advertisement

Editors Selection IGR 14-1

Clinical Examination Methods: Detecting Progression

Comment by Vincent Michael Patella on:

76870 Event-based analysis of visual field change can miss fast glaucoma progression detected by a combined structure and function index, Zhang C; Tatham AJ; Daga FB et al., Graefe's Archive for Clinical and Experimental Ophthalmology, 2018; 256: 1227-1234

Find related abstracts

Zhang and colleagues followed 135 eyes of 97 glaucoma patients for an average of 3.5 years to compare glaucoma progression as identified by a perimetric event analysis (Humphrey GPA) to progression events based upon rates of progression as estimated a) by OCT measures of RNFL thickness and b) by an index (RGC Index) combining standard automated perimetry (SAP) and OCT RNFL measurements. For RNFL thickness and RGC Index, eyes were deemed to have progressed if the slope was statistically significant (P < 0.05) and faster than average age-related loss, as determined from a group of 50 healthy eyes followed for an average of 1.9 years.

At baseline, median MD of the glaucoma eyes was -2.80 dB (IQR -4.76 to -1.26). Twenty-one eyes progressed by the RGC index but not by GPA, whereas only five eyes showed progression on GPA that was not detected with the RGC index. Eighteen eyes progressed by the RGC index but not by SDOCT, whereas eight eyes progressed by SDOCT but not RGC index. None of the healthy eyes showed perimetric progression on GPA.

The authors concluded that many glaucomatous eyes that were not found to be progressing by perimetric event analysis may actually have had fast rates of progression as detected by a combined index of structure and function.

This paper nicely demonstrates the advantages of analyses that combine data from complementary sources

Comments

1. This paper nicely demonstrates the advantages of analyses that combine data from complementary sources. Based upon the MD range cited above, most patients in this study had early glaucoma, which is exactly where OCT can contribute best in an OCT-Perimetry combined analysis. More importantly, the RGC index standardizes and simplifies the identification of glaucomatous progression based upon OCT & SAP.
2. Unfortunately, the compared metrics were not specificity matched, and without specificity matching, the sensitivities of diagnostic methods cannot be precisely compared, as authors from this group have recently affirmed.¹ However, we do see that GPA found zero progression events in 50 healthy controls, and RGC measurements showed nearly zero overlap between normal aging in the controls and the glaucoma eyes it found to progress. Thus, both methods appear to be operating at high specificities in this particular cohort. Therefore, I believe that the RGC Index would have been shown to be more sensitive than GPA, even at matched specificities. But we cannot use this paper to say by how much.
3. More generally, defining a rate of progression to be excessive when it simply exceeds average normal aging rates, as has been done in this study, may produce insufficient specificity in everyday clinical use. Instead, perhaps progression rates should only be considered to be outside normal limits when they exceed, e.g., the fastest 5% of observed normal aging rates, which would be consistent with our treatment of other diagnostic metrics.
4. Regardless, this is timely and important work, and the idea of combination analysis, be it perimetry plus OCT - or simply complementary combinations of, e.g., OCT metrics2 - cries out for practical clinical implementation.

Defining a rate of progression to be excessive when it simply exceeds average normal aging rates, as has been done in this study, may produce insufficient specificity in everyday clinical use

• ← Previous
• Next →