VIRAL KERATITIS

Emerging information indicates that ophthalmologists should now be considering cytomegalovirus (CMV) infection in their differential diagnosis for patients with endotheliitis or corneal transplant rejection, according to John Dart DM, FRCOphth, speaking at the 4th EuCornea Congress in Amsterdam. The importance of identifying epidemic keratoconjunctivitis (EKC), the most serious form of adenoviral ocular infection is well-known. Prof Dart, consultant ophthalmologist, Moorfields Eye Hospital, London, UK, provided an update on the management of EKC and discussed CMV as a recently described cause of endothelial decompensation and graft failure.

Strategies for preventing cross-infection are an essential component for managing EKC since the virus is so highly contagious and there is the potential for long-term debilitating and sight-threatening sequelae. In order to address the problem of nosocomial adenoviral spread at Moorfields Eye Hospital, outpatients with conjunctivitis are segregated in a separate waiting area and examination room and receive expedited care to minimise the time they may be exposing others to infection by their presence in the department. In addition, because shedding of adenovirus continues for up to two weeks in 10 per cent of patients, the current infection control policy requires infected medical staff to stay away from work for 14 days, and until the eyes are free of inflammation.

“With these two measures for isolating patients and employees, we have substantially cut down on the development of new cases of nosocomial adenovirus keratoconjunctivitis at our institution,” Prof Dart said. PCR is used for diagnosis of adenoviral ocular infection at Moorfields, and it is still considered the optimal technique. However, Prof Dart noted the now internationally available office-based immunoassay (AdenoPlus, RPS Sarasota, Florida USA) has high sensitivity and specificity according to the manufacturer’s studies, is cheaper than PCR, and provides immediate results.

Currently, there are no licensed antiviral medications for treating EKC or other forms of adenoviral ocular infection. Ganciclovir is probably not very effective against adenovirus even though findings from in vitro and animal studies suggest it has activity. However, two novel investigational treatments show promise. They are povidone-iodine 0.4 per cent/dexamethasone 0.1 per cent, and Auriclosene (NVC-422, NovaBay Pharmaceuticals).

Based on positive results reported with the use of dilute povidone-iodine, Prof Dart said that he would treat himself with povidoneiodine 0.5 per cent, prepared from the five per cent surgical solution, if he were to develop an adenoviral ocular infection. Topical corticosteroids are also used in the management of EKC, but their role is controversial. Results from a relatively small randomised controlled trial of patients with clinically diagnosed disease suggested corticosteroid treatment was safe and had a small benefit for reducing the severity of both membranous conjunctivitis and early corneal and conjunctival scarring. In addition, topical corticosteroid treatment has been shown to control chronic subepithelial infiltrates.

However, evidence in an animal model steroid treatment increases the viral load early in the disease and because of this Prof Dart said that he avoids using steroid for the first seven days when treating patients with adenoviral infection. Thereafter, he will prescribe intensive corticosteroid treatment, dexamethasone 0.1 per cent every one to two hours, in individuals who develop membranous conjunctivitis which is uncommon before a week. In addition, he prescribes a “safe” steroid, eg, fluorometholone, for use one to three times daily as treatment for patients with symptomatic subepithelial infiltrates. If the lesions are very persistent, cyclosporine drops may be a reasonable steroid-sparing alternative according to experience in three small case series, said Prof Dart.

CMV was first described as a cause of endotheliitis in Japan in 2006, and since then, the number of cases has been growing. While the reports are nearly all from Asia, Prof Dart believes ophthalmologists in Western countries should be thinking about CMV infection as a cause for unexplained and treatment-refractory endotheliitis or post-transplant rejection. “I have instituted a protocol for CMV diagnosis in these situations, and I would encourage others to consider it as well,” said Prof Dart. “Proper recognition of CMV endotheliitis will allow the initiation of effective antiviral treatment and hopefully prevent progression to corneal endothelial failure.”

Characteristic signs that should raise suspicion of CMV infection include localised corneal decompensation with linear (pigmented or clear) and coin-shaped keratic precipitates along with owl’s eye cells in the endothelium on confocal microscopy. Patients may also have elevated IOP. However, not all patients with CMV endotheliitis have the typical clinical features. Prior to initiating antiviral therapy, laboratory confirmation of the diagnosis should be made using RT-PCR or antibody testing of an aqueous sample. Treatment for CMV endotheliitis may involve combined systemic and topical anti- CMV agents. Published reports describe the use of intravenous ganciclovir, oral valacyclovir or oral valganciclovir and topical acyclovir or ganciclovir drops 0.3 per cent to two per cent. Prof Dart said topical ganciclovir 0.15 per cent gel might be a reasonable alternative.

Existing treatment with a corticosteroid or other immunosuppressive agent should be reduced. Patients should be monitored for treatment response with PCR and followed with measurements of endothelial cell count and pachymetry. Antiviral prophylaxis is indicated in patients who undergo a corneal graft procedure.