Predictors of long-term visual outcome after chemoreduction for management of intraocular retinoblastoma.

BACKGROUND: To determine the predictors of long-term visual outcome after chemoreduction for management of intraocular retinoblastoma.
DESIGN: Retrospective case series. PARTICIPANTS: One hundred and forty eyes of 96 new retinoblastoma patients.
METHODS: The clinical records were reviewed for demographical profile, tumour characteristics, treatment methods, treatment side effects and final visual outcome. Multivariate analysis was used to evaluate factors predictive of final visual acuity. MAIN OUTCOME MEASURE: The clinical data were analysed for main outcome measures of good vision (visual acuity ≥6/12) or ambulatory vision (visual acuity ≥6/60). The correlation among predictor variables was determined using Pearson's product moment correlation for continuous variables, and t-test and analysis of variance test for categorical baseline variables.
RESULTS: Follow-up ranged from 2 years to 14 years (median 75 months). The final mean logarithm of minimum angle of resolution visual acuity was 0.79. Ambulatory vision (≥6/60) was achieved in 100 of 140 (71%) eyes, and vision of ≥6/12 was seen in 52 eyes (37%). Extrafoveolar tumour and greater number of tumours in the eye were the only predictors of visual acuity ≥20/40. Greater number of tumours correlated with smaller mean basal tumour diameter. The mean basal tumour diameter was 9.8 mm in eyes with multiple tumours and 12.2 mm in eyes with single tumours (P = 0.03).
CONCLUSION: Long-term ambulatory vision (≥6/60) was achieved in the majority (71%) of retinoblastoma containing eyes not requiring enucleation after treatment with chemoreduction and adjunctive therapy. Absence of foveolar involvement and greater number of tumours were the predictor of long- term visual outcome of 6/12 or better.