Staph Bacteremia and Endocarditis
Staph Bacteremia and Endocarditis
ABSTRACT & COMMENTARY
Synopsis: Infectious endocarditis is not uncommon among hospitalized patients with Staphylococcus aureus (25%) and is associated with a five-fold increase in mortality.
Source: Fowler VG, et al. J Am Coll Cardiol 1997; 30:1072-1078.
Distinguishing patients with Staphylococcus aureus bacteremia (SB) from those with infectious endocarditis (IE) is difficult but of major clinical importance. Thus, Fowler and associates prospectively studied 103 consecutive patients with SB by transthoracic (TTE) and transesophageal (TEE) endocardiography. All the patients had fever, and at least one patient had a positive blood culture. These patients were followed-up for 12 weeks. IE was diagnosed by the Duke Criteria, which include clinical, echo, and pathologic evidence. Clinical or echo evidence of prior heart disease was found in 42 patients. Clinical evidence of IE was unusualoccurring in seven patients (5 peripheral emboli, 2 new murmurs), and only five patients met the Duke criteria for IE. Neither pre-existing heart disease nor clinical findings on presentation were useful for distinguishing SB from IE. TTE showed vegetations in seven, whereas TEE showed vegetations in 22. Definite IE by the Duke criteria was presented in 26 patients. The sensitivity and specificity of TTE was 32% and 100%, respectively, and the sensitivity and specificity of TEE was 100% and 99%, respectively (1 false-positive). Mortality was higher in those with IE (15%) vs. SB (3%, P = 0.03). Fowler et al conclude that IE is not uncommon among hospitalized patients with SB (25%) and is associated with a five-fold increase in mortality. TEE was superior to clinical evaluation and TTE for the detection of IE and should be part of the early evaluation of hospitalized patients with SB.
COMMENT BY MICHAEL H. CRAWFORD, MD
Although several studies, including one from my own institution, have shown the superiority of TEE vs. TTE for the diagnosis of IE, we have not been routinely performing TEE on all patients with SB. Because of the high specificity of TTE (100% in this study), we have performed TTE in SB patients without strong clinical evidence of IE and only done TEE if the TTE is positive or equivocal. This approach seemed justified given the frequency of SB in our patient population (county hospital) and the infrequent occurrence of IEless than the 25% frequency in this study. However, this study clearly shows the fallacy in our thinking, because the negative predictive value of TTE is only 81%. In fact, those with an equivocal TTE and those with a normal TTE had the same incidence of IE by TEEabout 20%. Fowler et al remarked that it was impossible to define a group by clinical or TTE findings in whom TEE was not necessary. The superiority of TEE was in the detection of small vegetations, abscesses, and valve perforations.
There are some limitations to this study. First, two-thirds of the patients had hospital-acquired SB, and almost all had known foci of infection. Second, there were few intravenous drug users, and few had metastatic foci of infection. Third, more than half had aortic and mitral IE, and only four had tricuspid valve disease. Fourth, only 59% of eligible patients were enrolled for a variety of reasons. Thus, there may be selection biases operant here, and the results may not apply to other populations such as community-acquired SB in IV drug users with prolonged illnesses. Also, the diagnosis of IE was largely based upon echo criteria since there was little pathologic data in this series. Using echo to test the accuracy of echo has conceptual problems. Therefore, the results of this study may be different in a longer duration of disease population with more surgery and death and consequently more pathologic material. TTE would likely perform better in such a population where more valve pathology would be expected as compared to the study population of early IE where TEE would be expected to be superior. Nevertheless, I am sufficiently impressed with this study, and I am going to modify my approach to hospitalized patients with SB and recommend early TEE. Remember, missed diagnosis is still the most common feature of successful law suits against physicians.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.