Vinod H. Ratageri, 
S.K. Kabra, 
S.N. Dwivedi* and 
V. Seth

CHILDHOOD asthma is an important and common cause of morbidity in children. In a recent study the prevalence of asthma in different parts of India varied between 4-20%(1). There is worldwide increase in asthma particularly in children in last 2 decades(2-4). Although precise reasons for this increase are unknown, it is likely that a number of environmental factors are at least partly responsible(5). The occurrence of severe asthma appears to have increased concurrently, as reflected by increase in rates of asthma mortality(6). What determines severity of illness is not very clear. It has been suggested that development of asthma is decided by genetic predisposition but its severity is determined by the environmental factors(7). Hence the severity of asthma may be changed if the environmental and other acquired factors can be identified. There are very few published studies which relate the various clinical parameters to the degree of severity of asthma. We planned a prospective study to determine the association of various factors with development and severity of asthma in children aged 5-15 years.

This case-control study was carried out between May ’96 to April ’98 at Pediatric Chest Clinic of All India Institute of Medical Sciences (AIIMS), New Delhi. AIIMS is a tertiary care referral hospital getting patients from all over north India. The sample size of the study was calculated on the basis of previous analysis (unpublished) which showed a significant association of severe asthma with passive smoking, wood burning and lack of breast-feeding. The lowest odds ratio for these factors was 3 for wood burning and prevalence of exposure in control was 20%. For power of 80%, a sample size of 60 children in each group was calculated. There were three groups of children in the study: (i) children with severe asthma, (ii) children with mild asthma and (iii) children without asthma (controls).

All consecutive children with mild and severe asthma attending the Pediatric Chest Clinic and aged between 5-15 years who were residents of Delhi were enrolled in the study. Diagnosis and classification was based on guidelines given by International Pediatric Asthma Consensus Group(8). Asthma was defined as recurrent cough/wheezing (three or more episodes) with reversible airway obstruction with aerosolized salbutamol inhalation. Mild asthma was defined as children with low-grade symptoms that did not interfere with sleep or life style of child. Episodes of cough and wheeze were occurring less than once per month. All such episodes were responsive to bronchodilator taken no more frequently than 2 or 3 times per week. There was no effect on growth and development. Severe asthma was defined as children with troublesome wheezing on most days and night. Their day to day activity and growth was affected due to disease and required frequent courses of oral/injectable steroids. In this group growth and development was affected.

Controls were selected from the patients attending the Pediatrics Out Patient Department of our hospital during the study period for some other problems and did not have recurrent wheezing at any time in past.

Children where diagnosis of asthma was doubtful or who had complications such as bronchiectasis were not enrolled in the study. Children with acute exacerbation were included only after the acute episode was over and fresh assessment classified them into mild/severe asthma.

Children who fulfilled the inclusion criteria were enrolled in the study after obtaining a informed consent from the parents. After the enrolment, detailed history and physical examination was recorded in each child. An attempt was made to elicit detailed information regarding various risk factors for asthma in all children and recorded in predesigned proforma. Age of children was taken in full completed years. Education of parents was recorded in total number of years studied. Past history of tuberculosis, bronchiolitis, pertussis and chorinic diarrhea was obtained and confirmed by asking symptoms of illness, investigations, treatment received and by looking at the records available with the patients. Similarly, the history of allergy such as chronic or recurrent rhinitis or atopic dermatitis and drug exposure was elicited. A detailed general and systemic examination was done including anthropometry (height, weight). For height and weight records, local norms published recently(9) were used.

Laboratory tests including complete hemogram, absolute eosinophil count, X-ray films of chest and paranasal sinuses, stool exmaination for ova/cyst, Mantoux test and detailed lung function test using portable spirometer (Cosmed PONY graphic) were done in all the children enrolled in the study. For air pollution, we used values obtained from a study conducted by the Pollution Monitoring Club of the Center for Science Education and Communication, University of Delhi. We took values available from nearest place from child’s house and also from nearest place of child’s school.

For each group of patients and controls descriptive statistics namely, mean and median, standard deviation, minimum (Q1) and maximum (Q3) range were calculated for all the quantitative parameters including basic parameter as well as factors associated with asthma. For qualitative variable, simple frequency distributions were recorded in each group. For comparison of quantitative parameters between the groups (parametric/ non-parametric) one way analysis of variance and multiple range test was carried out to identify pair of groups having significant difference. To compare the distribution of all three groups in relation to qualitative parameter; Chi square test/ Fisher exact test was applied. Results giving p values of less than 0.05 were considered significant. Exploratory multivariate analysis, mainly logistic regression analysis was carried out to assess the relationship between the factors associated with mild asthma and severe asthma after adjusting the effect of other parameters.

The total number of new cases registered in Pediatric Chest Clinic during the study period (May 96 to April 98) were 900 of which 711 (79%) children were diagnosed as asthma. Among these, children less than 5 years with asthma were 225 (32%) and children of 5-15 years of age group with asthma were 486 (68%). A total of 66 children with mild asthma and 80 with severe asthma were from Delhi and thus eligible for enrollment in the study. Six of mild asthma and twenty of severe asthma were not included in the study because either they were in acute exacerbation at the time of presentation and did not attend the clinic subsequently or were not able to perform pulmonary function test properly on spirometer.

The mean age of mild asthma, severe asthma and controls was 9.03, 9.23 and 10.08 years, respectively. There was no significant difference between the groups. There was male dominance in all three groups with male to female ratio of 2 : 1. There was no significant sex difference between the groups (Table I). The presenting complaints and precipitating factors in mild and severe asthmatic children are given in Table I.

Table I - Baseline Characteristics of Cases and Controls

Table II shows data for various risk factors associated with development of asthma and its severity. On univariate analysis, factors associated with development of asthma were family history of asthma (p = 0.003), lack of exclusive breastfeeding (p = 0.05), past history of bronchiolitis (p = 0.02) and tuberculosis (p = 0.03), associated allergic rhinitis (p = 0.04) and atopic dermatitis (p = 0.01).

Past history of tuberculosis was present in 11 patients (one child in control, 3 in mild asthma and 8 in severe asthma group). All these children were carrying records of positive Mantoux test and suggestive X-ray films (done by the attending pediatrician in past) suggesting pulmonary primary complex in past. No attempt was made to isolate acid fast bacilli from the sputum/gastric lavage of these children by the treating pediatrician. None of these children were assessed to have tuberculosis on X-ray film done at the time of enrolment in the study. Positive Mantoux test with PPDS 5 TU (induration more than 10 mm) was recorded in all of them at the time of enrollment. In addition to these 11 children, one control also had a positive Mantoux test without past history of tuberculosis. Tuberculosis was diagnosed and treated at about 36-60 months (mean 44 months) before presentation to our hospital. Eight children received 2 antitubercular drugs (isoniazid and rifampicin) for at-least 6 months (range 6-9 months). Four children received 3 drugs (isoniazid, rifampicin, pyrazinamide) for at least 4 months (range 4-7 months).

Factors associated with severe asthma were early onset of symptoms (p = 0.01), family history of asthma in grandparents (p = 0.04) and more than 10 cigarettes/day smoked by any family member (p = 0.02). No significant effect of air pollution, over crowding, pets, main cooking medium and passive smoking were found on either development of asthma or its severity (Table II).

Table II - Factors Associated with Severity of Asthma (Univariate Analysis)

Mean absolute eosinophil counts in control, mild and severe groups were 245, 533 and 687/mm3 respectively. There was significant difference between cases and controls. Though, there was an increase in counts in severe group, there was no significant difference between mild and severe groups.

There was no significant difference among cases and controls that were exposed to more than permissible levels of sulfur dioxide and nitrogen dioxide around home. There were increased number of children in severe group (38.3%) exposed to more than permissible levels of suspended particulate matter (SPM) than controls (25%) and mild group (23.3%) but statistically it was not significant (Table III). Similarly, around school, there was no significant difference between cases and control exposed to more than permissible level of sulfur dioxide, nitrogen dioxide and suspended particulate matter (Table III).

Table III - Air Pollution Levels Around Home/School in Cases and Controls

On multivariate analysis, only age of onset below 48 months was associated with severe asthma (OR 2.13, 95% CI, 1.00-4.54). Exclusive breastfeeding for >4 months was found to be protective factor for development of asthma (OR 0.25, 95% CI 0.08-0.70). Associated allergy (OR 7.5, 95% CI 1.64-34.4) and past history of tuberculosis/bronchiolitis (OR 5.26, 95% CI 1.70-16.20) were associated with development of asthma (Table IV).

Table IV - Multivariate Analysis for Factors Associated with Development of Asthma and its Severity

From our study, on univariate analysis, we found that factors associated with severity of disease were early age of onset of symptoms, family history of asthma in grandparents and number of cigarettes smoked (>10 cigarette/day) by any family member. The factors associated with development of asthma included family history of asthma, absence of exclusive breastfeeding, associated allergic disorders such as atopic dermatitis and allergic rhinitis, past history of tuberculosis and bronchiolitis. However, on multivariate analysis only factor found to be significant with severity of asthma was age of onset below 48 months. For development of asthma, absence of breast-feeding, past history of tuberculosis/bronchiolitis and associated allergic disorders (atopic dermatitis and rhinitis) emerged as significant factors. We included children with mild and severe asthma to select extremes of the spectrum of disease and to minimize the misclassification of disease. The chances of misclassification are more with moderate asthma due to overlap of symptoms.

Our results suggest that earlier the onset of symptoms more severe was the disease. In literature most studies showed no correlation between age of onset of wheezing in children and persistence of symptoms into later childhood/adulthood or severity(10). Lebowitz et al. have shown that children with the onset of symptoms later in childhood were at a higher risk for ongoing symptoms than those with onset in first two years of life(11).

In our study the effect of age on severity of asthma was significant even on multivariate analysis. One explanation for this finding in our study is that earlier the onset of asthma there are more chances of getting these children sensitized to various allergens. Sensitization to various allergens have been demonstrated to be associated with chronic and severe disease(12-14). However, a possibility of selection bias cannot be ruled out completely. Children with early onset of asthma that was chronic got selected in our study. To remove this bias a prospective follow-up of birth cohort may be required.

Results of the present study suggest that absence of exclusive breastfeeding was associated with development of asthma. Zeiger et al. critically evaluated 16 studies out of which 9 prospective studies showed benefit and 7 showed lack of effect(15). A recent study by Wright et al. has demonstrated that breast-feeding was most protective against wheezing lower respiratory tract illness early in life(16). Breastfeeding may decrease allergic sensitiza-tion by reducing both exposure and intestinal absorption of food allergens. The protective role of human breast milk immunoglobulins, especially serum IgA, in inhibiting absorption of antigenic substances has been documented in human neonate(17).

Past history of pulmonary tuberculosis was associated with development of asthma but not with severity. In order to decrease false diagnosis we considered a child to have past history of tuberculosis when a written record of suggestive X-ray film and reactive Mantoux test of more than 10 mm were available. These tests were carried by the pediatricians who attended the children before coming to our hospital. None of the children carried a record of positive AFB in sputum or gastric lavage. All of them had positive Mantoux test and normal X-ray film of chest at the time of enrollment in the study. However, in the absence of definite evidence of infection with Mycobacterium tuberculosis it is difficult to interpret its causal association with development of asthma. There are no studies to link past infection with Mycobacterium tuberculosis and asthma. In a study on 200 children presenting with acute wheeze, 20% of the patients were found to have active tuberculosis(18). Another study on bronchial challenge with Purified Protein Derivative (PPD) suggested persence of Type I Arthus type of IgE mediated allergic reaction to PPD in 59% in asthmatic and 12% in children with pulmonary tuberculosis. None of the children with extra pulmonary tuber-culosis and controls had abnormal bronchial challenge to PPD. On the basis of positive bronchial challenge with PPD and detection of PPD specific IgE in serum of children with asthma and pulmonary tuberculosis, it was concluded that Arthus type IgE mediated reaction was responsible for hyperreactivity of airways in some children with asthma(19). In a recent report, Shirakawa et al. observed an inverse association between tuberculin response and atopic disorders(20).

In our study bronchiolitis was associated with development of asthma. Numerous studies both retrospective and prospective have been performed evaluating the persistence of both clinical airway symptoms and the presence of functional airway abnormalities later in life(21-24). All these suggested a positive relation between bronchiolitis during infancy and development of recurrent wheezing in later life.

A strong association between history of allergic disease such as recurrent rhinitis/atopic dermatitis and development of asthma was found in the present study. Similar results have been reported in other studies(5,23).

In the present study pet ownership was found more often in severe group than in controls but it was not statistically significant. However, pets were present in only a small number of house-holds. A larger sample size may be required to study the effect of pets on severity of asthma. Strachan et al. showed that pet ownership was associated with severe asthma(25).

There was no significant effect of air pollution in either development of asthma or severity. But in this study we found that there was an increase in number of children exposed to more than permissible level of SPM in severe group but it was not statistically significant. A report on health of children in six cities in United States of America demonstrated an association of chronic cough and bronchitis with measures of particulate matter and less strongly with sulfur dioxide and nitrogen dioxide. However, no association was found with development of asthma(26). Similar results were found in another study from Germany(27). A recent study by ISAAC group suggests that air pollution is not a major risk factor for the development of asthma although it may exacerbate asthma in individuals(1).

There are certain limitations of present study. This is a hospital-based study at a tertiary care hospital leading to selection of more severe disease. We tried to minimize this selection bias by using the definition of disease strictly. Secondly, the sample size was small. To see the significant difference for factors with low prevalence such as pets at home, SPM in air and wood burning, a study on larger number of patients is required. Thirdly, for seeing the effect of environmental pollutants like SO2, NO2 and SPM, we used the data provided by Pollution Monitoring Club. We took the avail-able data for nearest point around school and home. Ideally air pollutant’s level should be done at home and school during the study period.

Contributors: VHR and SKK designed the study, collected and analyzed the data and prepared the manuscript and will act as guarantor for the paper. SND helped in designing the study and analysis of data. VS helped in designing the study.

Funding: None.
Competing interests: None stated.