Consecutive preschool children referred to the Pediatric Pulmonary Clinic, Meyer Children's Hospital, Rambam Medical Center, Haifa, over a 6-month period were recruited. Of 62 families offered participation in the study, seven refused. None of the children had experienced spirometry previously. Inclusion criteria were: 2.5–6.5 year-old children who were asthmatic according to GINA guidelines 15 with recurrent episodes of wheeze, cough and/or shortness of breath with clinical response to bronchodilator; normal chest auscultation and FEV₁ >75% of predicted for healthy preschool children 9 after saline inhalation. Exclusion criteria were: presence of other chronic respiratory conditions; emergency room visit in the past three months; respiratory infection in the past month; oral or inhaled steroids or other anti-inflammatory medication taken in the last week; bronchodilator taken within 24 hours prior to the test.

The Rambam Medical Center Ethics Board approved the study. Parental consent was obtained for each child.

Tests were performed in a designated room at the Pediatric Pulmonary Unit, Meyer Children's Hospital, Haifa, Israel. A parent and the investigating team (a pediatric pulmonary physician, respiratory physiologist and technician) were present throughout the test. MCT was performed according to published guidelines, 3, with doubling doses of fresh Methacholine solutions (0.06 to 8.00 mg/ml) dissolved in saline. Solutions were driven by compressed air of 5 l/min flow (giving a mean output of 0.4 ml/min), and nebulized using a Hudson nebulizer (Hudson RCI, Temecula, CA, USA). Inhalations were performed using a facemask while the child was sitting up straight and breathing normally. Nebulized Methacholine was inhaled for 2 minutes, with 5-minute intervals between doses, until the maximal concentration or the end point was reached. To ensure safety in light of the risk of airway closer, the MCH increment was only half the usual amount when transient wheeze or cough was noted, keeping in mind that the accumulative dose is affected by this manipulation. Oxygen saturation and heart rate were monitored continuously by pulse oximetry (Biox 3700e; Ohmeda). A single observer (LB) performed auscultation for 20 seconds over the trachea and two zones of both lungs (upper front and lower back) according to Springer et al. 7

The following indices were considered "end of test": appearance of audible wheeze, a fall of ≥5% in O₂-saturation, or an increase of ≥50% in respiratory rate and/or heart rate 7. At the "end of test", spirometric measurements were performed, followed by administration of nebulized Albuterol (2.5 mg).

Forced expiratory flow volume (FEFV) curves were measured with a ZAN100 commercial spirometer (ZAN Messgeraete GmbH, Oberthulba, Germany). Calibration was performed before the testing sessions. The curves were monitored on the computer screen to ensure best effort. Results were corrected to BTPS conditions. The software included an interactive animated computer game (SpiroGame^®) set by targets of the FEFV maneuver, combining forced inhalation preceding forced expiration, peak expiratory flow rate (PEFR) and forced vital capacity (FVC) with emphasis on prolonged expiration. 8 The targets were the extrapolated values derived from comparative data from older children, corrected for height. 16 An experienced pulmonary technician instructed each child how to operate the game. Teaching time was limited to 15 minutes. On-line rejection of curves was based on visual inspection for "non-cooperation" errors and included: poor effort; incomplete expiration; cough; glottis closure. Curves had to show a rapid rise to peak flow, and gradual, smooth decline of flow down to residual volume. Baseline maneuvers were repeated to visually obtain best possible efforts on at least 3 technically acceptable FEFV curves. After obtaining baseline spirometry, MCT was performed. A duplicate spirometry set was performed immediately after auscultation. PC₂₀-FEV₁ was determined off line by the provocative concentration that reduced FEV₁ by 20% from baseline. PC values were log-transformed before statistical analyses. Spirometry indices included FVC, FEV₁, PEFR, forced expiratory flow at 50% FVC (FEF₅₀), FEV₁/FVC ratio.

Three baseline spirometry curves were analyzed for acceptability criteria according to ATS/ERS guidelines 1718 and in comparison with similar data for preschool children 1119. These included: a) "Start of test" criteria: time to peak expiratory flow and backward extrapolated volume (Vbe) b) "End of test criteria": described by "total expiratory time" and the ratio of "no change in expiratory volume" to "total expiratory time" c) reproducibility (coefficient of variation) of the three baseline curves, calculated as SD/mean*100.

After inhalations, the curves were inspected visually online, and were analyzed offline in relation to baseline using paired t-test. Differences were considered significant when p < 0.05. The level of agreement between the dose at end of test and the dose of PC₂₀ were compared by Bland and Altman analysis (20).

Children's response to MCT (n = 36) is summarized in Figure 1 and Table 3. Average test time to reach PC₂₀-FEV₁ was 29 ± 11 minutes, while for PCW it was 41 ± 10 minutes (not including bronchodilator administration) (p < 0.001). The end point of the challenge was determined by the pediatric pulmonologist as positive in 35/36 children. One child did not display any of the determined criteria for PCW up to 8 mg/ml and was considered to have no BHR. The mean (± SD) concentration at PCW for the 36 children was 4.26± 3.31 mg/ml. Wheezing at the end point was observed in 26/36 children and in 9/36 the test was ended before the appearance of wheeze due to either oxygen desaturation or tachypnea accompanied by audible long expiration. Mean increase in heart rate at PCW was 25.5 ± 11% (range 10–42%); respiratory rate increased by 30.0 ± 21.1% (range 0–42%) and SaO₂ decreased by 6.3 ± 2.7% (range 2.3–10.3%).

PC₂₀-FEV₁ occurred at a mean concentration value of 1.96- ± 1.83 mg/ml. The one child who did not respond to MCH of up to 8 mg/ml by PCW (negative BHR) did not show a fall of 20% from baseline FEV₁ value either. The other 35 children exhibited a fall of 20% in FEV₁ from baseline values in response to MCH ≤8 mg/ml (Figure 1 and Table 3). A representative set of FEFV curves from a single patient that includes the predicted curve, baseline, PC₂₀-FEV₁ and end of test curves is shown in Figure 2.

At PC₂₀-FEV₁ there was a mean increase in heart rate of 13.5 ± 11.0%, respiratory rate increased by 15.4 ± 15.8% and SaO₂ decreased by 2.4 ± 2.1% from baseline level. These changes were significantly lower than those found at PCW (p < 0.01 for three parameters). The appearance of PC₂₀-FEV₁ occurred 2 concentrations earlier than PCW in 5 children, 1.5-concentrations earlier in 3 children, one concentration earlier in 17 children, 0.5 concentrations earlier in 3 children and at the same concentration as PCW in 7 children (Figure 1). The effects of MCH on the spirometry parameters are presented in Table 4. At PC₂₀-FEV₁, parameters were moderately decreased, while at end point, test parameters were markedly reduced. The severity of FEV₁ reduction at PCW was variable, ranging from 30.8 to 68.2% of baseline. The level of agreement between the dose at end of test (PCW) and the dose at PC₂₀20 is presented in Figure 3. Dotted lines represent 95% coefficient of variation values.

Bronchodilators improved FEV₁ by 43 ± 29% from PCW values and all respiratory symptoms disappeared shortly after bronchodilator administration.

We found that PCW occurred in our group at a mean concentration of 4.26 ± 3.31 mg/ml. PCW values in our study were much higher than the PCW (0.4 mg/ml) reported by Springer et al 7. The difference may be attributed to inclusion of more severe asthmatics in their study group.

We found that PC₂₀-FEV₁ occurred at a lower concentration than PCW in most subjects. This finding is in agreement with several other studies comparing PCW detection to PC₂₀-FEV₁ in school age children. 456714. However, in none of these studies were spirometry measurements carried out to the point of wheeze. We expected to find a good correlation between the two tests (PCW = 1.2195*PC₂₀-FEV₁ + 0.0288; R² = 0.9733; p < 0.005), yet the Blant and Altman analysis revealed that in children with higher mean provocation PCW dose (≥6 mg), the level of agreement between the methods was low, reflecting higher sensitivity of the PC₂₀ method, especially in mild airway reactivity (Figure 3).

We found that at PCW, FEFV curves visually seemed to be smaller and all parameters were reduced simultaneously (Figure 2), with a highly significant reduction in flows and volume parameters. The reduction in curve was gradual in most children, accompanied by an increase in respiratory symptoms (Table 2), and responded to bronchodilators, and hence was not considered to reflect fatigue. To further strengthen this point a representative curve of one child illustrating, a poor effort performed at teaching process vs. end of test curve is shown in figure 4. The poor-effort curve did not fulfill start of test criteria and is round while the "end of test curve" has an obstructed shape.

The reduced FVC and flows are most likely due to a severe degree of airway narrowing involving small to medium airways that may be accompanied by air trapping, partial closure of airways and elevation in FRC. Reduced FVC may also be due to increased glottic narrowing due to MCH irritation 2627, but the flow volume curve was not suggestive of upper airway obstruction (trimmed PEFR). Alternatively, the upper airways response to methacholine may contribute to the increase in total respiratory resistance 27. This pattern occurred in some cases before appearance of wheeze or other clinical end-points. Indeed, in 9/36 subjects, the test was terminated due to oxygen desaturation or tachypnea rather than wheeze. Similar to our results, Sprikkelman et al 28 reported that wheeze was detected in only 33% of 15 school-age asthmatic children at PC₂₀-FEV_1, and Springer et al 7 terminated the test without the presence of wheeze in 19.2% of young children. In this respect we would argue that FEV1 does make a contribution beyond simply asking the subject if they wheeze. Novitzki et al (4) found in 5–8 year-old children that FEV₁ is decreased by 33.3 ± 7.4% at PCW. Spence et al 29 reported a mean fall of 51 ± 14% from baseline FEV₁ when wheeze appeared in their asthmatic older subjects. Our results strengthen these prior findings, and suggest that spirometric PC₂₀-FEV₁ may be achieved with inhalation of lower MCH concentrations than those used to achieve wheeze.

Measuring PCW during tidal volume breathing has the advantage that no active cooperation on the child's part is needed. Therefore the success rate of PCW is higher than spirometry (44/55 children). However, using PC₂₀-FEV₁ (or PC₂₀-FEV_0.5) can preclude inhalation of higher concentrations of MCH used to achieve wheeze, leading to alarmingly diminished flows found at PCW and a significant shortening of test time relative to PCW.