Methacholine Challenge

Methacholine Challenge* Comparison of Two Methods Donald W. Cockcroft, MD, FCCP; Beth E. Davis; David C. Todd, MD; and Audrey J. Smycniuk

Background: Guidelines for the 2-min tidal-breathing and the five-breath dosimeter methods for methacholine challenge have recently been published by the American Thoracic Society (ATS). Although subjects are exposed to twice as much aerosol at any given concentration during the tidal-breathing method compared to the dosimeter method, they were thought to give equivalent results. Objective: To compare the 2-min tidal-breathing and the five-breath dosimeter methacholine challenges. Setting: Tertiary care university-based bronchoprovocation laboratory. Patients: Forty subjects with currently symptomatic asthma. Interventions: The two methacholine tests were done in random order on separate days at the same time of day at 1- to 7-day intervals. Results: The dosimeter provocation concentration of methacholine causing a 20% fall in FEV1 (PC20) was almost twice that of the tidal-breathing PC20: 2.4 mg/mL vs 1.3 mg/mL (paired t test, p < 0.00005). The difference was greater in those with mild airway hyperresponsiveness (AHR) [PC20 > 1.0 mg/mL; 3.2-fold] compared to those with moderate AHR (PC20 < 1.0 mg/mL; 1.6-fold) [p ⴝ 0.04]. Three subjects with mild asthma and mild AHR (tidal-breathing PC20, 1.9 to 4.3 mg/mL) had a nonmeasurable PC20 (> 32 mg/mL) with the dosimeter. Conclusions: The tidal-breathing method, which exposes the subject to twice as much aerosol at each concentration, produced approximately twice the response. The total lung capacity maneuvers with breathhold during the dosimeter method may inhibit the response in some patients with asthma. (CHEST 2005; 127:839 – 844) Key words: airway responsiveness; bronchoprovocation; methacholine; methods Abbreviations: AHR ⫽ airway hyperresponsiveness; AMMD ⫽ aerodynamic mass median diameter; ATS ⫽ American Thoracic Society; PC20 ⫽ provocative concentration of methacholine causing a 20% fall in FEV1; TLC ⫽ total lung capacity

challenge is a widely performed M ethacholine test for both diagnostic and research purposes. Recently, the American Thoracic Society (ATS) published guidelines1 for two of the more commonly used methods: the 2-min tidal breathing method,2 and the five-breath dosimeter method.3 At any given *From the Division of Respiratory Medicine, Department of Medicine, Royal University Hospital, University of Saskatchewan, Saskatoon, SK, Canada. Dr. Cockcroft is the Ferguson Professor of Respiratory Medicine. Supported by a grant from Methapharm Inc. Supported by the Lung Association of Saskatchewan. Manuscript received May 14, 2004; revision accepted August 6, 2004. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (e-mail: [email protected]). Correspondence to: Donald W. Cockcroft, MD, FCCP, Division of Respiratory Medicine, Royal University Hospital, Ellis Hall, Fifth Floor, Saskatoon, SK, S7N 0W8 Canada; e-mail: [email protected] www.chestjournal.org

concentration, subjects are exposed to approximately twice as much aerosol with the tidal-breathing method as compared to the dosimeter method. However, equivalent results have been demonstrated in two small studies, an older study4 using histamine, and a more recent study5 in which the dosing regimens for the two methods were not identical. We compared the two methods described in the ATS guidelines in a larger cohort of subjects using identical regimens.

Materials and Methods Subjects Subjects with asthma6 were recruited from the respiratory clinic at the Royal University Hospital and from advertisements. Inclusion criteria included an FEV1 ⱖ 65% predicted and a tidal-breathing provocative concentration of methacholine causCHEST / 127 / 3 / MARCH, 2005

839

ing a 20% fall in FEV1 (PC20) ⬍ 16 mg/mL. The study was approved by the University of Saskatchewan Ethics Committee, and signed informed consent was obtained. Study Design Subjects (n ⫽ 40) attended the laboratory on 2 days at the same time of day within 1 week. Inhaled albuterol (n ⫽ 39) or terbutaline (n ⫽ 1) was withheld for 6 h, and long-acting ␤2agonists (n ⫽ 3) for 36 h and inhaled corticosteroids (n ⫽ 9) were continued in the same dosage; no other asthma medications were used. Subjects had no relevant allergen exposure, respiratory tract infection, or change in asthma maintenance therapy for at least 4 weeks. Two methacholine challenges, one using each method, were performed in random order. Tidal-Breathing Method The 2-min tidal-breathing method was performed as outlined1,2,7; Bennett Twin jet nebulizers (Puritan-Bennett Corporation; Carlsbad, CA) were calibrated with 3 mL of saline solution by weighing before and after a 2-min period of nebulization to deliver a mass loss of 0.13 g/min approximating an output of 0.13 mL/min. Aerosols were directed to the subject via a loose-fitting facemask. Subjects wore a nose clip and inhaled each aerosol for 2 min of tidal breathing. Sterile isotonic saline solution was initially inhaled followed by doubling concentrations starting as outlined in previous publications.7,8 The available concentrations ran from 0.03 to 16 mg/mL. Full spirograms were initially done in triplicate. After the completion of each inhalation, single technically acceptable truncated spirograms were performed to obtain an FEV1 at both 30 s and 90 s. The dose step-up timing was kept constant with a 5-min time interval between the commencement of one concentration and the commencement of the next. The percentage fall in FEV1 was calculated from the lowest post-saline solution FEV1 to the lowest post-methacholine FEV1.2,7,9 The test was continued until the FEV1 had fallen by 20%, and the PC20 was calculated.10 Dosimeter Method The dosimeter method1,3 was performed using DeVilbiss 646 nebulizers (Sunrise Medical HHG; Somerset, PA) and a KoKo dosimeter (Pulmonary Data Services; Doylestown, PA). The DeVilbiss 646 nebulizers were calibrated to produce a mass loss of 9 mg per actuation, approximately equivalent to 9 ␮L per breath. Triplicate calibrations were done with 3 mL in the nebulizer by having the calibrator take 10 slow-breath activations from functional residual capacity to total lung capacity (TLC) followed by a 5-s breathhold and exhalation to the room. The five-breath dosimeter methacholine challenge was done using the identical methacholine solutions (0.03 to 16 mg/mL), starting concentration, FEV1 timing, dose step-up timing, PC20 calculation, etc., as the tidal-breathing method. Only the inhalation method was different. The method of inhalation was five slow inspiratory capacity (functional residual capacity to TLC) inhalations followed by a 5-s breathhold at TLC with exhalation to the room following each inhalation. Analysis The primary end point, the PC20, was log transformed and analyzed by paired t test using a computerized statistical program.11 840

Results Forty subjects meeting the inclusion criteria of an FEV1 ⱖ 65% and a tidal-breathing PC20 ⬍ 16 mg/mL completed the investigation without adverse events. A forty-first subject failed to return for the second methacholine challenge and was not included in the analysis. Demographics of the 40 subjects who completed both challenges are given in Table 1. Individual data points for the two methacholine challenges are compared in Figure 1. Three subjects with tidal-breathing PC20 values ranging from 1.9 to 4.3 mg/mL had a ⬍ 10% FEV1 fall at 16 mg/mL with the dosimeter method; this was extrapolated to a PC20 ⬎ 32 mg/mL, a conservative estimate that was used for the statistical analysis (ie, PC20 ⫽ 32). The dosimeter PC20 was almost twofold greater than the tidal-breathing PC20, with geometric mean values of 2.4 mg/mL (95% confidence interval, 1.5 to 3.7 mg/mL) and 1.3 mg/mL (95% confidence interval, 0.93 to 1.9 mg/mL), respectively (p ⬍ 0.00005). When the three subjects with extrapolated high PC20 values were omitted from the analysis, the geometric mean dosimeter PC20 remained 1.6-fold larger than the tidal-breathing PC20 (p ⬍ 0.00005). We made the post hoc observation that subjects with higher PC20 values appeared to have a greater difference between the two methods. Accordingly, the population was arbitrarily dichotomized into the 25 subjects with a tidal-breathing PC20 ⬎ 1 mg/mL (mild-to-borderline AHR) and 15 subjects with a tidal-breathing PC20 ⬍ 1 mg/mL (moderate-tomarked AHR). We compared the ratio of the dosimeter PC20 to the tidal-breathing PC20 and found a ratio of 3.2 for those with PC20 values ⬎ 1 mg/mL, and 1.6 mg/mL for those with PC20 ⬍ 1 mg/mL (unpaired t test, p ⫽ 0.04). When this analysis was repeated omitting the three subjects with extrapolated PC20 values, the ratios were 2.1 and 1.6, respectively (p ⫽ 0.19). Two of the three subjects with PC20 values extrapolated to ⬎ 32 mg/mL returned to the laboratory for repeat dosimeter PC20 determinations using subTLC inhalations and breathhold. These two subjects both had measurable dosimeter PC20 values with the sub-TLC inhalation method (12.0 mg/mL and 4.2 mg/mL). Discussion We have compared the two standardized methacholine challenges with the only differences in protocol being the method of aerosol generation and the method of inhalation. In the entire group, the geometric mean dosimeter PC20 was almost twice that of the tidal-breathing PC20. The discrepancy was Clinical Investigations

Table 1—Demographic Characteristics of the 40 Subjects Who Completed Both Challenges* Subjects

Gender

Age, yr

Height, Inches

FEV1, L

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Mean ⫾ SD

Male Male Female Female Male Female Male Female Female Male Female Female Male Male Female Female Male Female Male Female Female Female Male Female Female Male Male Male Female Female Female Male Male Female Female Female Female Female Female Female

57 29 38 23 26 26 24 19 26 21 25 24 25 27 20 42 21 21 26 24 22 21 49 21 19 25 24 46 23 64 23 32 31 19 44 39 22 28 22 21 28 ⫾ 11

66 70 66 66 71 66 72 65 63 73 66 63 73 72 68 64 71 61 74 62 62 69 70 65 66 67 70 67 66 62 63 74 71 68 65 65 67 64 68 64 67 ⫾ 4

2.42 3.80 3.04 2.98 3.79 3.39 4.48 2.98 3.58 3.76 3.41 3.09 3.64 4.36 4.14 2.99 4.83 2.80 3.81 2.69 3.36 3.54 2.59 3.44 3.88 4.00 3.50 3.10 3.20 2.45 2.75 4.30 3.25 3.00 1.95 1.93 3.37 2.99 3.77 2.95 3.33 ⫾ 0.65

% Predicted 73 85 96 84 82 97 93 83 110 76 97 94 75 92 108 103 101 88 77 84 103 91 65 97 106 94 76 83 90 113 83 89 72 78 66 67 92 91 101 86 89 ⫾ 12

Concomitant Medications A A A A A A A A A A A A A A A A A A A A A; B, 200 ␮g bid A; F, 125 ␮g bid A A A A; F/S, 250 ␮g qd A A A; F, 250 ␮g qd A; B, 200 ␮g bid T; F, 125 ␮g bid A; F, 125 ␮g bid A A; B/O, 100 ␮g bid A A; F/S, 500 bid A A A A; F, 250 ␮g bid

*A ⫽ albuterol (as needed); B ⫽ budesonide; F ⫽ fluticasone; S ⫽ salmeterol; T ⫽ terbutaline (as needed); O ⫽ formoterol.

greater (2.1-fold) in those with tidal-breathing PC20 ⬎ 1.0 mg/mL and less (1.6-fold) in those with tidal-breathing PC20 ⬍ 1 mg/mL. Adding to this discrepancy were an additional three subjects with tidal-breathing PC20 values approximately 2.0 to 4.0 mg/mL and dosimeter PC20 values unmeasurable at ⬎ 32 mg/mL. There are many potential reasons for these observations and some possibilities are outlined below. The dose of aerosol delivered to subjects would be the most obvious and likely most important difference between the two methods. The tidal-breathing method produces an aerosol of 0.13 mL/min for 2 min, giving a total output of 0.26 mL or 260 ␮L. www.chestjournal.org

When this is multiplied by the duty cycle (total duration of respiratory cycle/total breathing cycle time, approximately 0.35), the subject will be exposed to approximately 90 ␮L at each concentration. This compares to 45 ␮L (five breaths ⫻ 9 ␮L per breath) at each concentration for the dosimeter method. The approximate double dose administered by the tidal-breathing method would account for the PC20 being approximately one half that of the dosimeter method. The deposition and retention of aerosol are likely different between methods. It was empirically assumed1 that the slow inspiratory capacity breath and breathhold might result in a superior dose and CHEST / 127 / 3 / MARCH, 2005

841

Figure 1. Comparison of tidal-breathing PC20 and dosimeter PC20 values in study subjects. Results are plotted on a log scale. The three subjects with unmeasurable PC20 by dosimeter, extrapolated to ⬎ 32 mg/mL, are shown in dashed lines (p ⬍ 0.00005; paired t test).

aerosol retention, which might explain the apparent PC20 equivalence in the earlier histamine provocative concentration comparison between two methods.4 In that study,4 the two methods produced equivalent lung deposition of radiolabeled aerosol; however, the method of dosimeter calibration (see below) was not specified. The mean data suggest that greater aerosol retention is not a factor in our study. However, subjects with moderate-to-marked AHR showed less than the expected 2:1 ratio in provocative concentration values, suggesting greater dosimeter aerosol dose/deposition may have played a role in this group of subjects. A third consideration in the differences of the method is the effect of maximal inspiration on the measurement of airway responsiveness. Inhalations to TLC are recognized to be bronchodilator and bronchoprotector, particularly in subjects with airway responsiveness in the normal, borderline, or mild range.12 This is important enough so that delaying the time to the first spirogram (ie, until the 842

first inhalation to TLC) following methacholine inhalation to 3 min as opposed to 30 s results in a significantly lower PC20 values across all ranges of airway responsiveness.13 We kept the timing and the number of spirograms constant so this was not a factor in any difference between the two methods. However, the dosimeter method involves five inhalations to TLC with 5-s breathholds. Consequently, the subjects spend a total of 25 to 30 s breathholding at TLC in the 1.0 to 1.5 min before the first spirogram. The bronchodilator/bronchoprotective effect of this maneuver is the likely explanation for the greater discrepancy between the two methods seen in subjects with mild AHR. It also would explain the three subjects with mild-to-borderline AHR on the tidal-breathing method whose PC20 values were unmeasurable with the dosimeter method. Indeed, this was verified in two of these subjects by repeating the dosimeter PC20 with subTLC inhalations and obtaining measurable PC20 values. Further investigation into this observation is warranted. A fourth concern regarding comparative studies looking at the two methods is the mechanism of dosimeter nebulizer calibration. We have demonstrated large variations in apparent measured nebulizer output with dosimeter-nebulizer systems based on how the calibration is performed. As noted above, we routinely calibrate our dosimeter nebulizer systems by weighing the nebulizer before and after breath activation followed by exhalation into the room. When the nebulizers are operated at these running parameters and the nebulizer recalibrated by having the subject breath activate the nebulizer and exhale back into the nebulizer, the measured output (mass loss) is less by about a factor of one third, likely because of condensation from the exhaled breath within the nebulizer. The result is an underestimate of the mass loss.14 When the nebulizer is activated by manual activation (activation via a switch with no subject inhaling), the measured output, when operated at these parameters, is less by a factor of two thirds, almost certainly because the aerosol does not completely clear the dead space of the nebulizer and remains within. Conversely, if the nebulizer dosimeter operating parameters are adjusted to obtain a measured output of 9 ␮L per breath using the latter two methods, the operating characteristics have to be increased such that the true output is greater. It is possible that differences in nebulizer calibration in different studies could explain in whole or in part different comparative results. The two previous methods comparison investigations have suggested similar PC20 values by the two tests. Ryan et al4 examined histamine PC20 using Clinical Investigations

2-min tidal breathing method with a Wright nebulizer at a slightly higher output than usual (0.147 mL/min) and a very small particle size: aerodynamic mass median diameter of (AMMD) 0.87 ␮m. This was compared to a dosimeter method using a nebulizer (model 646; DeVilbiss; Minneapolis, MN) calibrated (means unspecified) to an output of 9 ␮L per breath and operated by a Rosenthal-French dosimeter. There were 10 subjects in this study. Geometric mean PC20 was 1.8 mg/mL with tidal breathing and 2.2 mg/mL with the dosimeter method. This study may not be comparable to ours for at least two reasons. The first is the method of dosimeter calibration was not specified, and this could have influenced the study. The second is that the very small Wright nebulizer particle size may be the reason for reduced effect produced by some (particularly Canadian-made) Wright nebulizers15; the small particle size aerosol may behave in part like a gas with a greater proportion of aerosol being exhaled. The reduced response produced by the Canadian Wright nebulizer is the major reason that we switched to the Bennett Twin jet nebulizer, which has a more appropriate particle size with the AMMD between 3 ␮m and 4 ␮m.16 The other methods comparison study5 compared the tidal-breathing method using doubling-dose steps of methacholine and a Canadian Wright nebulizer, and the dosimeter method using quadrupling-dose steps (a recommended ATS option) and the DeVilbiss 646 nebulizer. Eleven of the 12 subjects were evaluable with a geometric mean PC20 of 4.7 mg/mL for tidal breathing and 3.9 mg/mL for dosimeter. The different dosages make comparison difficult; however, since there is a significant cumulative effect of methacholine when administered at 5-min intervals,13 the cumulative effect should be smaller with quadrupling doses. If so, this should contribute to the dosimeter PC20 being higher (rather than lower) than with doubling doses; therefore, this is not a plausible explanation for why their study differs from ours. Their tidal-breathing method used a Canadian Wright nebulizer, some of which have been previously demonstrated to produce a markedly lower response than the English Wright nebulizer15 probably because of the very small particle size (AMMD ⬍ 1 ␮m). Thus, some Canadian Wright nebulizers may underestimate AHR. The particle size of the nebulizer used in the study by Wubbel et al5 is not specified. Likewise, the means of calibration of the dosimeter nebulizer system discussed above (DeVilbiss) was also not specified. Several other studies17–20 cannot be compared to ours because of methodologic differences or lack of documentation of nebulizer output. These results indicate that when the tests are done as outlined by the ATS, different cut points should www.chestjournal.org

be used for the different methods. The ATS has suggested 16 mg/mL as the cut point between normal and borderline, 4 mg/mL as the cut point between borderline and mild, and 1 mg/mL as the cut point between mild and moderate to marked.1 These are based on the tidal-breathing method.7 Based on the current study, the cut points should be increased to 32 mg/mL, 8 mg/mL, and 2 mg/mL, respectively, for the dosimeter method. Both methods were well tolerated. Subjects indicated a slight preference (not significant) in favor of the tidal-breathing method, which was perceived as easier; several subjects expressed no preference. From a scientific point of view, it has been believed that there was generally very little to choose between the two methods when properly standardized.1,3,4,7 However, our observation of several subjects with clinical asthma, mild AHR by the tidal-breathing method, and no measurable airway hyperresponsiveness by the dosimeter method would suggest that the bronchodilator/bronchoprotective effect of the dosimeter method itself may, in fact, reduce the sensitivity of the test and lead to some false-negative test results in subjects at the mild end of the AHR spectrum. This is the area where the majority of positive test results are likely to occur in clinical practice when testing subjects with symptoms and normal spirometry. ACKNOWLEDGMENT: The authors thank Jacquie Bramley for assisting in the preparation of this article.

References 1 Crapo RO, Casaburi R, Coates AL, et al. Guidelines for methacholine and exercise challenge testing-1999. Am J Respir Crit Care Med 2000; 161:309 –329 2 Cockcroft DW, Killian DN, Mellon JJA, et al. Bronchial reactivity to inhaled histamine: a method and clinical survey. Clin Allergy 1977; 7:235–243 3 Fabbri LM, Mapp CE, Hendrick DJ. Standardization of the dosimeter method for measurement of airway responsiveness in man. In: Hargreave FE, Woolcock AJ, eds. Airway responsiveness: measurement and interpretation, Mississauga, ON, Canada: Astra Pharmaceuticals, 1985; 29 –34 4 Ryan G, Dolovich MB, Roberts RS, et al. Standardization of inhalation provocation tests: two techniques of aerosol generation and inhalation compared. Am Rev Respir Dis 1981; 123:195–199 5 Wubbel C, Asmus MJ, Chesrown SE, et al. Comparison of the two ATS-recommended methods of methacholine challenge testing. Chest 2004; 125:453– 458 6 American Thoracic Society. Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease (COPD) and asthma. Am Rev Respir Dis 1987; 136:225–244 7 Juniper EF, Cockcroft DW, Hargreave FE. Histamine and methacholine inhalation tests: tidal breathing method; laboratory procedure and standardisation. 2nd ed. Lund, Sweden: AB Draco, 1994 8 Cockcroft DW, Marciniuk DD, Hurst TS, et al. Methacholine challenge: test-shortening procedures. Chest 2001; 120:1857– 1860 CHEST / 127 / 3 / MARCH, 2005

843

9 Davis BE, Cockcroft DW. Calculation of provocative concentration causing a 20% fall in FEV1: comparison of lowest vs highest post-challenge FEV1. Chest 2000; 117:881– 883 10 Cockcroft DW, Murdock KY, Mink JT. Determination of histamine PC20: comparison of linear and logarithmic interpolation. Chest 1983; 84:505–506 11 STATISTIX for Windows. Tallahassee, FL: Analytical Software, 1998 12 Kapsali T, Permutt S, Laube B, et al. Potent bronchoprotective effect of deep inspiration and its absence in asthma. J Appl Physiol 2000; 89:711–720 13 Simard B, Cockcroft DW, Davis BE, et al. Comparative influence of deep inspiration (D1) on methacholine (M) responsiveness (MR) in normal and asthmatic subjects: influence of variable time-intervals. Am J Respir Crit Care Med 2003; 167:A182 14 Todd DC, Davis BE, Smycniuk AJ, et al. Importance of method of dosimeter calibration on nebulizer output. Ann Allergy Asthma Immunol 2005; 94:45– 47 15 Juniper EF, Hargreave FE. Airway responsiveness assessed by aerosol inhalation tests: variability in results due to unex-

844

16

17

18

19

20

pected differences between calibrated nebulizers [abstract]. J Allergy Clin Immunol 1986; 77:171 Ryan G, Dolovich MB, Obminski G, et al. Standardization of inhalation provocation tests: influence of nebulizer output, particle size and method of inhalation. J Allergy Clin Immunol 1981; 67:156 –161 Beaupre A, Malo JL. Comparison of histamine bronchial challenges with the Wright nebulizer and the dosimeter. Clin Allergy 1979; 9:575–583 Asher MI, Bertrand JM, Beaudry PH, et al. Histamine inhalation challenge in children: a comparison of two methods. Ann Allergy 1983; 50:389 –392 Bennett JB, Davies RJ. A comparison of histamine and methacholine bronchial challenges using the DeVilbiss 646 nebulizer and the Rosenthal-French dosimeter. Br J Dis Chest 1987; 81:252–259 Siersted HC, Walker CM, O’Shaughnessy AD, et al. Comparison of two standardized methods of methacholine inhalation challenge in young adults. Eur Respir J 2000; 15:181–184

Clinical Investigations