Correlation Between Average Retinal Nerve Fiber Layer Thickness and Rim Area of the Spectral-Domain OCT with the Humphrey Visual Field Index in Eyes with Glaucoma
Andrei P. Martin, MD, Joseph Anthony Tumbocon, MD, and Noel Atienza, MD
Glaucoma is the leading cause of irreversible blindness worldwide. In 2010, there were 60.5 million people affected by glaucoma, and this will increase to 79.6 million by year 2020.1 Glaucoma is a progressive disease characterized by optic nerve head damage, peripapillary retinal nerve fiber layer (RNFL) loss, and characteristic visual field defects. Detection and monitoring of glaucoma patients is based on recognition of structural and functional changes.2-4 The spectral-domain optical coherence tomography (OCT) is currently being used in the diagnosis and assessment of structural changes in glaucomatous eyes. It quantitatively measures the thickness of the peripapillary retinal nerve fiber layer and the optic nerve head (ONH) rim area (RA). The global indices for this examination are the average RNFL thickness and rim area. The Humphrey Visual Field Index (VFI) is a relatively new global perimetric index that assesses the overall integrity of the visual field expressed as a percent of a normal age-adjusted visual field.5-6 In patients with advanced glaucoma, functional tests may not be reliable, with some patients unable to do visual field testing. Hence, there is a need for other indices to monitor the disease. In this study, we established the correlation of the OCT parameters; namely, average RNFL thickness and rim area value, with the visual field index of the Humphrey Visual Field Analyzer.
This was a cross-sectional study involving patients who underwent spectral-domain OCT of the optic nerve and Humphrey perimetry from January 2011 to October 2012 at the International Eye Institute, St. Luke’s Medical Center – Global City. Their medical records were retrieved and eyes diagnosed with glaucomatous optic neuropathy based on stereoscopic optic nerve head photographs, Cirrus spectral domain OCT RNFL and ONH measurements, and Humphrey standard automated perimetry were included in the study. Eyes without glaucomatous optic neuropathy or with other types of optic neuropathy; inadequate signal strength of 5/10 or below on spectral domain OCT imaging; high false positive (>33%), high false negative responses (>33%), and high fixation loss (>33%) on Humphrey visual field test; and patients with visual acuity of <20/200 who were unable to undergo standard automated perimetry were excluded. Each patient underwent eye examinations consisting of standard automated perimetry (Humphrey Field Analyzer, Carl Zeiss Meditec, Dublin CA) using the 30-2 SITA standard strategy, optical coherence tomography (Cirrus OCT, Carl Zeiss Meditec, Dublin, CA) using the RNFL and ONH:Optic Disc Cube 200×200 protocol, and optic nerve head photography using a Zeiss fundus camera and VISUPAC system. A glaucoma specialist independently interpreted all examination results based on glaucomatous optic nerve head changes and RNFL thinning on OCT, with corresponding functional changes in perimetry.
The average peripapillary RNFL thickness, optic nerve head rim area, and visual field index (VFI) of each eye were collated. The relationship between the average RNFL thickness and VFI, and optic nerve head rim area and VFI were analyzed using Spearman’s correlation coefficient, with a positive correlation being a value greater than 0. The Fisher r-to-z transformation was used to assess the significance of the difference between the two correlation coefficients. A value of p≤0.05 was considered statistically significant.
A total of 121 eyes of 85 glaucoma patients were included for analyses. The mean age of the patients was 63.7 ± 16.6 years, ranging from 12 to 94 years. 55.3% were male (n = 47) and 44.7% were female (n = 38). The mean average peripapillary RNFL thickness was 85.6 ± 5.7μm, ranging from 46 to 111μm. The highest average peripapillary RNFL thickness (111μm) was observed in a 31-year-old male, and the thinnest (46μm) in a 58-year-old male. The mean average RNFL thickness was similar for males and females (67.4 ± 13.6μm vs 68.6 ± 10.4μm; p = 0.58). The optic nerve head rim area ranged from 0.18 to 1.94 mm2 with a mean of 0.646 ± 0.3 mm2. The mean rim area was similar for males and females (0.64 ± 0.29 mm2 vs. 0.65 ± 0.25 mm2 , p = 0.92). The largest rim area (1.94 mm2) was observed in a 53-year-old male and the thinnest (0.18 mm2 ) in a 74-year-old female. The visual field index ranged from 0 to 98% with a mean of 56 ± 32%. The mean VFI was significantly higher in females than in males (65 ± 30% vs 48 ± 32%; p = 0.03). The highest VFI (98%) was observed in a 58-year-old female and the lowest (0%) was observed in five different males. Scatter plot showing the relationship between the average peripapillary RNFL thickness and VFI and the optic nerve head rim area with VFI are shown in Figures 1 and 2. A direct linear correlation was found for the average RNFL thickness and VFI (r = 0.35), and for the rim area and VFI (r = 0.15). The average RNFL thickness showed a greater correlation with the VFI than the rim area; however, this was not statistically significant (p = 0.10). Of the 121 eyes tested, 30.5% (n = 37) had VFI between 0 and 33%, 22.3% (n = 27) between 34 and 66%, and 47.1% (n = 57) between 67 and 100%. Using Spearman’s correlation coefficient, a positive correlation was established between RNFL thickness and VFI (r = 0.34), and between rim area and VFI (r = 0.02) in eyes with VFI values of 0-33% (Table 1). Although the RNFL thickness showed higher correlation coefficient with VFI than the rim area, the difference was not statistically significant (p = 0.16). A positive correlation was also established for the RNFL thickness and rim area with VFI in the 34-66% and 67-100% subgroups (Table 1). The difference between the 2 groups was also not statistically significant (p > 0.05).
The results from this study showed that in patients with glaucoma, the average RNFL thickness and the optic nerve head rim area obtained from the spectraldomain OCT demonstrated weak positive correlations with the visual field index obtained from standard automated perimetry of the Humphrey Visual FieldAnalyzer (Figures 1 and 2), similar to the findings of Sehi et al.7 This indicates a weak structure-function relationship. Nilforushan et al8 reported a higher correlation of rim area with VF thresholds compared to the RNFL. Our study, however, showed that the average RNFL thickness had a greater correlation with VFI than optic nerve head rim area, but it was not statistically significant (p >0.05). Though this study showed that, over-all, there was a correlation between the average RNFL thickness and rim area with the VFI, the correlation was not high. This was true specially for eyes with more significant visual field loss; in those with VFI of 0- 33%, the Spearman’s correlation coefficient was 0.34 and 0.02 for average RNFL thickness and rim area respectively. In earlier stages of glaucoma, the average RNFL thickness and rim area showed the highest correlation (r = 0.32 and 0.34 respectively), indicating that the OCT best predicts the VFI in earlier stages of the disease. The limitation of this study was the relatively small sample that included glaucomatous eyes. Moreover,
eyes with different co-morbidities that may decrease the VFI were also not excluded in this study. Further prospective studies involving non-glaucomatous and glaucoma suspects can be used as comparison groups.
Other parameters, such as the cup-to-disc ratio and Hoddap classification, can be studied to further divide
the eyes into subgroups. In conclusion, the average peripapillary RNFL thickness and optic nerve head rim area measured by the spectral-domain OCT have positive correlations with the standard automated perimetry visual field index of the Humphrey Visual Field Analyzer. These OCT parameters are weak indicators for VFI, indicating weak structure-function relationships in many cases of glaucoma.
We would like to acknowledge Ms. Ayra E. Martin for her contributions with the statistical analysis.
1. Quigley HA, Broman AT. The number of people with glaucoma worldwide in 2010 and 2020. Br J Ophthalmol 2006;90:262- 267.
2. Savini G, Carbonelli M, Barboni P. Spectral-domain optical coherence tomography for the diagnosis and follow-up of glaucoma. Curr Opin Ophthalmol 2011;22:115-23.
3. Nilforushan N, Nassiri N, Moghimi S, et al. Structure-function relationships between spectral-domain OCT and standard achromatic perimetry. Invest Ophthalmol Vis Sci 2012;53:
4. Leite MT, Zangwill LM, Weinreb RN, et al. Structure-function relationships using the Cirrus spectral domain ptical coherence tomograph and standard automated perimetry. J Glaucoma 2012;21:49-54.
5. Giraud JM, Fenolland JR, May F, et al. Analysis of a new visual field index, the VFI, in ocular hypertension and glaucoma. J Fr Ophthalmol 2010;33:2-9.
6. Sawada H, Fukuchi T, Abe H. Evaluation of the relationship between quality of vision and the visual function index in Japanese glaucoma patients. Graefes Arch Clin Exp Ophthalmol 2011;249:1721-7.
7. Sehi M, Zhang X, Greenfield DS, et al. Advanced Imaging for Glaucoma Study Group. Retinal nerve fiber layer atrophy is associated with visual field loss over time in glaucoma suspect and glaucomatous eyes. Am J Ophthalmol 2013;155:73-82.
8. Nilforushan N, Nassiri N, Moghimi S, et al. Structure-function relationships between spectral-domain OCT and standard achromatic perimetry. Invest Ophthalmol Vis Sci 2012;53:2740- 8.
9. Wollstein G, Schuman JS, Price LL, et. al. Optical coherence tomography longitudinal evaluation of retinal nerve fiber layer thickness in glaucoma. Arch Ophthalmol 2005;123:464-470.