Diagnostic accuracy of retinal optical coherence tomography in children with a newly diagnosed brain tumour

• Published in: Acta ophthalmologica (Oxford, England) England), 2023-09, Vol.101 (6), p.658-669
• Main Authors: Nuijts, Myrthe A., Stegeman, Inge, Porro, Giorgio L., Bennebroek, Carlien A. M., Seeters, Tom, Proudlock, Frank A., Schouten – van Meeteren, Antoinette Y. N., Imhof, Saskia M.
• Format: Article
• Language: English
• Subjects: Accuracy; Acuity; adolescent; Brain cancer; Brain tumors; brain tumour; Child; Cohort Studies; Cross-Sectional Studies; Female; Humans; Male; optical coherence tomography; Prospective Studies; Retina; Retinal Ganglion Cells - pathology; Sensitivity analysis; Tomography; Tomography, Optical Coherence - methods; Vision Disorders - pathology; Visual acuity; Visual field; visual function; Visual perception
• ISSN: 1755-375X; 1755-3768
• DOI: 10.1111/aos.15650

Purpose To estimate the diagnostic accuracy of circumpapillary retinal nerve fibre layer (RNFL) thickness and macular ganglion cell layer–inner plexiform layer (GCL‐IPL) thickness measurements to discriminate an abnormal visual function (i.e. abnormal age‐based visual acuity and/or visual field defect) in children with a newly diagnosed brain tumour. Methods This cross‐sectional analysis of a prospective longitudinal nationwide cohort study was conducted at four hospitals in the Netherlands, including the national referral centre for paediatric oncology. Patients aged 0–18 years with a newly diagnosed brain tumour and reliable visual acuity and/or visual field examination and optical coherence tomography were included. Diagnostic accuracy was evaluated with sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). Results Of 115 patients included in the study (67 [58.3%] male; median age 10.6 years [range, 0.2–17.8 years]), reliable RNFL thickness and GCL‐IPL thickness measurements were available in 92 patients (80.0%) and 84 patients (73.0%), respectively. The sensitivity for detecting an abnormal visual function was 74.5% for average RNFL thickness and 41.7% for average GCL‐IPL thickness at a specificity of 44.5% and 82.9%, respectively. The PPV and NPV were 33.0% and 82.6% for the average RNFL thickness and 57.1% and 82.2% for the average GCL‐IPL thickness. Conclusion An abnormal visual function was discriminated correctly by using the average RNFL thickness in seven out of ten patients and by using the average GCL‐IPL thickness in four out of ten patients. The relatively high NPVs signified that patients with normal average RNFL thickness and average GCL‐IPL thickness measurements had a relative high certainty of a normal visual function.