Which OCT parameters can best predict visual field progression in glaucoma?

Primary open-angle glaucoma (POAG) is characterized by progressive loss of retinal ganglion cells (RGC), resulting in visual field (VF) loss. Although standard automated perimetry can assess visual function, 25 to 35% of RGC must be lost prior to significant VF abnormalities [1]. Additionally, fixation losses, false positive, and false negative results on VF tests can cloud judgment of glaucoma severity [2]. Thus, accurate measurement of RGC structural changes is critical to predict VF changes. Optical coherence tomography (OCT) provides non-invasive imaging to assess structural glaucomatous damage and can guide management in patients. The retinal nerve fiber layer (RNFL) consists of RGC axons, and peripapillary RNFL (pRNFL) thinning has been strongly associated with VF progression in patients (HR 8.44; 95% CI: 3.30–21.61) based on Early Manifest Glaucoma Trial criteria [3, 4]. However, drawbacks persist with RNFL measurements. For example, monitoring of advanced glaucoma may be limited by a floor effect [5, 6], and high myopia can cause anatomical alterations such as peripapillary atrophy and optic nerve head tilt resulting in errors from OCT segmentation algorithms [7]. Here, we will discuss the performance of macular OCT and neuroretinal rim width, in predicting VF changes in glaucoma patients when compared to RNFL thickness (RNFLT) (Table 1).

Table 1.

Using optical coherence tomography parameters to monitor visual field progression in glaucoma.

Parameter	Study	Visual field type progression definition	Glaucoma stage, sample size, and results
pRNFLTa	Sehi et al. 2013 [4]	∙ First follow-up visit which reached a significant negative visual field index slope over time (p < 0.05)	∙ 310 glaucoma suspect or pre-perimetric eyes & 177 perimetric eyes
			∙ MD: −3.82 ± 4.28; −0.23 ± 1.04 dB, respectively
			∙ Average and superior RNFL losses were correlated with visual field index loss in multivariate Cox models.
	Yu et al. 2016 [3]	∙ 24-2 VFb	∙ 240 eyes of 139 patients with POAGc
			∙ MD4: −9.5 ± 9.1 dB
		∙ Early Manifest Glaucoma Trial or pointwise linear regression criteria	∙ Progressive RNFL thinning predicted incidence of VF progression using both event and trend-based analysis after controlling for baseline covariates.
mGCCTd	Anraku et al. 2014 [11]	∙ Fast progressors (MD < −0.4 dB/year) versus slow progressors (MD ≥ −0.4 dB/year)	∙ 56 POAG patients with non-advanced glaucoma
			∙ MD: −3.26 ± 3.0 dB
			∙ Baseline inferior mGCCT is associated with disease progression.
	Zhang et al. 2016 [10]	∙ Significant progression (p < 0.05) on Humphrey Progression Analysis or significant negative slope for VF index	∙ 277 eyes of 188 participants from Advanced Imaging for Glaucoma Study
			∙ MD: −4.76 ± 5.13 (non-progressors) and −4.99 ± 4.21 dB (progressors)
			∙ GCC focal loss volume was the most significant structural predicting factor for VF loss in a multivariable Cox model.
mGCIPLTe	Shin et al. 2019 [13]	∙ Three consecutive abnormal VFs	∙ 541 eyes of 357 glaucoma suspect patients
			∙ MD: −0.79 ± 1.34 dB
			∙ Eyes with progressive GCIPL thinning had a significantly higher risk of developing VF defects.
	Lee et al. 2017 [12]	∙ Mean deterioration of ≥3 dB compared with 2 baseline values, (observed twice)	∙ 65 POAG patients, non-progressors (38) and progressors (27)
		∙ “Likely progression” according to event analysis	∙ Mild (MD ≥ −6 dB) or moderate to advanced (MD < −6 dB)
			∙ Global and sectoral GCIPL thinning were significantly faster for POAG patients who were classified as progressors
	Shin et al. 2017 [14]	∙ Early manifest glaucoma trial criteria or linear regression analysis of the VF index	∙ 196 eyes of 123 POAG patients,
			∙ Patients divided into mild (MD ≥ −6dB) or moderate to advanced groups (MD < −6dB) based on VF defects
			∙ Rate of change of average GCIPL thinning was significantly higher in progressors compared to non-progressors. Rate of change in RNFL thinning did not differ significantly between progressors and non-progressors in moderate to advanced group.
	Shin et al. 2020 [15]	∙ Guided progression analysis	∙ 104 POAG patients with high myopia and 104 patients who were matched with VF severity-POAG without high myopia
			∙ MD: −6.36 ± 6.22 dB (high myopia) and −5.35 ± 5.11 dB (controls)
			∙ Highly myopic eyes with progressive GCIPL thinning were at higher risk for developing VF progression after adjusting for intraocular pressure
BMOf parameters	Pollet-Villard et al. 2014 [16]	∙ 24-2 VF	∙ 142 eyes of 142 subjects with glaucoma, glaucoma suspect, or controls
		∙ No progression analysis	∙ MD: −8.28 ± 8.64 dB (glaucoma), −1.56 ± 2.22 dB (suspect), −1.60 ± 3.02 dB (controls)
			∙ Structure-function relationships between BRO-MRWg and VF sensitivity were higher than with pRNFL thickness.
	Imamoglu et al. 2017 [17]	∙ 10-2 VF	∙ 33 eyes of 29 patients with advanced glaucoma
		∙ No progression analysis	∙ MD: −14.4 (−23.8 to −11)
			∙ Sectoral BMO-MRW measurements were highly correlated with VF sensitivities on the 10-2 test
	Choi et al. 2021 [18]	∙ Visual field index using the Humphrey Field Analyzer	∙ 121 eyes (73 with POAG and 48 normal eyes)
		∙ No progression analysis	∙ BMO-minimum rim area decreased more rapidly and preceded changes in RNFLT and visual field index during glaucoma progression

^aperipapillary retinal nerve fiber layer thickness.

^bvisual field.

^cprimary open angle glaucoma.

^dmacular ganglion cell complex thickness.

^emacular ganglion cell inner plexiform layer thickness.

^fBruch membrane opening.

^gBruch membrane opening-minimum rim width.

Macular ganglion cell complex thickness (mGCCT) may serve as a more accurate predictor for VF loss for two reasons. First, the macular OCT measures structural changes that topographically matches VF testing locations. Specifically, unlike the RNFLT that only monitors RGC axonal damage, mGCCT represents the sum of the thickness of RNFL, cell bodies and dendrites in the ganglion cell layer (GCL) and inner plexiform layer (IPL), respectively [1, 8]. Additionally, approximately 50% of the RGCs are concentrated in the macula [9]. Second, RNFLT is more sensitive to decentration error and the floor effects. RNFLT is inversely related to the radial distance from the center of the optic disc and can be affected by anatomic variations. Meanwhile, GCCT varies slowly outside of the parafoveal region and has a larger sampling circle (6 mm) compared to NFLT (3.4 mm) [10]. Evidently, several studies suggested the superior performance of the mGCCT comparing to the RNFLT for VF prediction. For instance, Anraku et al. found that inferior mGCCT, but not RNFLT, was the only prognostic factor for VF loss in POAG patients via a multivariate analysis (OR: 0.894; 95% CI: 0.825–0.970; p = 0.007) [11]. Patients without advanced glaucoma (MD > −12 dB) were divided into groups based on VF loss rates: fast progressors (mean deviation (MD) slope < −0.4 dB/year) and slow progressors (MD slope ≥ −0.4 dB/year). Baseline measurement of inferior hemifield mGCCT was significantly lower in fast progressors compared to slow progressors (68.0 ± 6.6 μm; 78.2 ± 11.6 μm; p = 0.002). Additionally, Zhang et al. showed that baseline mGCCT focal loss volume (FLV) but not pRNFLT FLV was the most accurate structural predictor of VF progression in a multivariate model (HR 1.36; p < 0.0001) [10].

Likewise, the ganglion cell inner plexiform layer (GCIPL) thickness (mGCIPLT), analyzing the dendrites of RGCs, demonstrates excellent performance in detecting VF progression, especially in eyes with anomalous optic nerves or advanced glaucoma. In early to moderate glaucoma, Lee et al. found that eyes with glaucoma progression – determined by serial red-free photography or VF tests - had a significantly higher rate of GCIPL and RNFL thinning - globally and in the affected hemifield - over time compared to nonprogressors [12]. Shin et al. found that progressive GCIPL thinning (HR: 7.130; 95% CI: 3.137–16.205) and RNFL thinning (HR: 7.525; 95% CI: 3.272–17.311) correlated with a higher risk of developing VF defects and were detected earlier compared to functional damage in glaucoma suspects [13]. However, in moderate to advanced glaucoma, while the rate of change of mGCIPLT was higher in progressors (−0.66 ± 0.30 μm/year) compared to nonprogressors (−0.31 ± 0.50 μm/year, p < 0.05), the rate of change of RNFL thickness was not significantly different in progressors (−0.26 ± 0.55 μm/year) versus nonprogressors (−0.33 ± 0.92 μm/year, p = 0.765) likely due to the “floor effect” [14]. In highly myopic eyes, GCIPL thinning (HR: 4.0; 95% CI: 1.76–9.09) but not RNFL thinning was significantly associated with VF progression [15]. This may be because myopia can cause thinner baseline RNFL and optic disc variation affecting RNFLT.

Additionally, novel parameters can analyze the neuroretinal rim using optic nerve head scans. These parameters rely on identification of the Bruch’s membrane (BM), a thin, acellular extracellular matrix located between the retina and choroid. Specifically, the BM opening – minimum rim width (BMO – MRW) measures the distance between the opening and the internal limiting membrane. Although not yet established as a prognostic factor for VF progression, previous studies have found a stronger correlation between sectoral BMO-MRW measurements and 24-2 VF sensitivity compared to RNFL measurements [16]. Additionally, temporal BRO-MRW measurements were significantly associated with their corresponding 10-2 VF sensitivities in advanced glaucoma [17]. Notably, Choi et al. found that BMO – minimum rim area had a faster and earlier decrease in early glaucoma compared to traditional metrics such as the RNFLT and visual field index [18].

Which OCT-based parameters are the best predictors for current and future functional damage in glaucoma patients? The answer is that it depends. Overall, RNFLT changes can predict VF progression; however, limitations such as the “floor effect”, myopia, sensitivities to decentration error, thickness changes outside the range of the VF test may prove challenging for accurate monitoring, especially in eyes with advanced glaucoma. Therefore, we propose three clinical implications: (1) Leverage all imaging: utilize RNFL, macular OCT and optic disc evaluation in tandem with VF for glaucoma diagnosis and monitoring. Be mindful that key diagnostic and prognostic parameters, such as disc hemorrhages and pallor, cannot be detected using OCT. (2) OCT macula can be helpful in early detection in eyes with anomalous nerves or myopia and in advanced stage-glaucoma when the RNFL reaches the floor. (3) Future studies can consider OCT angiography to evaluate optic disc perfusion, which could enhance the prognostic prediction, in combination with our current OCT parameters.

Author contributions

AS and HR contributed to literature review and preparation of the manuscript. JAH and TSV contributed to editing and revising the manuscript.

Competing interests

The authors declare no competing interests.