Sunteți pe pagina 1din 9

+ MODEL

Journal of Microbiology, Immunology and Infection xxx (xxxx) xxx

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.e-jmii.com

Original Article

Antibiotic exposure and asthma


development in children with allergic
rhinitis
Yi-Ching Lin a,b,c,d, Yen-Chun Chen b,e, Chang-Hung Kuo f,
Yu-Han Chang g, Hsin-Yi Huang g, Wei-Ju Yeh g, Ting-Yi Wu g,
Ming-Yii Huang h,i, Chih-Hsing Hung b,d,e,g,j,k,*

a
Department of Laboratory Medicine, Kaohsiung Medical University Hospital, Kaohsiung Medical
University, Kaohsiung, Taiwan
b
Department of Pediatrics, Kaohsiung Medical University Hospital, Kaohsiung Medical University,
Kaohsiung, Taiwan
c
Department of Laboratory Medicine, School of Medicine, College of Medicine, Kaohsiung Medical
University, Kaohsiung, Taiwan
d
Research Center for Environmental Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan
e
Department of Pediatrics, Kaohsiung Municipal Hsiao-Kang Hospital, Kaohsiung, Taiwan
f
Ta-Kuo Clinic, Kaohsiung, Taiwan
g
Teaching and Research Center of Kaohsiung Municipal Hsiao-Kang Hospital, Kaohsiung, Taiwan
h
Department of Radiation Oncology, Kaohsiung Medical University Hospital, Kaohsiung, Taiwan
i
Department of Radiation Oncology, College of Medicine, Kaohsiung Medical University, Kaohsiung,
Taiwan
j
Department of Pediatrics, School of Medicine, College of Medicine, Kaohsiung Medical University,
Kaohsiung, Taiwan
k
Graduate Institute of Medicine, College of Medicine, Kaohsiung Medical University, Kaohsiung,
Taiwan

Received 16 July 2018; received in revised form 8 December 2018; accepted 18 February 2019
Available online - - -

KEYWORDS Abstract Purpose: Early-life antibiotic use may be associated with asthma, yet whether this
Asthma; association also exists in patients with allergic rhinitis (AR) remains unknown. We investigated
Allergic rhinitis; the association between antibiotic exposure and asthma development in AR children.
Antibiotics; Methods: AR patients less than 18 year-old were enrolled from the Taiwan National Health In-
Children surance Database, which reported information from 2005 to 2010. The case group was defined
as having newly developed asthma, and the control group was defined as never having an
asthma diagnosis. The age of first exposure to antibiotic prescriptions and antibiotic exposure

* Corresponding author. Department of Pediatrics, Kaohsiung Medical University Hospital, Kaohsiung Medical University, #100, Tz-You 1st
Road, Kaohsiung, 807, Taiwan. Fax: þ886 7 3213931.
E-mail address: pedhung@gmail.com (C.-H. Hung).

https://doi.org/10.1016/j.jmii.2019.02.003
1684-1182/Copyright ª 2019, Taiwan Society of Microbiology. Published by Elsevier Taiwan LLC. This is an open access article under the CC
BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Please cite this article as: Lin Y-C et al., Antibiotic exposure and asthma development in children with allergic rhinitis, Journal of
Microbiology, Immunology and Infection, https://doi.org/10.1016/j.jmii.2019.02.003
+ MODEL
2 Y.-C. Lin et al.

records preceding 5 years before the first asthma diagnosis were obtained from drug prescrip-
tion records. The odds ratio (OR) was examined after adjusting for age, gender, resident ur-
banization, underlying medical disorders and medications.
Results: A total of 3236 AR patients with newly developed asthma and 9708 AR patients
without asthma were included in this study. Antibiotic exposure before the age of 3 years
was not associated with asthma development. Preceding 5-year antibiotic exposure increased
the risk of asthma development with a dose-response relationship, even for antibiotics with
low cumulative defined daily doses (adjusted OR 1.40, 95% CI 1.12e1.75). Preceding 5-year
exposure to penicillin and macrolide significantly increased the risk of asthma when diagnosed
before age 12 in AR patients, but this was not statistically significant when asthma diagnosed
after age 12.
Conclusion: Preceding 5-year antibiotic exposure, particularly to penicillin group of amoxicillin
and macrolides, is associated with the risk of asthma development before age 12 in AR chil-
dren.
Copyright ª 2019, Taiwan Society of Microbiology. Published by Elsevier Taiwan LLC. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).

Introduction independent risk factor for asthma in the primary care


study population, and patients with physician-diagnosed AR
Childhood asthma and rhinitis are complex diseases associ- had an almost 5-fold increased risk of receiving a future
ated with multiple genetic and environmental factors. asthma diagnosis.12 However, the risk factors for devel-
Several theories have been suggested in recent decades to oping asthma in AR patients remain unknown. The aim of
explain the increasing prevalence of asthma and allergic this study was to assess whether the use of antibiotics in
diseases in the industrialized world. In particular, the “hy- the previous 5 years is a risk factor for asthma development
giene hypothesis” has been proposed, which states that in a population of childhood AR. We also investigated the
growing up in a more hygienic environment with less mi- impact of different antibiotics on future asthma develop-
crobial exposure may promote the atopic T helper (Th) type ment in patients with AR.
2, whereas microbial pressure drives the immune system
toward balanced Th-1 and Th-2 immunity.1,2 Children of Methods
farmers are exposed to endotoxin from Gram-negative bac-
teria in early life, which may be the key element of less
hygienic environments, resulting in a lower prevalence of Data source
allergy and asthma.3 Exposure to different bacteria can
perturb the gut microbiota, which may alter the immuno- The present study was based on the National Health Insur-
logical balance and induce systemic inflammation.4 Although ance Research Database (NHIRD) of Taiwan, from January
immune development in young children is complex, anti- 1996 to December 2013, released from the Taiwan National
biotic consumption during infancy, when the commensal Health Research Institute for Research. The National Health
microbiota is developing, impairs the microbiota-mediated Insurance (NHI) program provides healthcare to 99% of the
mechanisms of immunological tolerance and predisposes 23.74 million Taiwan residents and is contracted with 97% of
children to inflammatory and allergic diseases.5,6 Early-life hospitals and clinics in Taiwan. The database provided one
infections may also be one of the major determinants of million random subjects for this study and consisted of pa-
asthma development. The progression from early-life in- tient records and details in prescription, including the
fections to asthma is the result of complex interactions be- encrypted patient identification numbers, medical facility
tween the specific type of infection and both genetic and registries, details of ambulatory care orders, inpatient or-
environmental factors.7 Antibiotics are frequently pre- ders, dental services, prescribed drugs, and physicians’
scribed for upper and lower respiratory tract infections in claims data. The diseases diagnosed were coded according
children.8 To date, epidemiologic studies have also sup- to the International Classification of Diseases, Ninth Revi-
ported an association between antibiotic use in early life sion, Clinical Modification (ICD-9-CM); drug prescription re-
and asthma development.9,10 Therefore, the understanding cords were coded under the Anatomical-Therapeutic-
of the relationship between antibiotic use and asthma is Chemical (ATC) Classification System. This study was
critical to clinicians and healthcare policy makers. approved by the Institutional Review Board of Kaohsiung
There is strong evidence that links allergic rhinitis (AR) Medical University Hospital (KMUHIRB-E(I)-20170113).
and asthma, but it is unclear whether this relationship is
causal. Based on primary healthcare data in German chil- Study population
dren with asthma, a considerably high prevalence of
several comorbid diseases is found in German children with We selected AR (ICD-9-CM code 477.8, 477.9) patients who
asthma.11 The diagnosis of AR was identified as an were less than 18 years old between January 1, 2005 and

Please cite this article as: Lin Y-C et al., Antibiotic exposure and asthma development in children with allergic rhinitis, Journal of
Microbiology, Immunology and Infection, https://doi.org/10.1016/j.jmii.2019.02.003
+ MODEL
Effect of antibiotics on childhood asthma 3

December 31, 2010 from the NHIRD database. The case Statistical analysis
group was defined as AR patients with newly developed
asthma (ICD-9-CM code 49390-49392), and the control group Pearson’s chi-square test or Fisher’s exact test was per-
was defined as AR patients without any history of asthma. formed to investigate the differences in categorical data
Patients who were diagnosed with asthma had at least two between patients from the case and control groups. An
prescriptions of anti-asthmatic drugs at different times independent t-test was conducted to evaluate the differ-
within a 1-year period. Anti-asthmatic drugs included ences in continuous data between the two groups. Condi-
inhaled selective b2-agonists (ATC code R03AC), inhaled tional logistic regression analysis was applied to determine
corticosteroids (ATC code R03BA), combined inhaled sal- the relationship between asthma and antibiotic exposure.
butamol/sodium cromoglycate (ATC code R03AK04), and The DDD recommended by the World Health Organization
combined inhaled selective b2-agonists/corticosteroids (WHO) was used to quantify antibiotic use. The cumulative
(ATC code R03AK06, R03AK07). The date of asthma diag- DDDs of antibiotics were categorized into low dose, mod-
nosis was defined as the index date in the present study. erate dose, and high dose in the study subjects. Various co-
Three non-asthmatic control children were matched to variables, including age, gender, resident urbanization,
each case child by age, gender, selected comorbid medical comorbidities, and co-medications, were adopted in the
disorders and infections, and the use of non-steroidal anti- statistical analysis model. In the analysis of antibiotic
inflammatory drugs (NSAIDs) by propensity score and was subtypes, we adjusted for the above variables but also for
randomly selected. To avoid any chronological overlap be- other different antibiotics. Odds ratios (ORs) and 95% con-
tween exposure and outcome, children who met our fidence intervals (CIs), using no antibiotic exposure as the
asthma definition, died, received their asthma diagnosis reference, were calculated to determine the risk of asthma
before an allergic rhinitis diagnosis, or were first diagnosed development in antibiotic users and in the dose-response
with asthma before age 2 years old were excluded. analysis. All statistical operations were conducted using the
SAS 9.3 statistical package. All p-values were two-sided,
and p-values < 0.05 were statistically significant.
Exposure to antibiotics

Drug usage information, including dosage, date of pre- Results


scription, usage days, and total number of pills, was
collected from the outpatient prescription database. The Demographic data of the study subjects
cumulative use of antibiotics was calculated as the cumu-
lative defined daily dose (DDD) to evaluate the dosage ef- The study entry flow chart is shown in Fig. 1. A total of 3236
fects. Antibiotic exposure was defined as at least one AR patients with newly developed asthma (case group) and
prescription of antibiotic (ATC code J01) within 5 years 9708 patients without an asthma diagnosis (control group)
preceding the date of asthma diagnosis for each case and were included in this study after 1:3 individual matching.
any antibiotic prescriptions during the same period for each Patient demographic characteristics between cases and
case-matched control. Three common antibiotics was controls at baseline are presented in Table 1. The mean age
analyzed in this study, including penicillins (ATC code of patients in the case group was 7.53  3.32 years, and
J01C), cephalosporins (ATC code J01D), and macrolides 44.1% were female. The distribution of age, gender,
(ATC code J01F). Other antibiotic classes were separately comorbidities, and co-medications were similar between
analyzed, such as quinolones, sulphonamides, glycopep- the cases and controls (Table 1).
tides, and lincosamides. Cases were then classified into
dosage categories based on the quartiles and deciles of the Relationship between early antibiotic exposure and
cumulative dosage distribution. asthma development in AR patients

Potential confounders In the present study, antibiotic exposure before 3 years old
was not associated with the development of asthma in AR
patients (adjusted OR 0.98, 95% CI 0.85e1.12; Table 2).
Potential confounders known to be associated with asthma
After age stratification of asthma diagnosis date, antibiotic
were investigated, including resident urbanization, atopic
exposure was no significant difference compared to the
dermatitis (ICD-9-CM code 691.8), chronic rhinitis (ICD-9-CM
controls in AR patients who were first diagnosed with
code 472.0), acute sinusitis (ICD-9-CM code 461.9), acute
asthma at 4e6 years old (adjusted OR 1.16, 95% CI
bronchitis (ICD-9-CM code 466), acute tonsillitis (ICD-9-CM
0.81e1.66), at 7e11 years old (adjusted OR 0.93, 95% CI
code 463), acute pharyngitis (ICD-9-CM code 462), pneu-
0.76e1.14), and more than 12 years (adjusted OR 0.93, 95%
monia (ICD-9-CM code 486), and the use of NSAIDs. NSAIDs
CI 0.66e1.30).
were often being used with antibiotics in the treatment of
infection in children. NSAIDs would induce clinical asthma
by the possibility of hypersensitivity and induction of severe Relationship between the cumulative DDDs of
airways obstruction.13 It has been reported that exposure antibiotics and asthma development in AR patients
of NSAIDs is associated with asthma risk.14,15 Therefore, the
use of NSAIDs was considered as an important covariate in There was a positive relationship between antibiotic
the present study. The use of NSAIDs was defined as NSAIDs exposure and the risk of asthma development (adjusted OR
use during 120 days before the index date. 1.74, 95% CI 1.41e2.17; Table 3). To further evaluate

Please cite this article as: Lin Y-C et al., Antibiotic exposure and asthma development in children with allergic rhinitis, Journal of
Microbiology, Immunology and Infection, https://doi.org/10.1016/j.jmii.2019.02.003
+ MODEL
4 Y.-C. Lin et al.

Figure 1. Flowchart of participant recruitment in this study. ICD-9-CM: International Classification of Diseases, Ninth Revision,
Clinical Modification NHIRD: National Health Insurance Research Database.

Table 1 Demographic characteristics between cases and controls in children with allergic rhinitis.
Case group Control group p value
(N Z 3236) (N Z 9708)
number (%) number (%)
Age, mean (standard deviation) 7.53 (3.32) 7.48 (3.30) 0.067
<4 years 511 (15.8) 1584 (16.3) 0.720
4e11 years 2407 (74.4) 7199 (74.2)
>Z12 years 318 (9.8) 925 (9.5)
Gender
Female 1426 (44.1) 4271 (44.0) 0.943
Male 1810 (55.9) 5437 (56.0)
Resident urbanization
Urban 889 (27.5) 2627 (27.1) 0.718
Suburban 1183 (36.6) 3626 (37.4)
Rural 1164 (36.0) 3455 (35.6)
Comorbidities
Atopic dermatitis 678 (21.0) 1926 (19.8) 0.172
Chronic rhinitis 763 (23.6) 2173 (22.4) 0.160
Acute sinusitis 2806 (86.7) 8475 (87.3) 0.387
Acute bronchitis 3143 (97.1) 9468 (97.5) 0.211
Acute tonsillitis 2470 (76.3) 7523 (77.5) 0.172
Acute pharyngitis 2718 (84.0) 8246 (84.9) 0.195
Pneumonia 179 (5.5) 473 (4.9) 0.138
Medication
NSAIDs 896 (27.7) 2661 (27.4) 0.759
NSAIDs: nonsteroidal anti-inflammatory drugs.

whether the relationship was dose-dependent, we divided 1.46e2.29 for overall antibiotics; adjusted OR 1.36, 95% CI
the study subjects into three subgroups based on the cu- 1.20e1.55 for penicillin; adjusted OR 1.65, 95% CI
mulative DDDs of antibiotics. The incidence of newly 1.47e1.84 for macrolides), and high dose (adjusted OR
developed asthma in the cases was significantly increased 2.28, 95% CI 1.82e2.86 for overall antibiotics; adjusted OR
in low dose (adjusted OR 1.40, 95% CI 1.12e1.75 for overall 1.65, 95% CI 1.45e1.88 for penicillin; adjusted OR 2.20, 95%
antibiotics; adjusted OR 1.60, 95% CI 1.43e1.79 for mac- CI 1.95e2.47 for macrolides) patients compared to the
rolides), moderate dose (adjusted OR 1.83, 95% CI controls. Otherwise, there was a significant trend toward

Please cite this article as: Lin Y-C et al., Antibiotic exposure and asthma development in children with allergic rhinitis, Journal of
Microbiology, Immunology and Infection, https://doi.org/10.1016/j.jmii.2019.02.003
+ MODEL
Effect of antibiotics on childhood asthma 5

Table 2 Antibiotic exposure before 3-year-old and the risk of asthma development in children with allergic rhinitis.
Age of index date Case group Control group Unadjusted model Adjusted modela
number (%) number (%) OR (95% CI) p value OR (95% CI) p value
(exposure/total) (exposure/total)
Overall 2803/3236 (86.6) 8467/9708 (87.2) 0.95 (0.84e1.07) 0.380 0.98 (0.85e1.12) 0.765
4e6 years 466/511 (91.2) 1435/1584 (90.6) 1.08 (0.76e1.53) 0.684 1.16 (0.81e1.66) 0.422
7e11 years 2258/2407 (93.8) 6792/7199 (94.3) 0.91 (0.75e1.10) 0.329 0.93 (0.76e1.14) 0.489
S12 years 79/318 (24.8) 240/925 (25.9) 0.94 (0.70e1.27) 0.698 0.93 (0.66e1.30) 0.652
a
Adjusted age, gender, resident urbanization, comorbidities and medication. OR: odds ratios; CI: confidence intervals.
p-values < 0.05 were statistically significant.

Table 3 Cumulative DDDs of antibiotics within 5-year preceding index date and the risk of asthma development in allergic
rhinitis children.
Group of antibiotics Case group Control group Unadjusted model Adjusted modela
Exposure number (%) Exposure number (%) OR (95% CI) p value OR (95% CI) p value
Any antibiotics
No 108 (3.3) 527 (5.4) 1.00 1.00
Yes 3128 (96.7) 9181 (94.6) 1.66 (1.35e2.05) <0.001 1.74 (1.41e2.17) <0.001
Cumulative DDDs
<8.08 870 (26.9) 3232 (33.3) 1.31 (1.05e1.64) 0.015 1.40 (1.12e1.75) 0.003
8.08e21.96 1046 (32.3) 3059 (31.5) 1.67 (1.34e2.08) <0.001 1.83 (1.46e2.29) <0.001
>21.96 1212 (37.5) 2890 (29.8) 2.05 (1.65e2.54) <0.001 2.28 (1.82e2.86) <0.001
p for trend <0.001 <0.001
Penicillins
No 682 (15.1) 935 (20.7) 1.00 1.00
Yes 3589 (84.9) 3589 (79.3) 1.38 (1.24e1.54) <0.001 1.31 (1.17e1.47) <0.001
Cumulative DDDs
<5.75 1143 (25.3) 1352 (29.9) 1.07 (0.94e1.21) 0.274 1.07 (0.94e1.22) 0.268
5.75e14.75 1260 (27.9) 1184 (26.2) 1.39 (1.23e1.57) <0.001 1.36 (1.20e1.55) <0.001
>14.75 1439 (31.8) 1053 (23.3) 1.74 (1.55e1.97) <0.001 1.65 (1.45e1.88) <0.001
p for trend <0.001 <0.001
Cephalosporins
No 618 (19.1) 1975 (20.3) 1.00 1.00
Yes 2618 (80.9) 7733 (79.7) 1.08 (0.97e1.19) 0.125 0.95 (0.85e1.06) 0.397
Cumulative DDDs
<2.55 581 (18.0) 1860 (19.2) 0.99 (0.87e1.13) 0.979 0.94 (0.82e1.08) 0.414
2.55e6.68 821 (25.4) 2599 (26.8) 1.01 (0.89e1.13) 0.877 0.91 (0.81e1.04) 0.182
>6.68 1216 (37.6) 3274 (33.7) 1.18 (1.06e1.32) 0.003 0.99 (0.88e1.12) 0.936
p for trend 0.001 0.965
Macrolides
No 1864 (41.2) 2617 (57.8) 1.00 1.00
Yes 2660 (58.8) 1907 (42.2) 1.76 (1.63e1.91) <0.001 1.77 (1.63e1.93) <0.001
Cumulative DDDs
<1.50 902 (19.9) 886 (19.6) 1.57 (1.40e1.75) <0.001 1.60 (1.43e1.79) <0.001
1.50e4.29 948 (21.0) 644 (14.2) 1.64 (1.47e1.83) <0.001 1.65 (1.47e1.84) <0.001
>4.29 810 (17.9) 377 (8.3) 2.20 (1.96e2.47) <0.001 2.20 (1.95e2.47) <0.001
p for trend <0.001 <0.001
Others
No 2527 (78.1) 7681 (79.1) 1.00 1.00
Yes 709 (21.9) 2027 (20.9) 1.06 (0.96e1.17) 0.214 0.97 (0.88e1.07) 0.584
Cumulative DDDs
<2.00 224 (6.9) 721 (7.4) 0.94 (0.80e1.10) 0.473 0.88 (0.75e1.03) 0.122
2.00e6.00 244 (7.5) 631 (6.5) 1.17 (1.00e1.37) 0.040 1.07 (0.91e1.25) 0.375
>6.00 241 (7.4) 675 (7.0) 1.08 (0.93e1.26) 0.297 0.97 (0.83e1.14) 0.767
p for trend 0.093 0.920
a
Adjusted age, gender, resident urbanization, comorbidities, medication and other subtype antibiotics. DDDs: defined daily doses;
OR: odds ratios; CI: confidence intervals. p-values < 0.05 were statistically significant.

Please cite this article as: Lin Y-C et al., Antibiotic exposure and asthma development in children with allergic rhinitis, Journal of
Microbiology, Immunology and Infection, https://doi.org/10.1016/j.jmii.2019.02.003
+ MODEL
6 Y.-C. Lin et al.

an increasing risk of asthma development with increasing 1.51e2.64), and erythromycin (adjusted OR 1.35, 95% CI
cumulative DDDs of any antibiotic analyzed (p for 1.23e1.47) was associated with asthma development in the
trend < 0.001). Furthermore, a dose-response relationship group of asthma patients first diagnosed before 12 years
between the occurrence of newly developed asthma and old. Only the preceding 5-year exposure of azithromycin
deciles of cumulative DDDs of antibiotics, including overall was associated with asthma patients diagnosed after 12
antibiotics, penicillins, and macrolides, was found (Fig. 2). years old (adjusted OR 3.46, 95% CI 1.37e8.70) (Table 4).

Relationship between antibiotic subtypes and Discussion


asthma development in AR patients
Atopic diseases have an inherited tendency to produce
Among 3236 AR children with newly developed asthma, immunoglobulin E (IgE) antibodies in response to small
3128 (96.7%) received an antibiotic prescription within 5 amounts of common environmental proteins, such as pol-
years preceding the index date, 2713 (83.8%) received at len, house dust mites, and food allergens. The presence of
least one penicillin prescription, 2618 (80.9%) received at atopy in an individual is associated with an increased risk of
least one cephalosporin prescription, 1866 (57.7%) received developing one or more atopic diseases e atopic derma-
at least one macrolide prescription, and 709 (21.9%) titis, allergic rhinoconjunctivitis, and asthma. Asthma is
received at least one other type of antibiotic prescription one of the most common chronic diseases of childhood and
(Table 4). After adjusting for age, gender, comorbidities, has an extensive impact on child health and develop-
medication, and the use of other antibiotics, the strengths ment.16,17 AR may contribute to asthma severity and in-
of the relationships between antibiotic subtypes and crease the risks of hospitalization and emergency
asthma development differed (adjusted OR 1.31, 95% CI department visits in children with asthma.18,19 Treatment
1.17e1.47 for penicillins; adjusted OR 1.32, 95% CI of comorbid AR reduces the odds of asthma-related
1.19e1.47 for amoxicillin; adjusted OR 1.77, 95% CI healthcare by up to 80%.20 However, why some patients
1.63e1.93 for macrolides; adjusted OR 2.97, 95% CI present with only one atopic disease (e.g., AR) in their
2.61e3.38 for azithromycin; adjusted OR 1.99, 95% CI lifetime, whereas other AR patients may develop asthma
1.53e2.59 for clarithromycin; and adjusted OR 1.32, 95% CI later in life, remains unknown.
1.22e1.44 for erythromycin) (Table 4). Antibiotic exposure in early childhood may contribute to
After age stratification of asthma diagnosis date, we the increasing asthma prevalence in industrialized coun-
found that the preceding 5-year exposure to overall anti- tries. Although a number of published studies have tested
biotics (adjusted OR 1.94, 95% CI 1.51e2.49), penicillin this hypothesis, the results have been conflicting. Marra
(adjusted OR 1.36, 95% CI 1.20e1.54), amoxicillin (adjusted et al. found that the use of antibiotics in the first year of
OR 1.37, 95% CI 1.22e1.54), macrolides (adjusted OR 1.83, life is associated with a small risk of developing asthma,
95% CI 1.68e2.00), azithromycin (adjusted OR 2.98, 95% CI and this risk increases with the number of courses of anti-
2.61e3.39), clarithromycin (adjusted OR 2.00, 95% CI biotic prescribed.21 When the number of courses of

Figure 2. Cumulative DDDs deciles (rank 1, 2, 3.10) within 5 years preceding index date and the risk of newly developed
asthma in children with allergic rhinitis. Data points were adjusted odds ratio (OR) with 95% confidence intervals (bars). No
antibiotic use as reference.

Please cite this article as: Lin Y-C et al., Antibiotic exposure and asthma development in children with allergic rhinitis, Journal of
Microbiology, Immunology and Infection, https://doi.org/10.1016/j.jmii.2019.02.003
+ MODEL
Effect of antibiotics on childhood asthma 7

Table 4 Classifications of antibiotic exposure within 5-year preceding index date and the risk of asthma development in
children with allergic rhinitis.
Age of index date/antibiotic type Case group Control group Unadjusted model Adjusted modela
Exposure number (%) OR (95% CI) p value OR (95% CI) p value
Overall N Z 3236 N Z 9708
Any antibiotics 3128 (96.7) 9181 (94.6) 1.66 (1.34e2.05) <0.001 1.74 (1.40e2.16) <0.001
Penicillins 2713 (83.8) 7659 (78.9) 1.38 (1.24e1.54) <0.001 1.31 (1.17e1.47) <0.001
Amoxicillin 2657 (82.1) 7449 (76.7) 1.39 (1.25e1.54) <0.001 1.32 (1.19e1.47) <0.001
Ampicillin 200 (6.2) 563 (5.8) 1.07 (0.90e1.26) 0.425 1.00 (0.84e1.18) 0.959
Penicillin 35 (1.1) 134 (1.4) 0.78 (0.53e1.13) 0.196 0.70 (0.48e1.03) 0.074
Cephalosporins 2618 (80.9) 7733 (79.7) 1.08 (0.97e1.19) 0.125 0.95 (0.85e1.06) 0.397
Macrolides 1866 (57.7) 4226 (43.5) 1.76 (1.63e1.91) <0.001 1.77 (1.63e1.93) <0.001
Azithromycin 506 (15.6) 576 (5.9) 2.93 (2.58e3.33) <0.001 2.97 (2.61e3.38) <0.001
Clarithromycin 97 (3.0) 147 (1.5) 2.01 (1.55e2.60) <0.001 1.99 (1.53e2.59) <0.001
Erythromycin 1397 (43.2) 3488 (35.9) 1.35 (1.24e1.46) <0.001 1.32 (1.22e1.44) <0.001
Others 709 (21.9) 2027 (20.9) 1.06 (0.96e1.17) 0.214 0.97 (0.88e1.07) 0.584
4e11 years old N Z 2918 N Z 8783
Any antibiotics 2839 (97.3) 8358 (95.2) 1.82 (1.43e2.33) <0.001 1.94 (1.51e2.49) <0.001
Penicillins 2504 (85.8) 7085 (80.7) 1.45 (1.29e1.62) <0.001 1.36 (1.20e1.54) <0.001
Amoxicillin 2455 (84.1) 6892 (78.5) 1.45 (1.30e1.62) <0.001 1.37 (1.22e1.54) <0.001
Ampicillin 177 (6.1) 510 (5.8) 1.04 (0.87e1.25) 0.606 0.98 (0.82e1.17) 0.832
Penicillin 35 (1.2) 123 (1.4) 0.85 (0.58e1.24) 0.416 0.76 (0.52e1.11) 0.163
Cephalosporins 2407 (82.5) 7110 (81.0) 1.10 (0.99e1.23) 0.065 0.96 (0.86e1.08) 0.581
Macrolides 1761 (60.3) 3979 (45.3) 1.83 (1.68e2.00) <0.001 1.83 (1.68e2.00) <0.001
Azithromycin 496 (17.0) 567 (6.5) 2.96 (2.60e3.37) <0.001 2.98 (2.61e3.39) <0.001
Clarithromycin 86 (2.9) 130 (1.5) 2.02 (1.53e2.66) <0.001 2.00 (1.51e2.64) <0.001
Erythromycin 1321 (45.3) 3283 (37.4) 1.38 (1.27e1.50) <0.001 1.35 (1.23e1.47) <0.001
Others 610 (20.9) 1741 (19.8) 1.06 (0.96e1.18) 0.206 0.96 (0.86e1.07) 0.533
12 years old N Z 318 N Z 925
Any antibiotics 289 (90.9) 823 (89.0) 1.23 (0.80e1.90) 0.340 1.20 (0.77e1.88) 0.401
Penicillins 209 (65.7) 574 (62.1) 1.17 (0.89e1.53) 0.243 1.15 (0.86e1.52) 0.326
Amoxicillin 202 (63.5) 557 (60.2) 1.15 (0.88e1.49) 0.297 1.13 (0.85e1.49) 0.383
Ampicillin 23 (7.2) 53 (5.7) 1.28 (0.77e2.13) 0.336 1.21 (0.72e2.03) 0.452
Penicillin 0 (0.0) 11 (1.2) NA NA
Cephalosporins 211 (66.4) 623 (67.4) 0.95 (0.73e1.25) 0.744 0.88 (0.66e1.17) 0.394
Macrolides 105 (33.0) 247 (26.7) 1.35 (1.02e1.78) 0.031 1.30 (0.98e1.73) 0.066
Azithromycin 10 (3.1) 9 (1.0) 3.30 (1.33e8.20) 0.010 3.46 (1.37e8.70) 0.008
Clarithromycin 11 (3.5) 17 (1.8) 1.91 (0.88e4.13) 0.098 1.78 (0.81e3.92) 0.149
Erythromycin 76 (23.9) 205 (22.2) 1.10 (0.81e1.49) 0.523 1.07 (0.78e1.46) 0.654
Others 99 (31.1) 286 (30.9) 1.01 (0.76e1.33) 0.943 0.97 (0.73e1.29) 0.841
a
Adjusted age, gender, resident urbanization, comorbidities, medication and other subtype antibiotics. OR: odds ratios; CI: confi-
dence intervals. NA: not applicable. p-values < 0.05 were statistically significant.

antibiotic increases, the risk of asthma also increases, with these changes. Mounting evidence shows that antibiotics
the highest risk being in children who receive more than influence the function of the immune system and the ability
four courses. Most antibiotics are associated with an to resist infection. The gut microbiota is important in
increased risk of asthma development, except for sulfon- human health, and dysbiosis of the intestinal ecosystem
amides.21 We found that antibiotic exposure before 3-year- contributes to the development of certain illnesses that can
old was not associated with the asthma development in AR be reversed by altering the microbiota via probiotic sup-
patient. However, our data suggested that antibiotic plementation. Commensal bacteria killed by antibiotics
exposure in the previous 5 years is associated with newly influence the epithelial barrier in many ways. Structural
diagnosed asthma in pediatric AR patients. The widespread components and metabolites of the intestinal microbiota
use of antibiotics in the past 80 years has saved millions of act on intestinal epithelial cells and local innate leukocytes
human lives and killed incalculable numbers of microbes, to maintain barrier integrity and regulate immune homeo-
both pathogenic and commensal. Human-associated mi- stasis.22 In the present study, penicillin groups and macro-
crobes perform several important functions and we are only lides were associated with the risk of asthma development
beginning to understand the ways in which antibiotics have in AR chideren. However, in penicillin groups, there was
reshaped their ecology and the functional consequences of only amoxicillin but not ampicillin nor penicillin

Please cite this article as: Lin Y-C et al., Antibiotic exposure and asthma development in children with allergic rhinitis, Journal of
Microbiology, Immunology and Infection, https://doi.org/10.1016/j.jmii.2019.02.003
+ MODEL
8 Y.-C. Lin et al.

significantly increased the risk of asthma development. It Conflicts of interest


has been reported that amoxicillin supplementation
effectively resulted in the reduction of ceco-colonic The authors have no conflicts of interest to disclose.
microbiota in pups.23 Korpela et al. found that penicillin
use was associated with a weaker shift in the microbiota
composition than macrolide use.24 Long-term probiotics Funding statement
intake on preschool children’s antibiotic use appeared to
protect some of the changes in the microbiota associated This study was supported by the Featured Areas Research
with penicillin use, but not those associated with macrolide Center Program within the framework of the higher edu-
use. The probiotic may be effective in protecting the cation Sprout Project by the Ministry of Education (MOE) in
microbiota against penicillin-associated disturbance and Taiwan, Ministry of Health and Welfare in Taiwan
increase in potential pathogens.24 The children using mac- (MOHW107-TDU-B-212-123006), and Kaohsiung Municipal
rolides had increased levels of Bacteroides spp., and an Hsiao-Kang Hospital Research Foundation (KMHKH-104-003,
elevated abundance of Bacteroides spp. has been KMHK-107-035, and KMHK-A-107-003).
demonstrated to be associated with increased serum IgE
levels in chideren.24,25 Financial disclosure
In the present study, we found a positive association
between asthma development and preceding 5-year anti- The authors have no financial relationships relevant to this
biotic exposure in AR children and that this association is article to disclose.
dose dependent. Moreover, there was a dose-response
relationship between the deciles of cumulative DDDs of
antibiotics and the occurrence of newly developed References
asthma. These results suggest that AR patients with higher
accumulated exposure to antibiotics have an increased 1. Liu AH, Murphy JR. Hygiene hypothesis: fact or fiction? J Al-
risk of asthma development. In addition, the preceding 5- lergy Clin Immunol 2003;111:471e8.
year exposure to different antibiotic subtypes, including 2. Gupta RS, Singh AM, Walkner M, Caruso D, Bryce PJ, Wang X,
et al. Hygiene factors associated with childhood food allergy
amoxicillin, azithromycin, clarithromycin, and erythro-
and asthma. Allergy Asthma Proc 2016;37:140e6.
mycin, increased the risk of asthma development in the 3. Braun-Fahrländer C, Riedler J, Herz U, Eder W, Waser M,
group of asthma patients who were diagnosed before 12 Grize L, et al. Environmental exposure to endotoxin and its
years old, but there was no statistical significance in the relation to asthma in school-age children. N Engl J Med 2002;
group of asthma patients diagnosed after 12 years old. 347:869e77.
These data suggest that younger AR patients are more 4. Sommer F, Backhed F. The gut microbiota–masters of host
susceptible to antibiotic exposure. We also analyzed the development and physiology. Nat Rev Microbiol 2013;11:227e38.
effects of three common antibiotic groups (penicillins, 5. Round JL, Mazmanian SK. The gut microbiota shapes intestinal
cephalosporins, and macrolides) on asthma development immune responses during health and disease. Nat Rev Immunol
in late childhood. Only two antibiotic groups (penicillins 2009;9:313e23.
6. Arrieta MC, Stiemsma LT, Dimitriu PA, Thorson L, Russell S,
and macrolides), but not cephalosporins or other antibi-
Yurist-Doutsch S, et al. Early infancy microbial and metabolic
otics, had a significant trend in increasing asthma risk. alterations affect risk of childhood asthma. Sci Transl Med
Furthermore, the association between antibiotic use and 2015;7:307ra152.
the risk of asthma development was highest in the group 7. Beigelman A, Bacharier LB. Early-life respiratory infections and
exposed to macrolides when the cumulative DDDs of asthma development: role in disease pathogenesis and poten-
macrolides were more than 4.29. tial targets for disease prevention. Curr Opin Allergy Clin
AR is one comorbidity of asthma that may contribute to Immunol 2016;16:172e8.
asthma severity and increase the risk of hospitalization. 8. McCaig LF, Besser RE, Hughes JM. Trends in antimicrobial
The allergic march refers to the “natural” history of atopic prescribing rates for children and adolescents. JAMA 2002;287:
disorders and concerns the development of atopic derma- 3096e102.
9. Kuo CH, Kuo HF, Huang CH, Yang SN, Lee MS, Hung CH. Early
titis in early childhood, progressing to other allergic dis-
life exposure to antibiotics and the risk of childhood allergic
eases in later childhood or adult life.26 Although data on diseases: an update from the perspective of the hygiene hy-
socioeconomic status, tobacco smoke exposure, the family pothesis. J Microbiol Immunol Infect 2013;46:320e9.
history of allergic skin diseases and other lifestyle-related 10. Pitter G, Ludvigsson JF, Romor P, Zanier L, Zanotti R,
risk factors were unavailable, we found that antibiotic Simonato L, et al. Antibiotic exposure in the first year of life
exposure in the previous 5 years, especially macrolide and later treated asthma, a population based birth cohort
accumulation, may be the “natural” history for the allergic study of 143,000 children. Eur J Epidemiol 2016;31:85e94.
march and a key factor for asthma development in AR pa- 11. Jacob L, Keil T, Kostev K. Comorbid disorders associated with
tients, especially younger children. asthma in children in Germany - National analysis of pediatric
In conclusion, the present study provides epidemiolog- primary care data. Pediatr Allergy Immunol 2016;27:861e6.
12. van den Nieuwenhof L, Schermer T, Bosch Y, Bousquet J,
ical evidence that exposure to antibiotics in the previous 5
Heijdra Y, Bor H, et al. Is physician-diagnosed allergic rhinitis a
years, especially a high accumulation of macrolides, in- risk factor for the development of asthma? Allergy 2010;65:
creases the risk of asthma development in pediatric AR 1049e55.
patients. The use of antibiotics should be restricted not 13. Joubert JR, Shephard E, Mouton W, Van Zyl L, Viljoen I. Non-
only early in life but also later in life in pediatric AR pa- steroid anti-inflammatory drugs in asthma: dangerous or useful
tients to prevent asthma development. therapy? Allergy 1985;40:202e7.

Please cite this article as: Lin Y-C et al., Antibiotic exposure and asthma development in children with allergic rhinitis, Journal of
Microbiology, Immunology and Infection, https://doi.org/10.1016/j.jmii.2019.02.003
+ MODEL
Effect of antibiotics on childhood asthma 9

14. Valkhoff VE, Schade R, t Jong GW, Romio S, Schuemie MJ, 21. Marra F, Marra CA, Richardson K, Lynd LD, Kozyrskyj A,
Arfe A, et al. Population-based analysis of non-steroidal anti- Patrick DM, et al. Antibiotic use in children is associated with
inflammatory drug use among children in four European increased risk of asthma. Pediatrics 2009;123:1003e10.
countries in the SOS project: what size of data platforms and 22. Goto Y, Kiyono H. Epithelial barrier an interface for the cross-
which study designs do we need to assess safety issues? BMC communication between gut flora and immune system.
Pediatr 2013;13:192. Immunol Rev 2012;245:147e63.
15. Sordillo JE, Scirica CV, Rifas-Shiman SL, Gillman MW, 23. Morel FB, Oozeer R, Piloquet H, Moyon T, Pagniez A, Knol J,
Bunyavanich S, Camargo CA, et al. Prenatal and infant exposure et al. Preweaning modulation of intestinal microbiota by oli-
to acetaminophen and ibuprofen and the risk for wheeze and gosaccharides or amoxicillin can contribute to programming of
asthma in children. J Allergy Clin Immunol 2015;135:441e8. adult microbiota in rats. Nutrition 2015;31:515e22.
16. Milligan KL, Matsui E, Sharma H. Asthma in urban children: 24. Korpela K, Salonen A, Virta LJ, Kumpu M, Kekkonen RA, de
epidemiology, environmental risk factors, and the public Vos WM. Lactobacillus rhamnosus GG intake modifies preschool
health domain. Curr Allergy Asthma Rep 2016;16:33. children’s intestinal microbiota, alleviates penicillin-
17. de Benedictis FM, Attanasi M. Asthma in childhood. Eur Respir associated changes, and reduces antibiotic use. PLoS One
Rev 2016;25:41e7. 2016;11:e0154012.
18. Thomas M, Kocevar VS, Zhang Q, Yin DD, Price D. Asthma- 25. Sepp E, Julge K, Mikelsaar M, Bjorksten B. Intestinal microbiota
related health care resource use among asthmatic children and immunoglobulin E responses in 5-year-old Estonian chil-
with and without concomitant allergic rhinitis. Pediatrics 2005; dren. Clin Exp Allergy 2005;35:1141e6.
115:129e34. 26. Reed CE. The natural history of asthma. J Allergy Clin Immunol
19. Sazonov Kocevar V, Thomas 3rd J, Jonsson L, Valovirta E, 2006;118:543e8. quiz 549e550.
Kristensen F, Yin DD, et al. Association between allergic rhinitis
and hospital resource use among asthmatic children in Norway.
Allergy 2005;60:338e42.
Appendix A. Supplementary data
20. Corren J, Manning BE, Thompson SF, Hennessy S, Strom BL.
Rhinitis therapy and the prevention of hospital care for Supplementary data to this article can be found online at
asthma. J Allergy Clin Immunol 2004;113:415e9. https://doi.org/10.1016/j.jmii.2019.02.003.

Please cite this article as: Lin Y-C et al., Antibiotic exposure and asthma development in children with allergic rhinitis, Journal of
Microbiology, Immunology and Infection, https://doi.org/10.1016/j.jmii.2019.02.003

S-ar putea să vă placă și