Md. Tischler et Pt. Vaitkus, THE ABILITY OF VEGETATION SIZE ON ECHOCARDIOGRAPHY TO PREDICT CLINICAL COMPLICATIONS - A METAANALYSIS, Journal of the American Society of Echocardiography, 10(5), 1997, pp. 562-568
To clarify whether echocardiographic detection of a vegetation 10 mm o
r larger in size in patients with left-sided infective endocarditis po
ses an increased risk for complications, we performed a meta-analysis
of English-language publications identified by a computerized search o
f the key words infective endocarditis and echocardiography. A pooled
odds ratio was calculated by using the Robins, Greenland, and Breslow
estimate of variance. The pooled odds ratio for increased risk of syst
emic embolization in the presence of a vegetation >10 mm (10 studies,
738 patients) was 2.80 (95% confidence interval [CI] 1.95 to 4.02; P <
0.01). The odds ratio of requiring valve-replacement surgery (seven s
tudies, 549 patients) was 2.95 (95% CI 1.90 to 4.58; p < 0.01). The od
ds ratio of death (six studies, 476 patients) was 1.55 (95% CI 0.92 to
2.60; p = 0.10). Thus this analysis supports the hypothesis that echo
cardiographically detected left-sided vegetations >10 mm pose a signif
icantly increased risk of (1) systemic embolization and (2) a need for
valve-replacement surgery than cases where either no or smaller veget
ations are detected.