Ganglion cell complex and retinal nerve fiber layer thickness in predicting visual outcome in brain tumors after surgery

Abstract

BACKGROUND

Anterior visual pathway lesions can cause severe visual loss. Optical coherence tomography (OCT) can detect ganglion cell complex (GCC) thinning, even with normal retinal nerve fiber layer (RNFL) thickness and subtle visual field changes.

AIM

To determine the proportion of eyes with RNFL and GCC thinning and their association with visual parameters in patients with brain tumors undergoing surgery.

METHODS

A prospective study was conducted on 37 patients (69 eyes) with peri-chiasmatic and chiasmatic brain tumors undergoing surgical decompression between February 2019 and June 2020 at a tertiary care institute. A comprehensive neuro-ophthalmological work-up, demographic and clinical profile documentation, and six-month postoperative follow-up weredone. Statistical analysis was significant at P < 0.05.

RESULTS

Mean age was 35.14 ± 11.98 years. The best and worst visual outcomes were associated with the craniopharyngioma and meningioma groups, respectively (P = 0.008). There was an increase in the proportion of eyes with RNFL damage in the inferior quadrant (P = 0.02). Maximum GCC thickness thinning was associated with severe visual impairment. The odds of having blindness in eyes with RNFL (inferior) and GCC loss were 0.96 (P = 0.003) and 0.95 (P = 0.03), respectively. GCC thickness showed a clinically positive correlation with visual acuity (r = -0.48, P < 0.001) and field defect (r = -0.27, P = 0.04) at six months postoperatively. The preoperative GCC thickness and the final postoperative visual outcome were plotted in an empirical ROC curve with area under the curve = 0.754. The cut-off value of RNFL (inferior) and GCC, beyond which blindness could be prevented, was 73 µm and 58 µm, respectively.

CONCLUSION

In chiasmal compression, RNFL and GCC thickness measurements using OCT can be a useful prognostic indicator for assessing visual recovery. An eye with structural damage, with significant RNFL and GCC loss, is a predictive factor of blindness. A minimum preoperative RNFL and GCC thickness of 73 µm and 58 µm, respectively, can preserve vision after surgery.

Key Words: Ganglion cell complex; Brain tumors; Retinal nerve fiber layer; Optical coherence tomography; Visual prognosis; Visual field; Chiasmatic compression

Core Tip: This study was conducted among perichiasmatic brain tumor patients in which optical coherence tomography was used to detect ganglion cell complex (GCC) and retinal nerve fibre layer (RNFL) damage as well as prognosticate visual outcome. A positive correlation was noted between GCC thickness, visual acuity and field defect at six months postoperatively. The predictors of blindness were reduced baseline GCC and RNFL thickness. A minimum cut-off value predicting preservation of vision was a RNFL thickness of 73 µm and a GCC thickness of 58 µm. Hence, it can guide early intervention by a neurosurgeon for vision preservation.

INTRODUCTION

Compressive optic neuropathies are one of the important anterior visual pathway disorders that can lead to severe impairment of vision, especially with delayed diagnosis. The sellar and parasellar tumors can cause compression of the optic chiasm, leading to retrograde damage of the adjacent retinal ganglion cell axons, resulting in visual field defects[1]. An endoscopic transsphenoidal approach for the excision of the tumor is usually preferred due to the lower complication rate[2]. Though difficult to assess, predicting visual outcomes after surgical intervention is imperative for preoperative counseling.

A study by Sriram et al[3] showed a 70.9% improvement in the visual field after surgery for sellar region tumors. It also concluded that patients of the younger age group, female gender, shorter duration of symptoms, and those undergoing tumor removal via transsphenoidal and transcranial approach had more significant improvement in the visual field after surgery. As per other published data, a good preoperative visual acuity, early detection, and management within one week were associated with improved postoperative visual outcomes[4,5]. Poor vision is the most common (54.8%) presenting feature, with a best corrected visual acuity worse than 20/400 in 40% of patients with the chiasmal syndrome; visual field assessment is not possible in the majority[6]. When they present with detectable visual field changes, there is irreversible damage to the optic nerve fibres. Hence, there is a need for a more sensitive parameter that can help with early detection and visual prognostication.

Optical coherence tomography (OCT) is a non-invasive tool that quantifies the thickness of the retinal nerve fiber layer (RNFL) and ganglion cell complex (GCC) in suspected patients of chiasmal compression syndrome with normal or unreliable perimetry reports[7]. In early cases, GCC damage can be picked up despite normal RNFL thickness and minimal visual field changes. Also, the field defect can be reversed with early decompression, even with persistent GCC loss in the postoperative period[8]. A study by Cennamo et al[9] reported cases of pituitary tumors with normal imaging and preserved visual function, which showed reduced thickness of GCC and RNFL in OCT scan[9]. In such cases, RNFL damage occurs by retrograde degeneration of the axons which can be detected subjectively from the degree of optic disc pallor and objectively by OCT measurements. It has been seen that preserved RNFL and GCL thickness corresponded to a better recovery of the field defects, whereas, a RNFL thickness of less than 85 microns did not lead to any improvement after surgery[8,10].

Hence, OCT can measure GCC and RNFL thickness to quantify the structural and functional axonal damage, much before subjective tests like automated perimetry. This can aid in timely surgical intervention and restore vision in the salvageable eyes. Also, OCT is essential for periodic follow-up and monitoring progression in cases with optic atrophy[7]. This study aimed to find the association of nerve fiber damage with the visual field defects and visual outcome in perichiasmatic brain tumor patient post-decompression. A probable hypothesis was given for a cut-off value for RNFL and GCC thickness for vision preservation.

MATERIALS AND METHODS

This prospective observational study was conducted at a tertiary care institute between February 2019 and June 2020. The patients between 12 and 60 years with brain tumors that caused compression of the optic chiasm on computed tomography or magnetic resonance imaging of the brain and presented with neurological and/or ocular symptoms, with or without visual field changes, were included in the study.

The exclusion criteria were vision less than the perception of light in both eyes, advanced cataract, glaucoma, high myopia, other causes of optic atrophy, and uncooperative patients. Study participants were consecutively enrolled during the study period to minimize selection bias. The sample size was estimated using the formula where Z_(1-α/2) = 1.96 for a 95% confidence, σ = 9 μm (standard deviation of GCC thickness based on Tieger et al[8]), and d = 3 μm (desired precision). Minimum Sample size came out to be 35.

Patients in the neurosurgery ward underwent a comprehensive ocular examination in the Ophthalmology clinic. Both eyes were included for subjective and objective assessment. Socio-demographic, clinical (OCT and visual fields), surgical (type, duration, complications), radiological (tumor dimensions, cause of compression), and histopathological (biopsy and immunohistochemistry) data were recorded. The study strictly adhered to the tenets of the Declaration of Helsinki. Approval from the institutional ethical committee was obtained.

All the patients enrolled in the study underwent OCT analysis. The measurements were done with the Cirrus^TM HD-OCT (500-20359) version 11.0.0.29946 (Carl Zeiss Meditec Inc, Dublin, CA 94568 United States). The RNFL was measured using an optic disc cube (200 × 200-line scan) and the thickness of the ganglion cell layer was measured using a macular cube (512 × 128-line scans). The RNFL thickness was calculated in a circle of diameter 3.46 mm with the optic nerve head in the center. The macular thickness was calculated on a 14.13 mm² elliptical annulus area centered on the fovea. The scans with signal strength > 6/10 were included for analysis. Eyes with low vision (< 6/60) were tested using fixation of the better eye after proper explanation. The mean thickness of all four quadrants of RNFL and macular GCL+ inner plexiform layer (IPL) thickness expressed in micrometers (µm) was documented. Masked investigators (Parija S and Chakraborty K), unaware of clinical or surgical outcomes, interpreted OCT and visual field assessment.

All patients underwent surgical decompression either by the transcranial or trans-nasal trans-sphenoidal route. Postoperatively, antibiotics, anti-epileptics, and drugs were administered to control raised intracranial pressure, pain, and cerebrospinal fluid leak. According to the histopathological reports, the tumors were further categorized into pituitary adenoma (PA), craniopharyngioma, meningioma and others.

All the neuro-ophthalmic tests that were done preoperatively were repeated at the time of discharge, after one and half, three, and six months of surgery. A comparative analysis was done on baseline and postoperative data.

Statistical analysis

All continuous variables have been expressed as mean ± SD (normal distribution) or median (interquartile range) for skewed data. The normality of data was calculated using the S-wilk/Kolmogorov-Smirnov test with values > 0.05 taken as normal and those < 0.05 as a skewed distribution. All categorical data have been presented as percentages. Pre and postoperative data were compared using a paired t-test for parametric and a Wilcoxon rank sum test for non-parametric ordinal variables. Association between more than three unrelated groups was determined using a one-way ANOVA test for parametric and a Kruskal Wallis test for non-parametric ordinal variables. Fischer’s exact/χ² test was done between 2 independent groups containing categorical variables. The strength of association was determined and expressed as Pearson’s coefficient. Logistic regression was done to calculate the odds ratio (OR) for predictive factors of blindness. A P-value < 0.05 was considered significant. Exploratory analysis of quadrant-wise and multi-timepoint analyses, primary outcomes (e.g., GCC-visual acuity correlation) were evaluated at P < 0.05 without adjustment, while secondary analyses applied stricter thresholds (P < 0.0125 for RNFL quadrants; P < 0.017 for timepoints) to account for multiple comparisons. Post-hoc Bonferroni-adjusted sensitivity analyses confirmed the robustness of key findings (e.g., inferior RNFL thinning, GCC prognostic value). Data were analyzed using STATA v13.

RESULTS

Demographic details and clinical presentation

A total of 37 participants were included in the study. Since two patients died, 35 patients were included for final analysis, and both eyes were evaluated. One eye had no light perception, and the patient was uncooperative for OCT evaluation. A total of 69 eyes were assessed objectively, and 70 eyes were subjectively evaluated. The mean age was 35.14 ± 11.98 years, with males comprising 17% and females 83%. Patients were from lower-middle (49%) socio-economic status according to the Kuppuswamy scale.

The complaints of headache, visual loss, and hormonal disturbances were present in 80%, 80%, and 77% of cases, respectively. The median time from the onset of clinical symptoms till the first radiological detection of the tumor was 12 (6-24) months. The median tumor volume was found to be 11.31 cc. (Table 1).

Table 1 Demographic data and clinical profile of study participants, n (%)/mean ± SD.

		Male (n = 6)	Female (n = 29)	Total (n = 35)	P value
Age (years)		34.67 ± 10.58	35.24 ± 12.42	35.14 ± 11.98	0.92
Residence					0.85
Local		5 (83.33)	25 (86.21)	30 (85.71)
Regional		1 (16.67)	4 (13.79)	5 (14.29)
Socio-economic status					0.19
Upper		1 (16.67)	1 (3.45)	2 (5.71)
Upper middle		3 (50.00)	11 (37.93)	14 (40.00)
Lower middle		1 (16.67)	16 (55.17)	17 (48.57)
Upper lower		1 (16.67)	1 (3.45)	2 (5.71)
Occupation					< 0.01
Farmer		1 (16.67)	0 (0.00)	1 (2.86)
Homemaker		1 (16.67)	21 (72.41)	22 (62.86)
Student		2 (33.33)	7 (24.14)	9 (25.71)
Business		2 (33.33)	0 (0.00)	2 (5.71)
Teacher		0 (0.00)	1 (3.45)	1 (2.86)
Chief complaints					0.82
Headache	Yes	5 (83.33)	23 (79.31)	28 (80.00)
	No	1 (16.67)	6 (20.69)	7 (20.00)
Visual loss	Yes	4 (66.67)	24 (82.76)	28 (80.00)	0.37
	No	2 (33.33)	5 (17.24)	7 (20.00)
Others (hormonal disturbances)	Yes	5 (83.33)	22 (75.86)	27 (77.14)	0.69
	No	1 (16.67)	7 (24.14)	8 (22.86)

On clinical-radiological assessment, most of the tumors were PA (43%) followed by craniopharyngioma (23%) (Figure 1). Optic chiasmatic compression was mainly seen from the inferior (33%) aspect, followed by the superior (25%) aspect. The surgical approach was via craniotomy in 54.3% and endoscopy in 45.7%, with a mean operative time of 7.77 ± 1.73 hours. Baseline visual acuity was worse for meningioma patients than for other categories (P = 0.013) (Figure 2).

Figure 1 Different types of brain tumors. SOL: Space-occupying lesion; ICA: Internal carotid artery.

Figure 2 Trend of visual acuity (logMAR) among various tumor categories. VA: Visual acuity.

Preoperative RNFL and tumors

Analysis of preoperative RNFL thickness showed the least thickness in the meningioma group and the highest thickness in the craniopharyngioma group, with no significant differences between the tumor categories (Table 2).

Table 2 Retinal nerve fiber layer thickness (microns) in various tumor categories, mean ± SD.

Tumour category	Retinal nerve fiber layer preoperative	At discharge	After 1.5 months	After 3 months	After 6 months
Pituitary adenoma	79.7 ± 27.9	80.7 ± 27.8	81.2 ± 27.5	87.6 ± 27.6	83.3 ± 28.2
Craniopharyngioma	102.0 ± 36.8	89.6 ± 34.2	67.5 ± 19.2	65.6 ± 18.2	67.8 ± 19.5
Meningioma	59.4 ± 12.9	62.8 ± 12.4	63.7 ± 11.1	60.0 ± 8.9	61.2 ± 9.9
Others	99.6 ± 32.2	89.1 ± 34.0	87.4 ± 31.8	86.9 ± 30.9	75.7 ± 27.0

Postoperative vision correlation of RNFL and GCC parameters

On comparing the data at baseline and final follow-up visit, it was seen that there was an increase in the proportion of eyes with reduced RNFL thickness in the superior, inferior, and nasal quadrants. The difference was statistically significant for the inferior quadrant by McNemar’s test (P < 0.05). There was a decrease in the proportion of eyes with reduced RNFL thickness in the temporal quadrant, which was statistically significant (P < 0.05). Similarly, after 6 months of surgery, there was a decrease in the proportion of eyes with reduced GCC thickness which was not statistically significant. (P > 0.05) (Tables 3 and 4).

Table 3 Distribution of retinal nerve fiber layer and ganglion cell complex thickness at baseline and follow-up visits, mean ± SD.

Retinal nerve fiber layer thickness (micron)	Pre-operative	At discharge	After 1.5 months	After 3 months	After 6 months
Superior	110.19 ± 60.94	107.16 ± 43.84	97.28 ± 31.19	97.38 ± 29.16	98.54 ± 31.98
Inferior	107.25 ± 53.04	102.84 ± 34.25	95.87 ± 32.16	93.58 ± 32.07	95.97 ± 31.04
Nasal	68.04 ± 31.85	65.79 ± 22.88	61.53 ± 16.13	60.78 ± 15.41	61.83 ± 16.58
Temporal	51.51 ± 19.25	50.72 ± 12.29	51.16 ± 12.1	51.38 ± 10.32	51.91 ± 9.97
Ganglion cell complex thickness (micron)	65.67 ± 14.85	64.85 ± 15.23	65.26 ± 13.69	65.43 ± 15.59	66.19 ± 15.62

Table 4 Proportion of eyes with retinal nerve fiber layer and ganglion cell complex thinning, before and after surgery, n (%).

			Preoperative			P value
			Normal	Thinning	Total
Postoperative	RNFL (superior)	Normal	35 (50.7)	4 (5.8)	39 (56.5)	0.25
		Thinning	8 (11.6)	22 (31.9)	30 (43.5)
		Total	43 (62.3)	26 (37.7)	69 (100)
	RNFL (inferior)	Normal	33 (47.8)	3 (4.3)	36 (52.2)	< 0.05
		Thinning	12 (17.4)	21 (30.4)	33 (47.8)
		Total	45 (65.2)	24 (34.8)	69 (100)
	RNFL (nasal)	Normal	42 (60.9)	6 (8.7)	48 (69.6)	0.22
		Thinning	11 (15.9)	10 (14.5)	21 (30.4)
		Total	53 (76.8)	16 (23.2)	69 (100)
	RNFL (temporal)	Normal	32 (46.4)	16 (23.2)	48 (69.6)	< 0.05
		Thinning	6 (8.7)	15 (21.7)	21 (30.4)
		Total	38 (55.1)	31 (44.9)	69 (100)
	GCC	Normal	20 (29.0)	12 (17.4)	32 (46.4)	0.09
		Thinning	5 (7.2)	32 (46.4)	37 (53.6)
		Total	25 (36.2)	44 (63.8)	69 (100)

GCC: Ganglion cell complex; RNFL: Retinal nerve fiber layer.

Preoperatively, RNFL damage in all the quadrants was significantly associated with visual impairment (P < 0.05). Postoperatively, RNFL damage in the superior and inferior quadrants was significantly associated with visual impairment (P < 0.05). The mean GCC thickness, both preoperatively and postoperatively, was significantly associated with visual impairment (P < 0.05) (Table 5).

Table 5 Association of visual impairment with retinal nerve fiber layer and ganglion cell complex thickness, at baseline and final follow-up, mean ± SD.

		Mild visual impairment (6/6-6/18)	Moderate impairment (6/18-6/36)	Severe impairment to blind (6/36-3/60)	P value
Preoperative
RNFL thickness in micron	Superior	128.06 ± 65.91	98.17 ± 24.37	66.06 ± 18.34	< 0.01
	Inferior	123.41 ± 54.42	103.67 ± 24.42	64.76 ± 27.67	< 0.01
	Nasal	74.96 ± 34.56	59.67 ± 18.14	52.29 ± 20.57	< 0.05
	Temporal	55.89 ± 21.23	42.83 ± 7.81	42.7 ± 11.08	< 0.05
Mean GCC thickness in micron		70.5 ± 13.6	58 ± 10.02	55.29 ± 13.51	< 0.01
Postoperative (six-month follow-up)
RNFL thickness in micron	Superior	105.98 ± 29.84	95.25 ± 28.06	63.18 ± 17.95	< 0.01
	Inferior	102.61 ± 27.51	101 ± 38.84	61.54 ± 23.1	< 0.01
	Nasal	63.2 ± 15.28	69.75 ± 9.18	52.18 ± 21.64	0.08
	Temporal	53.2 ± 9.19	46.5 ± 8.18	47.54 ± 12.89	0.12
Mean GCC thickness in micron		69.9 ± 12.67	67.25 ± 16.98	47.54 ± 16.26	< 0.01

GCC: Ganglion cell complex; RNFL: Retinal nerve fiber layer.

The change in RNFL thickness with change in visual acuity was plotted against time. The superior, inferior, and nasal quadrants of RNFL showed reduced thickness up to 1.5 months, after which there was improvement, but the values did not reach the preoperative thickness. The temporal quadrant thickness did not show much deviation, and the final thickness was more than the preoperative value. The mean GCC thickness initially showed a small dip after surgery but gradually increased over 6 months and attained a value greater than the baseline thickness (Figures 3 and 4).

Figure 3 Trend of visual acuity (logMAR) and retinal nerve fibre layer thickness at baseline and follow-up visits. A: Superior; B: Inferior; C: Nasal; D: Temporal quadrants. VA: Visual acuity; RNFL: Retinal nerve fiber layer.

Figure 4 Trend of visual acuity (logMAR) and ganglion cell complex thickness at baseline and follow-up visits. VA: Visual acuity; GCC: Ganglion cell complex.

The final postoperative RNFL thickness positively correlated with the final visual acuity, which was significant for the superior and inferior quadrants (P < 0.05) (Figure 5). There was also a significant positive correlation of preoperative GCC thickness with preoperative visual acuity (P < 0.05) but the correlation with preoperative field defect was not significant (Figure 6) The final postoperative GCC thickness showed a significant positive correlation with the final visual acuity (r = -0.48) and residual field damage (r = -0.27) (P < 0.05) (Figure 7).

Figure 5 Correlation of final postoperative retinal nerve fiber layer thickness with visual acuity (logMAR). A: Superior; B: Inferior; C: Nasal; D: Temporal quadrants. VA: Visual acuity; RNFL: Retinal nerve fiber layer.

Figure 6 Correlation between variables at baseline. A: Preoperative ganglion cell complex (GCC) thickness with preoperative field defect (mean deviation); B: Preoperative GCC thickness with preoperative visual acuity (logMAR). VA: Visual acuity; GCC: Ganglion cell complex; MSD: Mean standard deviation.

Figure 7 Correlation of ganglion cell complex thickness and visual field defects at 6 months visit. A: Final postoperative ganglion cell complex (GCC) thickness with final postoperative field defect (mean deviation); B: Final postoperative GCC thickness with final postoperative visual acuity (logMAR). VA: Visual acuity; GCC: Ganglion cell complex; MSD: Mean standard deviation.

In univariate analysis, the Odds of having blindness at presentation in patients with an atrophic RNFL in superior, inferior and temporal quadrants were found to be 0.95 (95%CI: 0.92-0.98), 0.96 (95%CI: 0.93-0.98), 0.93 (95%CI: 0.87-0.99), and in those with an atrophic GCC was 0.95 (95%CI: 0.91-0.99) with P < 0.001, P < 0.001, P < 0.03, and P < 0.02, respectively. The Odds of blindness after 6 months of surgery in patients with an atrophic inferior quadrant of RNFL and GCC were found to be 0.96 (95%CI: 0.94-0.98) and 0.95 (95%CI: 0.91-0.99) (P-value of 0.003 and 0.03), respectively (Table 6).

Table 6 Univariate analysis of factors predictive of blindness.

Variables		Preoperative		Final postoperative
		Cor (95%CI)	P value	Cor (95%CI)	P value
Age		0.99 (0.93-1.06)	0.915	1.00 (0.92-1.08)	0.991
Duration of symptoms		0.96 (0.86-1.06)	0.45	0.98 (0.89-1.08)	0.764
Preoperative mean RNFL thickness	Superior	0.95 (0.92-0.98)	< 0.01	0.96 (0.94-0.99)	< 0.01
	Inferior	0.96 (0.93-0.98)	< 0.01	0.96 (0.94-0.98)	< 0.01
	Nasal	0.97 (0.93-1.00)	0.06	0.98 (0.95-1.01)	0.273
	Temporal	0.93 (0.87-0.99)	< 0.05	0.95 (0.88-1.01)	0.11
Preoperative mean GCC thickness		0.95 (0.91-0.99)	< 0.05b	0.95 (0.91-0.99)	< 0.05
Tumor volume		1.00 (0.99-1.00)	0.128	1.00 (0.99-1.00)	0.09
Duration of surgery		1.61 (0.95-2.75)	0.078	1.57 (0.83-2.99)	0.166
Surgical approach		3.23 (0.55-18.96)	0.194	4 (0.39-40.1)	0.239

GCC: Ganglion cell complex; RNFL: Retinal nerve fiber layer.

The preoperative RNFL (inferior quadrant) thickness and the final postoperative visual outcome were plotted in an empirical receiver operating characteristic (ROC) curve with area under the curve (AUC) = 0.8076. Considering a specificity of 13.6% and sensitivity of 50.0%, the cut-off value of RNFL (inferior quadrant) thickness beyond which blindness could be prevented was 73 µm. The preoperative GCC thickness and the final postoperative visual outcome were plotted in an empirical ROC curve with AUC = 0.754. Considering a specificity of 22.03% and sensitivity of 60.0%, the cut-off value of GCC thickness beyond which blindness may be prevented was hypothesised to be 58 µm (Figure 8).

Figure 8 Receiver operating characteristic curve showing cut-offs for blindness with respect. A: Preoperative retinal nerve fiber layer thickness (inferior quadrant); B: Preoperative ganglion cell complex thickness. ROC: Receiver operating characteristic.

DISCUSSION

The RNFL and GCC have demonstrated definitive prognostic values in predicting post-operative visual outcomes in perichiasmatic tumors[11]. Recent studies have evaluated the role of GCC as a biomarker for the early detection of compressive optic neuropathy[12-14]. Danesh-Meyer et al[15] was the first to document a retinal OCT on patients with chiasmal compression. A strong correlation was noted between the RNFL thickness and visual field defects. The changes were more significant in the horizontal than the vertical sector, and there was a stronger correlation in the temporal than in the nasal sector. The axons that are compressed first are primarily those of the nasal ganglion cells, and as the tumor expands, the temporal ganglion cells also get affected. These pathophysiological changes that occur based on location and tumor compression contribute to varying injury patterns to ganglion cell axons[16,17].

The Buyuktepe et al[18] study reported significant changes in peripapillary RNFL thickness in all quadrants because of transsynaptic or transneuronal retrograde degeneration in chiasmal and post-geniculate tumors. The survey by Cennamo et al[9] reported that non-compressing pituitary macroadenomas had a preoperative RNFL thickness of 102.6 ± 15.6 µm and 100.8 ± 13.6 µm in the superior and inferior quadrants, respectively. Kurian et al[19] stated that the preoperative RNFL was 89.02 µm, which was thicker than in other studies (73-88 μm). In our patients, RNFL thickness in the superior quadrant was 110.19 ± 60.94 µm and in the inferior quadrant was 107.25 ± 53.04µm, respectively. This might be due to the diversity in age, duration of disease, severity of chiasmal compression, and tumor heterogeneity in our cohort.

Many studies have reported that GCC+IPL thickness analysis has a higher sensitivity in the early diagnosis of the optic nerve damage in chiasmal compression[5-7]. Banc et al[20] demonstrated sectoral GC-IPL thickness changes in patients with intracranial tumors affecting the visual pathway. In the Yum et al[21] study, the preoperative average GCC thickness was 68.76 ± 10.4 µm. Tieger et al[8] reported a preoperative mean GCC thickness of 67 ± 9 µm. Shetty et al[22] reported a higher preoperative mean GCC thickness of 71.6 ± 17.2 µm. In our study, preoperative mean GCC thickness was 65.67 ± 14.9 µm which increased to 66.19 ± 15.6 µm after six months postoperatively. These results support the hypothesis that patients with intracranial lesions affecting the anterior visual pathway can demonstrate variable degrees of compression, resulting in a wide range of GCC and RNFL thicknesses[5,6,23].

This study’s preoperative RNFL and GCC thickness were associated with final visual outcome, similar to various other studies[8-10,15,17,24]. Danesh-Meyer et al[15] explained that the pRNFL thickness reflected the surviving functional axons, and thereby, it could predict visual field improvement after decompression surgery. In their study of PA cases, Jacob et al[17] reported that the odds of complete visual field recovery were multiplied by 1.29 for each 1-µm increment in mean RNFL thickness. Lee et al[24] reported that the odds of complete visual field recovery were multiplied by 1.03 for each 1-µm increase in mean pRNFL thickness.

The study by Tieger et al[8] and Yum et al[21] reported nasal GCIPL thinning in patients with chiasmatic compression, even without field changes[8,21]. Ohkubo et al[25] and Monteiro et al[26] showed a positive correlation between macular GCC thickness and established field defects in patients with chiasmatic compression. Shetty et al[22] also reported a strong correlation between the nasal sector of GC-IPL and visual field defects in patients of PA. In our study, the preoperative and postoperative GCC thickness positively correlated with preoperative visual acuity and field defect, which were statistically significant (P < 0.5). These findings highlight that GCC is a strong biomarker to detect and prognosticate the visual outcome of brain tumor patients undergoing surgical decompression.

Several studies reported variable visual recovery ranging from 44% to 93%[4,27-29]. The study by Danesh-Meyer et al[10] reported improvement in visual acuity to more than 6/12 in 97% of eyes with thick RNFL and 72% with thin RNFL in 6 weeks (P = 0.02). A more extensive study by Wang et al[30] documented that the odds of visual recovery were associated with inferior RNFL, similar to our research. All the visual improvement was noted in the first six weeks following surgery, with no further changes later on. Kurian et al[19] reported no significant difference in RNFL thickness among the visual recovery group vs those without visual recovery. Jacob et al[17] concluded that the inferior quadrant of RNFL was a (OR: 6.31) predictor for field recovery in PA patients. On the contrary, Garcia et al[31] found that nasal RNFL thickness was a strong predictive factor (OR: 1.56) for visual field damage in such patients. Our study showed that patients with an atrophic GCC and RNFL thickness in the inferior quadrant at baseline had a significant chance of vision loss, 6 months after surgery.

The study by Kurian et al[19] reported RNFL thickness of 80 µm with 73.1% sensitivity and 62.5% specificity, respectively, as predictors for postoperative visual field recovery. Garcia et al[31] in a retrospective study using time domain OCT reported nasal RNFL thickness of 68.50 µm as being predictive, but with poor sensitivity and specificity of 61% and 50%, respectively. In our study, a preoperative RNFL (inferior) thickness of 73 µm (specificity of 13.6% and sensitivity of 50%), and a preoperative GCC thickness of 58 µm (specificity of 22% and sensitivity of 60%) was hypothesized as the cut-off value beyond which blindness can be prevented.

The strengths of the study were its prospective nature with six months of follow-up, including all subsets of peri-chiasmatic tumors, use of spectral domain OCT, comparative analysis of GCC, RNFL thickness and visual fields with determination of a cut-off value for vision preservation. While primary hypotheses were prespecified, the study’s secondary analyses involved multiple comparisons without universal adjustment. However, post-hoc corrections validated the main conclusions, suggesting that Type I error inflation was unlikely to alter clinical interpretations. The limitations include a small sample size and a lack of long-term follow-up. Hence, further studies are necessary with larger populations and longer follow-ups for validation and generalizability of data.

CONCLUSION

OCT can be used to quantify axonal loss, predict visual recovery, and prognosticate patients with peri-chiasmatic tumors. Preoperative GCC and RNFL thickness can objectively predict postoperative visual outcome. It can be utilized as a tool for the early detection of visual pathway injury before significant field loss. Early surgical decompression with a relatively preserved RNFL might save vision.