Cyclosporine-Induced Gingival Overgrowth in New Zealand White Rabbits (Oryctolagus cuniculus)

Abstract

A high incidence of gingival overgrowth occurred in a group of New Zealand White rabbits receiving daily cyclosporine (15 mg/kg IM) while on a retinoblastoma study. Over the course of 2 mo, rabbits presented with clinical signs of ptyalism (4 of 18 rabbits), inappetence (3 of 18), or both (3 of 18); facial dermatitis and erythema occurred secondary to ptyalism. Reducing the dose of cyclosporine to 10 mg/kg led to complete resolution of clinical signs in all but 2 rabbits, which then received azithromycin (62.5 mg PO once daily for 7 d), a common treatment for cyclosporine-induced gingival overgrowth in other species. After dose reduction and azithromycin treatment, clinical signs resolved and did not reoccur for the remainder of the study. Fourteen rabbits were necropsied at the end of the study, and gingival width was measured. Although some rabbits were clinically normal, the gingiva in all rabbits was grossly thickened. Rabbits on cyclosporine had molar gingiva that was significantly thicker (4.8 mm) than controls (2.5 mm) not treated with cyclosporine. Histologic analysis of the gingiva revealed mild to moderate gingival epithelial hyperplasia, hyperkeratosis, and mild inflammation. Gingival overgrowth is a known side effect of cyclosporine administration in other species but, to our knowledge, this report is the first description of the condition in rabbits. Because rabbits frequently are used in studies that involve systemic cyclosporine administration, clinicians are advised to include this possibility in their differential list for cases involving hypersalivation, facial dermatitis, or inappetence in rabbits.

Cyclosporine-induced gingival overgrowth (CIGO) or hyperplasia is a well-documented adverse effect of chronic high-dose cyclosporine therapy in dogs and humans.^{2,11,13,15,16,18-20} Cyclosporine is used frequently, and is often the drug of choice, for immunosuppressant therapy in whole-organ transplants, particularly in kidney, heart, and bone marrow recipients.¹² The incidence of CIGO in patients receiving cyclosporine therapy is reported to be 30% to 50% in humans and 30% to 40% in dogs.^13,17,19 The condition appears after patients have been on therapy for approximately 4 to 6 wk, and symptoms range from mild inflammation to severe inflammation and hemorrhagic gingiva.² The pathogenesis of CIGO is not understood entirely. Some of the predisposing factors include preexisting gingival inflammation, presence of dental plaque, deep periodontal pockets, and high doses of cyclosporine coupled with long durations of drug administration.^{10,13,15,17-19} Therefore the status of oral health prior to and during drug administration in combination with drug serum levels and duration are key factors in the onset of the condition.^13,17,19 Discontinuing or dramatically lowering the dose of the cyclosporine often results in resolution of clinical signs and lesions.^10,13,17 However, discontinuation of cyclosporine therapy or decreasing the dose is not always medically feasible, particularly in whole-organ transplant recipients.

Due to the large size of their globe, rabbits are commonly used for ophthalmologic research including ocular tumor studies. Human tumor cell lines are used often in these animal models, requiring aggressive immunosuppression to prevent transplant rejection. Although cyclosporine is formulated for topical ophthalmologic administration, sometimes the level of immunosuppression provided by this method is insufficient for adequate tumor growth to meet experimental needs^3,4. In these cases, and to achieve greater levels of immunosuppression, cyclosporine is administered systemically.^3,4 The present report shows that New Zealand White rabbits are prone to developing CIGO, similar to the lesions observed in humans and dogs.

A high incidence of inappetance, ptyalism (Figure 1), and facial dermatitis occurred among 10 New Zealand White rabbits used in a retinoblastoma study. We hypothesized that New Zealand White rabbits experience cyclosporine-induced gingival overgrowth when exposed to chronic (longer than 4 wk) high-dose (10 mg/kg) cyclosporine. In addition to reporting the clinical investigation, we describe risk factors for developing CIGO and discuss methods to alleviate it.

Figure 1.

A) Ptyalism in a NZW rabbit treated with cyclosporine (15 mg/kg). B) Cyclosporine-treated rabbit 2 d after decreasing dose from 15 mg/kg to 10 mg/kg.

Materials and Methods

Humane care and use of animals.

Rabbits were housed in an AAALAC-accredited facility and in compliance with the Guide for the Care and Use of Laboratory Animals.⁸ All procedures involving animal use were approved by the Institutional Animal Care and Use Committee at Emory University.

Animals.

Female SPF New Zealand White rabbits (age, 3 mo; average weight, 3 kg) were purchased from a commercial source (Myrtle's Rabbitry, Thompson Station, TN) and singly housed in standard caging (Lab Products, Seaford, DE). All rabbits were tested and found free of Psoroptes cuniculi, Cheyletiella parasitovorax, Listrophorus gibbus, Passalurus ambiguous, Taenia pisiformis, Eimeria stiedae, Eimeria perforans, Eimeria irresidum, Encephalitozoon cuniculi, other intestinal coccidia, Pasteurella multocida, Pasteurella pneumotropica, Bordetella bronchiseptica, Treponema cuniculi, Clostridium piliformis, and oral papilloma virus. Rabbits were maintained on a 12:12-h light:dark cycle and received commercial diet (Hi-Fiber Rabbit Diet 5326, Purina Mills, Richmond, IN), fresh fruits, and vegetables daily and reverse-osmosis water ad libitum.

Experimental procedures.

Rabbits in the first set of 10 clinical cases as well as the additional 18 rabbits involved in this clinical investigation were subjects in a retinoblastoma study. Each rabbit received a single intraocular injection (right eye; 30 µL) of a suspension of human retinoblastoma WERI-Rb cells (HTB 169, American Type Culture Collection, Manassas, VA). To facilitate tumor growth, the rabbits were immunosuppressed with daily injections of cyclosporine (Sandiummune 50 mg/mL, Novartis Pharmaceuticals, Cambridge, MA) on a projected schedule of 15 mg/kg IM for 4 wk and then 10 mg/kg IM for an additional 4 wk.⁴ Per the experimental protocol, 2 rabbits were euthanized every week beginning the second week after tumor cell inoculation. Standard endpoint guidelines from the Emory University Institutional Animal Care and Use Committee were applied to this research protocol. Under these guidelines, subjects were euthanized if they attained 25% weight loss from baseline weight when assigned to the protocol, experience signs of organ failure unresponsive to treatment, or fail to feed themselves for 4 d. Four age-matched control animals, which underwent no experimental manipulation other than receipt of 3 mL heparin (Heparin sodium injection, 1000 U/mL, Baxter Healthcare Corporation, Deerfield, IL) prior to euthanasia, were used for comparison. Rabbits were sedated with ketamine (30 mg/kg IM) and xylazine (4 mg/kg IM) and then euthanized with sodium pentobarbital (50 to 65 mg/kg IV).

According to our experimental plan, rabbits in the investigational cohort (n = 18) that began to exhibit clinical signs would undergo cyclosporine dose reduction from 15 mg/kg to 10 mg/kg initially. If the dose decrease did not result in resolution of clinical signs within 1 wk, the animal would receive azithromycin (62.5 mg PO) for 7 d. Animals that became inappetant or anorexic were offered supplemental preferred fruits and vegetables in addition to the normal diet.

Clinical and pathologic analysis.

For our clinical investigation, the schedule of experimental euthanasia was correlated with the duration and severity of clinical signs. The 2 animals with the most severe or longest-standing clinical signs were euthanized each week. At the request of the principal investigator, opioids and antiinflammatory agents were avoided in light of concerns regarding interference by these drugs with the research outcome and objectives of the original study. Instead, if pain, stress, or distress could not be alleviated by other means, animals were euthanized to prevent any additional or anticipated suffering. Other treatments consisted of providing soft, palatable foods; decreasing the dose of cyclosporine; and, in those that did not respond to the dose decrease, a short course azithromycin. A total of 14 investigational and 4 control animals underwent necropsy, and their skulls were harvested for gingival measurements. Measurements were taken at maximal thickness of the molar gingiva in the region of M1–P3 for the maxillary teeth and M1–P2 for the mandibular teeth. The total thickness of the gingiva and corresponding tooth was measured, and the width of the tooth was subtracted from the total to obtain an estimate of gingival thickness.

Tissue samples of the gingiva in all 4 quadrants were collected at necropsy in 4 experimental rabbits and 4 control animals and compared. In addition, the bilateral labial and buccal molar and incisor gingiva were evaluated histologically for evidence of inflammation, overgrowth, and other abnormalities in gingival architecture. Tissues were fixed in 10% buffered formalin for 4 d, decalcified (Cal-EX decalicifying solution, Fisher Scientific, Fair Lawn, NJ) for 72 h, embedded in paraffin, sectioned at 5 μm, and stained with hematoxylin and eosin.

Data analysis.

Subjects were separated into 3 groups for statistical analysis not only in an effort to maintain sufficient sample size but also identify when gingival overgrowth became significant. The first group consisted of control animals (n = 4) that had not received cyclosporine therapy, the second was an early-treatment group (n = 7) consisting of animals on cyclosporine therapy for 4 wk or less, and the third was a late-treatment group (n = 7) consisting of animals on cyclosporine therapy for more than 4 wk. Results are expressed as mean ± SEM. All data underwent ANOVA; multiple pairwise comparisons between means were performed by using the Tukey test for comparisons between group means (SigmaStat, Systat Software, Chicago, IL). Differences between groups were considered statistically significant when the P value was less than 0.05.

Results

This clinical investigation confirmed our hypothesis that New Zealand White rabbits are susceptible to CIGO similar to that in dogs and humans receiving chronic (4 wk or more) systemic cyclosporine. Retrospective review revealed that of the 18 rabbits comprising the cohort yielding the initial clinical cases, 10 animals exhibited clinical signs after 4 wk or more of cyclosporine (15 mg/kg) treatment. Among the 18 rabbits enrolled in our clinical investigation, signs and symptoms of CIGO had a 100% incidence rate in animals treated with 15 mg/kg cyclosporine daily for 2 wk or more. Clinical signs consisted of inappetence or ptyalism (or both), often with concurrent dermatitis around the muzzle and neck secondary to ptyalism. In severe cases, hyperemia or thickening of the gingiva around the incisor teeth (or both) was present. This incidence necessitated prematurely decreasing the dose of cyclosporine from 15 mg/kg to 10 mg/kg during week 3 rather than after week 4 as planned and as done with the clinical case cohort. All but 2 of the 18 subjects responded to decreasing the cyclosporine dose from 15 mg/kg to 10 mg/kg; these 2 rabbits subsequently responded to treatment with azithromycin. In addition, 1 rabbit had to be removed from the study due to respiratory infection with Bordatella bronchiseptica.

Gross pathology.

Postmortem examination revealed bilateral thickening of the mandibular and maxillary molar gingiva on the lingual and buccal surfaces in all cyclosporine-treated rabbits, including 4 rabbits from the initial cohort and 14 from the experimental cohort. The molar gingiva of the experimental animals appeared grossly thicker in width and depth as compared with that of the control animals (Figure 2). In control animals, the occlusal surface of the molar gingiva was (on average) 1 to 2 mm below the occlusal surface of the respective tooth. In experimental animals with thickened gingiva, the gingival border typically was either flush with the respective tooth's occlusal surface or overgrown past it.

Figure 2.

These images demonstrate the grossly visible thickening and disfigurement of the maxillary molar gingiva of experimental rabbits (right) versus the control animal (left). Note the lack of gingival hyperemia or thickening.

The mean thickness of the maxillary molar gingiva in the late treatment group (0.486 cm) was significantly (P < 0.05) greater than those of the control group (0.25 cm) and early-treatment group (0.31 cm). The thickness of the gingiva surrounding the mandibular molars in experimental animals did not differ from that of control animals or between the early- and late-treatment groups (Figure 3). Gingival thickness in 1 of the rabbits treated with azithromycin (the other animal could not be evaluated) was similar to that of other rabbits treated for the same length of time but not receiving azithromycin. Significant gross lesions were not apparent in any other organ.

Figure 3.

Gingival thickness in NZW rabbits. Early, cyclosporine therapy for < 4 wk; Late, cyclosporine therapy for > 4 wk; Oral location 1, left maxillary molar; oral location 2, left mandibular molar; oral location 3, right maxillary molar; oral location 4, right mandibular molar. *, Statistically significant difference (P < 0.05) between groups. Error bar, SEM.

Histology.

Tissue sections of the molars and incisors from 4 rabbits—2 rabbits from the early and 2 from the late group—were examined. Histologic examination of the gingiva from all animals revealed increased numbers of degenerate and viable keratinocytes (orthokeratosis and parakaratosis) in the stratified squamous epithelium (Figure 4). Hyperkeratinization varied in severity among experimental animals, which showed increased fibroblasts and collagen in the lamina propria as compared with those in the control rabbits (Figure 5). In addition, the lamina propria in 3 of the 4 animals evaluated had moderate multifocal heterophilic infiltrates. Lesions were more pronounced in the molar gingiva than incisor gingiva, correlating with the gross findings.

Figure 4.

A) Molar gingiva from a control animal. Note the smooth, organized stratified squamous epithelium compared with the disorganized, hyperkeratotic squamous epithelium of the experimental animals in panels B through D. Magnification, ×4. B) This photograph of the molar gingiva from rabbit 12 (7 wk of cyclosporine therapy) demonstrates severe thickening of the stratified squamous epithelium with hyperkeratosis. Magnification, ×4. C). Molar ginigva from rabbit 7 (6 wk of cyclosporine therapy with a 7-d course of azithromycin during week 3). Magnification, ×4. D) Orthokeratosis and parakeratosis of the squamous epithelium in severe lesions of animal shown in panel C. Hematoxylin and eosin stain; magnification, ×20.

Figure 5.

A) Photomicrograph of the lamina propria of rabbit 12, depicting an increase in connective tissue (specifically fibroblasts and collagen). B) Photomicrograph of the lamina propria in a control animal. Hematoxylin and eosin stain; magnification, ×20.

Discussion

Cyclosporine-induced gingival overgrowth after chronic high-dose cyclosporine therapy occurs spontaneously in dogs and humans and has been induced experimentally in rats, monkeys, and mongrel cats.¹⁰ This condition has not been described previously in rabbits. In our clinical investigation, CIGO was grossly apparent in 100% of rabbits that were on high-dose cyclosporine for 4 wk or longer, with lesions appearing as soon as 2 wk after treatment initiation in some animals. On gross examination, the lesions in rabbits were highly similar to those in humans and consisted of thickened overgrown gingiva surrounding the maxillary molar teeth, with less-pronounced thickening around the mandibular molars. In severe cases, the gingiva was thickened and the surface was irregular, in contrast to the smooth architecture of normal gingiva. Thickening occurred both labially and buccally but was more pronounced buccally. Lesions in humans have “marked lobulated enlargement,”¹⁶ affecting the incisor gingiva predominantly, and are more anterior than those in rabbits.^10,15-19 Human cases also show not only thickening but also associated gingivitis that can involve severe hemorrhagic lesions in addition to inflammation and irritation.^14,18 These lesions in humans are reported to be painful,¹⁶ and the inappetence in our rabbits implies that CIGO lesions may be painful in this species as well. Therefore pain management may be beneficial in animals that are anorexic or have a decreased appetite, those in which the cyclosporine dose cannot be decreased, and those in which other treatments are not an option. Furthermore, the wet fur around the muzzle and neck, possibly due to ptyalism and excessive dribbling when drinking, may be a sign of oral or gingival pain as well. The observed dermatitis and irritation most likely are secondary to long-standing, excessively moist skin and fur, possibly compounded with a secondary bacterial or yeast infection, although no abnormal pathologic microorganisms were cultured from skin.

In our rabbits, the most significant difference histologically between normal and affected animals was the difference in the thickness of the stratified squamous epithelium and cellular components of the lamina propria. The epithelium was significantly thicker in affected animals, due to orthokeratosis and parakeratosis. In severe cases, the stratification in this layer was lost and was replaced by a disorganized collection of similar cells. In humans, CIGO is characterized by an increase in connective tissue matrix especially with respect to type I collagen production.^13,17,19 Epithelial acanthosis has also been reported with CIGO lesions in humans,¹⁹ perhaps due to an enhanced keratinocyte lifespan rather than an increase in keratinocyte proliferation.¹³ Although the pathogenesis is not understood entirely, among humans, increased blood serum drug levels, prolonged drug administration, poor oral hygiene, gender, and age (children and adolescents are predisposed) can all contribute to a patient's susceptibility to developing lesions.^13-15,17

Cyclosporine toxicity, especially with respect to nephrotoxicity, in rabbits has been documented,^8,11 but none of the reports note gingival overgrowth as an adverse side effect. This lack may be due to either insufficient dose or duration of treatment or to the inability to discern the lesion antemortem or postmortem without disarticulating the jaw. Given the combination of nonspecific clinical signs and the inability to visualize the lesion, a clinician might attribute the observed signs to gastrointestinal abnormalities. For example, a previous case report⁷ describes clinical signs in a group of 153 rabbits with chronic high-dose cyclosporine therapy that were consistent with those in our animals. In the case report, the top differential diagnoses were associated with lower gastrointestinal disease. Several treatments were attempted with respect to anorexia and subsequent weight loss, but the only effective intervention was decreasing the dose, which decreased mortality.⁷ In that report, the authors primarily focused on the lower gastrointestinal tract and never addressed evaluation of the teeth or gingiva.

Possible treatment options to ameliorate CIGO include: decreasing or discontinuing the cyclosporine dose; improving dental hygiene; administering a short course of azithromycin; and surgical debridement of overgrown tissue. Although decreasing or discontinuing cyclosporine leads to clinical improvement or resolution in almost 100% of cases, this option may be detrimental to research requiring aggressive immunosuppression. Improving dental hygiene is not always an easy or viable option in veterinary research patients. A 3- to 5-d course of azithromycin typically results in reduction or even elimination of cyclosporine-induced gingival overgrowth in humans,^{1,5,6,9,16,20} with posttreatment duration of regression or resolution of 3 mo to 2 y. One study documents amelioration of CIGO lesions in humans as far as 3 mo after the use of azithromycin-containing toothpaste twice daily for 4 wk.¹ In addition, the success rate of treatment does not appear to diminish with subsequent treatments. In our experience, most animals showed a resolution of clinical signs with a dose reduction alone, and the 2 animals that did not do so subsequently responded to azithromycin therapy. Although they experienced significant clinical improvement, necropsy revealed that all rabbits on cyclosporine therapy had abnormally overgrown gingiva.

Here we report that CIGO can be a prominent side effect of high-dose cyclosporine therapy in New Zealand White rabbits. If untreated, this lesion can result in clinical signs of inappetance, ptyalism, and anorexia. Clinicians and researchers should keep this potential side effect in mind when designing research protocols using this drug and when treating animals undergoing cyclosporine therapy.