Abstract
Introduction
Retinal vein occlusion (RVO) has an estimated prevalence of up to 1.6% in the general population, making it the second most common retinal vascular disorder after diabetic retinopathy.1 –3 Branch RVO (BRVO) is the most prevalent form of RVO, with an incidence rate up to 4 times greater than that of central RVO (CRVO). 4 The pathogenesis is typically attributed to 1 of the following 3 mechanisms: compression of the branch retinal vein by a sclerotic retinal artery at an arteriovenous crossing site, degenerative venous wall changes, or hematologic disorders.5,6 Venous compression by a sclerotic artery is thought to be the most common mechanism of BRVO and may in be part related to the retinal vein and artery crossing site sharing the same adventitial sheath. 7
In older adults, systemic conditions that result in vascular abnormalities, such as systemic hypertension, hyperlipidemia, and diabetes mellitus (DM), are the most common risk factors for BRVO.8,9 Much less is known about the pathogenesis of BRVO in younger individuals (<50 years of age). Previous studies are scarce but mainly analyzed the systemic risk factors and abnormalities associated with BRVO in younger patients.10 –12 These studies focused on the differences in risk factors between younger patients with BRVO and controls, whereas our study focused on the difference between younger individuals and older individuals diagnosed with BRVO. We compared the presentations and outcomes of acute, treatment-naïve BRVO in younger patients (age <50 years at BRVO onset) vs older patients (age ≥50 years at BRVO onset) at a tertiary referral center.
Methods
The study protocol and data collection were approved by the Duke University School of Medicine Institutional Review Board and adhered to the US Health Insurance Portability and Accountability Act and all tenets of the Declaration of Helsinki. The Duke Enterprise Data Unified Content Explorer was used to search the electronic medical records of all patients who visited the Duke Eye Center from January 1, 2009, through June 30, 2018, to identify patients with acute, treatment-naïve BRVO. Inclusion criteria included BRVO with macular involvement, BRVO onset within 3 months of presentation, at least 12 months of follow-up, and lack of treatment before presentation to our institution. BRVO onset was assessed by patient onset of visual symptoms. Ethical approval for this study was obtained from the Duke Health Institutional Review Board (Pro00075701). Informed consent was not sought for the present study because of its retrospective nature.
Demographic data were extracted from the medical records of all patients who met the inclusion criteria. Basic demographic information recorded included age at BRVO onset, sex, race, and smoking status. The following clinical information was also abstracted: body mass index (BMI), hypertension, DM, hyperlipidemia, open-angle glaucoma (OAG), hypercoagulability workup, and laser treatments. Use of the following medications at presentation was documented: metformin, pioglitazone, angiotensin-converting enzyme inhibitors, beta blockers, hormone replacement therapy, aspirin 81 mg, aspirin 325 mg, fish oil, vitamin E, warfarin, clopidogrel, rivaroxaban, apixaban, and dabigatran.
From each patient’s first and final presentation at the institution, the following data were recorded for both eyes: Early Treatment Diabetic Retinopathy Study best-corrected visual acuity (VA) converted to logMAR notation, intraocular pressure (IOP) via a Tono-Pen, central subfield thickness (CST), subfoveal choroidal thickness (SFCT) via a Spectralis device (Heidelberg Engineering Inc), enhanced depth imaging optical coherence tomography (OCT), and the presence or absence of cystoid macular edema (CME) and subretinal fluid (SRF) on OCT. The first treatment received was recorded as well if there was fovea-involving intraretinal hemorrhage (IRH), vitreous hemorrhage (VH), or neovascularization of the iris (NVI), disc (NVD), or elsewhere (NVE). The number of antivascular endothelial growth factor (anti-VEGF) injections was logged as well as the number of intravitreal steroid injections performed during the first year of follow-up and at the final visit.
Extracted data were then compared between younger patients (age <50 years at BRVO onset) and older patients (age ≥50 years at BRVO onset). The cutoff age of 50 years was decided based on clinical experience and previous studies.13,14
Statistical analysis was completed using SAS software (version 9.4, SAS Institute, Inc). Continuous variables, such as BMI, logMAR acuity, IOP, CST, SFCT, number of intravitreal anti-VEGF injections 1 year after diagnosis, and number of injections at the final follow-up, were compared using the Wilcoxon rank sum test of the difference between medians. Nominal datapoints (eg, hypertension; diabetes; smoking status; various medications used; sex; race; presence of NVI, NVD, NVE, and VH; foveal IRH; CME; SRF; hypercoagulability workup) were analyzed using the Fisher exact test of the difference between proportions. Statistical significance was defined by a P value less than 0.05.
Results
Baseline Characteristics
Of the 302 patients who met inclusion criteria and had available demographic data necessary for analysis, 23 (7.6%) were younger than 50 years and 279 (92.4%) were 50 years or older at the time of BRVO onset. Table 1 shows the basic demographic data for both groups. The mean age was 41.52 years in the younger group and 70.77 years in the older group (P < .001). There was no significant difference in sex distribution or average BMI between the 2 groups (Table 1). There was a significant difference in racial distribution, with a larger proportion of White patients in the older cohort and a larger proportion of Asian patients in the younger cohort (P = .002).
Basic Demographics.
Abbreviation: BMI, body mass index.
Unless otherwise indicated due to limited data on chart review.
Statistically significant (P < .05).
Wilcoxon rank sum test of difference between medians.
Fisher exact test of difference between proportions.
Table 2 shows the baseline characteristics of the BRVO-affected eyes and fellow eyes in both age groups. Younger patients had significantly better logMAR VA in the fellow eye at presentation than the older group (0.20 ± 0.52 [Snellen 20/32] vs 0.49 ± 0.89 [20/62]; P < .001); however, there was no significant difference in the presenting VA in the eye with BRVO. At presentation, younger patients had a greater average SFCT than the older group in the affected eye (293.6 ± 112.3 vs 229.0 ± 75.97) and fellow eye (262.7 ± 78.94 vs 218.2 ± 75.22) (P = .019 and P = .033, respectively). The 2 cohorts had a similar IOP in the presenting eye and fellow eye (P = .072 and P = .855, respectively). There were no significant differences between the 2 cohorts in the CST in the affected eyes or fellow eyes. There was no significant difference in the proportions of BRVO-affected eyes presenting with NV (of the iris, disc, or elsewhere), CME, SRF, VH, or foveal IRH. There was also no difference between the younger group and older group in the proportion of BRVO-affected eyes with an enlarged foveal avascular zone at the baseline visit on fluorescein angiography.
Branch Retinal Vein Occlusion Characteristics at Presentation.
Abbreviations: CME, cystoid macular edema; CST, central subfield thickness; IOP, intraocular pressure; PRP, panretinal photocoagulation; SFCT, subfoveal choroidal thickness; SRF, subretinal fluid; VA, visual acuity.
Unless otherwise indicated due to limited data on chart review.
Statistically significant (P < .05).
Wilcoxon rank sum test of difference between medians.
Fisher exact test of difference between proportions.
Traditional risk factors recorded included hypertension, DM, positive smoking status, angle glaucoma, and hyperlipidemia (Table 3). Older patients and younger patients had similar rates of hypertension (70% younger vs 80% older; P = .288), DM (32% vs 31%; P = 1.000), smoking status (23% vs 22%; P = 1.000), and glaucoma (23% vs 30%; P = .628). Patients in the younger group had a lower rate of hyperlipidemia than patients in the older group (35% vs 59%; P = .028). In patients with hyperlipidemia, a higher proportion in the younger group than in the older group was on statin therapy (100% vs 66%; P = .047). More patients in the younger cohort than in the older cohort had a hypercoagulable workup, including tests for hyperhomocystinemia, protein C deficiency, factor V Leiden, and factor VIII, IX, and X (24% vs 2%; P < .001); however, there was no significant difference in hypercoagulability findings between the 2 age groups (Table 4).
Risk Factors.
Abbreviations: ACE, angiotension-converting enzyme; HRT, hormone replacement therapy.
Unless otherwise indicated due to limited data on chart review.
Fisher exact test of difference between proportions.
Statistically significant (P < .05).
Treatment and Clinical Outcomes.
Abbreviations: CME, cystoid macular edema; CST, central subfield thickness; FA, fluorescein angiography; FAZ, foveal avascular zone; IOP, intraocular pressure; IVTA, intravitreal triamcinolone acetonide; PRP, panretinal photocoagulation; RVO, retinal vein occlusion; SFCT, subfoveal choroidal thickness; SRF, subretinal fluid; VA, visual acuity.
Unless otherwise indicated due to limited data on chart review.
Statistically significant (P < .05).
Wilcoxon rank sum test of difference between medians.
Fisher exact test of difference between proportions.
With respect to medication use (Table 3), more patients in the older group used any aspirin (60% vs 23%; P = .002), anticoagulants (16% vs 0%; P = .032), and fish oil supplements (26% vs 5%; P = .020).
Clinical Outcomes
Table 4 shows the final follow-up eye examination data and treatment information. The mean follow-up was 189.6 ± 241 weeks for the younger cohort and 137 ± 164 weeks for the older cohort (P = .806). Younger patients had better logMAR VA in the BRVO-affected eye (0.51 ± 0.65 [Snellen 20/65] vs 1.01 ± 1.20 [Snellen 20/205]; P = .016) and the fellow eye (0.18 ± 0.47 [20/30] vs 0.59 ± 0.95 [20/78]; P < .001). Younger patients also had higher IOP in both the BRVO-affected eye (17.15 ± 3.87 mm Hg vs 14.99 ± 3.79 mm Hg; P = .013) and the fellow eye (17.70 ± 4.88 mm Hg vs 15.05 ± 3.58 mm Hg; P = .008). Younger patients also had a greater SFCT in the BRVO-affected eye (275.7 ± 92.74 µm vs 218.7 ± 76.88 µm; P = .027) and fellow eye (267.3 ± 70.29 µm vs 212.6 ± 73.77 µm; P = .006). No significant differences in the CST in the BRVO-affected eye or fellow eye were found.
No significant differences were found between the cohorts in the first treatment type used, number of intravitreal steroid or anti-VEGF injections 1 year after presentation or at the final follow-up, proportion of eyes with CME or SRF 1 year after presentation, proportion of patients who developed RVO in the fellow eye, and proportion of patients lost to follow-up.
Conclusions
This study of 302 patients with BRVO included 23 patients (7.6%) younger than 50 years. Except for hyperlipidemia, we found similar risk factor profiles between younger patients and their older counterparts. Specifically, the 2 cohorts had similar rates of hypertension, diabetes, glaucoma, and smoking. Previous studies have made similar conclusions, finding that systemic risk factors, except for diabetes, were associated with BRVO in patients of all ages. 11
The prevalence of hyperlipidemia increases with age,15 which is likely the reason there was a higher rate in the older cohort in our study. By this reasoning, the older cohort would be expected to have increased rates of hypertension and OAG, which was not reflected in our findings. However, in the younger group, a lower proportion of patients with hyperlipidemia were on statin therapy (P = .047). This may be the result of the higher rate of negative side effects, such as statin-induced myopathy, in older patients. 16 Hyperlipidemia is a known risk factor for BRVO, and previous studies have found that statin therapy is associated with a decreased risk for BRVO. 17 Thus, the lower rate of statin use among older patients than among younger patients may have contributed to the increased proportion of older patients with hyperlipidemia in this study.
The percentage of those in the US with obesity, prediabetes, and DM has increased across all age groups over the past few decades 18 ; thus, the increased number of younger patients with BRVO in this study could be a result of the increased incidence of systemic risk factors. It is possible that geographic variation in the incidence and prevalence of obesity played a role in our findings. Most patients in our sample were from North Carolina or surrounding states, whereas the study by Lam et al 11 was conducted in California. A larger comparative study of younger patients with BRVO across time would be needed to further explore this issue.
Previous work has shown that CRVO in younger patients may be associated with a variety of nontraditional risk factors, including inflammatory disorders, connective tissue disease, hypercoagulability, and contraceptive use.4,13 In contrast, our findings suggest that BRVO is associated with more common clinical risk factors, such as hyperlipidemia. Although hypercoagulability testing was unrevealing between the 2 age groups, only a minority of patients in both cohorts had a hypercoagulability evaluation.
Of the 302 patients in our study, a larger proportion of the younger patients with BRVO were of Asian descent compared with the older patients. In addition, there was a lower proportion of White patients in the younger group than in the older group. This could be the result of a variety of factors, including the ethnic makeup of the community, access to care, and possibly disease prevalence. A pooled analysis of population-based data has shown that the prevalence of BRVO may be highest in Asian individuals and lowest in White individuals. 19 Additional studies with larger samples may more reliably validate racial differences in the incidence of BRVO with respect to age.
At presentation, younger patients had better VA in the fellow eye than older patients, which is not surprising given the increased incidence of cataract and other age-related eye diseases in older individuals. Yet in the BRVO-affected eye, there was no significant difference between the 2 groups in VA at presentation. This contrasts with eyes with CRVO, in which younger patients typically present with better VA in the affected eye than their older counterparts. 20 As such, we theorize that age may not influence BRVO at presentation, as it may for CRVO. There was no significant difference between the younger cohort and older cohort with BRVO in the proportion presenting with intraocular neovascularization, CME, SRF, VH, or foveal IRH. Our study did, however, find that younger patients had a significantly greater SFCT than their older counterparts at presentation. A reduction in SFCT has been associated with aging, as reflected in our dataset. 21
Younger patients had better VA in both the BRVO-affected eye and unaffected fellow eye at the final follow-up. Even though the younger cohort and older cohort presented with similar VAs in the eye with BRVO, the VA at the final visit was significantly better in the younger cohort. Previous studies have found that younger patients have better and more rapid rehabilitation of retinal vasculature and retinal perfusion than older patients after receiving similar numbers of anti-VEGF injections, which may contribute to the better final VA in younger individuals with BRVO. 14 Nonetheless, between the 2 cohorts, there were no significant differences in the type of treatment, number of injections, CST, proportion with CME or SRF, or proportion with an enlarged foveal avascular zone. Thus, unlike CRVO, age may not be an adequate prognosticator of certain clinical outcomes or the number of injections needed; however, a better VA at the final follow-up was correlated with younger age at presentation.
Our current study has limitations inherent to a retrospective study. Although all patients received the standard of care for their BRVO diagnosis, there could have been variations in the follow-up interval, intervention, imaging, and other factors depending on physician recommendations. In addition, our dataset did not include every study variable for each patient because of the lack of availability and documentation of all desired information at various timepoints. Given the small sample of the younger patient cohort with BRVO, the generalizability of our findings may be limited.
Our findings suggest that younger patients with acute, treatment-naïve BRVO have risk factors similar to those of older patients and thus extensive testing for other etiologies, as is often recommended for younger persons with CRVO, may not yield significant findings. Treatment patterns in patients with BRVO were similar in the 2 age groups. Younger patients with BRVO may have better visual outcomes, suggesting that age may be a potential prognostic factor. Further investigation into the characteristics and outcomes of younger individuals with BRVO may yield additional insights.
Footnotes
Ethical Approval
Ethical approval for this study was obtained from the Duke Health Institutional Review Board (Pro00075701).
Statement of Informed Consent
Informed consent was not sought for the present study because of the retrospective nature of data collection.
Declaration of Conflicting Interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Dr. Thomas is a consultant to Alimera, Allergan, Avesis, Eyepoint, and Genentech, and Novartis. Dr. Fekrat received patent royalties from Alcon and research support from Genentech and Optos and is a consultant to Glaukos and Bausch + Lomb. None of the other authors reported financial disclosures or conflicts.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
