Predictors of Receiving Surgical Treatment for Neovascular Glaucoma in the California Medicare Population
1
David Geffen School of Medicine, University of California Los Angeles, Los Angeles, CA 90095, USA
2
Department of Ophthalmology, Stein and Doheny Eye Institutes, University of California Los Angeles, Los Angeles, CA 90095, USA
3
Department of Biostatistics, Fielding School of Public Health, University of California Los Angeles, Los Angeles, CA 90095, USA
*
Author to whom correspondence should be addressed.
J. Clin. Transl. Ophthalmol. 2026, 4(2), 15; https://doi.org/10.3390/jcto4020015
Submission received: 17 February 2026
/
Revised: 22 April 2026
/
Accepted: 3 June 2026
/
Published: 8 June 2026
Abstract
Background: Population-level predictors of intraocular pressure (IOP)-lowering surgery for neovascular glaucoma (NVG) are understudied. This study examines factors associated with IOP-lowering surgery in California (CA) Medicare beneficiaries with NVG. Methods: The study population included all 2019 CA Medicare beneficiaries with NVG. Covariates included age, sex, race/ethnicity, history of treatments for retinal ischemia, dual Medicare/Medicaid eligibility, Social Vulnerability Index score, and Charlson Comorbidity Index (CCI) score. Outcomes included incidence of trabeculectomy, tube shunt, minimally invasive glaucoma surgery, cyclophotocoagulation (CPC), or any IOP-lowering surgery. Results: Of 1843 beneficiaries, 264 (14.3%) had IOP-lowering surgeries. In multivariable logistic regression including all covariates, CCI ≥ 5 versus 0 was associated with lower odds of any IOP-lowering surgery and of each type of surgery except CPC (adjusted odds ratio [aOR] = 0.47, 95% confidence interval [CI] = 0.29, 0.75 for any versus no surgery; aOR = 1.35, CI = 0.51, 3.60 for CPC). Compared to Non-Hispanic White, racial and ethnic minorities had increased odds of trabeculectomy (aOR = 3.77, CI = 1.05, 13.57 for Black; aOR = 2.69, CI = 1.04, 6.92 for Hispanic) and tube shunt (aOR = 2.62, CI = 1.27, 5.41 for Other/Unknown). Beneficiaries 75–79 versus 65–69 years old had decreased odds of trabeculectomy (aOR = 0.21, CI = 0.05, 0.98). Conclusions: Among CA Medicare beneficiaries, higher systemic disease burden was associated with a decreased likelihood of surgery for NVG, while racial and ethnic minorities were more likely to undergo certain procedures. These findings suggest surgical risk stratification and treatment pattern disparities for individuals with NVG.
1. Introduction
Neovascular glaucoma (NVG) is an aggressive and often blinding secondary glaucoma that is characterized by neovascularization of the iris and angle with intraocular pressure (IOP) elevation [1]. NVG most often presents secondary to retinal ischemic diseases, including proliferative diabetic retinopathy (PDR), central retinal vein occlusion (RVO), and ocular ischemic syndrome (OIS) [2,3]. The occurrence of widespread retinal ischemia and a substantial loss of capillary perfusion trigger the release of various angiogenic growth factors, particularly vascular endothelial growth factor (VEGF), which in turn can lead to the development of abnormal new blood vessels [4]. To treat IOP elevation associated with NVG, the medical management of IOP reduction and treatment of retinal ischemia may be inadequate, and surgical intervention to reduce IOP is often indicated [5]. These procedures include filtration surgery such as trabeculectomy or tube shunts, cyclophotocoagulation (CPC) to reduce aqueous humor formation from the ciliary body, or, less often, minimally invasive glaucoma surgery (MIGS) [5,6,7]. Even with the prompt recognition and early treatment of retinal ischemia and elevated IOP, individuals with NVG often have a high risk of rapid progression toward irreversible blindness [8].
Previous studies have identified racial and ethnic disparities in the prevalence and treatment of NVG in individuals with retinal ischemia. A study in the American Academy of Ophthalmology (AAO) Intelligent Research in Sight (IRIS®) Registry revealed that individuals of Black and Asian race and ethnicity were least likely to receive an anti-VEGF injection within 12 months after RVO diagnosis compared to Non-Hispanic White individuals [9]. Additionally, a recent investigation in the AAO IRIS Registry found that the eyes of racial and ethnic minorities with retinal ischemia had a higher likelihood of NVG as well as a higher likelihood of IOP-lowering surgery for NVG [10]. While these studies suggest that there are likely racial and ethnic disparities in the treatment of retinal ischemia and IOP for individuals with NVG, there is a limited understanding of how additional systemic, demographic, and socioeconomic factors are associated with various types of surgical treatments for NVG in additional populations. To this end, the purpose of the present study was to examine systemic, demographic, and socioeconomic predictors associated with various surgical treatments for IOP reduction in individuals with NVG within the large and diverse California Medicare population.
2. Materials and Methods
2.1. Study Population
This cross-sectional retrospective study included all 2019 California Medicare beneficiaries ≥ 65 years of age who were enrolled in both Part A and Part B with an International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) code assessed based on ICD-10-CM codes for “glaucoma secondary to other eye disorders” and/or “other specified glaucoma” in conjunction with codes for retinal ischemia, including PDR, RVO, and/or OIS, as there was no specific ICD-10-CM code for NVG in 2019 [11]. Beneficiaries in the study population included all individuals with fee-for-service plans and a small proportion of beneficiaries with Medicare Advantage plans. This study was in accordance with the Declaration of Helsinki and its later amendments and was declared exempt from the Institutional Review Board review at the University of California Los Angeles.
2.2. Assessment of Covariates
Demographic, socioeconomic, and systemic factors examined as covariates included age, sex, race and ethnicity, history of treatments for retinal ischemia, dual Medicare/Medicaid eligibility, Centers for Disease Control and Prevention Social Vulnerability Index (SVI) score, and Charlson Comorbidity Index (CCI) score [12,13]. Age was classified into five-year groups based on beneficiary age in 2019: 65–69, 70–74, 75–79, 80–84, and 85+. Sex was classified as male or female. Race and ethnicity were classified as Non-Hispanic White, Black, Asian, Hispanic/Latino, and Other/Unknown based on the Research Triangle Institute Race Code [14]. History of anti-VEGF injection and panretinal photocoagulation in 2019 were assessed based on Current Procedural Terminology (CPT) code and coded as binary variables. Dual Medicare/Medicaid eligibility status was assessed as a binary variable as a proxy for low socioeconomic status. The SVI score was calculated using 16 variables downloaded from the 2018 American Community Survey based on theme classifications of socioeconomic status, household composition and disability, minority status and language, and housing type and transportation [15]. For the present study, the overall SVI score was included as a covariate, while theme-specific SVI scores were not. Each SVI score was calculated at the zip code level and categorized into quartiles for the 2019 California Medicare population [16]. For each zip code, percentile ranks were calculated for the 16 variables, the 4 themes, and the overall position. These ranks were summed and ordered to generate theme and overall percentile ranks. The results were categorized into quartiles based on the 25th, 50th, 75th, and 100th percentile values for the 2019 California Medicare population. Higher SVI scores indicated greater social vulnerability. The CCI score quantifies systemic disease burden by assigning a weighted score that reflects the presence or absence of various medical conditions, including myocardial infarction, congestive heart failure, peripheral vascular disease, cerebrovascular disease, dementia, chronic pulmonary disease, rheumatologic disease, peptic ulcer disease, cirrhosis, hepatic failure, immunosuppression, diabetes mellitus with or without complications, hemi-/paraplegia, chronic renal disease, malignant neoplasms, multiple myeloma/leukemia, lymphomas, metastatic solid tumor, and acquired immune deficiency syndrome (AIDS) [12]. Individual conditions for the CCI were assessed based on the ICD-10-CM code, as shown in Supplemental Table S1, and CCI scores were categorized into 0, 1–2, 3–4, and 5+, with a higher score representing higher systemic disease burden [17]. Three additional binary covariates were included as claims-based proxies for the prior treatment of retinal ischemia and socioeconomic status: dual Medicare–Medicaid eligibility, any intravitreal anti-VEGF injection, and any panretinal photocoagulation (PRP).
2.3. Outcome
The outcome of interest was the incidence of surgical treatments for IOP reduction, which were defined by the Current Procedural Terminology (CPT) code [18]. Surgeries that were examined included trabeculectomy, tube shunt, MIGS, and CPC, as shown in Supplemental Table S2. Any IOP-lowering surgery was defined as the presence of at least one of these types of surgeries. Although included in the primary outcome for the completeness of surgical capture within claims data, due to the lower likelihood of MIGS as a treatment modality for NVG in relation to the pathophysiology of the NVG disease process, a secondary outcome excluding MIGS from the definition of any IOP-lowering surgery was also created.
2.4. Statistical Analyses
Descriptive statistics were used to compare study covariates in beneficiaries with NVG with and without any IOP-lowering surgery using chi-squared tests. Logistic regression modeling was used to examine associations between study covariates with the incidence of each type of IOP-lowering surgery in the entire study population. First, unadjusted models were performed to separately examine associations between age, sex, race and ethnicity; CCI score; SVI score; and each type of IOP-lowering surgery in the study population. Next, multivariable regression models were performed by including all these covariates in the same model, with each type of surgery as a separate outcome, as well as any IOP-lowering surgery as an outcome. A secondary analysis was performed to examine associations between all study covariates and odds of any IOP-lowering surgery, excluding MIGS, in the entire study population, and additionally in beneficiaries with PDR and NVG only, and beneficiaries with RVO and NVG only. This analysis was not performed in beneficiaries with OIS and NVG only due to insufficient sample size. To further investigate the interaction between racial and ethnic differences in IOP-lowering surgery with additional demographic, socioeconomic, and systemic factors, stratified analysis by race and ethnicity was performed, including age, sex, CCI score, and SVI score as covariates and any IOP-lowering surgery as the outcome. To assess robustness to unmeasured confounding, we calculated E-values for primary adjusted associations, which represent the minimum association strength that an unmeasured confounder would need with both the exposure and the outcome, conditional on measured covariates, to fully explain the observed effect estimate [19]. All statistical analyses were performed using SAS version 9.4.
3. Results
Baseline characteristics of the study population are outlined in Table 1. The study population included 1843 beneficiaries with NVG, of whom 831 had PDR (45.1%), 1008 had RVO (54.7%), and 192 had OIS (10.4%). There were 264 (14.3%) beneficiaries who had any IOP-lowering surgery for NVG. The largest proportion of the study population was 70–74 years old (n = 425; 23.1%), male (n = 990; 53.7%), and Non-Hispanic White (n = 733; 39.8%), and had a CCI score of 5+ (n = 693; 37.6%). The mean SVI score was 0.63 ± 0.27 (scale 0–1). In beneficiaries with versus without any IOP-lowering surgery, there was a statistically significant difference in the CCI score distribution (p = 0.01), but no statistically significant differences in the distribution of other demographic and systemic factors. Among the study population, 858 (46.6%) were dually enrolled in Medicare and Medicaid, 828 (44.9%) had received any intravitreal anti-VEGF injection, and 280 (15.2%) had received PRP; in beneficiaries with versus without any IOP-lowering surgery, there were statistically significant differences in the distribution of intravitreal injection (p = 0.01) and PRP (p = 0.01), but not dual Medicaid eligibility.
Table 2 and Table A1 (Appendix A) describe results from univariable comparisons and multivariable logistic regression analyses of NVG surgery by study covariates, including age group, sex, race and ethnicity, CCI score, SVI quartiles, dual Medicare–Medicaid eligibility, intravitreal injection, and PRP. In unadjusted analyses, beneficiaries aged 75–79 had a significantly lower likelihood of trabeculectomy compared to those aged 65–69, which remained significant in the adjusted analyses (adjusted odds ratio [aOR] = 0.21; 95% CI = 0.05, 0.98). For the unadjusted analyses on race and ethnicity, Hispanic/Latino beneficiaries had significantly higher odds of trabeculectomy, and beneficiaries identifying as Other/Unknown had significantly higher odds of tube shunt. These findings remained significant in adjusted analyses (aOR = 2.69; 95% CI = 1.04, 6.92 for trabeculectomy in Hispanic/Latino; aOR = 2.62; 95% CI = 1.27, 5.41 for tube shunt in Other/Unknown), with the addition of Black beneficiaries having significantly higher odds of trabeculectomy compared to the Non-Hispanic White beneficiaries (aOR = 3.77; 95% CI = 1.05, 13.57). For CCI, there was a significantly lower likelihood of receiving any type of IOP-lowering surgery among beneficiaries with a CCI score of 5+ as well as a significantly lower likelihood of MIGS when compared to beneficiaries with a CCI score of 0. These findings were consistent in the adjusted analyses (aOR = 0.47; 95% CI = 0.29, 0.75; E-value = 3.68 for any IOP-lowering surgery; aOR = 0.26; 95% CI = 0.10, 0.72 for MIGS), with the addition of significantly lower likelihoods of trabeculectomy (aOR = 0.25; 95% CI = 0.08, 0.79) and tube shunt (aOR = 0.46; 95% CI = 0.25, 0.87) when compared to beneficiaries with CCI score of 0, but not of CPC (aOR = 1.35; 95% CI = 0.51, 3.60). Beneficiaries with CCI scores of 1–2 also had a lower likelihood of trabeculectomy (aOR = 0.24; 95% CI = 0.07, 0.81) compared to beneficiaries with a CCI score of 0 in the adjusted analyses. In secondary analyses excluding MIGS from the definition of IOP-lowering surgery (Appendix A Table A2), factors associated with decreased odds of any IOP-lowering surgery in the overall study population included female versus male sex (aOR = 0.73; 95% CI = 0.54, 0.97) and CCI 5+ versus 0 (aOR = 0.56; 95% CI = 0.34, 0.94). In beneficiaries with RVO and NVG only, ages 80–84 (aOR = 0.40, 95% CI = 0.20, 0.80) and 85+ (aOR = 0.50, 95% CI = 0.28, 0.90) versus 65–69 years old were associated with decreased odds of IOP-lowering surgery, while Hispanic versus Non-Hispanic White race and ethnicity was associated with increased odds (aOR = 1.70, 95% CI = 1.03, 2.78). In unadjusted analyses, both prior intravitreal anti-VEGF injection (OR = 2.18; 95% CI = 1.67, 2.84) and prior PRP (OR = 2.66; 95% CI = 1.96, 3.61) were associated with significantly higher odds of any IOP-lowering surgery, while dual Medicare–Medicaid eligibility was not significant. These associations remained significant in adjusted analyses for intravitreal anti-VEGF injection (aOR = 1.79; 95% CI = 1.34, 2.39; E-value = 2.98) and PRP (aOR = 2.03; 95% CI = 1.46, 2.83; E-value = 3.48), while dual Medicare–Medicaid eligibility remained non-significant.
Table 3 summarizes results from multivariable logistic regression analyses stratified by race and ethnicity. Stratified analysis was not performed for beneficiaries of Other/Unknown race and ethnicity due to a small sample size (n = 75). In Non-Hispanic White beneficiaries, those in the second quartile of SVI had a significantly increased likelihood of any IOP-lowering surgery (aOR = 1.94; 95% CI = 1.18, 3.20) when compared to beneficiaries in the first SVI quartile. In Asian and Hispanic/Latino beneficiaries, there was a significantly decreased likelihood of any IOP-lowering surgery associated with a CCI score of 5+ versus 0 (aOR = 0.18; 95% CI = 0.05, 0.63 for Asian; aOR = 0.30; 95% CI = 0.12, 0.73 for Hispanic/Latino).
4. Discussion
This study explored the demographic, socioeconomic, and systemic factors associated with IOP-lowering surgeries for NVG in California Medicare beneficiaries. We found that higher systemic disease burden was associated with lower odds of undergoing trabeculectomy, tube shunt, and MIGS, but not of CPC. Black and Hispanic/Latino beneficiaries had higher odds of undergoing trabeculectomy, and beneficiaries of Other/Unknown race and ethnicity had increased odds of tube shunt. Similar trends were observed after excluding MIGS from the definition of IOP-lowering surgery and in beneficiaries with NVG associated with RVO. Among Asian and Hispanic/Latino beneficiaries, higher systemic disease burden was associated with decreased odds of undergoing any IOP-lowering surgery compared to beneficiaries with lower systemic disease burden.
The insufficient management of retinal ischemia may exacerbate NVG, thereby potentially increasing the need for surgical interventions. Some previous studies have explored the racial disparities associated with retinal ischemia treatments, such as the studies using the AAO IRIS registry that report racial disparities in the treatment of retinal ischemia, including a higher prevalence of NVG and increased odds of IOP-lowering surgery for Hispanic/Latino and Black individuals compared to Non-Hispanic Whites [9,10]. For procedural treatments, one study evaluated racial disparities in the administration of anti-VEGF intravitreal injections to treat many retinal vascular diseases, including RVO, among commercially insured patients and found that Asian patients were significantly less likely than Non-Hispanic White patients to receive such treatments, while Black and Hispanic/Latino patients were not [20].
Our study expands on the existing findings by investigating additional demographic, socioeconomic, and systemic factors and their association with IOP-lowering surgeries to treat NVG in a diverse and elderly population, revealing that increased systemic disease burden is associated with a decreased likelihood of undergoing multiple types of IOP-lowering surgery except for CPC. We hypothesize that higher systemic disease burden likely reflects decreased ability to undergo invasive surgery rather than NVG severity, warranting future risk-stratification studies [8]. The significant association between prior intravitreal anti-VEGF injection or PRP and increased odds of IOP-lowering surgery likely reflects confounding by disease severity or better access to care, as patients who required these treatments for retinal ischemia likely had more advanced underlying disease and were therefore at a higher risk of progressing to NVG requiring surgical intervention. This association may also reflect that patients with greater engagement with the healthcare system are more likely to receive both retinal ischemia treatments and surgical intervention for NVG, suggesting that access to care may be an important upstream determinant of surgical treatment in this population. Notably, dual Medicare–Medicaid eligibility was not independently associated with IOP-lowering surgery after adjustment, suggesting that socioeconomic disparities in NVG surgical treatment may be more fully captured by area-level vulnerability metrics such as the SVI than by insurance status alone. Consistent with previous studies, we observed higher odds of IOP-lowering surgeries among beneficiaries from racial and ethnic minorities, potentially reflecting a delayed treatment of retinal ischemia, leading to worse NVG outcomes. Disparities in care for other glaucoma types, particularly among Black and Hispanic/Latino patients, may also contribute to these trends [10,21,22]. Additionally, previous studies have shown associations between the loss of follow-up or inconsistent adherence to glaucoma management among Black and Hispanic/Latino adults, when compared to Non-Hispanic White adults, and the loss of follow-up anti-VEGF injections for patients with retinal ischemia [9,23,24]. We observed no significant findings for CPC, likely due to its use in end-stage glaucoma and its shorter procedure duration, which may be preferable for patients with high systemic disease burden [25,26]. Moreover, the modest attenuation in the strength of association when MIGS was excluded as an outcome may reflect that MIGS is more likely to be utilized in cases of relatively lower clinical severity, where trabecular outflow pathways remain at least partially functional, compared to more advanced NVG, where filtration surgery or CPC are the preferred interventions. This is consistent with the known pathophysiological rationale for limiting MIGS use in NVG and further supports the interpretation that the primary findings are not materially driven by MIGS inclusion [6,7]. While some findings reached statistical significance, readers should consider the magnitude of effect sizes when interpreting the clinical relevance of individual associations.
Stratification by race and ethnicity showed that higher systemic disease burden was associated with a decreased likelihood of IOP-lowering surgery for Asian and Hispanic/Latino beneficiaries, while Non-Hispanic White beneficiaries with higher social vulnerability had increased odds of surgery. We have reported above that higher CCI scores were associated with a lower likelihood of IOP reduction surgeries except CPC. Such racial disparities could be due to the underutilization of ophthalmological care among Asians and Hispanics/Latinos when they are systematically ill compared to other racial and ethnic groups. This theory aligns with a previous study on diabetic patients that reports a reduced likelihood of eye exams among racial minorities [27]. There are some potential explanations regarding the observation that Non-Hispanic White beneficiaries with higher social vulnerability had increased odds of any surgery, while no statistically significant associations were observed by SVI score for racial and ethnic minorities. The relative contribution of socioeconomic status may be more prominent in Non-Hispanic White individuals, as those from racial and ethnic minority backgrounds have additional factors specifically related to their race and ethnicity that contribute more heavily to increased disparities in eye care. This is supported by previous studies that have demonstrated the surgical undertreatment of open-angle glaucoma (OAG) and a higher prevalence of OAG in racial and ethnic minorities in the Medicare population [28,29,30]. Overall, these findings suggest that associations between systemic and socioeconomic factors and the surgical treatment of NVG may vary by race and ethnicity, suggesting a need to comprehensively evaluate surgical needs on a personalized basis for individuals with NVG requiring IOP-lowering surgery. Addressing these disparities will also require upstream interventions. Community-based screening for elevated IOP and retinal pathology in high-risk populations, such as the Laroche Glaucoma Risk Calculator, has demonstrated feasibility and accuracy in underserved settings and may facilitate the earlier identification and treatment of retinal ischemia before progression to NVG [31].
This study has several limitations that should be acknowledged. First, as with most claims-based analyses, we lacked key clinical determinants that directly drive escalation to glaucoma surgery, most notably baseline IOP level, NVG severity, and prior treatments. As Medicare data are based on claims, there are no clinical data such as visual acuity and IOP available in the dataset. The classification of NVG severity by claims codes is also limited, as the definition of glaucoma severity for billing purposes is based on visual field loss and many patients with NVG have poor central vision from multifactorial causes [32]. As such, we anticipate that any coding of NVG severity may be inaccurate and accounting for this may lead to an increased risk of misclassification bias. Given that our database is cross-sectional, it would not be possible to determine the temporality of anti-VEGF injection and PRP with regards to IOP-lowering surgery, and the assessment of anti-VEGF and PRP on the risk of IOP-lowering surgery may be inaccurate. The absence of these measures raises the possibility of confounding by indication, because observed associations between demographic or systemic factors and the receipt of surgery may partly reflect unmeasured differences in presenting severity and IOP control rather than true differences in treatment pathways. This study is also limited by its focus on California Medicare beneficiaries, which may reduce generalizability to younger, privately insured or uninsured, and non-US patients, but provides insight into NVG treatment patterns in a diverse population. Reliance on Medicare claims data may result in the underreporting or misclassification of NVG, though this likely biases results toward the null, as the codes were originally recorded for purposes other than our research. Recently updated ICD-10 codes for NVG were not available as our data are from 2019 [33]. While we included SVI in our analyses, this metric is derived at the area level and may not reflect individual-level education, income, insurance benefit design, health literacy, transportation access, or caregiving support, and misclassification is therefore possible and could attenuate or distort associations. Future studies may benefit from other individual-level socioeconomic indicators, such as low-income subsidy status, to further validate these findings. While our analyses were hypothesis-driven with pre-specified covariates, the possibility of type I error due to multiple comparisons cannot be fully excluded; however, the association between higher CCI scores and a decreased likelihood of IOP-lowering surgery was consistent across multiple CCI categories, surgical subtypes, and both primary and secondary analytic approaches, supporting the robustness of this finding beyond any single p-value threshold. MIGS was included in the primary outcome to account for variation in surgical practice patterns and the inability to ascertain clinical indication from claims data, with a secondary analysis excluding MIGS performed to account for its lower relevance in active NVG management. The observational nature of the study also precludes causal inferences, and the lack of indicators related to glaucoma severity is a limitation to the study. We try to mitigate this by utilizing a diverse array of variables.
Factors beyond the scope of the present study that merit future investigation in relation to surgical treatment for NVG include geographic variables (e.g., urban vs. rural area, county of residence, distance to tertiary care) and provider-related factors (e.g., assessment by an ophthalmologist vs. an optometrist, healthcare system structure, and density of eye care providers). To partially address unmeasured clinical severity, future analyses could incorporate more detailed claims-based proxies (e.g., glaucoma medication burden, frequency of ophthalmology encounters, or time from NVG-related coding to surgery).
5. Conclusions
In summary, this study examined demographic, socioeconomic, and systemic factors associated with IOP-lowering surgeries for the treatment of NVG in the 2019 California Medicare population. Beneficiaries with a high systemic disease burden were less likely to receive IOP-lowering surgery for NVG, while racial and ethnic minorities were more likely to undergo certain procedures. Furthermore, poor socioeconomic status may be associated with increased IOP-lowering surgery in Non-Hispanic White beneficiaries with NVG. These multifaceted results emphasize the need for further in-depth investigations into the role of sociodemographic factors in NVG management to identify future areas for improvement and mitigate disparities in the care of retinal ischemia and its sequelae.
Supplementary Materials
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/jcto4020015/s1, Table S1: International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes for Charlson Comorbidity Index; Table S2: Current Procedural Terminology (CPT) codes for glaucoma surgery.
Author Contributions
Conceptualization, J.S.Y., K.K. and V.L.T.; Methodology, J.S.Y., D.P., F.Y. and V.L.T.; Software, K.K.; Formal analysis, J.S.Y., K.K., D.P., F.Y. and V.L.T.; Data curation, F.Y.; Writing—original draft, J.S.Y.; Writing—review and editing, J.S.Y., K.K., D.P., F.Y. and V.L.T.; Supervision, V.L.T.; Project administration, V.L.T.; Funding acquisition, V.L.T. All authors have read and agreed to the published version of the manuscript.
Funding
This work was supported by an unrestricted grant from Research to Prevent Blindness to the Department of Ophthalmology at UCLA. Victoria L. Tseng has research funding from the the Research to Prevent Blindness Career Development Award and the American Glaucoma Society Mid Career Physician Scientist Award. The remaining authors declare no funding or support.
Institutional Review Board Statement
This study was in accordance with the Declaration of Helsinki and its later amendments and was exempt from Institutional Review Board review and ethical approval at the University of California Los Angeles. This study was declared exempt from the Institutional Review Board at the University of California Los Angeles in 2023, IRB #23-000096, approval on 7 February 2023.
Informed Consent Statement
Patient consent was waived due to the retrospective nature of the study and the use of de-identified data.
Data Availability Statement
The datasets presented in this article are not readily available because the data are Medicare claims data obtained under a Data Use Agreement with the Centers for Medicare and Medicaid Services (CMS) and cannot be redistributed per the terms of that agreement.
Conflicts of Interest
The authors declare no conflicts of interest.
Abbreviations
The following abbreviations are used in this manuscript:
| AAO | American Academy of Ophthalmology |
| aOR | Adjusted odds ratio |
| AIDS | Acquired immune deficiency syndrome |
| CA | California |
| CCI | Charlson Comorbidity Index |
| CI | Confidence interval |
| CPC | Cyclophotocoagulation |
| CPT | Current Procedural Terminology |
| ICD-10-CM | International Classification of Diseases, 10th Revision, Clinical Modification |
| IOP | Intraocular pressure |
| IRIS® Registry | IRIS Registry (Intelligent Research in Sight) |
| MIGS | Minimally invasive glaucoma surgery |
| NVG | Neovascular glaucoma |
| OAG | Open-angle glaucoma |
| OIS | Ocular ischemic syndrome |
| OR | Odds ratio |
| PDR | Proliferative diabetic retinopathy |
| PRP | Panretinal photocoagulation |
| RVO | Retinal vein occlusion |
| SVI | Centers for Disease Control and Prevention Social Vulnerability Index |
| US | United States |
| VEGF | Vascular endothelial growth factor |
Appendix A
Table A1.
Predictors of each intraocular pressure (IOP)-lowering surgery to treat neovascular glaucoma (NVG) in the 2019 California Medicare population (n = 1843).
Table A1.
Predictors of each intraocular pressure (IOP)-lowering surgery to treat neovascular glaucoma (NVG) in the 2019 California Medicare population (n = 1843).
| Characteristic | Unadjusted Odds of Trabeculectomy | Unadjusted Odds of Tube Shunt | Unadjusted Odds of MIGS | Unadjusted Odds of CPC | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| OR | 95% CI | p | OR | 95% CI | p | OR | 95% CI | p | OR | 95% CI | p | |
| Age at index date (years) | ||||||||||||
| 65–69 | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| 70–74 | 1.06 | 0.48, 2.36 | 0.88 | 0.89 | 0.55, 1.44 | 0.64 | 1.28 | 0.56, 2.95 | 0.56 | 0.73 | 0.34, 1.55 | 0.41 |
| 75–79 | 0.22 | 0.05, 0.98 | 0.05 | 0.87 | 0.51, 1.47 | 0.60 | 1.48 | 0.62, 3.54 | 0.37 | 1.09 | 0.51, 2.29 | 0.83 |
| 80–84 | 0.37 | 0.10, 1.32 | 0.13 | 0.82 | 0.47, 1.43 | 0.48 | 1.37 | 0.55, 3.41 | 0.50 | 0.75 | 0.31, 1.77 | 0.51 |
| 85+ | 0.41 | 0.14, 1.18 | 0.10 | 0.50 | 0.29, 0.88 | 0.02 | 1.00 | 0.41, 2.44 | 0.99 | 1.34 | 0.69, 2.60 | 0.39 |
| Sex | ||||||||||||
| Male | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Female | 0.53 | 0.26, 1.08 | 0.08 | 0.82 | 0.58, 1.16 | 0.26 | 1.42 | 0.82, 2.45 | 0.21 | 0.92 | 0.57, 1.49 | 0.73 |
| Race/ethnicity | ||||||||||||
| Non-Hispanic White | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Black | 3.10 | 0.92, 10.45 | 0.07 | 1.65 | 0.82, 3.29 | 0.16 | 1.09 | 0.41, 2.87 | 0.87 | 1.01 | 0.38, 2.66 | 0.98 |
| Asian | 1.83 | 0.59, 5.66 | 0.29 | 1.50 | 0.87, 2.58 | 0.15 | 0.30 | 0.09, 1.01 | 0.05 | 0.87 | 0.41, 1.85 | 0.72 |
| Hispanic/Latino | 2.39 | 1.02, 5.57 | 0.04 | 1.46 | 0.96, 2.22 | 0.08 | 0.54 | 0.28, 1.04 | 0.07 | 0.84 | 0.48, 1.47 | 0.55 |
| Other/Unknown | 1.22 | 0.15, 9.93 | 0.85 | 2.83 | 1.39, 5.76 | 0.01 | 1.05 | 0.31, 3.54 | 0.94 | 0.98 | 0.29, 3.28 | 0.97 |
| Charlson Comorbidity Index score | ||||||||||||
| 0 | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| 1–2 | 0.37 | 0.11, 1.23 | 0.10 | 0.94 | 0.52, 1.70 | 0.83 | 0.58 | 0.24, 1.39 | 0.22 | 1.14 | 0.41, 3.11 | 0.81 |
| 3–4 | 1.05 | 0.37, 2.93 | 0.93 | 0.77 | 0.41, 1.42 | 0.40 | 0.96 | 0.42, 2.19 | 0.92 | 1.34 | 0.49, 3.64 | 0.57 |
| 5+ | 0.42 | 0.14, 1.26 | 0.12 | 0.55 | 0.30, 1.01 | 0.05 | 0.23 | 0.08, 0.61 | 0.01 | 1.33 | 0.51, 3.50 | 0.56 |
| Social Vulnerability Index (Quartiles) | ||||||||||||
| 0–0.41 (Q1) | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| 0.41–0.68 (Q2) | 0.89 | 0.34, 2.32 | 0.81 | 1.57 | 0.93, 2.65 | 0.09 | 1.24 | 0.59, 2.61 | 0.57 | 1.35 | 0.68, 2.67 | 0.39 |
| 0.68–0.87 (Q3) | 1.32 | 0.55, 3.17 | 0.53 | 1.45 | 0.86, 2.46 | 0.17 | 0.98 | 0.45, 2.15 | 0.97 | 1.47 | 0.75, 2.86 | 0.26 |
| 0.87–1 (Q4) | 0.68 | 0.24, 1.94 | 0.48 | 1.72 | 1.02, 2.88 | 0.04 | 0.87 | 0.39, 1.97 | 0.74 | 0.89 | 0.42, 1.90 | 0.77 |
| Characteristic | Adjusted Odds of Trabeculectomy | Adjusted Odds of Tube Shunt | Adjusted Odds of MIGS | Adjusted Odds of CPC | ||||||||
| OR | 95% CI | p | OR | 95% CI | p | OR | 95% CI | p | OR | 95% CI | p | |
| Age at index date (years) | ||||||||||||
| 65–69 | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| 70–74 | 1.06 | 0.47, 2.40 | 0.88 | 0.88 | 0.54, 1.43 | 0.61 | 1.11 | 0.48, 2.59 | 0.81 | 0.73 | 0.34, 1.57 | 0.42 |
| 75–79 | 0.21 | 0.05, 0.98 | 0.05 | 0.90 | 0.52, 1.54 | 0.69 | 1.22 | 0.50, 2.95 | 0.67 | 1.07 | 0.50, 2.30 | 0.85 |
| 80–84 | 0.36 | 0.10, 1.33 | 0.13 | 0.84 | 0.47, 1.48 | 0.54 | 1.06 | 0.41, 2.70 | 0.91 | 0.75 | 0.31, 1.80 | 0.52 |
| 85+ | 0.47 | 0.16, 1.40 | 0.17 | 0.57 | 0.32, 1.01 | 0.06 | 0.72 | 0.29, 1.82 | 0.49 | 1.36 | 0.67, 2.73 | 0.39 |
| Sex | ||||||||||||
| Male | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Female | 0.53 | 0.25, 1.12 | 0.10 | 0.82 | 0.57, 1.17 | 0.27 | 1.38 | 0.79, 2.43 | 0.26 | 0.88 | 0.54, 1.44 | 0.62 |
| Race/ethnicity | ||||||||||||
| Non-Hispanic White | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Black | 3.77 | 1.05, 13.57 | 0.04 | 1.62 | 0.79, 3.33 | 0.19 | 1.04 | 0.38, 2.88 | 0.94 | 1.06 | 0.39, 2.89 | 0.91 |
| Asian | 1.71 | 0.54, 5.41 | 0.36 | 1.46 | 0.84, 2.55 | 0.18 | 0.31 | 0.09, 1.03 | 0.06 | 0.84 | 0.39, 1.81 | 0.66 |
| Hispanic/Latino | 2.69 | 1.04, 6.92 | 0.04 | 1.30 | 0.81, 2.08 | 0.28 | 0.57 | 0.27, 1.19 | 0.14 | 0.87 | 0.47, 1.63 | 0.67 |
| Other/Unknown | 0.95 | 0.11, 7.88 | 0.96 | 2.62 | 1.27, 5.41 | 0.01 | 0.97 | 0.28, 3.33 | 0.96 | 1.05 | 0.31, 3.56 | 0.94 |
| Charlson Comorbidity Index score | ||||||||||||
| 0 | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| 1–2 | 0.24 | 0.07, 0.81 | 0.02 | 0.86 | 0.47, 1.58 | 0.63 | 0.62 | 0.26, 1.51 | 0.29 | 1.15 | 0.42, 3.17 | 0.79 |
| 3–4 | 0.66 | 0.22, 1.92 | 0.44 | 0.68 | 0.36, 1.29 | 0.24 | 1.10 | 0.47, 2.57 | 0.83 | 1.34 | 0.49, 3.68 | 0.57 |
| 5+ | 0.25 | 0.08, 0.79 | 0.02 | 0.46 | 0.25, 0.87 | 0.02 | 0.26 | 0.10, 0.72 | 0.01 | 1.35 | 0.51, 3.60 | 0.55 |
| Social Vulnerability Index (Quartiles) | ||||||||||||
| 0–0.41 (Q1) | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| 0.41–0.68 (Q2) | 0.79 | 0.30, 2.11 | 0.64 | 1.53 | 0.90, 2.60 | 0.12 | 1.31 | 0.62, 2.79 | 0.48 | 1.40 | 0.70, 2.78 | 0.34 |
| 0.68–0.87 (Q3) | 0.92 | 0.36, 2.35 | 0.86 | 1.33 | 0.76, 2.31 | 0.32 | 1.17 | 0.52, 2.64 | 0.71 | 1.58 | 0.78, 3.20 | 0.20 |
| 0.87–1 (Q4) | 0.39 | 0.13, 1.21 | 0.10 | 1.57 | 0.89, 2.77 | 0.12 | 1.15 | 0.47, 2.82 | 0.76 | 0.98 | 0.43, 2.23 | 0.97 |
Bold indicates statistical significance (p < 0.05). Adjusted for age, sex, race/ethnicity, Charlson Comorbidity Index, and Social Vulnerability Index. OR = odds ratio; CI = confidence interval.
Table A2.
Adjusted analysis of intraocular pressure (IOP)-lowering surgery excluding minimally invasive glaucoma surgery (MIGS) to treat neovascular glaucoma (NVG) in the 2019 California Medicare population (n = 1843).
Table A2.
Adjusted analysis of intraocular pressure (IOP)-lowering surgery excluding minimally invasive glaucoma surgery (MIGS) to treat neovascular glaucoma (NVG) in the 2019 California Medicare population (n = 1843).
| Characteristics | Adjusted Data for Any Surgery Among All NVG (n = 1843) | Adjusted Data Among NVG Patients with Proliferative Diabetic Retinopathy (PDR; n = 831) | Adjusted Data Among NVG Patients with Retinal Vein Occlusion (RVO; n = 1008) | ||||||
|---|---|---|---|---|---|---|---|---|---|
| OR | 95% CI | p | OR | 95% CI | p | OR | 95% CI | p | |
| Age at index date (years) | |||||||||
| 65–69 | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| 70–74 | 0.87 | 0.58, 1.29 | 0.49 | 1.16 | 0.71, 1.89 | 0.56 | 0.53 | 0.27, 1.03 | 0.06 |
| 75–79 | 0.75 | 0.47, 1.17 | 0.21 | 0.53 | 0.25, 1.15 | 0.11 | 0.65 | 0.35, 1.23 | 0.18 |
| 80–84 | 0.62 | 0.38, 1.02 | 0.06 | 0.57 | 0.23, 1.41 | 0.22 | 0.40 | 0.20, 0.80 | 0.01 |
| 85+ | 0.75 | 0.48, 1.16 | 0.19 | 1.03 | 0.41, 2.61 | 0.95 | 0.50 | 0.28, 0.90 | 0.02 |
| Sex | |||||||||
| Male | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Female | 0.73 | 0.54, 0.97 | 0.03 | 0.69 | 0.44, 1.08 | 0.10 | 0.82 | 0.55, 1.21 | 0.31 |
| Race/ethnicity | |||||||||
| Non-Hispanic White | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Black | 1.55 | 0.86, 2.78 | 0.15 | 2.17 | 0.89, 5.28 | 0.09 | 1.18 | 0.52, 2.71 | 0.69 |
| Asian | 1.20 | 0.76, 1.89 | 0.43 | 1.15 | 0.51, 2.57 | 0.74 | 1.01 | 0.55, 1.89 | 0.97 |
| Hispanic/Latino | 1.26 | 0.87, 1.82 | 0.22 | 1.28 | 0.70, 2.35 | 0.42 | 1.70 | 1.03, 2.78 | 0.04 |
| Other/Unknown | 1.61 | 0.82, 3.15 | 0.17 | 1.25 | 0.38, 4.07 | 0.71 | 2.21 | 0.98, 4.99 | 0.06 |
| Charlson Comorbidity Index score | |||||||||
| 0 | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| 1–2 | 0.72 | 0.43, 1.20 | 0.20 | N/A * | N/A * | N/A * | 0.82 | 0.46, 1.46 | 0.50 |
| 3–4 | 0.75 | 0.45, 1.27 | 0.28 | 0.98 | 0.55, 1.74 | 0.95 | 0.91 | 0.49, 1.69 | 0.76 |
| 5+ | 0.56 | 0.34, 0.94 | 0.03 | 0.62 | 0.36, 1.07 | 0.08 | 0.78 | 0.43, 1.44 | 0.43 |
| Social Vulnerability Index (Quartiles) | |||||||||
| 0–0.41 (Q1) | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| 0.41–0.68 (Q2) | 1.42 | 0.94, 2.16 | 0.10 | 1.19 | 0.57, 2.50 | 0.65 | 1.71 | 1.00, 2.93 | 0.05 |
| 0.68–0.87 (Q3) | 1.31 | 0.85, 2.03 | 0.22 | 1.18 | 0.58, 2.40 | 0.65 | 1.65 | 0.93, 2.91 | 0.09 |
| 0.87–1 (Q4) | 1.16 | 0.73, 1.83 | 0.54 | 1.19 | 0.59, 2.41 | 0.63 | 1.05 | 0.53, 2.05 | 0.89 |
Bold indicates statistical significance (p < 0.05). Adjusted for age, sex, race/ethnicity, Charlson Comorbidity Index, and Social Vulnerability Index. OR = odds ratio; CI = confidence interval. * Because PDR is coded under diabetes with complications on ICD-10-CM, no PDR patient could have a CCI of 0 or 1; therefore, those score categories are not represented in this subgroup.
References
- Rodrigues, G.B.; Abe, R.Y.; Zangalli, C.; Sodre, S.L.; Donini, F.A.; Costa, D.C.; Leite, A.; Felix, J.P.; Torigoe, M.; Diniz-Filho, A.; et al. Neovascular Glaucoma: A Review. Int. J. Retin. Vitr. 2016, 2, 26. [Google Scholar] [CrossRef]
- Sivak-Callcott, J.A.; O’Day, D.M.; Gass, J.D.M.; Tsai, J.C. Evidence-Based Recommendations for the Diagnosis and Treatment of Neovascular Glaucoma1 1The Authors Do Not Have Any Proprietary or Financial Interest in Any Products or Devices Discussed in This Study. Ophthalmology 2001, 108, 1767–1776. [Google Scholar] [CrossRef] [PubMed]
- Brown, G.C.; Magargal, L.E.; Schachat, A.; Shah, H. Neovascular Glaucoma: Etiologic Considerations. Ophthalmology 1984, 91, 315–320. [Google Scholar] [CrossRef] [PubMed]
- Calugaru, D. Etiology, Pathogenesis, and Diagnosis of Neovascular Glaucoma. Int. J. Ophthalmol. 2022, 15, 1005–1010. [Google Scholar] [CrossRef] [PubMed]
- Olmos, L.C.; Lee, R.K. Medical and Surgical Treatment of Neovascular Glaucoma. Int. Ophthalmol. Clin. 2011, 51, 27–36. [Google Scholar] [CrossRef]
- Iwasaki, K.; Arimura, S.; Takamura, Y.; Inatani, M. Clinical Practice Preferences for Glaucoma Surgery in Japan: A Survey of Japan Glaucoma Society Specialists. Jpn. J. Ophthalmol. 2020, 64, 385–391. [Google Scholar] [CrossRef]
- Fellman, R.L.; Mattox, C.; Singh, K.; Flowers, B.; Francis, B.A.; Robin, A.L.; Butler, M.R.; Shah, M.M.; Giaconi, J.A.; Sheybani, A.; et al. American Glaucoma Society Position Paper: Microinvasive Glaucoma Surgery. Ophthalmol. Glaucoma 2020, 3, 1–6. [Google Scholar] [CrossRef]
- Shazly, T.A.; Latina, M.A. Neovascular Glaucoma: Etiology, Diagnosis and Prognosis. Semin. Ophthalmol. 2009, 24, 113–121. [Google Scholar] [CrossRef]
- Haller, J.A.; Tomaiuolo, M.; Lucas, M.M.; Yang, C.C.; Hyman, L.; Lee, A.Y.; Lee, C.S.; Van Gelder, R.; Lorch, A.; Miller, J.W.; et al. Disparities in Retinal Vein Occlusion Presentation and Initiation of Anti-VEGF Therapy. Ophthalmol. Retin. 2024, 8, 657–665. [Google Scholar] [CrossRef]
- Tseng, V.L.; Pan, D.; Kitayama, K.; Yu, F.; Coleman, A.L. Racial and Ethnic Differences in the Prevalence and Treatment Patterns for Neovascular Glaucoma in the American Academy of Ophthalmology IRIS® Registry (Intelligent Research in Sight). Ophthalmol. Glaucoma 2024, 7, 615–623. [Google Scholar] [CrossRef]
- American Academy of Ophthalmology. ICD-10 Diagnosis Code for Neovascular Glaucoma. Available online: https://www.aao.org/practice-management/news-detail/icd-10-diagnosis-code-neovascular-glaucoma (accessed on 30 May 2025).
- Charlson, M.E.; Pompei, P.; Ales, K.L.; MacKenzie, C.R. A New Method of Classifying Prognostic Comorbidity in Longitudinal Studies: Development and Validation. J. Chronic Dis. 1987, 40, 373–383. [Google Scholar] [CrossRef]
- Centers for Disease Control and Prevention. CDC/ATSDR Social Vulnerability Index (SVI); Centers for Disease Control and Prevention: Atlanta, GA, USA, 2024. Available online: https://www.atsdr.cdc.gov/place-health/php/svi/index.html (accessed on 30 May 2025).
- Research Data Assistance Center. Research Triangle Institute (RTI) Race Code; RTI International: Durham, NC, USA; Available online: https://resdac.org/cms-data/variables/research-triangle-institute-rti-race-code (accessed on 30 May 2025).
- United States Census Bureau. Available online: https://www.census.gov/ (accessed on 12 July 2024).
- Tseng, V.L.; Kitayama, K.; Yu, F.; Pan, D.; Coleman, A.L. Social Vulnerability, Prevalence of Glaucoma, and Incidence of Glaucoma Surgery in the California Medicare Population. Ophthalmol. Glaucoma 2023, 6, 616–625. [Google Scholar] [CrossRef]
- National Center for Health Statistics. International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM); National Center for Health Statistics: Hyattsville, MD, USA, 2024. Available online: https://www.cdc.gov/nchs/icd/icd-10-cm/index.html (accessed on 30 May 2025).
- American Medical Association CPT. Available online: https://www.ama-assn.org/practice-management/cpt (accessed on 12 July 2024).
- VanderWeele, T.J.; Ding, P. Sensitivity Analysis in Observational Research: Introducing the E-Value. Ann. Intern Med. 2017, 167, 268–274. [Google Scholar] [CrossRef]
- Al-Moujahed, A.; Vail, D.; Do, D.V. Racial Differences in Anti-VEGF Intravitreal Injections Among Commercially Insured Beneficiaries. Ophthalmic Surg. Lasers Imaging Retin. 2021, 52, 208–217. [Google Scholar] [CrossRef] [PubMed]
- Wu, A.M.; Shen, L.Q. Racial Disparities Affecting Black Patients in Glaucoma Diagnosis and Management. Semin. Ophthalmol. 2023, 38, 65–75. [Google Scholar] [CrossRef] [PubMed]
- Varma, R.; Ying-Lai, M.; Francis, B.A.; Nguyen, B.B.T.; Deneen, J.; Wilson, M.R.; Azen, S.P. Prevalence of Open-Angle Glaucoma and Ocular Hypertension in Latinos: The Los Angeles Latino Eye Study. Ophthalmology 2004, 111, 1439–1448. [Google Scholar] [CrossRef]
- Murakami, Y. Racial and Ethnic Disparities in Adherence to Glaucoma Follow-up Visits in a County Hospital Population. Arch. Ophthalmol. 2011, 129, 872. [Google Scholar] [CrossRef]
- Cheng, B.T.; Tanna, A.P. Association of Race and Ethnicity with the Frequency of Outpatient Glaucoma Care. Ophthalmology 2022, 129, 585–587. [Google Scholar] [CrossRef] [PubMed]
- Pastor, S. Cyclophotocoagulation A Report by the American Academy of Ophthalmology. Ophthalmology 2001, 108, 2130–2138. [Google Scholar] [CrossRef]
- Shalaby, W.; Ganjei, A.; Wogu, B.; Myers, J.; Moster, M.; Razeghinejad, R.; Lee, D.; Kolomeyer, N.; Eid, T.; Katz, L.J.; et al. Outcomes of Ahmed Glaucoma Valve and Transscleral Cyclophotocoagulation in Neovascular Glaucoma. Indian J. Ophthalmol. 2022, 70, 1253. [Google Scholar] [CrossRef]
- Shi, Q.; Zhao, Y.; Fonseca, V.; Krousel-Wood, M.; Shi, L. Racial Disparity of Eye Examinations Among the U.S. Working-Age Population with Diabetes: 2002–2009. Diabetes Care 2014, 37, 1321–1328. [Google Scholar] [CrossRef] [PubMed]
- Gold, F.E.; Yadollahikhales, M.; Dersu, I.I. National Utilization of Glaucoma Laser and Surgical Management in Medicare Beneficiaries: Disparities and Accessibility. J. Glaucoma 2023, 32, 1038–1043. [Google Scholar] [CrossRef] [PubMed]
- Tseng, V.L.; Kitayama, K.; Yu, F.; Coleman, A.L. Prevalence and Severity of Glaucoma in the California Medicare Population. Am. J. Ophthalmol. 2024, 259, 25–34. [Google Scholar] [CrossRef]
- Javitt, J.C.; McBean, A.M.; Nicholson, G.A.; Babish, J.D.; Warren, J.L.; Krakauer, H. Undertreatment of Glaucoma among Black Americans. N. Engl. J. Med. 1991, 325, 1418–1422. [Google Scholar] [CrossRef]
- Laroche, D.; Rickford, K.; Mike, E.V.; Hunter, L.; Ede, E.; Ng, C.; Douglas, J. A Novel, Low-Cost Glaucoma Calculator to Identify Glaucoma Patients and Stratify Management. J. Ophthalmol. 2022, 2022, 5288726. [Google Scholar] [CrossRef] [PubMed]
- American Academy of Ophthalmology. Glaucoma ICD-10-CM Quick Reference Guide; American Academy of Ophthalmology: San Francisco, CA, USA, 2019; (Revised June 2019; effective 1 October 2020). [Google Scholar]
- American Academy of Ophthalmology. 2026 Quarterly Coding Updates; American Academy of Ophthalmology: San Francisco, CA, USA, 2026. [Google Scholar]
Table 1.
Baseline characteristics of beneficiaries with neovascular glaucoma (NVG) (n = 1843).
| Characteristics | No. (%) of Beneficiaries in Total Population; n = 1843 | No. (%) of Beneficiaries with Any Surgery for NVG; n = 264 | No. (%) of Beneficiaries Without Any Surgery; n = 1579 | p |
|---|---|---|---|---|
| Age at index date (years) | 0.37 | |||
| 65–69 | 416 (22.57) | 69 (26.14) | 347 (21.98) | |
| 70–74 | 425 (23.06) | 66 (25) | 359 (22.74) | |
| 75–79 | 312 (16.93) | 43 (16.29) | 269 (17.04) | |
| 80–84 | 276 (14.98) | 34 (12.88) | 242 (15.33) | |
| 85+ | 414 (22.46) | 52 (19.70) | 362 (22.93) | |
| Sex | 0.10 | |||
| Male | 990 (53.72) | 154 (58.33) | 836 (52.94) | |
| Female | 853 (46.28) | 110 (41.67) | 743 (47.06) | |
| Race/ethnicity | 0.59 | |||
| Non-Hispanic White | 733 (39.77) | 98 (37.12) | 635 (40.22) | |
| Black | 121 (6.57) | 21 (7.95) | 100 (6.33) | |
| Asian | 252 (13.67) | 34 (12.88) | 218 (13.81) | |
| Hispanic/Latino | 662 (35.92) | 97 (36.74) | 565 (35.78) | |
| Other/Unknown | 75 (4.07) | 14 (5.30) | 61 (3.86) | |
| Charlson Comorbidity Index score | 0.01 | |||
| 0 | 163 (8.84) | 32 (12.12) | 131 (8.30) | |
| 1–2 | 519 (28.16) | 74 (28.03) | 445 (28.18) | |
| 3–4 | 468 (25.39) | 79 (29.92) | 389 (24.64) | |
| 5+ | 693 (37.60) | 79 (29.92) | 614 (38.89) | |
| Social Vulnerability Index (Quartiles) | 0.31 | |||
| 0–0.41 (Q1) | 463 (25.12) | 55 (20.83) | 408 (25.84) | |
| 0.41–0.68 (Q2) | 462 (25.07) | 74 (28.03) | 388 (24.57) | |
| 0.68–0.87 (Q3) | 470 (25.50) | 71 (26.89) | 399 (25.27) | |
| 0.87–1 (Q4) | 448 (24.31) | 64 (24.24) | 384 (24.32) | |
| Dual Medicare–Medicaid Eligibility | 0.88 | |||
| No | 985 (53.45) | 140 (53.03) | 845 (53.51) | |
| Yes | 858 (46.55) | 124 (46.97) | 734 (46.49) | |
| Intravitreal Anti-Vascular Endothelial Growth Factor (VEGF) Injection | ||||
| No | 1015 (55.07) | 102 (38.64) | 913 (57.82) | 0.01 |
| Yes | 828 (44.93) | 162 (61.36) | 666 (42.18) | |
| Panretinal Photocoagulation | ||||
| No | 1563 (84.81) | 189 (71.59) | 1374 (87.02) | 0.01 |
| Yes | 280 (15.19) | 75 (28.41) | 205 (12.98) |
Bold indicates statistical significance (p < 0.05).
Table 2.
Predictors of any intraocular pressure (IOP)-lowering surgery to treat neovascular glaucoma (NVG) in the 2019 California Medicare population (n = 1843).
Table 2.
Predictors of any intraocular pressure (IOP)-lowering surgery to treat neovascular glaucoma (NVG) in the 2019 California Medicare population (n = 1843).
| Characteristic | Unadjusted Odds of Any Surgery | Adjusted Odds of Any Surgery | |||||
|---|---|---|---|---|---|---|---|
| OR | 95% CI | p | OR | 95% CI | p | E-Value | |
| Age at index date (years) | |||||||
| 65–69 | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| 70–74 | 0.92 | 0.64, 1.34 | 0.68 | 0.95 | 0.65, 1.40 | 0.81 | 1.29 |
| 75–79 | 0.80 | 0.53, 1.21 | 0.30 | 0.85 | 0.55, 1.30 | 0.44 | 1.63 |
| 80–84 | 0.71 | 0.45, 1.10 | 0.12 | 0.76 | 0.48, 1.21 | 0.25 | 1.96 |
| 85+ | 0.72 | 0.49, 1.07 | 0.10 | 0.82 | 0.54, 1.24 | 0.34 | 1.74 |
| Sex | |||||||
| Male | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Female | 0.80 | 0.62, 1.05 | 0.11 | 0.79 | 0.60, 1.04 | 0.09 | 1.85 |
| Race/ethnicity | |||||||
| Non-Hispanic White | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Black | 1.36 | 0.81, 2.28 | 0.24 | 1.50 | 0.87, 2.59 | 0.15 | 2.37 |
| Asian | 1.01 | 0.66, 1.54 | 0.96 | 1.05 | 0.67, 1.64 | 0.84 | 1.28 |
| Hispanic/Latino | 1.11 | 0.82, 1.51 | 0.49 | 1.09 | 0.74, 1.59 | 0.67 | 1.4 |
| Other/Unknown | 1.49 | 0.80, 2.76 | 0.21 | 1.51 | 0.79, 2.89 | 0.21 | 2.39 |
| Charlson Comorbidity Index score | |||||||
| 0 | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| 1–2 | 0.68 | 0.11, 1.23 | 0.10 | 0.60 | 0.37, 0.97 | 0.04 | 2.72 |
| 3–4 | 0.83 | 0.37, 2.93 | 0.43 | 0.73 | 0.45, 1.17 | 0.19 | 2.08 |
| 5+ | 0.53 | 0.14, 1.26 | 0.01 | 0.47 | 0.29, 0.75 | 0.01 | 3.68 |
| Social Vulnerability Index (Quartiles) | |||||||
| 0–0.41 (Q1) | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| 0.41–0.68 (Q2) | 1.41 | 0.97, 2.06 | 0.07 | 1.36 | 0.93, 2.01 | 0.12 | 2.06 |
| 0.68–0.87 (Q3) | 1.32 | 0.90, 1.93 | 0.15 | 1.32 | 0.88, 1.99 | 0.18 | 1.97 |
| 0.87–1 (Q4) | 1.24 | 0.84, 1.82 | 0.28 | 1.15 | 0.74, 1.79 | 0.53 | 1.57 |
| Dual Medicare–Medicaid Eligibility | |||||||
| No | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Yes | 1.02 | 0.79, 1.32 | 0.88 | 0.98 | 0.71, 1.36 | 0.92 | 1.16 |
| Intravitreal Anti-Vascular Endothelial Growth Factor (VEGF) Injection | |||||||
| No | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Yes | 2.18 | 1.67, 2.84 | 0.01 | 1.79 | 1.34, 2.39 | 0.01 | 2.98 |
| Panretinal Photocoagulation | |||||||
| No | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Yes | 2.66 | 1.96, 3.61 | 0.01 | 2.03 | 1.46, 2.83 | 0.01 | 3.48 |
Bold indicates statistical significance (p < 0.05); OR = odds ratio; CI = confidence interval.
Table 3.
Predictors of any intraocular pressure (IOP)-lowering surgery to treat neovascular glaucoma (NVG), stratified by race and ethnicity (n = 1843).
Table 3.
Predictors of any intraocular pressure (IOP)-lowering surgery to treat neovascular glaucoma (NVG), stratified by race and ethnicity (n = 1843).
| Characteristics | Non-Hispanic White | Black | Asian | Hispanic/Latino | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Adjusted OR | 95% CI | p | Adjusted OR | 95% CI | p | Adjusted OR | 95% CI | p | Adjusted OR | 95% CI | p | |
| Age at index date (years) | ||||||||||||
| 65–69 | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| 70–74 | 0.84 | 0.41, 1.73 | 0.64 | 0.60 | 0.14, 2.54 | 0.48 | 0.54 | 0.19, 1.58 | 0.26 | 1.12 | 0.65, 1.95 | 0.68 |
| 75–79 | 0.74 | 0.35, 1.54 | 0.42 | 0.38 | 0.08, 1.89 | 0.24 | 0.52 | 0.16, 1.72 | 0.28 | 1.01 | 0.51, 2.00 | 0.97 |
| 80–84 | 0.49 | 0.22, 1.13 | 0.09 | 0.57 | 0.14, 2.37 | 0.44 | 0.6 | 0.18, 1.99 | 0.40 | 0.89 | 0.42, 1.91 | 0.77 |
| 85+ | 0.88 | 0.46, 1.69 | 0.70 | 0.33 | 0.05, 2.11 | 0.24 | 0.44 | 0.14, 1.38 | 0.16 | 0.59 | 0.27, 1.31 | 0.20 |
| Sex | ||||||||||||
| Male | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Female | 0.84 | 0.54, 1.31 | 0.44 | 1.00 | 0.36, 2.75 | 0.10 | 0.78 | 0.35, 1.74 | 0.54 | 0.77 | 0.49, 1.21 | 0.27 |
| Charlson Comorbidity Index score | ||||||||||||
| 0 | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| 1–2 | 0.96 | 0.45, 2.04 | 0.92 | 0.58 | 0.08, 4.18 | 0.59 | 0.32 | 0.09, 1.12 | 0.08 | 0.44 | 0.18, 1.09 | 0.08 |
| 3–4 | 1.63 | 0.77, 3.45 | 0.20 | 0.43 | 0.06, 3.02 | 0.40 | 0.33 | 0.09, 1.14 | 0.08 | 0.47 | 0.19, 1.15 | 0.10 |
| 5+ | 0.89 | 0.42, 1.92 | 0.77 | 0.82 | 0.13, 5.01 | 0.83 | 0.18 | 0.05, 0.63 | 0.01 | 0.30 | 0.12, 0.73 | 0.01 |
| Social Vulnerability Index (Quartiles) | ||||||||||||
| 0–0.41 (Q1) | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| 0.41–0.68 (Q2) | 1.94 | 1.18, 3.20 | 0.01 | 0.56 | 0.10, 3.10 | 0.50 | 0.88 | 0.28, 2.72 | 0.82 | 0.88 | 0.37, 2.10 | 0.78 |
| 0.68–0.87 (Q3) | 0.93 | 0.47, 1.86 | 0.84 | 0.92 | 0.23, 3.74 | 0.91 | 1.74 | 0.62, 4.88 | 0.29 | 0.94 | 0.42, 2.07 | 0.88 |
| 0.87–1 (Q4) | 1.30 | 0.60, 2.83 | 0.51 | 0.73 | 0.17, 3.09 | 0.67 | 1.52 | 0.45, 5.11 | 0.50 | 0.78 | 0.36, 1.68 | 0.52 |
Bold indicates statistical significance (p < 0.05); Adjusted for age, sex, Charlson Comorbidity Index, and Social Vulnerability Index; OR = odds ratio; CI = confidence interval.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2026 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license.
Share and Cite
MDPI and ACS Style
Yun, J.S.; Kitayama, K.; Pan, D.; Yu, F.; Tseng, V.L. Predictors of Receiving Surgical Treatment for Neovascular Glaucoma in the California Medicare Population. J. Clin. Transl. Ophthalmol. 2026, 4, 15. https://doi.org/10.3390/jcto4020015
AMA Style
Yun JS, Kitayama K, Pan D, Yu F, Tseng VL. Predictors of Receiving Surgical Treatment for Neovascular Glaucoma in the California Medicare Population. Journal of Clinical & Translational Ophthalmology. 2026; 4(2):15. https://doi.org/10.3390/jcto4020015
Chicago/Turabian StyleYun, Justin S., Ken Kitayama, Deyu Pan, Fei Yu, and Victoria L. Tseng. 2026. "Predictors of Receiving Surgical Treatment for Neovascular Glaucoma in the California Medicare Population" Journal of Clinical & Translational Ophthalmology 4, no. 2: 15. https://doi.org/10.3390/jcto4020015
APA StyleYun, J. S., Kitayama, K., Pan, D., Yu, F., & Tseng, V. L. (2026). Predictors of Receiving Surgical Treatment for Neovascular Glaucoma in the California Medicare Population. Journal of Clinical & Translational Ophthalmology, 4(2), 15. https://doi.org/10.3390/jcto4020015
