COPD: can be diagnosed on a self-reported history, smoking history, age and maximum laryngeal height

Clinical bottom line (level 1b)

  1. Half of patients seeing a doctor have chronic obstructive airways disease.
  2. Chronic obstructive airways disease is more likely with
    • a self-reported history of chronic obstructive airways disease (LR + 7.3)
    • smoked > 40 pack-years (LR + 8.3)
    • aged 45 or more (LR + 1.3)
    • maximum laryngeal height 4 cm or less (LR + 2.8)
  3. Patients with all 4 features present have chronic obstructive airways disease (LR + 220) .
  4. Wheezing and laryngeal descent are not helpful at diagnosing or excluding chronic obstructive airways disease.
  5. Patients with none of these features are much less likely to have chronic obstructive airways disease (LR + 0.13) .
Straus et al: Journal of the American Medical Association 2000; 283 : 1853-1857
Expires April 2003

The study

Setting: 25 sites including primary care and hospitals, 14 countries

309 patients (aged mean 56, 57% male) with any of
  • known chronic obstructive airways disease (prior pulmonary function test results demonstrating FEV 1 < 5th percentile, FEV1-FVC less than 5th percentile, or FEV1-FVC ratio < 0.7; or patient self-report of a prior diagnosis of chronic obstructive airways disease, emphyesema, or chronic bronchitis; or patient taking inhaled bronchodilatros and/or inhaled steroids for long periods)
  • suspected of having chronic obstructive airways disease
  • neither known nor suspected of having chronic obstructive airways disease


Excluded if
  • purely reversible airways obstruction (i.e. asthma)
  • terminal illness whose goals of therapy were confined to comfort and dignity
  • aged < 18
  • respiratory distress so severe that bronchodilators could not be withheld safely until after spirometry
  • medically unstable from other causes (e.g. acute myocardial infarction, drug overdose)
  • unable to cooperate for clinical examination or spirometry
Independent blinded reference standard, applied in all patients from a consecutive appropriate spectrum.
Reference standard:
  • spirometry within 30 minutes of clinical examination: obstructive airways disease diagnosed if FEV1 and FEV1-FVC ratio less than fifth percentile
Diagnostic test: Clinical features
  • reported history of chronic obstructive airways disease
  • smoking history
  • laryngeal height (the distance between the top of the thyroid cartilage and the suprasternal notch)
  • larnygeal descent (the difference between the maximum (end of expiration) and minimum larnygeal height (end of inspiration)
  • wheezing

  • A multivariate analysis was used to adjust for related diagnostic elements. A model was created by selecting only adjusted LRs greater than 2.0 or less than 0.5.

The evidence

pre-test probability of all patients: chronic obstructive airways disease: 52%, (95% CI: 47% to 58%)
pre-test probability of patients without known chronic obstructive airways disease: 39%, (95% CI: 33% to 45%)

diagnostic test all patients: obstructive airways disease no obstructive airways disease LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
self-reported history of chronic OAD 7.3
(- to -) 		89% 0.5
(- to - ) 		35%
smoked > 40 pack-years 8.3
(- to - ) 		90% 46
(- to - )
aged 45 or more 1.3
(- to - ) 		58% 0.4
(- to - ) 		30%
maximum laryngeal height 4 cm or less 2.8
(- to - ) 		75% 0.8
(- to - ) 		46%
all 4 factors 220
(- to - ) 		100% 0.13
(- to - ) 		12%
total


diagnostic test patients with suspected chronic OAD: obstructive airways disease no obstructive airways disease LR+
(95% CI)		 post-test probability LR-
(95% CI)		 post-test probability
smoked > 40 pack-years 11.6
(- to - ) 		87% 0.90
(- to - ) 		35%
aged 45 or more 1.4
(- to - ) 		45% 0.50
(- to - ) 		23%
maximum laryngeal height 4 cm or less 3.6
(- to - ) 		68% 0.70
(- to - ) 		29%
all 3 factors 58.5
(- to - ) 		97% 0.32
(- to - ) 		16%
total

Comments

  1. 46 investigators participated in the study.
  2. Wheezing and laryngeal descent were not found to be helpful in diagnosing chronic obstructive airways disease.
  3. Insufficient original numbers were given in the article to recalculate likelihood ratios. No confidence intervals were given for multivariate likelihood ratios - the authors state this is due to the complexity of calculating confidence intervals from multivariate LRs.
  4. The model created is a clinical prediction rule and needs to be validated in an independent set of patients - the authors report a study is planned.

Citation

  1. Straus SE, McAlister FA, Sackett DL, et al: the accuracy of patient history, wheezing, and laryngeal measurements in diagnosing obstructive airway disease. Journal of the American Medical Association 2000; 283 : 1853-1857
Search Terms: hand-search
Contributor: Chris Ball, April 2000
Reviewer: Clare Wotton

Clinical Question.
Patient suspected COPD
Intervention or Exposure clinical features
Outcome COPD