Is Transesophageal Echocardiography Dispensable in Hospital-Acquired Staphylococcus aureus Bacteremia?

Staphylococcus aureus is an important cause of community, health care– associated, and hospital-acquired bacteremia. In a study from US hospitals that reviewed .6000 episodes of bloodstream infection, S. aureus was the most common pathogen, accounting for 23% of all episodes, and was more strongly associated with death than any other bacterial pathogen [1]. Although gram-negative bacteria are increasing as a cause of hospital-acquired bacteremia, S. aureus is still responsible for 10%– 15% of all cases for which intravascular catheters are the most common source [2]. The management guidelines of S. aureus bacteremia (SAB) recommend 2 and 4–6 weeks of antimicrobial therapy for uncomplicated and complicated bacteremia, respectively [3, 4]. The problem is that complicated bacteremia includes infectious endocarditis (IE), and an early diagnosis of this entity requires performance of an echocardiograph. For this reason, the recent guidelines for catheter-related SAB recommend transesophageal echocardiography (TEE) for guiding the duration of antibiotic therapy. This issue is under continuous debate because TEE is not widely available, could be associated with complications, and is an expensive technique. The recommendation to perform a TEE is derived from clinical experiences that select patients with a high risk of IE [5, 6]. For instance, in a study by Fowler et al [5], the prevalence of IE was 25%; however, the authors excluded 73 patients with SAB for whom TEE was not performed either because the patient or the attending physician refused it. It is possible that, in these cases, the prevalence of IE was lower [5]. Another series included mainly community-acquired SAB, and the authors also found a high prevalence (22%) of IE [6]. Indeed, the community acquisition of SAB has been recognized as a risk factor for IE and currently is generally accepted as an indication for TEE [6]. However, according to different reports, the prevalence of IE among patients with hospital-acquired SAB is 6%–9% [7, 8], and it is unknown whether it is possible to identify a subgroup of patients with a very low probability of IE for whom TEE is dispensable. The article by Kaasch et al [9] addresses this important question with use of results from 2 large, well-conducted and prospective cohorts of consecutive patients with hospital-acquired SAB from 2 hospitals in Europe (Invasive S. aureus Infection Cohort [INSTINCT]) and the United States (S. aureus Bacteremia Group [SABG]). To select patients with low risk of IE, the authors evaluated a criteria set applied 6–8 days after the first positive blood culture result. The criteria consisted of the following: (1) prolonged bacteremia, when .4 days elapsed between the first blood culture yielding S. aureus and the first negative result of a follow-up blood culture (‘‘documented’’) or when follow-up blood cultures were not performed (‘‘possible’’); (2) the presence of a permanent intracardiac device (eg, prosthetic heart valve, pacemaker, or cardioverter-defibrillator); (3) hemodialysis dependency; and (4) spinal infection (eg, vertebral osteomyelitis epidural, subdural, or intraspinal empyema; or abscess) or nonertebral osteomyelitis. Both cohorts were similar, except the rate of methicillinresistant strains was significantly lower in the INSTINCT cohort (15.5%) than in the SABG cohort (65.7%). The prevalence of IE in both cohorts was in agreement with previous reports of hospital-acquired SAB (4.3% and 9.3%, respectively). The main finding is that the negative predictive value of these criteria was 99.5% (100% in INSTINCT and 99.2% in SABG), which means that only 1 of 208 patients Received 22 March 2011; accepted 28 March 2011. Correspondence: Alex Soriano, MD, PhD, Department of Infectious Diseases, Hospital Clinic of Barcelona, C/Villarroel 170, Barcelona 08036, Spain (asoriano@clinic.ub.es). Clinical Infectious Diseases 2011;53(1):10–12 The Author 2011. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail: journals. permissions@oup.com. 1058-4838/2011/531-0002$14.00 DOI: 10.1093/cid/cir305