Keywords
atopy - children - EVW - MTW - phenotype - preschool - wheeze
Introduction
Preschool wheezers are twice more likely to visit hospitals and thrice more likely
to be admitted compared with older children.[1] An essential requirement for better management of these children is to classify
them as either episodic viral wheezers (EVWs) or multiple trigger wheezers (MTWs)
based on European Respiratory Society (ERS) Task Force recommendations 2008.[2] Although these phenotypes were found by some authors to change over time,[3]
[4]
[5] a recent study by Spycher et al indicates that they remain stable.[6] A significant number of them, particularly the MTW, are likely to have symptoms
consistent with asthma and reduced lung capacity later in life.[7]
[8] The identification of these phenotypes help in predicting long-term outcomes and
also to identify high-risk children who might benefit from secondary preventive interventions.
There are only few studies comparing the clinical and sociodemographic profiles of
these two phenotypes and none among Indian children. Hence, this study was performed
to identify the relative frequency of EVW and MTW in preschool children 1 to 5 years
of age with wheezing and to compare the relevant clinical and sociodemographic parameters
in the above phenotypes.
Methods
This cross-sectional study was conducted in the Department of Pediatrics, Jawaharlal
Institute of Postgraduate Medical Education and Research, Puducherry, a tertiary care
teaching institute from December 2015 to December 2017 after obtaining approval from
our institute ethics committee. Preschool children (1–5 years of age), who were on
follow-up in the childhood asthma clinic for at least 3 consecutive months and diagnosed
by a doctor to have recurrent wheeze participated in this study. Children with pneumonia,
history suggestive of gastroesophageal reflux disease, congenital heart disease, stridor,
history suggestive of foreign body, bronchopulmonary dysplasia, bronchiectasis, immunodeficiency,
and anatomic abnormalities leading to recurrent aspiration were not enrolled.
We recruited 165 children, who fulfilled the inclusion criteria during the study period
([Fig. 1]). The participants were classified into EVW and MTW as per ERS Task Force guidelines
2008. Children who had history of wheezing during discrete time periods, following
viral infections, with absence of wheeze between episodes were categorized as EVW.
Children who had history of discrete exacerbations of wheeze and also had symptoms
in between episodes of viral infections were categorized as MTW.[2]
For all participants, data on rural or urban location and parental literacy were recorded.
Socioeconomic status (SES) was classified based on modified Kuppuswamy scale 2014.[9] History regarding the age of onset, frequency of symptoms, bronchiolitis in the
past, frequency of wheeze exacerbations, triggers, seasonality, personal and family
history of atopy, exposure to environmental smoke, prematurity, low birth weight,
neonatal respiratory distress, breastfeeding status, immunization status, investigation,
and treatment details were noted. Weight and height were measured and classified according
to the World Health Organization guidelines.[10] Symptom control was assessed based on Global Initiative for Asthma (GINA) 2015 guidelines.[11] In our study, we classified the well-controlled symptom group in GINA 2015 as “well-controlled
symptom group” and the partly and uncontrolled groups together as “uncontrolled symptom
group.” The Institute Ethics Committee—Human Studies of our institute approved the
study (JIP/IEC/2015/22/776).
Statistical Analysis
Continuous variables were expressed as median with interquartile range (IQR). Categorical
variables were expressed as proportions. Mann–Whitney's U test was done for quantitative variables and chi-square test was used for categorical
variables. Multiple logistic regression analysis was done for factors associated with
symptom control for both EVW and MTW and also for all the participants combined. For
statistical analysis, Statistical Package for the Social Sciences (SPSS) version 19
by International Business Machines (IBM) Corporation was used.
Results
The median age of the study population was 42 months with an IQR of 27.5 to 51.5.
Boys were 110 (66.7%) in number; children belonging to rural areas were 98 (59.4%).
Two children (1.2%) belonged to lower, 38 (23%) belonged to upper lower, 72 (43.8%)
belonged to lower middle, 35 (21%) belonged to upper middle, and 18 (11%) belonged
to upper socioeconomic classes. Majority were born at term (91.5%) with normal birth
weight (78.1%), were exclusively breastfed (92.1%) in the first 6 months, and were
completely immunized (97%) for age as per national schedule. EVW was seen in 91 (55%)
and MTW in 74 (45%) children. Median age was significantly higher in the MTW group.
The two groups did not significantly differ in breastfeeding status, immunization
status, prematurity, low birth weight, neonatal respiratory distress, bronchiolitis
in infancy, age of onset, and duration of treatment ([Table 1]). The number of children on montelukast was higher in EVW group; those on inhaled
corticosteroids (ICSs) were higher in MTW group and this finding was statistically
significant. A significantly higher proportion of parental atopy and high absolute
eosinophil count (AEC) was seen in children with MTW. Seasonality of exacerbation
was present in 35 (47.3%) children with MTW. Viral infections in 67 (90.5%) children,
dust allergy in 46 (62.1%) children, environmental smoke exposure in 23 (31.1%) children,
and food allergens in 21 (28.3%) children were the common triggers identified in MTW
group. Out of 91 children with EVW, 79 (86.8%) had well-controlled symptoms, whereas
12 (13.2%) had uncontrolled symptoms. Among the 74 MTW, 58 (78.4%) had well-controlled
symptoms and 16 (21.6%) had uncontrolled symptoms. There was no significant difference
in the symptom control among EVW and MTW.
Table 1
General characteristics of the study population
|
Variable
|
Episodic wheezer n = 91 (%)
|
Multitrigger wheezer n = 74 (%)
|
p-Value (chi-square)
|
|
Median age in mo (interquartile range)
|
36 (24–48)
|
42 (36–54)
|
0.003[a]
|
|
Male gender
|
64 (70.3)
|
46 (62.2)
|
0.268
|
|
Rural residence
|
52 (57.1)
|
46 (62.2)
|
0.514
|
|
Socioeconomic status
|
Lower
|
2 (2.2)
|
0
|
0.772
|
|
Upper lower
|
19 (20.9)
|
19 (25.7)
|
|
Lower middle
|
39 (42.9)
|
33 (44.6)
|
|
Upper middle
|
21 (23.1)
|
14 (18.9)
|
|
Upper
|
10 (11)
|
8 (10.8)
|
|
Prematurity
|
8 (8.8)
|
6 (8.1)
|
0.876
|
|
Low birth weight
|
17 (18.7)
|
20 (27)
|
0.201
|
|
Exclusive breastfeeding for 6 mo
|
85 (93.4)
|
67 (90.5)
|
0.568
|
|
Median duration of breastfeeding in mo (interquartile range)
|
13 (12–18)
|
18 (12–18)
|
0.261[a]
|
|
Complete immunization status
|
88 (96.7)
|
72 (97.3)
|
0.463
|
|
Neonatal respiratory distress
|
5 (5.5)
|
9 (12.2)
|
0.126
|
|
Bronchiolitis in infancy
|
47 (51.6)
|
32 (43.2)
|
0.282
|
|
Median age of symptom onset in mo (interquartile range)
|
10 (6–18)
|
12 (5.75–25)
|
0.547[a]
|
|
Frequency of exacerbations
|
≤2 in 6 mo
|
17 (18.7)
|
8 (10.8)
|
0.161[b]
|
|
>2 in 6 mo
|
74 (81.3)
|
66 (89.2)
|
|
Controller medications
|
Montelukast
|
39 (42.9)
|
8 (10.8)
|
0.001[b]
|
|
ICS
|
49 (53.8)
|
66 (89.2)
|
|
None
|
3 (3.3)
|
0
|
|
Median duration of treatment in mo (interquartile range)
|
3 (3–6)
|
3.5 (3–8.25)
|
0.119[a]
|
|
Median absolute eosinophil count (interquartile range)[c]
|
292 (132.5–485)
|
850 (260–1,630)
|
0.001[a]
|
|
H/o atopic eczema
|
2 (2.2)
|
16 (21.6)
|
0.001[b]
|
|
H/o ocular/nasal allergy
|
5 (5.5)
|
34 (45.9)
|
0.001[b]
|
|
Parental atopy
|
12 (13.2)
|
44 (59.5)
|
0.001[b]
|
Abbreviations: H/o, history of; ICS, inhaled corticosteroid.
a
p-Value calculated using Mann–Whitney's U test.
b
pValue calculated using chi-square test.
c For 94 children.
On univariate analysis of factors associated with symptom control in EVW group ([Table 2]), it was found that children with either parent or both graduates had a higher percentage
of uncontrolled symptoms with an odds ratio (OR) of 8.5 (95% confidence interval [CI]:
2.2–32.1). Those with history of ocular/nasal allergies had more uncontrolled symptoms
(OR: 12.8, 95% CI: 1.8–87.3). Other factors such as gender, SES, anthropometric parameters,
prematurity, birth weight, breastfeeding, neonatal respiratory distress, bronchiolitis,
environmental smoke exposure, AEC, and type and duration of treatment were not significantly
associated. Among MTW ([Table 3]), only exclusive breastfeeding in the first 6 months of life predicted well-controlled
symptoms (OR: 6.1, 95% CI: 1.2–30.9).
Table 2
Sociodemographic, clinical, treatment-related factors, and comorbidities associated
with symptom control among episodic wheezers
|
Factors
|
Well controlled (n = 79)
|
Uncontrolled (n = 12)
|
Odds ratio (95% CI)
|
p-Value (chi-square)
|
|
Age > 3 y (%)
|
36 (45.6)
|
2 (16.7)
|
0.2 (0–1.1)
|
0.059
|
|
Male (%)
|
56 (70.9)
|
8 (66.7)
|
1.2 (0.3–4.4)
|
0.745
|
|
Rural (%)
|
46 (58.2)
|
6 (50)
|
0.7 (0.2–2.4)
|
0.592
|
|
Lower SES (%)
|
19 (24)
|
2 (16.7)
|
0.6 (0.1–3.1)
|
0.726
|
|
Father's education: graduate (%)
|
19 (24)
|
9 (75)
|
9.4 (2.3–38.6)
|
0.001
|
|
Mother's education: graduate (%)
|
19 (24)
|
8 (66.7)
|
6.3 (1.7–23.3)
|
0.005
|
|
Both parents' education: graduate (%)
|
15 (19)
|
8 (66.7)
|
8.5 (2.2–32.1)
|
0.001
|
|
Prematurity (%) (missing—1)
|
7 (9)
|
1 (8.3)
|
0.9 (0.1–8.2)
|
0.999
|
|
Low birth weight (%) (missing—4)
|
15 (20)
|
2 (16.7)
|
0.8 (0.1–4)
|
0.999
|
|
Exclusive breastfeeding (%)
|
75 (95)
|
10 (83.3)
|
3.7 (0.6–23.1)
|
0.177
|
|
Partial immunization (%)
|
3 (3.8)
|
0
|
NA
|
0.999
|
|
Neonatal RD (%)
|
4 (5)
|
1 (8.3)
|
1.7 (0.1–16.6)
|
0.515
|
|
Bronchiolitis in infancy (%) (missing—3)
|
40 (51.3)
|
7 (70)
|
2.2 (0.5–9.2)
|
0.327
|
|
Stunted (%)
|
6 (7.6)
|
2 (16.6)
|
2.4 (0.4–13.7)
|
0.284
|
|
Underweight (%)
|
20 (25.3)
|
2 (16.6)
|
0.6 (0.1–2.9)
|
0.723
|
|
H/o atopic eczema, rash, etc. (%)
|
2 (2.5)
|
0
|
NA
|
0.999
|
|
H/o nasal/ocular allergic symptoms (%)
|
2 (2.5)
|
3 (25)
|
12.8 (1.8–87.3)
|
0.015
|
|
Exposure to environmental smoke (%)
|
3 (3.8)
|
1 (8.3)
|
2.3 (0.2–24.1)
|
0.438
|
|
Family h/o allergy (%)
|
10 (12.6)
|
2 (16.7)
|
1.3 (0.2–7.2)
|
0.656
|
|
Treatment with inhaled corticosteroids (%) (missing—4)
|
43 (54.4)
|
6 (50)
|
1.3 (0.4–4.5)
|
0.634
|
|
Absolute eosinophil count > 500 (%) (missing—55)
|
5 (14.7)
|
0
|
NA
|
0.999
|
|
Duration of treatment > 3 mo (%) (missing—3)
|
31 (40.8)
|
5 (41.7)
|
1 (0.3–3.5)
|
0.999
|
Abbreviations: CI, confidence interval; H/o, history of; NA, not available; RD, respiratory
disease; SES, socioeconomic status.
Table 3
Sociodemographic, clinical, treatment-related factors associated with symptom control
among multitrigger wheezers
|
Factors
|
Well controlled (n = 58)
|
Uncontrolled (n = 16)
|
Odds ratio (95% CI)
|
p-Value (chi-square)
|
|
Age > 3 y (%)
|
37 (63.8)
|
11 (68.8)
|
1.2 (0.3–4)
|
0.713
|
|
Male (%)
|
37 (63.8)
|
9 (56.25)
|
1.3 (0.4–4.2)
|
0.582
|
|
Rural (%)
|
37 (63.8)
|
9 (56.25)
|
0.7 (0.2–2.2)
|
0.582
|
|
Lower SES (%)
|
14 (24.1)
|
5 (31.2)
|
1.4 (0.4–4.8)
|
0.538
|
|
Father's education: graduate (%)
|
15 (25.9)
|
3 (18.75)
|
0.6 (0.1–2.6)
|
0.746
|
|
Mother's education: graduate (%)
|
16 (27.6)
|
3 (18.8)
|
0.6 (0.1–2.4)
|
0.747
|
|
Both parents' education: graduate (%)
|
13 (22.4)
|
3 (18.8)
|
0.8 (0.2–3.2)
|
0.999
|
|
Prematurity (%)
|
5 (8.6)
|
1 (6.3)
|
0.7 (0–6.5)
|
0.999
|
|
Low birth weight (%) (missing—3)
|
18 (32)
|
3 (20)
|
0.5 (0.1–2.1)
|
0.527
|
|
Exclusive breastfeeding (%)
|
55 (94.8)
|
12 (75)
|
6.1 (1.2–30.9)
|
0.035
|
|
Partial immunization (%)
|
1 (1.7)
|
1(6.3)
|
3.8 (0.2–64.3)
|
0.388
|
|
Neonatal RD (%)
|
5 (8.6)
|
4 (25)
|
3.5 (0.8–15.1)
|
0.095
|
|
Bronchiolitis in infancy (%) (missing—1)
|
26 (45.6)
|
6 (37.5)
|
0.7 (0.2–2.2)
|
0.563
|
|
Stunted (%)
|
5 (8.6)
|
1 (6.3)
|
0.7 (0–6.5)
|
0.999
|
|
Underweight (%)
|
11 (19)
|
5 (31.3)
|
1.9 (0.5–6.7)
|
0.315
|
|
H/o skin atopy (%)
|
14 (24.1)
|
1 (6.3)
|
0.4 (0.1–2.2)
|
0.496
|
|
H/o nasal and ocular allergy (%)
|
26 (44.8)
|
8 (50)
|
1.2 (0.4–3.7)
|
0.713
|
|
Exposure to environmental smoke (%)
|
19 (32.8)
|
4 (25)
|
0.6 (0.2–2.4)
|
0.762
|
|
Family h/o allergy (%)
|
36 (62)
|
8 (50)
|
0.6 (0.2–1.8)
|
0.384
|
|
Treatment with inhaled corticosteroids (%) (missing—1)
|
50 (87.7)
|
15 (93.7)
|
0.4 (0–4.1)
|
0.676
|
|
Absolute eosinophil count > 500 (%) (missing—41)
|
19 (73)
|
3 (43)
|
0.2 (0–1.5)
|
0.186
|
|
Duration of treatment > 3 mo (%) (missing—1)
|
28 (49.1)
|
9 (56.3)
|
1.3 (0.4–4)
|
0.614
|
Abbreviations: CI, confidence interval; H/o, history of; RD, respiratory disease;
SES, socioeconomic status.
On univariate analysis among all participants ([Table 4]), wheezers with both parents being graduates had higher chance of having uncontrolled
symptoms (OR: 2.5, 95% CI: 1–5.9), and so were those who had nasal/ocular allergy
(OR: 2.5, 95% CI: 1–5.9) and those not exclusively breastfed (OR: 5, 95% CI: 1.5–16.4).
On multiple (binomial) logistic regression in the EVW group, absence of ocular/nasal
allergies predicted well-controlled symptoms (OR: 20.4, 95% CI: 1.9–216.6). On multiple
(binomial) logistic regression analysis among preschool wheezers in general, exclusive
breastfeeding in the first 6 months of life predicted well-controlled symptoms (OR:
4.1, 95% CI: 1.2–14.4).
Table 4
Sociodemographic, clinical, treatment-related factors, and comorbidities associated
with symptom control among wheezers in general
|
Factors
|
Well controlled (n = 137)
|
Uncontrolled (n = 28)
|
Odds ratio (95% CI)
|
p-Value (chi-square)
|
|
Age > 3 y (%)
|
73 (53.2)
|
13 (46.4)
|
0.7 (0.3–1.7)
|
0.508
|
|
Male (%)
|
93 (67.8)
|
17 (60.7)
|
1.3 (0.6–3.1)
|
0.463
|
|
Rural (%)
|
83 (61.3)
|
15 (53.5)
|
0.7 (0.3–1.7)
|
0.491
|
|
Lower SES (%)
|
33 (24)
|
7 (25)
|
1 (0.4–2.7)
|
0.918
|
|
Father's education: graduate (%)
|
34 (24.8)
|
12 (42.8)
|
2.2 (0.9–5.2)
|
0.052
|
|
Mother's education: graduate (%)
|
35 (25.5)
|
11 (39.2)
|
1.8 (0.8–4.4)
|
0.140
|
|
Both parents' education: graduate (%)
|
28 (20.4)
|
11 (39.2)
|
2.5 (1–5.9)
|
0.032
|
|
Prematurity (%) (missing—1)
|
12 (8.8)
|
2 (7.1)
|
0.8 (0.1–3.7)
|
0.999
|
|
Low birth weight (%) (missing—7)
|
33 (25.2)
|
5 (18.5)
|
0.6 (0.2–1.9)
|
0.460
|
|
Exclusive breastfeeding (%)
|
130 (95)
|
22 (78.5)
|
5 (1.5–16.4)
|
0.01
|
|
Partial immunization (%)
|
4 (2.9)
|
1 (3.5)
|
1.2 (0.1–11.4)
|
0.999
|
|
Neonatal RD (%)
|
9 (6.5)
|
5 (17.9)
|
3 (0.9–10)
|
0.06
|
|
Bronchiolitis in infancy (%) (missing—4)
|
66 (48.8)
|
13 (50)
|
1 (0.4–2.4)
|
0.917
|
|
Stunted (%)
|
11 (8)
|
3 (10.7)
|
1.3 (0.3–5.2)
|
0.709
|
|
Underweight (%)
|
31 (22.6)
|
7 (25)
|
1.1 (0.4–2.9)
|
0.786
|
|
H/o atopy (eczema, rash, etc.) (%)
|
16 (11.6)
|
2 (7.1)
|
0.5 (0.1–2.6)
|
0.741
|
|
H/o nasal and ocular allergic symptoms (%)
|
28 (20.4)
|
11 (39.2)
|
2.5 (1.–5.9)
|
0.032
|
|
Exposure to environmental smoke (%)
|
22 (16)
|
5 (17.8)
|
1.1 (0.3–3.3)
|
0.783
|
|
Family h/o allergy (%)
|
46 (33.5)
|
10 (35.7)
|
1. (0.4–2.5)
|
0.828
|
|
Treatment with inhaled corticosteroids (%) (missing—5)
|
93 (70.4)
|
21 (75)
|
0.8 (0.3–2)
|
0.629
|
|
Absolute eosinophil count > 500 (%) (missing—92)
|
24 (40)
|
3 (23)
|
0.4 (0.1–1.8)
|
0.348
|
|
Duration of treatment > 3 mo (%) (missing—4)
|
59 (44)
|
14 (50)
|
1.2 (0.5–2.8)
|
0.586
|
Abbreviations: CI, confidence interval; H/o, history of; RD, respiratory disease;
SES, socioeconomic status.
Discussion
From our study results, we find that EVW is the dominant phenotype. Children with
MTW were older than EVW, were mostly on ICS, and had more parental atopy, personal
atopy, and higher AEC. In the EVW group, symptoms were more likely to be significantly
well controlled in the absence of ocular or nasal allergy. In preschool wheezers as
a whole, exclusive breastfeeding in the first 6 months of life was associated with
well-controlled symptoms.
It is generally known that the proportion of MTW increases and EVW decreases over
time.[12]
[13] In a study involving 109 children in the age group of 2 to 6 years, Schultz et al
found that 35% had EVW and 65% had MTW at baseline and 31% had EVW, 47% had MTW, and
22% had no wheeze at 1-year follow-up.[4] We did not find significant difference in the age of onset, frequency of exacerbations,
or duration of treatment between EVW and MTW, but there was difference in the type
of treatment, AEC, and parental atopy. Mutti et al in their study on 55 children have
reported no significant difference between the age of onset, atopic status, and airway
eosinophilia in EVW compared with MTW.[14] Significantly higher AEC and family history of atopy were seen in MTW in our study
as MTW represents an atopic phenotype.[5]
[12]
[13]
[15]
[16]
[17]
[18] On univariate analysis, we found that more children whose parents were graduates
had uncontrolled symptoms. These children also had higher parental allergy and environmental
smoke exposure. Previous studies have found that parental allergies were independent
risk factors for developing persistent wheeze.[19]
[20] It is also known that lower parental education and low SES are associated with poorer
symptom control.[19]
[21] However, in our study, we find that parents who were graduates were a risk factor
for uncontrolled symptom on univariate analysis, but it was not confirmed in multiple
logistic regression indicating that the discrepancy may be probably due to confounders.
We also found that EVW without ocular/nasal allergy were more likely to have well-controlled
symptoms. It is a well-known fact that allergic rhinitis can have a negative effect
on symptom control.[22] Although atopy is more often associated with MTW, in our study, we found they may
be associated with EVW also and further serves to support the observations of the
ERS Task Force statement that there may be overlap between the two phenotypes.[2] Consistent with the conclusions of previous studies,[23]
[24]
[25] we also found that exclusive breastfeeding predicted well-controlled symptoms in
all preschool wheezers in general.
Fig. 1 Study flow diagram depicting the inflow of subjects into the study.
The limitations of our study are that the sample was drawn from an asthma clinic and
not from the community. It was a cross-sectional design and AEC was not available
for 92 children. Information regarding breastfeeding and immunization was obtained
based on history obtained from parents and information available in the case records
(for immunization).
Conclusion
To conclude, EVW is the dominant wheezing phenotype found in preschool children. Children
with MTW phenotype tend to be older than EVW phenotype and tend to have higher AECs.
Atopy in the form of ocular and nasal allergies can also be seen in children with
EVW and its absence predicts well-controlled symptoms. Exclusive breastfeeding predicts
well-controlled symptoms in children with preschool wheeze in general irrespective
of the phenotypes.
