Ventilation-Perfusion Mismatch after Methacholine Challenge in Patients with Mild Bronchial Asthma 1- 4
ROBERT RODRIGUEZ-RaiSIN, ANTONI FERRER,! DANIEL NAVAJAS, ALVAR G. N. AGUSTI, PETER D. WAGNER, and JOSEP ROCA
Introduction
Several experimental (1) and clinical studies (2-6) have demonstrated dissociation between maximal expiratory airflow rates and pulmonary gas exchange in both acute and chronic forms of bronchial asthma. Accordingly, it has been suggested that airflow rates are determined mostly by properties of the larger airways, whereas gas exchange is more related to peripheral airway obstruction (4). However, even when measured, arterial blood gases do not suffice in the analysis of the functional changes associated with bronchial asthma because of simultaneous changes in extrapulmonary determinants of arterial Pao, such as minute ventilation, cardiac output, and oxygenuptake (3-6). Little information concerning VAlQ inequality in bronchoprovocation challenge tests is available. Although it is known that these tests induce transient arterial hypoxemia or oxygen desaturation and/or abnormal lung scans (7-15), VA/Q inequality has only been investigateddirectly in exerciseinduced asthma in both adults (16) and children (17) and also after histamine in children only (18). By contrast, the pulmonary gas exchange response to methacholine (MTH) inhalation challenge, by far one of the most widely used tools for the diagnosis of asthma in both the clinicaland the epidemiologic setting, remains poorly documented to date. The aim of the present study was to investigate the pattern and time course of VA/Q inequality after challenge to better define the model of MTH bronchial challenge in patients with asymptomatic mild asthma. An additional question was whether an inhaled specific beta-adrenergic agent such as salbutamol may alter VAlQ mismatching induced by the MTH challenge test. Methods
Patients Sixteen patients (eight male, eight female) 16 88
SUMMARY ToInvestigate the effects of methacholine (MTH)ch.llenge on spirometry, lung mech.ndistributions, 18subjects 18to 58 yr of .ge Ics, respl...tory g...., .nd ventilation-perfusion (V with stable mild .sthma (FEV1 , 92 ± 5% [SEM] predicted; FEFu - 15 , 71 ± 7% predicted; respiratory system resistance (R...).t 4 Hz, 4.8 ± 0.4 cm H20/L- 1 s; Pa02' 88 ± 3 mm Hg; A.P0 2, 23 ± 3 mm distributions were unlmod.1 .nd rel.tlvely n.rrow In 12 petlents Hg) were recruited. sasellne V and modestly blmod.1 In the other four. The dlsperalon of pulmon.ry blood flow (log SO Q) was slightly enl.rged (0.71 ± 0.09) .nd th.t of ventll.tlon (log SO V) was norm.I (0.57 ± 0.04) (norm.' range, 0.3 to 0.8); an Index of overall V heterogeneity (OISPR-E·) was .Iao mildly abnorm.1 (5.3 ± 0.8) (norm.1 values < 3.0). After MTH ch.llenge, FEVh FEF25 - 15 , .nd Pa02 fell (to 82 ± 3.nd 35 ± 3% predicted, .nd to 71 ± 1 mm Hg, respectively), where.s R... (p < 0.001 each), minute ventll.tlon (p < 0.02), he.rt r.te (p < 0.01),.nd A.P0 2Incre.sed (p < 0.001).V rel.tlonshlps mildly to moder.tely wo....ned (log SO Q Incre.sed to 0.98 ± 0.04 [p < 0.01], log SO V to 0.79 ± 0.04, and OISP R-E· to 9.8 ± 0.8 [p < 0.001 each)). Qu.lltatlvely, the pattern of blood flow dlatrlbutlon was broadly unlmod.lln 13petlents .nd modestly blmod.lln three, of whom only one had • blmod.1 be..llne distribution. Pre-MTHto poat-MTH ch.llenge changes In R... were correlated to the ch.nges In both log SO V (r, 0.85) .nd DISP R-E· (r, 0.83) (p < 0.01 e.ch). After these me.surements (15 min .fter MTH), eight patients received placebo .nd eight received aalbutamol by InlMllatlon In ...ndomlzed double-blind fIIshlon. By 30 min .fter challenge, patients given placebo showed Improved maxlm.1 expl...tory ...tes, where.s R....nd gu exchange remained e_ntl.Uy unchanged from the 5-mln post-MTH levels. Salbutamol significantly Improved not only .Irflow ...tes but alao R... and all the Indices of pulmonary g.s exch.nge. Our results Indicate that .fter MTH challenge there Is dlaaocl.tlon between spirometry .nd gas exch.nge, Just .s Is often _n In moderate to severe .dult .sthm•• Although Ie.. severely, MTH .ppe.... to represent, .t le.st from. gas exchange viewpoint, • model of .sthma slmll.r to what h.s been observed during the n.tural cou.... of the dl..... over a much longer period of time. AM REV RESPIR DIS 1191; 144:81-14
Ala)
Ala
Ala
Ala
to 58 yr of age who fulfilled the criteria of the ATS for the diagnosis of asthma (19) and whose symptoms were well controlled at the time of the study, wererecruited from the community. The main anthropometric and baseline functional data are set out in table 1. Inclusion criteria were (1) age younger than 60 yr; (2) no respiratory infection or severe exacerbation of asthma within the preceding 6 wk; (3) FEV! greater than 1.5 L (and above 60070 predicted) after withholding both adrenergic agents and inhaled steroids for 12 h and oral methylxanthines for 24 h before performing spirometry; (4) maintenance therapy with aerosol specific beta-adrenergic, with or without inhaled beclomethasone/budesonide or theophylline preparations, but DQ treatment with oral steroids for 2 months prior to the study; and (5) absence of any systemic or cardiopulmonary disease other than asthma. No attempt was made to separate atopic from nonatopic patients. All were trained in forced spirometric maneuvers before data for the present study were collected. 1\\'0 subjects
(Received in originalform October 9, 1990 and in revised form February 22, 1991)
1 From the Servei de Pneumologia and Laboratori de Bioffsica i Bioenginyeria, Hospital Clinic, Facultat de Medicina, Universitat de Barcelona, Barcelona, Spain, and the Department of Medicine (Section of Physiology), University of California San Diego, La Jolla, California. 2 Supported by Grant CCA 8309185 from the Joint US-Spain Committee and by Grants PA85-0016 and PM88-0107 from the Comisi6n de Investigaci6n de Ciencia y Tecnologfa (Spain) and a grant from Allen Farmaceutica, Grupo Glaxo, Spain. 3 Presented in part at the Ann ual Meeting of the American Thoracic Society,Cincinnati, Ohio, May 1989. .. Correspondence and requests for reprints should be addressed to R. Rodriguez-Roisin, M.D., Servei de Pneumologia, Hospital Clinic, Villarroel 170, 08036-Barcelona, Spain. 5 Recipient of Postdoctoral Research Fellowship CCA 8309185from the Fundaci6 Bosch i Gimpera, Universitat de Barcelona.
89
GAS EXCHANGE RESPONSE AND METHACHOLINE CHALLENGE IN ASTHMA
(Patients 1 and 12)were regular smokers, and three were ex-smokers (Patients 8, 9, and 13). The study wasapproved by the ResearchCommittee on Human Investigations of the Hospital Clinic-Facultat de Medicina, Universitat de Barcelona, and written informed consent was obtained from each patient after the purpose, risks, and potential 0 f the study were explained and understood.
Procedures The day of the study, forced spirometry (FEV1 and FEF:15 - 7S) (Datospir-2000; Siebel-Med, Barcelona, Spain) according to ATS recommendations and total respiratory system resistance (Rrs) were performed in each patient. The latter was measured by forced oscillation applied at the mouth (20,21). Pseudorandom excitation (2 to 32 Hz; ± 1 em H10IL -1 s) was applied for three periods of 16 s each. Pressure and flow records were digitally highpass filtered (2 Hz, 3 dB) and divided into blocks, each one of 4 s duration, with 50070 overlapping. Rrs was computed by Spectra averaging (21 blocks) with an unbiased estimator to eliminate the bias error caused by breathing (21). The coherence function was also calculated, and data with coherence less than 0.85 were rejected. The analysis of Rrs was restricted to 4 and 12 Hz. In two measurements in which the coherence threshold was not achieved at 4 Hz, Rrs was computed by extrapolating the values obtained between 6 and 10 Hz. Predicted equations for forced spirometry were those of our own laboratory (22), and reference values for Rrs were those of Peslin and coworkers (23). Blood samples were anaerobically collected through a polyethylene catheter (Seldicath; Plastimed, Saint-Lou-La-Foret, France) inserted into the radical artery. Arterial Pol and Pco, were analyzed in duplicate using standard electrodes (IL 1302; Instrumentation Laboratories, Milano, Italy). Hemoglobin concentration was measured by an OSM-2 hemoximeter (Radiometer, Copenhagen, Denmark). A low dead space, low resistance, nonrebreathing valve (No. 1500; Hans Rudolph, Kansas City, MO) connected to a heated metal mixing chamber was used to collect the mixed expired gas. Oxygen uptake and COl consumption were calculated from measured mixed expired 0 1and COl by mass spectrometry (Multigas Monitor MS2; BOC-Medishield, London, UK). Minute ventilation (VE) and respiratory rate (f) were measured using a calibrated Wright spirometer (Respirometer MK8; BOC-Medical, Essex, UK). The alveolar-arterial 0 1difference (AaPo l ) was calculated according to the standard formula using the measured respiratory exchange ratio (R). A three-lead EKG, heart rate (HR), and systemic arterial pressures (psa) were continuously recorded throughout the whole study (HP 7830A Monitor and HP 7754B Recorder; Hewlett-Packard, Waltham, MA). Cardiac output (QT) was measured in duplicate or triplicate using a 5-mg bolus of indocyanine green (Waters Instruments Inc., Rochester, NY) injected into the peripheral venous line
used for inert gas infusion and sampled from the arterial line. VA/Q distributions were estimated by the multiple inert gas elimination technique (24). Specific features of the setup of this technique in our laboratory have been published extensively in earlier reports (25, 26). The inert gas concentrations in the mixed venous blood were calculated from the Fick principle using the mixed expired and arterial inert gas samples and the measured QT. Solubilities of inert gases were measured in each patient. Ventilation-perfusion distributions wereestimated from inert gas data using a least squares algorithm with enforced smoothing (27).The duplicate samples of each set of measurements were treated separately, resulting in two VA/Q distributions in each set, the final data being the average of variables determined from both distributions.
Design of the Study All studies were begun between 8:00 and 9:00 A.M. and wereperformed with patients breathing room air and seated in an armchair. Once all the hemodynamic and respiratory parameters were stable and the inert gas solution had been infused for approximately 45 min to allow for establishment of adequate steadystate conditions (seebelow), baseline measurements were done. All forced spirometric maneuvers were performed at least in duplicate during the entire study. To minimize the influence of both deep inspiration and forced expiration on the bronchomotor tone and gas exchange, all but 5-min post-MTH challenge sets consisted 0 f the following steps in sequence: (1) inert gas sampling and recording of 'IE, (2) respiratory blood and mixed expired gas sampling, (3) systemic hemodynamics and cardiac output measurement, (4) measurement of Rrs, and (5) forced spirometry. The patient was then challenged with MTH following the recommended standardized procedures (28). Particular details of the standardization of the MTH challenge in our laboratory have been reported in full elsewhere (10). In each patient, increasing MTH concentrations (from 0.1 to 25 mg/ml) were used until the cumulative dose causing a 25 to 30070 fall in FEV 1 was reached. Delivery of MTH was always done by the same observer (AF) in all patients using a hand-grip nondosimeter nebulizer (DeVilbiss 42; DeVilbiss Co., Somerset, PA). Five minutes after the challenge was completed, a set of measurements was made (challenge sample). For this set of measurements, FEV 1, first, and then Rrs were part of the MTH procedure such that the remaining variables were measured afterwards, in the same order as for the other three sets. To assess the degree of residual bronchoconstriction immediately after 5-min post-MTH measurements, FEV 1 was again measured (at approximately 15 min after MTH). Thereafter, three puffs of inhaled place bo (composition: trichlorofluoromethane, 5.7 g; dichlorofluoromethane, 14.6 g) or salbutamol (composition: the two former propellants plus oleic acid and salbutamol) (one puff = 100 J.1g) was inhaled, being delivered by the same ob-
server (AF) in randomized double-blind fashion. Complete sets of measurements were taken at 30 and 120 min after the completion of the challenge, corresponding to 15and 105 min after inhalation of placebo or salbutamol respectively. Maintenance of steady-state con: ditions at 5 min after the MTH challenge was demonstrated by stability (± 5070) of hemodynamic (HR and Psa) and spirometric (f and tidal volume) variables, and by agreement of mixed expired and arterial respiratory gases between duplicate measurements (within ± 5010). Moreover, the remaining sum of squares, an index of the goodness of the fit of the inert gas data, was 1.53 ± 0.12and exceeded six in two sets of data alone (6.12and 8.42, each), consistent with expected levels. All patients completed the study without complaints or major side effects. Inhaled salbutamol (two puffs) was administered to all patients after full completion of the study.
Analysis of Data Standard statistics were used to determine mean and SEM and to calculate linear correlations. Paired t tests were used to compare data before and after MTH. Comparisons of data obtained at 30 and 120 min after placebo or salbutamol with 5-min postchallenge samples were assessed by paired t test with Bonferroni's adjustment for multiple comparisons. Differences between salbutamol and placebo at 30 and 120 min after MTH were analyzed by unpaired t test with Bonferroni's adjustment.
Results
Baseline Conditions Before challenge, mean Ffiv.; FEF2s - 7s , Pao., and Pac02 were within predicted or normal values (tables 1 and 2), but Rrs was increased and AaPo 2 was slightly abnormal (> 20 mm Hg), There was a close correlation between Rrs and FEV l (either expressed in absolute values or as percent predicted); at 4 Hz, r = - 0.60 (p < 0.02) each; at 12 Hz, r = - 0.66 (p < 0.01) and - 0.57 (P < 0.02), respectively. Baseline VAlQ ratio distributions were mildly abnormal. An example of one patient is shown in figure 1. Although blood flow distributions were unimodal and relatively narrow in 12 patients, in the four remaining patients (patients 7, 9, 12, and 16), they were modestly bimodal; by contrast, ventilation distributions were unimodal and narrow in all patients. Overall, this was a population of patients with stable asthma, essentially asymptomatic under regular therapy, whose VA/Q ratio distributions on entry of the study were in keeping with data formerly reported in other series of patients with mild to moderate chronic asthma (29). Thus, most of the patients (> 75070) had
90
RODRIQUEZ-ROISIN, FERRER, NAVAJAS, MUSTI, WAGNER, AND AOCA TABLE 1 ANTHROPOMETRIC AND BASELINE FUNCTIONAL DATA ON ENTRY OF THE STUDY
Patient No.
Sex
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
F
F F M F F F
M F M
M M M F M
M
Mean SEM
FEV1
Age (yr)
Height (em)
31 19 45 17 16 58 40 18 35 24 53 45 52 27 57
153 162 157 179 157 155 149 165 168 164 171 171 162 150 169 158
2.10 2.93 2.90 4.81 1.61 3.03 1.65 2.62 2.93 3.88 3.78 3.38 1.83 1.87 3.02 2.09
36.6 3.8
161.9 2.1
2.78 0.23
48
FEF2a - 7a
(% pred)
(L)
Rrs* (em H20 L-, s)
Pa02 (mm Hg)
Pac02 (mm Hg)
AaP0 2 (mm Hg)
Cumulative Breath Units
88.2 100.9 96.0 99.1 86.9 103.7 76.0 81.8 72.1 93.7 98.3
63
5.52 3.17 4.30 2.69 6.18 3.46 3.34 5.46 3.27 3.17 4.46 4.26 9.45 5.94 4.63 4.73
73.5 91.7 97.4 73.5
32.3 34.2 37.7 38.0 41.0 38.1 36.2 37.7 32.0 36.0 34.5 35.4 37.5 35.9 37.5 30.9
18.6 15.6 13.6 8.0 17.7 1.8 40.0 16.2 49.8 18.2 17.8 31.9 35.2 16.3 18.2 46.8
468.0 231.0 580.0 825.0 8.0 1,325.0 5.5 58.0 130.0 55.5 45.5 105.0 138.0 330.5 455.0 130.5
71 6.7
4.63 0.42
88.1 2.7
36.0 0.7
22.9 3.4
305.7 89.9
(Us)
(% pred)
60 95
88 89
1.94 3.78 2.05 5.46 1.29 2.89 0.91 2.65 2.20 3.14 2.55 3.08 1.01 1.20 1.23 1.34
92 4.7
2.30 0.30
74 86 111 120 63 92 81 87 80 126 105 118 66 84
n
123 51 73 48 86 53 96 65 117 36 55 33
n.5
* At 4 Hz.
increased VAlo.inequality by at least one criterion.
Response to Methacholine Both a marked decrease in airflow rates (mean fallinFEV 1,33010, and in FEF2s-7s, 53010) and a substantial increase in mean Rrs (mean, 88010) were observed in all patients, indicating a substantial increase airway responsiveness to MTH. Both 'IE
and HR increased, but Psa, f, QT, and '102 remained unchanged. Whereas Pao, fell (range, 64.1 to 79.5 mm Hg) and AaPo2 increased (range, 16.9 to 54.8 mm Hg), Pac0 2 did not change. Concomitantly, there was mild to moderate to severe worsening of VA/o. relationships. The pattern of blood flow distribution was broadly unimodal in 13 patients (figure 1) and modestly bimodal
TABLE 2 EFFECTS OF METHACHOLINE CHALLENGE Before f, min- 1 \IE, L/min FEV1 , % pred FEF2&-7a , % pred Rrs, em H20 L-1 s HR, min- 1 Psa, mm Hg OT, Umin Pao 2 , mm Hg Paco2 , mm Hg AaP02 , mm Hg \102 , mllmin Shunt, % of OT Low VA/O, % of OT Log SO Q
0 Dead space, % of Log SD V
V DISP R-E*
VE
14.9 7.2 91.8 70.7 4.63 78.3 90.4 5.8 88.1 36.0 22.9 233.3 0.4 1.7 0.71 0.75 27.8 0.57 1.19 5.3
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
0.7 0.3 4.7 6.7 0.42 3.1 3.0 0.3 2.7 0.7 3.4 10.7 0.1 0.7 0.09 0.03 1.9 0.04 0.11 0.8
5 Minutes After 15.3 8.0 62.2 34.7 8.71 84.6 89.1 6.1 71.0 35.3 39.9 242.7 0.2 3.2 0.98 0.66 26.7 0.79 1.50 9.8
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
0.8 0.4 2.9 2.5 0.73 3.4 4.6 0.3 1.1 0.7 2.3 12.0 0.1 1.0 0.04 0.04 1.8 0.04 0.13 0.6
p Value
< 0.02 < 0.001 < 0.001 < 0.001 < 0.01 < 0.001 < 0.001
< 0.01 < 0.05 < 0.001 < 0.001 < 0.001
Definition of abbreviations: f • respiratory rate; VE • minute ventilation. HR • heart rate; Psa • mean systemic arterial pressure; Q-r • cardiac output; V02 • O2 uptake. Low VAlO • perfusion to lung units with low ventilation-perfusionratios « 0.1). Log SO Q • dispersion (second moment) of blood flow distribution on a log scale. Q • mean(first moment) VAlQ ratio of the perfusion distribution; Log SO V • dispersion (second moment)of ventilation distribution on a log scale; V• mean (first moment) VAJ6 ratio of the ventilation distribution; DISP R-E* • root mean squaredifference among measured retentions (R) and excretions (E) of the inert gases (except acetone) corrected for the dead space (30).
in the three remaining patients (Patients 6, 14, and 16), of whom only Patient 16 had a bimodal baseline distribution. The ventilation distributions also broadened but were never bimodal. The amount of dispersion of both distributions (log SO Q and log SO V) increased (range, 0.75 to 1.29 and 0.56 to 1.22, respectively) (normal range, 0.3 to 0.6 [29]), as did an overall index of VAlo. heterogeneity (OISP R-E*) (range, 5.13 to 12.9) (normal values < 3.0 [30]). Likewise, whereas Q decreased (range, 0.48 to 1.1), Vincreased (range, 0.74 to 7.55). By contrast, shunt, low and high VAlQ modes, and dead space did not change significantly. Although mean FEV 1 measured immediately after the completion of the postchallenge sampling sequence (just before giving placebo or salbutamol) showed that the degree of bronchoconstriction had already lessened (FEV 1 to 77 ± 4010 predicted or 16010 below baseline), this was still significantly lower than baseline (P < 0.(01). Pre-MTH to post-MTH challenge differences in Pao., in AaPo2, in log SO Q, in log SO V, and in OISP R-E* were closely correlated with their corresponding baseline prechallenge values (r, 0.92 [p < 0.001], 0.73 [p < 0.005], 0.86 [p < 0.001], 0.53, and 0.72 [p < 0.05 each], respectively). No correlations were shown between pre-MTH to post-MTH challenge differences in Rrs (at 4 and at 12 Hz) and in FEV 1 (expressed as percent predicted) (r, - 0.34 and - 0.32), respectively; by contrast, significant correlations were found when the changes in Pao, or in AaPo2 were plotted against
GAS EXCHANGE RESPONSE AND METHACHOLINE CHALLENGE IN ASTHMA 10
~
0.7
PaOl, 88 mm 119
Z
-J
0.6
..!,.
~
0.3
9 LL.
0.4
0
02 SHUNT
§
0.1
0%
rn
0
0
> a:
0.1
0.01
0
10.0
1.0
100.0
QOI
0
0.1
1.0
10D
looD
~
Z
0
0.7
~
~
05
0
Z
0.8
IgT:~1
"'Oz,H -Hg I.IIgSDQ,O.49 LogSDV, 0.41 asP R-E,·3.~
0.6
ClSP R-£~'O 0
«
S
I~~I
P.02 , 71 mIll 119 LogSOQ.125 LogSDV. 010
-J
0.3
t= Z
0.1
Response to Salbutamol
0.2
-J
~
0
0
0.1
0.01
0
1.0
100.0
10.0
0
0.01
0.1
1.0
10D
100.0
VA/a RATIO Fig. 1. Representative sequence of VAla distributions in Patient 2 treated with placebo. Immediately after challenge, there was mild to moderate VAla mismatch, as assessed by increases in the dispersion of blood flow (log SD a) and that of ventilation (log SD V), and in an index of overall VAla heterogeneity (DISP R-E*), and Pao2 fell. These findings were still present at 30 min after challenge. Note that after challenge both distributions broadened, more specifically, the perfusion distribution. Within 2 h, all variables had returned to baseline values.
those in log SO Q (r, - 0.80 and 0.90), in log SO V (r, -0.78 and 0.75), and in OISP R-E* (r, - 0.80 and 0.87) (p < 0.001 each), respectively. Moreover, although the differences in Rrs (at 4 and at 12 Hz) observed before and after challenge were not correlated with those in Pao, (r, -0.31 and -0.47), in AaPo 2 (r, 0.31 and 0.42), and in log SO Q (r, 0.29 and 0.48), they were closely correlated with those in log SO V (r, 0.65 and 0.78) and in OISP R-E* (r, 0.63 and 0.76) (p < 0.01 each), respectively (figure'2). Neither were correlations between changes in FEV 1 (expressed as percent predicted) and those in Pao, (r, 0.43), in AaPo 2 (r, -0.20), in log SO Q (r, -0.03), in log SO V (r, -0.35), and in OISP R-E* (r, -0.34) before and after challenge.
Time Course of MTH Challenge
lenge, subjects treated with placebo showed a significantly higher FEV 1 (P < 0.001) and FEF25 - 75 (P < 0.005) compared with values obtained after the initial postchallenge sampling sequence, which were both within a narrow range (approximately at 90070) of the prechallenge values; also, AaPo2 and QT were slightly reduced (P < 0.025 each). At 120 min, FEV 1 , FEF25 - 75 , and Pao2 were higher (P < 0.005 each), whereas Rrs (at 4 Hz)
...:.... ..
0.4
0
0.2
C>
9
ROBERT RODRIGUEZ-RaiSIN, ANTONI FERRER,! DANIEL NAVAJAS, ALVAR G. N. AGUSTI, PETER D. WAGNER, and JOSEP ROCA
Introduction
Several experimental (1) and clinical studies (2-6) have demonstrated dissociation between maximal expiratory airflow rates and pulmonary gas exchange in both acute and chronic forms of bronchial asthma. Accordingly, it has been suggested that airflow rates are determined mostly by properties of the larger airways, whereas gas exchange is more related to peripheral airway obstruction (4). However, even when measured, arterial blood gases do not suffice in the analysis of the functional changes associated with bronchial asthma because of simultaneous changes in extrapulmonary determinants of arterial Pao, such as minute ventilation, cardiac output, and oxygenuptake (3-6). Little information concerning VAlQ inequality in bronchoprovocation challenge tests is available. Although it is known that these tests induce transient arterial hypoxemia or oxygen desaturation and/or abnormal lung scans (7-15), VA/Q inequality has only been investigateddirectly in exerciseinduced asthma in both adults (16) and children (17) and also after histamine in children only (18). By contrast, the pulmonary gas exchange response to methacholine (MTH) inhalation challenge, by far one of the most widely used tools for the diagnosis of asthma in both the clinicaland the epidemiologic setting, remains poorly documented to date. The aim of the present study was to investigate the pattern and time course of VA/Q inequality after challenge to better define the model of MTH bronchial challenge in patients with asymptomatic mild asthma. An additional question was whether an inhaled specific beta-adrenergic agent such as salbutamol may alter VAlQ mismatching induced by the MTH challenge test. Methods
Patients Sixteen patients (eight male, eight female) 16 88
SUMMARY ToInvestigate the effects of methacholine (MTH)ch.llenge on spirometry, lung mech.ndistributions, 18subjects 18to 58 yr of .ge Ics, respl...tory g...., .nd ventilation-perfusion (V with stable mild .sthma (FEV1 , 92 ± 5% [SEM] predicted; FEFu - 15 , 71 ± 7% predicted; respiratory system resistance (R...).t 4 Hz, 4.8 ± 0.4 cm H20/L- 1 s; Pa02' 88 ± 3 mm Hg; A.P0 2, 23 ± 3 mm distributions were unlmod.1 .nd rel.tlvely n.rrow In 12 petlents Hg) were recruited. sasellne V and modestly blmod.1 In the other four. The dlsperalon of pulmon.ry blood flow (log SO Q) was slightly enl.rged (0.71 ± 0.09) .nd th.t of ventll.tlon (log SO V) was norm.I (0.57 ± 0.04) (norm.' range, 0.3 to 0.8); an Index of overall V heterogeneity (OISPR-E·) was .Iao mildly abnorm.1 (5.3 ± 0.8) (norm.1 values < 3.0). After MTH ch.llenge, FEVh FEF25 - 15 , .nd Pa02 fell (to 82 ± 3.nd 35 ± 3% predicted, .nd to 71 ± 1 mm Hg, respectively), where.s R... (p < 0.001 each), minute ventll.tlon (p < 0.02), he.rt r.te (p < 0.01),.nd A.P0 2Incre.sed (p < 0.001).V rel.tlonshlps mildly to moder.tely wo....ned (log SO Q Incre.sed to 0.98 ± 0.04 [p < 0.01], log SO V to 0.79 ± 0.04, and OISP R-E· to 9.8 ± 0.8 [p < 0.001 each)). Qu.lltatlvely, the pattern of blood flow dlatrlbutlon was broadly unlmod.lln 13petlents .nd modestly blmod.lln three, of whom only one had • blmod.1 be..llne distribution. Pre-MTHto poat-MTH ch.llenge changes In R... were correlated to the ch.nges In both log SO V (r, 0.85) .nd DISP R-E· (r, 0.83) (p < 0.01 e.ch). After these me.surements (15 min .fter MTH), eight patients received placebo .nd eight received aalbutamol by InlMllatlon In ...ndomlzed double-blind fIIshlon. By 30 min .fter challenge, patients given placebo showed Improved maxlm.1 expl...tory ...tes, where.s R....nd gu exchange remained e_ntl.Uy unchanged from the 5-mln post-MTH levels. Salbutamol significantly Improved not only .Irflow ...tes but alao R... and all the Indices of pulmonary g.s exch.nge. Our results Indicate that .fter MTH challenge there Is dlaaocl.tlon between spirometry .nd gas exch.nge, Just .s Is often _n In moderate to severe .dult .sthm•• Although Ie.. severely, MTH .ppe.... to represent, .t le.st from. gas exchange viewpoint, • model of .sthma slmll.r to what h.s been observed during the n.tural cou.... of the dl..... over a much longer period of time. AM REV RESPIR DIS 1191; 144:81-14
Ala)
Ala
Ala
Ala
to 58 yr of age who fulfilled the criteria of the ATS for the diagnosis of asthma (19) and whose symptoms were well controlled at the time of the study, wererecruited from the community. The main anthropometric and baseline functional data are set out in table 1. Inclusion criteria were (1) age younger than 60 yr; (2) no respiratory infection or severe exacerbation of asthma within the preceding 6 wk; (3) FEV! greater than 1.5 L (and above 60070 predicted) after withholding both adrenergic agents and inhaled steroids for 12 h and oral methylxanthines for 24 h before performing spirometry; (4) maintenance therapy with aerosol specific beta-adrenergic, with or without inhaled beclomethasone/budesonide or theophylline preparations, but DQ treatment with oral steroids for 2 months prior to the study; and (5) absence of any systemic or cardiopulmonary disease other than asthma. No attempt was made to separate atopic from nonatopic patients. All were trained in forced spirometric maneuvers before data for the present study were collected. 1\\'0 subjects
(Received in originalform October 9, 1990 and in revised form February 22, 1991)
1 From the Servei de Pneumologia and Laboratori de Bioffsica i Bioenginyeria, Hospital Clinic, Facultat de Medicina, Universitat de Barcelona, Barcelona, Spain, and the Department of Medicine (Section of Physiology), University of California San Diego, La Jolla, California. 2 Supported by Grant CCA 8309185 from the Joint US-Spain Committee and by Grants PA85-0016 and PM88-0107 from the Comisi6n de Investigaci6n de Ciencia y Tecnologfa (Spain) and a grant from Allen Farmaceutica, Grupo Glaxo, Spain. 3 Presented in part at the Ann ual Meeting of the American Thoracic Society,Cincinnati, Ohio, May 1989. .. Correspondence and requests for reprints should be addressed to R. Rodriguez-Roisin, M.D., Servei de Pneumologia, Hospital Clinic, Villarroel 170, 08036-Barcelona, Spain. 5 Recipient of Postdoctoral Research Fellowship CCA 8309185from the Fundaci6 Bosch i Gimpera, Universitat de Barcelona.
89
GAS EXCHANGE RESPONSE AND METHACHOLINE CHALLENGE IN ASTHMA
(Patients 1 and 12)were regular smokers, and three were ex-smokers (Patients 8, 9, and 13). The study wasapproved by the ResearchCommittee on Human Investigations of the Hospital Clinic-Facultat de Medicina, Universitat de Barcelona, and written informed consent was obtained from each patient after the purpose, risks, and potential 0 f the study were explained and understood.
Procedures The day of the study, forced spirometry (FEV1 and FEF:15 - 7S) (Datospir-2000; Siebel-Med, Barcelona, Spain) according to ATS recommendations and total respiratory system resistance (Rrs) were performed in each patient. The latter was measured by forced oscillation applied at the mouth (20,21). Pseudorandom excitation (2 to 32 Hz; ± 1 em H10IL -1 s) was applied for three periods of 16 s each. Pressure and flow records were digitally highpass filtered (2 Hz, 3 dB) and divided into blocks, each one of 4 s duration, with 50070 overlapping. Rrs was computed by Spectra averaging (21 blocks) with an unbiased estimator to eliminate the bias error caused by breathing (21). The coherence function was also calculated, and data with coherence less than 0.85 were rejected. The analysis of Rrs was restricted to 4 and 12 Hz. In two measurements in which the coherence threshold was not achieved at 4 Hz, Rrs was computed by extrapolating the values obtained between 6 and 10 Hz. Predicted equations for forced spirometry were those of our own laboratory (22), and reference values for Rrs were those of Peslin and coworkers (23). Blood samples were anaerobically collected through a polyethylene catheter (Seldicath; Plastimed, Saint-Lou-La-Foret, France) inserted into the radical artery. Arterial Pol and Pco, were analyzed in duplicate using standard electrodes (IL 1302; Instrumentation Laboratories, Milano, Italy). Hemoglobin concentration was measured by an OSM-2 hemoximeter (Radiometer, Copenhagen, Denmark). A low dead space, low resistance, nonrebreathing valve (No. 1500; Hans Rudolph, Kansas City, MO) connected to a heated metal mixing chamber was used to collect the mixed expired gas. Oxygen uptake and COl consumption were calculated from measured mixed expired 0 1and COl by mass spectrometry (Multigas Monitor MS2; BOC-Medishield, London, UK). Minute ventilation (VE) and respiratory rate (f) were measured using a calibrated Wright spirometer (Respirometer MK8; BOC-Medical, Essex, UK). The alveolar-arterial 0 1difference (AaPo l ) was calculated according to the standard formula using the measured respiratory exchange ratio (R). A three-lead EKG, heart rate (HR), and systemic arterial pressures (psa) were continuously recorded throughout the whole study (HP 7830A Monitor and HP 7754B Recorder; Hewlett-Packard, Waltham, MA). Cardiac output (QT) was measured in duplicate or triplicate using a 5-mg bolus of indocyanine green (Waters Instruments Inc., Rochester, NY) injected into the peripheral venous line
used for inert gas infusion and sampled from the arterial line. VA/Q distributions were estimated by the multiple inert gas elimination technique (24). Specific features of the setup of this technique in our laboratory have been published extensively in earlier reports (25, 26). The inert gas concentrations in the mixed venous blood were calculated from the Fick principle using the mixed expired and arterial inert gas samples and the measured QT. Solubilities of inert gases were measured in each patient. Ventilation-perfusion distributions wereestimated from inert gas data using a least squares algorithm with enforced smoothing (27).The duplicate samples of each set of measurements were treated separately, resulting in two VA/Q distributions in each set, the final data being the average of variables determined from both distributions.
Design of the Study All studies were begun between 8:00 and 9:00 A.M. and wereperformed with patients breathing room air and seated in an armchair. Once all the hemodynamic and respiratory parameters were stable and the inert gas solution had been infused for approximately 45 min to allow for establishment of adequate steadystate conditions (seebelow), baseline measurements were done. All forced spirometric maneuvers were performed at least in duplicate during the entire study. To minimize the influence of both deep inspiration and forced expiration on the bronchomotor tone and gas exchange, all but 5-min post-MTH challenge sets consisted 0 f the following steps in sequence: (1) inert gas sampling and recording of 'IE, (2) respiratory blood and mixed expired gas sampling, (3) systemic hemodynamics and cardiac output measurement, (4) measurement of Rrs, and (5) forced spirometry. The patient was then challenged with MTH following the recommended standardized procedures (28). Particular details of the standardization of the MTH challenge in our laboratory have been reported in full elsewhere (10). In each patient, increasing MTH concentrations (from 0.1 to 25 mg/ml) were used until the cumulative dose causing a 25 to 30070 fall in FEV 1 was reached. Delivery of MTH was always done by the same observer (AF) in all patients using a hand-grip nondosimeter nebulizer (DeVilbiss 42; DeVilbiss Co., Somerset, PA). Five minutes after the challenge was completed, a set of measurements was made (challenge sample). For this set of measurements, FEV 1, first, and then Rrs were part of the MTH procedure such that the remaining variables were measured afterwards, in the same order as for the other three sets. To assess the degree of residual bronchoconstriction immediately after 5-min post-MTH measurements, FEV 1 was again measured (at approximately 15 min after MTH). Thereafter, three puffs of inhaled place bo (composition: trichlorofluoromethane, 5.7 g; dichlorofluoromethane, 14.6 g) or salbutamol (composition: the two former propellants plus oleic acid and salbutamol) (one puff = 100 J.1g) was inhaled, being delivered by the same ob-
server (AF) in randomized double-blind fashion. Complete sets of measurements were taken at 30 and 120 min after the completion of the challenge, corresponding to 15and 105 min after inhalation of placebo or salbutamol respectively. Maintenance of steady-state con: ditions at 5 min after the MTH challenge was demonstrated by stability (± 5070) of hemodynamic (HR and Psa) and spirometric (f and tidal volume) variables, and by agreement of mixed expired and arterial respiratory gases between duplicate measurements (within ± 5010). Moreover, the remaining sum of squares, an index of the goodness of the fit of the inert gas data, was 1.53 ± 0.12and exceeded six in two sets of data alone (6.12and 8.42, each), consistent with expected levels. All patients completed the study without complaints or major side effects. Inhaled salbutamol (two puffs) was administered to all patients after full completion of the study.
Analysis of Data Standard statistics were used to determine mean and SEM and to calculate linear correlations. Paired t tests were used to compare data before and after MTH. Comparisons of data obtained at 30 and 120 min after placebo or salbutamol with 5-min postchallenge samples were assessed by paired t test with Bonferroni's adjustment for multiple comparisons. Differences between salbutamol and placebo at 30 and 120 min after MTH were analyzed by unpaired t test with Bonferroni's adjustment.
Results
Baseline Conditions Before challenge, mean Ffiv.; FEF2s - 7s , Pao., and Pac02 were within predicted or normal values (tables 1 and 2), but Rrs was increased and AaPo 2 was slightly abnormal (> 20 mm Hg), There was a close correlation between Rrs and FEV l (either expressed in absolute values or as percent predicted); at 4 Hz, r = - 0.60 (p < 0.02) each; at 12 Hz, r = - 0.66 (p < 0.01) and - 0.57 (P < 0.02), respectively. Baseline VAlQ ratio distributions were mildly abnormal. An example of one patient is shown in figure 1. Although blood flow distributions were unimodal and relatively narrow in 12 patients, in the four remaining patients (patients 7, 9, 12, and 16), they were modestly bimodal; by contrast, ventilation distributions were unimodal and narrow in all patients. Overall, this was a population of patients with stable asthma, essentially asymptomatic under regular therapy, whose VA/Q ratio distributions on entry of the study were in keeping with data formerly reported in other series of patients with mild to moderate chronic asthma (29). Thus, most of the patients (> 75070) had
90
RODRIQUEZ-ROISIN, FERRER, NAVAJAS, MUSTI, WAGNER, AND AOCA TABLE 1 ANTHROPOMETRIC AND BASELINE FUNCTIONAL DATA ON ENTRY OF THE STUDY
Patient No.
Sex
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
F
F F M F F F
M F M
M M M F M
M
Mean SEM
FEV1
Age (yr)
Height (em)
31 19 45 17 16 58 40 18 35 24 53 45 52 27 57
153 162 157 179 157 155 149 165 168 164 171 171 162 150 169 158
2.10 2.93 2.90 4.81 1.61 3.03 1.65 2.62 2.93 3.88 3.78 3.38 1.83 1.87 3.02 2.09
36.6 3.8
161.9 2.1
2.78 0.23
48
FEF2a - 7a
(% pred)
(L)
Rrs* (em H20 L-, s)
Pa02 (mm Hg)
Pac02 (mm Hg)
AaP0 2 (mm Hg)
Cumulative Breath Units
88.2 100.9 96.0 99.1 86.9 103.7 76.0 81.8 72.1 93.7 98.3
63
5.52 3.17 4.30 2.69 6.18 3.46 3.34 5.46 3.27 3.17 4.46 4.26 9.45 5.94 4.63 4.73
73.5 91.7 97.4 73.5
32.3 34.2 37.7 38.0 41.0 38.1 36.2 37.7 32.0 36.0 34.5 35.4 37.5 35.9 37.5 30.9
18.6 15.6 13.6 8.0 17.7 1.8 40.0 16.2 49.8 18.2 17.8 31.9 35.2 16.3 18.2 46.8
468.0 231.0 580.0 825.0 8.0 1,325.0 5.5 58.0 130.0 55.5 45.5 105.0 138.0 330.5 455.0 130.5
71 6.7
4.63 0.42
88.1 2.7
36.0 0.7
22.9 3.4
305.7 89.9
(Us)
(% pred)
60 95
88 89
1.94 3.78 2.05 5.46 1.29 2.89 0.91 2.65 2.20 3.14 2.55 3.08 1.01 1.20 1.23 1.34
92 4.7
2.30 0.30
74 86 111 120 63 92 81 87 80 126 105 118 66 84
n
123 51 73 48 86 53 96 65 117 36 55 33
n.5
* At 4 Hz.
increased VAlo.inequality by at least one criterion.
Response to Methacholine Both a marked decrease in airflow rates (mean fallinFEV 1,33010, and in FEF2s-7s, 53010) and a substantial increase in mean Rrs (mean, 88010) were observed in all patients, indicating a substantial increase airway responsiveness to MTH. Both 'IE
and HR increased, but Psa, f, QT, and '102 remained unchanged. Whereas Pao, fell (range, 64.1 to 79.5 mm Hg) and AaPo2 increased (range, 16.9 to 54.8 mm Hg), Pac0 2 did not change. Concomitantly, there was mild to moderate to severe worsening of VA/o. relationships. The pattern of blood flow distribution was broadly unimodal in 13 patients (figure 1) and modestly bimodal
TABLE 2 EFFECTS OF METHACHOLINE CHALLENGE Before f, min- 1 \IE, L/min FEV1 , % pred FEF2&-7a , % pred Rrs, em H20 L-1 s HR, min- 1 Psa, mm Hg OT, Umin Pao 2 , mm Hg Paco2 , mm Hg AaP02 , mm Hg \102 , mllmin Shunt, % of OT Low VA/O, % of OT Log SO Q
0 Dead space, % of Log SD V
V DISP R-E*
VE
14.9 7.2 91.8 70.7 4.63 78.3 90.4 5.8 88.1 36.0 22.9 233.3 0.4 1.7 0.71 0.75 27.8 0.57 1.19 5.3
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
0.7 0.3 4.7 6.7 0.42 3.1 3.0 0.3 2.7 0.7 3.4 10.7 0.1 0.7 0.09 0.03 1.9 0.04 0.11 0.8
5 Minutes After 15.3 8.0 62.2 34.7 8.71 84.6 89.1 6.1 71.0 35.3 39.9 242.7 0.2 3.2 0.98 0.66 26.7 0.79 1.50 9.8
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
0.8 0.4 2.9 2.5 0.73 3.4 4.6 0.3 1.1 0.7 2.3 12.0 0.1 1.0 0.04 0.04 1.8 0.04 0.13 0.6
p Value
< 0.02 < 0.001 < 0.001 < 0.001 < 0.01 < 0.001 < 0.001
< 0.01 < 0.05 < 0.001 < 0.001 < 0.001
Definition of abbreviations: f • respiratory rate; VE • minute ventilation. HR • heart rate; Psa • mean systemic arterial pressure; Q-r • cardiac output; V02 • O2 uptake. Low VAlO • perfusion to lung units with low ventilation-perfusionratios « 0.1). Log SO Q • dispersion (second moment) of blood flow distribution on a log scale. Q • mean(first moment) VAlQ ratio of the perfusion distribution; Log SO V • dispersion (second moment)of ventilation distribution on a log scale; V• mean (first moment) VAJ6 ratio of the ventilation distribution; DISP R-E* • root mean squaredifference among measured retentions (R) and excretions (E) of the inert gases (except acetone) corrected for the dead space (30).
in the three remaining patients (Patients 6, 14, and 16), of whom only Patient 16 had a bimodal baseline distribution. The ventilation distributions also broadened but were never bimodal. The amount of dispersion of both distributions (log SO Q and log SO V) increased (range, 0.75 to 1.29 and 0.56 to 1.22, respectively) (normal range, 0.3 to 0.6 [29]), as did an overall index of VAlo. heterogeneity (OISP R-E*) (range, 5.13 to 12.9) (normal values < 3.0 [30]). Likewise, whereas Q decreased (range, 0.48 to 1.1), Vincreased (range, 0.74 to 7.55). By contrast, shunt, low and high VAlQ modes, and dead space did not change significantly. Although mean FEV 1 measured immediately after the completion of the postchallenge sampling sequence (just before giving placebo or salbutamol) showed that the degree of bronchoconstriction had already lessened (FEV 1 to 77 ± 4010 predicted or 16010 below baseline), this was still significantly lower than baseline (P < 0.(01). Pre-MTH to post-MTH challenge differences in Pao., in AaPo2, in log SO Q, in log SO V, and in OISP R-E* were closely correlated with their corresponding baseline prechallenge values (r, 0.92 [p < 0.001], 0.73 [p < 0.005], 0.86 [p < 0.001], 0.53, and 0.72 [p < 0.05 each], respectively). No correlations were shown between pre-MTH to post-MTH challenge differences in Rrs (at 4 and at 12 Hz) and in FEV 1 (expressed as percent predicted) (r, - 0.34 and - 0.32), respectively; by contrast, significant correlations were found when the changes in Pao, or in AaPo2 were plotted against
GAS EXCHANGE RESPONSE AND METHACHOLINE CHALLENGE IN ASTHMA 10
~
0.7
PaOl, 88 mm 119
Z
-J
0.6
..!,.
~
0.3
9 LL.
0.4
0
02 SHUNT
§
0.1
0%
rn
0
0
> a:
0.1
0.01
0
10.0
1.0
100.0
QOI
0
0.1
1.0
10D
looD
~
Z
0
0.7
~
~
05
0
Z
0.8
IgT:~1
"'Oz,H -Hg I.IIgSDQ,O.49 LogSDV, 0.41 asP R-E,·3.~
0.6
ClSP R-£~'O 0
«
S
I~~I
P.02 , 71 mIll 119 LogSOQ.125 LogSDV. 010
-J
0.3
t= Z
0.1
Response to Salbutamol
0.2
-J
~
0
0
0.1
0.01
0
1.0
100.0
10.0
0
0.01
0.1
1.0
10D
100.0
VA/a RATIO Fig. 1. Representative sequence of VAla distributions in Patient 2 treated with placebo. Immediately after challenge, there was mild to moderate VAla mismatch, as assessed by increases in the dispersion of blood flow (log SD a) and that of ventilation (log SD V), and in an index of overall VAla heterogeneity (DISP R-E*), and Pao2 fell. These findings were still present at 30 min after challenge. Note that after challenge both distributions broadened, more specifically, the perfusion distribution. Within 2 h, all variables had returned to baseline values.
those in log SO Q (r, - 0.80 and 0.90), in log SO V (r, -0.78 and 0.75), and in OISP R-E* (r, - 0.80 and 0.87) (p < 0.001 each), respectively. Moreover, although the differences in Rrs (at 4 and at 12 Hz) observed before and after challenge were not correlated with those in Pao, (r, -0.31 and -0.47), in AaPo 2 (r, 0.31 and 0.42), and in log SO Q (r, 0.29 and 0.48), they were closely correlated with those in log SO V (r, 0.65 and 0.78) and in OISP R-E* (r, 0.63 and 0.76) (p < 0.01 each), respectively (figure'2). Neither were correlations between changes in FEV 1 (expressed as percent predicted) and those in Pao, (r, 0.43), in AaPo 2 (r, -0.20), in log SO Q (r, -0.03), in log SO V (r, -0.35), and in OISP R-E* (r, -0.34) before and after challenge.
Time Course of MTH Challenge
lenge, subjects treated with placebo showed a significantly higher FEV 1 (P < 0.001) and FEF25 - 75 (P < 0.005) compared with values obtained after the initial postchallenge sampling sequence, which were both within a narrow range (approximately at 90070) of the prechallenge values; also, AaPo2 and QT were slightly reduced (P < 0.025 each). At 120 min, FEV 1 , FEF25 - 75 , and Pao2 were higher (P < 0.005 each), whereas Rrs (at 4 Hz)
...:.... ..
0.4
0
0.2
C>
9
