Imagine this clinical scenario...
You are presented with a 5-year-old male Boxer dog with corneal oedema and blepharospasm. On ocular examination, you identify a spontaneous chronic corneal epithelial defect (SCCED), which you treat under local anaesthetic using a cotton bud to remove loose epithelial edges followed by a diamond burr debridement (DBD). You carry out the DBD after considering other interventional therapy. With the possibility that this may take weeks or months to heal, you consider using a bandage contact lens (BCL) as you have previously done so with other patients to help the healing process. You decide to consult the evidence to fact-check your assumptions about the use of BCLs.
Three studies were identified that are relevant to the research question. Existing studies published prior to 2015 were excluded as they did not make use of control or non-intervention groups. Wooff & Norman (2015) is a randomised prospective study, Dees et al. (2017) is a non-randomised prospective study, and Rivera-Viscal et al. (2024) is a case control study. All three papers studied the same outcome: non-retention of fluorescein, assessed objectively via an ophthalmic fluorescein examination. Wooff & Norman (2015) also carried out owner surveys to assess comfort level and contact lens retention.
Dees et al. (2017) studied 237 canine patients that had undergone DBD for treatment of SCCEDs in one eye. The dogs were split into 12 groups depending on which postoperative antibiotic was used, with 6/12 groups (129/237 patients) fitted with a BCL after surgery.
Rivera-Viscal et al. (2024) studied 68 client-owned dogs with at least a 7-day history of a fluorescein-retaining SCCED with non-adherent epithelium. All patients had their non-adherent epithelium removed with a cotton-tipped applicator followed by an anterior stromal puncture procedure. 33/68 dogs were fitted with a BCL and 35/68 were not. All dogs received topical antibiotic therapy, systemic pain management, and an Elizabethan collar.
Wooff & Norman (2015) studied 27 Boxer dogs presented to a referral ophthalmology centre and diagnosed with a SCCED in one eye. This was the only paper to reference Boxer dogs directly. A linear grid keratotomy was performed on all patients under sedation after the loose epithelial edges were removed using a scalpel or spatula. 14/27 dogs were randomly assigned to be fitted with a BCL.
The overall strength of evidence provided by the three studies is strong owing to their robust study design. However, there were some limitations to each of them.
Dees et al. (2017) had limitations around reporting of post-intervention results; checks by owners were approximate, time to healing was measured in bounds of 7 days rather than being exact, and patients were not examined daily. Additionally, the patients’ allocation into groups was not randomised and the owners and clinicians were not blinded, leading to the possibility of bias. Additionally, lens retention rate was 62%, which may limit the applicability of the BCL group. Some patients also had comorbidities which may have had an impact on clinical outcomes.
Rivera-Viscal et al. (2024) failed to record whether or not the BCL was present in each patient at the time of first recheck, when it is possible that some may not have been retained at this point. Additionally, not all dogs were placed on the same antibiotic regimen, and each procedure was not performed by the same clinician.
Wooff & Norman (2015) was the only study that was limited by a small sample size, with 27 dogs studied. There was also inconsistent follow-up with some dogs due to scheduling conflicts. Additionally, not all antimicrobial therapy was the same and some patients were on NSAIDs.
All three studies demonstrated a shorter healing time for SCCEDs when treated using a BCL.
Dees et al. (2017) found that use of a BCL that was retained significantly improved healing times for every group regardless of the other experimental protocol used. Patients were sorted into 12 groups with a variety of antibiotics, and every BCL group outperformed its non-BCL counterpart. The median healing in patients treated with use of a BCL was approximately 3 days shorter than patients treated without one.
Rivera-Viscal et al. (2024) found that BCL placement was associated with significantly greater odds of corneal healing by the first recheck timepoint, carried out at median 14 days.
Wooff & Norman (2015) found that dogs treated with the use of BCLs had a significant decrease in median corneal healing time (7 days) when compared to those without (10 days).
There is strong evidence to suggest that the application of a bandage contact lens post-intervention in the case of canine spontaneous chronic corneal epithelial defect shortens corneal healing time. Therefore, bandage contact lenses can be considered an effective method of speeding up the healing process in dogs that have undergone treatment for spontaneous chronic corneal epithelial defect. However it’s worth noting that Dees et al. (2017) reported a 62% retention rate for BCLs among the dogs they studied, so this method may not work in all cases.
Disclaimer
The application of evidence into practice should take into account multiple factors, not limited to: individual clinical expertise; patient’s circumstances and owner’s values; country, location or clinic where you work; the individual case in front of you; and the availability of therapies and resources.