Article Text

Download PDFPDF
Association of macular OCT and OCTA parameters with visual acuity in glaucoma
  1. Jo-Hsuan Wu,
  2. Sasan Moghimi,
  3. Takashi Nishida,
  4. Vahid Mohammadzadeh,
  5. Alireza Kamalipour,
  6. Linda M Zangwill,
  7. Robert N Weinreb
  1. Hamilton Glaucoma Center, Shiley Eye Institute and Viterbi Family Department of Ophthalmology, University of California, San Diego, La Jolla, California, USA
  1. Correspondence to Dr Robert N Weinreb, Hamilton Glaucoma Center, Shiley Eye Institute and Department of Ophthalmology, University of California at San Diego Department of Ophthalmology at the Shiley Eye Institute, La Jolla, CA 92093, USA; rweinreb{at}


Background/aims To investigate the association of macular optical coherence tomography (OCT)/OCT angiography (OCTA) parameters with visual acuity (VA) in glaucoma.

Methods 144 pseudophakic primary open-angle glaucoma eyes were included. Foveal (fVD), parafoveal (pf), perifoveal (perifVD) and whole-image vessel densities (wiVD) of superficial and deep layers, and their corresponding ganglion cell complex (GCC) thicknesses were obtained from OCTA 6×6 mm2 macula scans. Foveal avascular zone (FAZ) area, FAZ circumference and foveal density-300 (FD300) were measured. Correlation between OCT/OCTA parameters and Logarithm of the Minimum Angle of Resolution VA (logMAR VA) in early and moderate-advanced glaucoma was evaluated with age and Signal Strength Index-adjusted mixed models. Area under receiver operating characteristic (AUC) was used to evaluate discriminative power of OCT/OCTA for decreased VA (<20/25).

Results In early glaucoma (80 eyes), no parameter correlated with VA. In moderate-advanced glaucoma (64 eyes), greater FAZ area (β=0.228) and circumference (β=0.063) correlated with worse VA (p<0.05), but not FD300. fThinner sectoral and global GCC was associated with worse VA (β=0.002–0.003, p<0.05), except for inferior hemifield perifGCC and wiGCC. For VD, lower superior hemifield superficial perifVD and wiVD (β=0.007–0.008) and deep fVD (β=0.004) correlated with worse VA (p<0.05). OCT/OCTA parameters showed modest ability to discriminate decreased VA, with the superior hemifield performing better than the inferior hemifield. In early glaucoma, GCC and VD showed similar discrimination (AUC=0.67–0.77). In moderate-advanced glaucoma, fGCC and pfGCC yielded higher AUC (0.75–0.81) than VD (AUC=0.63–0.72).

Conclusions Some macular OCT/OCTA parameters were associated with VA in moderate-advanced, but not early glaucoma. These structural parameters may help identify glaucoma patients with impaired vision and reduced quality of life.

Trial registration number NCT00221897.

  • Glaucoma
  • Vision

Data availability statement

Data are available on reasonable request. The datasets generated and/or analysed during the current study are available from the corresponding author on reasonable request.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

What is already known on this topic

  • Although pertinent to our quality of life, visual acuity (VA) is often overlooked in glaucoma, and information about its relationship with other structural parameters is scarce.

What this study adds

  • Some macular optical coherence tomography (OCT)/OCT angiography (OCTA) parameters were associated with VA in moderate-advanced, but not early glaucoma. Most OCT/OCTA parameters showed modest discriminative power for glaucoma eyes with decreased VA, with parafoveal ganglion cell complex showing the best overall discrimination.

How this study might affect research, practice or policy

  • Through structural examination by OCT/OCTA, clinicians might be able to identify glaucoma patients at risk of impaired VA.


An accurate and timely evaluation of glaucomatous damage relies on both functional and structural examinations, and optical coherence tomography (OCT) and OCT angiography (OCTA) are increasingly used for diagnosing and monitoring glaucoma.1 2 Since macular damage directly affects central vision,3 clinical relevance of macular structural parameters have been particularly studied. Both macular ganglion cell complex (GCC) thinning and macular vessel density (VD) loss were found to correlate with glaucomatous visual field (VF) loss.4–8 Similar association has been found between some metrics of the foveal avascular zone (FAZ) and central VF defect,9 10 indicating these macular OCT and OCTA parameters are useful in evaluating glaucomatous vision impairment.

While VF loss is more characteristic of glaucoma, visual acuity (VA) is another important functional parameter affecting our vision-related quality of life (VrQOL).11–13 Since significant VA impairment usually does not present until later stages in glaucoma,14 it is often overlooked in the management of glaucoma. Nevertheless, prior studies have shown even mild-to-moderate decline in VA can result in decreased VrQOL and daily function,11 12 15 indicating the clinical relevance of this measure. Additionally, VA was found to have a significant effect on VrQOL in patients with mild glaucomatous damage in a recent study.12 This finding further suggests a possible role of VA in the early stages of glaucoma, which was not attended to previously.

Information about the relationship between structural changes and VA decline in glaucoma is scarce, although some studies have reported a correlation between VA and central VF.16 17 One study investigated the correlation between retinal thicknesses and VA,18 and found a stronger association in glaucoma eyes with more severe disease. Another study examined the relationship between VA and OCTA parameters, and found macular VD to be most promising in discriminating decreased VA; however, only advanced glaucoma eyes were evaluated.19 Since OCTA has demonstrated a good ability to detect glaucomatous change in early glaucoma, that is, non-inferior to OCT, whether this holds true for the detection of VA decline is also of interests.

Considering that decreased VA directly affects the patients’ VrQOL, it is possible that macular OCT and OCTA parameters may help identify glaucoma eyes at risk of impaired VA. In this study, we investigated the association between macular OCT and OCTA parameters with VA in eyes with different severities of glaucomatous damage.


Participants from the Diagnostic Innovations in Glaucoma Study (DIGS, details described previously20 21) meeting the below inclusion criteria were included in this cross-sectional study. Written informed consent was obtained from all participants. Briefly, all DIGS subjects underwent annual comprehensive ophthalmic examination in both eyes with slit-lamp biomicroscopy, dilated fundus examination, best-corrected VA, Goldmann applanation tonometry, and stereoscopic optic disc photography, and semiannual examination of standard automated perimetry (SAP), intraocular pressure (IOP) measurement and OCTA/OCT imaging. Gonioscopy and ultrasound pachymetry were performed at the first visit. Other demographic information, including age, race, systemic medical history, blood pressure and medication use, was also collected.

Inclusion criteria for this study were as follows: (1) age older than 18 years, (2) a diagnosis of primary open-angle glaucoma (POAG), (3) pseudophakic status. Exclusion criteria were as follows: (1) history of ocular trauma, (2) coexisting retinal pathologies, (3) non-glaucomatous optic neuropathy, (4) uveitis, (5) axial length >27 mm. Participants diagnosed with Parkinson’s disease, Alzheimer’s disease, dementia, or a history of stroke were also ineligible.

POAG was defined as eyes showing repeatable and reliable abnormal SAP results (fixation losses and false negatives ≤33% and false positives ≤33%) using the 24–2 Swedish Interactive Thresholding Algorithm with either a pattern SD outside the 95% normal limits or a glaucoma hemifield test result outside the 99% normal limit. Glaucoma severity was classified as early if the 24–2 VF mean deviation (MD) was greater than −6 dB, and moderate-severe if < −6 dB.

OCTA and spectral-domain OCT

OCTA and spectral domain-OCT imaging with the Avanti Angiovue system (Optovue, Fremont, California, USA) was performed on all patients,22 and non-HD 6 mm × 6 mm (304 A scans in each B-scan and 304-B scans acquired) macula scans centred on the fovea were acquired. The OCT/OCTA images were acquired simultaneously, and the OCT-based thickness analyses and OCTA-based vascular analyses were calculated from the same scan slab. The Angiovue software (V.2018.1.0.43) performed automatic segmentation with registration of the analysed regions,23 with the superficial VD measured from the internal limiting membrane to 10 µm above the inner plexiform layer, and the deep VD measured from 10 µm above the IPL to 10 µm below the outer plexiform layer. Automated projection artefacts removal was performed for calculation of VD in the deep layer.

The VD was calculated as the percentage of measured area occupied by flowing blood vessels. The following VD parameters calculated from the macula scans centred on the fovea were analysed in this study (figure 1A): (1) foveal VD (fVD), calculated from the area of a 1- mm diameter circle centred on the fovea; (2) parafoveal VD (pfVD), calculated from the annular region with an inner diameter of 1 mm and an outer diameter of 3 mm centred on the fovea; (3) perifoveal VD (perifVD), calculated the annular region with an inner diameter of 3 mm and an outer diameter of 6 mm centred on the fovea and (4) whole-image VD (wiVD), calculated from the entire macula scan.

Figure 1

Regions corresponding to the (A) foveal, parafoveal and perifoveal measurements and (B) foveal avascular zone (FAZ) and foveal density 300 (FD300) on the 6 mm × 6 mm macula scans.

Thickness measurements of GCC, consisting of the ganglion cell layer, internal plexiform layer and retinal nerve fibre layer (RNFL), was calculated using Angiovue from the macular cube image acquired from the OCTA scan. The fGCC, pfGCC, perifGCC, and wiGCC thicknesses were calculated from the same fovea-centred regions where fVD, pfVD, perifVD and wiVD were obtained.

For measurement of metrics associated with FAZ, the methods used in this study followed that used in a prior publication.24 Briefly, FAZ was defined as the region that is enclosed by the innermost macular arcade, and the Avanti Angiovue software automatically detected capillary-free area and calculated FAZ metrics based on the retinal slab. The following parameters were evaluated in this study: (1) FAZ area, (2) FAZ circumferences, (3) foveal density 300 (FD300), defined as the superficial VD of the 300 µm width ring surrounding the FAZ (figure 1B). For calculation of FAZ area, correction to consider the magnification effect was performed using the Littman formula (corrected FAZ area=FAZ area × 3.462 × 0.0130622 × [axial length − 1.82]2), which was derived from on the default axial length in the Avanti systems (23.95 mm).

OCTA image quality review was performed by trained graders according to the University of California, San Diego, USA, Imaging Data Evaluation and Analysis Reading Centre standard protocol. An image was considered poor-quality and excluded if any of the following was presented: (1) scan quality <4, (2) poor clarity, (3) residual motion artefacts visible as irregular vessel pattern on the en face angiogram, (4) image cropping or local weak signal, (5) off-centred fovea and (6) severe segmentation errors that could not be corrected.

Statistical analysis

Continuous variables were presented as mean (95% CI) and categorical variables as count (%). Categorical variables were compared using the χ2 test. Eye characteristics were compared using linear mixed-effects models to account for within-participant variability. Age and Signal Strength Index (SSI)-adjusted mixed modelling was performed to characterise the correlation between macular OCT/OCTA parameters, including their hemifield measurements, and logMAR VA. A positive β coefficient indicated worse logMAR VA. Axial length was additionally adjusted for FAZ area in the mixed model analysis. Receiver operating characteristic analysis was performed, and area under receiver operating characteristic (AUC) was calculated to examine the discriminative power of OCT/OCTA parameters for decreased VA (defined as LogMAR VA>0.10 or Snellen VA<20/25) due to glaucoma.18 25 Adjustment for possible within-subject correlation between eyes obtained from the same patients was performed in all analysis. Statistical analyses were performed using Stata V.16.0 (StataCorp), and statistical significance was defined as a p<0.05 for all analyses.


Clinical characteristics of the subjects are summarised in table 1. A total of 144 POAG eyes of 100 participants were included. The eyes were divided into early glaucoma group (80 eyes of 62 participants) and moderate-advanced glaucoma group (64 eyes of 53 participants), with the mean age (95% CI) of 78.7 (76.7 to 80.6) and 79.8 (77.8 to 81.7) years, respectively. There were significant differences in mean VF MD (early=−2.7 (95% CI −3.1 to –2.3) dB, moderate-advanced=−13.4 (95% CI -14.5 to −11.5) dB, p<0.001) and mean logMAR VA (early = 0.05 (95% CI 0.03 to 0.07), moderate-advanced=0.11 (95% CI 0.08 to 0.13), p=0.003). A greater percentage of eyes (38%) had decreased VA in the moderate-advanced glaucoma group (p<0.001). Most OCT/OCTA parameters were significantly different between the two groups (p<0.05 for all), except for FAZ area, FAZ circumference, and foveal measurements of superficial VD, deep VD and GCC.

Table 1

Demographics and baseline clinical characteristics of the subjects

Table 2 presents the age and SSI-adjusted associations of OCT/OCTA parameters with VA stratified by glaucoma severity. In the early glaucoma group, none of the OCT/OCTA parameters, including their hemifield measurements, demonstrated significant association with logMAR VA (p>0.05 for all). In moderate-advanced glaucoma eyes, greater FAZ area (β (95% CI) = 0.023 (0.002 to 0.043)], p=0.035, R2=0.11) and FAZ circumference (β (95% CI) = 0.063 (0.006 to 0.120), p=0.032, R2=0.11) were associated with worse logMAR VA, but FD300 was not. Lower measurements of almost all GCC thicknesses, including their superior hemifield thicknesses, were significantly associated with worse logMAR VA (range of β=0.002–0.003, p<0.05 for all, range of R2=0.11–0.33), except for the inferior hemifields of perifGCC and wiGCC (p>0.05). For superficial VD parameters, lower superior hemifield perifVD (β (95% CI) = 0.007 (0.000 to 0.014), p=0.045, R2=0.11) and lower superior hemifield wiVD (β (95% CI) = 0.008 (0.000 to 0.016), p=0.042, R2=0.12) were associated with worse logMAR VA. For deep VD, only fVD was associated with worse logMAR VA (β (95% CI) = 0.004 (0.000 to 0.007), p=0.049, R2=0.11). A supplemental analysis was performed with IOP included as an additional covariate in the mixed model (online supplemental table 1). Overall, the results were similar to those of the main analysis, and no significant effect of IOP on VA measurement was observed.

Supplemental material

Table 2

Mixed model analysis of OCT and OCTA parameters associated with logMAR visual acuity (VA) based on glaucoma severity

The AUC of OCT/OCTA parameters for discriminating between eyes with and without decreased VA are summarised in table 3. In early glaucoma, FAZ area yielded the highest AUC (0.72 (95% CI 0.50, 0.90)) among the FAZ metrics. For superficial and deep VD, parafoveal, perifoveal and whole-image measurements perfomed similarly across all sectors (AUC range=0.68–0.77). As for GCC, the parafoveal region showed the best discrimination (AUC range=0.72–0.76). In moderate-advanced glaucoma, all FAZ metrics showed similar discrimination (AUC range=0.66–0.67). For superficial and deep VD, measurements obtained from different regions again performed similarly (AUC range=0.63–0.72). While no regional differences in AUC were found among VD parameters, GCC thickness tended to perform better with foveal (AUC=0.79 (95% CI 0.68 to 0.90)) and parafoveal measurements (AUC range=0.75–0.81). Figure 2 shows the AUC of foveal and parafoveal OCT/OCTA measurements in moderate-advanced glaucoma. The discriminative power to detect decreased VA of most OCTA parameters was slightly weaker in the moderate-advanced group. While for GCC thicknesses, most parameters seemed to show a slightly higher AUC in moderate-advanced glaucoma. In both severity groups, the superior hemifield yielded higher AUC as compared with the inferior hemifield for most OCT/OCTA parameters.

Table 3

Receiver operating characteristic analysis of the discriminative power of OCT/OCTA parameters to detect decreased visual acuity

Figure 2

Area under the receiver operating characteristic (AUC) of (A) foveal and (B) parafoveal measurements for discrimination between eyes with and without decreased visual acuity in moderate-advanced glaucoma group. GCC, ganglion cell complex; VD, vessel density.


This study investigated the relationship between macular OCT and OCTA structural parameters with VA across different glaucoma severities. For early glaucoma eyes, none of the parameters were associated with VA. However, in moderate-advanced glaucoma eyes, lower values of most GCC thickness parameters were associated with worse VA. In contrast, most VD parameters were not significantly associated with VA. For discrimination between eyes with and without decreased VA, GCC and VD showed similar discriminatory ability in early glaucoma group. For moderate-advanced glaucoma, fGCC and perifGCC thickness overall and in the superior hemifield had the greatest power, whereas OCTA parameters had lower AUC.

These findings are consistent with prior studies reporting the association between macular OCT parameters and VA.18 26 In an earlier study, circumpapillary RNFL and macular GCC thicknesses were evaluated for correlation with VA in severity-stratified analysis.18 While the VF cut-off (MD = −12 dB) used for stratification in that study was different from ours, similar to our results, a significantly stronger association between VA and OCT parameters was found for the more severe group. Examining only advanced and severe (MD < −12 and −20 dB) glaucoma eyes, a recent OCTA study suggested macular VD, particularly deep nasal VD, as a promising indicator of VA.19 Macular GCC, however, was not significantly associated with VA in their study population, which might be due to the OCT measurement floor effect. As compared with functional parameters, structural parameters are more subjected to floor effect in late-stage glaucoma.26 27 Furthermore, the floor is detected earlier in OCT than in OCTA, particularly with a VF MD worse than −14 dB.28 In their study, a mean VF MD around −20 dB was reported. Whereas in our analysis, most moderate-advanced glaucoma eyes had a VF better than −14 dB, which may explain in part the stronger association observed between OCT and VA in this study.

Some other findings in this study also differ from prior results. The association between FAZ metrics and central VF has been shown previously.24 29 In the aforementioned OCTA study,19 both FAZ area and FD300 were significantly associated with VA. However, a similar association was not found for FD300 in our analysis. Without available evidence from other similar studies, it is most likely the discrepancy is also related to the differences in the severity of glaucoma in the study participants. Additionally, in the current analysis, deep VD was associated with VA only in the foveal area and did not outperform superficial VD. While in their study, deep VD was more strongly associated with VA as compared to superficial VD, with significant results obtained from the foveal, parafoveal and whole-image measurements.19 Since they examined only advanced-severe glaucoma, the authors hypothesised the findings may have resulted from an increased impairment of deep capillary perfusion (DCP) with glaucoma progression, as DCP is usually preserved until the later stages due to the anastomoses supply from superficial plexus.30 31

In most past reports, deep VD was reported to be less strongly associated with VA than superficial VD in glaucoma, particularly during the early-moderate stages,32–35 and whether the involvement of deep VD in glaucomatous damage alters as the disease progresses is unclear. Nevertheless, some studies which investigated other ocular conditions have also found a stronger correlation between VA and deep VD as compared with superficial VD,36–38 although the mechanism remains unknown and the association might be more attributed to the underlying pathologies. Overall, more studies are needed to clarify if there is a relationship between deep VD and VA in glaucoma, and if the role of DCP truly varies across different glaucoma severities.

Another intriguing finding of this study is the region-dependent association of VD obtained from different vascular layers, with deep VD demonstrating better correlation at the central/foveal region and superficial VD at the more peripheral and superior region. This is supported by a prior study that showed a better association between central VF and deep VD as compared with superficial VD in glaucoma,39 suggesting a primary effect of central/foveal DCP on central vision, which is related to VA. However, an association between VA and superficial VD obtained from the foveal and parafoveal regions has also been found in a previous OCTA study, in which a larger number of advanced glaucoma eyes were included.25

In this study, a modest discriminative power of superficial and deep VD was found for both severity groups, with a slightly weaker discrimination in the moderate-advanced glaucoma group. In contrast, OCT-measured GCC thicknesses, which performed similarly to OCTA in early glaucoma group, showed better discrimination than OCTA in moderate-advanced glaucoma. In addition, the parafoveal measurements yielded the best results for superficial VD and GCC parameters, indicating this region may be more helpful for this task. Results in a prior study examining VD also support the greater discriminative power of parafoveal measurements, although perifVD was not specifically examined.19 As mentioned previously, even mild-to-moderate decline in VA can cause impairment to VrQOL and daily function, and this should not be overlooked in glaucoma.11 12 15 Therefore, to identify structural parameters that may facilitate discrimination of patients at risk of decreased VA is clinically beneficial.

There are several limitations of this study. First, although all images went through quality review, the VD measurements are more variable the OCT thickness measurements.40 Analysis of deep VD, in particular, might be affected by projection artefacts that were not completely removed.41 Nevertheless, our result may serve as the basis for future investigation on the long-term association between VA and OCT/OCTA in glaucoma. Second, due to the limited number of available eyes, a detailed severity-stratified analysis (early/moderate/advanced) could not be performed, and some patients had both eyes included. Third, since data of the Early Treatment Diabetic Retinopathy Study letter score was not available, Snellen VA-converted logMAR VA was used in the current analysis, which was also the approach adopted by most prior studies.18 19 Last, to exclude the possible confounding effect of cataract on VA,42 only pseudophakic eyes were included, and most patients demonstrated mild-moderate VA decline. Whether the results would differ with inclusion of non-pseudophakic eyes or more eyes with severe VA decline is a subject for future study. However, this study still provides insights into the relationship between local structures and VA performance in its earlier course.

In summary, some macular OCT and OCTA parameters, particularly GCC thickness, FAZ area and FAZ circumference, showed statistically significant associations with VA in moderate-advanced, but not early glaucoma. The association of superficial and deep VD with VA varied by region, indicating the potentially differential involvement of local vasculatures in the decline of VA. Parafoveal measurements of GCC, especially in the superior hemifield, showed a greater ability to discriminate glaucoma eyes with decreased VA from those without it. These structural parameters may help to identify glaucoma patients at risk of impaired vision and reduced quality of life.

Data availability statement

Data are available on reasonable request. The datasets generated and/or analysed during the current study are available from the corresponding author on reasonable request.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and was approved by University of California San Diego Human Research Protection Program (NCT00221897). Participants gave informed consent to participate in the study before taking part.



  • J-HW and SM are joint first authors.

  • J-HW and SM contributed equally.

  • Contributors Concept and design: J-HW and SM; Acquisition and reviewing of data: J-HW, SM, TN, VM, AK and LMZ; Analysis or interpretation of data: J-HW, SM, TN, LMZ and RNW; Drafting of the manuscript: J-HW and SM; Critical revision of the manuscript: all authors; Obtained funding: SM, LMZ and RNW; Supervision: SM and RNW; Guarantor of the study: RNW.

  • Funding This work is supported by National Institutes of Health/National Eye Institute Grants R01EY029058, R01EY011008, R01EY019869, R01EY027510, R01EY026574, R01EY018926, P30EY022589; University of California Tobacco Related Disease Research Programme (T31IP1511), and an unrestricted grant from Research to Prevent Blindness (New York, NY).

  • Disclaimer The sponsor or funding organisation had no role in the design or conduct of this research.

  • Competing interests LMZ reported grants from the National Eye Institute; grants and nonfinancial support from Heidelberg Engineering, non-financial support from Carl Zeiss Meditec, Optovue and Topcon. RNW reported nonfinancial support from Heidelberg Engineering, Carl Zeiss Meditec, Konan Medical, Optovue, Centervue and Topcon; grants from the National Eye Institute; personal fees from Abbvie, Aerie Pharmaceuticals, Allergan, Equinox, Nicox, and Topcon; all outside the submitted work. No other disclosures were reported.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.