Conservative Management of Isolated Posthysterectomy Fever

This is the authors' abstract. We don't add key points for this paper.

A postoperative fever may signal a potentially serious bacterial infection, or it may be a natural reaction to the trauma of surgery and anesthesia and not require workup or treatment. Since 1993, a protocol has been in place at the Davis Grant Medical Center that limits routine studies and treatment for postoperative fever. Women who developed a postoperative fever greater than 38°C underwent a physical examination and were treated according to the findings of that examination. When the findings were normal, no further workup was performed. Clinical studies were ordered only when indicated by the examination or when the fever persisted beyond 72 postoperative hours. Antibiotic therapy was initiated according to the results of clinical findings and physical symptoms only. Patients with normal findings did not receive antibiotics. This article presents the results of a review of the charts of patients who underwent hysterectomy during a 18-month period according to this protocol. One hundred twelve women underwent hysterectomy (72 abdominal and 40 vaginal) during this period and met the criteria for the study (20 patients were ineligible). Fifty-one women had at least one postoperative febrile episode during the first 72 hours. Nine patients (8 percent) had fever persisting for 3 or more days. In 63 percent (32 of 51) of these patients, no apparent cause of fever was identified. Urinalysis was performed in 12 patients who reported dysuria. Cystitis was diagnosed in five of these women, three of whom also had fever. All were successfully treated with oral antibiotics. Seven patients were evaluated for vaginal cuff masses, five with pelvic ultrasound and two with computed tomography. Vaginal cuff cellulitis was diagnosed in one woman and vaginal cuff abscess in five, all of whom were treated with intravenous antibiotics. In two patients surgical drainage of the vaginal cuff abscess was necessary. One woman developed a wound infection that required readmission and treatment with intravenous antibiotics. Statistical analyses comparing patients with postoperative infection to patients without infection found the sensitivity and specificity of fever for postoperative infection to be 57 and 55 percent, respectively, resulting in a low predictive value of fever for infection of 7.8 percent. J Reprod Med 2000;45:572–576