Current issues of ACP Journal Club are published in Annals of Internal Medicine


Patients were most likely to die or require valve replacement during the first 2 years after infective endocarditis

ACP J Club. 1993 Mar-April;118:54. doi:10.7326/ACPJC-1993-118-2-054

Source Citation

Tornos MP, Permanyer-Miralda G, Olona M, et al. Long-term complications of native valve infective endocarditis in non-addicts. A 15-year follow-up study. Ann Intern Med. 1992 Oct 1;117:567-72.



To document the incidence and clinical manifestations of long-term cardiac complications after a first episode of native valve infective endocarditis in non-addicts.


Inception cohort followed for 15 years.


University-affiliated tertiary care medical center in Spain.


140 patients hospitalized from 1975 to 1990 with a first episode of native valve infective endocarditis and who were not addicted to parenteral drugs. All patients were to receive antimicrobial therapy for 4 to 6 weeks, and, in addition, 48 patients had valve replacement. During this treatment period, 15 patients died postoperatively, and 7 patients died from medical complications of the disease. 6 patients were lost to follow-up. The follow-up consisted of 112 patients (median age 47 y), 32 who had had valve replacement and 80 who had medical treatment alone.

Assessment of prognostic factors

Potential predictors for late cardiac surgery and cardiac mortality were age, sex, previous heart disease, time to onset of symptoms, microorganism, involved valve, vegetations, vegetation size, end-diastolic and end-systolic diameters and fractional shortening of the left ventricle at discharge, and surgery during the first 4 to 6 weeks after study entry.

Main outcome measures

Relapse, recurrence, need for late cardiac surgery, and cardiac mortality including sudden death.

Main results

Relapses occurred in 3 patients (3%, 95% CI 0.5% to 7.6%), all within 1 month of hospital discharge. Recurrences occurred in 5 patients (5%, CI 1.4% to 10.0%). 38 patients (47%) treated medically during the initial 6-week period needed late cardiac surgery, most often in the first 2 years of follow-up. Multiple logistic regression analysis showed that aortic valve involvement was an independent predictor of the need for late cardiac surgery (relative risk 3.0, CI 1.2 to 7.5). During follow-up, 16 patients died of cardiac causes: Most deaths were sudden (n = 7) or postoperative (n = 4) and occurred in the first 2 years. Survival was 90% at 2 years, 88% at 5 years, 81% at 10 years, and 61% at 15 years. No significant predictors of cardiac mortality were found, although the statistical power for this analysis was limited.


Mortality and the need for valve replacement were highest during the first 2 years after infective endocarditis. Aortic valve involvement was an independent predictor of the need for late cardiac surgery.

Source of funding: Not stated.

For article reprint: Dr. M.P. Tornos, Servei de Cardiologia, Hospital General Universitari Vall d'Hebron, Po Vall D'Hebron s/n, Barcelona 08035, Spain. FAX 011-34-3-274-6063.


In the past, infective endocarditis was almost uniformly fatal. Until now, few data have been available on the prognosis of this condition in the antibiotic era. This study, therefore, makes an important contribution. In particular, the long-term prognosis of infective endocarditis appears to be poorer than might be expected. Although the current low short-term mortality is reassuring, the 10-year probability of survival was only 81%, and only 38% of these patients survived prosthesis-free until the end of follow-up.

Clinicians should be cautious in generalizing the results of this well-done study to their own patients. First, the institution from which the cases were drawn was both a general hospital admitting community patients and a referral center. Second, the definition of endocarditis used was stricter than the conventional definition. Finally, although the authors sought variables that could serve as predictors for late cardiac surgery and cardiac mortality, the sample sizes available for these analyses were small. Thus, although it is apparent that aortic valve involvement is predictive of the need for late surgery, it remains uncertain whether the other variables examined are predictors for either late surgery or mortality.

Nevertheless, although the patients in this study may have suffered from more serious endocarditis than those seen by the average practitioner, these data make clear that infective endocarditis remains a serious disease in the antibiotic era, not only in short-term morbidity and mortality, but in long-term as well. Most complications occur within the 2 years after the infection and are more likely in the presence of aortic valve involvement. These data suggest that the indications for surgery in such patients may need to be re-examined, especially in patients with aortic valve involvement.

Brian L. Strom, MD, MPH
University of PennsylvaniaPhiladelphia, Pennsylvania, USA