Retinal nerve fibre layer thickness measurements in childhood glaucoma: the role of scanning laser polarimetry and optical coherence tomography

Abstract

Purpose: A central diagnostic tool in adult glaucoma is the peripapillary retinal nerve fibre layer (pRNFL) thickness. It can be assessed by scanning laser polarimetry (SLP) or optical coherence tomography (OCT). However, studies investigating the relevance of pRNFL measurements in children are rare. This study aims to compare the glaucoma diagnosing ability of SLP and OCT pRNFL thickness measurements in a paediatric population.

Methods: This retrospective study included 105 children (glaucoma: 22 (21.0%); healthy glaucoma suspects: 83 (79.0%)) aged 4-18 years, examined with SLP (GDxPro/ECC, Carl Zeiss Meditec) and spectral-domain OCT (SPECTRALIS®, Heidelberg Engineering). The thickness of pRNFL sectors was compared between diseased and healthy participants. Areas under the receiver-operating characteristic curves (AUC) and logistic regression results were used to compare the glaucoma discriminative capacity between SLP and OCT measurements.

Results: Using OCT, pRNFL thickness was decreased in the superior, nasal, and inferior quadrants of glaucoma patients compared to healthy controls (P < 0.001, each). With SLP, such a difference was only observed in the inferior quadrant (P = 0.011). A correlation between glaucoma diagnosis and OCT-measured pRNFL thickness was found in all quadrants (P < 0.001) other than the temporal. With SLP, a correlation was found for the total average thickness (P = 0.037) and inferior quadrant (P = 0.0019). Finally, the AUCs of OCT measurements were markedly higher than those of SLP (e.g., inferior quadrant: OCT 0.83, SLP 0.68).

Conclusion: pRNFL thickness measurements using both OCT and SLP, correlate notably with the presence of glaucoma. In general, the diagnostic performance of pRNFL thickness measurements seems higher for OCT than for SLP. Thus, pRNFL thickness measurements could provide important information, complementing conventional clinical and functional parameters in the diagnostic process of paediatric glaucoma.

Keywords: Childhood glaucoma; Imaging; Optical coherence tomography; Primary congenital glaucoma; Retinal nerve fibre layer; Scanning laser polarimetry.

Fig. 1

Distribution of glaucoma aetiologies. The diagram shows the relative distribution of glaucoma aetiologies in the presented cohort of paediatric glaucoma patients (n = 22) using the classification recommended by the Childhood Glaucoma Research Network (https://wga.one/wga/consensus-9/). PCG, primary congenital glaucoma; JOAG, juvenile primary open-angle glaucoma; GFC, glaucoma following cataract surgery; GNASD, glaucoma associated with non-acquired systemic disease or syndrome; GNAO, glaucoma associated with non-acquired ocular anomalies; GAC, glaucoma associated with acquired conditions

Fig. 2

ROC curves for SLP and OCT measurements of pRNFL thickness. Figure 2A shows the receiver-operator characteristic (ROC) curves for total peripapillary retinal nerve fibre layer (pRNFL) thickness measurements as well as thickness measurements in the superior, nasal, inferior and temporal quadrants with scanning laser polarimetry (SLP). Figure 2B show ROC for these parameters measured by optical coherence tomography (OCT). Figure 2C compares the best performer SLP and OCT parameter (inferior quadrant, both) and presents their respective area under the ROC curve (AUC)