VOP176.fm Page 93 Wednesday, May 30, 2001 5:17 PM

Veterinary Ophthalmology (2001) 4, 2, 93 – 98 Original Blackwell Artical Science, Ltd

Lipogranulomatous conjunctivitis: clinical findings from 21 eyes in 13 cats Robert A Read* and Julia Lucas† *Veterinary Ophthalmic Referrals, 102 Magill Rd, Norwood, SA 5067, Australia, †Idexx Veterinary Pathology Services – Adelaide, 33 Flemington St, Glenside, SA 5065, Australia

Address communications to: R. A. Read Tel.: + 61 881320533 Fax: + 61 881320633 e-mail: [email protected]

Abstract Objective The clinical and, where available, histologic features of lesions fitting a previous description of lipogranulomatous conjunctivitis were reviewed retrospectively from a series of cats. Animals studies Thirteen cats (21 eyes) with lesions consistent with a diagnosis of lipogranulomatous conjunctivitis were seen over a 4-year period in a referral practice in southern Australia. Affected cats ranged in age from 6 to 16 years (mean 11.2 ± 3.6 years). Results Twelve cats were domestic shorthair or longhair cats and one was a British shorthair cat. Five cats were white, four were bicolor, two were orange tabby and two were dilute orange tabby. Lesions were always found in the palpebral conjunctiva, adjacent to the eyelid margin. They affected upper eyelids only in five cases, and upper plus lower eyelids in eight cases. Lesions were bilateral in eight cases. Lesions were excised in eight cases, with resolution of associated irritation and no signs of lesion recurrence during a 4- to 21-month follow-up period. Summary Histologic features of tissues excised from surgically treated cases were consistent with a previous report of lipogranulomatous conjunctivitis and with descriptions of chalazia. A possible role of actinic radiation in the etiopathogenesis of lesions is discussed. Key Words: cat, chalazion, lipogranulomatous conjunctivitis, meibomian gland

IN TR OD U C TI ON

Lipogranulomatous conjunctivitis in cats has been reported previously with a description of the histologic findings from eight conjunctival biopsy specimens.1 Otherwise, it has been briefly mentioned in an atlas of veterinary ophthalmology.2 Lesions of similar appearance are included in two other atlases of veterinary ophthalmology under the descriptions of ‘chalazion’ and ‘subconjunctival masses’.3,4 One of the authors (RAR) noted a particular lesion distribution and signalment in cats with lesions consistent with lipogranulomatous conjunctivitis. The aim of this retrospective case series is to report clinical features of the disease and suggest a hypothesis regarding etiopathogenesis of this condition. MA TER I AL S AND M ETH ODS

The clinical and, where available, histologic features of lesions fitting a previous description of lipogranulomatous conjunctivitis were reviewed retrospectively from a series of © 2001 American College of Veterinary Ophthalmologists

cats seen over a 4-year period by one of the authors (RAR).1 All cases were referred to a private veterinary ophthalmic practice in Australia between 1996 and 2000. Clinical features recorded included signalment ( breed, age, sex, coat color), history prior to referral, appearance of lesions, distribution of lesions and any concurrent ocular or systemic disease. Response to medical or surgical treatment was recorded. Histologic findings available from some of the surgically treated cases were reviewed. R E S U LT S

Thirteen cats (21 eyes) with lesions consistent with a diagnosis of lipogranulomatous conjunctivitis were seen over a 4-year period in a referral practice in southern Australia (Table 1). Twelve were domestic shorthair or longhair cats, and one was a British shorthair cat. Nine animals were spayed females and four were neutered males. Affected cats ranged in age from 6 to 16 years (mean 11.2 ± 3.6 years). Five of the cats were white, four were bicolor but predominantly white (including all or a large proportion of the eyelids),

© 2001 American College of Veterinary Ophthalmologists, Veterinary Ophthalmology, 4, 93–98

Breed

Sex

Age in years

Color

Presenting problem

Eyelids affected by LG

Other ocular findings

Treatment

1

DSH

FS

14

Tabby & white

Blepharospasm & epiphora OD

Upper, OU

Lower entropion OU

LG resection and entropion correction OU

2

DSH

MN

6

Ginger & white

Epiphora OD

Upper & lower OU

3

DSH

FS

10

White

Mucoid discharge OD

Upper & lower OU

4

DSH

FS

12

White

Tumor, external upper eyelid OD

Upper, OU

5

DSH

FS

16

Black & white

Eyelid swelling OD

Upper & lower OU

6

DSH

FS

15

Cream

Blepharospasm & epiphora OD

Upper & lower, OD

7

DSH

MN

11

White

Blepharospasm & epiphora OD

8

DLH

FS

6

Cream

9

DSH

MN

14

10

British SH

FS

11

DSH

12

DSH

Histology ( LG)

Outcome

Follow-up

Yes

Resolution of all problems

21 months

LG resection OU

No

Resolution of all problems

12 months

LG resection OU

Yes

Resolution of all problems

16 months

SCC debulk and cryosurgery; LG lesions not treated

N/A

No signs of further irritation or SCC recurrence

14 months

LG resection OU

Yes

Resolution of all problems

9 months

Corneal ulcer OD

Topical antibiotics; anesthesia for LG resection avoided due to previous problems

N/A

Healing of ulcer and no signs of further irritation

3 months

Upper & lower OU

Low entropition, OD

LG resection and entropion correction declined

N/A

N/A

Lost to follow-up

Blepharospasm & epiphora OU

Upper & lower OD

Corneal ulcers OU

Topical antibiotics; LG resection delayed

N/A

Healing of ulcers and no further signs of irritation

5 months

Ginger

Mucoid discharge OD

Upper & lower OU

Central visual deficits and CNS signs following previous anesthetic

Topical antibiotics; anesthesia for LG resection avoided due to previous problems

N/A

Ongoing discharge & irritation

6 months

8

White

Blepharospasm & epiphora OS

Upper, OU

LG resection OU

No

Resolution of all problems

14 months

MN

9

Ginger

Mucoid discharge OD

Upper & lower, OD

Hyper-pigmented iris OS

LG resection OD

No

Resolution of all problems OD

7 months

FS

9

White

Nictitans tumor OS

Upper, OS

9 mm diameter SCC, nictitans OS

SCC and LG resection OS

Yes

Resolution of all problems

4 months

20 mm diameter SCC upper eyelid skin OD

DSH, domestic short hair; DLH, domestic long hair; FS, female spayed; MN, male neutered; OD, right eye; OS, left eye; OU, both eyes; LG, lipogranuloma; N/A, not applicable; SCC, squamous cell carcinoma.

READ AND LUCAS

Case Number

VOP176.fm Page 94 Wednesday, May 30, 2001 5:17 PM

94

Table 1 Summary of signalment, lipogranuloma distribution, other ocular disease, treatment, and outcome for cats in this series

VOP176.fm Page 95 Wednesday, May 30, 2001 5:17 PM

FELINE LIPOGRANULOMATOUS CONJUNCTIVITIS

Figure 1. Small isolated lipogranuloma, upper palpebral conjunctiva: 14-year-old domestic cat (case 1).

Figure 2. Row of large lipogranulomatous nodules, upper palpebral conjunctiva: 16-year-old domestic cat (case 13).

two were orange tabby (‘ginger’); and two were dilute orange tabby (‘cream’). Most cases had a history of chronic ocular discomfort (blepharospasm, excess lacrimation), although occasionally the lesions appeared to be incidental

95

Figure 3. Row of large lipogranulomatous nodules, upper and lower palpebral conjunctiva: 11-year-old domestic cat (case 3).

to other ocular disease. In all cases lesions were smooth, nonulcerated, cream or white conjunctival nodules varying from one or two lesions 1 mm in diameter (Fig. 1) to overlapping rows of relatively large bulbous lesions up to 5 mm in diameter (Figs 2 and 3). Lesions were always found in the palpebral conjunctiva adjacent to the eyelid margin and affected the upper eyelids only in five cases, and upper plus lower eyelids in eight cases. Lesions were bilateral in eight cats. Concurrent ophthalmic diseases included lower eyelid entropion (three eyes), corneal erosions (three eyes) and squamous cell carcinoma (two eyes). Lipogranulomatous lesions were not found elsewhere in these cats; however, extensive diagnostic testing for disorders of lipid metabolism was not performed. Prereferral treatment was generally limited to topical and systemic antibiotic and corticosteroid preparations. Postreferral treatment was surgical, medical, or conservative. Surgical treatment was recommended where lesions appeared to be a source of irritation. Surgical treatment involved a modification of a previously described conjunctival resection technique used to treat distichiasis in dogs.5,6 Two incisions were made in palpebral conjunctiva parallel to the eyelid margin and on either side of the row(s) of lesions. A strip of up to 5 mm of conjunctival and subconjunctival tissue containing the lesions was then excised. Surgical sites were not sutured and postoperative management consisted of topical antibiotics for 7 days. Eight cases (five bilateral, three unilateral) underwent surgery.

© 2001 American College of Veterinary Ophthalmologists, Veterinary Ophthalmology, 4, 93 – 98

VOP176.fm Page 96 Wednesday, May 30, 2001 5:17 PM

96

READ AND LUCAS

Figure 4. Histologic section of excised lesion showing large, clear spaces presumed to be lakes of lipid, periglandular fibrosis and granulomatous inflammation of the lamina propria (case 3, H&E, ×25 objective).

Figure 5. Histologic section of excised lesion showing mixed inflammatory cells surrounding but not extending into meibomian glands (case 3, H&E, ×40 objective).

Figure 6. Histologic section of excised lesion showing inflammatory infiltrate including macrophages and giant cells containing clear material presumed to be lipid droplets (case 3, H&E, ×40 objective).

© 2001 American College of Veterinary Ophthalmologists, Veterinary Ophthalmology, 4, 93–98

VOP176.fm Page 97 Wednesday, May 30, 2001 5:17 PM

FELINE LIPOGRANULOMATOUS CONJUNCTIVITIS

This was associated with resolution of all signs of irritation and no signs of significant recurrence during a 4- to 21month follow-up period. Medical management with topical antibiotic ointments was used in three cases where irritation from lesions appeared to be present but where surgery was declined due to financial constraints or risks associated with general anesthesia. Medical management appeared to reduce but not eliminate irritation associated with lesions. Conservative management was chosen for one case where lesions showed no signs of irritation and were judged incidental to squamous cell carcinoma involving the upper eyelid skin. Surgery was recommended for the remaining case but this and medical management were declined and the case was lost to follow-up.

Histologic findings Excised tissue from four of the eight surgically treated cases was fixed in formalin and was paraffin embedded. Sections were cut at 4 microns and stained with hematoxylin and eosin for histologic examination. Tissue specimens examined ranged from 3 to 13 mm in length and contained varying numbers of meibomian glands. Conjunctival epithelium was generally ragged and many of the epithelial cells were degenerate or necrotic. The epithelium varied from cuboidal to nonkeratinizing squamous in nature. Goblet cell hyperplasia was observed in one case. The lamina propria contained a mild to moderate inflammatory infiltrate at the epithelial junction. The inflammatory cells were predominantly lymphocytes and neutrophils with low numbers of plasma cells. There was exocytosis of neutrophils into the epithelium. Large, clear spaces presumed to be lakes of lipid, and lined by flattened macrophages were present in the lamina propria (Fig. 4). Inflammatory cell infiltration surrounded these areas, and varied from scattered foci of mixed inflammatory cells to severe infiltration obliterating normal fibrous architecture. The inflammatory infiltration did not extend into meibomian glands (Fig. 5). The inflammatory infiltrate was composed of numerous macrophages, epithelioid macrophages and giant cells together with low numbers of plasma cells and lymphocytes. The cytoplasm of the macrophages and giant cells contained numerous tiny round droplets presumed to be lipid material ( Fig. 6). In two cases the lamina propria contained lymphoid aggregates, but these were not closely associated with the large clear spaces. Some or all of the meibomian glands in each of the sections had varying degrees of periglandular fibrosis. The sebaceocytes in the glands were variably hyperplastic and dysplastic, and there were low numbers of lymphocytes and neutrophils in the central duct of some glands. The duct epithelium was generally normal cuboidal epithelium at the level present in sections. However, keratinizing squamous epithelium was present in occasional meibomian gland ducts in sections from two cases. Some of these glands had ruptured with leakage of sebaceous material into surrounding

97

lamina propria. A focus of inflammatory cells comprising moderate numbers of macrophages, lymphocytes and plasma cells was associated with free sebaceous material. D I S C U S S IO N

Lipogranulomatous conjunctivitis in cats was previously reported as possibly arising from lysis of local lipid-containing cells.1 Trauma and inflammation were proposed as possible causes of leakage of lipid-rich secretion into surrounding tissues. Adipocytes in connective tissues, mucosal goblet cells, and meibomian glands were suggested as possible sources of lipid. Conjunctival connective tissue adipocytes are thought to give rise to lipogranulomatous lesions in humans, and consequently these cells were suggested as possibly being involved in similar lesions in cats. Goblet cells appear unlikely sources of lipid as they secrete predominantly mucopolysaccharide, and the lipogranuloma sections where periodic acid Schiff (PAS) staining was used were negative for this material. Meibomian glands were identified in two of eight biopsy sections examined by these workers but not in the remainder of sections. Conclusions regarding the origin of the lipid material were limited, because the location of the conjunctival biopsy sites was not recorded. Lipogranuloma formation in various tissues has been associated with systemic diseases, such as hyperlipidemia, hypercholesterolemia, and xanthomatosis in humans.7 This appears to be an unlikely cause of lipogranulomatous conjunctivitis in cats as none of the cases reported here or in the previous series had lipogranulomas noted at other sites, or evidence of other disease leading to or resulting from disorders of lipid metabolism.1 Ophthalmic ointments have been proposed as another source of lipid material in lipogranulomatous conjunctivitis in cats.1 This would appear unlikely in the cases reported here given the specific conjunctival location of lesions, although prereferral treatment with ophthalmic ointments was commonplace. In the 21 eyes reported here, all lesions were in the vicinity of the meibomian glands, and meibomian glands were identified in all excised lesion tissue specimens examined histologically. Therefore, it is possible that these glands were the source of lipid material in these animals. The appearance of obvious meibomian gland elements in sections from biopsied or excised lesions will depend largely on the depth and orientation of excised tissue. A lack of meibomian gland elements in previous reports does not rule out their involvement, if the specific site of conjunctival biopsy has not been identified. The cases described here have several key features in common with the previous histologic description of lipogranulomatous conjunctivitis in cats.1 The presence of large droplets of presumed free lipid in the conjunctival lamina propria and granulomatous inflammation were considered the most consistent findings. Goblet cell hyperplasia was

© 2001 American College of Veterinary Ophthalmologists, Veterinary Ophthalmology, 4, 93 – 98

VOP176.fm Page 98 Wednesday, May 30, 2001 5:17 PM

98

READ AND LUCAS

observed in one case but was not a constant feature. In contrast to the previous description, our cases did not exhibit concurrent ocular neoplasia in eyelid lesion sections examined histologically. However, one of the cases had a squamous cell carcinoma affecting the nictitans conjunctiva, and another had a squamous cell carcinoma affecting upper eyelid skin but not the eyelid margin or palpebral conjunctiva. Meibomian gland duct epithelium is normally cuboidal apart from the initial segment and orifice that comprise keratinized squamous epithelium.8,9 Squamous metaplasia of duct epithelium in the salivary gland has been described in hypovitaminosis A, and squamous metaplasia of the conjunctival epithelium is a response to chronic insult including actinic radiation.8 The fact that all affected eyes reported here had relatively little or no pigmentation of the eyelid margins may be significant. In the absence of any evidence of nutritional imbalance, keratinizing squamous epithelium in more proximal sections of the meibomian gland ducts may represent an extension of chronic actinic radiation damage, and may lead to obstruction of meibomian gland openings, accumulation of secretions, eventual gland rupture and associated inflammation. Biopsy specimens including areas of eyelid margin were not available in this retrospective study but would have been helpful in determining whether there were histologic changes involving the meibomian gland orifices. Two of the four animals on which histologic examination of lesions was performed had meibomian gland rupture associated with keratinizing squamous duct epithelium. The higher frequency of lesions involving the upper eyelids in the cases reported here may be explained by the increased development of the meibomian glands at that site relative to the lower lid.10 The granulomatous inflammation in these cats appears most likely to be a reaction to sebaceous secretions from damaged or ruptured meibomian glands and may represent a form of chalazion.11 If actinic radiation is involved in the pathogenesis of meibomian gland damage, then risk factors contributing to increased disease frequency would be expected to include low levels of eyelid pigmentation, outdoor access, high regional actinic radiation levels, and age. Information on outdoor access was not obtained in the cases described here. Southern Australia does however, have high levels of solar radiation for a large proportion of the year. During a similar 4-year period in an ophthalmic referral practice in the UK (not renowned for high solar radiation levels), one of the authors (RAR) was presented with far fewer cats with lesions characteristic of lipogranulomatous conjunctivitis. All cases reported here were 6 or more years of age. A similar range of 6 to 16 years of age was found in the previous report of lipogranulomatous conjunctivitis. Details of eyelid pigmentation or geographic location of cases was not included.1 Associated ocular diseases in the cases described here appeared to be secondary to irritation from the lipogranulomatous nodules, entropion, or corneal erosions. The diagnosis of squamous cell carcinoma in two cases may also be associated with poorly pigmented adnexa and actinic radiation. The

previously suggested association of lipogranulomatous conjunctivitis with ocular or periocular malignant neoplasia may have been a reflection of the diseases prompting biopsy collection and submission.1 The response to surgical excision of lesions was excellent. Resolution of all signs of irritation and no signs of recurrence during the follow-up period were reported in all cases receiving this treatment. Management with topically applied antibiotic preparations appeared to ease ocular irritation and was an acceptable but less effective alternative treatment when surgery was declined. Excision of the majority of meibomian glandular tissue might be expected to lead to problems with an inadequate lipid component of the precorneal tear film. Polarized light biomicroscopy would be one way of attempting to assess the effects of this procedure on qualitative tear film properties.12 C O N C LU S IO N

Lipogranulomatous conjunctivitis in cats appears to be specifically associated with the meibomian glands and may represent a form of chalazion. The occurrence in only middle-aged or older cats with minimal eyelid pigmentation in this case series suggests a possible role for solar ultraviolet radiation damage in the etiopathogenesis of this disease. Surgical excision of lesions was found to be curative where lesions were causing ocular irritation. R E FE R E N C E S 1. Kerlin RL, Dubielzig RR. Lipogranulomatous conjunctivitis in cats. Veterinary and Comparative Ophthalmology 1997; 7: 177–179. 2. Walde I, Schaffer EH, Kostlin RG. Atlas of Ophthalmology in Dogs and Cats. BC Decker, Philadelphia, 1990. 3. Ketring KL, Glaze MB. Atlas of Feline Ophthalmology. Veterinary Learning Systems, Trenton, 1994. 4. Barnett KC. A Colour Atlas of Veterinary Ophthalmology. Wolfe Publishing, Ipswich, 1990. 5. Spreull JSA. Surgery of the eyelids in small animals. Veterinary Annual 1982; 22: 286 – 290. 6. Long RD. Treatment of distichiasis by conjunctival resection. Journal of Small Animal Practice 1991; 32: 146–148. 7. Townsend DJ, Jacobiec FA. Section IX – Orbit. In: Principles and Practice of Ophthalmology. Volume 3: Clinical Practice (eds Albert DM, Jacobiec FA) W.B. Saunders, Philadelphia, 1987; 2094–2096. 8. Jubb KVF, Kennedy PC, Palmer NC. Pathology of Domestic Animals, 4th edn. Academic Press, San Diego, 1993. 9. Jester JV, Nicolaides N, Smith RE. Meibomian gland studies: histologic and ultrasound investigations. Investigative Ophthalmology and Visual Science 1981; 20: 537–547. 10. Samuelson DA. Ophthalmic anatomy. In: Veterinary Ophthalmology, 3rd edn (ed. Gelatt KN ) Lippincott, Williams & Wilkins, Philadelphia, 1999; 31–150. 11. Bedford PGC. Diseases and surgery of the canine eyelid. In: Veterinary Ophthalmology, 3rd edn (ed. Gelatt KN), Lippincott, Williams & Wilkins, Philadelphia, 1999; 535–568. 12. Carrington SD, Bedford PGC, Guillon J-P et al. Polarized light biomicroscopic observations on the pre-corneal tear film. 3. The normal tear film of the cat. Journal of Small Animal Practice 1987; 28: 821– 826.

© 2001 American College of Veterinary Ophthalmologists, Veterinary Ophthalmology, 4, 93–98