Conduction Abnormalities in Infectious Endocarditis

Abstract & Commentary

Synopsis: ECG conduction changes are not uncommon in infectious endocarditis and are associated with invasive infection and a higher mortality.

Source: Meine TJ, et al. Am Heart J. 2001;142: 280-285.

The development of cardiac conduction abnormalities has long been considered an ominous sign in patients with infectious endocarditis, but there are little objective data on this complication in the echocardiographic era. Thus, Meine and colleagues prospectively evaluated cardiac-conduction defects in patients with suspected endocarditis to determine their incidence and prognostic value. The Duke criteria were used to identify 137 patients with definite or possible endocarditis and an interpretable ECG. New conduction abnormalities or ones of indeterminant age were evaluated in this study, which included echocardiography. ECG conduction changes were observed in 36 of the 137 patients (26%), which included AV blocks and intraventricular blocks. Conduction abnormalities were more common in men and those with a prosthetic valve. Conduction abnormalities were not related to type of organism, but they were more common in patients with evidence of invasive infection. Although surgery was not more frequent in those with conduction abnormalities, mortality was higher (31% vs 15%; P < .04) or RR = 2.6; CI 1.01-6.18). Also, mortality was highest in those with intraventricular block (41%). Meine et al concluded that ECG conduction changes are not uncommon in infectious endocarditis and are associated with invasive infection and a higher mortality.

Comment by Michael H. Crawford, MD

Textbooks and older articles emphasize the dire prognosis of complete heart block in patients with infectious endocarditis, especially of the aortic valve. In this study, complete heart block was unusual (4%) and all these patients survived. However, this paper shows that any conduction abnormality, even first-degree AV block (one third of the conduction abnormalities) is associated with an increased mortality (31%) and intraventricular blocks such as bundle-branch block has the highest mortality (41%). Also, conduction abnormalities on ECG identified patients more likely to have invasive infection on echocardiography, which supports the pathophysiologic hypothesis that the conduction abnormalities are often due to extension of the infection into the conduction system. This would explain the higher incidence of conduction abnormalities in patients with prosthetic valves, since infection is most likely in the sewing ring rather than on the leaflets.

Despite modern diagnostic techniques, broad spectrum antibiotics, and advances in surgery, endocarditis remains a lethal disease with an overall 20% mortality in this study, and twice that in patients with conduction abnormalities. So should surgery be performed earlier in those with conduction abnormalities? It would seem to make sense, but in this study only 5% of the patients had surgery. Thus, the role of surgery cannot be ascertained. At this point, a conduction abnormality that is new or of indeterminant age should prompt consideration of surgery and be a factor if surgery is being contemplated, but it is not a sufficient reason alone for surgery.