A Survey of Seasonal and Perennial Allergic Conjunctivitis in Children in Southwest China: A Case-Control Study

Background Seasonal allergic conjunctivitis (SAC) and perennial allergic conjunctivitis (PAC) were the most usual types of allergic conjunctivitis. House dust mites were the most common sensitization agents for SAC and PAC. This study aimed to explore SAC and PAC risk factors in children. Methods We recruited 176 children suffering from SAC or PAC and 131 control subjects in southwest China. A questionnaire , several eye exams and the skin prick test (SPT) were performed.The scores of the symptoms/signs were recorded. Results The percentage of children that had ever been breastfed in the case group was lower than in the control group (P<0.05). The rate of parental allergy history in the case group was higher than for the control group (P<0.01). Compared with the control group, the case group was more likely to have other systemic allergic diseases (P<0.01).The incidence rate of adenoidal hypertrophy (ATH) in case group was higher than control group (P<0.05). The ocular symptoms and signs scores had signicant correlation to the course (P<0.05), but no correlation to the SPT results (P>0.05). Conclusion

always accompanied by symptoms of allergic rhinitis [8]. Dust mites are widely distributed, particularly in the southwest of China, because of the humid climate. Dust mite species include Dermatophagoides farinae and Dermatophagoides pteronyssinus [9]. They are the most common sensitising agents for SAC and PAC [10,11].
In this study, we administereda survey to the subjects of 307 subjects aged from 4 to 8 years in southwest China; 176 suffered from SAC or PAC, and 131 were normal control subjects. All were studied for more details to explore the risk factors of AC in children and to analyse the related factors affecting the severity of clinical symptoms and signs, with the aim of providing more help for clinical diagnosis and treatment in the future.

Materials And Methods
The study consisted of a disease-speci c questionnaire, an ophthalmologic examination and a SPT.

Patients
The study included 176 subjects suffering from SAC or PAC, con rmed by a positive history and positive skin tests, and 131 normal control subjects in the Children's Hospital of Chongqing Medical University (Chongqing, China) from July 2016 to July 2017. Informed consent was obtained from both parents. The diagnosis of ocular allergy was based on clinical history and signs and symptoms, with the support of a positive SPT. All the subjects had not been treated in the previous 4 weeks with topical or systemic H1 receptor antagonists, topical cromolyn or glucocorticosteroids.
The disease-speci c questionnaire and ophthalmologic examination Since the subjects were too young to respond adequately to a questionnaire, all the questionnaires were answered by the children's both parents or their guardians. The questionnaire included questions about the children's basic demographics, breastfeeding history, parental history of allergic disease, the children's experiences of systemic allergic disease, the season of the disease onset and the symptoms. Our ophthalmologists performed eye exams, including a primary ophthalmologic examination, slit lamp observation, corneal uorescent staining and tear break-up time (BUT). The corneal uorescein staining scores were modi ed according to the National Eye Institute grading scale (Table 1) [7].The symptom score and sign score measurements were adapted from Macy M. S. Wu et al. (Table 1) [6,12]. The SPTs were performed using commercial extracts of 13 common inhalation allergens:D. farina,D. pteronyssinus,cockroaches,saccharomycetes, penicillium, dog fur, cat hair,duck feather, birch pollen, artemisia pollen,maize pollen, cotton wooland cigarettes.The negative control solution was a phenolated glycerol-saline solution. The positive control solution was 10 mg/ml histamine hydrochloride (ALK-ABELLO Laboratories,Hrsholm, Denmark). All patients enrolled in the study had discontinued antiallergic drugs for at least 4 weeks prior to the test.The skin reaction was graded 20 minutes later. The skin oedema and erythema that developed were graded from zero to four degrees by comparing the size with positive and negative controls [13].Grade 1 is 25% of the area of histamine-induced wheal, Grade 2 is 50%, Grade 3 is 100% and Grade 4 is 200%. Grades 2, 3, and 4 were considered as positive skin reactions [14].

Statistics
The inter-group indicators of the case and control groups were described and compared. The quantitative data were described as mean ± standard deviation (SD) and compared by the t-test. The rank data were described by the median (quartile spacing), and the rank sum Wilcoxon or Kruskal-Wallis tests were used for inter-group comparisons. Qualitative data were described by frequency (percentage), and the Chi-Square test was used to compare the groups. All hypotheses were tested on both sides of the P value. A P value of 0.05 was set to be statistically signi cant. The con dence interval was 95%. SPSS software (version 21.0) was used for all statistical calculations (SPSS,Inc.,Chicago, IL, USA).

Clinical features
The mean age at rst examination was 5.99 ± 5.98 years. There was no signi cant difference in age between the case and control group. The AC morbidity was higher in male than in female children (P < 0.01) ( Table 2).

Breastfeeding history
The parents were asked if the children were exclusively breastfed and about its duration. The relationship between breastfeeding and AC outcomes was investigated; a small number of children in the case group were ever breastfed, and the percentage was far lower than in the control group (19.85% vs. 63.64%) (P < 0.01). Among those who were ever breastfed, the mean duration of exclusive breastfeeding in the case group was shorter than in the control group (P < 0.05) ( Table 3). We explored the correlation between duration of exclusive breastfeeding and the age of AC onset, but no positive result was found. We also explored the correlation between the duration of exclusive breastfeeding and the degree of allergic reaction to different house dust allergens found by the SPT, but no positive result was found. Parental history of allergic disease The parents were asked about their history of allergic diseases, such as allergic rhinitis, eczema and urticaria. The rate of parental allergy history in the case group was higher than in the control group (P < 0.01, both in the father and the mother) ( Table 4). The proportion with allergicrhinitis was the highest at 18.69% (n = 20) in the father and 24.18% (n = 22) in the mother. Eight children in the case group had parents who both had a history of allergic rhinitis.

Systemic allergic disease and AC
We studied the relationship between systemic allergic diseases and AC in children. In the case group, allergic rhinitis in 85.80% of the group was the most closely related to AC, followed by eczema (76.14%).
The third most common allergic disease was asthma (65.34%). Atopicdermatitis and urticaria papulosa were also associated with AC. Compared with the control group, these ndings all had signi cant differences (all P < 0.01). Surprisingly, we found some patients had ATH. Meantime, although the proportion was not high at 6.25%, there was a statistical difference between the case group and the control group (P < 0.05) ( Table 5).

Eye symptoms
The top three symptoms of SAC and PAC were eye rubbing, itching, and blinking. More than half of children in the case group suffered from the top three symptoms.The redness was the forth most common sign which present in 26.7% of the case group, it also could not be ignored as it was hard to distinguish from conjunctivitis.

Ocular signs
The top three clinical signs of SAC and PAC were chemosis, tarsal conjunctival papillary hypertrophy and bulbar conjunctival hyperaemia. However, they all occurred in no more than half of the case group children. Interestingly, discolouration was speci c and accounted for 21% (n = 37) of the cases, slightly lower than the top three signs. Limbal hypertrophy was also speci c, and the proportion was 13.6% (n = 24), which was lower than discoloration but it still always happened. The keratitis and mucus secretions were uncommon, with a proportion of 8.0% and 9.7%, respectively; as previously reported, corneal involvement rarely happened [5]Moreover, only 2.84% (n = 5) of the patients had scales and scurf of the eyelid skin, and 2.27% (n = 4) cases had meibomian gland obstruction.

Relationship between the ocular symptom/sign scores and the disease duration
The mean clinical course in the case group was 3.09 ± 2.92 months. We found signi cant positive correlations between the ocular symptom/sign scores and the disease duration (Table 6). Relationship between the ocular symptom/sign scores and the SPT results for two Dust mite allergens The SPT results in the control group were all negative. In the case group, the results of the skin reaction to D. pteronyssinus were in accord with the D. farinaresults; the goodness of t was up to 81.25% (n = 143). Grade 3 was the main SPT result for both the dust mite allergens (The results are displayed in supplementary Table.1).There is no correlation between the ocular symptom/sign scores and the grading of the SPT results for the response to these two dust mite allergens(The results are shown in supplementary Table 2).

Discussion
In this study, SAC mostly happened in the spring and the autumn,but seldom in the winter.
The SPT results showed that SAC and PAC patients more frequently and severely displayed allergic reactions to dust mite allergens and mites maybe the most common allergen in children with allergic conjunctivitis. Because dust mites like warm and wet weather, so dust mites bred more in the spring and autumn than in the winter.
No correlation was found between the ocular symptom/sign scores and the SPT grades in response to dust mite allergens. But we found signi cantly positive correlations between ocular symptom/sign scores and disease duration. The longer the disease lasted, the higher the patients' symptom or sign scores. Lasting moderate or severe allergic reaction to dust mite allergens may be responsible for the ocular symptoms and signs. Therefore, to avoid treatment di culty and the aggravation of symptoms and signs caused by the delay in diagnosis, timely antiallergic treatment is recommended.
The results showed that the number of children in the case group who had ever been exclusively breastfed was lower than in the control group. The mean duration of exclusive breastfeeding in the case group was shorter than in the control group. These results indicated an association between a history of breastfeeding with a lower rate of allergic eye diseases.We speculate that exclusive breastfeeding may play an important protective role in the AC.As reported by Kull et al. that breastfeeding for four months or more could reduce the risk of eczema and onset of the allergy [15]. In addition, we found that parental allergy history in the case group was signi cantly higher than in the control group. Allergic rhinitis was most common among their parents. Therefore, prolonged breastfeeding could be a particularly recommended way for infants to reduce the risk of onset of AC, especially with a parental allergy history.
Systemic allergic diseases were closely related with AC. According to our study, allergic rhinitis was the most common, followed by eczema, asthma and urticaria papulosa. Investigatingthe systemic allergic history was necessary in the clinic, especially for children without typical symptoms and signs or children were too young to express themselves. This would help with diagnosis and provide appropriate treatment.
Interestingly, children with ATH were more common in the case group, which suggested thatallergic conjunctivitis may be related to ATH, a result consistent with previous researches [16][17][18].The conjunctiva are located in the upper extremity of the respiratory system, and the nasolacrimal duct is a drainage system into the nose [19]. Allergens and allergic mediators drain to the nose by this pathway, generating nasal symptoms. The conjunctiva and the nose make up an entire system [1,2,[5][6][7][8]. This can also be explained, allergic rhinitis and conjunctivitis were always co-existent and persistent to repeatedly happen. SAC and PAC were considered to be associated with type I hypersensitivity reactions [20]. Xiaowen Zhang et al. found that the rate of IgE presenting in the adenoids or tonsils was signi cantly higher than in the serum of childhood ATH, which suggests a role for local atopy [21,22]. Allergy control may play a role in reducing the rate of adenotonsillectomy in children suffering from allergic reactions caused by ATH [23]. We assumed that in children suffering from PAC combined with ATH, effective PAC control could alleviate the ATH symptoms. Children with ATH would be suggested to have an ophthalmic exam to determine if they are suffering from AC, in order to give more suitable synchronous treatments.
The ocular surface in ammation was usually driven by mast cells, which led to rubbing eyes, itching, blinking and redness in the acute phase [20].The symptoms of SAC and PAC in children were typical, mainly including eye rubbing, itching, blinking and redness. Nearly half the children experienced these symptoms. The top three clinical signs of AC were chemosis, tarsal conjunctival papillary hypertrophy, and bulbar conjunctival hyperaemia, but they were not speci c. Discoloration, limbal hypertrophy, mucus secretions and keratitis were the characteristic signs. Discoloration and limbal hypertrophy always occurred and could make the eye circumference become thickened and opaque. Keratitis and mucus secretions were rare and often happened when eye rubbing was uncontrollable [5].As children are different from adults, they were not be able to express their feelings accurately and have a variety of clinical manifestations. We always have di culties in clinical diagnosis, so the more we learn about the characteristics of symptoms and signs of SAC and PAC in children, the more professional the decisions we will make concerning diagnosis and treatment.
This study also had some limitations. First, the sample of subjects in this study was small. In the further research, we need to recruit more to verify our positive ndings. Second, we used SPT instead of conjunctival provocation test which is an established diagnostic procedure for allergic conjunctivitis.
Because the conjunctival provocation test is not usually used in the clinic regarding the relative risks for children [24]. SPT has higher accuracy in the diagnosis compared with serum-speci c IgE in vitro [2,25]. Even so, conjunctival provocation test, serum-speci c IgE or SPT can only be considered as a diagnostic tool for evaluation the allergic status of individuals, but not to a diagnosis tool [26]. Third, since our subjects were too young to accomplish such a questionnaire adequately, all the questionnaires were answered by their parents. But even parents can have limited understanding of subjects' routines. Last but not the least, our results showed all subjects in the control group were negative to the SPT test considering dust mite allergen, allergic conjunctivitis patients more frequently displayed allergic reactions to it. It suggested that allergic background made subjects more sensitive to dust mite allergen rather than dust mite causing seasonal and perennial allergic conjunctivitis.

Conclusion
In this study, we explored the risk factors in children with SAC and PAC, and found such as history of breastfeeding, parental history of allergic disease and systemic allergic disease were closely related to the incidence of allergic conjunctivitis. Ophthalmologist need pay more attention to the children's systemic symptoms/signs and family histories, and SPT results to provide a rm diagnosis and timely treatments. for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.