Lag time to diagnosis as a predictor of mortality in children with extraocular retinoblastoma: Experience from a developing country

Abstract

Aim: To define the prevalence of mortality and identify factors associated with mortality in pediatric patients with extraocular retinoblastoma attending the tertiary hospital in Indonesia.

Methods: We retrospectively collected medical records from 2013 to 2019 of patients who were diagnosed with extraocular retinoblastoma. Cox proportional hazard regression analysis with 95% confidence interval (CI) was used to evaluate the association of mortality predictors with patient outcomes (Hazard Ratio [HR], 95% CI) with significance set as p < .05.

Results: Overall, 60 patients were included in this study for a retrospective chart review, with 55% males and 45% females. The median age at diagnosis was 13 (5-24) months. About 60% of the patients did not survive, while 2-year survival probability was 45%. The overall median survival time was 21.5 (7.25-40.75) months. Predictors of mortality were laterality (unilateral/bilateral): HR 2.15 (95% CI: 1.07-4.28; p = .03), nutritional status: HR 2.65 (95% CI: 1.34-5.25; p = .05), and lag time to diagnosis: HR 3.12 (95% CI: 1.56-6.2; p = .001).

Conclusion: Laterality, nutritional status and lag time to diagnosis were identified to be mortality predictors in extraocular retinoblastoma. The 2-year survival for children with extraocular retinoblastoma was 45% with 21 months for median survival.

Keywords: death; developing country; extraocular retinoblastoma; lag time to diagnosis; mortality predictors.