Infective endocarditis: transoesophageal echocardiography was better at excluding perivalvular abscesses.

Clinical bottom line (level 4)

  1. Around a third of patients with infective endocarditis had perivalvular abscesses.
  2. Both transthoracic and transoesophageal echocardiography could help diagnose perivalvular abscesses.
  3. Transoesophageal echocardiography was better than transthoracic at excluding perivalvular abscesses, but neither could safely rule them out.
Daniel et al: New England Journal of Medicine 1991; 324: 795-800
Expires July 2003

The study

Setting: university hospital, Germany, 1984 to 1989

118 patients (aged range 17 to 74 years; mean 48, 68% male) acute infective endocarditis

Excluded if
  • no histological confirmation (n=28)



    • Non-independent ?blinded reference standard, applied in all patients from a consecutive appropriate spectrum.
    Reference standard:
    • abscess diagnosed by surgery (n=109) or autopsy (n=11)
    Diagnostic test: transthoracic and transoesophageal echocardiography- positive if definite regions of reduced echodensity and echolucent cavity were considered abscesses

    The evidence

    pre-test probability of abscess: 38%, (95% CI: 30% to 47%)

    diagnostic test perivalvular abscess abscess LR+
    (95% CI)    		 post-test probability LR-
    (95% CI)    		 post-test probability
    transoesophageal echocardiography positive (TEE) 40 4 16
    (6.2 to 42)     		91% 0.14
    (0.065 to 0.29)     		8%
    transthoracic echocardiography (TTE) 13 1 21
    (2.8 to 150)     		93% 0.73
    (0.61 to 0.87)     		31%
    total 46 74

    • There were 46 abscesses in 44 patients.
    • A total of 137 valves were infected by endocarditis- 59 aortic valves, 41 mitral valves, 3 tricuspid valves and 34 prosthetic valves.

    Comments

    1. All echocardiograms were evaluated by two independent physicians- the operator of TEE evaluated the TEE and TTE studies immediately after the procedure, and a second physician evaluated the videotape recordings before autopsy or surgery without the knowledge of the first evaluation. In cases of disagreement, the observers re-evaluated the studies together and consensus was reached.
    2. The surgeons were aware of the echocardiogram findings before the procedure. Uncertain whether this influenced the decision to perform surgery.

    Citation

    1. Daniel WG, Mugge A, Martin RP, et al: Improvement in the diagnosis of abscesses associated with endocarditis by transoesophageal echocardiography. New England Journal of Medicine 1991; 324: 795-800
    Search Terms: infective, endocarditis and echocardiography
    Contributor: Sumit Dhingra, Carl Heneghan and Chris Ball, July 2000
    Reviewer:

    Clinical Question.
    Patient infective endocarditis
    Intervention or Exposure transoesophageal echo, TEE, TOE
    Comparison transthoracic echo, TTE
    Outcome diagnosis of perivalvular abscess