Bronchial Provocation Testing: The Future

Sandra D. Anderson; John D. Brannan


Curr Opin Allergy Clin Immunol. 2011;11(1):46-52. 

In This Article

Clinical Trials and Bronchial Provocation Tests

One study[9••] raises some important questions on BPTs and their role in excluding a diagnosis of asthma or identifying EIB. This was a phase III trial in 375 patients not taking ICS (mean age 24.3 years ± 10.2 FEV1% predicted 93.6 ± 10) with mild symptoms of asthma but without a definite diagnosis, and with EIB defined as a 10% fall FEV1 on at least one of two exercise tests. The study revealed some findings that may change our approach to ruling out asthma and choosing patients for clinical trials. The participants performed two standardized 8-min exercise tests – a methacholine challenge, and a mannitol challenge – over a period of 2.6 ± 3.2 days. The results of those with a positive test are summarized in Table 2. The asthma diagnosis was made by a respiratory physician on the basis of history, questionnaire, spirometry, response to bronchodilator, skin-prick tests and the two exercise tests, but the physician was blinded to the mannitol and methacholine test results. First, methacholine (41.6%) was no more sensitive than mannitol (44.8%) to detect BHR in this population, with a concordance of 69% for the methacholine and mannitol test results. Second, of the 240 participants assigned a diagnosis of asthma by a respiratory physician at the end of the study, 118 or 49% were negative to methacholine at 16 mg/ml. Third, a negative methacholine test was common and occurred in 45% of patients with EIB. Of participants with a negative methacholine, 19.7% had a fall in FEV1 after exercise at least 20% of baseline (Table 3).[9••]

The high frequency of negative methacholine test results in this large and well examined group suggests that a methacholine BPT should no longer be relied on as a rule-out test for asthma. This study is important as the participants in this trial, without a definite diagnosis, with a normal FEV1, are the very ones likely to be referred for BPT to confirm or exclude asthma.

The reason for these findings is not clear, but it does raise the question as to whether airway calibre has played an important role historically in giving methacholine and histamine a reputation for being highly sensitive to detect BHR. A low FEV1 has long been recognized as a risk factor for BHR to methacholine.[48] What has not been generally recognized is that these pharmacological stimuli may not be as sensitive to detect BHR in those presenting with a normal or high FEV1, particularly early in the development of asthma. In the previous studies reporting unexpected negative responses to pharmacological agents in school children and athletes, the FEV1 was normal to high.[7,49] Some of the discordant findings between exercise and methacholine test results may also be explained by differences in potency of the mediators of bronchoconstriction. For example, PGD2 and LTC4 released with exercise and mannitol, either alone or together, are very much more potent than histamine or methacholine alone.[50]

Another important issue arising from this phase III trial concerns the responses to mannitol in relation to the response to exercise.[9••] As may be expected, in a group of patients without a definite diagnosis of asthma, the EIB recorded in 43.5% of patients was mild, with a median value for percentage fall in FEV1 of 15.5% (Table 2). What was unexpected was that the sensitivity of mannitol to detect EIB, defined as a 10% fall in FEV1 on at least one of two tests, was only 59.8% (Table 3). The reason for this low sensitivity is unclear but may relate to the different mechanisms whereby exercise provokes airway narrowing,[13] or to the site of action of the mannitol.[27•] Mannitol was, however, more useful than one exercise test in identifying BHR. Thus the frequency of a positive mannitol test was 1.41 times that of a positive exercise test (% fall in FEV1 ≥10%) on the first test, and the frequency of an at least 15% fall in FEV1 after mannitol was 1.65 times greater than at least 15% fall on exercise. The relationship between the airway response to mannitol and exercise, however, was not as strong as that documented in established asthmatic patients.[51] Further, as with methacholine, mannitol did not identify 21.4% who had more than 20% fall in FEV1 after at least one of two exercise tests and this too was unexpected (Table 3). These data inform us that in those with mild symptoms of asthma not taking medications, bronchial responsiveness can vary over a short period. This was demonstrated by variability in presence and severity of EIB when two tests were performed over a few days.[52••] This variability could not be accounted for by differences in the intensity of exercise, baseline FEV1, exposure to aeroallergens to which the person was allergic, or to the condition of the inspired air during exercise.[52••]

In a recent study[24••] in children clinically stable with a history of EIB, 33 exercised on a treadmill and on a separate day they performed a mannitol test with 25 children completing both tests. Of the 25, 10 had EIB, defined as at least 15% fall in FEV1 after exercise, and 13 had a positive mannitol test with a 15% fall (PD15 84 mg, CI 26–266), with nine children being positive to both tests. Seven of the children negative to mannitol were taking ICS and seven of the children negative to exercise. Six of the 10 children had more than 30% fall in FEV1 after exercise, whereas no child has this degree of reduction in FEV1 after mannitol. The positive predictive value of the mannitol for EIB was 69% and the negative predictive value 91%. There was a significant relationship between reactivity to mannitol, expressed as log RDR, and the percentage fall in FEV1 after exercise (r p = 0.66, P < 0.001) for the whole group.

Cowan et al.[41] reported responses to mannitol, exercise and FeNO in 39 patients with exercise-induced wheeze. Mannitol was also more sensitive in identifying BHR in this group than exercise in that 92% were positive to mannitol but only 36% to bicycle exercise. All of those with high FeNO (66.0 ppb) and 84% of those with low FeNO (17.8 ppb) were positive to mannitol. Further, there was no significant difference in the PD15 to mannitol between the low FeNO (119 mg, 64–222) and high FeNO [PD15 143 mg (95–215)] group. Only those with high FeNO were given 4 weeks of fluticasone (250 mcg twice daily) and 75% of these patients had a significant reduction (≥1 doubling dose) in BHR to mannitol.[41] The results suggest that in future, those with a PD15 but a low FeNO should not be excluded from treatment with inhaled steroids.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: