Comparison of Responses to Methacholine and Cold Air in Patients Suspected of Having Asthma: Conclusion
We therefore believe that in patients thought to have asthma, although responses to cold air and methacholine are related, the relationship is by no means perfect, response to one agent explaining something on the order of 25 percent of the variation in response to the other. These results are not entirely unprecedented. In their studies of established asthmatic patients, both Weiss et al and Tessier et al found correlations between methacholine and cold-air challenges that were only slightly better than ours, although their cold-air challenges used the more elaborate dose-response technique, and their results were plotted in a log-log format. Response to one agent is not equivalent to response to the other. This is particularly evident when one considers that in our series 17 patients who had positive methacholine responses had negative cold-air testing, and 12 who had PC20 >8 mg/ml had positive cold-air tests. there
While it could be argued that in the former group cold-air sensitivity was missed because these patients did not achieve adequate levels of ventilation during the cold-air test, the ventilation that they attained did not differ from those who had positive cold-air tests (Table 2). Among patients with PC^ <8 mg/ml, we observed negative cold-air responses in one subject who achieved a ventilation of 118 L/min during the test and another who exceeded her predicted MBC. Inadequacy of the ventilatory challenge, on the other hand, may have resulted in underestimation of meth-acholine-negative, cold-air-positive patients, one of whom showed a 36 percent decline in FEVj after a test in which the ventilation was 40 L/min.
Patients who had positive cold-air responses and negative methacholine responses did not in general show borderline methacholine responses. Only 4/12 of these patients had PC20 of <16 mg/ml, and only one of these had a cold-air response of greater than 20 percent of the initial FEVx; the other three patients with such large cold-air responses had PCao ^16 mg/ml.
It follows that if these tests are used to diagnose asthma in patients suspected of having the disease, the populations identified will differ according to which test is used. In our hands, there was little difference in test sensitivity; each missed a similar number of patients who had positive responses to the other. We conclude that neither test constitutes a “gold standard” for the diagnosis of asthma.