Infective endocarditis: microscopic haematuria was a helpful test.

Clinical bottom line (level 4)

  1. Infective endocarditis was common in patients that are suspected to have it.
  2. Microscopic haematuria did not diagnose or safely exclude infective endocarditis in high risk patients.
  3. Both the aortic and mitral valves were commonly affected.
Benn et al: Journal of Internal Medicine 1997; 242: 15-22
Expires July 2003

The study

Setting: county, Denmark, 1984 to 1993

106 patients (aged range 15 to 83 years; mean 55, 70% male) suspected infective endocarditis

Excluded if
  • autopsy performed with no signs of infective endocarditis
  • repeated negative blood cultures



    • Independent unblinded reference standard, applied in all patients from a consecutive ?appropriate spectrum.
    Reference standard:
    • infective endocarditis- positive if fulfilled by the diagnostic criteria by von Reyn
    Diagnostic test: microscopic haematuria (urine stix or microscopy)

    The evidence

    pre-test probability of infective endocarditis: 59%, (95% CI: 49% to 68%)

    differential diagnosis number of patients prevalence
    (95% CI)
    aortic valve 27 44%
    (31% to 56%)
    mitral valve 26 42%
    (30% to 54%)
    aortic and mitral valve 7 12%
    (3.4% to 19%)
    tricuspid 2 3.2%
    (0.0% to 7.6%)
    fever >37.5 ° C 58 94%
    (87% to 100%)
    tachycardia 32 52%
    (39% to 64%)
    new or changed murmur 29 47%
    (34% to 59%)
    breathlessness 27 44%
    (32% to 56%)
    cerebral emboli 11 18%
    (8.2% to 27%)
    petechia 5 8.1%
    (1.3% to 15%)
    splinters 5 8.1%
    (1.3% to 15%)


    diagnostic test infective endocarditis no infective endocarditis LR+
    (95% CI)    		 post-test probability LR-
    (95% CI)    		 post-test probability
    microscopic haematuria 44 7 4.5
    (2.2 to 9.0)     		86% 0.35
    (0.23 to 0.52)     		33%
    total 62 44

    Comments

    1. No information on the frequency of the clinical findings in the patients subsequently found not to have infective endocarditis. Other studies have shown that splinter haemorrhages are not a clinically useful sign.

    Citation

    1. Benn M, et al: Infective endocarditis, 1984 through 1993: a clinical and microbiological survey. Journal of Internal Medicine 1997; 242: 15-22
    Search Terms: explode infective and endocarditis in Medline
    Contributor: Carl Heneghan, Sumit Dhingra and Chris Ball, July 2000
    Reviewer:

    Clinical Question.
    Patient high risk patients
    Intervention or Exposure microscopic haematuria
    Outcome diagnosis of infective endocarditis