Scand J Prim Health Care 1991: 9: 232-238
Do Physical Signs Reflect the Degree of Airflow Obstruction in Patients with Asthma or Chronic Obstructive Pulmonary Disease? C. P. van SCHAYCK', C. van WEEL', H. J. M. HARBERS' and C. L. A. van 'HERWAARDEN
Scand J Prim Health Care Downloaded from informahealthcare.com by University of Laval on 03/04/15 For personal use only.
'Department of General PracticelFamily Medicine, Nijmegen University, =Department of Pulmonary Direases, Nijmegen University, The Netherlands
Schayck CP van, Wee1 C van, Harbers KIM, Henvaarden CLA. van. Do physical signs reflect the degree of airflow obstr&ion in patients with asthma or chronic obstructive pulmonary disease? Scand J Prim Health Care 1991; 9 2328. The aim of the study was to relate the physical signs of the chest to the degree of airflow obstruction in asthma and COPD. Methods: 113 patients with COPD and 76 patients with asthma were recruited from general practice. A standard physical examination of the chest was performed by trained medical students. Physical signs were related to the degree of airflow obstnrction. Results: the signs correlating closely with the degree of airflow obstruction were: a prolonged expiratory phase, low-standing diaghragm, decreased expiratory breath sounds, noisy inspiratory sounds, and &creased diaphragmatic excursions. There was a fair correlation between the number of physical signs and the degree of airflow obstruction. This was especially the case in asthma (r=0.62), but it was less clear in COPD (r=0.45). Sensitivity of separate physical signs to detect airflow obstruction was less than 50%, hut at least one of the signs was present in 70% of the patients with obstruction. Specificity of separate signs was more than 85%, apart from wheezing in asthma. Conclusion: the combination of physical signs can offer relevant information in monitoring the severity of airflow obstruction in asthma and COPD. Key words: airflow obstruction, physical signs, COPD, astma.
C. P. van Schayck, Department of General PracticelFamily Medicine, P. 0. Box 9101, 6500 HB Nijmegen, The Netherlands.
Asthma and chronic obstructive pulmonary disease (COPD) are among the most common chronic disorders that are treated in general practice (1, 2). Airflow obstruction is one of the diagnostic features of asthma and COPD. Physical examination of the chest is used regularly in the diagnosis of respiratory diseases, particularly in general practice. Physical examination is mentioned in textbook as an essential part of diagnosis and evaluation in this respect (3). However, the introduction of other diagnostic techniques (spirometry, radiology) has degraded physical examination more and more to a 'bedside ritual'. There are contradictory reports on the correlation between physical signs and more 'advanced' diagnostic tests such as spirometry (4-7). The aim of the present study was to relate the Scand J Prim Health Care 1991; 9
commonly observed physical signs in the examination of the chest to the degree of airflow obstruction (expressed as FEV, = Forced Expiratory Volume in one second) in patients with asthma and COPD .
METHODS This study was part of a long-term intervention study of asthma and COPD (8). Patients Patients with a history and symptoms suggestive of asthma or COPD, treated in general practice, were invited to participate in the study. The selection of
Do physical signs reflect the degree patients has previously been described in detail (8) The criteria for inclusion in the study were: - age 30 years or over; - absence of important concomitant morbidity;
- moderate airflow obstruction (FEV, at least 50%
Scand J Prim Health Care Downloaded from informahealthcare.com by University of Laval on 03/04/15 For personal use only.
of the predicted value (9)) and/or bronchial hyperresponsiveness (PC,, 5 8 mg/ml histamine).
233
Study design All 189 patients were subjected to a physical examination of the chest just before the testing of respiratory function. The examination took place during an exacerbation-free period. All medication was discontinued for a period of 8 or more hours before the examination. This part of the study was performed by three trained medical students (just before graduation; special training at the skills’ laboratory of the University of Nijmegen (16)). A standardized procedure was followed in the physical examination of the chest:
With respect to this latter aspect, the patients were selected by the principal researcher. A standard respiratory history (8, 10) was taken from all the patients who were considered for the study, and spirometry was carried out by means of the Microspiro HI-298, Chest Corporation (Tokyo, Japan) (11). - the examination consisted of inspection, percussion, and auscultation, and was performed while The degree of bronchial responsiveness (Histamine the patients were in an upright position; Challenge Test) was assessed according to Cockcroft - the auscultation was performed during unforced (12). deep breathing with the mouth open; 189 patients were selected. Random samples of patients who refused to participate or who were - the following signs were examined (way of assessment): excluded showed that no selection bias had been introduced in the selection procedure (8). Their con- 1 shape of the chest (barrel shape yeslno) dition was diagnosed ns ‘asthma’ (76 patients) or 2 low level of diaphragm (lower than eleventh thoracic vertebra: yedno) ‘COPD’ (1 13 patients), according to the criteria of the American Thoracic Society (13). The diagnosis 3 decreased diaphragmatic excursions (less than 2 fingers wide: yesho) was based o n a combination of symptoms (MRCECCS questionnaire), lung function (before and 60 4 expiratory breath sounds (decreased yeslno) minutes after 400 pg salbutamol and 80 pg ipratro- 5 expiratory phase (prolonged yes/no) pium bromide), and bronchial responsiveness. Lung 6 wheezing (yedno) function was assessed five times, bronchial respon- 7 inspiratory sounds, audible without auscultation siveness three times in the year preceding this study. (yesho) COPD was defined as the combination of : 1) per- 8 fine crackles (yesho) sistent bronchial obstruction (FEV, 5 85% of the 9 coarse crackles (yesho) predicted value during at least three of five measurements) and 2 ) chronic cough or chronic sputum production for at least three months for at least two Analysis consecutive years. Asthma was defined as the com- The interrelation of physical signs was tested in a bination of 1) reversible airway obstruction (FEV, factor analysis. The correlation between the presincrease 2 15%, 60 minutes after inhalation of both ence of a physical sign and bronchial obstruction was 400 pg salbutamol and 80 pg ipratropium bromide computed by comparing the mean lung function of during at least three of five measurements); 2) bron- those with and without that particular physical sign. chial hyperresponsiveness (PC,, 5 8 mg hista- Differences in FEV, were tested by means of the mine/ml during all three measurements); 3) dys- non-paired Student’s t-test. Percentages of expnoea, and 4) wheezing and/or allergy. Although plained variance in lung function by separate physseparate features of asthma and COPD were not ical signs were compared by means of analyses of mutually exclusive, the combination of features was: variance (univariate analyses). To study the combino patient with asthma had COPD as well, and vice nation of physical signs, the FEV, was correlated versa (14, 15). with the number of signs found in every patient by All patients gave informed consent. The study was means of linear regression analysis. These statistical approved by the Ethics Committee of the Faculty of procedures (ANOVA and linear regression) allowed Medicine, University of Nijmegen. us to use obstruction as a continuous variable: perScund I Prim Heulth Cure 1991; 9
234
C . P. van Schayck et al.
Table I. Clinical characteristics of the patients. Standard deviations in brackets. Asthma
COPD
Total
113 53 (13) 62/38 88/12 19
189 52 (13)
~-
76
Scand J Prim Health Care Downloaded from informahealthcare.com by University of Laval on 03/04/15 For personal use only.
Number Age ( y e a d Maleifemale (YO) Smokednon-smoker (%) Pack years Allergy (Y) Mean FEV, (1) Mean FEV, % p.v., Mean PCm
51 (13) 46/54
* ** **
71/29
13
*t
35 2.13 (0.73) 68 (23) 7
*
***
56/44
18 2.40 (0.79) 74 (18) 23
81/19 16 24 2.29 (0.78) 72 (21) 16
'
p.v. = predicted value (9) Differences between asthma and COPD. * pCO.05 ** pco.01
*** pco.005 centage of predicted value. This has an advantage over using arbitrary boundaries of obstruction. However, in order to calculate the sensitivity and specificity of physical signs for airflow obstruction, patients had to be categorized into obstructive and non-obstructive patients. Airflow obstruction was defined as FEV, less than predicted value minus 1.5 standard deviation (range FEV,: 69.7-74.2Y0 of predicted). Expressing bronchial obstruction in SD is more sensible than YO of predicted value, because the deviation from normal is made independent of the distribution of obstruction in the whole population (especially kurtosis-independent). The likelihood ratio of a borderline of 1.5 SD was higher than that of a borderline of 1 or 2 SD. Furthermore, on theoretical grounds 1.5 SD is probably preferred,
because 5-10% of a standard population is then defined as ill (in a Gauss distribution of airflow obstruction), which is in accordance with the prevalence of asthma and COPD in the general population.
RESULTS The clinical characteristics of the patients are shown in Table I. Patients with asthma and COPD differed significantly in sex ratio, smoking, and, because of *I.explained variance
COPD
asthma
Table 11. Percentage of patients with physical signs in asthma and CO PD Number of patients
Asthma
COPD
76
113
Percentage of patients with:
Barrel-shaped chest Low-standing diaphragm Decreased diaphragmatic excursions Decreased expiratory breath sounds Prolonged expiratory phase Wheezing Noisy inspiratory sound Fine crackies Coarse crackles
3 18 17 26 33 33 15 0 0
At least one physical sign
61
Scand J Prim Health Care 1991; 9
4
16 19 25 21 24 12
15 2 58
w probnged expiratory phase low-standing diaphragm decreased expiratory breath sounds without auscultatlOn decreased diaphragmatic excursions wheezing fine crackles
0noisy insparatory sound
0 Fig. I. The relation of several physical signs with the degree of airflow obstruction in patients with asthma and COPD (Yo explained variance of separate physical signs in univariate analysis). Significant difference in FEV, between patients with and without the sign studied. * P < 0.05 ** P < 0.01 *** P < 0.005
Do physical signs reflect the degree FEY, '/.p.v
4
-
i:;
Jru ' CC""
= - 3 7 8 . * 8 3 I3
I=
..
t
\ X
Scand J Prim Health Care Downloaded from informahealthcare.com by University of Laval on 03/04/15 For personal use only.
0
'
.
\ i
.
\
-P
4
OL5
p'oooo1
\
'.
I 0
L
I
0
I
2
I
A
asthma: = - 6 4 ~ ..8492 r 1062 p c 0 0001 I
L
I
I
I
I
6 0 number of signs
Ffg. 2. Correlation between the FEV, % of the predicted value (dependent variable) and the number of signs (independent variable) in patients with asthma and COPD (regression analysis: y = FEV, % p . ~ .x, = number of signs).
definition criteria, also in allergy, lung function, and bronchial responsiveness (PC?"). Table I1 lists the number of signs present. The presence of physical signs could be demonstrated in 111 patients (59%). Wheezing, a prolonged expiratory phase, and decreased expiratory sounds were the signs that occurred most frequently. Only seven patients had a barrel-shaped chest, two patients had coarse crackles, while fine crackles were not present in any of the patients with asthma. Therefore it was difficult to draw any conclusions from these signs. By means of factor analysis the physical signs could be divided in two groups: 1) barrel-shaped chest, low standing diaphragm, decreased diaphragmatic excursions, decreased expiratory breath sounds (load varying from 0.38 to 0.76); and 2) wheezing and a noisy inspiratory sound (load varying from 0.48 to 0.63). A prolonged expiratory phase could be placed in both groups. Figure 1 shows that the percentages of explained variance (r-square) in FEV, of the separate physical signs were higher in the asthma than in the C O P D patients. The differences in explained variance between the physical signs were larger in asthma as well. A prolonged expiratory phase appeared to be the most important physical sign in asthma (27%).
235
Other signs of any importance in asthma were a low level of the diaphragm (21%) and decreased expiratory sounds (18%). The most important sighns in C O P D were decreased expiratory sounds and decreased diaphragm excursions (both 10%). Apart from wheezing in asthma, patients with more o r less airflow obstruction could be distinguished by means of all signs (p
Do Physical Signs Reflect the Degree of Airflow Obstruction in Patients with Asthma or Chronic Obstructive Pulmonary Disease? C. P. van SCHAYCK', C. van WEEL', H. J. M. HARBERS' and C. L. A. van 'HERWAARDEN
Scand J Prim Health Care Downloaded from informahealthcare.com by University of Laval on 03/04/15 For personal use only.
'Department of General PracticelFamily Medicine, Nijmegen University, =Department of Pulmonary Direases, Nijmegen University, The Netherlands
Schayck CP van, Wee1 C van, Harbers KIM, Henvaarden CLA. van. Do physical signs reflect the degree of airflow obstr&ion in patients with asthma or chronic obstructive pulmonary disease? Scand J Prim Health Care 1991; 9 2328. The aim of the study was to relate the physical signs of the chest to the degree of airflow obstruction in asthma and COPD. Methods: 113 patients with COPD and 76 patients with asthma were recruited from general practice. A standard physical examination of the chest was performed by trained medical students. Physical signs were related to the degree of airflow obstnrction. Results: the signs correlating closely with the degree of airflow obstruction were: a prolonged expiratory phase, low-standing diaghragm, decreased expiratory breath sounds, noisy inspiratory sounds, and &creased diaphragmatic excursions. There was a fair correlation between the number of physical signs and the degree of airflow obstruction. This was especially the case in asthma (r=0.62), but it was less clear in COPD (r=0.45). Sensitivity of separate physical signs to detect airflow obstruction was less than 50%, hut at least one of the signs was present in 70% of the patients with obstruction. Specificity of separate signs was more than 85%, apart from wheezing in asthma. Conclusion: the combination of physical signs can offer relevant information in monitoring the severity of airflow obstruction in asthma and COPD. Key words: airflow obstruction, physical signs, COPD, astma.
C. P. van Schayck, Department of General PracticelFamily Medicine, P. 0. Box 9101, 6500 HB Nijmegen, The Netherlands.
Asthma and chronic obstructive pulmonary disease (COPD) are among the most common chronic disorders that are treated in general practice (1, 2). Airflow obstruction is one of the diagnostic features of asthma and COPD. Physical examination of the chest is used regularly in the diagnosis of respiratory diseases, particularly in general practice. Physical examination is mentioned in textbook as an essential part of diagnosis and evaluation in this respect (3). However, the introduction of other diagnostic techniques (spirometry, radiology) has degraded physical examination more and more to a 'bedside ritual'. There are contradictory reports on the correlation between physical signs and more 'advanced' diagnostic tests such as spirometry (4-7). The aim of the present study was to relate the Scand J Prim Health Care 1991; 9
commonly observed physical signs in the examination of the chest to the degree of airflow obstruction (expressed as FEV, = Forced Expiratory Volume in one second) in patients with asthma and COPD .
METHODS This study was part of a long-term intervention study of asthma and COPD (8). Patients Patients with a history and symptoms suggestive of asthma or COPD, treated in general practice, were invited to participate in the study. The selection of
Do physical signs reflect the degree patients has previously been described in detail (8) The criteria for inclusion in the study were: - age 30 years or over; - absence of important concomitant morbidity;
- moderate airflow obstruction (FEV, at least 50%
Scand J Prim Health Care Downloaded from informahealthcare.com by University of Laval on 03/04/15 For personal use only.
of the predicted value (9)) and/or bronchial hyperresponsiveness (PC,, 5 8 mg/ml histamine).
233
Study design All 189 patients were subjected to a physical examination of the chest just before the testing of respiratory function. The examination took place during an exacerbation-free period. All medication was discontinued for a period of 8 or more hours before the examination. This part of the study was performed by three trained medical students (just before graduation; special training at the skills’ laboratory of the University of Nijmegen (16)). A standardized procedure was followed in the physical examination of the chest:
With respect to this latter aspect, the patients were selected by the principal researcher. A standard respiratory history (8, 10) was taken from all the patients who were considered for the study, and spirometry was carried out by means of the Microspiro HI-298, Chest Corporation (Tokyo, Japan) (11). - the examination consisted of inspection, percussion, and auscultation, and was performed while The degree of bronchial responsiveness (Histamine the patients were in an upright position; Challenge Test) was assessed according to Cockcroft - the auscultation was performed during unforced (12). deep breathing with the mouth open; 189 patients were selected. Random samples of patients who refused to participate or who were - the following signs were examined (way of assessment): excluded showed that no selection bias had been introduced in the selection procedure (8). Their con- 1 shape of the chest (barrel shape yeslno) dition was diagnosed ns ‘asthma’ (76 patients) or 2 low level of diaphragm (lower than eleventh thoracic vertebra: yedno) ‘COPD’ (1 13 patients), according to the criteria of the American Thoracic Society (13). The diagnosis 3 decreased diaphragmatic excursions (less than 2 fingers wide: yesho) was based o n a combination of symptoms (MRCECCS questionnaire), lung function (before and 60 4 expiratory breath sounds (decreased yeslno) minutes after 400 pg salbutamol and 80 pg ipratro- 5 expiratory phase (prolonged yes/no) pium bromide), and bronchial responsiveness. Lung 6 wheezing (yedno) function was assessed five times, bronchial respon- 7 inspiratory sounds, audible without auscultation siveness three times in the year preceding this study. (yesho) COPD was defined as the combination of : 1) per- 8 fine crackles (yesho) sistent bronchial obstruction (FEV, 5 85% of the 9 coarse crackles (yesho) predicted value during at least three of five measurements) and 2 ) chronic cough or chronic sputum production for at least three months for at least two Analysis consecutive years. Asthma was defined as the com- The interrelation of physical signs was tested in a bination of 1) reversible airway obstruction (FEV, factor analysis. The correlation between the presincrease 2 15%, 60 minutes after inhalation of both ence of a physical sign and bronchial obstruction was 400 pg salbutamol and 80 pg ipratropium bromide computed by comparing the mean lung function of during at least three of five measurements); 2) bron- those with and without that particular physical sign. chial hyperresponsiveness (PC,, 5 8 mg hista- Differences in FEV, were tested by means of the mine/ml during all three measurements); 3) dys- non-paired Student’s t-test. Percentages of expnoea, and 4) wheezing and/or allergy. Although plained variance in lung function by separate physseparate features of asthma and COPD were not ical signs were compared by means of analyses of mutually exclusive, the combination of features was: variance (univariate analyses). To study the combino patient with asthma had COPD as well, and vice nation of physical signs, the FEV, was correlated versa (14, 15). with the number of signs found in every patient by All patients gave informed consent. The study was means of linear regression analysis. These statistical approved by the Ethics Committee of the Faculty of procedures (ANOVA and linear regression) allowed Medicine, University of Nijmegen. us to use obstruction as a continuous variable: perScund I Prim Heulth Cure 1991; 9
234
C . P. van Schayck et al.
Table I. Clinical characteristics of the patients. Standard deviations in brackets. Asthma
COPD
Total
113 53 (13) 62/38 88/12 19
189 52 (13)
~-
76
Scand J Prim Health Care Downloaded from informahealthcare.com by University of Laval on 03/04/15 For personal use only.
Number Age ( y e a d Maleifemale (YO) Smokednon-smoker (%) Pack years Allergy (Y) Mean FEV, (1) Mean FEV, % p.v., Mean PCm
51 (13) 46/54
* ** **
71/29
13
*t
35 2.13 (0.73) 68 (23) 7
*
***
56/44
18 2.40 (0.79) 74 (18) 23
81/19 16 24 2.29 (0.78) 72 (21) 16
'
p.v. = predicted value (9) Differences between asthma and COPD. * pCO.05 ** pco.01
*** pco.005 centage of predicted value. This has an advantage over using arbitrary boundaries of obstruction. However, in order to calculate the sensitivity and specificity of physical signs for airflow obstruction, patients had to be categorized into obstructive and non-obstructive patients. Airflow obstruction was defined as FEV, less than predicted value minus 1.5 standard deviation (range FEV,: 69.7-74.2Y0 of predicted). Expressing bronchial obstruction in SD is more sensible than YO of predicted value, because the deviation from normal is made independent of the distribution of obstruction in the whole population (especially kurtosis-independent). The likelihood ratio of a borderline of 1.5 SD was higher than that of a borderline of 1 or 2 SD. Furthermore, on theoretical grounds 1.5 SD is probably preferred,
because 5-10% of a standard population is then defined as ill (in a Gauss distribution of airflow obstruction), which is in accordance with the prevalence of asthma and COPD in the general population.
RESULTS The clinical characteristics of the patients are shown in Table I. Patients with asthma and COPD differed significantly in sex ratio, smoking, and, because of *I.explained variance
COPD
asthma
Table 11. Percentage of patients with physical signs in asthma and CO PD Number of patients
Asthma
COPD
76
113
Percentage of patients with:
Barrel-shaped chest Low-standing diaphragm Decreased diaphragmatic excursions Decreased expiratory breath sounds Prolonged expiratory phase Wheezing Noisy inspiratory sound Fine crackies Coarse crackles
3 18 17 26 33 33 15 0 0
At least one physical sign
61
Scand J Prim Health Care 1991; 9
4
16 19 25 21 24 12
15 2 58
w probnged expiratory phase low-standing diaphragm decreased expiratory breath sounds without auscultatlOn decreased diaphragmatic excursions wheezing fine crackles
0noisy insparatory sound
0 Fig. I. The relation of several physical signs with the degree of airflow obstruction in patients with asthma and COPD (Yo explained variance of separate physical signs in univariate analysis). Significant difference in FEV, between patients with and without the sign studied. * P < 0.05 ** P < 0.01 *** P < 0.005
Do physical signs reflect the degree FEY, '/.p.v
4
-
i:;
Jru ' CC""
= - 3 7 8 . * 8 3 I3
I=
..
t
\ X
Scand J Prim Health Care Downloaded from informahealthcare.com by University of Laval on 03/04/15 For personal use only.
0
'
.
\ i
.
\
-P
4
OL5
p'oooo1
\
'.
I 0
L
I
0
I
2
I
A
asthma: = - 6 4 ~ ..8492 r 1062 p c 0 0001 I
L
I
I
I
I
6 0 number of signs
Ffg. 2. Correlation between the FEV, % of the predicted value (dependent variable) and the number of signs (independent variable) in patients with asthma and COPD (regression analysis: y = FEV, % p . ~ .x, = number of signs).
definition criteria, also in allergy, lung function, and bronchial responsiveness (PC?"). Table I1 lists the number of signs present. The presence of physical signs could be demonstrated in 111 patients (59%). Wheezing, a prolonged expiratory phase, and decreased expiratory sounds were the signs that occurred most frequently. Only seven patients had a barrel-shaped chest, two patients had coarse crackles, while fine crackles were not present in any of the patients with asthma. Therefore it was difficult to draw any conclusions from these signs. By means of factor analysis the physical signs could be divided in two groups: 1) barrel-shaped chest, low standing diaphragm, decreased diaphragmatic excursions, decreased expiratory breath sounds (load varying from 0.38 to 0.76); and 2) wheezing and a noisy inspiratory sound (load varying from 0.48 to 0.63). A prolonged expiratory phase could be placed in both groups. Figure 1 shows that the percentages of explained variance (r-square) in FEV, of the separate physical signs were higher in the asthma than in the C O P D patients. The differences in explained variance between the physical signs were larger in asthma as well. A prolonged expiratory phase appeared to be the most important physical sign in asthma (27%).
235
Other signs of any importance in asthma were a low level of the diaphragm (21%) and decreased expiratory sounds (18%). The most important sighns in C O P D were decreased expiratory sounds and decreased diaphragm excursions (both 10%). Apart from wheezing in asthma, patients with more o r less airflow obstruction could be distinguished by means of all signs (p
![[Usefulness of the Chronic Obstructive Pulmonary Disease Assessment Test in chronic obstructive pulmonary disease with severe airflow limitation].](/assets/img/hold.png)
