Follicular Conjunctivitis in Dogs: A Retrospective Study (2007-2022)
Follicular Conjunctivitis in Dogs: A Retrospective Study (2007-2022)
Follicular Conjunctivitis in Dogs: A Retrospective Study (2007-2022)
DOI: 10.1111/vop.13155
ORIGINAL REPORT
Irene Cerrada1 | Marta Leiva1,2 | Rita Vilao1 | Teresa Peña1,2 | Jose Ríos3
1
Servei d'Oftalmologia, Fundació
Hospital Clínic Veterinari, Universitat Abstract
Autònoma de Barcelona, Bellaterra, Objective: To evaluate the incidence, clinical features, treatment, and outcome
Spain
2
of canine follicular conjunctivitis (CFC).
Departament de Medicina i Cirurgia
Animals, Facultat de Veterinària,
Procedure: Medical records of dogs diagnosed with CFC were reviewed. Data re-
Universitat Autònoma de Barcelona, corded included signalment, duration of clinical signs and treatment details prior
Bellaterra, Spain to presentation, concurrent ocular/systemic diseases, ocular clinical signs, cytol-
3
Laboratory of Biostatistics &
ogy, treatment, follow-up, and outcome. Blepharospasm, signs of self-trauma,
Epidemiology, Facultat de Medicina,
Universitat Autònoma de Barcelona, hyperemia, chemosis, ocular discharge, and follicle location and severity (0.5–4)
Bellaterra, Spain were retrospectively evaluated. Based on severity, treatment consisted of topical
0.1% diclofenac or 0.1% dexamethasone sodium eyedrops. Dogs were classified
Correspondence
Marta Leiva, Servei d'Oftalmologia, into young (YD < 18 months) and adult (AD ≥ 18 months).
Fundació Hospital Clínic Veterinari, Results: One hundred and fifty-three dogs (276 eyes) were included in the study:
Universitat Autònoma de Barcelona,
Bellaterra, Spain.
83YD (54%) and 70AD (46%). Males and bilateral disease were over-represented
Email: marta.leiva@uab.cat in both groups. Severity was associated with young age (p = .032) and bilateral-
ity (p = .025), and not with dermatological diseases (p > .05). No differences in
follicular location were observed except for more frequent involvement of the
nictitating membrane (MN) in YD (p = .02). Response to treatment was faster in
AD (p = .001), with complete resolution in 80.6% of the eyes (100/124) at 1 month.
YD treated with diclofenac showed faster resolution than those treated with 0.1%
dexamethasone (p = .009).
Conclusions: Although CFC is a bilateral ocular disease occurring at any age, the
clinical presentation is influenced by age. Follicular conjunctivitis in adult dogs
is less sever, less commonly affects the NM, and responds more quickly to topical
treatment. One month of topical diclofenac may be adequate for mild cases, and
1 month of topical 0.1% dexamethasone is recommended as initial therapy for
moderate to severely affected cases.
KEYWORDS
allergic conjunctivitis, atopy, follicles, papillae, self-limiting conjunctivitis, vernal
conjunctivitis
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any
medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
© 2023 The Authors. Veterinary Ophthalmology published by Wiley Periodicals LLC on behalf of American College of Veterinary Ophthalmologists.
14635224, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/vop.13155 by Turkey Cochrane Evidence Aid, Wiley Online Library on [20/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2 CERRADA et al.
14635224, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/vop.13155 by Turkey Cochrane Evidence Aid, Wiley Online Library on [20/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
CERRADA et al. 3
F I G U R E 1 Grading of severity according to the number of follicles, from left to right: very mild (grade 0.5), mild (grade 1), moderate
(grade 2), severe (grade 3) to very severe (grade 4).
was assigned to palpebral conjunctiva, bulbar conjunctiva Dogs were classified into YD (83 individuals; 54%)
and palpebral or bulbar aspects of the NM. and AD (70 individuals; 46%). In both groups, males (48;
Concomitant systemic diseases were classified as der- 57.8%YD vs. 38; 54.3%AD) and bilateral presentation (69;
matological and non-dermatological. Both groups of dis- 83.1%YD vs. 54; 77.1%AD) were over-represented. There
eases were diagnosed after a complete examination by a was no statistically significant over-representation of
board-certified dermatologist or internist according to the brachycephalic breeds in either group (p = .557).
diagnostic protocols established for each disease. Previous ocular history was less commonly reported in
The selected dogs were divided into two age YD (17 eyes; 11.18% eyes) compared to AD (28 eyes; 22.58%
groups, young dogs (YD < 18 months) and adult dogs eyes) (p = .014). Comorbidities were detected in 47 ani-
(AD ≥ 18 months). mals (84 eyes), of which 17 were classified as non-derma-
tological and 30 as dermatological. Of the 84 affected eyes,
39/152 (25.65%) were YD and 45/124 (36.30%) were AD.
2.3 | Statistical analysis Non-dermatological conditions consisted of food allergy
(4 dogs, 5 eyes/276; 1.80%), gastritis (1 dog, 2 eyes/276;
Quantitative variables were described as median, and their 0.72%), gastroenteritis (3 dogs, 6 eyes/276; 2.17%), epi-
95% confidence intervals (95% CI) and between-group lepsy (1 dog, 2 eyes/276; 0.72%), chronic kidney disease (1
analysis was performed using the Mann–Whitney U test. dog, 2 eyes/276; 0.72%), leishmaniasis (4 dogs, 8 eyes/276;
In the case of qualitative variables, the comparison be- 2.90%), traumatic pneumothorax (1 dog, 1 eye/276; 0.36%),
tween groups was performed with Fisher's Exact test and immune-mediated sacculitis (1 dog, 1 eye/276; 0.36%),
they were described by the absolute frequency and percent- and intrahepatic shunt (1 dog, 2 eyes/276; 0.72%). Atopy
age over the study group. Grade of severity was described was the most common dermatological condition in both
as qualitative variable and analyzed according to ordinal groups and especially in AD (18 dogs, 33/124 eyes; 26.61%
variable with a Mann–Whitney U test. A two-sided Type I AD vs. 10 dogs, 18/152 eyes; 11.84% YD) (p = .033). Other
error of 5% was established for statistical significance. spss dermatological conditions included demodicosis (1 dog, 2
Version 26 (IBM Corp. Armonk, NY. USA) was used for all eyes/276; 0.72%) and zinc-responsive dermatosis (1 dog, 2
statistical analyses. Dogs with incomplete information for eyes/276; 0.72%).
one parameter were excluded for that specific variable. Topical treatment prior to presentation included
broad-spectrum antibiotics in 59/276 eyes (21.37%), dexa-
methasone in 26/276 (9.42%), NSAIDs in 14/276 (5.07%),
3 | R E S U LTS and 0.2% cyclosporine in 9/276 (3.26%). In addition, five
dogs were under systemic treatment for concurrent sys-
A total of 153 dogs (276 eyes) met the inclusion criteria. temic diseases: amoxicillin-clavulanic acid and prednis-
There were 72 entire males, 49 entire females, 11 neutered olone, allopurinol, lokivetmab, oclacitinib maleate, and
males and 21 spayed females. The median age was 3 years, trimethoprim-sulfamethoxazole, one dog each.
ranging from 10 months to 9 years. Several pure and mixed The most common presenting complaints were red-
breeds were included, with mixed breeds being over- ness, blepharospasm and ocular discharge, varying in
represented (29 animals, 18.95%), followed by 13 Golden duration from 4 days to 6 months. FC was an incidental
Retrievers (8.5%), eight Labrador Retrievers (5.2%), six finding in some patients with concurrent ocular disease—
English Bulldogs (3.9%), eight English Setters (3.9%), five in those patients, the present complaint varied from aniso-
Border Collies (3.2%), five Weimaraners (3.2%), and 81 coria to oral pain. Table 1 summarizes the ocular clinical
dogs of various pure breeds. signs, location and severity of the follicles recorded. All
|
14635224, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/vop.13155 by Turkey Cochrane Evidence Aid, Wiley Online Library on [20/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4 CERRADA et al.
parameters were recorded for all animals, except for the Conjunctival cytology was performed in only five
location of the follicles on the bulbar surface of the NM, eyes (5/276; 1.8%). All eyes were severely affected and
which was assessed only in 238/276 eyes. showed a predominantly lymphoplasmacytic cellular
The severity of follicles was statistically greater in YD infiltrate.
(more 2 and 3 grades) than in AD (more 0.5 and 1 grades) All dogs received topical anti-inflammatory med-
(p = .032) and was not associated with atopy in either ication for the treatment of the FC, either 0.1% di-
group (p = .892). Considering all dogs, YD and AD, uni- clofenac sodium solution (Voltaren®, Théa Laboratories,
lateral cases showed less severity (more 0.5 and 1 grades) Barcelona, Spain) or 0.1% dexamethasone solution
than bilateral cases (more 2, 3, and 4 grades) (p = .025). (Maxidex®, Alcon Laboratories, Barcelona, Spain).
The distribution of the follicles was very similar in both Based on the more limited effects of NSAIDs on the syn-
groups except for the more frequent involvement of the thesis and action of inflammatory mediators compared
bulbar surface of the NM in YD (62 eyes; 46.27% vs. 27 to corticosteroids, diclofenac was used in all the cases
eyes; 25.96% AD). In addition, statistical differences be- graded as mild (grades 0.5 or 1) or when a concomi-
tween the two groups were observed only for this particu- tant ocular condition precluded the use of corticoste-
lar location (p = .020). roids [corneal ulcers (three eyes)]. For YD, 38 eyes were
Concurrent ocular diseases included chronic dry treated with diclofenac (25%) versus 114 eyes treated
eye (10/276 eyes; 3.62%), anterior uveitis (5/276 eyes; with steroids, and for AD, 53 eyes were treated with di-
1.81%), corneal ulcers (3/276 eyes; 1.08%), optic neuritis clofenac (42.74%) versus 71 eyes treated with steroids
atrophy, periocular abscess, and phthisis bulbi (1/276 (p = .255). The posology differed according to the sever-
eyes; 0.36%, each). ity and ranged from 1 drop twice a day to 1 drop four
|
14635224, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/vop.13155 by Turkey Cochrane Evidence Aid, Wiley Online Library on [20/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
CERRADA et al. 5
times a day in the affected eye until the first follow-up and histology. Bilaterality is a common feature of the three
visit, which was scheduled at 1 month at our institution. diseases,10,27 but the clinical hallmark differs between
If signs of FC persisted, medication was continued until them. CFC is manifested by follicles4 (translucent nodu-
the next review at 2 months (1 month after the first fol- lar structures outlined by capillaries), VC by giant papillae
low-up). If the signs of FC had resolved, treatment was (inflammatory thickening with central blood vessels), and
either discontinued (diclofenac) or tapered and discon- trachoma by both features. In addition, VC and trachoma
tinued over 2 weeks (dexamethasone). may induce formation of pseudomembranes and scarring,
At the first follow-up (1 month), resolution of clinical features that have not been previously reported in CFC
signs was observed in 100 eyes of AD (80.64%) and in 94 literature, nor observed in our study.10,18,28 Although the
eyes of YD (61.84%) (p = .001). In both groups, the healing mechanisms of scar formation are not fully understood,
time was shorter in eyes treated with diclofenac [median chronicity, severity, etiologic agent, differences in the in-
26 days (95% CI 7; 30) in YD and median 21 days (95% CI flammatory infiltrate and necrosis grading, may play a
14; 45) in AD] than in those treated with dexamethasone role in its development. In trachoma, conjunctival scar-
[median 30 days (95% CI 30; 50) in YD and median 30 days ring has been found to result from follicular necrosis, in-
(95% CI 30; 45) in AD], nevertheless, this difference was ducing “star-shaped” scars and semilunar limbal scars,
statistically significant only in YD (p = .009). or Herbert's pits.7 The latter is formed by the thinning of
the superior cornea as the follicles heal.18 In VC, scarring
is shown as punctate limbal epithelial keratopathy, or
4 | DI S C USSION Trantas’ dots.10 Those consist of clumps of necrotic eosin-
ophils, neutrophils, and epithelial cells collected in crypts
This is the first retrospective characterization of CFC, eval- at the junction of the cornea and conjunctiva.29
uating the clinical features, comorbidities, and response to The distribution of follicles varies among diseases,
treatment. Published data on this disease are scarce and being more prevalent on the limbal and/or dorsal bulbar
textbook based, therefore the discussion is mainly based conjunctiva in VC10; and on the palpebral conjunctiva and
on human literature. Among all types of human conjunc- fornices in trachoma. In the present study, the NM was
tivitis, vernal allergic conjunctivitis (VC) and trachoma more frequently involved than the other regions, followed
show the greatest similarity to CFC.10,20,24 by the palpebral conjunctiva. Interestingly, a recent study
Young human patients are more commonly affected in dogs localized pronounced allergic reactions more fre-
by VC or trachoma and young dogs by FC.1,4,5,25 Indeed, quently in the ventral conjunctiva and NM, where most
in the present study, follicles were more common in dogs allergens accumulate.30 This finding would suggest an al-
younger than 18 months, mirroring the young age of pa- lergic origin for CFC in our study.
tients with VC (5–25 years, with an onset at 6–7 years of Although the pathogenesis of allergic conjunctivitis
age).10 Additionally, VC tends to resolve spontaneously historically described follicle formation,1,31,32 in a recent
after puberty, when the presumed overexpression of con- study follicles are not a common feature.30 This could
junctival sex hormones (estrogen and progesterone) be- question whether “allergy” is a truly trigger for follicle
gins to decrease.10,25,26 This may explain why young dogs formation. Despite the aforementioned publication, we
in our study were more frequently and more severely af- understand that atopy should be considered as one of
fected than adults. the potential etiologies for follicle formation. Further
Contrary to what has been described for VC, where studies are needed to establish the etiopathogenesis of
boys are more likely to be affected than girls, this study this common condition. In this study, atopy was the
found no gender predisposition for CFC. However, gen- most common systemic comorbidity, with a higher inci-
der susceptibility becomes equal after puberty in both dence in AD, possibly explained by the usual late onset
conditions.10 of canine atopy (between 6 months and 3 years of age).33
Anatomical factors such as skull type and eyelid anat- Similarly, allergic conjunctivitis has been associated
omy, may play a role in the conjunctival response to poten- with atopy, in both humans and dogs, with incidences
tially harmful substances.1,2 In the herein, brachycephalic of 25%–45%34,35 and 94%,30 respectively. Similarly, VC,
breeds showed no predisposition to CFC, which could considered an allergic disease, has classically been asso-
be explained by the physiologic exophthalmia which, by ciated with atopic diseases such as asthma, allergic rhi-
moving the eyelids forward, reduces the conjunctival sac, nitis, or atopic dermatitis.36–38 Notwithstanding, recent
thus diminishing the possibility of harvesting foreign bod- studies suggest a different pathogenesis from the classi-
ies or contact with allergens. cally considered Type 1 IgE-mediated hypersensitivity,39
The differences between CFC, VC and trachoma are with skin tests often negative in people affected with
mainly reflected in the clinical presentation, pathogenesis, VC. Trachoma, which has a bacterial etiopathogenesis
|
14635224, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/vop.13155 by Turkey Cochrane Evidence Aid, Wiley Online Library on [20/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
6 CERRADA et al.
(Chlamydia trachomatis), has not been associated with In veterinary medicine, follicular debridement has
atopy. been supported by some authors for refractory cases, al-
Histopathology facilitates the diagnosis of these three though it may exacerbate the situation by predisposing to
diseases. Although CFC has not been histologically chronic conjunctivitis.24 Certainly, there are no published
characterized, VC and trachoma have been extensively data on its benefit. In human medicine, cryotherapy and
described. The conjunctival giant papillae, commonly CO2 laser have been described for refractory VC cases,
seen in VC, are formed by deposition of mixed cellu- although their long-term efficacy is unclear.10,42,43 In the
lar infiltration and new collagen, whereas the follicles authors' opinion, the integrity of the mucosa-associated
seen in trachoma are composed mainly of lymphoid lymphoid tissue should be maintained, therefore debride-
tissue.4,10,18,32,40,41 ment was avoided in this study.
Diagnosis of these three diseases is mainly based on In the herein, clinical resolution at 1 month follow-up
clinical criteria and conjunctival cytology. Mast cells, ba- was 74.05%, being higher in AD (80.64% eyes) than in
sophils and eosinophils are predominant in VC,10 whereas YD (61.84% eyes). The results showed a faster resolution
intracytoplasmic inclusions of C. trachomatis are the hall- in YD treated with diclofenac sodium compared to dexa-
mark of trachoma. Contrarily to these human diseases, methasone. It should be noted that most of the animals
the cellular response in CFC is predominantly lymphop- treated with the NSAIDs were mildly affected, thus the
lasmacytic with a low eosinophilic component,1,4 which results may be related to a low severity of the condition
was consistent with the cytology results in this study. In rather than to a major anti-inflammatory effect of the
addition, biopsy, although more invasive, has been proven NSAIDs.
to be a valuable diagnostic tool. Other less commonly used This study has several limitations owing to its retro-
diagnostic methods encompass cell culture, immunoflu- spective design. Limitations include the variation of treat-
orescence, enzyme immunoassay, serology, PCR, allergy ment, the short follow-up period, the lack of data for some
testing, tear cytology, and evaluation of tear IgA levels and specific parameters (follicular location), and the potential
proinflammatory cytokines.10 bias in the severity criteria—the assessment of the num-
In human medicine, VC and trachoma may resolve ber of follicles is subjective and needs to be compared with
without treatment. If needed, topical treatment will de- previous apprehended scores. Undoubtedly, a prospective
pend on the etiology, severity, and systemic and ocular study, with a more sophisticated and standardized grad-
concomitant diseases. The therapeutic approach is very ing system, may help to better characterize the disease. It
similar for CFC and VC, consisting of anti-inflammatories, is also worth noting that in a large population, mild cases
immune-modulators and mast cell stabilizers. In terms of of CFC may be underdiagnosed, which could influence
anti-inflammatory agents, topical corticosteroids are con- the results.
sidered the gold standard treatment for CFC, whereas in In summary, canine follicular conjunctivitis is a pre-
VC they are reserved only for severe or refractory cases. dominantly bilateral ocular disease affecting dogs of all
NSAIDs have also been used, although usually for mild ages, although age induces variability in its clinical pre-
cases.10,42,43 The use of immunosuppressive agents, such sentation. While YD are more commonly and severely
as topical cyclosporine, montelukast, tacrolimus, siroli- affected, have bilateral presentation and a more frequent
mus, and mitomycin-C has been described for refractory NM involvement, AD show milder severity and a higher
cases in humans.10,42,43 Mast cell stabilizers and antihista- incidence of concomitant atopy. One month of topical
mines are frequently prescribed for VC, being of question- diclofenac may be adequate for the mildly affected cases,
able efficacy for CFC.1,4 Occlusive treatments (patching, while 1 month of topical 0.1% dexamethasone is recom-
occlusive goggles or tarsorrhaphy) and contact lenses (to mended as a starting point treatment for the moderately
avoid friction from the enlarged palpebral papillae) are to severely affected cases.
also used in human ophthalmology to provide relief.10,42,43
Avoidance of the allergen with saline conjunctival irriga- CONFLICT OF INTEREST STATEMENT
tion or artificial tears has been described in VC10,42,43 and The authors declare no conflicts of interest.
CFC32 to reduce follicle formation and is of particular in-
terest in dogs with deep fornices where large amounts of ORCID
irritants or allergens can be housed. Irene Cerrada https://orcid.org/0000-0002-0665-0202
In humans, systemic treatment is infrequently used Marta Leiva https://orcid.org/0000-0003-2378-0946
because of its side effects. The veterinary literature only Rita Vilao https://orcid.org/0000-0002-6513-0397
recommends systemic therapy when there is concurrent Teresa Peña https://orcid.org/0000-0001-8809-5566
skin disease.1,4,32 Jose Ríos https://orcid.org/0000-0002-0716-8784
|
14635224, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/vop.13155 by Turkey Cochrane Evidence Aid, Wiley Online Library on [20/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
CERRADA et al. 7
14635224, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/vop.13155 by Turkey Cochrane Evidence Aid, Wiley Online Library on [20/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
8 CERRADA et al.