Quantitative study of normal bronchial mucus glands.
Chronic bronchitis has been recognised as a major public health problem in U.K. since 1950. The formulation of a definition and the adoption of the standardized questionnaires developed by the Medical Research Council have allowed comparison of its prevalence in many different countries. There are however geographical variations in the prevalence of chronic bronchitis which persists even after accounting for the differences in smoking habits. These variations have been explained by variations in the level of air pollution and ethnic differences.
Most observers agree that the production of sputum is associated with enlargement of mucus glands. Various methods for quantitation of mucus gland hypertrophy are available , , , 
Estimation of Reid index provides quantitative data on mucus secreting elements. This study was undertaken to obtain such a baseline data of a population for future studies on chronic bronchitis, in subjects from Bombay.
Thirty six lung specimens were collected randomly from the autopsies performed at the King Edward VII Memorial Hospital, Bombay. The patients had no chronic respiratory disease as judged by the history and clinical symptoms. Sections were taken for the histological study from the main and lobar bronchi of the right and left lungs. Totally 7 sections were taken from each case. Paraffin sections, 5 u in thickness, were prepared and stained with haematoxylin and eosin. A point was chosen where the basement membrane of the mucosa and perichondrium were parallel, preferably close to the centre of the cartilage plate. The thickness of the bronchial wall as well as the mucus gland layer was estimated at this point using an ocular micrometer. On an average, three measurements were made for each bronchus. From these measurements gland to wall ratio was calculated. Mean of G: W ratio obtained at the three sites was taken to be the representative Reid index for that bronchus. While taking the measurements, recommendations of Thurlbeck and Angus were followed, in that, the glands at the edge of the cartilage were not measured.
All the cases were between 20 and 70 years of age. These were arbitrarily classified into three age groups to find out if age had any influence. The range of Reid indices obtained at each site in these groups is depicted in Fig. 1, while the mean index with S.D. and S.E. [Table 1].
Group I (age 20-30 years)
This consisted of 13 cases. The Reid indices in these cases ranged from 0.186 to 0.435. The mean index for the group, for each site, varied from 0.293 to 0.3-10 with the mean of 0.301.
Group II (age 36-50 years)
There were 12 cases in this group. The Reid indices varied from 0.142 to 0.530. The variation is little wider than the previous group, but the mean index for each site varied from 0.293 to 0.383 with a mean of 0.313. Thus the mean index is very close to the previous group. Group III (age 51 years and above)
This group comprised of 11 cases. The range of Reid indices observed was 0.134 to 0.552. The mean indices for the group varied from 0.264 to 0.381 with an average of 0.340. Though the range is as wide as the previous group the mean is higher than the previous group. It is however not statistically significant.
Age and Reid Index
Analysis of variance has shown that there is no significant increase in the Reid index with age though there is a tendency towards higher values with increasing age (F = 1.046; not significant).
Site and Reid Index
No statistically significant difference was found between the Reid indices of the main and lobar bronchi of any given patient though minor variation from site to site was noticed (F = 1.587; not significant). Combined analysis of variance for age and site also did not show significant variation (F = 1.126).
The subjects of this study were basically from the lower middle class with a wide range of occupations representing a city population in general.
The method adopted in the present study has certain drawbacks. Firstly, in many bronchi it is not very easy to find suitable areas where the cartilage is parallel to the mucus membrane. In the present series, however, this difficulty was encountered in only 6 out of 252 bronchi studied. Secondly, the point at which the measurement is made is arbitrary and the ratio may vary considerably from place to place in the bronchial wall. Thirdly, the method gives no indication of total quantity of gland tissue, much of which may be between the cartilage plates.
The Reid index, however, has the advantage that it is a quick method and gives results comparable with those obtained by Dunnill's point counting method. Scott noted a close correlation between these two methods. Thus, inspite of certain drawbacks the Reid index appears to be a quick and a reliable method which can be used in any laboratory.
The results in the present series show higher values than those noted by Reid for the nonbronchitic subjects. In her series the values for nonbronchitics varied from 0.14 to 0.36 with a mean of 0.26. In the present series it varies from 0.11 to 0.59 with a mean of 0.32. These higher indices could be because this study was done in a hospital which drains a population residing in an area with a high degree of air pollution. Takizawa and Thurlbeck reported still higher indices with a range of 0.31-0.58 and a mean of 0.42. Similar data is of available from any other city in India for comparison. However, clinical studies of Joshi et al have shown that the incidence of chronic bronchitis in an industrialised population of north India is comparable with areas of low air pollution in Europe and North America.
Significantly higher incidence of chronic bronchitis in smokers as compared to non-smokers has been reported, by many investigators., ,  In our study, the data regarding smoking habits of the subjects was not available, however, no significant case to case variation in the Reid indices was observed.
Hernandez and co-workers reported significantly higher Reid indices and absolute thickness of bronchial glands with increasing age. In the present series no statistically significant increase in the Reid index with age was noted, though a tendency towards higher values with increasing age was seen. This is in keeping with the results of Field and co-workers.
Restrepo and Heard concluded that considerable variation exists in the bronchial mucus glands within the same bronchial tree. They reported only absolute gland areas, ignoring the variations that occur as a result of difference in the size of the bronchi at different sites. The proportion of gland to bronchial wall could remain constant despite marked variation in absolute measurements if the bronchial wall size changed proportionately with the change in the glands. Thurlbeck and Angus measured the Reid index at several sites within the bronchial tree and concluded that although variation exists, one section in most cases can be representative of the entire bronchial tree. These findings are corroborated by the present study in which no statistically significant difference in the Reid indices of various sites was noted. On the whole, no particular site can be considered to be more representative than the other and sections from the mainstem bronchi alone may suffice for the evaluation of mucus gland hypertrophy. These bronchi are easiest to obtain at autopsy and are probably the appropriate ones to use in epidemiologic studies.
We thank Mrs. K. D. Lotlikar, M.Sc., Biostatistician, for carrying out statistical analysis of the data.