Corneal erosions heal by coordinated remodeling of epithelial elements. During the initial stages of healing, CECs at the margins of the lesion (ie, wound) develop a migratory phenotype similar to that seen during EMT, which is the transformation of sessile epithelial cells into migrating mesenchymal cells.1–4 The process of EMT involves the loss of intercellular adhesion mediated by desmosomes and adherens junctions and increased protease production that facilitates remodeling of the extracellular matrix and release of epithelial cells from organized epithelium, which results in enhanced cellular motility.5–7 Our group8 and others9–13 have reported that there are EMT-like changes in cellular morphology and motility during cutaneous and corneal wound healing in healthy tissues. Growth factors such as TGF-B appear to play key roles in corneal wound healing and can stimulate expression of many transcription factors, such as members of the snail family, that modulate EMT.14,15
Nonhealing (ie, refractory) corneal ulcers (also called indolent ulcers, spontaneous chronic corneal epithelial defects, and Boxer ulcers) commonly affect middle-aged to older dogs, with the average age of onset being 8 to 9 years.16–18 Medical treatment alone is often ineffective, and because the cornea fails to undergo normal wound healing processes, defects can take up to 180 days to heal.16,19 Refractory corneal ulcers are characterized histologically by a nonadherent epithelial lip, epithelial dysmaturation, and suppurative stromal infiltrates.16,19 Although there does not appear to be an underlying dystrophy of the basement membrane associated with this disease, alterations have been detected in the CECs and the underlying stroma.16
Subsequent to corneal abrasion, typical canine CECs at the wound edges thicken, lose their hemidesmosomal attachments to the basement membrane, and rearrange their cytoskeletal elements. The cells enlarge and the epithelial sheet migrates in a monolayer to cover the corneal defect.20,21 Because these cells remain in contact with adjacent cells and with the underlying extracellular matrix, coordinated movement of the entire cell sheet is required for correct wound closure.20 The CECs in refractory ulcers do not migrate properly across the surface of the wound. Our group reported in another study8 that hallmarks of EMT (eg, loss of intercellular adhesion, increased motility, and loss of epithelial cellular morphology) during typical corneal wound healing are not found in refractory ulcers. Epithelial cells at the leading edge of refractory ulcers retain intercellular adhesion complexes, do not increase protease expression, and do not develop a migratory phenotype.8 There is also no detectable expression of the transcription factors required for CEC migration.8,a Stromal alterations have also been detected in refractory ulcers and are likely involved in the underlying pathophysiologic changes that result in this disease.16 An abnormal nerve plexus (with tangled, disorganized regions of subepithelial and epithelial hyperinnervation surrounding the leading edge of the wound) and a superficial hyalinized acellular zone have been described histologically in clinical samples of refractory ulcers and are characteristic of this disease.19,22
No established, consistently effective treatments have been described for refractory corneal ulcers in dogs,16,23 largely because the underlying mechanisms are not clearly understood. Medical treatment includes topically applied ophthalmic preparations of antimicrobials, atropine, exogenous epidermal growth factors, and glycosaminoglycans, used alone or in combination.16,23 Surgical treatment includes grid or punctate keratotomy, lamellar keratectomy, or keratectomy with a conjunctival graft.16,23,24 These medical and surgical treatments have variable success rates, and additional risks are associated with anesthetics and surgical approaches; this makes treatment (which may be ineffective) potentially expensive and frustrating for clients. Anecdotal evidence has suggested an improved rate of corneal wound healing via treatment with tetracyclines, and tetracyclines have been used to treat corneal wounds in humans25; however, no prospective studies have been published that critically evaluate the effectiveness of tetracyclines in refractory ulcers in dogs. Therefore, the objective of the study reported here was to determine whether adjunctive treatment with tetracycline analogues would decrease the time to complete wound healing of refractory corneal ulcers in dogs.
Materials and Methods
Animals—Eighty-nine adult dogs of various breeds (45 males and 44 females; mean ± SD age, 9.4 ± 2.3 years) were initially enrolled in the study. The dogs had corneal ulcers in either eye deemed refractory on the basis of ophthalmologic examination and clinical history. Exclusion criteria included presence of corneal abrasions that did not meet the inclusion criteria, a diagnosis of underlying ophthalmic disease, and known sensitivity to tetracycline or cephalexin. Dogs enrolled in the study were not treated with other medications at any time during study participation. The use of dogs in this study adhered to the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research26 and was reviewed and monitored by The Ohio State University Clinical Trials Office. Written consent was obtained from all owners prior to study-related treatment.
Study design and treatments—The multicenter trial was performed at a university veterinary teaching hospital and 2 referral hospitals. Each clinic received supplies of doxycycline tablets,b oxytetracycline ophthalmic ointment (containing oxytetracycline and polymyxin B),c cephalexin tabletsd (systemic control treatment), and an ophthalmic ointment containing neomycin, polymyxin B, and bacitracine (ie, triple antibiotic ointment; topical control treatment). Neither cephalexin nor the triple antibiotic ointment has immunomodulating properties.27,28 There is no commercially available oxytetracycline preparation for oral administration, nor is there a commercially available ophthalmic preparation of doxycycline. Rather than compounding these drugs, the investigators chose to use doxycycline for oral administration and oxytetracycline ophthalmic ointment, formulations that are available to most practicing veterinarians. Drugs used in this study were obtained from single lots and purchased from 1 veterinary supply source.f
All study drugs were provided without charge to clients. Drugs were labeled according to treatment group (A, B, or C) and were dispensed to clients by the pharmacy staff of participating clinics such that owners and ophthalmologists would be unaware of the treatment regimen each dog was assigned to receive. Despite these efforts, both clinicians and owners were easily able to discern the treatment group to which animals were assigned because of identifying characteristics of the drugs; therefore, bias could not be completely eliminated from the study results.
Refractory ulcers were diagnosed by the attending ophthalmologist using the following criteria: a superficial epithelial defect that did not have evidence of infection, was surrounded by a rim of loose epithelium, did not vascularize at a rate similar to that expected for corneal wounds, and retained fluorescein dye within the defect as well as under the rim of the loose epithelium. To confirm the diagnosis of a refractory ulcer and ensure that all ulcers were in similar stages prior to treatment, dogs underwent Schirmer tear testing, fluorescein staining, corneal debridement, and grid keratotomy before being assigned to a treatment group by 1 investigator (HLC) according to a randomization table.
The debridement and keratotomy procedures were all performed in the same manner by board-certified veterinary ophthalmologists who participated in the study. A topical anestheticg was applied to the corneal surface, and nonadherent epithelium was gently removed from the edges of the ulcer with sterile cotton swabs. The surface of the debrided ulcer was lightly scratched with a sterile 25-gauge needle. Horizontal and vertical scratches were created to form a grid pattern with the scratches extending slightly into the unaffected epithelium. To alleviate pain and prevent self-inflicted injury, ophthalmologists were permitted to treat the dogs with atropine,h a commercially available NSAID,i or 5.0% NaCl solution and to use Elizabethan collars at their discretion. These treatments are considered within the realm of standard treatment for refractory ulcers.16,29
Dogs were assigned to 1 of 3 treatment groups according to a randomization table created prior to recruitment for the study. Twenty-nine dogs received doxycycline (5 mg/kg [2.27 mg/lb], PO, q 12 h) with topically applied triple antibiotic ointment (q 8 h), 31 received cephalexin (22 mg/kg [10 mg/lb], PO, q 12 h) with topically applied oxytetracycline ophthalmic ointment (q 8 h), and 29 received a control treatment of cephalexin (22 mg/kg, PO, q 12 h) with topically applied triple antibiotic ointment (q 8 h).27 All doses were considered standard treatment doses. Healing times during the 6-week study period were measured from the time of debridement and grid keratotomy (time 0).
Assessment of response to treatment—Attending ophthalmologists used standardized forms to record signalment, clinical signs (including clinical assessment of ocular pain and discomfort), size and location of the ulcer, treatment regimen, and any evidence of adverse effects. Each dog was treated for 2 to 6 weeks with the assigned drugs; treatment progress and corneal wound healing were assessed via measurements obtained every 2 weeks by use of calipers and evaluation of photographs. Photographs were sent to The Ohio State University for analysis by use of a public domain image processing and analysis programj to determine approximate ulcer area; these data were used to support and verify measurements obtained for dogs at each clinic. Photographs were analyzed by a technician who was unaware of the treatment group assignment of each dog. Time to complete resolution of the corneal ulcer and clinical evidence of decreased pain (eg, decreased evidence of blepharospasm, redness, and tear production; reduced corneal edema) were used as indicators for the effectiveness of the drugs. To monitor treatment compliance, a form was provided to owners with instructions to record administration of the drugs daily. At the end of their participation in the study, owners were also asked to complete a questionnaire on missed treatments and adverse effects (including behavioral changes), patient comfort (as assessed by the owner), or other problems or improvements that might have been attributable to the treatments.
After the study was initiated, another treatment group was added that consisted of 29 dogs that received cephalexin (22 mg/kg, PO, q 12 h) with topically applied oxytetracycline ophthalmic ointment (q 8 h) after corneal debridement and grid keratotomy; in addition, a polyxylon soft contact lensk was applied to the corneal surface once by the attending veterinary ophthalmologist. The authors recognize that there was no appropriate control group (which would have consisted of dogs that were treated by use of orally administered cephalexin, topically applied triple antibiotic ointment, and a polyxylon soft contact lens) for this treatment group. The experimental protocol described for other treatment groups was followed for this group of dogs, and the resulting data were interesting enough to warrant inclusion and discussion in this report; however, because an appropriate control group was not established, the authors determined that it was not appropriate to include this treatment group in the statistical analysis.
Statistical analysis—All applicable data were entered into a spreadsheet, and descriptive statistics were calculated for continuous and categorical variables. Continuous variables were then converted into categorical variables to facilitate analysis. The dependent or outcome variable was ulcer completely healed (0, no; 1, yes) at the 2-week and 4-week time points after treatments were initiated. Univariate analysis was completed by use of commercially available statistical analysis software.l All independent variables were tested with the outcome variable to determine whether there was an association by calculating χ2 tests of homogeneity, ORs, and associated 95% CIs. Variables that met a critical A value ≤ 0.25 were inserted into the multivariate logistic regression model, and a backward-stepwise procedure was used to determine the final model at an A value of ≤ 0.05 similar to the method described by Hosmer and Lemeshow.30 Parameter estimates and likelihood ratio statistics were evaluated during the modeling. Odds ratios and associated 95% CIs were estimated from the final model.
Results
Photographic analyses confirmed the results of ophthalmologic examinations (Figure 1). There were no significant differences in the distribution of age, sex, or breed size categories among the treatment groups. Nonstudy drugs (ie, atropine, commercially available NSAIDs, and 5.0% NaCl solution) had a similar distribution among all treatment groups, and none of the nonstudy drugs were found to influence the time to complete wound healing. Although Elizabethan collars were provided for all dogs, the use of these was not monitored or evaluated. Owner compliance was reported to be > 90% in all treatment groups, and no adverse effects were reported by owners or detected by ophthalmologists during the study. All owners of dogs assigned to treatment groups that received orally administered cephalexin with topically applied oxytetracycline ointment (with or without a polyxylon soft contact lens) reported that their dogs appeared more comfortable than they had prior to treatment and that the dogs responded positively to the treatment. Assessment of the results of owner questionnaires indicated that more owners of dogs assigned to the treatment groups that did not receive topical oxytetracycline ointment perceived no change in their dogs' comfort or detected no obvious improvement in ulcer healing, compared with owners of dogs that received this treatment.
Photographs of a refractory corneal ulcer in a dog (A) and the same cornea 2 weeks after treatment (B). In panel A, the ulcer area is demarcated by fluorescein (green) uptake in the central region of the cornea. The photograph in panel B was obtained 2 weeks after the dog had been treated via corneal debridement and grid keratotomy and was assigned to receive cephalexin (22 mg/kg [10 mg/lb], PO, q 12 h) with topically applied oxytetracycline ophthalmic ointment (q 8 h). The cornea did not stain with fluorescein, and the refractory ulcer had resolved. The owner reported that the dog did not have signs of discomfort after 2 weeks of treatment.
Citation: Journal of the American Veterinary Medical Association 237, 4; 10.2460/javma.237.4.378
Univariate analysis—Dogs were categorized for analysis according to breed size (ie, large, medium, or small) on the basis of a previously published report.31 Small-breed dogs included Boston Terriers (n = 5), Dachshund (1), French Bulldog (1), Jack Russell Terriers (3), Miniature Pinschers (3), Miniature Schnauzer (1), Shih Tzu (1), Yorkshire Terrier (2), and mixed-breed dogs (6). Medium-breed dogs included American Bulldogs (n = 2), American Staffordshire Terrier (1), Boxers (23), Elkhound (1), Welsh Corgis (6), and mixed-breed dogs (8). Large-breed dogs included Bearded Collie (n = 1), German Shorthaired Pointer (1), Golden Retrievers (3), Siberian Husky (1), Labrador Retrievers (5), Newfoundland (1), Rottweilers (2), Samoyeds (2), and mixed-breed dogs (9). Corneal ulcers of small-breed dogs healed significantly (P < 0.001) faster than those of large-breed dogs (Figure 2). The proportion of Boxers enrolled in the study was significantly (P < 0.001) greater than that of any other breed. No significant differences attributable to age or sex were detected in healing times among any of the treatment groups.
Healing times for refractory corneal ulcers in 89 dogs categorized on the basis of breed size.31 After corneal debridement and grid keratotomy, dogs (small breed [white bars], n = 23; medium breed [gray bars], 41; and large breed [black bars], 25) were assigned to receive 1 of 3 treatment regimens for up to 6 weeks: doxycycline (5 mg/kg [2.27 mg/lb], PO, q 12 h) with topically applied triple antibiotic ointment (q 8 h), cephalexin (22 mg/kg, PO, q 12 h) with topically applied oxytetracycline ophthalmic ointment (q 8 h), or a control treatment of cephalexin (22 mg/kg, PO, q 12 h) with topically applied triple antibiotic ointment (q 8 h). Corneal ulcers in small-breed dogs were significantly (P < 0.001) more likely to heal within 2 weeks after treatment initiation, regardless of treatment, than were ulcers of medium- or large-breed dogs. Healing times > 6 weeks indicated corneal ulcers that had not resolved by the end of the study period.
Citation: Journal of the American Veterinary Medical Association 237, 4; 10.2460/javma.237.4.378
Interestingly, the size (area) of the ulcer prior to treatment did not significantly alter the corneal healing time in any of the treatment groups, such that large ulcers were as likely to heal within each measured time period as were small ulcers. In addition, ulcer location within the cornea did not significantly influence the healing time in any of the treatment groups.
Mean healing time for corneal ulcers in dogs that received orally administered doxycycline with topically applied triple antibiotic ointment was longer than that in dogs that received orally administered cephalexin with topically applied oxytetracycline ointment and shorter than that in dogs that received the control treatment; however, these differences were not significant (Figure 3). Of 29 dogs that received orally administered doxycycline with topically applied triple antibiotic ointment, corneal ulcers of 14 (48.3%) healed within 2 weeks and corneal ulcers of another 14 (48.3%) healed within 4 weeks. The corneal ulcer of 1 dog required ≥ 6 weeks to heal.
Corneal wound healing at 2 (white bars), 4 (black bars), or > 6 (gray bars) weeks after debridement and grid keratotomy in 89 dogs randomly assigned to receive 1 of 3 treatment regimens for up to 6 weeks (A) and in 29 dogs assigned to receive cephalexin PO with topically applied oxytetracycline ophthalmic ointment and a polyxylon soft contact lens (B). There was a significant (P < 0.001) increase in corneal wound healing at 2 weeks for dogs that received oxytetracycline ophthalmic ointment, compared with that for dogs that received the control treatment. Dogs treated with oxytetracycline ophthalmic ointment plus a soft contact lens (B) were not included in statistical analysis because no appropriate control group was assigned; however, this treatment outcome was remarkably successful, and dogs in this group appeared to have the most rapid corneal wound healing times. Values reported represent the percentage of dogs with complete resolution of refractory corneal ulcers at 2 or 4 weeks and percentage of dogs with corneal ulcers that had not resolved by the end of the 6-week study period.
Citation: Journal of the American Veterinary Medical Association 237, 4; 10.2460/javma.237.4.378
A significant (P < 0.001) decrease in healing time was found in dogs assigned to receive orally administered cephalexin with topically applied oxytetracycline ointment, compared with healing time in dogs that received the control treatment (Figure 3). Of 31 dogs in the former group, corneal ulcers in 23 (74.2%) healed within 2 weeks and corneal ulcers in 7 (22.6%) healed within 4 weeks; of 29 dogs in the control treatment group, corneal ulcers in 3 (10.3%) healed within 2 weeks and corneal ulcers in 9 (31.0%) healed within 4 weeks. Only 1 of 31 dogs in the group that received orally administered cephalexin with topically applied oxytetracycline ointment had a corneal ulcer that did not heal within the 6-week treatment period; 17 of 29 (58.6%) dogs that received the control treatment had corneal ulcers that did not heal within 6 weeks.
Multivariate analysis—All variables that resulted in a value of A ≤ 0.25 for univariate analysis were considered for multivariate analysis. These variables were used in a backward-stepwise multivariate logistic regression model, and results of the final analysis were summarized (Table 1). The 2 variables that remained in the model at a critical value of A ≤ 0.05 were breed size and treatment group. Corneal ulcers in small-breed dogs were > 7 times as likely (OR, 7.3; 95% CI, 1.3 to 50.6) to heal within 2 weeks as were those of medium- or large-breed dogs. When analyses were controlled for breed size, corneal ulcers in dogs that received orally administered cephalexin with topically applied oxytetracycline ointment were 0.15 times as likely (OR, 0.15; 95% CI, 0.03 to 0.63) to heal within 2 weeks as were corneal ulcers in dogs that received orally administered doxycycline with topically applied triple antibiotic ointment, and they were 50 times as likely (OR, 0.02; 95% CI, 0.003 to 0.135) to heal as were corneal ulcers in dogs in the control treatment group during the same time period.
Results of multivariate analysis of the likelihood of complete corneal wound healing within 2 weeks in 89 dogs with refractory corneal ulcers.
| Variable | OR | 95% CI | χ2 | P value |
|---|---|---|---|---|
| Small breed* | 7.3 | 1.3–50.6 | 6.78 | 0.030 |
| Control group treatment | 0.02 | 0.003–0.135 | 14.64 | < 0.001 |
| Doxycycline PO treatment | 0.15 | 0.03–0.63 | 6.32 | 0.0134 |
Dogs were assigned to receive 1 of 3 treatment regimens until resolution of the refractory ulcer or for up to 6 weeks if ulcers remained unresolved: doxycycline (5 mg/kg [2.27 mg/lb], PO, q 12 h) with topically applied triple antibiotic ointment (q 8 h), cephalexin (22 mg/kg [10 mg/lb], PO, q 12 h) with topically applied oxytetracycline ophthalmic ointment (q 8 h), or a control treatment of cephalexin (22 mg/kg, PO, q 12 h) with topically applied triple antibiotic ointment (q 8 h).
Dogs were classified as belonging to small, medium, or large breeds on the basis of a published report.31
Outcome for dogs not included in statistical analysis—Dogs were categorized according to breed size (ie, large, medium, or small) on the basis of a previously published report.31 Small-breed dogs included Boston Terrier (n = 1), Jack Russell Terriers (2), Lhasa Apso (1), Shih Tzu (1), Silky Terrier (1), Toy Fox Terrier (1), and mixed-breed dogs (3). Medium-breed dogs included Boxers (n = 6), English Springer Spaniel (1), Keeshond (1), Welsh Corgi (1), and mixed-breed dogs (3). Large-breed dogs included Doberman Pinscher (n = 1), German Shorthaired Pointer (1), Golden Retrievers (2), Labrador Retriever (1), and mixed-breed dogs (2). Within 2 weeks, corneal ulcers had completely resolved in 25 of 29 (86.2%) dogs that received orally administered cephalexin with topically applied oxytetracycline ophthalmic ointment and a polyxylon soft contact lens. Among dogs in this group, 3 (10.3%) dogs required 2 to 4 weeks for complete corneal wound healing. Similar to results for dogs that received orally administered cephalexin and topically applied oxytetracycline ointment without a contact lens, only 1 dog had a corneal ulcer that did not resolve within the 6-week treatment period. Retention times for contact lenses were not recorded.
Discussion
To the authors' knowledge, no established, consistently effective medical treatments for refractory corneal ulcers in dogs have been described.16 Superficial keratectomy is reported24 to have a high success rate and is commonly performed by many veterinary ophthalmologists. However, risks associated with this surgical treatment include the need for general anesthesia, increased likelihood for corneal scarring, and increased costs. In 1 study,32 investigators reported that in 8 dogs, 100% of corneal ulcers resolved within 2 to 3 weeks after surgical procedures (ie, thermal cautery) that did not require general anesthesia. It should be mentioned that the authors of that study cautioned that this technique should be reserved for difficult refractory ulcers that have not responded to other treatments because thermal cautery procedures can be associated with increased complications and corneal scarring.32 We found no clinical reports in which topically or systemically administered tetracycline was investigated as a treatment for refractory corneal ulcers in dogs (alone or as an adjunct to surgery). Grid keratotomy is among the most commonly performed surgical procedures used by veterinary ophthalmologists to treat refractory corneal ulcers in dogs and was therefore the surgical method selected for use in the study reported here. The objective of the present study was to evaluate the effectiveness of topically and orally administered tetracycline-related drugs as adjunctive medical treatment for refractory corneal ulcers in dogs in a randomized and controlled clinical trial.
Analysis of the results of the present study indicated that corneal ulcers in small-breed dogs were significantly (P < 0.001) more likely to heal within 2 weeks than were those in medium-breed and large-breed dogs. A possible explanation for this is that larger dogs may spend more time outdoors. Another study33 revealed that exposure to UV radiation can delay cutaneous wound healing; the same might be true for corneal wound healing. Furthermore, there may be breed-related physiologic characteristics that result in more rapid or delayed healing of corneal ulcers. For instance, brachycephalic (predominantly small-breed) dogs blink less frequently than do mesaticephalic or dolichocephalic dogs.34 Of the small-breed dogs in the present study, at least 32 of 89 (35.9%) could be classified as brachycephalic; because several of the small-breed dogs were also mixed-breed dogs, the exact percentage may have been higher. Eyelid excursions over the cornea may disrupt migrating CECs and subsequently delay healing.35 In other studies,16,36 it was reported that Boxers are predisposed to refractory corneal ulcers, and our results support these conclusions. Significantly (P < 0.001) more Boxers (23/89 [25.8%]) were enrolled in the present study than were dogs of any other breed. The reason that Boxers are at a greater risk for developing refractory corneal ulcers has not been elucidated.
Topical tetracycline analogues are reported25,37,38 to be nontoxic and have a low likelihood of negatively affecting the health of patients; these drugs are also relatively inexpensive. Adverse effects, particularly gastrointestinal tract disturbances, have been reported27 in dogs receiving systemic treatment with doxycycline. Although these adverse effects are generally mild, they can infrequently become more severe and potentially become fatal.27
As stated previously, tetracycline has been used in humans to promote corneal wound healing, without reports of adverse effects.25 Results of the study reported here provide in vivo evidence supporting the hypothesis that adjunctive tetracycline treatment can improve the rate of corneal reepithelialization in dogs after grid keratotomy. Dogs with refractory corneal ulcers that were treated with topically administered oxytetracycline ophthalmic ointment had a significant decrease in the time to complete wound closure, compared with the time to complete wound closure for the control treatment group. Although doxycycline administered orally resulted in more rapid corneal wound healing, compared with that for the control treatment of cephalexin administered orally, values did not differ significantly. Doxycycline was not available as an ophthalmic preparation, and oxytetracycline was not available for oral administration; an objective of this study was to use commercially available drugs that would be easily accessible to veterinarians in clinical practice. The authors recognize this as a limitation of the study because we could not determine whether the differences observed between these treatment groups were attributable to the analogue of tetracycline (doxycycline vs oxytetracycline) or the method of treatment (tablet administered orally vs topically applied ophthalmic ointment).
The control drug treatments were chosen because they were readily available to veterinary clinicians, and the authors believed they were the best option available to minimize any confounding factors as well as the number of treatment groups needed. Although the most appropriate control treatments would have been sham treatments consisting of the inert vehicles used for administration of the tetracycline medications, this was not an ethical option. Topically applied antimicrobials are indicated for the prevention of infection during treatment of all corneal ulcers29; therefore, if the control treatment for topically applied medication had been the ointment vehicle, the ophthalmologists would also have had to prescribe an additional topical antimicrobial. This would have further confounded the data and increased the number of treatment groups needed. The authors recognize that because topically applied triple antibiotic ointment was chosen as a control treatment, it cannot be definitively stated that oxytetracycline improved the time to complete healing for corneal wounds in dogs in the present study; an alternative possibility is that triple antibiotic ointment delayed the healing process.
The 29 dogs treated by use of orally administered cephalexin with topically applied oxytetracycline ophthalmic ointment and a polyxylon soft contact lens were not included in the statistical analysis because of the lack of an appropriate control treatment group. However, use of this treatment regimen was particularly successful, and it appeared that use of a soft contact lens as a bandage, in conjunction with oxytetracycline ophthalmic ointment, resulted in the most rapid wound healing of refractory corneal ulcers. We could not ascertain whether this result was attributable to pressure applied to the corneal epithelium from the contact lens, protection of migrating CECs against traction elicited by excursions of the third eyelid and eyelid, or improved drug retention (ie, prolonged exposure of CECs to oxytetracycline because of the contact lens). It is the authors' opinion that all of these factors likely influenced reepithelialization when a soft contact lens was used.
Reepithelialization is initiated by endogenous signals, including soluble growth factors such as TGF-B family members. Activation of TGF-B signaling pathways is sufficient to induce EMT-like changes in CECs.14,39,40 Inactive forms of TGF-B are secreted by most cells41 and become activated in the extracellular environment, then they bind to cell surface receptors to activate a myriad of intracellular signaling cascades.42,43 Many mechanisms initiated during typical corneal wound healing are driven by growth factors secreted by CECs and stromal cells; CECs also possess appropriate receptors, including TGF-B receptors, for these factors.15 Growth factors such as members of the TGF-B family appear to play key roles in corneal wound healing, and analysis of the results of some studies44–46 suggests that exogenous growth factors can also enhance this repair process.
In a small study,m investigators found decreased concentrations of TGF-B in tears of dogs with refractive corneal ulcers. Another study8 conducted by our laboratory group revealed that the tetracycline analogue oxytetracycline upregulates TGF-B in CECs in vitro, which results in increased expression of specific transcription factors. This promotes migration of CECs, which complete corneal wound healing. Tetracyclines have protease-inhibiting properties; however, the ability of oxytetracycline to promote migration of CECs in vitro is independent of this function.8 In another report,47 investigators concluded that MMP-2 and MMP-9 are not responsible for the pathophysiology of refractory ulcers in dogs. Those results suggest that inhibition of MMPs is unrelated to the improved rate of corneal wound healing caused by treatment with tetracycline analogues. The authors believe that, although not directly tested in the study reported here, the combined results of our group's in vitro8 and in vivo studies support the hypothesis that tetracycline promotes corneal reepithelialization through upregulation of specific growth factors, such as TGF-B family members and transcription factors.
Reports48,49 published in the 1960s indicated that ophthalmic ointments delay corneal wound healing. Although these statements were not definitively supported by the results of scientific studies, this belief has become dogma within many ophthalmology practices. It is clear that results of early studies50–52 on the effects of nonemulsive ointments on corneal wound healing were inconsistent. Furthermore, the formulations of commercially available ophthalmic ointments sold at the present time differ considerably from those investigated in the aforementioned early studies; authors of a more recent study53 that examined the effects of ophthalmic ointments on corneal wound healing in rats, rabbits, and monkeys concluded that there was no evidence to indicate that ophthalmic ointments, alone or in combination with the ophthalmic drugs tested, interfered with wound healing.
The authors of several studies18,23,24 examined the impact of debridement and grid keratotomy on refractory ulcers in dogs. The interval required for complete wound healing after the surgical procedures were performed in those experiments ranged from 13 to 25 days. It is difficult to explain why the healing times in the present study were longer in all groups, compared with those in previous reports. Dogs in the control treatment group had a markedly increased time to complete wound closure, compared with that reported in the literature, and this makes it difficult to fully evaluate results for tetracycline analogues as adjunctive treatments in the study reported here. Although information in the published literature indicates that it is unlikely, there is a possibility that the ointment used as a carrier in the ophthalmic formulations for all groups slowed corneal wound healing. In addition, because both clinicians and owners could easily discern the treatment group to which dogs were assigned, bias could not be completely removed from the present study.
Potential detrimental effects of various antimicrobials on corneal reepithelialization have been examined in several studies.54–57 A group examined toxic effects and reepithelialization of canine CECs in vitro following treatment with various concentrations of a combination of neomycin, polymyxin B, and gramicidin.54 It was concluded that these antimicrobials were in the midrange for effects on corneal wound healing and toxic effects, compared with results for chloramphenicol, tobramycin, gentamicin sulfate, cefazolin, and ciproflozacin. It should be mentioned that CECs were continually exposed to these drugs for 24 to 96 hours,54 which is a longer time period than CECs would be exposed to the drugs within the tear film in vivo. Another study55 in which investigators tested the effects of 8 topically applied antimicrobial solutions on the rate of CEC-mediated wound healing in rabbits revealed that the combination of neomycin, polymyxin B, and gramicidin had the least detrimental effect among the drugs evaluated.
The triple antibiotic ointment used in the present study contained bacitracin instead of gramicidin. A study56 in which investigators evaluated the in vivo effects of neomycin and bacitracin on CEC-mediated wound healing in rabbits revealed that the reepithelialization process was not delayed by use of those drugs at concentrations similar to the concentrations used in the study reported here. Topical ophthalmic application of triple antibiotic solution (ie, neomycin, polymyxin B, and bacitracin) is a common adjunctive treatment for refractory corneal ulcers in dogs, with no reported adverse effects.29 Although topical ointments remain in contact with the corneal epithelium for longer periods than do the equivalent solutions, if the triple antibiotic ointment used in the present study had substantial toxic effects, the literature would likely have revealed toxic effects for triple antibiotic solution.
The observed differences in the time to complete wound healing between tetracycline analogues administered topically or orally is likely attributable to drug availability at the corneal epithelium; the topical preparation is applied directly to the corneal epithelium, whereas the orally administered drug likely reaches the epithelium in lower concentrations via the tear film. Alternatively, corneal ulcers in dogs in the doxycycline treatment group may have had longer healing times because of a detrimental effect of the topically applied triple antibiotic ointment or because different tetracycline analogues exert different effects on the cornea. Although the drugs used in the triple antibiotic ointment have the potential to delay corneal wound healing or to cause toxic effects,54 it has yet to be determined whether the use of this medication impairs corneal wound healing in dogs.
In the study reported here, dogs that received orally administered doxycycline with topically applied triple antibiotic had an intermediate healing time, compared with healing times of dogs in the other treatment groups. If the triple antibiotic ointment interfered with corneal wound healing, similar healing times could be expected among all dogs that received this topical treatment. On the basis of our data and data published in another report,8 we believe that the adjunctive treatments with tetracycline analogues were responsible for improved corneal wound healing times. However, there is no definitive evidence to indicate that the triple antibiotic ointment did not impede cell migration. Further studies are needed to determine the effects of this medication on in vivo corneal wound healing.
The impact of the findings in the study reported here involves improved healing times and increased overall comfort, as assessed by both clients and clinicians, in dogs affected with refractory corneal ulcers. These experiments provide clinical evidence of the effectiveness of tetracycline analogues as adjunctive treatments after corneal debridement and grid keratotomy of refractory corneal ulcers in dogs. Topical application of oxytetracycline ophthalmic ointment or systemic (oral) administration of doxycycline appears to provide a safe, reliable, and cost-effective adjunctive treatment for refractory corneal ulcers.
Future studies to evaluate the necessity of a grid keratotomy for refractory corneal ulcers prior to medical management by use of tetracycline analogues may be of value. The information reported in the present study suggests that other studies to evaluate the effects of soft contact lenses on corneal wound healing in combination with oxytetracycline or a triple antibiotic ointment would be of substantial value.
ABBREVIATIONS
| CEC | Corneal epithelial cell |
| CI | Confidence interval |
| EMT | Epithelial-mesenchymal transition |
| MMP | Matrix metalloproteinase |
| OR | Odds ratio |
| TGF | Transforming growth factor |
Chandler HL, Kusewitt DF, Colitz CMH. The slug transcription factor modulates corneal epithelial cell migration (abstr). Invest Ophthalmol Vis Sci 2004;45:3759.
Doxycycline, Watson Pharmaceuticals Inc, Corona, Calif.
Terramycin, Pfizer Inc, New York, NY.
Cephalexin, Karalex Pharma, Woodcliff Lake, NJ.
Triple antibiotic ophthalmic ointment, Fougera, Melville, NY.
MWI Veterinary Supply, Meridian, Idaho.
Proparacaine, Akorn Inc, Lake Forest, Ill.
Atropine, Bausch & Lomb, Tampa, Fla.
Rimadyl, Pfizer Inc, New York, NY.
NIH Image, Research Service Branch, US National Institutes of Health, Bethesda, Md. Available at: rsb.info.nih.gov/nih-image/. Accessed Jun 8, 2008.
Acrivet Inc, Salt Lake City, Utah.
PROC GENMOD, SAS, version 9.1.3, SAS Institute Inc, Cary, NC.
Jurk I, Gilger B, Malok E. TGF-beta levels in tears of normal dogs and dogs with refractory corneal ulcers (abstr), in Proceedings. Am Coll Vet Ophthalmol 2000;66.
References
- 1.
Birchmeier W, Birchmeier C. Epithelial-mesenchymal transitions in development and tumor progression. EXS 1995;74:1–15.
- 2.
Boyer B, Roche S, Denoyelle M, et al. Src and Ras are involved in separate pathways in epithelial cell scattering. EMBO J 1997;16:5904–5913.
- 3.
Hemavathy K, Guru SC, Harris J, et al. Human slug is a repressor that localizes to sites of active transcription. Mol Cell Biol 2000;20:5087–5095.
- 4.
Hemavathy K, Ashraf SI, Ip YT. Snail/slug family of repressors: slowly going into the fast lane of development and cancer. Gene 2000;257:1–12.
- 5.
Savagner P. Leaving the neighborhood: molecular mechanisms involved during epithelial-mesenchymal transition. Bioessays 2001;23:912–923.
- 6.
Caulin C, Scholl FG, Frontelo P, et al. Chronic exposure of cultured transformed mouse epidermal cell to transforming growth factor-beta 1 induces an epithelial-mesenchymal transdifferentiation and a spindle tumoral phenotype. Cell Growth Differ 1995;6:1027–1035.
- 7.
Bement WM, Forscher P, Mooseker MS. A novel cytoskeletal structure involved in purse string wound closure and cell polarity maintenance. J Cell Biol 1993;121:565–578.
- 8.↑
Chandler HL, Colitz CMH, Lu P, et al. The role of the slug transcription factor in cell migration during corneal re-epithelialization in the dog. Exp Eye Res 2007;84:400–411.
- 9.
Savagner P, Kusewitt DF, Carver EA, et al. Developmental transcription factor slug is required for effective re-epithelialization by adult keratinocytes. J Cell Physiol 2005;202:858–866.
- 10.
Nakamura K, Kurosaka D, Yoshino M, et al. Injured corneal epithelial cells promote myodifferentiation of corneal fibroblasts. Invest Ophthalmol Vis Sci 2002;43:2603–2608.
- 11.
SundarRaj N, Rizzo JD, Anderson SC, et al. Expression of vimentin by rabbit corneal epithelial cells during wound repair. Cell Tissue Res 1992;267:347–356.
- 12.
Yu FX, Gipson IK, Guo Y. Differential gene expression in healing rat corneal epithelium. Invest Ophthalmol Vis Sci 1995;36:1997–2007.
- 13.
Suzuki K, Tanaka T, Enoki M, et al. Coordinated reassembly of the basement membrane and junctional proteins during corneal epithelial wound healing. Invest Ophthalmol Vis Sci 2000;41:2495–2500.
- 14.
Han G, Lu SL, Li AG, et al. Distinct mechanisms of TGF-beta1-mediated epithelial-to-mesenchymal transition and metastasis during skin carcinogenesis. J Clin Invest 2005;115:1714–1723.
- 15.↑
Imanishi J, Kamiyama K, Iguchi I, et al. Growth factors: importance in wound healing and maintenance of transparency of the cornea. Prog Retin Eye Res 2000;19:113–129.
- 16.↑
Gilger BC, Bentley E, Ollivier FJ. Diseases and surgery of the canine cornea and sclera. In: Gelatt KN, ed. Veterinary ophthalmology. 4th ed. Ames, Iowa: Blackwell Publishing, 2007;701–705.
- 17.
Kirschner SE, Niyo Y, Betts DM. Idiopathic persistent corneal erosions: clinical and pathological findings in 18 dogs. J Am Anim Hosp Assoc 1989;25:84–90.
- 18.
Champagne E, Munger RJ. Multiple punctate keratotomy for the treatment of recurrent epithelial erosions in dogs. J Am Anim Hosp Assoc 1992;28:213–216.
- 19.
Bentley E, Abrams GA, Covitz D, et al. Morphology and immuno-histochemistry of spontaneous chronic corneal epithelial defects (SCCED) in dogs. Invest Ophthalmol Vis Sci 2001;42:2262–2269.
- 20.↑
Schultz GS. Modulation of corneal wound healing. In: Krachmer JH, Mannis MJ, Holland EJ, eds. Cornea. St Louis: Mosby, 1997;183.
- 21.
Apple DJ. Cornea. In: Apple DJ, Rabb MF, eds. Ocular pathology. 5th ed. St Louis: Mosby, 1998;59–116.
- 22.
Murphy CJ, Marfurt CF, McDermott A, et al. Spontaneous chronic corneal epithelial defects (SCCED) in dogs: clinical features, innervation, and effect of topical SP, with or without IGF-1. Invest Ophthalmol Vis Sci 2001;9:277–285.
- 23.
Morgan RV, Abrams KL. A comparison of six different therapies for persistent corneal erosions in dogs and cats. Vet Comp Ophthalmol 1994;4:38–43.
- 24.↑
Stanley R, Hardman C, Johnson B. Results of grid keratotomy, superficial keratectomy, and debridement for the management of persistent corneal erosions in 92 dogs. Vet Ophthalmol 1998;1:233–238.
- 25.↑
Perry HD, Kenyon KR, Lamberts DW, et al. Systemic tetracycline hydrochloride as adjunctive therapy in the treatment of persistent epithelial defects. Ophthalmology 1986;93:1320–1322.
- 26.↑
Association for Research in Vision and Ophthalmology. Statement for the use of animals in ophthalmic and visual research. Available at: www.arvo.org/eweb/dynamicpage.aspx?site=arvo2&webcode=AnimalsResearch. Accessed Jul 5, 2007.
- 28.
Van Vlem B, Vanholder R, De Paepe P, et al. Immunomodulating effects of antibiotics: literature review. Infection 1996;24:275–291.
- 29.↑
Maggs DJ. Cornea and sclera. In: Maggs DJ, Miller PE, Ofri R, eds. Slatter's fundamentals of veterinary ophthalmology. 4th ed. St Louis: Saunders Elsevier, 2008;175–202.
- 30.↑
Hosmer DW, Lemeshow S. Applied logistic regression. 2nd ed. New York: Wiley-Interscience Publication, 2000;31–43.
- 31.↑
Bras ID, Colitz CMH, Saville WJA, et al. Posterior capsular opacification in diabetic and nondiabetic canine patients following cataract surgery. Vet Ophthalmol 2006;9:317–327.
- 32.↑
Bentley E, Murphy CJ. Thermal cautery of the cornea for treatment of spontaneous chronic corneal epithelial defects in dogs and horses. J Am Vet Med Assoc 2004;224:250–253.
- 33.↑
Brockis JG, Elliott M, Lissiman M. The rate of healing of wounds treated with ultra-violet light. Aust N Z J Surg 1965;35:108–114.
- 34.↑
Gelatt KN, Gelatt JP. Chapter 8. In: Small animal ophthalmic surgery. Boston: Butterworth-Heinemann, 2001;191.
- 35.↑
Cosar CB, Cohen EJ, Rapuano CJ, et al. Tarsorrhaphy: clinical experience from a cornea practice. Cornea 2001;20:787–791.
- 36.
Gelatt K, Samuelson D. Recurrent corneal erosions and epithelial dystrophy in the Boxer dog. J Am Anim Hosp Assoc 1982;19:453–460.
- 37.
Seedor JA, Perry HD, McNamara TF, et al. Systemic tetracycline treatment of alkali-induced corneal ulceration in rabbits. Arch Ophthalmol 1987;105:268–271.
- 38.
Shlopov BV, Stuart JM, Gumanovskaya ML, et al. Regulation of cartilage collagenase by doxycycline. J Rheumatol 2001;28:835–842.
- 39.
Nawshad A, Lagamba D, Polad A, et al. Transforming growth factor-beta signaling during epithelial-mesenchymal transformation: implications for embryogenesis and tumor metastasis. Cells Tissues Organs 2005;179:11–23.
- 40.
de Iongh RU, Wederell E, Lovicu FJ, et al. Transforming growth factor-beta-induced epithelial-mesenchymal transition in the lens: a model for cataract formation. Cells Tissues Organs 2005;179:43–55.
- 41.↑
Lawrence DA. Transforming growth factor-beta: a general review. Eur Cytokine Netw 1996;7:363–374.
- 42.
Kolodziejczyk SMH, Hall BK. Signal transduction and TGF-beta superfamily receptors. Biochem Cell Biol 1996;74:299–314.
- 43.
Xu J, Lamouille S, Derynck R. TGF-beta-induced epithelial to mesenchymal transition. Cell Res 2009;19:156–172.
- 44.
Klenkler B, Sheardown H. Growth factors in the anterior segment: role in tissue maintenance, wound healing and ocular pathology. Exp Eye Res 2004;79:677–688.
- 45.
Schultz G, Khaw PT, Oxford K, et al. Growth factors and ocular wound healing. Eye 1994;8:184–187.
- 46.
Schultz GS, Strelow S, Stern GA, et al. Treatment of alkali-injured rabbit corneas with a synthetic inhibitor of matrix metalloproteinases. Invest Ophthalmol Vis Sci 1992;33:3325–3331.
- 47.↑
Carter RT, Kambampati R, Murphy CJ, et al. Expression of matrix metalloproteinase 2 and 9 in experimentally wounded canine corneas and spontaneous chronic corneal epithelial defects. Cornea 2007;26:1213–1219.
- 48.
Havener WH. Chapters 1 and 6. In: Ocular pharmacology. 2nd ed. St Louis: Mosby, 1966;9,244.
- 49.
Newell FW. Chapters 3 and 11. In: Ophthalmology: principles and concepts. 2nd ed. St Louis: Mosby, 1969;116.
- 50.
Leopold IH, Steele WH. Influence of local application of sulfonamide compounds and their vehicles on regeneration of corneal epithelium. Arch Ophthal 1945;33:463–467.
- 51.
Friedenwald JS, Buschke W. The influence of some experimental variables on the epithelial movements in the healing of corneal wounds. J Cell Comp Physiol 1944;23:95–107.
- 52.
Heerema JC, Friedenwald JS. Retardation of wound healing in the corneal epithelium by lanolin. Am J Ophthalmol 1950;33:1421–1427.
- 53.↑
Hanna C, Fraufelder FT, Cable M, et al. The effects of ophthalmic ointments on corneal wound healing. Am J Ophthalmol 1973;76:193–200.
- 54.↑
Hendrix DV, Ward DA, Barnhill MA. Effects of antibiotics on morphologic characteristics and migration of canine corneal epithelial cells in tissue culture. Am J Vet Res 2001;62:1664–1669.
- 55.↑
Stern GA, Schemmer GB, Farber RD, et al. Effect of topical antibiotic solutions on corneal epithelial wound healing. Arch Ophthalmol 1983;101:644–647.
- 56.↑
Petroutsos G, Guimaraes R, Giraud J, et al. Antibiotics and corneal epithelial wound healing. Arch Ophthalmol 1983;101:1775–1778.
- 57.
Nelson JD, Silverman V, Lima PH, et al. Corneal epithelial wound healing: a tissue culture assay on the effect of antibiotics. Curr Eye Res 1990;9:277–285.
