Association of Macular Optical Coherence Tomography Measures and Deficits in Facial Recognition in Patients With Glaucoma

Abstract

This study investigates the association between facial recognition and macular structural damage, as measured by spectral-domain ocular coherence tomography retinal ganglion cell and inner plexiform layer.

We recently reported that glaucomatous macular damage, as measured by 10-2 perimetry, is directly correlated with diminished facial recognition in eyes with good central vision.¹ However, it is possible that glaucomatous macular damage impairs several dimensions of visual function required for facial recognition and is not limited to impairment related to Humphrey visual field defects. This study investigates the direct association between facial recognition and macular structural damage, as measured by spectral-domain ocular coherence tomography (SD-OCT) retinal ganglion cell and inner plexiform layer (RGC+). Because global damage is more likely to impair central visual tasks, such as contrast and vernier acuity, we also determined if the pattern of glaucomatous macular damage had a differential association with facial recognition.

Methods

The institutional review board of Columbia University Medical Center approved the study and its methods. All study methods adhered to the tenets of the Declaration of Helsinki.² Eligible consecutive patients with primary open-angle glaucoma in 1 or both eyes were recruited between March 25, 2019, and April 30, 2019. Verbal informed consent was obtained from all patients. Methods have been previously described.¹ In brief, all participants underwent a comprehensive ophthalmologic examination, including SD-OCT and standard automated perimetry (24-2 and 10-2 programs) using the Swedish Interactive Threshold Algorithm (Carl Zeiss Meditec). Glaucoma was defined as characteristic optic nerve damage on stereoscopic examination with localized or diffuse retinal nerve fiber layer thinning on SD-OCT and with corresponding visual field defects. Facial recognition was assessed using the Cambridge Face Memory Test. Macular damage in the better seeing eye (determined by mean RGC+) was categorized as focal, diffuse, or mixed (methodology is demonstrated in the decision tree in the Figure). Two-sided P values were significant at less than .05. Analysis began September 2020.

Figure. Decision Map for Determining Focal, Diffuse, and Mixed Types of Macular Damage Based on Retinal Ganglion Cell and Inner Plexiform Layer Sectors on Spectral-Domain Optical Coherence Tomography (OCT).

Results

Of 68 better eyes, 54 (79%) had structural macular damage and 14 (21%) did not. There was no difference between eyes with and without macular damage with respect to age (1.5 years; 95% CI, −8.6 to 5.6; P = .67), number of antiglaucoma medications (0.4; 95% CI, −1.0 to 0.02; P = .19), logMAR visual acuity (0.03; 95% CI, −0.08 to 0.02; P = .25), proportion with early cataract (0.10; 95% CI, −0.18 to 0.40; P = .46), spherical equivalence (2.87 diopters; 95% CI, −0.65 to 2.99; P = .28), or significant astigmatism (0.15 diopters; 95% CI, −0.39 to 0.09; P = .21).

In the multivariable analyses for the better eye, there was a positive association between facial recognition and mean RGC+ (β = 0.19; 95% CI, 0.01-0.37; P = .04), even after adjusting for the 10-2 visual field as well as age, visual acuity, and number and type of drops. Multivariable analyses are summarized in the Table. Patients with diffuse macular damage recognized fewer faces (47.9; 95% CI, 43.6-52.0) than those with focal macular damage (50.4; 95% CI, 47.4-53.4) or those without macular damage (51.6; 95% CI, 48.0-55.3), although the difference only approached significance at P = .051.

Table. Multivariate Analyses for Better Eyes.

Potential determinant	β Coefficient (95% CI)	P value
Mean RGC+, μm	0.19 (0.01 to 0.37)	.04a
Age, y	−0.22 (−0.37 to 0.08)	.003a
Central visual acuity, logMAR	−6.65 (−24.6 to 11.3)	.46
10-2 VF MD, dB	0.64 (0.31 to 0.97)	<.001a
No. of antiglaucoma drops	1.68 (−1.04 to 4.40)	.22
Drop use
β-Blocker	−7.00 (−10.90 to −3.01)	.001a
α-Agonist	−5.01 (−9.45 to −0.56)	.03a
Prostaglandin analogue	−4.80 (−9.52 to −0.07)	.047a

Abbreviations: MD, mean deviation; RGC+, retinal ganglion cell and inner plexiform layer; VF, visual field.

^aP values were statistically significant at less than .05. Univariable and multivariable linear regression analyses using ordinary least squares were conducted to determine the association between macular function in better eyes and facial recognition. All potential confounders with P < .20 in the univariable analyses were then evaluated by a linear multivariable regression model.

Discussion

In the present study, we show that decreased mean RGC+ in the better-seeing eye was independently associated with diminished facial recognition, even after adjusting for central visual field loss. This suggests that field loss may not be the only visual deficit resulting from macular damage and that the resultant functional losses may impair facial recognition. Our study also suggests that diffuse rather than focal structural macular damage in the better seeing eye may be associated with diminished facial recognition, although this will need to be validated in future studies with larger sample sizes. These findings generally support previous work showing functional differences between patients with diffuse and focal macular damage.^3,4,5 It also suggests that the global effect of diffuse damage appears to result in worsened visual function.