Early Predictors of Long-term Outcomes in Childhood Glaucoma

Purpose: A data-based staging system for childhood glaucoma is likely to improve the prediction of visual acuity and intraocular pressure (IOP) control outcomes. We investigated early clinical factors associated with poor long-term visual and IOP control outcomes in early-onset glaucoma as the initial steps to constructing a severity staging system. Design: Statistical modeling of retrospective case series data. Participants: Glaucoma patients younger than 3 years of age who presented to Bascom Palmer Eye Institute between 1990 and 2010 with at least 5 years of follow up. Methods: Statistical modeling of retrospective case series data from first, second, third, 3-year, 5-year and final visits. Main Outcome Measures: Association of early clinical characteristics to final IOP-control outcomes and visual acuities. Results: Twenty-six eyes of 15 children were included. Nine of 15 (60%) of patients were male. Mean age at initial presentation: 9.98±10.55 months. Mean duration between initial and final visits: 11.13±3.55 years. By the third visit (mean 6.69mo after presentation), presence of nystagmus, anterior segment dysgenesis (ASD) or having failed angle surgery increased the final LogMAR visual acuity by 0.76 (P=0.0516), 0.64 (P=0.0618) and 0.58 (P=0.0159), respectively. At year 3, failed amblyopia therapy, failed angle surgery, nystagmus, media opacity (MO), or ASD increased the final LogMAR by 1.30, 1.34, 1.21, 0.85, 0.64, respectively (all P<0.02). Failed angle surgery or MO increased the chance of uncontrolled IOP at the final visit (proportional odds ratio of 6.77 and 12.88, respectively). Conclusions: In this pilot study of a modest size cohort, the presence of nystagmus, ASD or failed angle surgery early in the course of infantile-onset glaucoma management predicted poor final visual outcome, whereas failed angle surgery and presence of MO predicted poor final IOP control. These predictors of poor outcomes will serve as the initial steps in constructing a severity staging system.