Abstract
Asthma is characterised by an increased airway smooth muscle (ASM) area (ASMarea) within the airway wall. The present study examined the relationship of factors including severity and duration of asthma to ASMarea.
The perimeter of the basement membrane (PBM) and ASMarea were measured on transverse sections of large and small airways from post mortem cases of fatal (n = 107) and nonfatal asthma (n = 37) and from control subjects (n = 69). The thickness of ASM (ASMarea/PBM) was compared between asthma groups using multivariate linear regression.
When all airways were considered together, ASMarea/PBM (in millimetres) was increased in nonfatal (median 0.04; interquartile range 0.013–0.051; p = 0.034) and fatal cases of asthma (0.048; 0.025–0.078; p<0.001) compared with controls (0.036; 0.024–0.042). Compared with cases of nonfatal asthma, ASMarea/PBM was greater in cases of fatal asthma in large (p<0.001) and medium (p<0.001), but not small, airways. ASMarea/PBM was not related to duration of asthma, age of onset of asthma, sex or smoking. No effect due to study centre, other than that due to sampling strategy, was found.
The thickness of the ASM layer is increased in asthma and is related to the severity of asthma but not its duration.
The airway smooth muscle (ASM) layer area (ASMarea) seen on transverse sections is increased up to five-fold in cases of asthma 1. The post mortem study of Carroll et al. 2 showed that the thickness of the ASM layer in people with clinically mild or moderately severe asthma was less than that in patients who died of asthma and who had clinically severe disease. This suggests a relationship between asthma severity and ASMarea. In biopsy studies, the only structural change that distinguished severe from mild/moderate cases of asthma was the ASMarea 3, 4. In addition to clinical severity, a number of other factors, such as sex, age, cigarette smoking, treatment, age at onset of asthma and its duration, might be related to the ASMarea. Bai et al. 5 examined the effects of age on ASMarea and the relative area fractions of smooth muscle and extracellular matrix by comparing younger (17–23 yrs) and older (40–49 yrs) adults with fatal asthma. The amount of ASM was greater in older versus younger subjects with fatal asthma after correction for the fraction of smooth muscle within the ASM layer.
The Melbourne longitudinal study of asthma followed patients with mild and severe asthma from childhood 6. In general, individuals tended to show stable clinical severity from childhood through to middle age. Similarly, reductions in lung function have been shown to track with persistence of symptoms from childhood to adulthood 7. These studies suggest that the severity of asthma may be determined early in life. Therefore, we hypothesised that the ASMarea would not be related to age or duration of asthma but to severity of asthma. This hypothesis was tested by examining the thickness of the ASM layer in a large number of cases of asthma and control subjects from six study centres.
MATERIAL AND METHODS
Further details regarding methods are available in the online supplementary material.
Subjects
Post mortem tissues from subjects with or without asthma contained in six tissue banks (table 1⇓) were studied. These included a prospective study of fatal asthma 2; a study of asthma management and mortality 8; a study of airway development 9; a study of fatal asthma 10; a study from New Zealand of fatal asthma 11; and a Canadian study of fatal asthma 12. Subjects were classified as: control (death due to nonrespiratory causes; history of asthma excluded); nonfatal asthma (death due to nonrespiratory causes; history of asthma confirmed); or fatal asthma (cause of death consistent with a fatal attack of asthma; history of previous asthma confirmed). Approval for these studies and for the collaborative analysis was obtained from the Institutional Ethics Committees of the participating centres and from the Sir Charles Gairdner Hospital (Nedlands, Australia) Human Research Ethics Committee. Demographic data, including age at death, sex and, where available, smoking history, frequency of asthma symptoms, duration of asthma, age of onset of asthma and current treatment requirements were recorded. Smoking was categorised as: ever-smoker (current or ex-smoker); or never-smoker (at time of death).
Sources and preparation of tissue for study
Asthma severity score
Those with a history of asthma and with available information were assigned a category for asthma severity, unrelated to the cause of death, based on guidelines from the Global Initiative for Asthma (GINA) 13. Subjects were classified as having severe asthma if they were using oral corticosteroids, reported hospitalisations for asthma (ever) or had daily symptoms. Subjects were classified as having moderate asthma if they had none of the above, but had symptoms on most days or nights (>3 days·week−1), used regular inhaled corticosteroids or used reliever medications on most days or nights. The remaining cases were, therefore, classified as mild.
Tissue preparation and measurement
Post mortem tissues (ranging from whole lungs to dissected airways or parenchyma) were fixed by inflation or immersion. Airways had been sampled systematically in some centres 2, 9, 12, and at the discretion of the pathologist, for diagnostic purposes, at others 8, 10, 11. Cases were excluded if there was macroscopic or microscopic evidence of lung injury or disease (such as pneumonia). Sections of large and small airways were cut at a thickness of either 4 or 5 μm and stained with haematoxylin and eosin or using either Masson’s or Gomori’s trichrome technique. Planimetry was perfomed on each transverse section of airway in order to measure the length of the basement membrane (PBM), and either planimetry or point counts were used to measure the ASMarea 14.
Analysis
Analyses were conducted using SPSS version 13.0 (SPSS, Chicago IL, USA), primarily for ASMarea corrected for PBM as a marker of airway size. As ASMarea/PBM was not normally distributed, univariate differences in log10ASMarea/PBM between asthma categories were tested using ANOVA and Tukey’s test post hoc. Correlations were tested using Pearson’s correlation. Associations between categories were tested using the Chi-squared test. In multivariate analyses, general linear models were used to examine the effects of study centre, age, duration of asthma, age of onset of asthma, sex, asthma group and PBM on ASMarea/PBM (logarithmically transformed). Analyses were also repeated for airways categorised as small (PBM<4 mm), medium (PBM 4–10 mm) or large (PBM>10 mm) airways. Probabilities of <5% were considered significant and adjusted r2 are presented.
RESULTS
The baseline characteristics of the cases are shown in table 2⇓. There were 213 subjects with airways available for analysis. Clinical data were not available for all cases (table 2⇓). The age distribution was similar for the three case groups, but there were relatively more females in the asthma groups (p = 0.013). In the 57 cases for whom data were available, there was a strong relationship between nonfatal asthma and clinically mild or moderate asthma, and fatal asthma and clinically severe asthma (Chi-squared = 15.3; p<0.001). There were no significant differences between groups with regard to smoking (information available in 99 cases), age, age at onset or duration of asthma.
Subject characteristics
When all airways were considered together (fig. 1a⇓), the thickness (ASMarea/PBM) of ASM was increased in the nonfatal (median 0.040 mm; interquartile range 0.013–0.051 mm; p = 0.034) and fatal cases of asthma (0.048 mm; 0.025–0.078 mm; p<0.001) compared with controls (0.036 mm; 0.024–0.042 mm). Fatal and nonfatal cases of asthma were not significantly different (p = 0.105). ASMarea/PBM increased with airway size and the relationships of ASMarea/PBM and asthma group for different airway sizes are shown in figure 1b–d⇓. In nonfatal cases of asthma, compared with controls, ASMarea/PBM was significantly increased in large (p = 0.001), medium (p = 0.003) and small airways (p = 0.001). In cases of fatal asthma, ASMarea/PBM was greater compared with controls for each airway size group (p<0.001 for all comparisons) and compared with nonfatal cases of asthma for large and medium airways (p<0.001), but not for small airways (p = 0.399). Univariate analysis showed an effect of centre, but not sex or cigarette smoking. ASMarea/PBM was associated weakly with age (r2 = 0.166; p = 0.041) and with duration of asthma (r2 = 0.226; p = 0.032), but not with age at onset of asthma (r2 = 0.081; p = 0.78). These relationships and differences between groups were similar when airways were grouped by size. Analysis by asthma group showed an effect of age (fig. 2⇓) in the nonfatal cases of asthma (r2 = 0.387; p<0.001). However no effect of age was apparent for controls or fatal asthma cases.
Area of the airway smooth muscle (ASMarea) relative to basement membrane perimeter (PBM) in controls (C) and cases of nonfatal asthma (NFA) and fatal asthma (FA) in: a) all airways combined; b) small airways (PBM <4 mm); c) medium airways (PBM 4–10 mm); and d) large airways (PBM >10 mm). Boxes represent median and interquartile range and whiskers represent 95% confidence interval. ○: outliers. ***: p<0.001; ###: p = 0.001; #: p = 0.034; ¶: p = 0.003.
Area of the airway smooth muscle (ASMarea) relative to basement membrane perimeter (PBM) by age for asthma cases categorised as control (▵), nonfatal asthma (•) or fatal asthma (▾).
The results of the general linear models are summarised in table 3⇓. The primary model explained 50% of the variance in ASMarea/PBM. Only subject group (control or nonfatal or fatal asthma) showed a significant effect on ASMarea/PBM. There was no significant effect of age, sex or smoking, and there were no significant interactions between age, asthma group or smoking. In separate models, no effects of duration of asthma or age at onset of asthma were observed on ASMarea/PBM. If PBM was included in the model, it had a strong independent effect on ASMarea/PBM (p<0.001), but did not significantly change the effects of other variables. When cases of nonfatal asthma were analysed separately, there was a significant but weak effect of age on ASMarea/PBM for all airways (r2 = 0.06; p = 0.044) and a stronger effect if the analysis was confined to adults aged 21–60 yrs (r2 = 0.255; p = 0.046). There was no effect of age in the fatal asthma or control groups.
Parameter estimates for general linear model for airway smooth muscle area/basement membrane perimeter
DISCUSSION
In the present study, the thickness of the ASM layer (ASMarea/PBM) was increased in cases of asthma compared with controls, and increased in cases of fatal asthma (clinically severe) compared with nonfatal cases of asthma (mild or moderate clinical severity). In multivariate analyses, there was no relation to duration of asthma, age at onset of asthma, smoking or sex. These findings suggest that the increased thickness of the ASM layer is present at an early stage in the natural history of asthma and may determine clinical severity.
We found a strong relationship between clinical severity and asthma group (nonfatal or fatal asthma). The underlying severity of asthma is generally constant over long periods 6, 7, and the data from the present study suggest that ASM thickness and asthma severity are associated. Most previous studies have examined ASM in cases categorised as fatal or nonfatal asthma, but have not reported grades of clinical severity 1. In mild-to-moderate cases studied during life, only small biopsy specimens can be obtained 3, 15. These studies have consistently shown that the relative ASMarea is increased in asthma, and at least three studies have shown that the amount of ASM distinguished mild/moderate from severe cases of asthma 3, 4, 16. The clinical severity of asthma in the present study was based upon GINA guidelines 13 and used available data on these post mortem cases. Data on lung function were not evaluable for enough subjects to be included in the model. Therefore, severity may have been generally underestimated since cases were categorised by their most severe criteria. The presence of lung function abnormalities would have placed more subjects into more severe categories. However, even in the absence of lung function data, the nonfatal cases that were scored as severe showed similar amounts of ASM to those that were mild to moderate (fig. E1 of online supplementary material).
The thickness of the ASM layer was found to be independent of age at onset and duration of asthma. This raises the possibility that the thickness of the ASM layer is determined early in the course of asthma. Although asthma severity is related to lung function 6, 7, the relationships between lung function, severity of asthma and ASM thickness could not be assessed in the present study. Others have found a relation between duration of asthma and lung function 17, and between age and airway wall thickness, with a similar trend for ASM 5. In the present study, a relationship between ASM thickness and age was found on univariate analysis in the nonfatal asthma group. There was a small overall effect of age in the same group in the general linear model. It is possible that the effects of age are small relative to the effect of severity and are, therefore, only evident in cases of mild-to-moderate severity.
The effects of age were not observed if PBM (the marker of airway size) was included in the model. Although we found that the results were similar if analyses were confined to specific airway size groups, there is potential for confounding by airway size since small airways have a thinner layer of ASM than do larger airways. Children, having smaller airways, have a thinner ASM layer in airways at the same anatomical site as adults 9. Although the present study included cases over a wide age range (10–>80 yrs), there were only 13 children aged <15 yrs, too few for separate analysis. Analyses confined to adults showed a stronger effect of age in the nonfatal asthma cases. Therefore, a small effect of ageing on the thickness of the ASM layer cannot be excluded. There was a small increase in ASMarea with age in the nonfatal asthma group. The lack of change with age in the control group suggests that ageing per se does not increase the thickness of the ASM. The effect in the nonfatal asthma group may have been apparent because subtle changes, possibly related to matrix deposition or ASM cell enlargement or proliferation over time, are discernible against a background of mild remodelling, whereas, in the fatal cases of asthma, these subtle changes may be masked by the marked remodelling seen with such cases.
The findings of the present study and the stability of asthma severity from early childhood to middle age seen in longitudinal studies suggest that the increased thickness of the ASM layer in asthma is an early event that changes little with the duration of asthma and possibly very little with age. The increase in the thickness of the ASM layer might occur at the time of, or prior to, the appearance of the first symptoms of asthma. One study has suggested that some of the inflammatory and remodelling changes (increased thickness of the basement membrane) in childhood asthma develop at an age of 2–5 yrs 18, and a recent study 19 has shown increased area fraction of ASM within the airway wall and increased numbers of ASM cells in children aged 8–14 yrs with asthma, compared with controls. Prospective studies are required to determine the origins of increased ASM thickness in asthma.
The data examined in the present study came from six centres, each of which collected, fixed, embedded and sectioned the tissue samples, and measured the airway dimensions in the sections (table 1⇑). Therefore, it is possible that the effect of the centre itself might account for differences between groups, and, on univariate analysis, there was a significant effect of centre on ASMarea/PBM. This effect was found to be due to differences in the sampling strategies used at each centre. Some centres opportunistically sampled predominantly small airways 8, 10, 11, whereas others systematically sampled large and small airways 2, 9, 12. The Perth (Australia) centre sampled all available small airways on tissue blocks, whereas the Calgary (Canada) centre sampled only small airways along three axial paths, resulting in relatively fewer small airways than large airways compared with the other centres. Differences between centres were present only in the small airways, which were affected by these sampling strategies 20. However, the relationship between ASMarea and PBM was almost identical for the Perth and Calgary centres, both of which sampled all available large and small airways 20. There were also differences in fixation, measurement and staining techniques between the centres. Some lungs were fixed by inflation and others without inflation. This has been shown to have no significant effect on airway wall area or PBM 20, 21. The use of point counts or planimetry gives the same result provided appropriate calibration is carried out and the same area of interest is measured. Centres that participated in the present study routinely calibrate measurement devices. The area of interest that is included in measurements of point counts or planimetry is affected by tissue thickness (due to overlap), magnification and tissue staining. Preliminary experiments determined that the ASM layer is equally well defined by different stains, and by both planimetry and point counts in tissue sections of 3–4 μm in thickness at a magnification of 40×. In all cases, the ASM layer was defined by the inner and outer edges of the muscle, rather than connective tissue.
Another potential limitation to the present study is the lack of clinical information on all subjects, as shown in table 2⇑. The general linear model, which included all subjects, examined the effects of age (but not duration or age of onset of asthma) and case group (control or nonfatal or fatal asthma) on the thickness of the ASM layer (ASMarea/PBM). Like the models limited to smaller numbers by the availability of data on duration of asthma, age of onset of asthma and smoking, no effect of time on ASM thickness was observed.
Smoking has been shown to adversely affect asthma in a number of ways. Subjects with asthma who smoke show reduced responses to corticosteroids 22, increased symptoms 23, reduced lung function and a greater decline in lung function 24, 25. However, we did not detect any effect of smoking on ASM thickness.
We also found no effect of age of onset or duration of asthma on the thickness of the ASM layer. Previous studies have shown an effect of duration of asthma on lung function 17. It has also been suggested that the length of time that asthma goes untreated is related inversely to lung function 26. Therefore, to the extent that fixed airflow obstruction reflects ASM thickness 16, these studies raise the possibility that ASM remodelling may increase with time, especially time without treatment. However, there are no prospective studies of ASM dimensions to support this view. In children with asthma, treatment with inhaled corticosteroids did not improve lung function decline 27. It has been shown that treatment with high-dose inhaled corticosteroids can reduce the thickness of the reticular basement membrane 27, 28. The degree to which treatment can reverse remodelling of the ASM is unknown. The presence of inflammation and fibrosis in asthma and the association of persistent inflammation with the development of fibrosis have led to the view that inflammation is the primary event in asthma, and results eventually in airway remodelling, including subepithelial deposition of extracellular matrix proteins, such as collagen, and increased ASM. The present study suggests the view that the increased ASM may be independent of the inflammation, or, at least, that persistent inflammation does not result in gradually increasing thickness of ASM in asthma.
In conclusion, although the thickness of the ASM layer is related to the clinical severity of asthma, it does not seem to be related to duration of asthma. This suggests that remodelling of ASM is an early event in the natural history of asthma and may be a factor that determines asthma severity.
Support statement
This study was supported by the National Health and Medical Research Council (NHMRC; Canberra, Australia) (ID 346301). The paediatric growth study (Sydney) and the Victorian Asthma Mortality Study were supported by the NHMRC. T.R. Bai was supported by the Canadian Institutes of Health Research (Ottawa, Canada) (grant No 1371894). The São Paulo asthma study was supported by the Brazilian National Research Council (Brasilia, Brazil). A.L. James was supported by the NHMRC (ID 446800).
Statement of interest
A statement of interest for M.J. Abramson can be found at www.erj.ersjournals.com/misc/statements.dtl
Acknowledgments
Tissue specimens from deceased cases were made available by the Office (Southbank, Australia) of the Victorian State Coroner, G. Johnstone, for the Victorian Asthma Mortality Study.
Footnotes
-
This article has supplementary material accessible from www.erj.ersjournals.com
- Received November 30, 2008.
- Accepted March 5, 2009.
- © ERS Journals Ltd
References
