A prospective study evaluating the correlation between local weather conditions, pollen counts and pruritus of dogs with atopic dermatitis

Abstract

Background

Canine atopic dermatitis (cAD) is a hereditary, generally pruritic and predominantly T‐cell‐driven inflammatory skin disease, involving an interplay between skin barrier abnormalities, allergen sensitisation and microbial dysbiosis. The individual immunological response is predominantly against environmental allergens, including mite antigens; mould spores; and pollen from grasses, trees and weeds. Airborne pollens show fluctuating patterns during the year.

Objective

The aim of this prospective study was to evaluate the influence of local pollen concentrations and weather conditions on the clinical signs of atopic dogs, and to investigate any possible correlations with the results of intradermal testing (IDT).

Materials and Methods

Thirty‐seven privately owned atopic dogs in Bavaria were surveyed from 1 April to 30 November 2021. Owners were asked to record pruritus using a validated Visual Analog Scale (PVAS) score and the weekly medication of their dog. Furthermore, weather data, including pollen count, rainfall, relative humidity, hours of sunshine and temperature from the dog's location were collected daily.

Results

Of the evaluated parameters, only humidity and medication scores correlated positively with the PVAS scores of the atopic dogs. There was no correlation between specific pollen counts and PVAS scores of dogs with positive IDT reactions to that pollen.

Conclusion and Clinical Relevance

The outcome of this study highlights the importance of a careful interpretation of positive IDT results in dogs with cAD and questions the validity of airborne pollen trap methodology in representing pollen exposure for dogs at ground level.

Short abstract

Background – Canine atopic dermatitis (cAD) is a hereditary, generally pruritic and predominantly T‐cell‐driven inflammatory skin disease, involving an interplay between skin barrier abnormalities, allergen sensitisation and microbial dysbiosis. The individual immunological response is predominantly against environmental allergens, including mite antigens; mould spores; and pollen from grasses, trees and weeds. Airborne pollens show fluctuating patterns during the year. Objective – The aim of this prospective study was to evaluate the influence of local pollen concentrations and weather conditions on the clinical signs of atopic dogs, and to investigate any possible correlations with the results of intradermal testing (IDT). Conclusion and Clinical Relevance – The outcome of this study highlights the importance of a careful interpretation of positive IDT results in dogs with cAD and questions the validity of airborne pollen trap methodology in representing pollen exposure for dogs at ground level.

INTRODUCTION

Canine atopic dermatitis (cAD) is a clinical syndrome which as a result of predisposing genetic factors and environmental conditions manifests as an inflammatory and pruritic skin disease. ¹ The individual immunological response comprises cell‐mediated and humoral components, and in most cases is associated with immunoglobulin (Ig)E antibodies against environmental allergens, including dust and storage mites; mould spores; and pollens from grasses, trees and weeds. ² , ³ Canine atopic dermatitis may also be triggered by food allergens. ¹ Cutaneous hypersensitivity is influenced by the balance between a variety of T‐cell responses and inflammatory mediators, such as T‐helper 2 (Th2), Th1, Th17 and regulatory cytokines. ⁴ , ⁵ In the acute stages of cAD, a Th2 response seems to predominate; however, in the chronic stages, there is a mixed reaction with the addition of Th1 and Th17 cytokine activation. ⁵

To date, there is no reliable diagnostic test for cAD. The diagnosis of cAD is based on a detailed history, clinical examination and exclusion of other differential diagnoses, such as flea allergy dermatitis (FAD) or infestation with other ectoparasites, food allergy, bacterial infections and Malassezia overgrowth. ⁶ Once a clinical diagnosis of cAD has been made, two treatment options are available: symptomatic therapy with antipruritic drugs or allergen immunotherapy (AIT). Currently, AIT is the only available allergen‐specific treatment for cAD and is considered a safe and effective treatment option that can modify the individual's immune response. ⁷ , ⁸ , ⁹ , ¹⁰ Allergens to be included in the immunotherapy extract are identified by intradermal testing (IDT) or serum allergen‐specific IgE testing (SAT) and selected in the context of the patient's history. ¹¹ , ¹²

Airborne pollens show fluctuating patterns during the year. These fluctuations are the result of a number of environmental factors. In temperate climates, primary pollen production by individual plants occurs only during certain times of the year. Other important factors refer to the local weather conditions, ¹³ such as temperature and rainfall. ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ Further meteorological parameters such as relative humidity and sunshine hours also influence daily and diurnal variation of environmental pollen concentration. ¹⁶ To the best of the authors' knowledge, the influence of such parameters have not been evaluated in cAD.

The aim of this study was to evaluate the impact of local pollen concentrations and weather conditions on the clinical signs of atopic dogs and to investigate whether there was a correlation with the results of intradermal testing in these cases.

MATERIALS AND METHODS

Ethics

According to German law, this study is an observational study without direct interventions possibly causing harm to the participating dogs, and consequently did not need formal approval. We obtained informed consent from participating owners, and all data were anonymised.

Patient selection

For this prospective study, 37 dogs previously diagnosed with AD by clinical examination, history and ruling out differential diagnoses at the University of Munich's Centre for Clinical Veterinary Medicine were followed from 1 April to 30 November 2021. Nonskin‐related comorbidities were permitted. All participating dogs were on routine ectoparasite control. The involvement of food allergens was ruled out by the lack of a response to an elimination diet with a commercial hydrolysed or home‐cooked novel protein diet over a period of at least 8 weeks. Dogs that were positive on a food challenge were excluded from the study.

Assessment of clinical signs and weather data

For the evaluation of the impact of local pollen concentrations and weather conditions on the clinical signs of atopic dogs, owners of the enrolled atopic dogs were surveyed by phone every 7 days to determine the subjective mean pruritus score using a validated pruritus Visual Analog Scale (PVAS), ²¹ , ²² and to record the current topical and systemic treatment for each week. If dogs were on allergen‐specific immunotherapy they had to be treated for >1 year on a stable protocol that was not changed during the study. These data were used to calculate a weekly medication score as reported previously (Table 1). ²³ , ²⁴ , ²⁵

TABLE 1.

Medication score.

Treatment	Points
No medication	0
Topical medication (except tacrolimus and hydrocortisone aceponate)	5
Antihistamines and essential fatty acids	10
Oral antibiotics	30
Glucocorticoids >1 mg/kg daily	40
Glucocorticoids 0.5–1 mg/kg daily	30
Glucocorticoids 0.2–0.5 mg/kg daily	20
Glucocorticoids or topical hydrocortisone aceponate <0.2 mg/kg daily	10
Ciclosporin >5 mg/kg daily	40
Ciclosporin 2.5–5 mg/kg daily	30
Ciclosporin 1.25–2.5 mg/kg daily	20
Topical tacrolimus or ciclosporin <1.25 mg/kg daily	10
Lokivetmab injection	40
Oclacitinib 0.4–0.6 mg/kg twice daily	40
Oclacitinib 0.4–0.6 mg/kg daily	30
Oclacitinib <0.4–0.6 mg/kg daily	20

Weather data were collected daily from the weather station nearest to the individual dog's location and a mean weekly value for each of the parameters was calculated, including pollen count, rainfall, relative humidity, hours of sunshine and temperature. All weather data were calculated for each week, which was collected online from the German Meteorological Service, except the pollen count of individual plants which were provided by the Bavarian government (Bayrisches Landesamt für Gesundheit und Lebensmittelsicherheit).

Intradermal testing

Before the study, IDT was performed during a flare of clinical signs for each individual dog. Intradermal testing was performed and interpreted as described previously. ²⁶ For this study, results ≥2 were included as positive for each specific allergen.

Injectable glucocorticoids were withdrawn for ≥6 weeks and oral glucocorticoids and ciclosporin for ≥4 weeks before IDT to minimise drug influence on test results. Topical glucocorticoids, oclacitinib and oral antihistamines were withdrawn 2 weeks before intradermal testing. Lokivetmab was not withdrawn.

Statistical methods

The D'Agostino test of skewness was used to check the normality of distribution of PVAS scores. In addition, a Shapiro–Wilk Normality Test was used to check the normality of residual distribution in the multivariable ordinary least squares model. Owing to the skewed distribution of PVAS scores (skew = 0.49271, p‐value <0.001) and non‐normally distributed residuals (p < 0.001), the relationship between PVAS scores and predictors (humidity, hours of sunshine, pollen counts and rainfall) was studied via multivariable median‐based regression. False discovery rate correction of p‐values for multiple testing was applied, and only FDR‐corrected p‐values were considered. The PVAS scores of dogs with positive IDT reactions to a specific pollen allergen were correlated with the pollen concentration of that specific pollen.

RESULTS

Initially, 37 dogs were included in the study. Of these, four dogs were subsequently excluded from the study owing to low compliance, and three dogs were euthanised during the study owing to unrelated conditions. One of these dogs was euthanised during the last 4 weeks of the study, and data collected until this time were included, leading to a total number of 32 dogs.

In total, 17 female dogs (12 spayed) and 15 male dogs (eight neutered) with the diagnosis of cAD were included in the study. Their age ranged from 3 to 15 years, with a mean of 5 years. Breeds included were French bulldog (n = 5), Labrador retriever (n = 4), Golden retriever, German shepherd dog (n = 2 each), continental bulldog, Boxer, Rottweiler, shi tzu, Boston terrier, Berger Blanc Suisse, pug, Bracco Italiano, English setter, Griffon Fauve de Bretagne and miniature poodle (n = 1 each), as well as mixed‐breed dogs (n = 8). Weather data as well as pruritus and medication scores are listed in Table S1 in Supporting information. Results of the intradermal tests are given in Table S2.

Airborne grass, weed and tree pollen concentrations had no correlation with PVAS scores of atopic dogs sensitised to each specific pollen, as presented in Table 2. Among the collected weather parameters, only humidity had a significant positive correlation with pruritus (β = 0.07 [95% CI = 0.03–0.10], p < 0.001). Medication scores also correlated significantly with PVAS scores (β = 0.03 [0.01–0.05], p = 0.001).

TABLE 2.

Results of the multivariable median‐based regression.

Predictor	Beta	95% CI	p‐value	q‐value a
Relative humidity	0.07	0.03, 0.10	<0.001	0.006
Medication	0.03	0.01, 0.05	0.001	0.007
Ambrosia elatior	−2.8	−5.0, −0.51	0.016	0.070
Fagus	0.24	0.03, 0.45	0.027	0.088
Temperature	0.05	0.00, 0.11	0.073	0.2
Rumex acetosa	−0.53	−1.2, 0.13	0.12	0.3
Hours of sunshine	0.09	−0.03, 0.21	0.2	0.3
Artemisia vulgaris	0.20	−0.32, 0.72	0.5	0.7
Rainfall	0.02	−0.03, 0.06	0.5	0.7
Grass mix	0.02	−0.03, 0.06	0.5	0.7
Corylus	0.05	−0.42, 0.52	0.8	>0.9
Chenopodium album	−0.55	−6.4, 5.3	0.9	>0.9
Populus	0.00	−0.01, 0.01	>0.9	>0.9

Abbreviation: CI, confidence interval.

^a False discovery rate correction for multiple testing.

DISCUSSION

In this study, we analysed the relationship between mean weekly atmospheric pollen concentrations, environmental factors and their impact on pruritus of dogs with cAD presumably sensitised to these pollens, as indicated by positive IDT results to specific pollen allergens. The number of positive IDT results in the present study was similar to those reported in other larger studies reporting IDT results. ²⁷ , ²⁸ Pruritus, considered a major sign of cAD, was present in all dogs. ²⁹ , ³⁰ Consequently, it was chosen as a reliable parameter to monitor the severity of cAD in individual dogs alongside the medication scores.

We found that of the environmental parameters evaluated, only relative humidity was correlated with the pruritus scores of affected dogs, this is in concordance with a previous report in humans sensitised to grass pollen. ³¹ Relative humidity can play a significant role in aeroallergen release. The thin and delicate pollen wall of Poaceae pollen grains can easily burst as a result of high atmospheric humidity and release allergenic proteins. ³² , ³³ Conflicting results have been found about whether outdoor humidity is positively or inversely associated with AD prevalence in humans and mice. Higher humidity may increase trans‐epidermal water loss in atopic skin. ³⁴ A murine model showed that low humidity can induce epidermal DNA synthesis and mast cell degranulation leading to epidermal hyperplasia in response to barrier disruption. ³⁵ By contrast, air temperature, rainfall and hours of sunshine had no impact on the pruritus of the dogs with cAD in our study, even though they are reported influential meteorological factors for atmospheric pollen concentration. Air temperature for example is important for anther opening, pollen release and atmospheric pollen dispersal. ³⁶ It has also been found that daily rainfall did not significantly influence the occurrence of pollen grains in the atmosphere. ³⁷ These factors may be less effective in increasing the concentration of airborne allergens and consequently may not trigger clinical signs as readily as atmospheric humidity in dogs. Climate change and global warming contribute to increasingly extended pollination seasons resulting in wide overlaps between seasons, which can lead to difficulties in identifying the specific pollen to be included in the allergen extract for immunotherapy. At the same time, climate change leads to increased air pollution with carbon dioxide, which alters plant growth and the timing of pollen release. Urban carbon dioxide concentrations are approximately 30% higher compared with rural areas, ³⁸ which may explain why dogs in urban areas react more strongly to pollen allergens in intradermal tests. ²⁶

Pruritus scores correlated with medication scores in our study, which was expected owing to the fact that all selected medications listed in the used scoring system have been reported as successful treatment options for pruritus in atopic dogs. ⁸ , ¹⁰ , ³⁹ Consequently, increased pruritus led to increased administration of antipruritic medication.

We were not able to demonstrate a correlation between mean weekly atmospheric pollen concentrations and pruritus of dogs with cAD, and furthermore, there was no direct correlation between elevations in specific pollen counts and individual sensitivities as indicated by positive IDT. This is in contrast to what has been reported previously in humans. ³¹ However, the correlation between atmospheric pollen concentration and manifestation of allergic disease in humans was not consistently evident, ³³ , ⁴⁰ , ⁴¹ and is often reported as lower than expected. ³² , ⁴² , ⁴³ , ⁴⁴ , ⁴⁵ Another possible explanation may be that the positive reactions to seasonal allergens in our study population may represent subclinical sensitisation or false positive IDT results, even though the time of testing of the dogs in our study was chosen during their allergic flares. Additionally, pollen concentrations are usually measured with pollen traps located on roofs, as occurred in our study, and therefore may not reflect the true pollen exposure of atopic dogs on the ground. Pollen concentrations on the feet of people walking in an urban area were four‐fold higher than on the shoulders, back and legs. ⁴⁶

Allergenic activity of pollens has also been found outside the pollination period which has led to the suggestion that the pollen trap count is not a reliable measure of the total allergen exposure. ⁴⁷ , ⁴⁸ , ⁴⁹ Small aero carriers loaded with allergens can remain airborne longer than intact pollen grains, that are detected by classical pollen traps. ⁵⁰ Cabrera et al. observed more occurrences of high Phl p1 concentrations (a Timothy grass pollen allergen) when pollen counts were low outside the grass pollen season, which could explain why even low levels of 2–9 pollen grains/m³ can result in clinical signs of atopic disease. ³¹ Additionally, high levels of Phl p5 were detected in indoor dust outside the grass pollen season; the authors hypothesised that pollen allergens were carried indoors via footwear. ⁵¹ This could contribute to the fact that indoor life is one of the main predisposing factors for the development of cAD. ³⁰ With severe clinical signs all year round resulting from, for example, a hypersensitivity against dust mite antigens, seasonal peaks may not be observed as easily. In addition, not every positive reaction on an allergen test has clinical relevance for the individual atopic dog. In a French study, there was no evidence for a link between sensitisation to grass pollen and seasonality of clinical signs of atopic dogs. ⁵² Only 16.5% of atopic dogs suffered from seasonal pruritus and of those between 0% and 20% reacted to grass pollen. By contrast, 208 dogs included in that study had no seasonal pruritus and between 12.2% and 31.1% of those were sensitised to grass pollen. ⁵² In another more recent study, a poor correlation was found between the IDT results of dogs and their clinical history. However, the assessed parameters, such as pruritus and seasonality, were identified retrospectively and a prospective study was recommended to confirm those findings. ²⁶

This study has a few limitations. We did not include the first months of the year, and therefore, we were not able to detect pollen counts of trees such as hazel, which are already detectable in the early months of the year. Another limitation is the recruitment of dogs living in Bavaria only. Different geographical locations exhibit different weather conditions, which influence atmospheric pollen concentration. We included pollen and weather data from weather stations situated as close as possible to the dogs' home to evaluate the most relevant data for each individual dog, yet the pollen counts in the dogs' backyards or walking areas may have differed to a degree from the pollen counts measured in the nearest pollen traps a few kilometres away. This may have been particularly important for tree pollens that are not carried as far by the wind as grass or weed pollens. Further studies are needed to evaluate the real pollen exposure of dogs.

CONCLUSION

This study showed a significant correlation between relative humidity and PVAS of dogs with cAD. However, we did not demonstrate a significant correlation between airborne pollen concentrations and PVAS of dogs with cAD sensitised to the specific pollen allergens as indicated by IDT. The outcome of our study highlights the importance of a careful interpretation of IDT results and questions the value of airborne pollen capture systems representing the real pollen exposure in dogs sensitised to those pollens.

AUTHOR CONTRIBUTIONS

L. Widorn: Writing – original draft; data curation; investigation; project administration; writing – review and editing. Y. Zabolotski: Formal analysis; software; visualization; validation. Ralf S. Mueller: Conceptualization; methodology; supervision; writing – review and editing.

FUNDING INFORMATION

Self‐funded.

CONFLICT OF INTEREST STATEMENT

No conflicts of interest have been declared.

Supporting information

Table S1.

Table S2.

Widorn L, Zabolotski Y, Mueller RS. A prospective study evaluating the correlation between local weather conditions, pollen counts and pruritus of dogs with atopic dermatitis. Vet Dermatol. 2024;35:500–507. 10.1111/vde.13268