Background and aim of the study: Staphylococcus aureus is a leading cause of bacteremia and is often associated with endocarditis. The diagnosis of endocarditis may be missed when relying on clinical risk prediction, and this has led others to recommend transesophageal echocardiography (TEE) for diagnosis in most cases of S. aureus bacteremia (SAB). The study aim was to determine the likelihood of finding vegetations on TEE in patients with SAB in a suburban teaching hospital setting, and to identify risk factors predictive of vegetation on TEE.
Methods: All cases of SAB at Walter Reed Army Medical Center between January 2000 and May 2003 were evaluated. The prevalence of vegetations was determined in those cases selected for TEE. Potential risk factors for endocarditis were analyzed by review of medical records.

Results: A total of 176 patients had documented SAB during the time frame of the study, and 64 of these had TEE performed. Among the latter patients, 14% had a previously unidentified vegetation discovered by TEE. Patients with vegetation on TEE were as likely as those without vegetation to have nosocomial bacteremia, an alternate source of infection, and lack of valvular disease by prior surface echocardiography. Patients with a vegetation were significantly older (mean age 68.4 ± 10.9 versus 54.6 ± 19.6 years; p = 0.04).
Conclusion: TEE identified a significant number of vegetations resulting from SAB. The clinical risk profile and transthoracic echocardiography did not reliably exclude vegetation. These findings support the liberal use of TEE for the diagnosis of SAB.

The Journal of Heart Valve Disease 2005;14:23-28