Time-related distribution, risk factors and prognostic influence of embolism in patients with left-sided infective endocarditis

https://doi.org/10.1016/j.ijcard.2005.07.016Get rights and content

Abstract

Background

Few studies evaluated systemic arterial embolism after beginning of symptoms of infective endocarditis in a large series of patients.

Methods

We studied 629 patients with left-sided infective endocarditis, aged 37.9 ± 17.3 years, 396(63%) men and 233(37%) women. Endocarditis occurred on native valves in 405(64.4%) patients and on prosthetic heart valves in 224(35.6%). Infecting microorganisms were streptococci in 297(47.3%) patients, Staphylococcus aureus in 77(12.3%), Staphylococcus epidermidis in 56(8.9%), enterococci in 51(8.1%), Gram-negative bacteria in 33(5.2%), fungi in 9(1.4%) and other microorganisms in 27(4.2%). In 79(12.6%) patients blood cultures were negative.

Results

146 embolic events occurred in 133(21.1%) out of 629 patients; in 63(47.4%) of them emboli affected the central nervous system, in 57(42.9%) affected peripheral organs and in 13(9.7%) affected both the central nervous system and peripheral organs. Embolism occurred between beginning of symptoms of endocarditis and antimicrobial therapy in 56(42.1%) patients and on the day therapy started in 18(13.5%); 109(81.9%) embolic events occurred up to the 15th day of antimicrobial therapy.

Embolic risk was higher in S. aureus endocarditis (relative risk 2.97); in patients with a mitral (relative risk 2.4) or aortic (relative risk 3.3) prosthetic valve and vegetations on echocardiography. Embolic risk was lower in patients with a longer duration of symptoms. The death risk doubled in patients with embolism (relative risk 2.01).

Conclusions

Embolic events were more frequently early events after beginning of symptoms of infective endocarditis. Embolic risk was higher in S. aureus endocarditis and in patients with prosthetic heart valves and vegetations on echocardiography.

Introduction

Systemic arterial embolism which is among the common extracardiac complications in patients with infective endocarditis [1], [2], [3], was reported to occur from 12.9% to 46% of the cases in recent experiences [4], [5], [6], [7], [8], [9], [10], [11], [12], [13] and may carry significant morbidity and mortality [1], [2], [3], [14].

Significant proportion of embolic events in clinical practice occurs before patients seek medical care, therefore prior to institution of effective medical treatment [1] as re-emphasized in studies of 288 patients [10] and 80 patients [11]. In addition, the frequency of emboli was demonstrated to decrease after the beginning of antimicrobial therapy [4], [13], [15]. These data substantiate the hypothesis that arterial embolism is more frequently an early event in the course of the disease. Few recent studies have systematically analyzed the occurrence of systemic arterial embolism in a large series of patients with infective endocarditis in a time-related fashion, more specifically after the beginning of the symptoms of endocarditis.

We evaluated the occurrence of symptomatic systemic arterial embolism after the beginning of symptoms of endocarditis in a large series of patients with infective endocarditis, in order to ascertain its frequency, the time-related distribution and risk factors.

Section snippets

Study design

We retrospectively analyzed an inception cohort of 757 patients with 824 episodes of infective endocarditis treated in a tertiary 423-bed university hospital dedicated mainly to cardiology from October 1978 through November 2001.

The diagnosis of infective endocarditis was made according to previously published criteria [16], [17], [18] and required (1) clinical findings compatible with the disease (fever, predisposing heart disease, new regurgitant murmur, and vascular phenomena); (2) the

Frequency of embolic events and clinical characteristics of the patients with embolism

One hundred forty-six embolic events occurred in 133 (21.1%) out of the 629 patients; 89 (67%) of them were men and 44 (33%) women. Thirteen patients (9.8%) had more than one episode of arterial embolism. Subsequent analysis will be based on the first embolic event.

Embolism involved the central nervous system in 63 (47.4%) patients, peripheral organs in 57 (42.8%) patients, and both central nervous system and peripheral organs in 13 (9.8%) patients.

Embolism to peripheral organs (n = 57) affected

Discussion

The frequency of a first symptomatic embolic event in the present series was 21.1% (133 out of 629 patients), a rate that is in the broad range of the more recently reported rates of embolism in patients with infective endocarditis between 12.9% to 46% of the cases in recent experiences [4], [5], [6], [7], [8], [9], [10], [11], [12], [13].

Recurrence of embolism was found in 9.8% (6 out 133) of the patients, suggesting that this occurrence was a less frequent event with current treatment. This

References (28)

  • E. Mylonakis et al.

    Infective endocarditis in adults

    N Engl J Med

    (2001)
  • Horstkotte D (chairperson), Follath F, Gutschik E, et al. Task Force Members on Infective Endocarditis of the European...
  • P. Tornos et al.

    Infective endocarditis in Europe: lessons from the Euro heart survey

    Heart

    (2005)
  • G. Di Salvo et al.

    Endocarditis in the elderly: clinical, echocardiographic, and prognostic features

    Eur Heart J

    (2003)
  • Cited by (65)

    • Coronary Embolism: A Systematic Review

      2020, Cardiovascular Revascularization Medicine
    • Clinical and echocardiographic predictors of embolism in infective endocarditis: systematic review and meta-analysis

      2019, Clinical Microbiology and Infection
      Citation Excerpt :

      Among 11 studies including patients with EEs occurring at various times and reporting data, the overall proportion of patients with EEs whose embolism occurred before initiation of antibiotics was 53.3% (95% CI 38.0%–68.1%; see Supplementary material, Fig. S2) [3,6,8,11,13,26,28,30,43,44,56]. Of 34 studies pooled in meta-analyses of echocardiographic characteristics, six used transthoracic echocardiography [24,27,30,36,58,59], six used trans-oesophageal echocardiography [6,8,21,26,34,52], 16 used both [3,5,11,22,23,31–33,37,44,46,48,49,53,56,57], and six did not specify the modality [13,28,45,47,51,55]. The Supplementary material (Table S2) summarizes Newcastle–Ottawa scale assessments of study quality.

    View all citing articles on Scopus
    View full text