Extensive fever workup produces low yield in determining infectious etiology

Objective: The study was undertaken to evaluate the use of a fever workup in women undergoing benign gynecologic procedures. Study design: A retrospective chart review was performed at Jackson Memorial Hospital between 1994 and 2000. Information was abstracted from hospital and clinic records. Fever criteria was defined as 1 temperature equal to or greater than 101.5, or 2 equal to or greater than 100.4, at least 4 hours apart within a 24-hour period. Patients undergoing additional intraoperative procedures leading to increased febrile morbidity were excluded. Data abstracted included patient demographics, procedure, complications, antibiotic use, and extent of fever workup. Statistical analysis used was 2-sample t tests, Wilcoxon rank test, chi(2) test, and multivar.iate logistic regression. Alpha level =.05. Results: The charts of 505 patients were reviewed, and 147 patients met fever criteria. All patients underwent surgery for benign conditions, abdominal hysterectomy being the most common (90%). The study population was divided into 2 groups: the noninfectious group and infectious group. These groups were determined by wound infection, pelvic abscess, blood or urine culture, ultrasound, and chest roentgen. Both groups were found to be similar with respect to demographics, surgical procedures, and postoperative complications, with the exception of body mass index (28.4 vs 31.7) and length of hospital stay (3.9 vs 5.3). Results from fever workups included positive results blood cultures (9.7%), urine culture (18.8%), and chest roetgens (14%) in this study population. We found no association between positive urine analysis and urine culture. When comparing both groups, a statistically significant difference was found with regard to maximum temperature elevation, number of days febrile, and postoperative day of maximum temperature (P < .05). Conclusion: The extensive fever workup was not frequently positive in this study population. Its use and cost-effectiveness should be questioned. Therefore, the fever workup should be tailored to the individual patient. (c) 2005 Elsevier Inc. All rights reserved.