Distilled Water Challenges in Asthmatic Children

Distilled Water Challenges in Asthmatic Children* Comparison of Different Protocols Irmgard Eichler; M.D.; Manfred GOtz, M.D.; Jasminka Zarkooic, M.D.;

and Andrea KOfinger

Inhalation of ultrasonically nebulized distilled water (UNDW) appears a promising candidate for routine challenge testing in bronchial asthma. We have compared two different methods of application of UNDW in 12 asthmatic children with a positive response to methacholine provocation (MCh), in an attempt to increase UNDW sensitivity and to establish standard testing protocols. In addition, resuits from UNDW challenges were compared to responses to inhalation of jet-nebulized distilled water ONDW) and cold air (CACh). Compared to MCh, the sensitivity of continuously or intermittently (iUNDW) inhaled UNDW was 67 percent or 75 percent, respectively, when a positive response was defined by a ~20 percent fall in FEV., but was higher when definition of a positive response was based on restilts from Bow volume curves. Sensitivity of continuous or intermittent inhalation of

is, at present, no method for testing bronchial T here responsiveness that is optimally suited for young

children. Methacholine provocation (MCh) is considered the criterion standard for testing because of its high sensitivity, but it is not a specific test for asthma, 1 and it requires a considerable amount of patient cooperation and time. Cold air and exercise challenges also require maximum patient effort, which is a disadvantage with young children whose cooperation is unpredictable. Inhalation ofultrasonically nebulized distilled water (UNDW) is a potent bronchoconstrictor in patients with bronchial asthma and has been proposed for measuring bronchial responsiveness. It was found to be highly specific for asthma, especially when compared to methacholine provocation.2-5 The UNDW inhalation requires considerably less patient cooperation than other testing procedures and is well tolerated since tidal breathing is used. Another advantage is low cost. Hence, UNDW challenge could become the test of choice in children. Compared to pharmacologic

challenges, UND~ a natural stimulus, is appealing as a potentially useful method in the evaluation of bronchodilator medications because of the avoidance of drug-interactions. However, UNDW challenge cannot yet be recommended for routine screening because *From the Department of Pediatrics, Vienna University Hospital, Vienna, Austria. Manuscript received October 31; revision accepted January 24.

JNDW was lower than for UNDW. The UNDW inhalations were better tolerated than CACh. Following stepwise iUNDW challenge, there was a clear reaction plateau for aU variables measured. Results indicate that testing protocols with iUNDW inhalations over 6 to 10 min (corresponding to 7 to 11 ml water inhalation) yield the maximum sensitivity attainable with UNDW challenges, and require a minimum of patient and investigator effort. (Cheat 1992; 102:753-58) BT = aerosol tem~rature before challenge; c = continuous; CACh = cold air challenge; i = intermittently· JNDW = jet nebulized distilled water; MCh = methacholine provocation; MEFV=maximum expiratory 80w volume; PT=aerosol temperature after challenge; UNDW = ultrasonically nebulized distilled water

of its apparent lack of sensitivity. 1.4.5 The purposes of this study were as follow: (a) to compare two different methods of application of UNDW in asthmatic children, in an attempt to increase UNDW sensitivity and to establish standard testing protocols; (b) to determine whether a reaction plateau can be demonstrated during intermittent UNDW inhalation in asthmatic children; (c) to assess whether bronchoconstriction comparable to UNDW challenges can be obtained using jet-nebulized distilled water ONDW); and (d) to compare the response to inhalation of distilled water with the response to cold air. METHODS

Subjects Fifteen asthmatic patients, eight girls and seven boys. aged 11.5±2.9 years (mean±SD; range: 7.75 to 17 years) with posithe results from MCh entered the study: AU patients had asth.a previously diagnosed by a physician and gave a history of episodic dyspnea, wheezing, and coughing. Asthma was precipitated by exercise in six patients, by exposure to known a1lelltens in four patients, and by cold, foggy weather in three patients. Standardized skin prick testing, radioallergosorhent tests (RAST) with the seven most common inhalant allergens (Dermatophagoides pteronyssinus, cat dander, birch pollen, mixed grasses, mugwort, Cladosporium, and Alternaria), and total plasma-IgE (PRIST) were performed on all patients. Seven patients had at least one positive skin prick test reaction, eight patients had at least one positive class 3 to 4 RAST result and total 19E plasma levels> 100 kUIL (range: 102 to > 1000 kUIL). None received systemic corticosteroid treatment, and all patients were asymptomatic at the time of the study and had not CHEST I 102 I 3 I SEPTEMBER, 1992

753

taken oral or inhaled drugs for at least 24 h prior to any challenge. Informed parental consent was obtained before the study. Study Design

Each subject was tested on six different days at the same time of the day. The six different inhalation tests were conducted at least 48 h apart, within two weeks in a randomized order. Before each challenge, baseline pulmonary function tests (PITs) were performed: thoracic gas volume, specific airway resistance and its reciprocal, and specific conductance (Sgaw) were obtained by body plethysmography as described elsewhere. e Forced vital capacity, FEV. and maximum expiratory Bow at 50 percent (MEF50) and the remaining 25 percent (MEF25) of vital capacity were recorded in the form ofa maximum expiratory Bow volume (M EFV) curve. The best of three efforts was used for calculations. Results were expressed as percent predicted normals based on accepted reference standards. 1 After determination of baseline pulmonary function, the bronchial challenge tests were started. The changes in PF (~PF) in response to bronchial challenges were determined as: ~PF = (baseline PF - postchallenge PF}lbaseline PF x 100 For each challenge, a drop in FEV. of C!;20 percent and in Sgaw of C!;40 percent from baseline was considered a positive result. If the last postchallenge PIT after each inhalation test revealed a FEV. of less than 100 percent baseline or less than 80 percent of predicted value, salbutamol (200 tJ.g) was administered by pressurized aerosol inhaler. Side effects, such as cough or inability to continue with the inhalation test were recorded. Inhalation 'lests

Methacholine Challenge 1Ut: Initially, subjects inhaled normal

saline solution nebulized by a Pari-Inhalierboy (Pari, Starnberg, Germany) for 2 min while breathing normally. Immediately, and 3 min after inhalation, MEFV curves and plethysmography were repeated. As no subject had a fall of FEV. ~20 percent with the diluent, the test was continued and subjects inhaled methacholine chloride, prepared from a fresh bottle, in the same wa~ Given the high nebulizer output of0.38 mVmin, the concentrations used were low: 0.03, 0.06, 0.125, 0.25, 0.5, 1.0, 2.0, and 4.0 mg methacholine! ml. Immediately, and 3 min after each inhalation, MEFV curves and plethysmography were repeated. The test was stopped when the highest dose was inhaled or when a fall in FEV. ~20 percent had been reached. For each subject, the FEV. was plotted against the log of the methacholine concentration, and the provocative concentration of methacholine required for a 20 percent fall in FEV I (PC20) was determined from this line. In a similar way, the provocative concentration required for a 40 percent fall in Sgaw (PC40) was determined. Distilled WJter Challenges: Aerosols of distilled water were generated either by an ultrasonic nebulizer (UNDW challenge) or by a jet nebulizer ONDW challenge). The temperature of the aerosols was recorded inside the mouthpiece immediately before (B1), and after (P1) each challenge with a digital thermometer. Volume output was determined by weighing the nebulizer cannisters initially and at the end ofeach step or after the continous inhalation, respectively, on a precision balance Sartorius 2250 (accuracy: ± 1 mg). By summing the cannister output at each of the individual steps, we obtained a cumulative dose of water nebulized. The water volume output of the ultrasonic nebulizer was 1.1 ± 0.2 mVmin, and aerosolized particle size was set at 10 tJ.m. Water volume output of the jet nebulizer was 0.38 mllmin, and the aerosolized particle size was between 0.5 and 5 fLm, according to company specifications. The jet nebulizer was connected to a cooling box to maintain the temperature of the aerosol at a constant level. Both UNDW and JNDW challenges were performed continuously (cUND~ cJNDW) or intermittently (iUND~ iJNDW). The

754

patients, with noseclips attached, tidally breathed through a mouthpiece with a three-way valve to permit the expired air to be exhaled into the atmosphere. Continuous Inhalatfon: Subjects inhaled the aerosols continuously over 10 min. Immediatel~ 2, 5, and 10 min after the inhalation, MEFV curves and plethysmography were repeated. Intennittent Inhalation: This type of challenge was performed using sequential timed steps of 2, 4, 8, 6, 5, and another 5 min, separated from each other only by the short period oftime necessary to repeat PIT During this interval, the output cannister was weighed by an assistant. These measurements were used to construct cumulative dose response-curves. Recovery was documented by repeating MEFV curves and plethysmography immediately, 2, 5, and 10 min after the last 5-min step. Cold Air Challengu: CACh was performed using a respiratory heat exchange system described in detail." An inhaled air temperature of - urc (- 8° to - 12' was maintained during challenges. During cold-air inhalation for 4 min, the patients were continuously coached to hyperventilate at 75 percent of maximal voluntary ventilation, as calculated from FEV. and monitored by a calibrated rotameter. Spirometry and plethysmography were repeated 3 min after CACh was completed. Statistical Analysis

Statistical evaluation was performed using the Friedman test (multiple analysis of variance for nonnormally distributed data). Spearman's rank correlation test was employed to assess the association between variables. A p
Twelve patients completed the six different challenges, and none developed symptoms or signs of respiratory distress. The challenges were well tolerated except for a mild cough at the beginning of UNDW inhalations. Three patients had to be excluded because of inability to hyperventilate as required for CACh (two patients) or for lack of compliance (one patient). Baseline results from PFrs and maximum responses of PF to challenge tests are summarized in Table 1. There were no significant day-to-day differences in baseline PF assessed before the challenges. The MEF50 and MEF25 showed a trend toward lower baseline values. All 12 patients had a positive reaction to MCh challenge (Table 1). Following cUNDW challenge, four patients had a drop of less than 20 percent in FEV. (7, 9, 11, and 15 percent) and a drop of less than 40 percent in Sgaw (0, 19, 20, and 22 percent). The remaining eight patients had a mean drop in FEV. of 44.63 ± 19.0 percent and a mean drop in Sgaw of 66.75 ± 20.21 percent. Three ofthe "nonresponders" also had a drop in MEF25 <30 percent, the fourth had a drop in MEF25 of 62 percent. Mean MEF25 of the nine "responding" patients was 50.33 ± 20.30, and BT was 16.9±2.7°C and PT, 20.6±2.5°C. Following iUNDW challenge, three patients were "nonresponders" with a drop in FEV. of 6, 11, and 11.5 percent, respectively, and a drop in Sgaw of 0, 10, and 20 percent, respectively. All three were also Distilled Water ChaIenges in Asthmatic Children (EIchler et 81)

Table l-Baulta of Pulmonary Function Tatajrom 12 ABthmatic Children Before and After Six Different BronclaitJl Challenga· Postchallenge,

Baseline,

% Decrease from

FEV.: cUNDW 94.33± 18.25 iUNDW 9O.95±17.87 cJNDW 9O.16±13.80 iJNDW 9O.10± 18.85 CaCh 89.46 ± 19.14 MCR 93.96± 19.71 (p=O.974) Sgaw: cUNDW 58.24 ± 13.98 iUNDW 55.39±9.00 cJNDW 56.00±15.08 iJNDW 62.01 ± 15.26 CaCh 54.40 ± 12.72 MCR 54.36±17.28 (p=0.776) MEF50: cUNDW 59.15±22.99 iUNDW 58.68±21.83 cJNDW 57.68± 17.02 iJNDW 56.83±24.25 CaCh 51.52±22.00 MCR 57.34±21.67 (p = 0.748) MEF25: cUNDW 46.47 ± 26. 18 iUNDW 5O.19±23.14 cJNDW 43.72± 13.59 iJNDW 45.78±21.44 eaCh 49.62±33.35 MCR 54.80±21.88 (p=O.412)

34.79±23.06 38.18±24.48 27.42±22.95 21.19± 15.84 23.28 ± 19.56 PC20 (FEV.) 0.16±0.16mglml 50.28 ± 30.18 57.07±29.74 37.77±28.22 56.75 ± 27.31 39.88 ± 30.33 PC40(sGAW) 0.14±0.13 mglml 47.05±24.58 55.99±21.34 32.92±22.58 37.06±21.21 3O.31±27.92

% Predicted

Baseline

No. of Responders n112, % 8 (67)

9 (75)

6 (50) 5 (42) 6 (50)

12 (100) 8 9 5 6 8

(67) (75) (50) (50) (67)

12 (100)

39.20±21.67 56.57±15.68 30.80± 17.04 36.63± 19.98 35.87±29.76

*Results are mean± ISD. Positive response was defined as a ~20% decrease in FEV. or a ~40% fall in sGAW. The p values relate to differences in baseline values between challenges.

nonresponders in cUNDW challenge. The remaining nine patients had a mean drop in FEV. of50.67 ± 17.56 percent and in Sgaw of 66.78 ± 22.32 percent. However, all 12 patients had a ~30 percent drop in MEF25, and 10 patients in MEFSO. For the three nonresponders, decreases in MEF25 (MEF50) were 45 percent (26 percent), 62 percent (30 percent), and 67 percent (39 percent), respectivel)'. The BT was 17.2± 1.5°C and Pr, 23.3±2.2°C. Cumulative dose-response curves for individual patients were constructed by expressing the induced changes after each inhalation step as a percentage of individual maximum response. Thus, interindividual differences were reduced. The time course for the mean responses to iUNDW challenges is illustrated in Figure 1. Five of the nine responders to iUNDW inhalation had their maximum reaction after 2 to 6 min of inhalation. The remaining four patients took 20 •to 25 min to reach their maximum reaction. However, all responders showed a positive reaction (fall in FEV1 ~20 percent and in Sgaw ~40 percent) already within 2 to 6 min ofinhalation, ie, after inhalation of6.6± 1.14 ml water. There was a nonsignificant trend toward diminished responses for FEV1 and Sgaw after 14 to 20 min inhalation. Results from PFTs after 10 min recovery following cUNDW and iUNDW challenges are summarized in Table 2. Following cJNDW challenge, six patients had a drop in FEV. <20 percent (7.5, 11, 12, 14, 16, and 18 percent). The mean drop in FEV. of the remaining six responders was 40.67 ± 19.39 percent. Only five of our 12 patients had a drop in Sgaw >40 percent (mean:

sGAW

FEV 1 -t

100 -,-

100-

80

80

60

60

40

40

20 0

1. Mean cumulative doseresponse curves of FEV., Sga\lv, MEF50, and MEF25 during and after iUNDW challenge from nine asthmatic iUNDW-responders. Besuits are expressed as percent of individual maximum reaction. The iUNDW challenge was stopped after 30 min (vertical line). Recovery was recorded for 10 min. Vertical bars represent ISD. FIGURE

100

2

6

14

20 25 30

time (minutes)

2

5

10

MEF50

-/-

2

6

14

20 25 30

time ( minutes}

2

5

10

2

5

10

MEF25

-t 100-

eo

80

60

60

40

20

40 2

6

14 20 25 30

time(minutes)

2

5

10

20

2

6

14

20 25 30

tm. (mirut.s)

CHEST I 102 I 3 I SEPTEMBER, 1992

755

Table I-Baulta ofPulmorasf1J Function Tem[rom 12 AathrntJtic Children * Sgaw

cUNDW: b: 94.33± 18.25 r: 68.95±24.10t 5: 91.86 ± 20.77 iUNDW: b: 9O.95± 17.87 r: 71.76±24.42 5: 95.19± 15.47

MEF50

MEF25

DISCUSSION

58.24± 13.98 35.57±20.37 84.64±29.50

59.15±22.99 46.47±26.18 36.14±20.51t 30.95± 16.28* 59.02±26.55 49.22±28.81

55.39±9.00 58.68±21.83 5O.19±23.14 37.13± 18.24t 36.91 ±21.49t 31.44±22.16 88.59±32.39 63.85 ± 29.61 57.52±35.29

*b indicates before challenge; r, after 10 min of recovery; and s, after administration of 200 IA-g salbutamol. Data are presented as percent of predicted nonnal values. Results are mean ± SD. tp
65.60±20.62 percent). The maximum fall in FEVI occurred immediately after the inhalation (seven patients), after 2 min (one patient), after 5 min (three patients), or after 10 min (one patient). The BT was 11.8±3.rc and PT, 8.5±0.2°C. Following iJNDW challenge, seven patients had a drop in FEV I <20 percent (4, 9, 11, 11, 11.5, 15, and 16 percent). The mean drop in FEV. of the remaining five responders was 46.40 ± 20.55 percent. Six patients had a drop in Sgaw <40 percent (15, 17, 21, 21, 34, and 37.5 percent). In the remaining six patients, fall in Sgaw was 64.50± 19.92 percent. Maximum fall in FEV. occurred after 2 min (three patients), after 6 min (one patient), after 14 min (three patients), after 20 min (three patients), or after 30 min (two patients). The BTwas 9.5±3.1°C, and PT, 11.0±3.3°C. Six patients responded to CACh, six were nonresponders with respect to FEV1 (drop in FEV1 1.5, 6.5, 10, 12, 13, and 18 percent, respectively); mean maximal drop in the remaining six patients was 37.00± 11.95. With respect to Sgaw and MEF25, eight patients were responders with a mean maximal drop of 52.63±23.18 percent and 51.88± 18.43 percent, respectively. There was no significant association between baseline values and changes induced by any of the six different inhalation tests (p>0.1). There was a signi6cant correlation between cUNDW- and iUNDW-induced maximum responses for FEV I (r.=0.69, p=0.02) and Sgaw (r.=0.59, p = 0.04), though not for MEF50 (r. =0.35, p =0.36) or MEF25 (r. = 0.39, p = 0.27). No significant correlation was obtained between UNDW- and JNDW-induced maximum responses for FEV., Sgaw, MEF50 and MEF25 (neither continuous nor intermittent inhalations); also, there was no correlation between responses to distilled water inhalations and CACh (p>0.1). There was a significant negative correlation between PC20 (MCh) and maximum response for FEV. during 758

iUNDW (r.= -0.78, p=O.OO45), and to a lesser degree, after cUNDW (rs = - 0.57, p = 0.047), but not during/after cJND~ iJNDW or CACho In our study, MCh was used as the criterion standard for measuring nonspecific bronchial hyperresponsiveness in asthmatic children. Against this standard, the sensitivity of cUNDW was 67 percent and that of iUNDW was 75 percent. This is in broad agreement with other authors who reported sensitivities of between 52 and 68 percent for iUNDW challenge. 4,5,9 In contrast, Anderson et al3 found an iUNDW sensitivity of 100 percent. We have no ready explanation for the discrepancy between the latter report and our present results; age differences in study groups and the relatively small number of children we have studied may have played a role; the volumes ofdistilled water employed in our study were equal to or greater than those used by previous investigators, including those of Anderson et al. 3 Previous studies have shown that in children with asthma, obstruction is primarily localized in peripheral airways}o.ll This concept is supported by our present data showing low baseline levels for MEF25 and MEF50, but not for FEV h since MEF25 and MEF50 reflect predominantly peripheral airway resistance. l i During iUNDW challenges, 100 percent and 83 percent of patients had a 2: 30 percent fall in MEF25 and MEF50, respectively. This may indicate that inhalation ofiUNDW affects primarily peripheral airways. Perhaps, for iUNDW challenges, a fall in MEF25 or MEF50, rather than a decrease in FEV1, should be considered the variable of choice for the diagnosis of asthma. In contrast to previous authors,4,9 we found a significant association between results from MCh and iUNDW challenges in our patient group. Since we have only measured a biologic response, this association is still compatible with the current concept that these tests induce bronchoconstriction by different pathophysiologic mechanisms. Methacholine induces bronchospasm by acting on acetylcholine receptors in bronchial smooth muscle, but there are a number of possible mechanisms by which inhalation of UNDW may induce bronchoconstriction, including osmotic changes on the epithelial surface,13 stimulation of airway mast cells with subsequent release of mediators,14,15 or alterations in bronchial epithelial permeability with stimulation of subepithelial neuroreceptors. 16.17 Sterk and Be}18 have recently suggested that bronchial hyperresponsiveness in asthma is characterized by two distinct abnormalities: a leftward shift of the dose-response curve for inhaled stimuli, termed hypersensitivity, which can be quantified by PC20 valDistilled Water ChaIenges In Asthmatic Children (Eichler et 81)

ues, and an increase in the maximal response, reflecting the extent to which the airways can narrow when exposed to high doses of inhaled stimuli. These authors 18 raised the question whether measures of the shift in the dose-response curve (eg, PC20) are the most relevant parameters for diagnosing bronchial hyperresponsiveness. Macklem 19 argued that the clinical severity of airway obstruction is also a function of the maximal attainable degree of constriction. Hence, determining the maximal response to a standardized bronchial provocation procedure in addition to sensitivity measurements may be useful for characterizing bronchial hyperresponsiveness. I8-23 Several authors21 ,24,25 found a plateau in the doseresponse curves for histamine and methacholine in normal subjects and in some patients with mild asthma. The mechanisms responsible for this limited maximal airway narrowing are not entirely understood. 18.26 A plateau response was also shown for CACh20 and exercise provocation in asthmatic children. ~ However, conventional challenge tests using histamine or cholinergic drugs21 ,25 fail to elicit a reaction plateau in adult asthmatics, presumably because the plateau lies beyond the limits ofsubjectively tolerable obstruction. 20 In our group of patients, there was a clear reaction plateau during iUNDW challenge (Fig 1). This may. be another potential advantage ofiUNDW challenges. However, our findings need further confirmation from other studies with larger numbers of subjects, before maximal response to iUNDW challenge could be recommended as an accurate measure of airway responsiveness. The time course for the responses to iUNDW challenges shows that in responding asthmatic children, clinically relevant bronchoconstriction (fall in FEV. ~20 percent and in Sgaw ~40 percent) was obtained within the first few minutes of inhalation; extended inhalations did not produce signi6cantly more obstruction. This is in agreement with recent results by Smith and Anderson. 28 Therefore, in children, test protocols with inhalation periods of no more than 6 to 10 min (corresponding to 6.6 to 11.0 ml water inhalation) appear sufficient when assessment of bronchial sensitivity (as opposed to maximal airway narrowing) is requested. In all responders to cUNDW challenge, a positive

test result was obtained immediately after inhalation of UNDW was stopped, even though some of these patients had a further increase in response later during the recovery period. Hence, a single measurement of PF immediately after termination of cUNDW challenge appears sufficient to obtain a positive test result without loss of sensitivity. As discussed above, data from iUNDW challenge indicate that the duration of continuous inhalation might be shortened to 5 to 6

min, as was initially described in adults. 2 This would make cUNDW a very rapid test procedure. Bronchoconstriction persisted for at least 10 min after iUNDW or cUNDW inhalations were stopped (Table 2 and Fig 1). Therefore, patients responding to UNDW challenges should be given inhaled bronchodilating agents after termination of the test; this is indicated because late reactions have been described. 29 There was a significant correlation between maximum responses to cUNDW and iUNDW challenges for FEV. and Sga~ but not for MEF50 and MEF25. It is unclear whether this is solely attributable to the large intraindividual variability in MEF50 and MEF25 recordings or indicates different deposition patterns in peripheral airways for the two different inhalation procedures. Jet nebulizers are available in most doctors' offices and hospitals; thus, JNDW challenge would be an inexpensive and rapid method for broad screening tests for asthma. We found no significant correlation between results from UNDW and JNDW challenges. In our patient group, the sensitivity of JNDW challenges was less than that of UNDW Also, no correlation was found between PC20 and either iJNDW or cJNDW challenges. These data confirm Allegra and Bianco's assumption2 that nonultrasonic nebulizers do not yield results comparable to UNDW challenges. At present, JNDW is not a candidate for routine provocation testing. The CACh produced a positive response in 50 percent and 67 percent of patients for FEV1and Sga~ respectivel~ Similar results were reported by Reisman et al,30 who found a sensitivity of 61 percent when a positive response for CACh was defined as a ~20 percent decrease in FEV•. However, our results differ from the sensitivity of 100 percent reported by Zach et al,8 although we used the same testing procedure. The difference may reflect different patient selection criteria: 83 percent of the patients studied by Zach et aI, 8 but only 50 percent of our asthmatic children gave a positive history of wheezing after exercise. Another reason may be that a 20 percent decrease in FEV. was too stringent a requirement for a positive response. Accepting a 10 percent decrease in FEVI would increase the sensitivity of CACh to 75 percent in our patients. However, compared to UNDW challenge, CACh was less well tolerated because it requires maximum patient effort and cooperation and is thus less suitable for children. No association was found between results from UND~ eACh, and MCh. This is in agreement with previous reports,4.31 and again, suggests a different pathophysiologic mechanism of action for each challenge. In conclusion, inhalation of UNDW proved a well CHEST I 102 I 3 I SEPTEMBER. 1992

757

tolerated, easily administered test procedure for children, requiring a minimum of patient and investigator effort. The iUNDW may be slightly more sensitive than cUNDW challenge but not as sensitive as MCh and thus cannot be relied upon as the only test for detecting bronchial hyperresponsiveness. A reaction plateau was demonstrated during iUNDW challenge. Inhalation of UNDW may affect predominantly small to medium-sized airways. Future clinical trials will establish the definition of a positive response to UNDW that is best stuited to obtain maximum sensitivity and specificity for the diagnosis of bronchial asthma. ACKNOWLEDGMENT: We thank Brigitte Pomianek and Marianne Koller fOr their technical assistance, and H. Herber, Ph. D., for the PRIST and BAST studies.

REFERENCES 1 Hopp RJ, Christy J, Bewtra AIC, Nair NM, Townley RG. Incorporation and analysis of ultrasonically nebulized distilled water cbalIenges in an epidemiologic study of asthma and bronchial reactivi~ Ann Allergy 1988; 60:129-33 2 Allegra L, Bianco S. Nonspecific broncbo-reactivity obtained with an ultrasonic aerosol of disilled water. Eur J Respir Dis 1980; 61 (suppll06):41-49 3 Anderson SD, Schoeffel RE, Finney M. Evaluation of ultrasonically nebulised solutions for provocation testing in patients with asthma. Thorax 1983; 38:284-91 4 Galdes-Sebaldt M, McLaughlin FJ, Levison H. Comparison of cold air, ultrasonic mist, and methacholine inhalations as tests of bronchial reactivity in normal and asthmatic children. J Pediatr 1985; 107:526-30 5 Bascom B, Bleecker ER. Bronchoconstriction induced by distilled water: sensitivity in asthmatics and relationship to exerciseinduced bronchospasm. Am Rev Respir Dis 1986; 134:248-53 6 DuBois AS, Botelho SY, Comroe JH. A new method for measuring airway resistance in man using a body plethysmograph: values in normal subjects and in patients with respiratory disease. J Clio Invest 19S6; 35:321-35 7 Zapletal A, Samanek M, Paull: Lung function in children and adolescents: methods, reference values. Herzog H, ed. Progress in respiration research. Basel: Karger, \'0122; 1987 8 Zach M, Polgar G, Kump H, Kroisel E Cold air cbalIenge of airway hyperreactivity in children: practical application and theoretical aspects. Pediatr Res 1984; 18:469-78 9 Foresi A, Mattoli S, Corbo GM, Polidori G, Ciappi G. Comparison of bronchial responses to ultrasonically nebulized distilled water, exercise, and methacholine in asthma. Cbest 1986;

13

14 15

16 17

18

29 20 21

22 23

24

25

26

27 28

29

90:822-26

10 Hill OJ, Landau L1, McNicol KN, Phelan PD. Asthma: the physiological and clinical spectrum in cbildbood: respiratory function studies in its assessment. Arch Dis Child 1972; 47:87481 11 Zapletal A, Motoyama EK, Gibson LE, Bouhuys A. Pulmonary mechanics in asthma and cystic fibrosis. Pediatrics 1971; 48:6472 12 Hyatt RE, Rodarte JR, Mead J, Wtlson TA. Changes in lung

758

30

mechanics: flow-volume relations. In Macklem Pr, Permutt S, eds. The lung in the transition between health and disease. New York: Marcel Dekker Inc, 1979; 73-112 Erschenbacher WL, Boushey HA, Sheppard D. Alteration in osmolarity of inhaled aerosols causes bronchoconstriction and cough, but absence of a permeant anion causes cough alone. Am Rev Respir Dis 1984; 129:211-15 Rimmer J, Bryant DH. Effect of hypo- and hyper-osmolarity on basophil histamine release. Coo Allergy 1986; 16:221 Bascom R, Kagey-Sobotka A, Lichtenstein LM. Histamine release (HR) from human basophils and mast cells using hypoosmolar solutions. Am Rev Respir Dis 1985; 131:30 Hogg JC, EgglestOD PA. Is asthma an epithelial disease? Am Rev Respir Dis 1984; 129:207-OS Borland CDR, Chamberlain AT, Higenbotom 1W The effect of ultrasonically nebulized distilled water on lung epithelial permeability. Chest 1983; 83:935-36 Sterk PJ, Bel EH. Bronchial hyperresponsiveness: the need for a distinction between hypersensitivity and excessive airway narrowing. Eur Respir J 1989; 2:267-74 Macklem PT. Bronchial hyporesponsiveness. Chest 1985; 87: 15859 Zach MS, Polgar G. Cold air challenge ofairway hyperreactivity in children: dose-response interrelation with a reaction plateau. J Allergy Coo Immunoll987; 80:9-17 Woolcock AJ, Salome CM, Yan K. The shape of the doseresponse curve to histamine in asthmatic and normal subjects. Am Rev Respir Dis 1984; 130:71-75 Sterk PJ, TImmers Me, Dijbnan JH. Maximal airway narrowing in humans in vivo: histamine compared with methacholine. Am Rev Respir Dis 1986; 134:714-18 Cockcroft DW Nonallergic airway responsiveness. J Allergy Clin Immunoll988; 81:111-19 Sterk PJ, Daniel EE, Zamel N, Hargreave FE. Limited bronchoconstriction to methacholine using partial Row-volume curves in nonasthmatic subjects. Am Rev Respir Dis 1985; 132:272-77 Townley RG, Ryo UY, Kolotkin KM, Kang B. Bronchial sensitivity to methacholine in current and former asthmatic and allergic rhinitis patients and control subjects. J Allergy Clin Immunol 1975; 56:429-42 Moreno RH, Hogg Je, Pare PD. Mechanics ofairway narrowing. Am Rev Respir Dis 1986; 133:1171-80 Silverman M, Andenon SD. Standardization of exercise tests in asthmatic children. Arch Dis Child 1972; 47:882-89 Smith CM, Anderson SD. Inhalation challenge using hypertonic saline in asthmatic subjects: a comparison with responses to hyperpnoea, methacholine and water. Eur Respir J 1990; 3:14451 Mattoli S, Foresi A, Corbo GM, Valente S, Patalano F, Clappi G. Increase in brondUal responsiveness to methacholine and late response after the inhalation of ultrasonically nebulized distilled water. Chest 1986; 90:726-32 Reisman J, Mappa L. de Benedictis F, McLaughlin J, Levison H. Cold air challenge in children with asthma. Pediatr Pulmonol

1987; 3:251-54 31 Lemire TS, Hopp RJ, Bewtra AK, Nair NM, Townley RG.

Comparison of ultrasonically nebulized distilled water and coldair hyperventilation challenges in asthmatic patients. Chest

1989; 95:958-61