Histamine Challenge in Young Children Using Computerized Lung Sounds Analysis* Raphael Beck, M.D., F.C.C.R; Ursula Dickson, M.B.; Mark D. Montgomery, M.D., F.C.C.R; and Ian Mitchell, M.B., F.C.C.R The use of the histamine challenge test (HCT) for the diagnosis of asthma has so far been limited to older children who can perform spirometry consistently. Recently, wheeze detection by tracheal auscultation with analog recording into a tape recorder has been utilized in young children in place of spirometry. Wheezing can also be identi6ed using computerized lung sounds analysis (LSA) by a typical pattern on spectral analysis. Our aim was to develop a practical computerized system in which the response to histamine could be identified in an objective manner and documented on hard cop~ Lung sounds were recorded with a Hewlett-Packard HP 21050A contact sensor placed over the right upper anterior chest. Sounds were amplified, band-61tered (SO to 2,000 Hz), and digitized at a sampling rate of 5.5 kHz into a Macintosh SE computer, and spectral LSA was performed. To validate our method, six older children (ages 9 to 16 years) with mild or moderate asthma underwent HCT. The identi6cation of typical wheezing pattern (discrete, high-amplitude power peaks) on LSA was compared to 20 percent fall in FEV1 (PC20) and symptoms (cough, wheeze, chest tightness). In five children, the histamine concentration required to produce the characteristic wheezing pattern on LSA was half that required to
produce a 20 percent fall in FEV•. In the sixth patient, wheezing on LSA and PC20 occurred at the same histamine concentration. To determine the technique's applicability . to young children, we then studied six young asthmatic children (age 2 to 5 years). All children showed the wheezing pattern at a histamine concentration of 25 percent or 50 percent (one or two steps prior) to that producing symptoms (cough, wheeze, chest tightness) or wheezing on tracheal auscultation. Six age- and sex-matched nonasthmatic children (control subjects) did not show this pattern on LSA and had no symptoms or tracheal wheeze with HCT. We describe a sensitive method enabling application of HCT to young children who are unable to perform spirometry. This method is as sensitive as, and often more sensitive than, conventional PC20 with spirometry or tracheal auscultation. (Chest 1992; 102:759-63)
(or methacholine) challenge is a widely H istamine used, sensitive, and specific method for the
rate and reproducible results on spirometry, limiting applicability of this test to older children and adults. An alternative technique recently described is the use of tracheal auscultation to monitor response to histamine." Histamine is given until wheezes are detected (pew) by a stethoscope or microphone placed over the trachea. Modern computerized lung sounds (LS) recording and analysis techniquess .6 enable easy and quick recognition of wheezes by combining "on line" auscultation with appearance of a typical pattern of discrete, high-amplitude sound peaks on spectral analysis. 5 .6 By using a sensitive recording device coupled to an amplifier and filter, much softer sounds than can be detected by ear can be demonstrated. The wheezing pattern during histamine challenge can therefore be identified before wheezing becomes obvious. Our aim was to develop a simple-to-perform, practical tool for performing histamine challenge in young children and infants that would offer objective results.
diagnosis and evaluation of asthma. 1 Nonspecific airway hyperreactivity is demonstrated by inhaling increasing concentrations of nebulized histamine or methacholine until a 20 percent drop in FEV 1 occurs (PC20). Extensive experience over the past 30 years has shown the test to be safe, easily performed, and reproducible, and to correlate well with degree of asthma severi~2 Some authors3 prefer histamine over methacholine because of its shorter duration of action and its being a physiologic mediator of bronchoconstriction. This test is in widespread use in most pulmonary function laboratories, but accuracy of results depends on the patient's ability to perform technically acceptable and reproducible spirometry. Young children, usually under the age of six to eight years, are not able to consistently perform the forced vital capacity maneuver necessary to generate accu-
*From the Division of Respiratory Medicine, Department of Paediatrics, University ofCalgary, The Alberta Children's Hospital~ Calgary, Alberta, Canada. This study was supported in part by a University of Calgary Research Grant. Manuscript received September 23; revision accepted January 2.
=
=
RCT histamine challenge test; LS lung sounds; LSA = lung sounds analysis; Pe20 = provocation concentration of histamine producing 20 percent fall in FEV.; FFf=fast Fourier transform; Pew = provocation concentration of histamine producing tracheal wheeze; PCLsA = provocation concentration of histamine producing a positive result on LSA
MATERIALS AND METHODS
Histamine Challenge This was perfonned by a standard protocol, using tidal breathin~ of continuously nebulized histamine solution, adapted from CockCHEST I 102 I 3 I SEPTEMBER, 1992
759
IV; dry roBing seal spirometer) befOre starting. after Inhaling
nebulized 0.9 percent saIiDe solution (baseline) and after each dose ofhistamlne. until a 00 percent drop from baseline FEV, occurred. or to a maximal histamine dose ofa mgtml.
....._+-......- _...._ -....
Sound Recordlng and Analym Sy6tem8
~~
Figure 1 scbematically depicts the recording system. Sounds were recorded via a Hewlett-Packard HP 210s0A contact sensor placed over the right upper anterior chest. Placement site was chosen so as to be over a large airway as far away from heart sounds as possible. Sounds were ampli6ed (x 10 to x 10(0). band-filtered between 50 and 2,000 Hz (fouN:hannelinstrumentation amplifier/ &Iter. custom built), and digitized (8-bit anaIog-to-digltal converter) at a sampling rate of 5.5 kHz Into a Macintosh SE computer. P1aybaclc screening of sound segments and wave-form and spectral analysis by fast Fourier transfOrm (FFI') were perfOrmed using the SoundWave software on the recorded sounds prior to the next histamine dose. A histamine response was considered positive wben a typical wheezing pattern of high amplitude. discrete power spikes"-" was identified on spectral (FFI') analysis.
•
FIGURE 1. Schematic description of computerized lung sounds recording and analysis system (see text for details). croft' Subjects inhaled the histamine fOr 2 min. starting at a dose of0.03 mgtmI. Histamine concentration was doubled progressively every 5 min to a maximal concentration of a mgtml (Ie, the new dose was started 3 min after the previous one finisbed~ ~ used a Hudson Up-Draft Nebumlst 1700 nebulizer driven with 4 Umln of air (mean mass diameter, 1.5 ....m (85 percent) geometric standard deviation. 1.69). Older children performed spirometry (Gould 5000
.........
sell.. O.9a
~
J
II
l
j
I~
lA \
...
\f'-. ~
Zl
)$'
\ k '\ v 'In
IUS
.n
U1
Zl
HI.taml".O 125 mg/ml
10n
136$
IOlt
lUI
tU.llmlnl '2 0 m,/ml
IA
IV ~ VIA
"
f
s
j
\'\
... f.-.-,
~Ll .51
Zl
II U\
A
V
h
1029
r~'\ .51
Ins
--.
""IA~
lttzl
....... ·4
se".. O.9a
J j
$en. .
'Y\
I"
Zl
'n
0."
\ ....
"'-
."
)$'
'JU
Zl
""- .S1
.n
Hfatemln.O .25 mg/mt
! FIGURE 2. Ell8II'Iples of positive Her by computerized LSA. In each case, note the appearance of discrete high-amplitude power peaks. typical of wheezing lung sounds.
780
1
j
II
r-
~
151
~
II
\J V V'""'"",- ..J\...,.....N au
IOU
II
V\,..
VV
Zl
:In
A
6U
""'V\.-... 10ft
~
IUS
Table I-Patient CharGCtefVtica and BeaultI ofPhaae 1: VGlitltdion ofMetlwd COIIIpGring Con~ BCT and Computeriz,ed LSA Patient
Age/Sex
Asthma Severity
l6IF
Mild
2
l3IF
Moderate
3
141M
Moderate
4 5
111M 9IM
Mild Moderate
6
111M
Mild
Treatment 8eclomethasoQe, salbutamol Theophylline, beclomethasone, salbutamol Theophylline, salbutamol Salbutamol Theophylline, beClomethasone, salbutamol Beclomethasone
PC20, mwml Histamine
LSA, mwml Histamine
0.5
0.25
0.03
0.03
0.06
0.03
2.0 0.25
1.0 0.125
4.0
2.0
lMng Sounds Analysis Method
Study Design
After recording and storing of LS, initial screening of expiratory LS was performed in the time interval until the next LS recording was due. This was performed manually by the researcher, using the SoundWave software, according to a previously described method. 6 Each expiration was divided into 200-ms sequential sound segments. These were analyzed in succession, starting at the beginning of the expiration, by performing spectral analysis (FIT, 1,024 data points) Oil each of the segments. The resulting spectra (Fig 2) were e~ined by the researcher for the presence of peaks typical of wheezes, according to previously published criteria. 6 The main criteria for wbeeze identmcation were as follows: (1) discrete, narrow peak of power; (2) the peak is distinctly separate from its surrounding sound spectrum; (3) peak amplitude at least threefold larger on one side and twice larger on the other side than the base it arises from; and (4) at least one aspect of the peak has to show a sharp rise, whereas on the other side, the slope of the peak may be graded. If, on playback, a wheezing sound was heard, that section of LS was analyzed; otherwise one or two expirations were chosen at random for analysis. This method enabled analysis of one to two expirations over the 5 min before the next LS recording was performed. Systematic analysis of all recorded sounds was performed at a later date and served to confirm the findings of the screening process, or identify wheezing patterns which were missed.
Phase 1: \hlidation of Method: This was done by comparing appearance of wheezing pattern on LS analysis (LSA) to PC20 by spirometry. Six children attending the Alberta Cbildrens Hospital Asthma Clinic (age 9 to 16 years, four male subjects), with known mild or moderate asthma, as judged by the severity of their symptoms and the amount and type of medication taken (Table 1), were studied. Following histamine inhalation, spirometry was performed, followed by computerized LSA, so that PC20 and PCLSA could be compared. The challenge was terminated when a 20 percent drop in FEVI occurred or when the suhject became symptomatic (cough, wheeze, chest tightness).' Phase 2: ApplicabilUy to Younger Children: Here, computerized LSA was compared to tracheal auscultation· during histamine challenge in astJimatic children and in children with no asthma (control subjects). Six children who were being followed up in our Asthma Clinic (age two to five yean, three girls), with mild or moderate asthma (Table 2) were studied. They were unable to perfonn spirometry and had the Her performed according ~o the above protocol, with computerized LSA and tracheal ausculta&n performed after each histamine dose. Six age- and sex-matc~ children without asthma (control subjects) were studied in an identical manner. The test was terminated when wheezing was identified on tracheal auscultation· or when clinical symptoms (cough, wheeze, chest tightness) occurred.
Table I-Patient ClaarGCtefVtica and BeaultI ofPhaae 2: Applictmon to Yourag ~~nm COIIIpGring HCT by Tracheal Auacultation arad Computeriz,ed LSA
Patient
Age/Sex
Asthma Severity
1
3IF
Mild
2
5IF
Mild
3
4IF
Moderate
4
9/M
Moderate
5
31M
Mild
6
31M
Moderate
7-12
2-5
Control subjects
Treatment Salbutamol, astemizole Salbutamol, beclomethasone Salbutamol, cromoglycate Salbutamol, beclomethasone Theophylline, salbutamol Salbutamol, cromoglycate, beclomethasone None
LSA, mglml Histamine 0.125
Tracheal Auscultation, mglml Histamine 0
2.0
4.0
0.125
0.125
0.03
0.03
0.5
1.0
0.06
0.06
>8.0
>8.0
CHEST I 102 I 3 I SEPTEMBER, 1992
781
Ethics
The study was approved by the University of Cal~ary Conjoint Medical Ethics Committee and the Alberta Children·s Hospital Research Committee. Informed consent was wven by all parents and by the older children (phase 1) prior to the study. RESULTS
Figure 2 shows four examples ofa positive histamine challenge, with appearance ofhigh-amplitude discrete power spikes on spectral analysis, typical ofa wheezing pattern. 5•6 Table 1 summarizes results of phase 1. All subjects had a positive histamine challenge, with PC20 ranging from mild (Nos. 4, 6) to severe (Nos. 2, 3) airway reactivity. In all subjects, a typical wheezing pattern on LSA was present at the time 20 percent fall in FEV. occurred. Moreover, in five of six subjects, LSA was positive for wheezing one dilution prior (at one half the histamine concentration) to the PC20. Table 2 summarizes results of phase 2. Five of the six children had a positive histamine challenge by tracheal auscultation," with PCw ranging from mild (No.2) to severe (Nos. 4, 6) airway reactivity. In two of them, typical wheezing pattern on LSA was detected one step prior (at one half the histamine concentration) to PCw. One subject (No.1), had a typical wheezing pattern on LSA at a histamine dose of 0.125 mg/ml, but no tracheal wheezing could be detected with the next three histamine doses. At this point, the child became symptomatic (cough, congestion, irritability), the study was terminated, and salbutamol was administered, resulting in quick relief. Six age- and sex-matched children without asthma or respiratory symptoms for one month prior to being studied (control subjects) had a negative HCT by both tracheal auscultation and computerized LSA. The screening process proved accurate in 10 of 12 (83.3 percent) subjects. In two patients (No.4, Table 1; No.2, Table 2), later complete analysis revealed wheezes at one histamine concentration lower than identified by the screening process. DISCUSSION
Histamine or methacholine challenge is the best available test for confirmation of the diagnosis of asthma. Studies have shown it to be superior in terms of sensitivity and comparable in specificity to other available tests which demonstrate airway hyperreactivity, such as cold air inhalation or exercise challenge. 3 •7 It is also easier and quicker to administer and is therefore the most widespread procedure used in the diagnosis of asthma. Unfortunately, its use has been limited to adults and older children who can perform spirometry consistently. In children under school age, respiratory symptoms are very common, and distinguishing between airway hyperreactivity and other causes is crucial for optimal management. 762
Until recently, no test for airway hyperreactivity could be performed on this large population of patients. Development of "infant pulmonary functions" with the "squeeze technique" enabled application of HCT to infants and young children. 8 However, this method requires specialized equipment and personnel and sedation of the child; also, it is labor-intensive and time consuming and therefore difficult to perform as a routine test in a pulmonary function lab. An alternative recent technique enabling application of HCT to young children is tracheal auscultation... This method correlates well with asthma severity,9 similar to conventional HCT. It entails attaching a microphone (or stethoscope) over the trachea and recording or listening to the sounds after each histamine dose. This method is easy to use and does not require much equipment, which makes it very practical. Its principle, which is to induce wheezing with histamine, is also its main shortcoming, since it would be preferable for a test not to make the subject symptomatic. Computerized LSA offers a few advantages over the tracheal auscultation technique. The use of sensitive recording devices coupled to an amplifier and a filter which eliminates much of the background noise interference enables identification of much softer sounds. We have also found that the spectral LSA will identify wheezing sounds which are not apparent on playback, even after amplification and filtration. Therefore, many asthmatic patients undergoing HCT by this method can be identified before actual wheezes (ie, symptoms) appear. In our stud~ five of six children were identified by LSA prior to the 20 percent fall in FEV. and three of six were identified before wheezes were identified by tracheal auscultation. This method requires more training and expertise than the tracheal auscultation technique, but can well be mastered by pulmonary function technicians. The equipment is not unduly sophisticated or very expensive. This test is now in clinical use in our pulmonary function laboratory. ACKNOWLEDGMENTS: The authors wish to thank Diane Conley and Sue Hegland for technical assistance with Her, and Valerie McGrath for secretarial assistance in preparing the manuscript. REFERENCES
1 Cockcroft DW Measurement ofairway responsiveness to inhaled histamine or methacholine: method of continuous aerosol generation and tidal breathing inhalation. In: Hargreave FE, Woolcock AJ, eds. Airway responsiveness. Mississauga, Canada: Astra, 1985 2 Har~reave FE, Ryan G, Thomson NC, O·Byrne PM, Latimer K, Juniper GF. Bronchial responsiveness to histamine or methacholine in asthma: measurement and clinical significance. J Allergy Clio Immunoll981; 68:347-55 3 Mellis CM, Kattan M, Keens TG, Levison H. Comparative study of histamine and exercise challenges in asthmatic children. Am Rev Respir Dis 1987; 117:911-15 4 Avital A, Bar-Yishay E, Springer C, Godfrey S. Bronchial Histamine Challenge in Young Children (Beck et aJ)
provocation tests in young children using tracheal auscultation.
J Pediatr 1988; 112:591-94
5 Gavriely N, Palty Y, Alroy G, Grotberg JB. Measurement and theory of wheezing breath sounds. J Physioll984; 57:481-92 6 Beck R, Gavriely N. The reproducibility of forced expiratory wheezes. Am Rev Respir Dis 1990; 14:1418-22 7 Filuk RB, Serretta C, Anthomsen NR. Comparison of responses to methacholine and cold air in patients suspected of having
asthma. Chest 1989; 95:948-52 8 LesouefPN, Geelhoed GC, Turner OJ, Morgan EG, Landau LI. Response of normal infants to inhaled histamine. Am Rev Respir Dis 1989; 139:62-66 9 Noviski N, Cohen L, Springer C, Bar-Yisbay E, Avital A, Godfrey S. Bronchial provocation determined by breath sounds compared with lung function. Arch Dis Child 1991; 66:952-55
Plan to Attend ACCP's
58th Annual SCientific Assembly Chicago ~..---.......~ :::~ OCtober 25-29, 1992
cA
CHEST I 102 I 3 I SEPTEMBER, 1992
713
produce a 20 percent fall in FEV•. In the sixth patient, wheezing on LSA and PC20 occurred at the same histamine concentration. To determine the technique's applicability . to young children, we then studied six young asthmatic children (age 2 to 5 years). All children showed the wheezing pattern at a histamine concentration of 25 percent or 50 percent (one or two steps prior) to that producing symptoms (cough, wheeze, chest tightness) or wheezing on tracheal auscultation. Six age- and sex-matched nonasthmatic children (control subjects) did not show this pattern on LSA and had no symptoms or tracheal wheeze with HCT. We describe a sensitive method enabling application of HCT to young children who are unable to perform spirometry. This method is as sensitive as, and often more sensitive than, conventional PC20 with spirometry or tracheal auscultation. (Chest 1992; 102:759-63)
(or methacholine) challenge is a widely H istamine used, sensitive, and specific method for the
rate and reproducible results on spirometry, limiting applicability of this test to older children and adults. An alternative technique recently described is the use of tracheal auscultation to monitor response to histamine." Histamine is given until wheezes are detected (pew) by a stethoscope or microphone placed over the trachea. Modern computerized lung sounds (LS) recording and analysis techniquess .6 enable easy and quick recognition of wheezes by combining "on line" auscultation with appearance of a typical pattern of discrete, high-amplitude sound peaks on spectral analysis. 5 .6 By using a sensitive recording device coupled to an amplifier and filter, much softer sounds than can be detected by ear can be demonstrated. The wheezing pattern during histamine challenge can therefore be identified before wheezing becomes obvious. Our aim was to develop a simple-to-perform, practical tool for performing histamine challenge in young children and infants that would offer objective results.
diagnosis and evaluation of asthma. 1 Nonspecific airway hyperreactivity is demonstrated by inhaling increasing concentrations of nebulized histamine or methacholine until a 20 percent drop in FEV 1 occurs (PC20). Extensive experience over the past 30 years has shown the test to be safe, easily performed, and reproducible, and to correlate well with degree of asthma severi~2 Some authors3 prefer histamine over methacholine because of its shorter duration of action and its being a physiologic mediator of bronchoconstriction. This test is in widespread use in most pulmonary function laboratories, but accuracy of results depends on the patient's ability to perform technically acceptable and reproducible spirometry. Young children, usually under the age of six to eight years, are not able to consistently perform the forced vital capacity maneuver necessary to generate accu-
*From the Division of Respiratory Medicine, Department of Paediatrics, University ofCalgary, The Alberta Children's Hospital~ Calgary, Alberta, Canada. This study was supported in part by a University of Calgary Research Grant. Manuscript received September 23; revision accepted January 2.
=
=
RCT histamine challenge test; LS lung sounds; LSA = lung sounds analysis; Pe20 = provocation concentration of histamine producing 20 percent fall in FEV.; FFf=fast Fourier transform; Pew = provocation concentration of histamine producing tracheal wheeze; PCLsA = provocation concentration of histamine producing a positive result on LSA
MATERIALS AND METHODS
Histamine Challenge This was perfonned by a standard protocol, using tidal breathin~ of continuously nebulized histamine solution, adapted from CockCHEST I 102 I 3 I SEPTEMBER, 1992
759
IV; dry roBing seal spirometer) befOre starting. after Inhaling
nebulized 0.9 percent saIiDe solution (baseline) and after each dose ofhistamlne. until a 00 percent drop from baseline FEV, occurred. or to a maximal histamine dose ofa mgtml.
....._+-......- _...._ -....
Sound Recordlng and Analym Sy6tem8
~~
Figure 1 scbematically depicts the recording system. Sounds were recorded via a Hewlett-Packard HP 210s0A contact sensor placed over the right upper anterior chest. Placement site was chosen so as to be over a large airway as far away from heart sounds as possible. Sounds were ampli6ed (x 10 to x 10(0). band-filtered between 50 and 2,000 Hz (fouN:hannelinstrumentation amplifier/ &Iter. custom built), and digitized (8-bit anaIog-to-digltal converter) at a sampling rate of 5.5 kHz Into a Macintosh SE computer. P1aybaclc screening of sound segments and wave-form and spectral analysis by fast Fourier transfOrm (FFI') were perfOrmed using the SoundWave software on the recorded sounds prior to the next histamine dose. A histamine response was considered positive wben a typical wheezing pattern of high amplitude. discrete power spikes"-" was identified on spectral (FFI') analysis.
•
FIGURE 1. Schematic description of computerized lung sounds recording and analysis system (see text for details). croft' Subjects inhaled the histamine fOr 2 min. starting at a dose of0.03 mgtmI. Histamine concentration was doubled progressively every 5 min to a maximal concentration of a mgtml (Ie, the new dose was started 3 min after the previous one finisbed~ ~ used a Hudson Up-Draft Nebumlst 1700 nebulizer driven with 4 Umln of air (mean mass diameter, 1.5 ....m (85 percent) geometric standard deviation. 1.69). Older children performed spirometry (Gould 5000
.........
sell.. O.9a
~
J
II
l
j
I~
lA \
...
\f'-. ~
Zl
)$'
\ k '\ v 'In
IUS
.n
U1
Zl
HI.taml".O 125 mg/ml
10n
136$
IOlt
lUI
tU.llmlnl '2 0 m,/ml
IA
IV ~ VIA
"
f
s
j
\'\
... f.-.-,
~Ll .51
Zl
II U\
A
V
h
1029
r~'\ .51
Ins
--.
""IA~
lttzl
....... ·4
se".. O.9a
J j
$en. .
'Y\
I"
Zl
'n
0."
\ ....
"'-
."
)$'
'JU
Zl
""- .S1
.n
Hfatemln.O .25 mg/mt
! FIGURE 2. Ell8II'Iples of positive Her by computerized LSA. In each case, note the appearance of discrete high-amplitude power peaks. typical of wheezing lung sounds.
780
1
j
II
r-
~
151
~
II
\J V V'""'"",- ..J\...,.....N au
IOU
II
V\,..
VV
Zl
:In
A
6U
""'V\.-... 10ft
~
IUS
Table I-Patient CharGCtefVtica and BeaultI ofPhaae 1: VGlitltdion ofMetlwd COIIIpGring Con~ BCT and Computeriz,ed LSA Patient
Age/Sex
Asthma Severity
l6IF
Mild
2
l3IF
Moderate
3
141M
Moderate
4 5
111M 9IM
Mild Moderate
6
111M
Mild
Treatment 8eclomethasoQe, salbutamol Theophylline, beclomethasone, salbutamol Theophylline, salbutamol Salbutamol Theophylline, beClomethasone, salbutamol Beclomethasone
PC20, mwml Histamine
LSA, mwml Histamine
0.5
0.25
0.03
0.03
0.06
0.03
2.0 0.25
1.0 0.125
4.0
2.0
lMng Sounds Analysis Method
Study Design
After recording and storing of LS, initial screening of expiratory LS was performed in the time interval until the next LS recording was due. This was performed manually by the researcher, using the SoundWave software, according to a previously described method. 6 Each expiration was divided into 200-ms sequential sound segments. These were analyzed in succession, starting at the beginning of the expiration, by performing spectral analysis (FIT, 1,024 data points) Oil each of the segments. The resulting spectra (Fig 2) were e~ined by the researcher for the presence of peaks typical of wheezes, according to previously published criteria. 6 The main criteria for wbeeze identmcation were as follows: (1) discrete, narrow peak of power; (2) the peak is distinctly separate from its surrounding sound spectrum; (3) peak amplitude at least threefold larger on one side and twice larger on the other side than the base it arises from; and (4) at least one aspect of the peak has to show a sharp rise, whereas on the other side, the slope of the peak may be graded. If, on playback, a wheezing sound was heard, that section of LS was analyzed; otherwise one or two expirations were chosen at random for analysis. This method enabled analysis of one to two expirations over the 5 min before the next LS recording was performed. Systematic analysis of all recorded sounds was performed at a later date and served to confirm the findings of the screening process, or identify wheezing patterns which were missed.
Phase 1: \hlidation of Method: This was done by comparing appearance of wheezing pattern on LS analysis (LSA) to PC20 by spirometry. Six children attending the Alberta Cbildrens Hospital Asthma Clinic (age 9 to 16 years, four male subjects), with known mild or moderate asthma, as judged by the severity of their symptoms and the amount and type of medication taken (Table 1), were studied. Following histamine inhalation, spirometry was performed, followed by computerized LSA, so that PC20 and PCLSA could be compared. The challenge was terminated when a 20 percent drop in FEVI occurred or when the suhject became symptomatic (cough, wheeze, chest tightness).' Phase 2: ApplicabilUy to Younger Children: Here, computerized LSA was compared to tracheal auscultation· during histamine challenge in astJimatic children and in children with no asthma (control subjects). Six children who were being followed up in our Asthma Clinic (age two to five yean, three girls), with mild or moderate asthma (Table 2) were studied. They were unable to perfonn spirometry and had the Her performed according ~o the above protocol, with computerized LSA and tracheal ausculta&n performed after each histamine dose. Six age- and sex-matc~ children without asthma (control subjects) were studied in an identical manner. The test was terminated when wheezing was identified on tracheal auscultation· or when clinical symptoms (cough, wheeze, chest tightness) occurred.
Table I-Patient ClaarGCtefVtica and BeaultI ofPhaae 2: Applictmon to Yourag ~~nm COIIIpGring HCT by Tracheal Auacultation arad Computeriz,ed LSA
Patient
Age/Sex
Asthma Severity
1
3IF
Mild
2
5IF
Mild
3
4IF
Moderate
4
9/M
Moderate
5
31M
Mild
6
31M
Moderate
7-12
2-5
Control subjects
Treatment Salbutamol, astemizole Salbutamol, beclomethasone Salbutamol, cromoglycate Salbutamol, beclomethasone Theophylline, salbutamol Salbutamol, cromoglycate, beclomethasone None
LSA, mglml Histamine 0.125
Tracheal Auscultation, mglml Histamine 0
2.0
4.0
0.125
0.125
0.03
0.03
0.5
1.0
0.06
0.06
>8.0
>8.0
CHEST I 102 I 3 I SEPTEMBER, 1992
781
Ethics
The study was approved by the University of Cal~ary Conjoint Medical Ethics Committee and the Alberta Children·s Hospital Research Committee. Informed consent was wven by all parents and by the older children (phase 1) prior to the study. RESULTS
Figure 2 shows four examples ofa positive histamine challenge, with appearance ofhigh-amplitude discrete power spikes on spectral analysis, typical ofa wheezing pattern. 5•6 Table 1 summarizes results of phase 1. All subjects had a positive histamine challenge, with PC20 ranging from mild (Nos. 4, 6) to severe (Nos. 2, 3) airway reactivity. In all subjects, a typical wheezing pattern on LSA was present at the time 20 percent fall in FEV. occurred. Moreover, in five of six subjects, LSA was positive for wheezing one dilution prior (at one half the histamine concentration) to the PC20. Table 2 summarizes results of phase 2. Five of the six children had a positive histamine challenge by tracheal auscultation," with PCw ranging from mild (No.2) to severe (Nos. 4, 6) airway reactivity. In two of them, typical wheezing pattern on LSA was detected one step prior (at one half the histamine concentration) to PCw. One subject (No.1), had a typical wheezing pattern on LSA at a histamine dose of 0.125 mg/ml, but no tracheal wheezing could be detected with the next three histamine doses. At this point, the child became symptomatic (cough, congestion, irritability), the study was terminated, and salbutamol was administered, resulting in quick relief. Six age- and sex-matched children without asthma or respiratory symptoms for one month prior to being studied (control subjects) had a negative HCT by both tracheal auscultation and computerized LSA. The screening process proved accurate in 10 of 12 (83.3 percent) subjects. In two patients (No.4, Table 1; No.2, Table 2), later complete analysis revealed wheezes at one histamine concentration lower than identified by the screening process. DISCUSSION
Histamine or methacholine challenge is the best available test for confirmation of the diagnosis of asthma. Studies have shown it to be superior in terms of sensitivity and comparable in specificity to other available tests which demonstrate airway hyperreactivity, such as cold air inhalation or exercise challenge. 3 •7 It is also easier and quicker to administer and is therefore the most widespread procedure used in the diagnosis of asthma. Unfortunately, its use has been limited to adults and older children who can perform spirometry consistently. In children under school age, respiratory symptoms are very common, and distinguishing between airway hyperreactivity and other causes is crucial for optimal management. 762
Until recently, no test for airway hyperreactivity could be performed on this large population of patients. Development of "infant pulmonary functions" with the "squeeze technique" enabled application of HCT to infants and young children. 8 However, this method requires specialized equipment and personnel and sedation of the child; also, it is labor-intensive and time consuming and therefore difficult to perform as a routine test in a pulmonary function lab. An alternative recent technique enabling application of HCT to young children is tracheal auscultation... This method correlates well with asthma severity,9 similar to conventional HCT. It entails attaching a microphone (or stethoscope) over the trachea and recording or listening to the sounds after each histamine dose. This method is easy to use and does not require much equipment, which makes it very practical. Its principle, which is to induce wheezing with histamine, is also its main shortcoming, since it would be preferable for a test not to make the subject symptomatic. Computerized LSA offers a few advantages over the tracheal auscultation technique. The use of sensitive recording devices coupled to an amplifier and a filter which eliminates much of the background noise interference enables identification of much softer sounds. We have also found that the spectral LSA will identify wheezing sounds which are not apparent on playback, even after amplification and filtration. Therefore, many asthmatic patients undergoing HCT by this method can be identified before actual wheezes (ie, symptoms) appear. In our stud~ five of six children were identified by LSA prior to the 20 percent fall in FEV. and three of six were identified before wheezes were identified by tracheal auscultation. This method requires more training and expertise than the tracheal auscultation technique, but can well be mastered by pulmonary function technicians. The equipment is not unduly sophisticated or very expensive. This test is now in clinical use in our pulmonary function laboratory. ACKNOWLEDGMENTS: The authors wish to thank Diane Conley and Sue Hegland for technical assistance with Her, and Valerie McGrath for secretarial assistance in preparing the manuscript. REFERENCES
1 Cockcroft DW Measurement ofairway responsiveness to inhaled histamine or methacholine: method of continuous aerosol generation and tidal breathing inhalation. In: Hargreave FE, Woolcock AJ, eds. Airway responsiveness. Mississauga, Canada: Astra, 1985 2 Har~reave FE, Ryan G, Thomson NC, O·Byrne PM, Latimer K, Juniper GF. Bronchial responsiveness to histamine or methacholine in asthma: measurement and clinical significance. J Allergy Clio Immunoll981; 68:347-55 3 Mellis CM, Kattan M, Keens TG, Levison H. Comparative study of histamine and exercise challenges in asthmatic children. Am Rev Respir Dis 1987; 117:911-15 4 Avital A, Bar-Yishay E, Springer C, Godfrey S. Bronchial Histamine Challenge in Young Children (Beck et aJ)
provocation tests in young children using tracheal auscultation.
J Pediatr 1988; 112:591-94
5 Gavriely N, Palty Y, Alroy G, Grotberg JB. Measurement and theory of wheezing breath sounds. J Physioll984; 57:481-92 6 Beck R, Gavriely N. The reproducibility of forced expiratory wheezes. Am Rev Respir Dis 1990; 14:1418-22 7 Filuk RB, Serretta C, Anthomsen NR. Comparison of responses to methacholine and cold air in patients suspected of having
asthma. Chest 1989; 95:948-52 8 LesouefPN, Geelhoed GC, Turner OJ, Morgan EG, Landau LI. Response of normal infants to inhaled histamine. Am Rev Respir Dis 1989; 139:62-66 9 Noviski N, Cohen L, Springer C, Bar-Yisbay E, Avital A, Godfrey S. Bronchial provocation determined by breath sounds compared with lung function. Arch Dis Child 1991; 66:952-55
Plan to Attend ACCP's
58th Annual SCientific Assembly Chicago ~..---.......~ :::~ OCtober 25-29, 1992
cA
CHEST I 102 I 3 I SEPTEMBER, 1992
713
