advertisement
PURPOSE: To investigate the impact of typical scan score (TSS) on discriminating glaucomatous and healthy eyes by scanning laser polarimetry and spectral domain optical coherence tomography (SD-OCT) in 32 peripapillary sectors. PATIENTS AND METHODS: One hundred two glaucoma patients and 32 healthy controls underwent standard automated perimetry, 24-hour intraocular pressure profile, optic disc photography, GDxVCC, and SD-OCT measurements. For controls, only very typical scans (TSS=100) were accepted. Glaucoma patients were divided into 3 subgroups (very typical: TSS=100; typical: 99(greater-than or equal to)TSS(greater-than or equal to)80, atypical: TSS<80). Receiver operating characteristic curves were constructed for mean retinal nerve fiber layer values, sector data, and nerve fiber indicator (NFI). Sensitivity was estimated at (greater-than or equal to)90% specificity to compare the discriminating ability of each imaging modality. RESULTS: For discrimination between healthy and glaucomatous eyes with very typical scans, the NFI and inferior sector analyses 26 to 27 demonstrated the highest sensitivity at (greater-than or equal to)90% specificity in GDxVCC and SD-OCT, respectively. For the typical and atypical groups, sensitivity at (greater-than or equal to)90% specificity decreased for all 32 peripapillary sectors on an average by 10.9% and 17.9% for GDxVCC and by 4.9% and 0.8% for SD-OCT. For GDxVCC, diagnostic performance of peripapillary sectors decreased with lower TSS, especially in temporosuperior and inferotemporal sectors (sensitivity at (greater-than or equal to)90% specificity decreased by 55.3% and by 37.8% in the atypical group). CONCLUSIONS: Diagnostic accuracy is comparable for SD-OCT and GDxVCC if typical scans (TSS=100) are investigated. Decreasing TSS is associated with a decrease in diagnostic accuracy for discriminating healthy and glaucomatous eyes by scanning laser polarimetry. NFI is less influenced than the global or sector retinal nerve fiber layer thickness. The TSS score should be included in the standard printout. Diagnostic accuracy of SD-OCT is barely influenced by low TSS.
L. M. Hoesl.
6.9.1.2 Confocal Scanning Laser Polarimetry (Part of: 6 Clinical examination methods > 6.9 Computerized image analysis > 6.9.1 Laser scanning)
6.9.2.2 Posterior (Part of: 6 Clinical examination methods > 6.9 Computerized image analysis > 6.9.2 Optical coherence tomography)