Background: Transient Staphylococcus aureus bacteremia (TSAB) is frequently encountered in hospitalized patients, yet physicians lack guidance in the management of these patients.
Methods: We retrospectively reviewed all cases of S. aureus bacteremia at Detroit Receiving Hospital from January 1, 2006 to December 31, 2006 and compared patients with TSAB to patients with persistent S. aureus bacteremia (PSAB).
Results: There were 112 patients in the study: 35 patients with TSAB and 77 patients with PSAB. Injection drug use was less common among patients with TSAB than those with PSAB (P = 0.001). More patients with TSAB had surgical wounds (P = 0.033); metastatic infection was more common among patients with PSAB (P < 0.001). The time to positivity for blood cultures was longer for TSAB (27.03 hours) than for PSAB (18.42 hours; P = 0.014) and patients with TSAB were more likely to become bacteremic more than 48 hours after admission (P = 0.046). Vascular catheters and skin/soft tissue/bone infection were the most frequent portals of entry of bacteremia for patients with TSAB; primary bloodstream infection was more common in PSAB (P < 0.001). The rates of readmission, relapse/recurrence of SAB and death were similar for the 2 groups.
Conclusion: Compared to patients with PSAB, patients with TSAB are more likely to develop TSAB as a hospital-acquired infection and are at very low risk of metastatic infection. In addition, the prolonged time for their blood cultures to become positive and their rapid clearance of bacteremia make endovascular infection unlikely. Our findings suggest that physicians faced with patients with TSAB may be confident in providing a relatively short course of antibiotic therapy.
Summary: Patients with transient Staphylococcus aureus bacteremia have different characteristics from patients with persistent S. aureus bacteremia. These differences may be used to help identify patients who might benefit from relatively short courses of therapy.
From the Wayne State University, Detroit, MI.
Correspondence to: Donald P. Levine, MD, Wayne State University Health Center, Suite 5C, 4201 St Antoine, Detroit, MI 48201. E-mail: firstname.lastname@example.org.
GR has no conflicts of interest to disclose.
DPL has received research support from Cubist, Theravance, Nabi, Astellas, 3M, and Johnson & Johnson; has served as a consultant to Cubist, Pfizer, Forest and Rib-X; and has served on speaker’s bureaus for Cubist, Novartis, Astellas, and Forest.