Comparison of Humphrey Matrix Frequency Doubling Technology to standard automated perimetry in neuro-ophthalmic disease
Michael K Yoon1, Thomas N Hwang2, Shelley Day3, Jenny Hong1, Travis Porco4, Timothy J McCulley5
1 Department of Ophthalmology, University of California San Francisco, San Francisco, California, USA
2 Department of Ophthalmology, University of California San Francisco, San Francisco, California, Department of Ophthalmology, Kaiser Permanente, Redwood City, California, USA
3 Department of Ophthalmology, Duke University, Durham, North Carolina, USA
4 Department of Ophthalmology, Epidemiology and Biostatistics, University of California San Francisco, San Francisco, California, USA
5 The Wilmer Ophthalmological Institute, Johns Hopkins School of Medicine, Baltimore, MD, USA; King Khaled Eye Specialist Hospital, Riyadh, Saudi Arabia
Timothy J McCulley
The Wilmer Eye Institute, Johns Hopkins School of Medicine, 600 North Wolfe Street, Wilmer 110, Baltimore, MD 21287, USA
Source of Support: Supported in part by an unrestricted grant to the Wilmer Ophthalmological Institute from Research to Prevent Blindness Inc, New York,, Conflict of Interest: None
Purpose: We compared Humphrey Matrix FDT 30-2 (FDT) and Humphrey Visual Field Analyzer 30-2 SITA standard (SAP) in the assessment of anterior (optic nerve or chiasm) and posterior (retro-chiasmal) afferent visual pathway defects.
Materials and Methods: In this retrospective comparative study, the charts of 37 patients (16 males, range 13-84 years, mean 72.1), with neuro-ophthalmic visual field defects who were tested with both FDT and SAP, were reviewed. Two masked graders assessed the concordance and extent of field defects between the perimeters. The mean concordance between anterior and posterior disease was compared using the Wilcoxon rank sum test. The mean deviation (MD) and pattern standard deviation (PSD) of each perimeter were correlated with the Spearman coefficient.
Results: Twenty-eight patients had anterior and nine had posterior disease. Most had a fair or good concordance (89.3% anterior, 88.9% posterior). When comparing anterior to posterior disease, the mean concordance of the defects of the two parameters was not statistically different (P = 0.94 and P = 0.61 for total deviation and pattern deviation, respectively). The MD and PSD between perimeters had a significant correlation.
Conclusions: Our series, using 30-2 field analysis, demonstrates fair to good correlation between FDT and SAP in the majority of patients. In roughly 10% findings between FDT and SAP were discordant. This difference was similar for anterior and posterior disease.