Threat to fixation and vision‐related quality of life in patients with open‐angle glaucoma

Abstract

Purpose

We aimed to investigate the association between overlapping threat to fixation (TTF) and vision‐related quality of life (VRQoL) in patients with manifest open‐angle glaucoma (OAG).

Methods

Overall, 316 patients with manifest OAG were recruited from the glaucoma outpatient department of Skåne University Hospital, Malmö, Sweden, between April and December 2018. VRQoL was assessed using the Swedish version of the National Eye Institute Visual Function Questionnaire‐25 (VFQ‐25). The results were analysed using Rasch analysis. Overlapping TTF was defined as glaucomatous visual field loss (VFL) involving at least one of the four most central points, depressed at p < 1% on the total deviation probability map of Humphrey visual fields, and present at corresponding points in both eyes. The relationship between VRQoL and overlapping TTF was examined using linear regression analysis.

Results

In univariable linear regression analysis, overlapping TTF was significantly associated with VRQoL scores on the Visual Functioning Scale (VFS) and Socioemotional Scale (SES). However, this correlation was not significant in multivariable linear regression analysis after adjusting for potentially confounding factors (all p > 0.05). Integrated visual field mean deviation (IVF‐MD) and best‐corrected visual acuity in the better and worse eyes were significantly correlated with Rasch‐calibrated VFQ‐25 scores. Collectively, these factors accounted for 37% and 34% of the variance in VFS and SES scores, respectively.

Conclusion

Overlapping TTF was not an independent predictor of VRQoL after controlling for VFL severity. By contrast, IVF‐MD emerged as the most significant predictor of VRQoL.

1. INTRODUCTION

Glaucoma is a progressive optic neuropathy and among the leading global causes of irreversible visual impairment (Tham et al., 2014). All glaucoma treatments are primarily aimed at preserving the patient's visual function and quality of life (QoL) (‘European Glaucoma Society Terminology and Guidelines for Glaucoma, 5th Edition’, 2021; Johannesson et al., 2024). Although vision‐related QoL (VRQoL) evaluation has been increasingly incorporated into glaucoma research over the past two decades (Biggerstaff & Lin, 2018; Khachatryan et al., 2021; Quaranta et al., 2016), this domain remains insufficiently explored.

The concept of threat to fixation (TTF) has been used to assess the damage to the central visual field (VF). In grading different stages of glaucoma, TTF has been considered an aggravating factor (Hodapp et al., 1993; Mills et al., 2006), resulting in the exclusion of patients with TTF or their differential management (Lascaratos et al., 2013; Leske et al., 1999). Since then, several studies have evaluated the correlation between various localisations of VF loss (VFL) and VRQoL (Black et al., 2011; Cheng et al., 2015; Chun et al., 2019; Murata et al., 2013; Sumi et al., 2003). However, comparison across these studies is challenging due to the variability in TTF definitions and QoL instruments employed. The impact of TTF on the lifetime risk of blindness has been previously assessed, demonstrating that TTF is not an independent predictor of blindness (Peters, Bengtsson, & Heijl, 2015). A more recent study presenting results from the Glaucoma Intensive Treatment Study suggested that the presence of TTF did not affect VRQoL in a relatively large cohort, which primarily consisted of individuals with early glaucoma (Peters et al., 2023). Another study also recently reported that TTF does not independently predict VRQoL in individuals with more advanced disease (Gazanchian & Jansonius, 2024). Therefore, we aimed to further assess the relationship between VRQoL and the presence of overlapping TTF in all stages of manifest open‐angle glaucoma (OAG).

2. METHODS

This study was approved by the regional Ethical Review Board of Lund, Sweden (approval number: Dnr 2018/241), and conducted in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from all participants. The present analysis is a cross‐sectional cohort study including patients from a previously conducted prospective cross‐sectional study assessing the prevalence of Charles Bonnet syndrome in patients with glaucoma (Peters et al., 2022). Specifically, the process for identifying eligible patients has been previously described in detail (Peters et al., 2022). Briefly, adult patients aged ≥18 years with manifest OAG—including primary open‐angle glaucoma (POAG) and pseudoexfoliation glaucoma (PEXG)—who were followed at the glaucoma outpatient department of Skåne University Hospital in Malmö, Sweden, were considered for inclusion. The POAG group included patients with an intraocular pressure (IOP) within the normal range and those with higher IOP levels. Only patients with a manifest OAG diagnosis for at least 1 year, who attended the department between 1 April 2018 and 31 December 2018, were eligible for inclusion.

Patients without reliable VF measurements, those with other neurological or ophthalmological conditions that could affect the VF and those with advanced macular disease, dementia or other neurological diseases were excluded. Additionally, patients without recent visual acuity (VA) available and those with a VA <0.1 on the Snellen scale in one or both eyes (unrelated to glaucoma) were excluded (Figure S1). Data on VRQoL were available for all 337 patients participating in the previous study. For the present cross‐sectional cohort study, we evaluated these 337 patients for inclusion and assessed their VF measurements for the presence of TTF (Figure 1). Patients with overlapping TTF were required to have bilateral glaucoma to ensure that TTF due to other conditions did not interfere with the evaluation of TTF as a risk factor for low VRQoL. Therefore, patients with unilateral glaucoma but overlapping TTF were excluded from the final cohort analysis.

FIGURE 1.

Flow chart illustrating the distribution of TTF and overlapping TTF, as well as the localisation of overlapping TTF within the study population. Patients without overlapping TTF–including those without TTF–were classified as group 1, while those with overlapping TTF were designated as group 2. Patients with overlapping TTF needed to have bilateral glaucoma to ensure that TTF due to other conditions did not interfere with our evaluation of TTF as a risk factor for low VRQoL. Therefore, patients with unilateral glaucoma but overlapping TTF were excluded from the final cohort analysis. *Reasons for the other non‐glaucomatous eye exhibiting reduced sensitivity in the TTF region: Cataract (nine patients), posterior capsular opacification (one patient), corneal disease (five patients) and unknown causes (six patients). TTF, threat to fixation.

Patients' visual function was assessed using best‐corrected VA (BCVA) and VF measurements from the clinical visit closest to the date of completion of the National Eye Institute Visual Function Questionnaire‐25 (VFQ‐25). Additional information on age, sex, glaucoma type (POAG or PEXG) and glaucoma laterality was obtained from clinical records.

All included patients were required to have reliable VF tests performed with a Humphrey Field Analyser (Carl Zeiss Meditec, Dublin, CA, USA), using the 24‐2 or 30‐2 Swedish Interactive Thresholding Algorithm (SITA) standard or fast program. VF tests were considered reliable if the value of the point corresponding to the blind spot was <10 dB and the false positive rate was <15%. Tests were excluded if they were clinically unreliable (e.g. poor cooperation or inability to follow the testing protocol), even if they met the formal reliability criteria. The presence of TTF was assessed in all eyes. TTF was defined as a VF defect in one or more of the four most central points, depressed at p < 1% on the total deviation probability map of Humphrey 24‐2 or 30‐2 SITA standard or fast VFs. These points were located 3° below or above the horizontal meridian and 3° nasally or temporally along the vertical meridian. Patients with bilateral glaucomatous VF defects were classified as having overlapping TTF if at least one such point was present at the corresponding test location in both eyes. Overlapping TTF localisation was noted as present in the superior hemifield, inferior hemifield or both.

The study population was categorised into groups 1 (which included patients without overlapping TTF) and 2 (comprising all patients with overlapping TTF) (Figure 1). Glaucoma was defined by the presence of a VF defect consistent with glaucomatous damage (e.g. arcuate scotoma, nasal step or paracentral scotoma) observed in the same area on at least two consecutive VF tests. Additionally, the Glaucoma Hemifield Test had to be classified as ‘outside normal limits’ on at least one test and at least ‘borderline’ on the other. Reliable VF tests closest to the date of completion of the VFQ‐25 were selected for inclusion in the study. Mean deviation (MD) in decibels (dB) was recorded for both the better and worse eyes. The integrated VF MD (IVF‐MD) was calculated by selecting the highest sensitivity value from either eye at each corresponding point in the VF (Nelson‐Quigg et al., 2000). Subsequently, the unweighted average of these values was calculated to obtain the IVF‐MD. These calculations were performed automatically using a program developed by our research group. Based on the IVF‐MD, patients were categorised into different glaucoma stages according to the definitions provided by Mills and co‐workers (Mills et al., 2006).

BCVA was extracted from the clinical records. As VA was measured using Snellen charts, data on BCVA in the clinical records were only available in Snellen equivalents. VA values noted as counting fingers and hand movements were converted to Snellen equivalents of 0.01 and 0.005, respectively. Eyes without a corresponding Snellen equivalent (e.g. light perception) were labelled accordingly and excluded from the calculation of mean VA. To facilitate easier comparison between studies and accessibility to clinicians, we converted the Snellen decimals to logMAR and approxETDRS (Gregori et al., 2010). When interpreting these figures, it is important to note that Snellen charts were used to measure BCVA and that the conversion to logMAR and approxETDRS still contains the inherent flaw of Snellen measurement, particularly in measuring low VA.

VRQoL was assessed using the validated VFQ‐25. The original VFQ‐25 (Mangione et al., 2001), developed by RAND and sponsored by the National Eye Institute, has been translated into Swedish and validated previously (Hyman et al., 2005). This Swedish version was modified by removing the two questions related to driving, as requested by the Ethical Review Board of Lund, and subsequently distributed via post and self‐administered.

2.1. Rasch analysis of VFQ‐25

Rasch analysis was conducted using Winsteps software version 4.7.0.0 (Winsteps, Chicago, IL, USA). The VFQ‐25 results were analysed using the visual functioning and socioemotional models, as previously described by Marella and co‐workers (Marella et al., 2010).

The initial fit of the VFQ‐25 to the Rasch model revealed disordered thresholds in the social functioning subscale. Participants could not differentiate between categories 0 (‘stopped doing so because of eyesight’) and 1 (‘extreme difficulty’). This issue was resolved by combining these two categories. All items included in the models subsequently demonstrated satisfactory fit, with item infit (information‐weighted fit) mean square values <1.4 (Wright & Linacre, 1994) and point‐measure correlation values >0.5. Unidimensionality was assessed using principal component analysis of the residuals. Both scales demonstrated unidimensionality, with variance explained by the measure exceeding 70% and aligning closely with the variance expected from the models.

Differential item function (DIF) testing was conducted to assess whether subgroups within the sample responded differently to individual items. Specifically, DIF was evaluated based on median age (< 78 vs. ≥78 years), sex and general health. General health was assessed using the first item of the VFQ‐25, which asks respondents to provide a subjective rating of their general health. Participants were categorised as having at least ‘good’ health or reporting worse general health. Notable DIF was defined as a difference between subgroups >1.0 logits (Pesudovs et al., 2010). No notable DIF was detected in the Rasch analysis.

Separate person measures were calculated for each participant, reflecting visual ability based on the two Rasch models. A linear transformation was applied to rescale the Rasch estimates (logits) to positive values, where 0 and 100 represented the lowest and highest possible visual ability, respectively, within the study population. These transformed values—referred to as Rasch‐calibrated VFQ‐25 scores—were used in all subsequent analyses.

2.2. Statistical analysis

The relationship between VRQoL, as defined by the Rasch‐calibrated VFQ‐25 scores, and the presence and localisation of overlapping TTF was initially explored using univariable linear regression analysis. For this purpose, the different potential localisations of overlapping TTF (superiorly only, inferiorly only and both hemifields) were decoded into dichotomous variables with categories 0 and 1. Potential confounding variables, such as IVF‐MD, BCVA in the better and worse eyes, age and sex, were included in a multivariable linear regression model to further evaluate the relationship between overlapping TTF and VRQoL. Clinical characteristics of the two groups—those with and without overlapping TTF—were compared using the Mann–Whitney U test and t‐test for non‐normally and normally distributed data, respectively. Categorical variables were analysed using Fisher's exact test. All statistical analyses were performed using IBM SPSS Statistics for Windows, version 27 (IBM Corp., Armonk, N.Y., USA), with statistical significance set at p < 0.05.

3. RESULTS

VFs from 337 patients were evaluated for the presence of TTF. TTF in at least one paracentral point of one eye was observed in 305 patients (90.5%). Only 32 patients (9.5%) had no TTF in either eye (Figure 1). Twenty‐one patients with unilateral glaucoma and overlapping TTF were excluded from all further analyses. In those cases, one of the eyes was exhibiting reduced sensitivity in the TTF region because of causes unrelated to glaucoma, such as cataract (nine patients), posterior capsular opacification (one patient), corneal disease (five patients) and unknown causes (six patients). A total of 316 patients were included in the final analyses. The median (interquartile range [IQR]) MD of the better and worse eyes was −4.69 (−11.47 to −1.51) and −15.31 (−24.29 to −8.10) dB, respectively. Based on the IVF‐MD, 198 (63%), 53 (17%), 36 (11%) and 29 (9%) patients were classified as having early, moderate, advanced and severe glaucoma, respectively (Figure 1).

Among the 316 patients, 170 (54%) were classified into group 1 (no overlapping TTF), while 146 (46%) were classified into group 2 (overlapping TTF). Overlapping TTF was found in both hemifields in over 50% of patients (Figure 1). Patients with overlapping TTF had more advanced stages of glaucoma than those without overlapping TTF (Table 1). All patients with overlapping TTF had bilateral glaucoma. Table 1 presents a comparison of the clinical characteristics between patients without (group 1) and with (group 2) overlapping TTF.

TABLE 1.

Comparison of the clinical characteristics of patients without and with overlapping TTF.

	Group 1: Without overlapping TTF (n = 170)	Group 2: Overlapping TTF (n = 146)	p‐value
Age, years
Mean (±SD)	76 ± 8	80 ± 8	0.554 a
Sex, n (%)
Female	84 (49)	89 (61)	0.042 b
Male	86 (51)	57 (39)
Type of glaucoma, n (%)
POAG	106 (62)	99 (68)	0.345 b
PEXG	64 (38)	47 (32)
Glaucoma laterality, n (%)
One eye	104 (61)	0 (0)	< 0.001 b
Both eyes	66 (39)	146 (100)
VFI (%), median (IQR)
Better eye	98 (4)	68 (39)	< 0.001 c
Worse eye	73 (40)	20 (44)	< 0.001 c
BCVA, median (IQR)
Snellen decimal
Better eye	1.00 (0.20)	0.80 (0.40)	< 0.001 c
Worse eye	0.70 (0.33)	0.50 (0.50)	< 0.001 c
logMar
Better eye	0.00 (0.09)	0.10 (0.22)	< 0.001 c
Worse eye	0.10 (0.14)	0.26 (0.50)	< 0.001 c
approxETDRS
Better eye	85 (5)	80 (11)	< 0.001 c
Worse eye	77 (10)	72 (25)	< 0.001 c
MD (dB), median (IQR)
Better eye	−1.82 (2.61)	−11.55 (12.96)	< 0.001 c
Worse eye	−9.43 (12.30)	−21.55 (11.96)	< 0.001 c
IVF‐MD (dB), median (IQR)	−1.58 (2.48)	−11.00 (12.22)	< 0.001 c

Abbreviations: approxETDRS, approximated Early Treatment Diabetic Retinopathy Study letters; BCVA, best‐corrected visual acuity; dB, decibels; IVF, integrated visual field; IQR, interquartile range; MD, mean deviation; PEXG, pseudoexfoliation glaucoma; POAG, primary open‐angle glaucoma; SD, standard deviation; TTF, threat to fixation; VFI, visual field index.

^a t‐test.

^b Fisher's exact test, 2‐sided.

^c Mann–Whitney U test.

Most patients without overlapping TTF (159 of 170, 94%) had early glaucoma according to the IVF‐MD, whereas the group with overlapping TTF included patients across all stages of glaucoma (Figure 2). Only 10 and 1 patients with moderate and advanced glaucoma, respectively, had no overlapping TTF. All patients with severe glaucoma exhibited overlapping TTF. The median time between VF measurement and VRQoL assessment was 2 (range 0–10) months.

FIGURE 2.

Localisation of overlapping TTF in relation to the glaucoma stage, as defined by IVF‐MD. Most patients without overlapping TTF had early glaucoma, while overlapping TTF was present across all stages of glaucoma. Notably, all patients with severe glaucoma exhibited overlapping TTF. dB, decibels; IVF‐MD, integrated visual field mean deviation; TTF, threat to fixation.

The median Rasch‐calibrated VFQ‐25 score on the Visual Functioning Scale (VFS) was 53 (IQR: 34–68) and 66 (IQR: 57–77) for patients with and without overlapping TTF, respectively. On the Socioemotional scale (SES), the median Rasch‐calibrated VRQoL score was 57 (IQR: 41–72) for patients with overlapping TTF and 73 (IQR: 63–80) for patients without overlapping TTF.

Figure 3 shows the Rasch‐calibrated VFQ‐25 scores for subgroups of patients with early glaucoma—those without overlapping TTF, those with overlapping TTF only in the inferior hemifield, those with overlapping TTF only in the superior hemifield and those with overlapping TTF in both hemifields.

FIGURE 3.

Rasch‐calibrated VFQ‐25 scores on the Visual Functioning and Socioemotional Scales in relation to the presence and localisation of overlapping TTF in the subgroup of patients with early glaucoma. No significant differences were observed among the four groups after adjusting for IVF‐MD and visual acuity. IVF‐MD, integrated visual field mean deviation; NEI VFQ‐25, National Eye Institute Visual Function Questionnaire‐25; QoL, quality of life; TTF, threat to fixation.

In univariable linear regression analysis, both the presence of overlapping TTF (VFS: R ² _adjusted = 0.153, 95% confidence interval [CI]: 0.081–0.225, p < 0.001; SES: R ² _adjusted = 0.158, 95% CI: 0.085–0.231, p < 0.001) and its localisation (VFS: R ² _adjusted = 0.199, 95% CI: 0.121–0.277; p < 0.001; SES: R ² _adjusted = 0.202, 95% CI: 0.124–0.280, p < 0.001) were identified as risk factors for lower VRQoL score. However, when adjusting for potential confounding factors in multivariable linear regression analysis, neither the presence (Table 2) nor the localisation of overlapping TTF (Table 3) remained independent risk factors. Instead, IVF‐MD emerged as the most important risk factor for low VRQoL on both scales. In the subgroup of patients with early glaucoma, BCVA in the worse eye was the main predictor of VRQoL.

TABLE 2.

Multivariable linear regression analysis: TTF adjusted for potentially confounding factors and Rasch‐calibrated VRQoL scores on the Visual Functioning and Socioemotional Scales (n = 316).

	Reference	Visual Functioning Scale			Socioemotional Scale
		R 2 adjusted	95% CI	p‐value	R 2 adjusted	95% CI	p‐value
TTF a	No			0.648			0.568
Other factors
IVF‐MD	Per 1 dB	0.284	0.202, 0.366	<0.001	0.272	0.191, 0.354	<0.001
BCVAworse eye	Per 0.1 Snellen Scale	0.074	0.020, 0.128	<0.001	0.055	0.007, 0.103	<0.001
BCVAbetter eye	Per 0.1 Snellen Scale	0.020	−0.010, 0.050	0.001	0.010	−0.011, 0.031	0.030
Sex	Male			0.141	0.009	−0.011, 0.029	0.042
Age	Years			0.541			0.748
Model summary b		0.374	0.292, 0.456	<0.001	0.340	0.257, 0.423	<0.001

Abbreviations: BCVA, best‐corrected visual acuity; CI, confidence interval; dB, decibel; IVF‐MD, integrated visual field mean deviation; TTF, threat to fixation; VRQoL, vision‐related quality of life.

^a Adjusted for IVF‐MD, BCVA_{better eye}, BCVA_{worse eye}, age and sex.

^b Model including all significant factors from the multivariable linear regression analysis.

TABLE 3.

Multivariable linear regression analysis: TTF localisation adjusted for confounding factors and Rasch‐calibrated VRQoL scores on the Visual Functioning and Socioemotional Scales evaluated for the subgroup of patients with early glaucoma (n = 198).

Explanatory factor(s)	Reference	Visual Functioning Scale			Socioemotional Scale
		R 2 adjusted	95% CI	p‐value	R 2 adjusted	95% CI	p‐value
TTF localisation a	No TTF
TTF superiorly only a				0.271			0.211
TTF inferiorly only a				0.840			0.868
TTF in both hemifields a				0.666			0.421
Other factors
BCVAworse eye	Per 0.1 Snellen Scale	0.118	0.053, 0.183	< 0.001	0.118	0.053, 0.183	< 0.001
IVF‐MD	Per dB	0.037	−0.003, 0.077	< 0.001	0.031	−0.006, 0.068	0.009
BCVAbetter eye	Per 0.1 Snellen Scale			0.132			0.111
Age	Years			0.925			0.514
Model summary b		0.151			0.145

Abbreviations: BCVA, best‐corrected visual acuity; CI, confidence interval; dB, decibel; IVF‐MD, integrated visual field mean deviation; MD, mean deviation; TTF, threat to fixation; VRQoL, vision‐related quality of life.

^a Adjusted for IVF‐MD, BCVA_{better eye}, BCVA_{worse eye} and age.

^b Model including all significant factors from the multivariable linear regression analysis.

4. DISCUSSION

The present cross‐sectional study evaluated VRQoL in patients across all stages of glaucoma, including over 200 individuals with bilateral disease. This provided an opportunity to further investigate whether TTF is an independent predictor of reduced VRQoL in patients with glaucoma. Our results suggest that overlapping TTF is common in patients with glaucoma and occurs in nearly all individuals with advanced VF defects, as indicated by IVF‐MD. TTF was not an independent risk factor for low VRQoL in our study, as measured by the VFQ‐25. Instead, IVF‐MD emerged as the primary independent predictor of VRQoL, with BCVA in the better and worse eyes also contributing to the prediction.

TTF is more common in patients with advanced glaucoma (Sihota et al., 2007; Sullivan‐Mee et al., 2016) and has been suggested to increase the risk of further VFL near fixation, which is associated with central acuity loss (Membrey et al., 2000). However, TTF has not been identified as an independent risk factor for lifetime blindness (Peters, Bengtsson, & Heijl, 2015), challenging the long‐held assumption that TTF presence indicates more severe glaucoma.

More advanced glaucomatous VFL in the better eye is a well‐established predictor of low VRQoL (Okamoto et al., 2014; Peters, Heijl, et al., 2015). While TTF could still reflect a more severe condition if independently correlated with worse VRQoL, the impact of different VFL locations on patients' QoL has been evaluated in several studies, though the results remain inconclusive (Abe et al., 2016; Black et al., 2011; Cheng et al., 2015; Chun et al., 2019; Murata et al., 2013; Sawada et al., 2014; Sumi et al., 2003; Sun et al., 2016). The localisation of VFL in the central lower hemifield has been reported to be correlated with VRQoL in some studies; however, their definitions of central and peripheral VF areas differed from our definition of TTF (Abe et al., 2016; Sun et al., 2016). A study by Sumi and colleagues revealed that VFL in the lower central hemifield within 5° of the fixation point, combined with a low better eye VA, was the most significant predictor for visual disability (Sumi et al., 2003). In our previous study, TTF did not independently predict VRQoL in patients with predominantly early glaucoma (Peters et al., 2023), and the current study confirms these findings in those with more advanced stages of glaucoma. Differences in the study populations, glaucoma types and the questionnaires used restricted a direct comparison between the study by Sumi et al. and ours. However, our results align with the findings of a recent study evaluating the impact of TTF on 269 patients across all stages of glaucoma (Gazanchian & Jansonius, 2024). Our study included a higher proportion of patients with severe glaucomatous VFL than that of the abovementioned study. Approximately 14% of our study population had a better eye MD of less than −18 dB—an important threshold previously identified as significantly associated with VRQoL measured by the VFQ‐25 (Peters, Heijl, et al., 2015).

In our study, most patients with moderate and advanced glaucoma and all those with severe glaucoma exhibited overlapping TTF, which was predominantly located in both hemifields. Therefore, analysing whether the specific localisation of the TTF in one or both hemifields had any additional impact on VRQoL within these subgroups was not possible. To further explore this, we examined how TTF could impact VRQoL by applying a stricter TTF definition: specifically, a defect depth of <10 dB. However, TTF did not emerge as an independent predictor of VRQoL even with this definition (results not shown). Only the subgroup of patients with early glaucoma was sufficient to allow the analysis of the relationship between the localisation of overlapping TTF and VRQoL scores. Among these patients, no significant differences were observed between the localisation groups after adjusting for IVF‐MD and BCVA. This finding aligns with that of a previous study conducted in a cohort of patients with primarily early glaucoma (Peters et al., 2023). One important priority for patients with glaucoma is maintaining outdoor mobility, including the ability to drive (Aspinall et al., 2008), with driving identified as one of the most affected domains of QoL (Khachatryan et al., 2021). VF defects in the lower central hemifield have been associated with driving difficulties (Chun et al., 2019). However, many patients with glaucoma remain unaware of visual symptoms while driving (Kunimatsu‐Sanuki et al., 2025). Nonetheless, assessing the impact of overlapping TTF on patients' subjective experiences related to driving ability would have been valuable.

This study's strengths include a relatively large sample size encompassing patients with all stages of manifest OAG and a high proportion of participants (>1/3) with moderate or more severe glaucomatous VF defects, as measured by IVF‐MD. Over 65% of the included patients had bilateral glaucoma. Another strength is the comparable size of the groups with and without overlapping TTF, which was advantageous for statistical analyses. The use of Rasch analysis to evaluate the VFQ‐25 results is another study strength. Furthermore, only a small number of participants had a better eye BCVA <0.3, making this less of a confounder in our study evaluating TTF as a potential predictor of VRQoL.

This study has some limitations. First, driving‐related QoL questions were excluded, as required by the Ethical Review Board of Lund, potentially omitting an important aspect of VRQoL. Second, the VRQoL was assessed using only the VFQ‐25, which is not a glaucoma‐specific instrument. Using a disease‐specific instrument to assess VRQoL might have yielded different results. However, Rasch analysis demonstrated good targeting and unidimensionality for the two subscales of the VFQ‐25, indicating that both visual and socioemotional aspects of QoL were effectively evaluated in this study.

In conclusion, our findings further support the assumption that TTF is not independently correlated with VRQoL, whereas IVF defects and VA in the better and worse eyes are important independent predictors of VRQoL in patients with manifest OAG. To the best of our knowledge, this study represents the largest investigation to date assessing the relationship between overlapping TTF and VRQoL. Future studies should incorporate glaucoma‐specific instruments and evaluate whether TTF influences patients' perceived driving experiences and related QoL. Our study results highlight the importance of focusing on IVF defects and VA to improve VRQoL, as well as the need for further research to explore the impact of TTF on daily activities such as driving.

CONFLICT OF INTEREST STATEMENT

No conflicting relationship exists for any of the authors.

DISCLOSURE

Dorothea Peters reports receiving speaker honoraria from Santen Pharma AB, Thea Nordic and Low Vision International; she has also served as a consultant for Santen Pharma AB and AbbVie. Additionally, she serves as an unpaid board member of the Swedish Glaucoma Society. Charlotte Almer has nothing to disclose. All authors confirm that they meet the current ICMJE criteria for authorship.

Supporting information

Data S1.

Figure S1. Flow chart demonstrating the process of purposeful sampling for the cohort study.