Background: Intravitreal ganciclovir is widely used to achieve effective local antiviral concentrations for cytomegalovirus (CMV) retinitis; however, to our knowledge, standardized dosing strategies have not been established, and the reported regimens vary considerably across studies. In this study, we evaluated dose-dependent treatment outcomes of intravitreal ganciclovir for CMV retinitis.
Methods: The PubMed, Embase, Cochrane, and Scopus databases were searched through November 2025. Eligible studies included intravitreal ganciclovir monotherapy, with or without systemic antiviral therapy. Cumulative first-week intravitreal dose was calculated and classified as low dose (<4,000 µg) or high dose (≥4,000 µg). The pooled proportions for resolution, visual outcomes, recurrence, and retinal detachment were estimated using a random-effects model.
Results: Eighteen studies comprising 1132 eyes were included across all outcomes. The pooled proportion of anatomical resolution was 89% (95% confidence interval, 0.77-0.95), and 74% of eyes maintained stable or improved vision. Recurrence and retinal detachment occurred in 12% and 9% of the eyes, respectively. High-dose regimens achieved a significantly higher resolution than low-dose regimens (94% vs. 73%, p=0.019). Visual outcomes did not differ according to dose (77% vs. 73%, p=0.646). Recurrence also showed no dose-dependent difference (14% vs. 8%, p=0.654) and was observed predominantly in patients before the introduction of highly active antiretroviral therapy. The retinal detachment rates were similar (9% vs. 10%, p=0.780).
Conclusion: Initial intravitreal dosing at ≥4,000 µg within the first week achieved better retinitis resolution, supporting the benefit of a higher local ganciclovir concentration in the treatment of CMV retinitis.
Keywords: Cytomegalovirus retinitis; Dose-response relationship; Intravitreal ganciclovir; Review; Treatment outcome.