Early diagnosis of primary open-angle glaucoma using isolated-check visual evoked potential with blue-on-yellow perimetry

Abstract

Background

Early detection of primary open-angle glaucoma (POAG) is crucial, as current diagnostic methods often miss early-stage damage. Combining blue-on-yellow perimetry with isolated-check visual evoked potential (Ic-VEP) could enhance the sensitivity and specificity of early POAG detection.

Objective

To evaluate the diagnostic efficacy of Ic-VEP combined with blue-on-yellow perimetry for early POAG detection.

Methods

This study included 66 POAG patients and 35 healthy controls, all of whom underwent comprehensive ophthalmologic assessments, including intraocular pressure (IOP), fundus examination, and optical coherence tomography (OCT) to measure retinal nerve fiber layer (RNFL) and ganglion cell complex (GCC) thickness. Ic-VEP and Humphrey 24 − 2 blue-on-yellow perimetry were performed to assess the sensitivity, specificity, and ROC curve for early POAG diagnosis. The correlation between Ic-VEP. results, GCC thickness, and visual field loss was also analyzed.

Results

Ic-VEP demonstrated 74% sensitivity and 91% specificity for detecting POAG, with an area under the ROC curve (AUC) of 0.785, indicating reliable diagnostic performance. The Ic-VEP signal-to-noise ratio (SNR) showed significant correlation with the mean deviation (MD) of blue-on-yellow perimetry and GCC thickness (p < 0.05). The Kappa coefficient for consistency between Ic-VEP and blue-on-yellow perimetry was 0.226 in early-stage POAG, increasing to 0.672 in moderate to severe stages suggesting enhanced diagnostic value in later stages.

Conclusion

Combining Ic-VEP with blue-on-yellow perimetry shows promise for enhancing the early diagnosis of POAG, with Ic-VEP’s high specificity (91%) complementing the sensitivity of perimetry. This approach could lead to earlier diagnosis and improved patient outcomes.

Introduction

Primary open-angle glaucoma (POAG) is a progressive and irreversible optic neuropathy that is one of the leading causes of blindness worldwide. This condition primarily affects the retinal ganglion cells (RGCs) and the optic nerve fiber layer, resulting in gradual visual field loss if not detected early and managed appropriately. The macular region contains the highest density of RGCs, and early glaucomatous damage often affects the macular ganglion cells before observable glaucomatous optic nerve head (ONH) changes (e.g., rim thinning or notching) or visual field defects [1–3]. RGCs are primarily classified into M (magnocellular) and P (parvocellular) cells, with M cells, which have larger cell bodies, being more susceptible to early damage in glaucoma. While magnocellular (M) cells, which are more prevalent in the peripheral retina, are particularly vulnerable in early glaucoma, the central macular region, assessed by blue-on-yellow perimetry and Ic-VEP, also contains M-cell pathways that exhibit early functional deficits [4–6].

Early and accurate detection of POAG is crucial for preventing irreversible vision loss. Traditional diagnostic approaches, such as optical coherence tomography (OCT), focus on measuring the thickness of the retinal nerve fiber layer (RNFL) and the macular ganglion cell complex. However, these methods may not detect subtle early changes, and functional impairments can precede structural loss. Recent studies have demonstrated that glaucoma patients may experience specific color vision deficits, particularly in the blue-yellow spectrum [7], even before conventional visual field defects are noticeable. This makes blue-on-yellow perimetry, a psychophysical test that uses a blue stimulus on a yellow background to assess short-wavelength-sensitive pathways, an effective tool in glaucoma detection [8–10]. Despite its advantages, blue-on-yellow perimetry is influenced by subjective factors, such as patient attention and cooperation, limiting its reliability [11]. As an objective alternative, the isolated-check visual evoked potential (Ic-VEP) test measures the electrical response of the visual cortex to stimuli that specifically target the M-cell pathway. This test has shown potential for detecting functional damage early in the disease process [12–14]. Combining Ic-VEP with blue-on-yellow perimetry offers a comprehensive approach for early POAG detection by assessing functional changes in the visual pathway, while optical coherence tomography (OCT) complements these tests by evaluating structural changes in the retinal nerve fiber layer (RNFL) and ganglion cell complex (GCC). This study aims to evaluate the combined diagnostic value of Ic-VEP and blue-on-yellow perimetry in patients with primary open-angle glaucoma, providing new insights into early detection strategies for this sight-threatening disease.

Methods

Study enrollment

Study introduction

Participants with primary open-angle glaucoma (POAG) and healthy controls were recruited from Shenzhen People’s Hospital and Shenzhen Eye Hospital between 2017 and 2018 for a controlled analysis. Informed consent was obtained from all participants, ensuring ethical compliance.

Study objective

This study aimed to evaluate the diagnostic performance of isolated-check visual evoked potential (Ic-VEP) and blue-on-yellow perimetry in detecting POAG across its clinical spectrum. Medication use, including IOP-lowering drugs (e.g., prostaglandin analogs, beta-blockers, or carbonic anhydrase inhibitors), was documented but not controlled to reflect real-world clinical conditions.

POAG classification

POAG was classified as early, moderate, or severe using the Hodapp-Anderson-Parrish (HAP) criteria [15, 16]. The study included 30 eyes with early POAG (mean deviation ≥ −6 dB) and 36 eyes with moderate to severe POAG (mean deviation < −6 dB).

Ophthalmic examinations

All participants underwent comprehensive ophthalmic examinations, including best-corrected visual acuity (BCVA), slit-lamp anterior segment examination, fundus examination (including lens assessment), gonioscopy, Goldmann applanation tonometry for intraocular pressure (IOP), blue-on-yellow perimetry, isolated-check visual evoked potential (Ic-VEP), and optical coherence tomography (OCT). Astigmatism was limited to ≤ 2.0 diopters to minimize optical distortions during blue-on-yellow perimetry and Ic-VEP testing.

General inclusion criteria

General inclusion criteria for all participants were: age 20–80 years, BCVA better than 0.1 (logarithm of the minimum angle of resolution, logMAR), spherical refraction within − 6 to + 3 diopters, clear ocular media, normal anterior segment on slit-lamp examination, and pupil size between 2.5 and 4.0 mm.

POAG inclusion criteria

Eyes were classified into early POAG, moderate to severe POAG, and control groups based on the HAP criteria for staging and the 2014 Chinese Glaucoma Consensus for diagnosis [15, 16]. Inclusion criteria for the POAG group were: (1) characteristic glaucomatous fundus changes, such as retinal nerve fiber layer (RNFL) defects or optic nerve head (ONH) rim thinning/notching, with or without visual field damage; (2) an open anterior chamber angle confirmed by gonioscopy; and (3) exclusion of other causes of elevated IOP, fundus, or visual field damage. Patients with normal-tension glaucoma (NTG, IOP ≤ 21 mmHg) or treated POAG with stabilized IOP (≤ 21 mmHg) were included to reflect the clinical spectrum of POAG [15, 16].

Exclusion criteria

Exclusion criteria for all participants included a history of mental health conditions, alcohol or drug abuse, or inability to cooperate with functional testing (e.g., perimetry or Ic-VEP) to ensure reliable test results. Additional exclusions were primary angle-closure glaucoma, secondary glaucoma, non-glaucomatous optic nerve involvement, conditions affecting vision or macular thickness, moderate or severe cataracts or other opacities interfering with fundus observation, difficult fixation, pupil size < 2.0 mm, recent eye infection (within 3 months), history of ocular trauma or surgery, or prolonged systemic or local corticosteroid therapy.

Control group criteria

Participants in the control group met the following criteria: (1) no family history of glaucoma; (2) IOP < 21 mmHg; (3) normal visual field (MD ≥ −2 dB, no significant defects per HAP criteria); (4) normal ONH (cup-to-disc ratio ≤ 0.4, no rim thinning/notching); and (5) normal RNFL thickness on OCT (no sectors in red or yellow zones, Spectralis, Heidelberg Engineering). Subjects with mental health conditions or alcohol/drug abuse were also excluded from the control group to ensure reliable performance during functional tests, as these conditions could impair attention, fixation, or response accuracy in blue-on-yellow perimetry and Ic-VEP.

Blue-on-yellow perimetry

Participants requiring near vision correction for perimetry were provided with appropriate refractive lenses and examined using the Humphrey 750 perimeter (Carl Zeiss Meditec, Inc., USA) with a 24 − 2 blue-on-yellow SITA-Standard protocol, selected for its standardized methodology and widespread clinical use. The parameters were: (1) a hemispherical field-of-view screen with a fixation distance of 330 mm; (2) a Goldmann V projective light source (blue light, 440 nm wavelength, viewing area 0.43–2 mm²); (3) a broad-spectrum yellow background light (530 nm wavelength); (4) visual target stimulation duration of 100–200 ms; (5) automatic physiological blind spot surveillance combined with micro-video monitoring; and (6) a testing range of 30 degrees covering 76 points in the central visual field, with 6-degree spacing between points. Results were considered reliable if fixation loss, false-positive rate, and false-negative rate were each ≤ 10%. Participants with unreliable results were trained and retested until meeting reliability standards [9, 10]. Mean sensitivity (MS) was recorded for statistical analysis. Blue-on-yellow visual field abnormalities were defined as: (1) three or more adjacent points with a light sensitivity decrease of ≥ 7 dB; (2) two or more adjacent points with a decrease of ≥ 7 dB in one point and ≥ 10 dB in another; (3) a difference in light sensitivity of ≥ 10 dB between two or more adjacent points above or below the nasal horizontal meridian; or (4) three or more adjacent points with a sensitivity deviation probability of < 5%. These criteria excluded the four points immediately above and below the physiological blind spot and the visual field edge [15, 16].

Isolated-Check Visual Evoked Potential (Ic-VEP)

Ic-VEP was conducted using the Cordia electrophysiological instrument (Huzhou, China) with an isolated-check pattern targeting the magnocellular pathway in the macular region to assess early functional deficits in POAG. During Ic-VEP testing, the participant’s gaze was aligned with the center of the stimulus pattern (approximately 10–15 degrees) on the Cordia electrophysiological instrument (Huzhou), positioned 70 cm from the eye. Fixation was monitored to ensure the visual axis remained centered on the stimulus, minimizing eye movement artifacts and ensuring accurate measurement of the magnocellular pathway response [14, 17, 18]. The stimulation protocol lasted 96 s, divided into eight cycles, each initiated with a voice prompt. Cycles with significant noise or eye-movement artifacts were automatically rejected by the system, and measurements were repeated until reliable. Two examinations were performed per participant, with a minimum 10-minute interval, and only reliable results were included in the analysis. A signal-to-noise ratio (SNR) ≥ 1 indicated a measurable brain response exceeding the 95% confidence interval of baseline noise, calculated as the ratio of the fundamental component’s amplitude to the radius of a confidence circle. An SNR < 1 suggested potential magnocellular pathway impairment, indicative of glaucomatous damage, and prompted repeated checks to confirm reliability.

Statistical analyses

SPSS 24.0 was used to describe the results. Chi-square test was used for qualitative data. Mean standard deviation (xs) was used for quantitative data. Differences in characteristics between glaucoma patients and normal subjects were compared using the independent sample t test. The difference was statistically significant with P < 0.05.

Result

General profile of subjects

The study included 66 patients (66 eyes, one eye per patient) with POAG, comprising 30 with early POAG and 36 with moderate to severe POAG, and 35 healthy controls (35 eyes). Table 1 summarizes the demographic and clinical characteristics. No significant differences were observed in age, gender, or spherical equivalent between groups (p > 0.05). However, the cup-to-disc ratio was significantly higher in both POAG groups compared to controls (p < 0.05). The moderate to severe POAG group had a mean IOP of 17.4 ± 4.3 mmHg, indicating that some patients had IOP > 21 mmHg at diagnosis, which was stabilized through treatment, or had normal-tension glaucoma, consistent with the inclusion criteria (see Methods).

Table 1.

Demographic and clinical characteristics of participants with early POAG, moderate to severe POAG, and healthy controls

Variables	Early POAG patients (n = 30)	Moderate to Severe POAG (n = 36)	Control group (n = 35)
Age(years)	36 ± 5.3	43 ± 8.3	25 ± 5.6
Gender(male/female)	17/13	24/12	16/19
Spherical eauivalent(D)	−2.26 ± 0.98	−1.86 ± 0.54	−1.26 ± 0.51
BCVA (logMAR, mean ± SD)	0.87 ± 0.18*	0.85 ± 0.21*	0.92 ± 0.13
IOP (mmHg, mean ± SD)	16.2 ± 3.8	17.4 ± 4.3*	15.7 ± 2.6
Cup-to-Disc Ratio	0.53 ± 0.15*	0.82 ± 0.14*	0.31 ± 0.14

Compared with the control group, * P < 0.05

Sensitivity and specificity of Ic-VEP in the detection of primary glaucoma

The diagnostic performance of Ic-VEP in detecting primary open-angle glaucoma (POAG) was evaluated using receiver operating characteristic (ROC) curve analysis and contingency tables. The results demonstrated that Ic-VEP exhibited a sensitivity of 74.2% and specificity of 91.4% for distinguishing all POAG cases (n = 66) from healthy controls (n = 35), with a positive predictive value (PPV) of 94.2% and negative predictive value (NPV) of 65.3% (Table 2).

Table 2.

Diagnostic performance of Ic-VEP in detecting POAG compared to controls

Group	Sensitivity (%)	Specificity (%)	PPV (%)	NPV (%)
All POAG (n = 66)	74.2	91.4	94.2	65.3
Early POAG (n = 30)	70.0	91.4	87.5	78.0
Moderate to Severe POAG (n = 36)	86.1	91.4	91.2	86.5

Ic-VEP diagnostic performance for POAG vs. controls (n = 35), showing sensitivity, specificity, PPV (Positive Predictive Value), and NPV (Negative Predictive Value) for all POAG (n = 66), early POAG (n = 30), and moderate to severe POAG (n = 36). AUC for all POAG: 0.785 (95% CI: 0.606–0.965, p < 0.05; Fig. 1)

When stratified by disease severity, Ic-VEP showed a sensitivity of 70.0% in early POAG (n = 30) and 86.1% in moderate to severe POAG (n = 36), while maintaining a consistent specificity of 91.4% across all stages. The area under the ROC curve (AUC) for Ic-VEP was 0.785 (95% CI: 0.606–0.965, p < 0.05; Fig. 1), indicating moderate diagnostic accuracy. The higher sensitivity in advanced stages suggests that Ic-VEP may better capture functional deficits as glaucomatous damage progresses.

Fig. 1.

Ic-VEP ROC curve in diagnosis of primary open angle glaucoma

Notably, the Kappa statistic revealed weak agreement (κ = 0.226, p < 0.05) between Ic-VEP and blue-on-yellow perimetry in early POAG, which improved to substantial agreement (κ = 0.672, p < 0.05) in moderate to severe stages. This discrepancy suggests that Ic-VEP may detect subtle macular dysfunction earlier than perimetry, while both modalities align more closely in advanced disease.

The consistency of Ic-VEP qualitative examination with GCC and damage of blue-yellow visual field

The consistency between Ic-VEP results and structural (GCC thickness) or functional (blue-on-yellow perimetry) measures was assessed using Kappa statistics.

In early POAG, Ic-VEP showed moderate agreement with GCC damage (κ = 0.507, p < 0.05; Table 3), with 15 true-positive and 8 true-negative cases. In advanced POAG, agreement improved (κ = 0.620, p < 0.01; Table 4), with 30 true-positive and 3 true-negative cases, indicating stronger concordance as structural damage became more pronounced.

Table 3.

Comparison of the damage consistency between Ic-VEP with GCC in macular region in early POAG group

Ic-VEP	GCC
	Positive	Negative
Positive	15	3
Negative	4	8
Sum up	19	11

Table 4.

Comparison of damage consistency between Ic-VEP with GCC in macula in advanced POAG group

Ic-VEP	GCC
	Positive	Negative
Positive	30	1
Negative	2	3
Sum up	32	4

For functional consistency, Ic-VEP and blue-on-yellow perimetry exhibited moderate agreement in early POAG (κ = 0.552, p < 0.05; Table 5) and advanced POAG (κ = 0.442, p < 0.01; Table 6). The higher Kappa values in early disease for perimetry (vs. GCC) suggest that functional tests may correlate better at initial stages, while structural changes become more evident later.

Table 5.

Comparison of the damage consistency between Ic-VEP with blue-yellow visual field in early POAG group

Ic-VEP	Blue-yellow visual field
	Positive	Negative
Positive	17	2
Negative	4	7
Sum up	21	9

Table 6.

Comparison of damage consistency between Ic-VEP with blue-yellow visual field in advanced POAG group

Ic-VEP	Blue-yellow visual field
	Positive	Negative
Positive	30	1
Negative	3	2
Sum up	33	3

Discussion

This study demonstrated that Ic-VEP, combined with blue-on-yellow perimetry, offers a sensitivity of 74% and specificity of 91% for detecting primary open-angle glaucoma (POAG), with an AUC of 0.785 (95% CI: 0.606–0.965, p < 0.05). These findings are consistent with prior studies highlighting the high specificity of Ic-VEP for detecting functional deficits in glaucoma [12, 14, 19]. In China alone, there are an estimated 22.1 million glaucoma patients aged 40 or older, with 12 million suffering from open-angle glaucoma [20]. Open-angle glaucoma is marked by a range of clinical signs, including elevated intraocular pressure (IOP), optic nerve head changes, an open anterior chamber angle, and progressive visual field loss. Despite the presence of early visual impairment in primary open-angle glaucoma (POAG), it is often overlooked due to subtle clinical symptoms. Previous studies have reported that in cases of glaucoma, approximately 50% of retinal nerve fibers can become damaged before significant visual field loss is detectable [1]. Thus, early and accurate diagnosis is crucial for effective disease management.

Standard automated perimetry (SAP), commonly known as white-on-white perimetry (W/WP), is widely used for glaucoma diagnosis but has significant limitations in detecting early-stage disease [13]. This is because SAP primarily detects changes in visual fields after substantial damage to the optic nerve has occurred. For example, Quigley’s studies have shown that typical visual field deficits are evident only after 40–50% of retinal ganglion cells (RGCs) are lost [21]. Additionally, a decrease in retinal sensitivity by 5 dB corresponds to a loss of approximately 20% of retinal ganglion cells [22]. This underscores the need for more sensitive diagnostic methods capable of detecting glaucomatous damage earlier.

Retinal color vision relies on three types of cone cells—red, green, and blue—which are part of distinct visual pathways. Research has found that blue cone cells and their neural pathways tend to suffer damage earlier in glaucoma and ocular hypertension than other cones [23, 24]. This vulnerability is likely due to the relatively low density of blue cones and the thicker axons of nerve fibers transmitting blue cone signals. The blue-yellow perimetry (B/YP) test is designed to combine color vision assessment with visual field testing, using blue light as the stimulus and a yellow background to test retinal sensitivity [7]. The yellow background neutralizes the sensitivity of the medium- and long-wavelength cones (red and green) and rod cells, allowing for the selective isolation of the short-wavelength-sensitive (blue) pathways, which comprise about 6% of RGCs [7] This makes the B/YP test highly sensitive and specific for detecting glaucomatous visual field loss [8].

Blue-on-yellow perimetry offers significant advantages over standard white-on-white perimetry by selectively targeting the short-wavelength-sensitive (blue) pathways, which are mediated by approximately 6% of retinal ganglion cells and are particularly vulnerable in early glaucoma [10, 25]. The use of a blue stimulus (440 nm) on a yellow background (530 nm) isolates these pathways, neutralizing the sensitivity of medium- and long-wavelength cones (red and green) and rod cells. This enhances the detection of early functional deficits, often before significant structural loss is evident on OCT or standard perimetry, making it a valuable tool for early POAG diagnosis. Visual signals are transmitted by different types of RGCs, broadly categorized based on their size into large M (magnocellular) cells and smaller P (parvocellular) cells [6]. Glaucoma primarily leads to the irreversible loss of RGCs, with early-stage damage often involving M cells due to their peripheral location, which correlates with the pattern of early visual field changes seen in glaucoma [5]. Recognizing the early vulnerability of M cells, the Cordia visual electrophysiology instrument uses grating pattern stimulation to selectively assess M-cell function. This approach could offer valuable insights for the early diagnosis of open-angle glaucoma. In our study, the sensitivity and specificity of the isolated-check visual evoked potential (Ic-VEP) for detecting primary open-angle glaucoma were 74% and 91%, respectively. The area under the ROC curve was 0.785 (95% confidence interval: 0.606–0.965, p < 0.05), demonstrating its utility in diagnosing primary open-angle glaucoma [14]. Variations in the reported positive rates of Ic-VEP in detecting primary open-angle glaucoma across studies can be attributed to differences in the definition of POAG, variations in the operator’s experience and equipment used, and the levels of strain and fatigue in patients [17].

Furthermore, when assessing POAG classification, the consistency between Ic-VEP signal-to-noise ratio (SNR) and mean deviation (MD) values showed a Kappa value of 0.226 for early-stage POAG, which increased to 0.672 in moderate and advanced stages. This indicates that Ic-VEP may have lower concordance with blue-yellow perimetry in early POAG but shows improved agreement in later stages. The study found that Ic-VEP SNR was more affected in early POAG cases detected by blue-yellow perimetry; thus, patients with mild visual field defects on perimetry may still show significant changes in Ic-VEP results, indicating intermediate stages of glaucoma. The discrepancies between the Ic-VEP and blue-yellow perimetry results in POAG diagnosis may be due to: (1) Ic-VEP’s ability to monitor macular center activity and detect subtle changes in electrical conduction in retinal ganglion cells, which increases its sensitivity to macular visual field changes; (2) Ic-VEP SNR demonstrating moderate changes in early POAG without significant alterations in the M and P cells of retinal ganglion cells. Moreover, early damage in POAG may affect the neurofibrillar layer (NGF), where RGC axons are embedded. Loss of axons in this layer, even if total NGF thickness remains unchanged, can be masked by intact axons in the superficial layers, making these changes difficult to detect with fundus photography or standard visual field exams [5]. At this stage, Ic-VEP can detect abnormal electrical conduction in RGCs, proving more sensitive than conventional visual field tests [26].

In conclusion, while perimetry remains a cornerstone for glaucoma evaluation, its accuracy can be compromised by psychological, physiological, and operator-related factors. The isolated-check visual evoked potential (Ic-VEP) offers a more objective and reliable measure of visual function, demonstrating considerable potential for the early detection of primary open-angle glaucoma (POAG). This approach is crucial as it allows for earlier intervention before substantial visual field loss occurs, potentially preventing irreversible blindness.

In conclusion, the combination of Ic-VEP and blue-on-yellow perimetry demonstrates promising diagnostic performance for early POAG, with a sensitivity of 74% and specificity of 91%, offering a robust diagnostic combination for detecting functional deficits in the central visual field before significant visual field loss is evident on standard perimetry. The high specificity of Ic-VEP, combined with the sensitivity of blue-on-yellow perimetry, suggests a complementary approach for detecting functional deficits in the early stages of glaucoma.

A limitation of this study is that both blue-on-yellow perimetry (24 − 2 protocol) and Ic-VEP primarily assess the central visual field (approximately 24 and 10–15 degrees, respectively), which may not fully capture peripheral magnocellular damage prevalent in early glaucoma. However, the central retina, including the macula, is also affected early in POAG, and our findings demonstrate that Ic-VEP and blue-on-yellow perimetry are sensitive to these changes, with Ic-VEP showing a specificity of 91%.

Future studies should incorporate peripheral field testing, such as the 30 − 2 perimetry protocol, to evaluate the full extent of glaucomatous damage. Despite these promising findings, larger multi-center studies with diverse populations, standardized protocols, and advanced imaging integration are needed to further validate the clinical efficacy of Ic-VEP and optimize its use in clinical practice for the early and accurate management of POAG [27].

Authors’ contributions

Q.L. supervised the study and is the first and corresponding author. D.L. and M.Z. were involved in the conception and design of the study, performed the experiments, and participated in data acquisition. Both D.L. and M.Z. contributed equally to the design and execution of the experiments, as well as data collection and analysis. Q.L. led the data interpretation and provided critical guidance throughout the study.All authors participated in drafting and revising the manuscript for important intellectual content. Q.L. provided final approval of the version to be submitted. All authors agree to be accountable for the accuracy and integrity of the work presented.

Funding

The study was funded by Science, Technology and Innovation Commission of Shenzhen (JCYJ20220530151805011) and (JCYJ20240813104304007).

Data availability

All data generated or analysed during this study are included in this published article.

Declarations

Ethics approval and consent to participate

The study was conducted following the tenets of the Declaration of Helsinki. The study involved human participants and was approved by the Institutional Review Board and Ethics Committee of Shenzhen People’s Hospital and Shenzhen Eye Hospital (2024SZES00237). Written informed consent was obtained from the patient for participation in this study.

Consent for publication

Not Applicable.

Competing interests

The authors declare no competing interests.