Prevalence and Onset of Pediatric Sickle Cell Retinopathy.

PURPOSE: Children with sickle cell hemoglobinopathy (SCH) can demonstrate proliferative retinopathy with vision loss, but lack of consensus exists regarding screening regimens. We sought to determine the prevalence, age at onset, and risk factors associated with sickle cell retinopathy (SCR) to inform development of screening guidelines for asymptomatic children.

DESIGN

Retrospective cohort study.

PARTICIPANTS

Children with SCH over a 4-year period.

METHODS

Prevalences of any retinopathy, nonproliferative retinopathy (NPR), and proliferative retinopathy (PR), determined as proportions of all children examined, were calculated. Subgroup analyses were completed by SCH genotype. Ages at first diagnosis were reported using standard descriptive statistics. The association of potential risk factors with retinopathy were assessed using univariate and multivariate linear and logistic regression.

MAIN OUTCOME MEASURES: Outcomes were prevalence, age at onset, and type of SCR, based on examination by an ophthalmologist. Markers of SCH severity (number of emergency room or hospital admissions for crises, number of blood transfusions, hydroxyurea therapy, transcranial Doppler-confirmed cerebral vasculopathy), genotype, gender, and race were evaluated as SCR risk factors.

RESULTS

Of 398 children (mean age, 9.6±4.6 years; range 0-18 years), 208 (52%) showed sickle cell homozygote (SS) genotype, 113 (28%) showed sickle cell hemoglobin C (SC) genotype, and 77 (19%) showed trait genotype. Forty-eight children (12.1%) demonstrated SCR, 44 of 398 children (11.1%; 95% confidence interval, 8.3%-14.5%) demonstrated NPR, and 9 of 398 children (2.3%; 95% confidence interval, 1.2%-4.2%) demonstrated PR. Prevalence was higher for SC than SS genotype for NPR (21% vs. 9%) and PR (5% vs. 1%); onset for SC genotype was earlier than that for SS genotype for NPR (youngest diagnosis 4.8 vs. 6.1 years) and PR (12.2 vs. 15.4 years). No other risk factors were associated significantly with SCR.

CONCLUSIONS

Clinical markers of SCH severity assessed were not associated with SCR and are not necessary for screening guidelines. Based on our study and literature review, although screening could begin at age 5 years for NPR, screening of children without ophthalmologic symptoms to identify treatment-requiring PR could begin later, at 9 years of age for SC and 13 years of age for SS.