Dyspnoea: clinical examination may help to determine cause.

Clinical bottom line (level 1a)

  1. In patients with dyspnoea, clinical examination can help determine the cause with an accuracy of between 66% and 92%.
Mulrow et al: Journal of General Internal Medicine 1993; 8: 383-392
Expires September 2003

The study

Systematic review of published studies of
  • Patients: dyspnoea
  • Intervention: utility of clinical examination
  • Outcome: diagnosis of cause
Articles found in ? using MEDLINE, 1966 to 1991 (search terms: dyspnea, differential diagnosis, decision making, diagnostic, heart disease, sensitivity and specificity, predictive value, and decision. Specific examination manoeuvers (Valsalva and hepato- or abdominojugular) were exploded ) and other articles were found from references in pertinent chapters of major textbooks, contacts with expert colleagues, and citations noted from the reference sections of relevant articles

Selection criteria: as above
Appraisal criteria: three independent reviewers identified relevant articles- detailed in text
Articles excluded if: dyspnea not presenting symptom, subject pool limited to a single diagnosis, bedside manoeuvres not utilised to determine the cause of dyspnoea, five or less patients involved

Five studies were included. Four studies evaluated the accuracy of the initial clinician's examination, one study assessed accuracy of history alone

The evidence


diagnostic test heart failure no heart failure LR
(95% CI)
history of dyspnoea with exertion 1.3
(1.0 to 1.6)
history of orthopnoea 2.0
(1.0 to 3.9)
history of nocturnal dyspnoea 1.9
(0.8 to 4.7)
history of peripheral oedema 1.6
(0.6 to 4.3)
total

  • Accuracies of the clinical examination ranged from 66% to 92%.
  • Clinical examination included: previous chronic obstructive lung disease, smoker, cough, previous asthma, wheezing, throat clearing, postnasal drip,, dyspnoea with exertion, orthopnoea, nocturnal dyspnoea, peripheral oedema and wheeze or crackles on physical examination.
  • Valsalva manoeuvre was evaluated in three studies, and likelihood ratios ranged from 2.1 to 8.8.
  • No likelihood ratios were available for other aspects of history or physical examination.

Comments

  1. The search strategy is restricted to Medline and chapters of "major" textbooks and is likely to have introduced some bias in the selection of papers reviewed.
  2. A number of clinical signs were included without the precision or reproducibility of these measurements being evaluated.

Citation

  1. Mulrow CD, Lucey CR, Farnett LE: Discriminating causes of dyspnea through clinical examination. Journal of General Internal Medicine 1993; 8: 383-392
Contributor: Clare Wotton and Musab Hayatli, May 2000
Reviewer: Ross Lawrenson

Clinical Question.
Patient dyspnoea
Intervention or Exposure clinical examination
Outcome cause