Presence of vegetations on echocardiography did not predict embolic events in native valve infective endocarditis
ACP J Club. 1991 July-Aug;115:27. doi:10.7326/ACPJC-1991-115-1-027
Steckelberg JM, Murphy JG, Ballard D, et al. Emboli in infective endocarditis: the prognostic value of echocardiography. Ann Intern Med. 1991 Apr 15;114:635-40.
To determine whether vegetations visualized on 2-dimensional echocardiography are predictive of embolic events in patients with infective endocarditis, and to assess the timing of emboli relative to the initiation of antimicrobial therapy.
Historical inception cohort assembled at the time of initiation of effective antimicrobial therapy.
Tertiary referral center: Mayo Clinic, Rochester, Minnesota, USA.
Patients diagnosed with active, left-sided, native valve infective endocarditis between 1978 and 1987 were identified from a medical index. 207 patients (median age 60 y) who had a 2-dimensional echocardiographic examination within 72 hours of starting antimicrobial treatment were included.
Assessment of prognostic factors
Vegetation status was classified as absent, indeterminate, or definite by 2 investigators blinded to the patient's clinical course.
Main outcome measure
Embolic event occurring during the period of risk (beginning with the initiation of therapy and ending with successful completion of therapy, embolic event, surgical valve replacement, death, or hospital discharge), as determined by review of medical records.
Vegetations were present in 79 patients (38%), absent in 82 patients (40%), and indeterminate in 46 patients (22%). The crude incidence rate of first embolic events was 6.2 per 1000 patient-days (95% CI 4.2 to 9.2). The incidence of first embolic events was 7.1 per 1000 patient-days in patients with vegetations and 4.9 per 1000 patient-days in patients without vegetations, with an incidence rate ratio of 1.4 (CI 0.6 to 3.3, P > 0.2). No increased risk for embolus was seen with increasing vegetation size (P > 0.2). The observed effect of vegetations on embolic rate varied depending on the causative microorganism, ranging from no effect (relative risk 1.0, CI 0.2 to 3.9) in patients with Staphylococcus aureus endocarditis to a relative risk of 6.9 (CI 1.1 to 42) in patients with viridans streptococcal endocarditis (P < 0.05). The rate of first embolic events diminished over time, falling from 13 per 1000 patient-days during the first week of therapy to < 1.2 per 1000 patient-days after completion of week 2 of therapy (P < 0.001).
The presence of vegetations on echocardiography was not predictive of future embolic events in patients with left-sided native valve infective endocarditis. The rate of embolic events declined with time after initiation of antimicrobial treatment.
Source of funding: In part, Merck Sharp and Dohme/Society for Epidemiologic Research Clinical Epidemiology Fellowship Program.
Address for article reprint: Dr. J.M. Steckelberg, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, USA.
Echocardiography allows the identification of valvular vegetations in approximately 50% to 75% of patients with infective endocarditis. Most studies have reported that patients in whom vegetations were seen on echocardiography had a higher risk for complications, including congestive heart failure, embolization, valve surgery, and death. Because of this, early valve replacement has been recommended by some investigators if vegetations are seen on echocardiography.
This is a well-designed historical cohort study with more patients included than have been in previous studies in this area. Patients were restricted to those with native valve, left-sided endocarditis. Presence of vegetations on echocardiography and embolic events were assessed by blinded investigators and a clinically relevant time frame was used, starting with the initiation of effective antibiotic treatment. In addition, complete follow-up was obtained. The results provide important information regarding the marked reduction of the embolic event rate during antimicrobial therapy and support a more conservative approach to patients with vegetations.
Some caution is needed, however, regarding the conclusion that vegetations on echocardiography were not predictive of future embolic events. The power to show differences in a cohort study depends on the number of events of interest observed. Only 27 patients (13%) had an embolic event, and 7 of those had an inconclusive echocardiogram. In all previous studies, a higher risk for embolism existed when vegetations were detected, although in some of these studies the differences did not reach statistical significance.
Similar caution is needed in interpreting the results of the stratified analysis, which suggest a relation between vegetations and risk for embolism in patients with viridans streptococci endocarditis. Misclassification of only 1 patient could potentially change the results. This finding has not been previously reported and needs to be verified in future studies.
Fernando T. Lanas, MD
Universidad de la FronteraTemuco, Chile