Abstract
Purpose :
Varicella zoster virus (VZV) causes herpes zoster ophthalmicus by reactivation of latent virus, which leads to ocular complications, characterized by keratitis
or iridocyclitis. Although diagnosis is generally made by its acute skin episode, effectiveness of viral genome-based diagnosis for VZV-associated keratitis has remained obscure. We sought to evaluate how real-time PCR of VZV may improve its diagnostic efficacy as well as predict prognosis for better management of chronic and refractory cases.
Methods :
The medical records of 563 consecutive case series with suspected VZV-associated keratitis or infectious keratitis were retrospectively reviewed. All the patient underwent real-time PCR of VZV using corneal scraping samples at the Tottori University Medical Hospital between November 2005 and September 2016. Diagnostic accuracy of real-time PCR and clinical signs were determined by receiver operating characteristic analysis. Prognosis, described as the persistence of active disease and complications, were assessed for their associations using logistic regression and cox proportional hazard model.
Results :
For the VZV-associated keratitis (38 eyes), VZV copy number was 1.9×107± 5.9×107 copy. Significant associations with the VZV amount was observed for facial skin eruption (OR; 4.1, 95% CI: 2.5- 6.5, P=0.000) and extraocular muscle paralysis (OR; 2.0, 95% CI: 1.2- 3.4, P=0.007), followed by IOP elevation, dendritic lesion, and mutton-fat keratic precipitate. VZV amount and facial skin eruption showed highest diagnostic accuracy calculated as area under the curve (AUC) of 0.92 and 0.96, respectively. The mean disease duration of VZV-associated keratitis was 108 days. Higher amount of VZV (per log) and the presence of iridocyclitis significantly prolonged the duration of active disease with hazard ratio of 0.75 (P=0.003) and 0.18 (P=0.002).
Conclusions :
Quantification of VZV genome is effective for predicting the persistence of active disease as well as diagnosis of VZV keratitis.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.