M Thanassi1. 1. Yale University School of Medicine, Department of Surgery, New Haven, CT 06519-1315, USA.
Abstract
OBJECTIVE: To determine how often the results of urine and blood cultures led to changes in antibiotic therapy for patients discharged from the hospital with the diagnosis of pyelonephritis. METHODS: A retrospective chart review was performed of consecutively admitted patients, 10-90 years old, with an ICD-9 discharge diagnosis of acute pyelonephritis. All patients were admitted to a university-based, tertiary care center and a large HMO medical center from 1993 to 1994. The association of urine and blood culture results with a change in antibiotic therapy was assessed. RESULTS: Of the 194 patients who met inclusion criteria, 189 (97%) had urine cultures obtained at the time of admission and 139 (71%) had blood cultures obtained. Ampicillin, gentamicin, or both were given as initial antibiotics 81% of the time, and isolated organisms from urine or blood were sensitive to the empiric antibiotics 95% of the time. Most (171/189; 90%) urine cultures were positive, but only 9 (5%) of these led to a change in antibiotic therapy. 80% of the urinary pathogens were Escherichia coli, 5% Enterococcus, 5% Proteus, and 4% Klebsiella. Only 40 (29%) of the 139 blood cultures were positive; none prompted a change in antibiotics. There were no cases in which blood and urine cultures grew different pathogens. CONCLUSIONS: Urine cultures are useful in directing antibiotic therapy in patients with the discharge diagnosis of acute pyelonephritis and support a change in therapy in 5% of cases. Among the patients in this study, blood cultures results did not lead to changes in antibiotic therapy. These findings warrant prospective, multicenter evaluation.
OBJECTIVE: To determine how often the results of urine and blood cultures led to changes in antibiotic therapy for patients discharged from the hospital with the diagnosis of pyelonephritis. METHODS: A retrospective chart review was performed of consecutively admitted patients, 10-90 years old, with an ICD-9 discharge diagnosis of acute pyelonephritis. All patients were admitted to a university-based, tertiary care center and a large HMO medical center from 1993 to 1994. The association of urine and blood culture results with a change in antibiotic therapy was assessed. RESULTS: Of the 194 patients who met inclusion criteria, 189 (97%) had urine cultures obtained at the time of admission and 139 (71%) had blood cultures obtained. Ampicillin, gentamicin, or both were given as initial antibiotics 81% of the time, and isolated organisms from urine or blood were sensitive to the empiric antibiotics 95% of the time. Most (171/189; 90%) urine cultures were positive, but only 9 (5%) of these led to a change in antibiotic therapy. 80% of the urinary pathogens were Escherichia coli, 5% Enterococcus, 5% Proteus, and 4% Klebsiella. Only 40 (29%) of the 139 blood cultures were positive; none prompted a change in antibiotics. There were no cases in which blood and urine cultures grew different pathogens. CONCLUSIONS: Urine cultures are useful in directing antibiotic therapy in patients with the discharge diagnosis of acute pyelonephritis and support a change in therapy in 5% of cases. Among the patients in this study, blood cultures results did not lead to changes in antibiotic therapy. These findings warrant prospective, multicenter evaluation.