Introduction: This study aimed to investigate intraocular pressure (IOP)-independent factors associated with the progression of primary open-angle glaucoma (POAG) with IOP ≤15 mm Hg. Methods: POAG patients with maximum IOP ≤15 mm Hg at the Kyoto University Hospital between January 2011 and August 2021 were retrospectively enrolled. We evaluated effects of various factors on the rate of mean deviation (MD) changes in the visual field (VF) examinations using a linear mixed model. These factors included hypertension, diabetes mellitus (DM), hyperlipidemia (HL), cardiovascular disease, arrhythmia, disc hemorrhage, sleep apnea syndrome, orthopedic diseases, and malignant tumors. Results: In total, 98 eyes from 68 patients were included. The baseline MD was −9.74 ± 7.85 dB. The mean rate of MD change and IOP during the observation period were −0.28 ± 0.04 dB/year and 11.8 ± 1.0 mm Hg, respectively. Comorbidity of DM or HL showed a significant positive association with the rate of MD change (β = 0.35, p = 0.0006 and β = 0.18, p = 0.036, respectively) in the model adjusted for age, sex, axial length, mean IOP, and standard deviation of IOP during the observation period. However, no significant association of DM or HL was found after adjusting for central corneal thickness. Conclusion: This study suggests that DM or HL is associated with VF deterioration in glaucoma with lower IOP, but the association may be due to differences in IOP characteristics.

Glaucoma is a progressive optic neuropathy and is a leading cause of blindness worldwide [1, 2]. Higher intraocular pressure (IOP) is an important risk factor for disease development and progression [3, 4]. However, according to the Tajimi Study report [5], many patients with glaucoma in Japan had IOP within the normal limits. Approximately 80% of glaucoma patients aged over 40 years have primary open-angle glaucoma (POAG); of them, 90% have normal-tension glaucoma (NTG). Except for IOP, it is known that age [6], myopia [7], history of diabetes mellitus (DM) [8], and family history of glaucoma [9] are risk factors for the development of POAG. Factors associated with POAG progression include age [10], long-term IOP fluctuations [11, 12], and presence or history of disc hemorrhage (DH) [12, 13]. In addition, an inverse association was reported between cigarette smoking and visual field (VF) deterioration [13]. However, IOP is currently the only treatable target of glaucoma, and there is no effective treatment for suppressing VF deterioration other than lowering IOP, even in cases with normal IOP [14]. Therefore, identifying risk factors other than IOP will help to clarify the pathogenesis of glaucoma in patients with lower IOP and develop new treatment options. It will also play an important role in the early intervention for VF deterioration and in improving the quality of life of patients with glaucoma.

Myopia has been reported to be a risk factor for the development of POAG, but no evidence has been published suggesting its role in disease progression [7]. Thus, separately investigating the risk factors for the development and progression of glaucoma allows for the correct assessment of the underlying pathology. Compared with studies on glaucoma development, few studies have investigated the factors associated with its progression. Therefore, this study investigated the factors associated with glaucoma progression in patients with a lower IOP to further elucidate the pathogenesis of glaucoma.

Patients (Cohort Selection)

This retrospective study adhered to the tenets of the Declaration of Helsinki, and Institutional Review Board (IRB)/Ethics Committee approval was obtained from the Kyoto University Graduate School of Medicine (approval number [R2652]). The requirement for written informed consent was waived due to the retrospective design of this study. Instead, we publicly disclosed results of this retrospective research on our faculty’s website, while also providing the subjects an opportunity to opt out of the study. Under this condition, the Institutional Review Board and Ethics Committee of Kyoto University Graduate School of Medicine approved the waiver for informed consent.

Patients diagnosed with POAG at Kyoto University Hospital between January 2011 and August 2021 were recruited. The inclusion criteria were as follows: (1) the number of VF tests conducted on the Humphrey field analyzer (HFA: SITA-Standard 24-2 testing protocol, Carl Zeiss Meditec, Dublin, CA, USA) was 4 or more; and (2) maximum IOP ≤15 mm Hg during the observation period including both before and after treatment. The exclusion criteria were as follows: (1) history of diseases that might affect VF deterioration, except for POAG, such as epiretinal membrane, branch retinal vein occlusion, central retinal vein occlusion, diabetic retinopathy, and stroke; (2) history of the intraocular surgeries other than cataract surgery; and (3) axial length (AL) ≥28 mm. Cases with AL ≥28 mm were excluded to omit cases of pathological myopia.

Data Collection

All subjects underwent a comprehensive ophthalmic examination at baseline involving IOP measurements with a Goldmann applanation tonometer and uncorrected and best corrected visual acuity with a Landolt chart at 5 m. AL was measured using IOLMaster 500 (Carl Zeiss Meditec), and central corneal thickness (CCT) was measured with SP-3000 (Tomey, Tokyo, Japan). Circumpapillary retinal nerve fiber layer thickness (cpRNFLT) was analyzed by optical coherence tomography (Spectralis HRA + OCT System; Heidelberg Engineering, Heidelberg, Germany) by merging circular scans (3.46 mm in diameter) centered on the optic disc. Data on the history and comorbidities of the patients were collected from medical records.

VF Examination

Only reliable VF test results were included in the analysis. They included the fixation loss rate, false-positive rate, and false-negative rates <20%. A glaucomatous VF defect was defined according to the Anderson-Patella criteria [15]: glaucoma hemifield test results outside normal limits, more than three significant (p < 0.05) and one highly significant (p < 0.01) nonedge contiguous points on the same side of the horizontal meridian as in the pattern deviation plot, or pattern standard deviation < 5% in an otherwise normal VF.

Statistical Analysis

Continuous variables were expressed as the mean ± standard deviation (SD). The results of linear mixed models are described in terms of the mean and standard error. We used linear mixed models to investigate the effects of risk factors on the rates of change in the mean deviation (MD), with reference to previous studies [16, 17]. Mean IOP and SD of IOP for the modeling refers to the IOP of all visits, including pre- and posttreatment values. Among the variables, all diseases were self-reported except for DH, which was detected based on fundus photography (all cases diagnosed with DH had fundus records of when DH occurred).

Equation 1 describes the corrections applied to the data as follows:
MDijt=β0+β1*TIME+ζ0j+ζ1j*TIME+ζ0i|j+ζ1i|j*TIME+εijt
(1)
where MDijt = individual measurements at visit t; β0, β1 = fixed-effects coefficients; ζ0j, ζ1j = random patient effects associated with the intercept and time slope; ζ0i|j, z1i|j = random effects associated with the inclusion of both eyes of a single subject; and εijt = residual. TIME was defined as the time elapsed since the first VF test. The same was true for TIME in the following equation.
To evaluate the effects of age, sex, AL, mean IOP, and SD of IOP on the rate of change in MD during the observation period, we used a linear mixed model adjusted for age, sex, AL, mean IOP, and SD of IOP. Equation 2 describes the adjustment formula for linear mixed modeling as follows:
MDijt=β0+β1*TIME+β2*age+β3*gender+β4*AL+β5*meanIOP+β6*SDofIOP+β7*age*TIME+β8*gender*TIME+β9*AL*TIME+β10*meanIOP*TIME+β11*SDofIOP*TIME+ζ0j+ζ1j*TIME+ζ0i|j+ζ1i|j*TIME+εijt
(2)
To account for the influence of risk factors other than age, sex, AL, mean IOP, and SD of the IOP, we used a linear mixed model. Equation 3 shows an example of the evaluation of the influence of DM on the rate of MD change without correction as follows:
MDijt=β0+β1*TIME+β2*DM+β3*DM*TIME+ζ0j+ζ1j*TIME+ζ0i|j+ζ1i|j*TIME+εijt
(3)
DM*TIME refers to the interaction between DM and TIME. Equation 4 shows an example of the evaluation of the influence of DM on the rate of MD change after adjusting for age, sex, AL, mean IOP, and SD of IOP:
MDijt=β0+β1*TIME+β2*DM+β3*DM*TIME+β4*age+β5*age*TIME+β6*gender+β7*gender*TIME+β8*AL+β9*AL*TIME+β10*meanIOP+β11*meanIOP*TIME+β12*SDofIOP+β13*SDofIOP*TIME+ζ0j+ζ1j*TIME+ζ0i|j+ζ1i|j*TIME+εijt
(4)

Equations 3 and 4 were applied to the following other risk factors: hypertension, hyperlipidemia (HL), cardiovascular disease, arrhythmia, DH, sleep apnea syndrome (SAS), orthopedic diseases, and malignant tumors. Orthopedic diseases included osteoporosis, fracture, and spondylosis deformans. Similar analyses described by equations 1–4 for the assessment of longitudinal MD changes were applied to the cpRNFLT measurements.

All p values presented are 2-sided values. Statistical significance was defined as p < 0.05. All statistical analyses were performed using SAS (SAS® OnDemand for Academics; SAS Institute, Cary, NC, USA) statistical software and R software version 4.2.2 (R Foundation for Statistical Computing, Vienna, Austria).

In total, 98 eyes of 68 patients were included. Of them, cpRNFLT measurement results were available for 69 eyes of 49 patients. Table 1 presents the baseline patient characteristics. The observation period was 6.41 ± 2.78 years (median, 6.18 years). The mean IOP during the observation period was 11.8 ± 1.0 mm Hg (median, 11.8 mm Hg), and the SD of IOP was 1.4 ± 0.3 (median, 1.41 mm Hg). The baseline MD was −9.74 ± 7.85 dB (median, −8.24 dB), and the rate of MD change during the observation period was −0.28 ± 0.04 dB/year (median, −0.23 dB/year).

Table 1.

Demographics of patients

Mean±SDRange
Patients/eyes 68/98  
Observation period, years 6.41±2.78 1.19–10.74 
Age, years 58.1±13.8 25–76 
Sex (male, female) 57/41  
Eye (right/left) 48/50  
Spherical equivalent refraction, D −3.06±3.07 −12.25 to 2.13 
CCT, μm 519.0±33.1 425–575 
AL, mm 25.07±1.34 22.53–27.77 
Maximum IOP, mm Hg 14.3±0.9 11–15 
Mean IOP, mm Hg 11.8±1.0 8.9–14.3 
SD of IOP, mm Hg 1.4±0.3 0–2.4 
Number of HFA examination 8.8±5.7 4–35 
MD of HFA, dB −9.74±7.85 −25.67 to 0.88 
The rate of MD change, dB/year* −0.28±0.04 −2.86 to 1.84 
Number of glaucoma medications 1.6±1.0 0–4 
Proportion of patients with β-blockers, % 43.9  
Number of OCT examination 7.2±7.7 0–34 
Mean±SDRange
Patients/eyes 68/98  
Observation period, years 6.41±2.78 1.19–10.74 
Age, years 58.1±13.8 25–76 
Sex (male, female) 57/41  
Eye (right/left) 48/50  
Spherical equivalent refraction, D −3.06±3.07 −12.25 to 2.13 
CCT, μm 519.0±33.1 425–575 
AL, mm 25.07±1.34 22.53–27.77 
Maximum IOP, mm Hg 14.3±0.9 11–15 
Mean IOP, mm Hg 11.8±1.0 8.9–14.3 
SD of IOP, mm Hg 1.4±0.3 0–2.4 
Number of HFA examination 8.8±5.7 4–35 
MD of HFA, dB −9.74±7.85 −25.67 to 0.88 
The rate of MD change, dB/year* −0.28±0.04 −2.86 to 1.84 
Number of glaucoma medications 1.6±1.0 0–4 
Proportion of patients with β-blockers, % 43.9  
Number of OCT examination 7.2±7.7 0–34 

HFA, Humphrey field analyzer; IOP, intraocular pressure; MD, mean deviation; OCT, optical coherence tomography; SD, standard deviation.

*We calculated the rate of MD change by linear mixed model.

The linear mixed models revealed the influence of age, sex, AL, mean IOP, and SD of IOP on the rate of change in MD (Table 2). Of these factors, age had a significant negative effect (β = −0.0071, p = 0.037), while AL had a significant positive influence (β = 0.087, p = 0.0055) on MD changes. The effects of the CCT were also evaluated (see online suppl. Table S1; for all online suppl. material, see https://doi.org/10.1159/000536314).

Table 2.

Multivariate linear mixed model to evaluate the effect of age, gender, AL, mean IOP, and SD of IOP on the rate of MD change

Multivariate linear mixed model
mean±SE [95% CI]p value
Baseline β0’ 11.0±21.0 [−30.8 to 53.0] 0.60 
β1’ – year −2.30±1.06 [−4.43 to −0.19] 0.033 
β2’ – age −0.085±0.066 [−0.22 to 0.044] 0.20 
β3’ – gender 1.44±1.66 [−1.86 to 4.74] 0.39 
β4’ – AL −0.83±0.67 [−2.14 to 0.48] 0.21 
β5’ – mean IOP 0.52±0.92 [−1.31 to 2.35] 0.57 
β6’ – SD of IOP −1.32±2.56 [−6.40 to 3.75] 0.61 
β7’ – age*TIME −0.0071±0.0034 [−0.014 to −0.0004] 0.037 
β8’ – gender*TIME 0.15±0.076 [0.0021–0.31] 0.047 
β9’ – AL*TIME 0.087±0.030 [0.027–0.15] 0.0055 
β10’ – mean IOP*TIME 0.041±0.047 [−0.053 to 0.13] 0.39 
β11’ – SD of IOP*TIME −0.22±0.13 [−0.47 to 0.028] 0.081 
Multivariate linear mixed model
mean±SE [95% CI]p value
Baseline β0’ 11.0±21.0 [−30.8 to 53.0] 0.60 
β1’ – year −2.30±1.06 [−4.43 to −0.19] 0.033 
β2’ – age −0.085±0.066 [−0.22 to 0.044] 0.20 
β3’ – gender 1.44±1.66 [−1.86 to 4.74] 0.39 
β4’ – AL −0.83±0.67 [−2.14 to 0.48] 0.21 
β5’ – mean IOP 0.52±0.92 [−1.31 to 2.35] 0.57 
β6’ – SD of IOP −1.32±2.56 [−6.40 to 3.75] 0.61 
β7’ – age*TIME −0.0071±0.0034 [−0.014 to −0.0004] 0.037 
β8’ – gender*TIME 0.15±0.076 [0.0021–0.31] 0.047 
β9’ – AL*TIME 0.087±0.030 [0.027–0.15] 0.0055 
β10’ – mean IOP*TIME 0.041±0.047 [−0.053 to 0.13] 0.39 
β11’ – SD of IOP*TIME −0.22±0.13 [−0.47 to 0.028] 0.081 

AL, axial length; IOP, intraocular pressure; MD, mean deviation of the Humphrey field analyzer; SD, standard deviation; SE, standard error.

Using linear mixed models, we analyzed the influence of comorbidities (Table 3). In the uncorrected model, DM had a significant positive correlation with the rate of MD change (β = 0.28, p = 0.0058), whereas orthopedic diseases had a significant negative correlation (β = −0.21, p = 0.018). In the model corrected for age, sex, AL, mean IOP, and SD of IOP, DM, and HL were significantly and positively associated with the rate of MD change (β = 0.35, p = 0.0006; β = 0.18, p = 0.036, respectively). Online supplementary Table S2 shows the effect of comorbidities, considering the impact of CCT. The models, including when they were adjusted for CCT, detected no significant associations of DM or HL with the rate of MD change. The results of the linear mixed models for the evaluation of longitudinal cpRNFLT changes are described in online supplementary Tables S3, S4, S5, and S6. We found that HL negatively impacted the rate of MD change in cpRNFLT.

Table 3.

Univariate and multivariate linear mixed model to evaluate the effect of the factors on the rate of MD change

Rate of MD change (dB/year)Patients, nEyes, nUnivariate linear mixed modelMultivariate linear mixed model
β3’’ (mean±SE [95% CI])p valueβ3’’’ (mean±SE [95% CI])p value
DM 10 17 0.28±0.097 [0.085–0.47] 0.0058 0.35±0.097 [0.16–0.54] 0.0006 
HT 19 26 0.0074±0.089 [−0.17 to 0.19] 0.93 0.019±0.089 [−0.16 to 0.20] 0.83 
HL 14 21 0.11±0.091 [−0.070 to 0.30] 0.22 0.18±0.084 [0.012–0.35] 0.036 
CVD 16 22 0.013±0.093 [−0.17 to 0.20] 0.89 0.087±0.093 [−0.18 to 0.19] 0.93 
Arrhythmia 12 −0.10±0.12 [−0.35 to 0.14] 0.41 −0.0007±0.11 [−0.23 to 0.23] 1.00 
SAS −0.15±0.23 [−0.62 to 0.32] 0.52 −0.18±0.20 [−0.59 to 0.23] 0.38 
Orthopedic disease 18 27 −0.21±0.087 [−0.39 to −0.038] 0.018 −0.12±0.086 [−0.29 to 0.055] 0.18 
Malignant tumor 10 14 0.027±0.12 [−0.22 to 0.27] 0.83 0.18±0.11 [−0.046 to 0.41] 0.12 
DH 15 19 −0.12±0.096 [−0.31 to 0.0075] 0.23 −0.048±0.093 [−0.23 to 0.14] 0.61 
Rate of MD change (dB/year)Patients, nEyes, nUnivariate linear mixed modelMultivariate linear mixed model
β3’’ (mean±SE [95% CI])p valueβ3’’’ (mean±SE [95% CI])p value
DM 10 17 0.28±0.097 [0.085–0.47] 0.0058 0.35±0.097 [0.16–0.54] 0.0006 
HT 19 26 0.0074±0.089 [−0.17 to 0.19] 0.93 0.019±0.089 [−0.16 to 0.20] 0.83 
HL 14 21 0.11±0.091 [−0.070 to 0.30] 0.22 0.18±0.084 [0.012–0.35] 0.036 
CVD 16 22 0.013±0.093 [−0.17 to 0.20] 0.89 0.087±0.093 [−0.18 to 0.19] 0.93 
Arrhythmia 12 −0.10±0.12 [−0.35 to 0.14] 0.41 −0.0007±0.11 [−0.23 to 0.23] 1.00 
SAS −0.15±0.23 [−0.62 to 0.32] 0.52 −0.18±0.20 [−0.59 to 0.23] 0.38 
Orthopedic disease 18 27 −0.21±0.087 [−0.39 to −0.038] 0.018 −0.12±0.086 [−0.29 to 0.055] 0.18 
Malignant tumor 10 14 0.027±0.12 [−0.22 to 0.27] 0.83 0.18±0.11 [−0.046 to 0.41] 0.12 
DH 15 19 −0.12±0.096 [−0.31 to 0.0075] 0.23 −0.048±0.093 [−0.23 to 0.14] 0.61 

We adjusted for age, gender, AL, mean IOP, and SD of IOP in the multivariate model.

CVD, cardiovascular disease; DH, disc hemorrhage; DM, diabetes mellitus; HL, hyperlipidemia; HT, hypertension; MD, mean deviation of the Humphrey field analyzer; SAS, sleep apnea syndrome; SE, standard error.

Generalized estimation equations were used to compare the baseline characteristics between the groups with and without DM or HL (Tables 4, 5). Table 4 shows that the male ratio was significantly higher in the group with DM than in the group without DM (p = 0.0051). In addition, the SD of IOP was significantly lower in the group with DM than in the group without DM (p = 0.024). As shown in Table 5, age, CCT, and baseline MD were significantly higher in the group with HL than in the group without HL (p = 0.0088, 0.027, and 0.034, respectively).

Table 4.

Comparison of characteristics of patients between two groups divided by presence of DM

DM (+)DM (−)p value
mean±SEmean±SE
Subjects 10 58  
Eye 17 81  
Male (eye) 15 42 0.0051 
Age, years 63.6±8.8 56.9±14.5 0.051 
AL, mm 24.49±1.14 25.20±1.36 0.08 
Mean IOP, mm Hg 11.4±1.1 11.9±1.0 0.24 
SD of IOP, mm Hg 1.3±0.2 1.4±0.4 0.024 
Max IOP, mm Hg 13.8±1.2 14.4±0.8 0.072 
CCT, μm 538.8±38.3 515.6±31.2 0.068 
Baseline MD of HFA, dB −7.45±8.86 −10.22±7.60 0.33 
DM (+)DM (−)p value
mean±SEmean±SE
Subjects 10 58  
Eye 17 81  
Male (eye) 15 42 0.0051 
Age, years 63.6±8.8 56.9±14.5 0.051 
AL, mm 24.49±1.14 25.20±1.36 0.08 
Mean IOP, mm Hg 11.4±1.1 11.9±1.0 0.24 
SD of IOP, mm Hg 1.3±0.2 1.4±0.4 0.024 
Max IOP, mm Hg 13.8±1.2 14.4±0.8 0.072 
CCT, μm 538.8±38.3 515.6±31.2 0.068 
Baseline MD of HFA, dB −7.45±8.86 −10.22±7.60 0.33 

AL, axial length; CCT, central corneal thickness; DM, diabetes mellitus; IOP, intraocular pressure; MD, mean deviation of the Humphrey field analyzer; SD, standard deviation; SE, standard error.

Table 5.

Comparison of characteristics of patients between two groups divided by presence of HL

HL (+)HL (−)p value
mean±SEmean±SE
Subjects 14 54  
Eye 21 77  
Male gender (reference, female) 13 44 0.75 
Age, years 64.6±8.7 56.3±14.5 0.009 
AL, mm 24.87±1.17 25.13±1.39 0.49 
Mean IOP, mm Hg 11.8±0.9 11.8±1.0 0.76 
SD of IOP, mm Hg 1.4±0.3 1.4±0.4 0.74 
Max IOP, mm Hg 14.2±1.0 14.3±0.9 0.73 
CCT, μm 533.5±30.8 515.1±32.8 0.027 
Baseline MD of HFA, dB −6.75±6.26 −10.55±8.08 0.034 
HL (+)HL (−)p value
mean±SEmean±SE
Subjects 14 54  
Eye 21 77  
Male gender (reference, female) 13 44 0.75 
Age, years 64.6±8.7 56.3±14.5 0.009 
AL, mm 24.87±1.17 25.13±1.39 0.49 
Mean IOP, mm Hg 11.8±0.9 11.8±1.0 0.76 
SD of IOP, mm Hg 1.4±0.3 1.4±0.4 0.74 
Max IOP, mm Hg 14.2±1.0 14.3±0.9 0.73 
CCT, μm 533.5±30.8 515.1±32.8 0.027 
Baseline MD of HFA, dB −6.75±6.26 −10.55±8.08 0.034 

AL, axial length; CCT, central corneal thickness; HL, hyperlipidemia; IOP, intraocular pressure; MD, mean deviation of the Humphrey field analyzer; SD, standard deviation; SE, standard error.

This study showed that age, AL, DM, orthopedic disease, and HL may be involved in VF deterioration. HL is a lifestyle-related disease that has become a major health concern in recent years. Although there has been much debate regarding its association with the development of POAG, little has been reported regarding its association with POAG progression. In meta-analyses wherein HL significantly increased the incidence of POAG [18], subgroup analysis revealed no significant association between NTG development and HL. It has also been reported that dyslipidemia may not be significantly associated with NTG development [19]. Although HL has been suggested to increase IOP [20], further research is needed to determine the association between HL and glaucoma progression and development in patients with lower IOP. In this study, we found a significant positive effect of HL on the rate of MD change, which may be due to the action of statins. Statins, which are used to treat HL, may exert a protective effect against glaucoma [21]. Although we could not access the details of the treatment used in the HL cases in this study, it is possible that the use of statins may have contributed to these results. On the contrary, we also researched the effect of HL on the rate of change in cpRNFLT. We found that HL had a significant negative effect on the rate of change in cpRNFLT. RNFL thickness is associated with VF deterioration, and RNFL thinning may be involved in the rate of MD change [22], which conflicts with the effect of HL on the rate of MD change. Further studies are warranted to accurately investigate the effect of HL on glaucoma with lower IOP; however, the seemingly protective effect on VF deterioration may be due to the IOP, which will be discussed later.

The relationship between POAG and DM has been the subject of several studies. However, it is not fully understood how they influence each other. A meta-analysis by Zhou et al. [8] concluded that DM was a risk factor for the development of POAG. However, few studies have investigated the effect of DM on glaucoma progression. This study aimed to determine the effect of DM on VF deterioration in POAG patients with low IOP. The results showed a significant positive effect of DM on VF deterioration. This finding suggests that DM may not be a risk factor for VF deterioration in patients with POAG. A hypothesis supporting this result is that VF deterioration in many patients with DM is not due to glaucoma but is a non-glaucomatous progression due to diabetic neuropathy. According to Chihara et al. [23], patients with very early diabetes are prone to nerve fiber layer loss in the retina because of reduced protein synthesis before the onset of clinically evident microvascular abnormalities [24]. In this study, patients diagnosed with diabetic retinopathy were excluded; therefore, patients with diabetes who had progressed to the point of obvious vascular damage were excluded. Therefore, it is possible that many patients with diabetes had neuropathy that did not progress to microvascular disorders, which may have led to the above results. In addition, the mechanisms that do not involve elevated IOP as a risk factor for POAG in patients with diabetes have not been considered. In contrast, it has been hypothesized that DM may have a neuroprotective effect. This hypothesis suggests that diabetes-induced disruption of the blood-retinal barrier leads to leakage of neuroprotective vascular endothelial growth factor and induces preischemic survival mechanisms, and that diabetes-induced glycation of connective tissue reinforces liquid crystals and protects nerve fibers [25]. Therefore, the association between POAG and DM remains controversial. Thus, further investigation of the association between POAG and DM is warranted.

This study also found that orthopedic diseases may have a significant negative impact on POAG progression. Although we could not find any reports of an association between orthopedic diseases and POAG, the association between vitamin D and POAG has also received attention in recent years. Several studies have suggested that decreased serum 25-hydroxy vitamin D levels may be a risk factor for the development or progression of POAG [26‒28]. The results of the present study are consistent with these findings because patients with diseases such as bone fractures and scoliosis are often deficient in vitamin D. Several hypotheses have been proposed regarding the function of vitamin D in POAG, one of which is its action via the immune system. Recent studies have shown that imbalances in the immune system are a major cause of neurodegenerative injury to optic nerve axons and ganglion cell bodies [29, 30]. As vitamin D has a significant effect on the regulation of immune cell function, it may play an important role in the protection of the optic nerve. Note that there are still many questions regarding the relationship between POAG and vitamin D. It has been hypothesized that low vitamin D levels in patients with glaucoma may be due to reduced external activity caused by the disease. In addition, the age-corrected model used in this study was not statistically significant. As aging is positively correlated with decreased vitamin D levels, it is possible that the present results reflect the effects of aging.

There was no negative effect of mean IOP on the rate of MD changes in this study. However, there may have been a potential effect of CCT on IOP. Tables 4 and 5 suggest that DM and HL had marginally significant positive effects on CCT. The CCT of patients with diabetes is significantly thicker than that of those without [31]. It is also known that the thicker the central cornea, the higher the IOP [26]. Therefore, the actual baseline IOP may differ between patients with and without a history of DM or HL. In the first group, the actual IOP was lower than the measured IOP, whereas in the second group, the actual IOP was higher than the measured IOP. Therefore, VF deterioration in the latter group may have been more negatively affected by IOP, resulting in a significantly positive effect of DM and HL on the rate of MD change.

The finding that aging is a risk factor for glaucoma progression and that myopia is not a risk factor is consistent with those of previous reports [7, 10]. A possible reason why aging is related to the progression of glaucoma is that the elderly are more susceptible to elevated IOP and that RNFL thins with age, even in non-glaucomatous eyes, and, to a greater extent, with advancing age [32]. A nonlinear correlation exists between function and structure in glaucomatous eyes [33]. That is, with age-related thinning of the RNFL, VF deterioration progresses significantly with slight thinning; therefore, aging can be a risk factor for glaucoma progression. It has been suggested that myopia is a risk factor for the development of glaucoma but may be a protective factor against its progression [7]. Therefore, the finding that a longer AL is associated with less progressive VF deterioration is consistent with those of previous reports.

This study had some limitations. First, this was a single-center retrospective cohort study. Therefore, the sample size was small, and the results were difficult to interpret because causal relationships were not known. Second, participants were not interviewed systematically. As we only used information derived from patients’ medical records, it is possible that some patients who were suffering from the disease in question were not interviewed. In addition, we did not have access to the detailed oral medication status of the participants. Therefore, it is necessary to consider that oral medications may have influenced the results. Third, there were cases in which the IOP was controlled by medication. In this study, we included patients with a lower IOP during the observation period. However, we also included patients whose IOP was controlled using glaucoma medication. Therefore, it is possible that disease progression was controlled by the drugs. A multicenter prospective study that investigated the risk factors for glaucoma progression in untreated patients with NTG reported that DH, long-term IOP changes, and vertical cup-to-disc ratio contribute to progression [12]. In addition, because the study did not evaluate daily variations in IOP measurement methods, it is not possible to conclude that the subjects had a lower IOP. Further prospective studies with larger sample sizes are needed to identify risk factors for POAG progression.

In conclusion, this study proposes that DM or HL may be associated with VF deterioration in patients having glaucoma with lower IOP. However, this association may be due to differences in IOP characteristics, such as CCT or SD, rather than blood flow reduction related to atherosclerosis or disturbance of the microcirculation caused by DM or HL. Further research is needed to understand the pathology of glaucoma with low IOP and to develop new treatment options other than reducing IOP.

This retrospective study adhered to the tenets of the Declaration of Helsinki. Institutional Review Board (IRB)/Ethics Committee approval was obtained from the Kyoto University Graduate School of Medicine (approval number [R2652]). This retrospective study adhered to the tenets of the Declaration of Helsinki, and Institutional Review Board (IRB)/Ethics Committee approval was obtained from the Kyoto University Graduate School of Medicine (approval number [R2652]). The requirement for written informed consent was waived due to the retrospective design of this study. Instead, we publicly disclosed results of this retrospective research on our faculty’s website, while also providing the subjects an opportunity to opt out of the study. Under this condition, the Institutional Review Board and Ethics Committee of Kyoto University Graduate School of Medicine approved the waiver for informed consent.

The authors have no conflicts of interest to declare.

This study was not supported by any sponsor or funder.

Y.U. and K.S. contributed to the study design, data analysis, and interpretation. K.S., T.K., H.O.I., M.M., T.H., and S.N. contributed to data acquisition and drafting of the manuscript. A.T. supervised the project. All authors approved the final version of the manuscript.

The data that support the findings of this study are not publicly available owing to the presence of information that could compromise the privacy of research participants but are available from the corresponding author (K.S.) upon reasonable request.

1.
GBD 2019 Blindness and Vision Impairment CollaboratorsVision Loss Expert Group of the Global Burden of Disease Study
.
Causes of blindness and vision impairment in 2020 and trends over 30 years, and prevalence of avoidable blindness in relation to VISION 2020: the Right to Sight: an analysis for the Global Burden of Disease Study
.
Lancet Glob Health
.
2021
;
9
(
2
):
e144
60
. .
2.
Morizane
Y
,
Morimoto
N
,
Fujiwara
A
,
Kawasaki
R
,
Yamashita
H
,
Ogura
Y
, et al
.
Incidence and causes of visual impairment in Japan: the first nation-wide complete enumeration survey of newly certified visually impaired individuals
.
Jpn J Ophthalmol
.
2019
;
63
(
1
):
26
33
. .
3.
Ocular Hypertension Treatment Study Group
,
European Glaucoma Prevention Study Group
,
Gordon
MO
,
Torri
V
,
Miglior
S
,
Beiser
JA
, et al
.
Validated prediction model for the development of primary open-angle glaucoma in individuals with ocular hypertension
.
Ophthalmology
.
2007
;
114
(
1
):
10
9
. .
4.
Jay
JL
,
Murdoch
JR
.
The rate of visual field loss in untreated primary open angle glaucoma
.
Br J Ophthalmol
.
1993
;
77
(
3
):
176
8
. .
5.
Yamamoto
T
,
Iwase
A
,
Araie
M
,
Suzuki
Y
,
Abe
H
,
Shirato
S
, et al
.
The Tajimi Study report 2: prevalence of primary angle closure and secondary glaucoma in a Japanese population
.
Ophthalmology
.
2005
;
112
(
10
):
1661
9
. .
6.
Wormald
RPL
,
Basauri
E
,
Wright
LA
,
Evans
JR
.
The African Caribbean eye survey: risk factors for glaucoma in a sample of African Caribbean people living in London
.
Eye
.
1994
;
8
(
3
):
315
20
. .
7.
Wu
J
,
Hao
J
,
Du
Y
,
Cao
K
,
Lin
C
,
Sun
R
, et al
.
The association between myopia and primary open-angle glaucoma: a systematic review and meta-analysis
.
Ophthalmic Res
.
2022
;
65
(
4
):
387
97
. .
8.
Zhou
M
,
Wang
W
,
Huang
W
,
Zhang
X
.
Diabetes mellitus as a risk factor for open-angle glaucoma: a systematic review and meta-analysis
.
PLoS One
.
2014
;
9
(
8
):
e102972
. .
9.
O’Brien
JM
,
Salowe
RJ
,
Fertig
R
,
Salinas
J
,
Pistilli
M
,
Sankar
PS
, et al
.
Family history in the primary open-angle African American glaucoma genetics study cohort
.
Am J Ophthalmol
.
2018
;
192
:
239
47
. .
10.
Heijl
A
,
Bengtsson
B
,
Hyman
L
,
Leske
MC
;
Early Manifest Glaucoma Trial Group
.
Natural history of open-angle glaucoma
.
Ophthalmology
.
2009
;
116
(
12
):
2271
6
. .
11.
Guo
ZZ
,
Chang
K
,
Wei
X
.
Intraocular pressure fluctuation and the risk of glaucomatous damage deterioration: a meta-analysis
.
Int J Ophthalmol
.
2019
;
12
(
1
):
123
8
. .
12.
Sakata
R
,
Yoshitomi
T
,
Iwase
A
,
Matsumoto
C
,
Higashide
T
,
Shirakashi
M
, et al
.
Factors associated with progression of Japanese open-angle glaucoma with lower normal intraocular pressure
.
Ophthalmology
.
2019
;
126
(
8
):
1107
16
. .
13.
Founti
P
,
Bunce
C
,
Khawaja
AP
,
Doré
CJ
,
Mohamed-Noriega
J
,
Garway-Heath
DF
, et al
.
Risk factors for visual field deterioration in the United Kingdom glaucoma treatment study
.
Ophthalmology
.
2020
;
127
(
12
):
1642
51
. .
14.
Collaborative Normal-Tension Glaucoma Study Group
.
Comparison of glaucomatous progression between untreated patients with normal-tension glaucoma and patients with therapeutically reduced intraocular pressures
.
Am J Ophthalmol
.
1998
;
126
(
4
):
487
97
.
15.
Anderson
DR
,
Patella
VM
.
Automated static perimetry
. 2nd ed.
Mosby
;
1999
.
16.
Suda
K
,
Hangai
M
,
Akagi
T
,
Noma
H
,
Kimura
Y
,
Hasegawa
T
, et al
.
Comparison of longitudinal changes in functional and structural measures for evaluating progression of glaucomatous optic neuropathy
.
Invest Ophthalmol Vis Sci
.
2015
;
56
(
9
):
5477
84
. .
17.
Suda
K
,
Akagi
T
,
Nakanishi
H
,
Noma
H
,
Ikeda
HO
,
Kameda
T
, et al
.
Evaluation of structure-function relationships in longitudinal changes of glaucoma using the Spectralis OCT follow-up mode
.
Sci Rep
.
2018
;
8
(
1
):
17158
. .
18.
Wang
S
,
Bao
X
.
Hyperlipidemia, blood lipid level, and the risk of glaucoma: a meta-analysis
.
Invest Ophthalmol Vis Sci
.
2019
;
60
(
4
):
1028
43
. .
19.
Kim
M
,
Jeoung
JW
,
Park
KH
,
Oh
WH
,
Choi
HJ
,
Kim
DM
.
Metabolic syndrome as a risk factor in normal-tension glaucoma
.
Acta Ophthalmol
.
2014
;
92
(
8
):
e637
43
. .
20.
Madjedi
KM
,
Stuart
KV
,
Chua
SYL
,
Luben
RN
,
Warwick
A
,
Pasquale
LR
, et al
.
The association between serum lipids and intraocular pressure in 2 Large United Kingdom cohorts
.
Ophthalmology
.
2022
;
129
(
9
):
986
96
. .
21.
Mcgwin
G
,
Mcneal
S
,
Owsley
C
,
Girkin
C
,
Epstein
D
,
Lee
PP
.
Statins and other cholesterol-lowering medications and the presence of glaucoma
.
Arch Ophthalmol
.
2004
;
122
(
6
):
822
6
. .
22.
Swaminathan
SS
,
Jammal
AA
,
Berchuck
SI
,
Medeiros
FA
.
Rapid initial OCT RNFL thinning is predictive of faster visual field loss during extended follow-up in glaucoma
.
Am J Ophthalmol
.
2021
;
229
:
100
7
. .
23.
Chihara
E
.
Impairment of protein synthesis in the retinal tissue in diabetic rabbits: secondary reduction of fast axonal transport
.
J Neurochem
.
1981
;
37
(
1
):
247
250
. .
24.
Levine
SR
,
Sapieha
P
,
Dutta
S
,
Sun
JK
,
Gardner
TW
.
It is time for a moonshot to find “Cures” for diabetic retinal disease
.
Prog Retin Eye Res
.
2022
;
90
:
101051
. .
25.
Quigley
HA
.
Can diabetes be good for glaucoma? Why can’t we believe our own eyes (or data)
.
Arch Ophthalmol
.
2009
;
127
(
2
):
227
9
. .
26.
Kim
HT
,
Kim
JM
,
Kim
JH
,
Lee
MY
,
Won
YS
,
Lee
JY
, et al
.
The relationship between vitamin D and glaucoma: a Kangbuk Samsung health study
.
Korean J Ophthalmol
.
2016
;
30
(
6
):
426
33
. .
27.
Lv
Y
,
Yao
Q
,
Ma
W
,
Liu
H
,
Ji
J
,
Li
X
.
Associations of vitamin D deficiency and vitamin D receptor (Cdx-2, Fok I, Bsm I and Taq I) polymorphisms with the risk of primary open-angle glaucoma
.
BMC Ophthalmol
.
2016
;
16
:
116
. .
28.
Goncalves
A
,
Milea
D
,
Gohier
P
,
Jallet
G
,
Leruez
S
,
Baskaran
M
, et al
.
Serum vitamin D status is associated with the presence but not the severity of primary open angle glaucoma
.
Maturitas
.
2015
;
81
(
4
):
470
4
. .
29.
Tsai
T
,
Reinehr
S
,
Maliha
AM
,
Joachim
SC
.
Immune mediated degeneration and possible protection in glaucoma
.
Front Neurosci
.
2019
;
13
:
931
. .
30.
Wang
L
,
Wei
X
.
T cell-mediated autoimmunity in glaucoma neurodegeneration
.
Front Immunol
.
2021
;
12
:
803485
. .
31.
Kim
YJ
,
Kim
TG
.
The effects of type 2 diabetes mellitus on the corneal endothelium and central corneal thickness
.
Sci Rep
.
2021
;
11
(
1
):
8324
. .
32.
Peng
PH
,
Hsu
SY
,
Wang
WS
,
Ko
ML
.
Age and axial length on peripapillary retinal nerve fiber layer thickness measured by optical coherence tomography in nonglaucomatous Taiwanese participants
.
PLoS One
.
2017
;
12
(
6
):
e0179320
. .
33.
Hood
DC
,
Kardon
RH
.
A framework for comparing structural and functional measures of glaucomatous damage
.
Prog Retin Eye Res
.
2007
;
26
(
6
):
688
710
. .