Diagnostic capability of lamina cribrosa thickness by enhanced depth imaging and factors affecting thickness in patients with glaucoma.

OBJECTIVE: To determine the diagnostic ability of laminar thickness obtained using enhanced depth imaging (EDI) of the Heidelberg Spectralis optical coherence tomography (OCT) system (Heidelberg Engineering, Heidelberg, Germany) and the factors related to laminar thickness in patients with glaucoma.

DESIGN: Evaluation of a diagnostic test.

PARTICIPANTS: A total of 144 patients with glaucoma (68 with primary open-angle glaucoma [POAG], 76 with normal-tension glaucoma [NTG]), and 65 healthy controls.

METHODS: All patients underwent retinal nerve fiber layer (RNFL) thickness and optic nerve head (ONH) scanning with the Spectralis OCT system. Mean laminar thickness was defined at the center of the mid-superior, central, and mid-inferior horizontal B-scans of the ONH, and the mean of 3 laminar thickness measurements was used for the diagnostic test. Receiver operating characteristic (ROC) curves were obtained for average and quadrant RNFL thickness and mean laminar thickness. Areas under the ROC curve (AUCs), 95% confidence intervals (CIs), and sensitivities at a fixed specificity (90%) were calculated. Factors related to mean laminar thickness were analyzed by univariate and multivariate regression analyses in patients with glaucoma.

MAIN OUTCOME MEASURES: Comparison of diagnostic ability using AUCs.

RESULTS: Mean laminar thickness had the largest AUCs among all patients with glaucoma (AUC, 0.980; 95% CI, 0.966-0.993) and patients with NTG (AUC, 0.989; 95% CI, 0.980-0.994). The AUC of the mean laminar thickness of patients with NTG was significantly different from the AUC of the inferior RNFL thickness (AUC, 0.947; 95% CI, 0.929-0.965; P = 0.047), which had the largest AUC among the RNFL thickness parameters. The AUCs for discrimination between early glaucomatous eyes and normal eyes showed significant differences between the NTG group (AUC, 0.981; 95% CI, 0.968-0.992) and all patients with glaucoma (AUC, 0.941; 95% CI, 0.931-0.952; P = 0.007). The factor significantly associated with laminar thickness in both univariate and multivariate regression was NTG diagnosis (P = 0.001).

CONCLUSIONS: The diagnostic ability of laminar thickness was comparable to that of peripapillary RNFL thickness in patients with glaucoma overall and better than peripapillary RNFL thickness in patients with early NTG.