Brief Communications Abnormalities in Airway Smooth Muscle in Fatal Asthma
A Comparison between Trachea and Bronchus 1 - 3 TONY R. BAI
We have recently reported the first systematic study of in vitro asthmatic tracheal smooth muscle (TSM) responsiveness(1).The TSM function of sevenpatients who died during relatively sudden, severe asthma attacks outside the hospital was compared with the TSM function of control subjects with normal lungs. In these two groups, spontaneous mechanical activity, electrical stimulation of cholinergic and nonadrenergic noncholinergic (NANC) nerves, and responses to acetylcholine (ACh), histamine, isoproterenol, and theophylline were compared. An increased maximal response to histamine and ACh was noted, without alteration in EC so, and the relaxation response to isoproterenol, and possibly theophylline, was decreased. Cholinergic and NANC nerves functioned normally, and there were no abnormalities in spontaneous mechanical behavior. As the site of airways obstruction is in more peripheral airways in asthma and as TSM properties are different from bronchial smooth muscle (2), subsegmental bronchi (fourth generation) werealso studied from the same patients; these data have now been compared with bronchi from five subjects with normal lungs and with the tracheal data. Full details of methodology, including descriptive details of the seven asthmatics studied, have been previously published (1). In summary, subsegmental bronchi from the anterior segment of the right or left upper lobe were dissected free of surrounding parenchyma and cut into 2- to 2.S-cm spirals in a helical fashion. Four to eight spirals from each subject were suspended in 6-ml organ baths in oxygenated modified Krebs solution maintained at 37° C, and isometric tension was measured using standard techniques. Bronchi were stretched three times to 3-g tension then incubated for 180 min, with washes every 10 min, under I-g passive tension, which in preliminary length-tension experiments in control tissues had yielded optimal force generation with electrical field stimulation (EFS). All bronchi were studied on the day of collection and each spiral was used for only one experiment. The bronchi were initially treated with 0.3 mM ACh, which was washed out as soon as the contractile response had plateaued. All spirals that exhibited a contractile response to ACh were included for further study. 'Iracheal preparation had been judged viable if isometric contractile responses to both EFS and supermaximal concentrations of ACh were greater than O.S g. Cholinergic contractile and nonadrenergic inhibitory neural responses were
SUMMARY Tracheal smooth muscle from seven cases of fatal asthma demonstrated an Increased contractile response to histamine, acetylcholine, and electrical stimulation of Intrinsic cholinergic nerves; Impaired relaxation to Isoproterenol, and possibly theophylline, was also evident (1).Fourth generation bronchial spirals from the same patients were also studied, and these results were compared with those of the trachea and normal bronchi (n 5). In contrast to trachea, contractile responses In asthmatic bronchi to acetylcholine, histamine, and cholinergic nerve stimulation were similar to those In control bronchI. The potency of Isoprenaline (ICso) was reduced 9.4-fold (p < 0.003), similar to trachea (4.5-fold), whereas theophylline responses were normal. The discrepant results obtained may reflect differences In the disease proCess, Including rates of postmortem change, at the two anatomic sites. AM REV RESPIR DIS 1991; 143:441-443
=
studied with EFS (70 V, O.S ms). Cholinergic nerves were studied with 10-s pulse trains at S-min intervals and inhibitory responses with continuous graded field stimulation (1). Histamine, isoproterenol, and theophylline responses were studied using cumulative techniques. Relaxant responses were studied after precontraction of tissues with histamine to 600/0 of the maximal ACh response. At the end of the experiments, the strips were blot-dried, weighed, and then submitted for histologic examination. The asthmatic data were compared with the data of five control subjects with normal lungs by two-way analysis of variance, followed by the least squares mean multiple comparison procedure (SAS statistical package, SAS Institute, Cary, NC). All results are mean ± SEM or geometric mean ± log SEM.
The normal bronchi were obtained at autopsy 6.8 ± 1.1 h after death from three men and two women (age, 48 ± 8.9 yr): the asthmatic bronchi were from a similar age group and autopsy time after death (1). Most of the asthmatic bronchi contained intraluminal exudate and mucus plugs, and the mucosae were erythematous, with histologic features similar to the asthmatic trachea (1).Spontaneous tone was minimal and not different between the two groups, and periodic spontaneous contractions developed during incubation in a minority of spirals in both groups. Forty of 46 bronchial preparations from asthmatics wereresponsiveto ACh, similar to the proportion of tracheal preparations judged viable (50 of 60 preparations). A summary of bronchus and tracheal data is included in table 1. The mean weight of the asthmatic bronchial spirals was 72.8 ± 11.3 mg and of the normal spirals, 70.14 ± 11.2 mg. Mean cumulative concentration and frequency-response curves to ACh, histamine, and EFS are illustrated in figures 1, 2, and
3, respectively, and mean data tabulated are shown in table 1. There were no significant differences in maximal tension or EC so after ACh or histamine, and the two curves are not significantly different by two-way ANOVA. Maximal responses to histamine for one asthmatic bronchus (46 gig) was> 2 SD above the mean for normal tissue. Only two of seven asthmatic bronchi showed contractile responses to cholinergic EFS in contrast to all normal bronchi. However, one of the asthmatic responders showed a maximal tension (21.4gig) > 2 SD above the mean for the normal subjects. Again, the two curves are not significantly different by two-way ANOVA. No NANC inhibitory response was apparent in normal or asthmatic tissue under baseline conditions or after contraction to 60070 of maximal ACh tension with histamine. Maximal relaxation of tissues precontracted with histamine induced by theophylline and isoproterenolwasnot different betweengroups. Mean cumulative theophylline and isoproterenol concentration-response curves are illustrated in figures 4 and 5, and IC so values are
(Received in original form January 31, 1990 and in revised form August 17, 1990) 1 From the Department of Respiratory Medicine, Green Lane Hospital and the Department of Pharmacology, University of Auckland School of Medicine, Auckland, New Zealand. 2 Supported by the Auckland Medical Research Foundation and the Auckland Asthma Society. 3 Correspondence and requests for reprints should be addressed to Tony R. Bai, M.D., UBC Pulmonary Research Lab, St. Paul's Hospital, 1081 Burrard Street, Vancouver, BC V6Z 1Y6, Canada.
441
442
BRIEF COMMUNICATION
TABLE 1 ASTHMATIC BRONCHIAL AND TRACHEAL SMOOTH MUSCLE RESPONSES Drug at 50% max imal effact" (EC•• or IC••)
Maximal Tension
(gig tissue)
C SEM n Bronchus A SEM n
EFS
ACh
7.6 5
21.4 3.7 5
13.6 2.6 5
3.5 3.0 7
15.7 3.5 7
19.1 6.6 6
1.1
Histamine
NS
NS
C SEM n
47.3 4.5 29
62 .7 7.5 29
33 .6 3.6 27
Trachea A SEM n
71.1 12.6 7
101.9 17.4 7
61 .3 13.2 7
P Value
P Value
NS
0.041
0.012
ACh (1lM)
Histamine (1lM)
Isoproterenol (nM)
1.7
3.4
16.4
17.7
5
5
5
5
3.49
1.8
7
6
NS
7
NS
< 0.0001
6.6
12.6
27 5.7 7
0.001
154
NS
Theophylline (1lM)
30 .8 7 NS 5.7
21
14
60.3
32 .0
7
7
0.Q1
0.041
Definition of abbrev iations : C • control; A • asthmatic; EFS • aleel'icallield stimulation; ACh = acetylcholine; NS • not signilicant. " Valuasa,e geomet,ic mean.
tabulated in table 1. There were no significant differences in response to theophylline, but isoproterenol was significantly less potent in asthmatic tissues at all data points except IC ,o: IC 3o, P = 0.049; IC so, p = 0.003; IC,o, p < 0.0001; IC.o, p < 0.0001. Mean log., IC so values for isoproterenol were 7.79 ± 0.25 and 6.81 ± 0.15 for normal and asthmatic tissues, respectively. For theophylline, -log,o ICsovalues were 4.75 ± 0.28 (normal) and 4.51 ± 0.18 (asthmatic). The ratio of IC so values for isoproterenol in asthmatic versus normal tissue was 9.4 in bronchus compared with 4.5 in trachea. The correlation coefficient
(r) comparing isoproterenol responses (IC,o IC. o) within the same individual in asthmatic bronchus and trachea was 0.79 (P < 0.001).
* * Systematic differences in responses of asthmatic bronchus compared with trachea were obtained in this study. In particular, most of
the asthmatic bronchi were hyporesponsive to ACh and field stimulation of intrinsic cholinergic neural elements, whereas maximal responses to these stimuli were increased in trachea (I). The variability in responsiveness was much greater in asthmatic bronchus compared with trachea. There was a trend toward increased responsiveness to histamine in asthmatic bronchi that was nonsignificant and could be a reflection of the small number of control tissues studied (type II error) as well as the variability in responsiveness . An increased responsiveness to histamine was the most apparent contractile abnormality in the tracheal study. An increased contractile response is perhaps to be expected in asthmatic bronchi because smooth muscle volume is increased (3), although it is likely that airway responsiveness is regulated by a series of factors other than smooth muscle characteristics such as hypertrophy and/or hyperplasia. In contrast to the contractile data, the responses to the relaxant agonists isoproterenol and theophylline were in broad agreement with the tracheal results, the striking abnormality being impaired relaxation to adrenoceptor agonists. The theophylline results in trachea showed an unusual dose-response relationship, with borderline significantdifferences at the IC 30 and IC so data points alone. The bronchi showed no trend at all toward differences in responses to theophylline, casting further doubt on the significance of the tracheal data. The results obtained in bronchi therefore substantiated those obtained by Goldie and colleagues (4) in postmortem bronchi from cases of fatal asthma. The sensitivity to the cholinergic contractile agonist carbachol showed a small decrease in their study; maximal responses were similar, as was hista-
26 24
22
12
20 10
24
18
I
21
;;
18
8
~
14
12
10
o 65
4
16
0.5
32
64
. Iog,o ACETYLCHOLINE (M) FRFQUENCY (HZ)
Fig . 1. Mean cumulative acetylcholine concentrationresponse curves of bronch ial spirals from seven subjects with fatal asthma (open circles) and from five normal subjects (closed circles) . Bars are 1 SEM. The ordinate is induced tens ion, gram per gram bronchial tissue.
- Iog,o HISTAMINE 1M)
Fig. 2. Mean cumulative histamine concentrationresponse curves of bronchial spirals from six SUbjects with fatal asthma (open circles) and from five normal subjects (closed circles). Bars are 1 SEM . The ord inate is induced tension, gram per gram bronchial tissue.
Fig . 3. Mean frequency-response curves after electrical field stimulation of cholinergic nerves in bronchial spirals from seven subjects with fatal asthma (open circles) and from five normal subjects (closed circles). Bars are 1 SEM . The ordinate is induced tension , gram per gram bronchial tissue .
443
BRIEF COMMUNICATION
'0
'0
30
50
70
70
!lO
!lO
• )GO,OTHEOPHYLLINE 1M)
Fig. 4. Mean cumulative theophyllinelog" concent ration-response curves of bronchial sp irals from seven asthmat ics (open circles) and five normal subjects (cIossd circles). All preparationswereprecontractedwith histamine. Responses were calculatedas a percentage of the maximal relaxation produced by theophylline in each preparation. Horizontal bars represent SEM of mean concentrations producing10. 30, SO. 70, and 90% of the maximal response .
mine sensitivity and maximal response. Responses to isoproterenol and theophylline were almost identical to our results. In contrast, in the only other systematic study, Whicker and coworkers (5) showed decreased sensitivity to both carbachol and histamine as well as similar maximal responses in bronchial specimens obtained at thoracotomy from mild to moderate asthmatics (seven of eight of whom were smokers). Responses to relaxant agonists were normal. How can the discrepant results between trachea and bronchus be explained? We hypothesize that TSM is less susceptible to the confounding effects of autolysis, and hence, the tracheal rather than the bronchial results are mo re representative of the in vitro properties of asthmatic airway smooth muscle in fatal asthma. In support of th is hypothesis, it is known that autolysis occurs more rapidly in diseased lungs than in healthy lungs postmortem (6, 7), probably because of a number of factors, including disease-related antemortem pulmonary leukocytosis and alterations in cell membrane permeability. This process may be more advanced peripherally than centrally, in part because intraluminal exudate is commonly more pronounced peripherally, as was found in our patients, which may hasten autolysis. In addition, central airways cool more quickly upon mortuary refrigeration, arresting autolysis (6). Further evidence for a more pronounced peripheral, postmortem effect influencing our results was the absence of a NANC. response in both normal and asthmatic bronchi, whereas this response was well preserved in thoracotomy tissue (8).
.10010 ISOPRENALINE (M)
Fig. 5. Mean cumulative isoproterenol concentrationresponse curves of bronchial spirals from seven asthmatics (open circles) and five normal subjects (closed circles). All preparations were precontracted with histamine. Responses were calculated as a percentage of the maximal relaxation produced by isoproterenol in each preparation. Horizontal bars represent SEM of mean concentrations producing10,30, SO, 70.and 90% of the maximal response.
Detailed length-tension studies were not performed in asthmatic bronchi, both normal and asthmatic tissues being studied under the same l-g passive tension that has proved optimal for force generation in preliminary length-tension experiments in normal tissues. Hence, the asthmatic tissues may have been at a suboptimal length for force generation. However, tracheae in both groups were also studied under the same passive tension (2 g), and asthmatic trachea showed increased contractile responses. Furthermore, Whicker and colleagues (5) showed no difference in length-tension characteristics in a similar study in bronchi, albeit in less severe asthmatics. More detailed analysis of the mechanical properties of asthmatic airway smooth muscle is still necessary before further conclusions can be inferred. Alternatively, discrepant results between tracheal and bronchial smooth muscle studies may in part be explained by the now wellrecognized regional heterogeneity in the intrinsic properties of airway smooth muscle (2). Previous studies have documented a number of differences in the functional in vitro properties of central and peripheral airways in various mammalian species (9). Histamine is more effective in peripheral airways , whereas muscarinic stimulation is relatively greater in trachea. A progressive 350010 increase in maximal force of contraction, normalized for cross-sectional area and smooth muscle content, is observed from Generations I through 5 for canine airway smooth muscle. However, such studies are unlikely to explain the differences observed between paired
trachea and bronchi in our study, which have been compared with appropriate control samples. Qualitative assessment of histologic sections of bronchial spirals showed features similar to those found in tracheal strips (I). Although a relationship between in vitro responsiveness and airway wall inflammation is possible, previous studies have not shown a correlation (10).Similarly, although we have not studied the influence of epithelial removal in bronchial spirals, our studies in normal trachea of epithelial removal did not replicate the findings found in asthmatic trachea (ll). In conclusion, postmortem bronchial and tracheal smooth muscle in fatal asthma demonstrate in vitro a clear reduction in adrenergic responsiveness that requires further investigation (I), but the contractile properties of the muscle differ at the two sites and the tra cheal results may more accurately reflect in vivo airway responsiveness.
Acknowledgment F. W. Prasad provided technical assistan ce.
References I. Bai TR. Abnormalities in airway smooth muscle in fatal asthma. Am Rev Respir Dis 1990; 141:552-7. 2. Daniel EE , Kannan M, Davis C, Posey-Daniel V. Ultrastructural studies of the neuromuscular control of human tracheal and bronchial muscle. Respir Physiol 1986; 63:109-28. 3. Hossain S. Quantitative measurement of bronchi muscle in men with asthma. Am Rev Respir Dis 1973; 107:99-109. 4. Goldie RG, Spina D, Henry PJ , Lulich KM, Paterson JW.ln vitro responsiveness of human asthmatic bronchus to ca rbachol, histamine, beta adrenoceptor agonists and theophylline. Br J Clin Pharamacol 1986; 22:669-76. 5. Whicker S, Armour C, Black J . Responsive ness of bronchial smooth muscle from asthmatic subjects to contractile and relaxant agonists, Pul mon Pharmacol 1988; 1:25-31. 6. Bachofen M, Weibel ER, Roos B. Postmortem fixat ion of human lungs for electron microscopy. Am Rev Respir Dis 1975; 111:247-56. 7. Lee RMK , Rossman CM, O'Brodovich H. Assessment of postmortem respiratory ciliary motil ity and ultrastructure. Am Rev Respir Dis 1987; 136:445-7. 8. Taylor SM , Pare PD, Armour CL, Hogg JC, Schellenberg RR . Airway reacti vity in chronic obstructive pulmonary disease. Am Rev Respir Dis 1985; 132:30-5. 9. Leff AR. Endogenous regulation of bronchomotor tone. Am Rev Respir Dis 1988; 137:1198-216. 10. Armour CL , Black JL, Berend N, Woolcock AJ. The relationship between bronchial hyperresponsi veness to methacholine and airway smooth muscle structure and reactivity. Respir PhysioI1984; 58:223-33. 11. Bai TR. Effect of epithelial removal on con tractile and relaxant responses in postmortem human trachealis muscle (abstract). Am Rev Respir Dis 1990; 141:A289.
A Comparison between Trachea and Bronchus 1 - 3 TONY R. BAI
We have recently reported the first systematic study of in vitro asthmatic tracheal smooth muscle (TSM) responsiveness(1).The TSM function of sevenpatients who died during relatively sudden, severe asthma attacks outside the hospital was compared with the TSM function of control subjects with normal lungs. In these two groups, spontaneous mechanical activity, electrical stimulation of cholinergic and nonadrenergic noncholinergic (NANC) nerves, and responses to acetylcholine (ACh), histamine, isoproterenol, and theophylline were compared. An increased maximal response to histamine and ACh was noted, without alteration in EC so, and the relaxation response to isoproterenol, and possibly theophylline, was decreased. Cholinergic and NANC nerves functioned normally, and there were no abnormalities in spontaneous mechanical behavior. As the site of airways obstruction is in more peripheral airways in asthma and as TSM properties are different from bronchial smooth muscle (2), subsegmental bronchi (fourth generation) werealso studied from the same patients; these data have now been compared with bronchi from five subjects with normal lungs and with the tracheal data. Full details of methodology, including descriptive details of the seven asthmatics studied, have been previously published (1). In summary, subsegmental bronchi from the anterior segment of the right or left upper lobe were dissected free of surrounding parenchyma and cut into 2- to 2.S-cm spirals in a helical fashion. Four to eight spirals from each subject were suspended in 6-ml organ baths in oxygenated modified Krebs solution maintained at 37° C, and isometric tension was measured using standard techniques. Bronchi were stretched three times to 3-g tension then incubated for 180 min, with washes every 10 min, under I-g passive tension, which in preliminary length-tension experiments in control tissues had yielded optimal force generation with electrical field stimulation (EFS). All bronchi were studied on the day of collection and each spiral was used for only one experiment. The bronchi were initially treated with 0.3 mM ACh, which was washed out as soon as the contractile response had plateaued. All spirals that exhibited a contractile response to ACh were included for further study. 'Iracheal preparation had been judged viable if isometric contractile responses to both EFS and supermaximal concentrations of ACh were greater than O.S g. Cholinergic contractile and nonadrenergic inhibitory neural responses were
SUMMARY Tracheal smooth muscle from seven cases of fatal asthma demonstrated an Increased contractile response to histamine, acetylcholine, and electrical stimulation of Intrinsic cholinergic nerves; Impaired relaxation to Isoproterenol, and possibly theophylline, was also evident (1).Fourth generation bronchial spirals from the same patients were also studied, and these results were compared with those of the trachea and normal bronchi (n 5). In contrast to trachea, contractile responses In asthmatic bronchi to acetylcholine, histamine, and cholinergic nerve stimulation were similar to those In control bronchI. The potency of Isoprenaline (ICso) was reduced 9.4-fold (p < 0.003), similar to trachea (4.5-fold), whereas theophylline responses were normal. The discrepant results obtained may reflect differences In the disease proCess, Including rates of postmortem change, at the two anatomic sites. AM REV RESPIR DIS 1991; 143:441-443
=
studied with EFS (70 V, O.S ms). Cholinergic nerves were studied with 10-s pulse trains at S-min intervals and inhibitory responses with continuous graded field stimulation (1). Histamine, isoproterenol, and theophylline responses were studied using cumulative techniques. Relaxant responses were studied after precontraction of tissues with histamine to 600/0 of the maximal ACh response. At the end of the experiments, the strips were blot-dried, weighed, and then submitted for histologic examination. The asthmatic data were compared with the data of five control subjects with normal lungs by two-way analysis of variance, followed by the least squares mean multiple comparison procedure (SAS statistical package, SAS Institute, Cary, NC). All results are mean ± SEM or geometric mean ± log SEM.
The normal bronchi were obtained at autopsy 6.8 ± 1.1 h after death from three men and two women (age, 48 ± 8.9 yr): the asthmatic bronchi were from a similar age group and autopsy time after death (1). Most of the asthmatic bronchi contained intraluminal exudate and mucus plugs, and the mucosae were erythematous, with histologic features similar to the asthmatic trachea (1).Spontaneous tone was minimal and not different between the two groups, and periodic spontaneous contractions developed during incubation in a minority of spirals in both groups. Forty of 46 bronchial preparations from asthmatics wereresponsiveto ACh, similar to the proportion of tracheal preparations judged viable (50 of 60 preparations). A summary of bronchus and tracheal data is included in table 1. The mean weight of the asthmatic bronchial spirals was 72.8 ± 11.3 mg and of the normal spirals, 70.14 ± 11.2 mg. Mean cumulative concentration and frequency-response curves to ACh, histamine, and EFS are illustrated in figures 1, 2, and
3, respectively, and mean data tabulated are shown in table 1. There were no significant differences in maximal tension or EC so after ACh or histamine, and the two curves are not significantly different by two-way ANOVA. Maximal responses to histamine for one asthmatic bronchus (46 gig) was> 2 SD above the mean for normal tissue. Only two of seven asthmatic bronchi showed contractile responses to cholinergic EFS in contrast to all normal bronchi. However, one of the asthmatic responders showed a maximal tension (21.4gig) > 2 SD above the mean for the normal subjects. Again, the two curves are not significantly different by two-way ANOVA. No NANC inhibitory response was apparent in normal or asthmatic tissue under baseline conditions or after contraction to 60070 of maximal ACh tension with histamine. Maximal relaxation of tissues precontracted with histamine induced by theophylline and isoproterenolwasnot different betweengroups. Mean cumulative theophylline and isoproterenol concentration-response curves are illustrated in figures 4 and 5, and IC so values are
(Received in original form January 31, 1990 and in revised form August 17, 1990) 1 From the Department of Respiratory Medicine, Green Lane Hospital and the Department of Pharmacology, University of Auckland School of Medicine, Auckland, New Zealand. 2 Supported by the Auckland Medical Research Foundation and the Auckland Asthma Society. 3 Correspondence and requests for reprints should be addressed to Tony R. Bai, M.D., UBC Pulmonary Research Lab, St. Paul's Hospital, 1081 Burrard Street, Vancouver, BC V6Z 1Y6, Canada.
441
442
BRIEF COMMUNICATION
TABLE 1 ASTHMATIC BRONCHIAL AND TRACHEAL SMOOTH MUSCLE RESPONSES Drug at 50% max imal effact" (EC•• or IC••)
Maximal Tension
(gig tissue)
C SEM n Bronchus A SEM n
EFS
ACh
7.6 5
21.4 3.7 5
13.6 2.6 5
3.5 3.0 7
15.7 3.5 7
19.1 6.6 6
1.1
Histamine
NS
NS
C SEM n
47.3 4.5 29
62 .7 7.5 29
33 .6 3.6 27
Trachea A SEM n
71.1 12.6 7
101.9 17.4 7
61 .3 13.2 7
P Value
P Value
NS
0.041
0.012
ACh (1lM)
Histamine (1lM)
Isoproterenol (nM)
1.7
3.4
16.4
17.7
5
5
5
5
3.49
1.8
7
6
NS
7
NS
< 0.0001
6.6
12.6
27 5.7 7
0.001
154
NS
Theophylline (1lM)
30 .8 7 NS 5.7
21
14
60.3
32 .0
7
7
0.Q1
0.041
Definition of abbrev iations : C • control; A • asthmatic; EFS • aleel'icallield stimulation; ACh = acetylcholine; NS • not signilicant. " Valuasa,e geomet,ic mean.
tabulated in table 1. There were no significant differences in response to theophylline, but isoproterenol was significantly less potent in asthmatic tissues at all data points except IC ,o: IC 3o, P = 0.049; IC so, p = 0.003; IC,o, p < 0.0001; IC.o, p < 0.0001. Mean log., IC so values for isoproterenol were 7.79 ± 0.25 and 6.81 ± 0.15 for normal and asthmatic tissues, respectively. For theophylline, -log,o ICsovalues were 4.75 ± 0.28 (normal) and 4.51 ± 0.18 (asthmatic). The ratio of IC so values for isoproterenol in asthmatic versus normal tissue was 9.4 in bronchus compared with 4.5 in trachea. The correlation coefficient
(r) comparing isoproterenol responses (IC,o IC. o) within the same individual in asthmatic bronchus and trachea was 0.79 (P < 0.001).
* * Systematic differences in responses of asthmatic bronchus compared with trachea were obtained in this study. In particular, most of
the asthmatic bronchi were hyporesponsive to ACh and field stimulation of intrinsic cholinergic neural elements, whereas maximal responses to these stimuli were increased in trachea (I). The variability in responsiveness was much greater in asthmatic bronchus compared with trachea. There was a trend toward increased responsiveness to histamine in asthmatic bronchi that was nonsignificant and could be a reflection of the small number of control tissues studied (type II error) as well as the variability in responsiveness . An increased responsiveness to histamine was the most apparent contractile abnormality in the tracheal study. An increased contractile response is perhaps to be expected in asthmatic bronchi because smooth muscle volume is increased (3), although it is likely that airway responsiveness is regulated by a series of factors other than smooth muscle characteristics such as hypertrophy and/or hyperplasia. In contrast to the contractile data, the responses to the relaxant agonists isoproterenol and theophylline were in broad agreement with the tracheal results, the striking abnormality being impaired relaxation to adrenoceptor agonists. The theophylline results in trachea showed an unusual dose-response relationship, with borderline significantdifferences at the IC 30 and IC so data points alone. The bronchi showed no trend at all toward differences in responses to theophylline, casting further doubt on the significance of the tracheal data. The results obtained in bronchi therefore substantiated those obtained by Goldie and colleagues (4) in postmortem bronchi from cases of fatal asthma. The sensitivity to the cholinergic contractile agonist carbachol showed a small decrease in their study; maximal responses were similar, as was hista-
26 24
22
12
20 10
24
18
I
21
;;
18
8
~
14
12
10
o 65
4
16
0.5
32
64
. Iog,o ACETYLCHOLINE (M) FRFQUENCY (HZ)
Fig . 1. Mean cumulative acetylcholine concentrationresponse curves of bronch ial spirals from seven subjects with fatal asthma (open circles) and from five normal subjects (closed circles) . Bars are 1 SEM. The ordinate is induced tens ion, gram per gram bronchial tissue.
- Iog,o HISTAMINE 1M)
Fig. 2. Mean cumulative histamine concentrationresponse curves of bronchial spirals from six SUbjects with fatal asthma (open circles) and from five normal subjects (closed circles). Bars are 1 SEM . The ord inate is induced tension, gram per gram bronchial tissue.
Fig . 3. Mean frequency-response curves after electrical field stimulation of cholinergic nerves in bronchial spirals from seven subjects with fatal asthma (open circles) and from five normal subjects (closed circles). Bars are 1 SEM . The ordinate is induced tension , gram per gram bronchial tissue .
443
BRIEF COMMUNICATION
'0
'0
30
50
70
70
!lO
!lO
• )GO,OTHEOPHYLLINE 1M)
Fig. 4. Mean cumulative theophyllinelog" concent ration-response curves of bronchial sp irals from seven asthmat ics (open circles) and five normal subjects (cIossd circles). All preparationswereprecontractedwith histamine. Responses were calculatedas a percentage of the maximal relaxation produced by theophylline in each preparation. Horizontal bars represent SEM of mean concentrations producing10. 30, SO. 70, and 90% of the maximal response .
mine sensitivity and maximal response. Responses to isoproterenol and theophylline were almost identical to our results. In contrast, in the only other systematic study, Whicker and coworkers (5) showed decreased sensitivity to both carbachol and histamine as well as similar maximal responses in bronchial specimens obtained at thoracotomy from mild to moderate asthmatics (seven of eight of whom were smokers). Responses to relaxant agonists were normal. How can the discrepant results between trachea and bronchus be explained? We hypothesize that TSM is less susceptible to the confounding effects of autolysis, and hence, the tracheal rather than the bronchial results are mo re representative of the in vitro properties of asthmatic airway smooth muscle in fatal asthma. In support of th is hypothesis, it is known that autolysis occurs more rapidly in diseased lungs than in healthy lungs postmortem (6, 7), probably because of a number of factors, including disease-related antemortem pulmonary leukocytosis and alterations in cell membrane permeability. This process may be more advanced peripherally than centrally, in part because intraluminal exudate is commonly more pronounced peripherally, as was found in our patients, which may hasten autolysis. In addition, central airways cool more quickly upon mortuary refrigeration, arresting autolysis (6). Further evidence for a more pronounced peripheral, postmortem effect influencing our results was the absence of a NANC. response in both normal and asthmatic bronchi, whereas this response was well preserved in thoracotomy tissue (8).
.10010 ISOPRENALINE (M)
Fig. 5. Mean cumulative isoproterenol concentrationresponse curves of bronchial spirals from seven asthmatics (open circles) and five normal subjects (closed circles). All preparations were precontracted with histamine. Responses were calculated as a percentage of the maximal relaxation produced by isoproterenol in each preparation. Horizontal bars represent SEM of mean concentrations producing10,30, SO, 70.and 90% of the maximal response.
Detailed length-tension studies were not performed in asthmatic bronchi, both normal and asthmatic tissues being studied under the same l-g passive tension that has proved optimal for force generation in preliminary length-tension experiments in normal tissues. Hence, the asthmatic tissues may have been at a suboptimal length for force generation. However, tracheae in both groups were also studied under the same passive tension (2 g), and asthmatic trachea showed increased contractile responses. Furthermore, Whicker and colleagues (5) showed no difference in length-tension characteristics in a similar study in bronchi, albeit in less severe asthmatics. More detailed analysis of the mechanical properties of asthmatic airway smooth muscle is still necessary before further conclusions can be inferred. Alternatively, discrepant results between tracheal and bronchial smooth muscle studies may in part be explained by the now wellrecognized regional heterogeneity in the intrinsic properties of airway smooth muscle (2). Previous studies have documented a number of differences in the functional in vitro properties of central and peripheral airways in various mammalian species (9). Histamine is more effective in peripheral airways , whereas muscarinic stimulation is relatively greater in trachea. A progressive 350010 increase in maximal force of contraction, normalized for cross-sectional area and smooth muscle content, is observed from Generations I through 5 for canine airway smooth muscle. However, such studies are unlikely to explain the differences observed between paired
trachea and bronchi in our study, which have been compared with appropriate control samples. Qualitative assessment of histologic sections of bronchial spirals showed features similar to those found in tracheal strips (I). Although a relationship between in vitro responsiveness and airway wall inflammation is possible, previous studies have not shown a correlation (10).Similarly, although we have not studied the influence of epithelial removal in bronchial spirals, our studies in normal trachea of epithelial removal did not replicate the findings found in asthmatic trachea (ll). In conclusion, postmortem bronchial and tracheal smooth muscle in fatal asthma demonstrate in vitro a clear reduction in adrenergic responsiveness that requires further investigation (I), but the contractile properties of the muscle differ at the two sites and the tra cheal results may more accurately reflect in vivo airway responsiveness.
Acknowledgment F. W. Prasad provided technical assistan ce.
References I. Bai TR. Abnormalities in airway smooth muscle in fatal asthma. Am Rev Respir Dis 1990; 141:552-7. 2. Daniel EE , Kannan M, Davis C, Posey-Daniel V. Ultrastructural studies of the neuromuscular control of human tracheal and bronchial muscle. Respir Physiol 1986; 63:109-28. 3. Hossain S. Quantitative measurement of bronchi muscle in men with asthma. Am Rev Respir Dis 1973; 107:99-109. 4. Goldie RG, Spina D, Henry PJ , Lulich KM, Paterson JW.ln vitro responsiveness of human asthmatic bronchus to ca rbachol, histamine, beta adrenoceptor agonists and theophylline. Br J Clin Pharamacol 1986; 22:669-76. 5. Whicker S, Armour C, Black J . Responsive ness of bronchial smooth muscle from asthmatic subjects to contractile and relaxant agonists, Pul mon Pharmacol 1988; 1:25-31. 6. Bachofen M, Weibel ER, Roos B. Postmortem fixat ion of human lungs for electron microscopy. Am Rev Respir Dis 1975; 111:247-56. 7. Lee RMK , Rossman CM, O'Brodovich H. Assessment of postmortem respiratory ciliary motil ity and ultrastructure. Am Rev Respir Dis 1987; 136:445-7. 8. Taylor SM , Pare PD, Armour CL, Hogg JC, Schellenberg RR . Airway reacti vity in chronic obstructive pulmonary disease. Am Rev Respir Dis 1985; 132:30-5. 9. Leff AR. Endogenous regulation of bronchomotor tone. Am Rev Respir Dis 1988; 137:1198-216. 10. Armour CL , Black JL, Berend N, Woolcock AJ. The relationship between bronchial hyperresponsi veness to methacholine and airway smooth muscle structure and reactivity. Respir PhysioI1984; 58:223-33. 11. Bai TR. Effect of epithelial removal on con tractile and relaxant responses in postmortem human trachealis muscle (abstract). Am Rev Respir Dis 1990; 141:A289.
