Related Subject Headings
- Surgery
- Sensitivity and Specificity
- Risk Factors
- Prospective Studies
- Multivariate Analysis
- Middle Aged
- Male
- Humans
- Female
- Discriminant Analysis