Specific conductance criteria for a positive methacholine challenge test: are the American Thoracic Society guidelines rather generous?

BACKGROUND: American Thoracic Society (ATS) guidelines for methacholine challenge testing (MCT) discuss specific airways conductance (sG(aw)) as a surrogate marker for forced expiratory volume in the first second (FEV(1)) to diagnose airways obstruction. The guidelines suggest a cutoff value of 45% drop in sG(aw) to diagnose a positive MCT. However, there is no available evidence that supports this cutoff value of 45%. We conducted this study to examine the relationship between FEV(1) and sG(aw) during MCT.

METHODS: One-hundred thirty-eight patients who had both sG(aw) and FEV(1) measured during MCT between April 2003 and March 2004 were retrospectively evaluated. The tests were done according to the ATS guidelines. Data were first analyzed using linear regression modeling, comparing the change in FEV(1) to changes in sG(aw). Then the sensitivity and specificity were generated for different cut points, using receiver operating characteristic analysis.

RESULTS: Thirty-eight patients had a positive MCT based on ATS FEV(1) criteria. A decrease of 20% in FEV(1) correlated with a drop of 56% in sG(aw) (95% confidence interval 52% to 60%, r(2) 0.35, P < .001). Using 20% decline from baseline in FEV(1) at different PC(20) (provocational concentration that produced a > or = 20% FEV(1) decrease) values (4 mg/mL, 8 mg/mL, and 16 mg/mL), we then analyzed the sensitivity, specificity, positive predictive value, and negative predictive value of the 45% decline in sG(aw) and compared it with a 56% decline in sG(aw). Using receiver operating characteristic analysis, we were able to find that a cutoff of 51-52% performed better than either of the 2 values.

CONCLUSIONS: Our study suggests that the ATS suggested cutoff value of 45% decline in sG(aw) to diagnose a positive MCT may be rather generous, and a decline of 51% from baseline may provide a more accurate measure of airway hyper-responsiveness. Further studies using well defined subjects with and without asthma should be done to better assess the test characteristics of sG(aw).