Clinical characteristics of the three subject groups are shown in Table 1. Sputum eosinophils were insignificantly increased in patients with CVA compared with those with NAC (p = 0.076). Log Dmin was significantly lower in the CVA group than in the NAC group (0.23 ± 0.73 U vs 1.40 ± 0.49 U, p < 0.0001). Log C5 was not significantly different between 11 patients with CVA and 9 patients with NAC (GERD, n = 3; SBS, n = 2; PICC, n = 1; unexplained cough, n = 3) [Table 1].
Representative CT images are shown in Figure 2. In the CVA group, all three indexes of airway (read about airway remodeling) wall thickness were significantly greater than in control subjects (Table 2). WA/BSA and T/VBSA in the NAC group were also significantly greater than in control subjects, and WA% was insignificantly increased. WA/BSA (Fig 3) and T/VBSA in the NAC group were less than in CVA group (Table 2). There was no difference in Ai/BSA among the three groups. When the NAC patients were separately analyzed by the cause of cough, there was a significant difference in WA/BSA and T/VBSA among control subjects and subgroups of NAC (p = 0.016 and p = 0.003, Kruskal-Wallis test). SBS and unexplained cough subgroups showed a significant increase in WA/BSA (Fig 4) and T/VBSA (p < 0.01 and p < 0.05, Dunn posttest) compared with healthy control subjects. PICC and GERD subgroups were not different from control subjects.
In the patients with NAC who underwent capsaicin cough sensitivity test, C5 was inversely correlated with WA% (n = 9, r = — 0.75, p = 0.034) [Fig 5; Table 3]. Other clinical indexes, such as age, disease duration, pulmonary function, and sputum cell differentials, did not correlate with any of the CT indexes in either group of cough patients (Table 3).
Discussion
This is the first study that has quantitatively evaluated airway wall thickness using CT in patients with CVA and those with NAC. Airway walls were thickened in patients with CVA, and to a lesser degree in those with NAC compared with healthy control subjects. Cough sensitivity was significantly correlated with one index of wall thickness in patients with NAC.
Our present result that the whole airway wall was thickened in patients with CVA compared to control subjects is consistent with our previous endobronchial biopsy studies that have shown airway eosinophilic inflammation, and thickening of airway subbasement membrane in patients with CVA. Furthermore, based on our findings of increased vascularity, vessel size, and goblet-cell area in CVA in comparison with healthy control subjects, the whole airway wall thickening on CT in patients with CVA may be ascribed to these structural changes.
Video below provides information about CVA:
The airway walls of patients with NAC (SBS, n = 8; GERD, n = 3; PICC, n = 3; GERD and SBS, n = 1; and unexplained [idiopathic] chronic cough, n = 11) were also thickened compared with those of healthy control subjects in this study. Previous biopsy studies have denoted the presence of airway inflammation and remodeling in patients with NAC. Boulet and colleagues described increase in epithelial shedding, mononuclear cells, submucosal fibrosis in the bronchial biopsy sample taken from patients with NAC that could be due to GERD (n = 6), postnasal drip syndrome (n = 4), both conditions (n = 5), and undiagnosed condition (n = 4). Our study revealed an increase of submucosal mast cells, subbasement membrane thickening, hypervascularity, goblet-cell hyperplasia, and an increase in airway smooth-muscle area in patients with NAC (postnasal drip/rhinitis, n = 6; GERD, n = 5; bronchiectasis, n = 3; and unidentified condition, n = 19). Our finding of airway wall thickening in NAC is presumably explained as a net result of these pathologic changes, as is in CVA. In addition to the wall thickening in the NAC group as a whole, patients with SBS and those with unexplained chronic cough separately analyzed showed thickened airway walls when compared with healthy control subjects. Lack of significant difference in wall thickness between control subjects and GERD or PICC subgroups may be due to small sample size. To our knowledge, this is the first attempt to clarify the airway wall dimensions as assessed by CT in various causes of chronic cough. Although PICC and GERD subgroups were small, we did not exclude these subgroups or another patient having both GERD and SBS from analysis, since they are considered important causes of NAC. Further studies involving larger number of patients are needed.
Cough hypersensitivity is a characteristic feature of chronic dry cough, especially in patients without airway hyperresponsiveness. Thus far, only one study has investigated on an association of airway structural changes with cough sensitivity. In that study, goblet-cell hyperplasia and epithelial shedding were associated with cough hypersensitivity in patients with NAC. In the present study, log C5 was significantly correlated inversely with WA% in patients with NAC who underwent cough sensitivity testing (n = 9, r = — 0.75, p = 0.034). Although we cannot draw a definitive conclusion because half of the patients with NAC were missed from analysis, and no similar correlation was observed in the CVA group, this association in NAC group may propose a possibility that coughing itself causes airway wall thickening, or conversely airway wall thickening and growth factors involved in the process lead to cough hypersensitivity in NAC. We consider the lack of association in CVA may be due to less involvement of cough hypersensitivity in the pathophysiology of CVA than in NAC, as has been suggested by previous studies.
For the first time to our knowledge, we have shown that the increase of airway wall thickness in CVA group was greater than in NAC group. Airway luminal area was not different between the two groups. The greater wall thickening in CVA may have been resulted from thickened outer airway wall that is reported in case of fatal asthma but cannot be assessed by endobronchial biopsy. This may not be plausible in CVA, a milder phenotype of asthma, however. The difference in the whole airway wall thickness between the two conditions cannot be explained by our biopsy study in which most of the airway components of asthmatic coughers and NAC patients show similar degrees of remodeling changes. Although we need to consider possible disparities between the previous biopsy study and the present CT study, such as causes, duration, and severity of cough, this difference between the two groups is important when the development of airway remodeling in asthma is considered, since the two chronic cough groups may be indistinguishable symptomatically and functionally except the presence or absence of airway hyperresponsiveness.
Values of WA/BSA of healthy control subjects (14.8 ± 3.8 mm2/m2) in the present study were higher than those of control subjects shown in our earlier study (11.2 ± 3.0 mm2/m2). In the present study, we used our automatic program that started from seeding a pixel in the airway lumen followed by using a full-width, half-maximal method. In some of our earlier studies, inner and outer edges of the airway walls were manually traced to obtain outer airway area and Ai on a workstation. The discrepancy between the present and previous studies may be mostly due to the methodologic difference. We previously compared WA/BSA measured automatically and WA/BSA measured manually in 26 asthmatic patients (13 men; age, 55 ± 15 years; FEV1; 90.7 ± 20.6% of predicted). The correlation between WA/BSA automated and WA/BSA manual was strong (r = 0.89, p < 0.0001). However, there was a difference between the two measurements (WA/ BSA automated, 20.8 ± 6.9 mm2/m2; WA/BSA manual, 17.3 ± 5.9 mm2/m2), which may explain the discrepancy in WA/BSA in the current and earlier studies. When employing the same automatic program, patients with CVA in the present study who were all symptomatic and untreated showed similar values of WA/BSA (22.8 ± 6.7 mm2/m2) to those of steroid-nai’ve, mild asthmatic patients (21.1 ± 7.6 mm2/m2).
As a limitation of the present study, mechanisms of the whole airway wall thickening in CVA and NAC were not clarified. Association of eosinophilic airway inflammation to wall thickening of the large airway has not been demonstrated in a recent study of nonasthmatic eosinophilic bronchitis patients, neither was it in our study, in which sputum eosinophil count did not correlate with airway wall dimensions in either group of patients. It is also unclear to what extent reversible components contributed to the whole airway wall thickening in both conditions. Further studies, such as changes of the wall thickness in response to inhaled corticosteroid therapy, might give some insights into this issue. Finally, as a possible common mechanism of the airway wall thickening in chronic coughers, bouts of coughing may play a role, as was mentioned above. Accumulated evidence suggests that exposure to a variety of mechanical forces stimulates cell growth and modulates extracellular matrix deposition.
We conclude that central airway wall is thickened in patients with CVA and, to a lesser degree, in those with NAC. Wall thickening may be associated with cough hypersensitivity in NAC. Present findings give insights into the development of airway remodeling and its physiologic consequence in chronic cough and asthma. Other interesting articles about studies connected with asthma you may find on our site https://onlineasthmainhalers.com.
Figure 2. Representative CT images of the selected bronchus (arrows). Top left, A: Healthy control subject. Top right, B: Chronic cough due to GERD. Bottom, C: CVA.
Figure 3. WA/BSA in healthy control subjects (HC), patients with CVA, and patients with NAC. Bars indicate means.
Figure 4. WA/BSA in healthy control subject and subgroups of NAC; p = 0.016, Kruskal-Wallis test; *p < 0.05, Dunn posttest. Bars indicate medians. Unx = unexplained chronic cough. See Figure 3 legend for expansion of abbreviation.
Figure 5. Correlation between WA % and cough sensitivity in patients with NAC.
Table 1—Characteristics of the Three Subject Groups
| Characteristics | CVA(n = 27) | NAC (n = 26) | Healthy Control (n = 15) | p Value# |
| Male/female gender, No. | 8/19 | 7/19 | 7/8 | NS |
| Age, yr | 58 ± 14 | 55 ± 15 | 52 ± 13 | NS** |
| Disease duration, yr | 5.2 ± 8.2 | 2.6 ± 3.6 | ||
| Ex-smokerf | 4 (15) | 5 (19) | 2(13) | NS |
| IgE, IU/mL | 218 ± 419 | 64 ± 88 | NE | NS |
| Atopic/nonatopic, No. | 12/15 | 9/ | NE | NS |
| Sputum cell differentials, %{ | ||||
| Eosinophils | 3.5 ± 6.2 | 0.4 ± 0.6 | NS | |
| Neutrophils | 70.9 ± 15.7 | 67.4 ± 21.6 | NS | |
| Lymphocytes | 0.26 ± 0.32 | 0.18 ± 0.21 | NS | |
| Macrophages | 24.6 ± 17.0 | 30.5 ± 22.4 | NS | |
| FEV1/FVC, % | 80 ± 8.2 | 81 ± 6.8 | 81 ± 5.8 | NS** |
| FEV1, % predicted | 107 ± 17 | 108 ± 13 | 108 ± 11 | NS** |
| Log C5, ^mol/L | ||||
| Men§ | 1.37 ± 0.75 | 1.52 ± 0.90 | NE | NS |
| Women|| | 1.00 ± 0.80 | 0.51 ± 0.62 | NE | NS |
Table 2—Outcomes of CT Indices
| Variables | CVA (n = 27) | NAC (n = 26) | Healthy Control (n = 15) | p Value | |||
| Analysis of Variance | Multiple Comparisons | ||||||
| CVA vs Healthy Control | NAC vs Healthy Control | CVA vs NAC | |||||
| WA/BSA, mm2/m2 | 22.8 ± 6.7 | 19.7 ± 4.8 | 14.8 ± 3.8 | 0.0002 | < 0.0001 | 0.008т | 0.048 |
| WA%, % | 65.0 ± 8.1 | 62.0 ± 5.7 | 58.1 ± 7.5 | 0.023 | 0.004 | 0.099 | 0.12 |
| T/VBSA, mm/m | 1.4 ± 0.2 | 1.2 ± 0.2 | 1.0 ± 0.2 | < 0.0001 | < 0.0001 | 0.001 | 0.006 |
| Ai/BSA, mm2/m2 | 12.9 ± 6.4 | 11.0 ± 3.8 | 12.5 ± 4.5 | 0.74 | |||
Table 3—Correlation Coefficients (r) Between CT and Clinical Indices
| Variables | CVA | NAC | ||
| WA/BSA | WA% | WA/BSA | WA% | |
| Age | 0.28 | 0.02 | 0.13 | 0.08 |
| Disease duration | 0.19 | – 0.22 | – 0.20 | 0.05 |
| Sputum eosinophils* | – 0.10 | – 0.40 | 0.34 | – 0.08 |
| FEVj/FVC | – 0.31 | 0.03 | – 0.18 | 0.13 |
| Log Dmin | – 0.31 | – 0.27 | – 0.06 | – 0.03 |
| Log C5| | – 0.06 | – 0.16 | 0.11 | – 0.75§ |