The value of infectious diseases consultation in Staphylococcus aureus bacteremia

Abstract

Background: Staphylococcus aureus bacteremia results in substantial mortality. Infectious diseases specialist consultation can improve adherence to evidence-based management of S. aureus bacteremia, but its effect on mortality is unclear.

Methods: A 2-year prospective cohort study of patients with S. aureus bacteremia was performed at a large tertiary care hospital. Patients who died within 2 days of diagnosis were excluded. Independent risk factors for 28-day mortality were determined.

Results: Among 341 patients with S. aureus bacteremia, 189 (55%) were male, 196 (58%) were Caucasian, 185 (54%) had methicillin-resistant S. aureus, 108 (32%) had nosocomial bacteremia, and 231 (68%) had a central venous catheter at the time of diagnosis. The median age was 56 years (range 22-95 years). A total of 111 patients (33%) had an infectious diseases consultation. Fifty-four patients (16%) died within 28 days after diagnosis. Factors associated with mortality were intensive care unit admission 48 hours or less after the first positive blood culture (adjusted hazard ratio, 4.65; 95% confidence interval [CI], 2.65-8.18), cirrhosis (adjusted hazard ratio, 4.44; 95% CI, 2.40-8.20), and advanced age (adjusted hazard ratio, 1.27 per every 10 years of age; 95% CI, 1.08-1.50). Infectious diseases consultation was associated with a 56% reduction in 28-day mortality (adjusted hazard ratio, 0.44; 95% CI, 0.22-0.89).

Conclusion: Only one third of patients with S. aureus bacteremia in this cohort had an infectious diseases specialist consultation. Infectious diseases consultation was independently associated with a reduction in 28-day mortality. Routine infectious diseases consultation should be considered for patients with S. aureus bacteremia, especially those with greater severity of illness or multiple comorbidities.

Figure 1. Extended Kaplan-Meier Curve of S. aureus Bacteremia Based on Infectious Diseases Consultation

Note. ID, infectious diseases. Extended Kaplan-Meier curves did not correspond to fixed patient cohorts, but allowed for the cohorts to be updated at each event time. The patient without an infectious disease consultation was included in one curve, until consultation was obtained and the patient is then included in the consultation curve. Adjusted hazard ratio for 28-day mortality was 0.44 (P =.022) and the crude hazard ratio for 29-365 day mortality was 0.88 (P =.579).