Abstract
Purpose:
This work aims to investigate real-world treatment patterns and outcomes in eyes with branch retinal vein occlusion in the antivascular endothelial growth factor (anti-VEGF) era.
Methods:
A retrospective, nonrandomized, comparative study was conducted on eyes diagnosed with branch retinal vein occlusion at a single tertiary center between 2009 and 2017. Medical history, treatment patterns, and visual acuity outcomes were examined. Subanalysis was performed for eyes that met the eligibility criteria for the BRAVO (Ranibizumab for the Treatment of Macular Edema Following Branch Retinal Vein Occlusion) trial.
Results:
A total of 315 eyes were included, of which 244 were treatment naive. In all eyes, the most common first treatment was the following: intravitreal bevacizumab (38.4%), aflibercept (15.1%), ranibizumab (8.1%), sectoral scatter laser (6.2%), and triamcinolone (3.1%). At 1 year, treatment-naive eyes had received an average of 2.43 anti-VEGF injections. During follow-up, treatment-naive eyes gained an average of 0.21 Early Treatment Diabetic Retinopathy Study lines. Forty eyes that met BRAVO trial criteria received an average of 5.05 anti-VEGF injections in the first year and gained an average of 1.83 Early Treatment Diabetic Retinopathy Study lines.
Conclusions:
This real-world cohort received fewer anti-VEGF injections at year 1 and experienced less improvement in visual acuity during the course of treatment than clinical trial participants. Trial-eligible patients received more injections and had greater visual gains than those who would not have been eligible for the trial.
Keywords: branch retinal vein occlusion, clinical outcomes, real-world data, retinal vein occlusion, treatment patterns
Introduction
Retinal vein occlusion (RVO) is the second most frequent retinal vascular disease after diabetic retinopathy, affecting 16 million individuals worldwide. 1 Branch RVO (BRVO), the more common subtype of RVO, has a reported annual incidence of up to 1.6% in the general population. 2 -4 BRVO is thought to be caused by degenerative changes in a branch retinal artery wall and compression of a branch retinal vein at an arteriovenous crossing. Development of BRVO may also be facilitated by hypercoagulability. 2,5
In BRVO, the resulting retinal ischemia increases levels of vascular endothelial growth factor (VEGF) and other cytokines, which in turn promote vascular permeability. This precipitates other complications, particularly cystoid macular edema (CME) and retinal neovascularization, 6 -8 which are often associated with reduced visual acuity (VA). 2,5 Given the critical role of VEGF in the pathophysiology of BRVO, intravitreal anti-VEGF injections continue to be a mainstay of treatment as highlighted in the American Academy of Ophthalmology’s Retinal Vein Occlusion Preferred Practice Pattern guidelines. 9 The 3 available anti-VEGF medications are bevacizumab (Avastin, Genentech), ranibizumab (Lucentis, Genentech), and aflibercept (Eylea, Regeneron).
Several clinical trials have evaluated the efficacy of anti-VEGF therapy in BRVO. The Ranibizumab for the Treatment of Macular Edema Following Branch Retinal Vein Occlusion (BRAVO) trial showed that treatment with ranibizumab was effective compared with sham in treating CME due to BRVO. 10 The Efficacy and Safety of Ranibizumab With or Without Laser in Comparison to Laser in Branch Retinal Vein Occlusion (BRIGHTER) and the Ranibizumab Dose Comparison (0.5 mg and 2.0 mg) and the Role of Laser in the Management of Retinal Vein Occlusion (RELATE) trials further demonstrated that ranibizumab injections were more effective at improving VA and decreasing central subfield thickness (CST) than grid-pattern laser photocoagulation alone. 11,12 The Study to Assess the Clinical Efficacy and Safety of Intravitreal Aflibercept Injection in Patients With Branch Retinal Vein Occlusion (VIBRANT) trial similarly showed superior outcomes for aflibercept injections compared with laser. 13,14 Regarding the relative effectiveness of the anti-VEGF medications, the Macular Edema due to Branch Retinal Vein Occlusion (MARVEL) trial showed similar improvements in VA and CST at 6 months between those treated with bevacizumab and those treated with ranibizumab. 15
Although these randomized clinical trials established the safety and efficacy of anti-VEGF injections in the treatment of BRVO, 16 there is also evidence that real-world practice patterns and outcomes may differ from the controlled environments of clinical trials. A recent study examined cases of CME due to RVO and noted that in clinical practice, anti-VEGF injections were administered less frequently and resulted in suboptimal improvement in VA. 17 Further studies have shown similar findings after examining real-world anti-VEGF use and efficacy in Germany, 18 Denmark, 19 Portugal, 20 and the United Kingdom. 21,22
Given that practice patterns and outcomes in RVO often differ from those reported in clinical trials, we sought to assess them in a relatively large cohort of well-characterized eyes with BRVO treated by 22 academic retina specialists in a tertiary center. We further compared practice patterns and outcomes in a subset of clinical trial-eligible BRVO eyes with the cohorts treated in the context of multicenter clinical trials.
Methods
With approval from the Duke Institutional Review Board and in alignment with the tenets of the Declaration of Helsinki, a retrospective medical-record review was conducted. All patients diagnosed with BRVO at a single tertiary academic referral center between January 2009 and June 2017 were identified. All patients were followed at our institution after referral and diagnosis. They were identified using the Duke Enterprise Data Unified Content Explorer (Duke University Health System). Individuals were excluded if treatment history or date of initial BRVO diagnosis was unknown.
Demographic information was collected including age, hypertension, diabetes, glaucoma, medications, and prior eye surgery. Details of ocular examination and diagnostic workup including gonioscopy, fluorescein angiography (FA), and optical coherence tomography (OCT) were noted. The method of determining fovea-involving intraretinal hemorrhage (IRH) in this cohort is detailed elsewhere, 23 but in brief, 2 independent masked graders evaluated for IRH on color fundus photographs, and a third masked grader analyzed them further using multimodal imaging. For patients that either had BRVO in both eyes or subsequently developed BRVO in the fellow eye, the eye that developed BRVO first was used for analysis. Various modes and dates of treatment were also recorded. Corrected VA and intraocular pressure (IOP) were collected at the baseline and final visits. The baseline visit was considered the presenting visit for BRVO and the final visit was the final examination available.
Descriptive statistics were calculated for the overall cohort as well as subgroups of treatment-naive eyes, nontreatment-naive eyes, and eyes that met BRAVO clinical trial inclusion criteria (trial-eligible subgroup). 10 The BRAVO trial criteria were used because that trial was the first to assess the effectiveness of anti-VEGF treatment for BRVO and offered a clear comparison of anti-VEGF treatment vs sham. In the case of missing data, the overall number of available data points was noted. Corrected VA was converted from Snellen equivalent to the logarithm of the minimum angle of resolution (logMAR). All data were analyzed with Python version 3.7.6.
Results
Patient Characteristics
A total of 315 eyes with BRVO of 315 patients were identified. Seventy-one had received treatment elsewhere prior to presentation (nontreatment naive), and 244 had not (treatment naive). The median age was 69 years (interquartile range [IQR], 61-78 years), and 58% were female. Sixty-seven percent of patients identified as White and 23% as African-American. The median time from initial to final follow-up was 18 months (IQR, 6-36 months). Medical, ocular, and medication history are summarized in Table 1.
Table 1.
Medical, Ocular, and Medication-Use History of Patients.
| n (%) | N | |
|---|---|---|
| Medical and ocular history | ||
| Hypertension | 241 (79.8) | 302 |
| Diabetes | 95 (32.4) | 293 |
| Glaucoma | 89 (30.7) | 290 |
| Smoking | 54 (22.5) | 240 |
| Lens status at baseline | 280 | |
| Phakic | 174 (62.1) | |
| Pseudophakic | 106 (37.9) | |
| Medication use | ||
| ACE inhibitor | 96 (32.2) | 289 |
| β Blocker | 124 (42.9) | 289 |
| Oral contraceptive | 3 (1.1) | 267 |
| Hormone replacement therapy | 16 (5.6) | 286 |
| Aspirin, 81 mg | 145 (50.2) | 289 |
| Aspirin, 325 mg | 28 (9.9) | 283 |
| Warfarin | 24 (8.5) | 284 |
| Clopidogrel | 28 (9.8) | 286 |
| Rivaroxaban | 6 (2.1) | 282 |
| Apixaban | 12 (4.3) | 282 |
| Dabigatran | 3 (1.1) | 283 |
| Fish oil | 60 (21.1) | 284 |
| Metformin | 51 (17.6) | 289 |
| Pioglitazone | 8 (2.8) | 284 |
Abbreviation: ACE, angiotensin-converting enzyme.
BRVO Characteristics
The median duration of symptoms related to BRVO prior to presentation was 2 months for treatment-naive eyes and 17 months for nontreatment-naive eyes. On presentation, vitreous hemorrhage was found in 6.2% of eyes, and foveal IRH was found in 20%. Neovascularization of the iris, angle, disc, or elsewhere was noted in 0%, 1.2%, 2.1%, and 6.7% of cases, respectively. Of the patients for whom OCT was performed, CME was documented in 68.1% and 51.4% of eyes at baseline and final follow-up, respectively. Subretinal fluid was noted in 19.6% of eyes at baseline visit and in 5.4% of eyes at final visit. The average CST at initial and final follow-up were 390 µm and 324 μm, respectively. There were no differences in presenting characteristics between treatment-naive and nontreatment-naive groups (all P > .05).
Treatment History of Nontreatment-Naive Eyes
There were 71 nontreatment-naive eyes. The most common treatments administered prior to presentation were intravitreal anti-VEGF injections (71.2%). Bevacizumab, ranibizumab, and aflibercept were administered to 49.2%, 17.2%, and 13.6% of eyes, respectively. Among those presenting after October 2014—when aflibercept was approved for BRVO treatment—21.0% had previously received aflibercept, whereas none of the patients presenting before October 2014 had received aflibercept. There were no other differences in treatment history among nontreatment-naive eyes presenting before and after October 2014.
In addition to anti-VEGF injections, the most common treatments given prior to presentation were grid-pattern laser (29.0%) and sectoral scatter laser (27.9%), followed by steroids (10.8%) in the form of posterior sub-Tenon triamcinolone, intravitreal triamcinolone, and intravitreal dexamethasone (Ozurdex, Allergan).
Imaging and Laboratory Workup
OCT was the most commonly (96.8%) performed imaging modality at presentation. In 77.3% of eyes, FA was also performed, and 33.2% of eyes had additional FAs during follow-up. Laboratory workup to identify an underlying etiology of BRVO was performed in 9.0% of treatment-naive eyes and 4.4% of nontreatment-naive eyes. Full details of the diagnostic workup performed on presentation are outlined in Table 2.
Table 2.
Diagnostic Procedures Performed.
| All eyes | Nontreatment-naive eyes | Treatment-naive eyes | ||||
|---|---|---|---|---|---|---|
| n (%) | N | n (%) | N | n (%) | N | |
| Workup for etiology of BRVO | 21 (7.9) | 267 | 3 (4.4) | 68 | 18 (9.0) | 199 |
| Gonioscopy on presentation | 6 (2.9) | 209 | 3 (5.6) | 54 | 3 (1.9) | 155 |
| FA obtained on presentation | 269 | 69 | 200 | |||
| Yes | 208 (77.3) | 51 (73.9) | 157 (78.5) | |||
| 30° without sweeps | 44 (16.4) | 13 (18.8) | 31 (15.5) | |||
| 30° with sweeps | 110 (40.9) | 23 (33.3) | 87 (43.5) | |||
| UWF | 54 (20.1) | 15 (21.7) | 39 (19.5) | |||
| Additional FA during follow-up | 93 (33.2) | 280 | 19 (28.4) | 67 | 74 (34.7) | 213 |
| OCT obtained on presentation | 271 (96.8) | 280 | 67 (97.1) | 69 | 204 (96.7) | 211 |
Abbreviations: BRVO, branch retinal vein occlusion; FA, fluorescein angiography; OCT, optical coherence tomography; UWF, ultra-widefield (200°).
Patterns in Treatment Practices
Table 3 details the treatment patterns in eyes with BRVO. On the day of presentation, 42.9% of all eyes received treatment. This proportion was similar for treatment-naive (43.5%) and nontreatment-naïve (41.2%) eyes. The most common first treatment administered was intravitreal bevacizumab (38.4%), followed by aflibercept (15.1%) and ranibizumab (8.1%). Among patients presenting after the approval of aflibercept for BRVO in October 2014, intravitreal bevacizumab, aflibercept, and ranibizumab were the initial treatments in 32.9%, 24.2%, and 6% of eyes, respectively. Among patients presenting prior to October 2014, intravitreal bevacizumab and ranibizumab, respectively, were the initial treatments in 45.9% and 11.0% of eyes. None were started on aflibercept.
Table 3.
Treatment.
| All eyes (N = 315) | Nontreatment-naive eyes (n = 71) | Treatment-naive eyes (n = 244) | |
|---|---|---|---|
| % | |||
| First treatment | |||
| Bevacizumab | 38.4 | 30.5 | 40.7 |
| Ranibizumab | 8.1 | 10.2 | 7.5 |
| Aflibercept | 15.1 | 22.0 | 13.1 |
| PST | 0.0 | 0.0 | 0.0 |
| Intravitreal triamcinolone | 3.1 | 8.5 | 1.5 |
| Dexamethasone implant | 0.4 | 0.0 | 0.5 |
| Grid-pattern laser | 1.9 | 5.1 | 1.0 |
| Intravitreal tPA | 0.4 | 0.0 | 0.5 |
| Fluocinolone implant | 0.0 | 0.0 | 0.0 |
| First intravitreal anti-VEGF | |||
| Bevacizumab | 35.6 | 28.2 | 37.7 |
| Ranibizumab | 8.6 | 11.3 | 7.8 |
| Aflibercept | 12.4 | 19.7 | 10.2 |
| Mean ± SD (range) | |||
| Total No. grid-pattern laser | 0.26 ± 0.62 (0-4) | 0.24 ± 0.32 (0-2) | 0.27 ± 0.65 (0-4) |
| Total No. PST | 0.03 ± 0.22 (0-3) | 0.06 ± 0.24 (0-1) | 0.02 ± 0.22 (0-3) |
| Total No. IVTA | 0.29 ± 1.73 (0-27) | 0.70 ± 3.39 (0-27) | 0.17 ± 0.61 (0-4) |
| Total No. dexamethasone | 0.06 ± 0.45 (0-5) | 0.10 ± 0.46 (0-3) | 0.05 ± 0.44 (0-5) |
| Total No. bevacizumab | 1.67 ± 3.45 (0-25) | 0.97 ± 1.97 (0-9) | 1.88 ± 3.75 (0-25) |
| Total No. ranibizumab | 1.36 ± 4.66 (0-42) | 0.75 ± 3.33 (0-26) | 1.53 ± 5.00 (0-42) |
| Total No. aflibercept | 1.96 ± 4.50 (0-28) | 1.57 ± 3.83 (0-19) | 2.08 ± 4.70 (0-28) |
| Total No. anti-VEGF at 1 ya | 2.41 ± 3.25 (0-12) | 2.37 ± 3.03 (0-12) | 2.43 ± 3.32 (0-12) |
| Total No. anti-VEGF at final follow-up | 5.09 ± 7.82 (0-54) | 4.03 ± 6.26 (0-27) | 5.41 ± 8.21 (0-54) |
Abbreviations: anti-VEGF, antivascular endothelial growth factor; IVTA, intravitreal triamcinolone acetonide; PST, posterior sub-Tenon triamcinolone; tPA, tissue plasminogen activator.
a Comparison includes only those with 1 or more years of follow-up.
Triamcinolone and grid-pattern laser were the first treatments initiated in 3.1% and 1.9% of eyes, respectively. Fewer than 1% of eyes were initiated on dexamethasone or intravitreal tissue plasminogen activator as the first treatment. The remaining 25% of eyes were not initiated on a treatment at presentation for a variety of reasons including presence of a chronic BRVO for which visual improvement may not have been thought likely, occurrence of noncentral edema with good vision for which a clinical decision was made to monitor closely, patient deferring treatment, insurance hurdles, and transportation issues. For 267 eyes with at least 1 year of follow-up, the average number of anti-VEGF injections received at 1 year was 2.41 (SD, 3.25). Sectoral scatter laser was also performed at the presenting visit in 6.2% of all eyes, and 7.5% and 1.7% of treatment-naive and nontreatment-naive eyes, respectively.
Visual Outcomes
For treatment-naive eyes, the mean logMAR VA at the baseline visit and final follow-up was the same at 0.43 (baseline SD, 0.45; final SD, 0.56; Snellen equivalent, 20/54). For the nontreatment-naive eyes, the mean logMAR VA at baseline was 0.61 (SD, 0.62; Snellen, 20/81) and at final follow-up was 0.70 (SD, 0.67; Snellen, 20/100). Among treatment-naive eyes, 19.6% gained 3 or more lines over a median follow-up of 17 months. Among nontreatment-naive eyes, 12.8% experienced an improvement of at least 3 lines over a median follow-up of 21 months. Baseline and final IOP measurements as well as the mean number of lines gained for each group are listed in Table 4.
Table 4.
Visual Acuity and Intraocular Pressure Outcomes.
| All eyes (N = 315) |
Nontreatment-naive eyes (n = 71) |
Treatment-naive eyes (n = 244) |
|
|---|---|---|---|
| On presentation, mean ± SD | |||
| logMAR VA (Snellen) | 0.48 ± 0.51 (20/60) | 0.61 ± 0.62 (20/81) | 0.43 ± 0.45 (20/54) |
| IOP | 14.98 ± 4.01 | 15.43 ± 4.78 | 14.81 ± 3.69 |
| Final visit, mean ± SD | |||
| logMAR VA (Snellen) | 0.48 ± 0.59 (20/60) | 0.70 ± 0.67 (20/100) | 0.43 ± 0.56 (20/54) |
| IOP | 15.59 ± 3.81 | 15.71 ± 4.36 | 15.57 ± 3.67 |
| Mean No. lines gained,a mean ± SD | –0.09 ± 4.56 | –1.19 ± 4.37 | 0.21 ± 4.58 |
| Gained > 3, % | 18.14 | 12.77 | 19.64 |
| Lost ≤ 3, % | 11.63 | 21.28 | 8.93 |
Abbreviations: IOP, intraocular pressure; VA, visual acuity.
a Mean number of lines gained from presentation to final visit for all eyes within each group.
Treatment Burden and Visual Outcomes of Ranibizumab for the Treatment of Macular Edema in BRVO Trial–Eligible Eyes
Forty treatment-naive eyes were identified in our cohort that met eligibility criteria for BRAVO. In this subgroup, the median age was 73 years (IQR, 64-78 years), the proportion of women was 47.5%, and the median duration of follow-up was 26 months (IQR, 17-35 months). Of the 40 eyes, the first treatment administered was bevacizumab in 19 (47.5%), ranibizumab in 5 (12.5%), aflibercept in 5 (12.5%), triamcinolone in 2 (5%), dexamethasone in 1 (2.5%), and grid-pattern laser in 2 (5%). The remaining 6 (15%) were initially observed without treatment. The average number of anti-VEGF injections received at 1 year was 5.05 (SD, 3.81). The average baseline VA was 0.57 logMAR (SD, 0.22; Snellen, 20/74), and the average VA at 1 year was 0.37 (SD, 0.27; Snellen, 20/47). Thirty-seven percent of these eyes gained at least 3 lines during this time, and 3% worsened by at least 3 lines. The mean number of lines gained on the Snellen chart at 1 year was 1.83 (SD, 2.74) or an increase in corrected VA letter score of 9.15. A comparison of the total number of injections and mean change in VA at 1 year between our trial-eligible cohort and various clinical trials is shown in Figure 1.
Figure 1.
Comparison of mean number of injections at 1 year and mean change in visual acuity (letters) for branch retinal vein occlusion across trials and studies. BCVA, indicates best-corrected visual acuity; BRAVO, Ranibizumab for the Treatment of Macular Edema Following Branch Retinal Vein Occlusion trial; VIBRANT, Study to Assess the Clinical Efficacy and Safety of Intravitreal Aflibercept Injection in Patients With Branch Retinal Vein Occlusion trial.
Conclusions
We report on a large cohort of BRVO eyes evaluated by 22 retina specialists at a tertiary referral center spanning 8 years. Treatment patterns and clinical outcomes resulting from real-world clinical decision-making are highlighted. The main findings are as follows: (1) The most commonly administered first treatment was intravitreal bevacizumab at 38.4%; (2) treatment-naive eyes had received an average of 2.43 anti-VEGF injections at 1 year; (3) from initial presentation to final visit, treatment-naive eyes had a slight increase in VA with an average of 0.21 lines gained from initial presentation to final visit; (4) at 1 year, eyes that would have been eligible for the BRAVO trial received an average of 5.05 anti-VEGF injections and gained an average of 1.83 lines, which were less than that of eyes actually in the BRAVO clinical trial.
Among the entire cohort, the most common treatment to be administered first for management of BRVO was intravitreal bevacizumab, followed by ranibizumab and aflibercept. This is consistent with the results of the annual Preferences and Trends survey conducted in 2018 by the American Society of Retina Specialists that showed that bevacizumab is the most commonly used first-line treatment among retina specialists in the United States. 24 However, there are no definitive data establishing one anti-VEGF medication as clearly superior to another. The MARVEL trial showed that efficacy and safety were comparable between ranibizumab and bevacizumab. 15 Although the Retinal Vein Occlusion Preferred Practice Pattern mentions anti-VEGF as a class of medications used as first-line treatment, it does not specify whether one medication is better. 9 Given that there are differing costs to these medications and other economic considerations, 25,26 shared decision-making between retina specialists and patients and awareness of the practices of others can help in deciding which anti-VEGF medication to use.
In our study, the entire cohort of eyes presenting with BRVO received an average of 2.41 anti-VEGF injections over the first year of follow-up. While treatment strategies varied among retina specialists at our institution, a treat-and-extend approach was most commonly used. This 2.41 injections at 1 year is fewer than has been reported in clinical trials. For example, in the BRAVO trial, patients were randomly assigned to receive intravitreal ranibizumab injections monthly for 6 months and then as needed for the subsequent 6 months. The mean number of injections at 6 months was 5.7 and at 12 months was 8.4. 27 In the MARVEL trial, patients received an initial baseline injection and then monthly as-needed injections, based on predefined criteria. 15 At 1 year in MARVEL, the mean number of intravitreal anti-VEGF injections stratified by central retinal thickness (CRT) was 2.91 for CRT less than 500 µm and 3.4 for CRT greater than 500 µm. 28 Patients randomly assigned to the intravitreal aflibercept arm of the VIBRANT trial received injections every 4 weeks for the first 20 weeks and then every 8 weeks until 48 weeks, and thus received an average of 9 injections of intravitreal aflibercept during this time. 13
These differences in treatment frequency and number may be explained in part by undertreatment in real-world cohorts, institutional follow-up practices, differences in patient populations, the proportion of chronic BRVO patients who received fewer injections despite the presence of CME, and other factors. In any case, these differences are important to characterize because intravitreal anti-VEGF injections have been noted to have rare but potentially serious complications. Various trials have noted injection-related cataract, 10,13 epiretinal membrane, 15 increased IOP, 15 vitreous hemorrhage, 27 endophthalmitis, 10 and retinal detachment. 10 These potential complications are often taken into account when considering treatment and frequency of injections and are counterbalanced with the therapeutic effects of treatment to optimize outcomes.
For treatment-naive eyes in our study, mean VA on presentation and at the final visit was approximately the same at 20/54. The mean number of lines gained for this group was small (0.21, or corrected VA letter score of 1.05), with 19.6% of eyes having gained at least 3 lines at final follow-up and 8.9% having lost at least 3 lines. Given that the median follow-up period for treatment-naive eyes in our cohort was 17 months, we could not directly compare the VA improvements with those of clinical trials, which generally have reported 6- or 12-month outcomes. Nonetheless, a mean corrected VA letter score improvement of 1.05 at a median follow-up of 17 months stood in stark contrast to the results of several major trials that have shown marked improvements in VA at the 6- or 12-month time point. Various trials have reported mean improvement of corrected VA letter score of 17.1 with aflibercept, 13 12.1 to 18.3 with ranibizumab, 12,15 and 15.6 with bevacizumab. 15 Given that these trials had treatment protocols involving monthly injections initially that resulted in greater numbers of injections, it could be possible that real-world undertreatment with anti-VEGF may be contributing to these differences in outcomes.
Another possibility is a potential ceiling effect of the relatively good VA at baseline (20/54), especially since BRAVO trial patients’ VA at 1 year was about 20/40, which is not too different from the present study. In addition, another likely source of the difference is the characteristics of our entire cohort of patients at presentation. Many would not have qualified for clinical trial enrollment based on comorbidities, disease presentation, VA, and other factors.
To explore this further, we conducted a subgroup analysis on 40 treatment-naive eyes that would have met eligibility criteria for enrollment in BRAVO 10 and analyzed 12-month VA outcomes. By doing so, we sought to address 2 confounding factors in the comparison of our cohort with those of clinical trials: (1) eligibility for trials based on clinical and medical history characteristics, and (2) VA outcomes at 12 months, so that improvements can be assessed using the same time interval.
In this subgroup, 19 were started on bevacizumab, 5 on ranibizumab, 5 on aflibercept, 2 on triamcinolone, 1 on dexamethasone, and 2 with grid-pattern laser, whereas in BRAVO, all eyes received ranibizumab. 10 We found that in our trial-eligible subgroup, the average number of anti-VEGF injections at 1 year was 5.05, more than double that for our entire treatment-naive cohort (2.43) but still less than that administered in BRAVO (8.4). 27 In addition, the mean improvement in letter score for our trial-eligible group at 1 year was 9.15, also greater than the 1.05 lines gained by the total treatment-naive group in our cohort but still less than that in BRAVO (16.4 for the group receiving 0.3 mg doses and 18.3 for the group receiving 0.5 mg doses). 27 Although less than clinical trials, the greater improvement in our trial-eligible subgroup compared with our overall cohort suggested that results from BRVO clinical trials may not be fully generalizable to real-world BRVO eyes. Moreover, given that the average number of injections and improvement in VA in our trial-eligible subgroup were still less than that of clinical trials, the difference in outcomes may also have been attributed to undertreatment in real-world clinical settings. Thus, perhaps more closely following trial treatment protocols may be of greater visual benefit for those with BRVO.
The present study has limitations inherent to its retrospective design. A causal relationship between VA improvement and overall number of treatments could not be determined. Not all eyes in our cohort had complete clinical data at each time point, which may have introduced some selection bias. Not all eyes had VA data at 12 months, which prevented a direct comparison of VA improvement in our entire cohort with those of clinical trials. However, this potential confounder was addressed in our subgroup analysis of trial-eligible eyes that had 12-month VA outcomes. Another potential confounder was lens status. At baseline, 62% were phakic. Some of these patients may have undergone cataract surgery over time, and because we did not control for lens status, this may have affected final VA outcomes. In addition, since BRVO is a heterogeneous clinical condition, nuances in clinical decision-making may not have been fully captured. Differences in diagnostic or treatment patterns may exist at different institutions or practice settings, which would not be elucidated here since we examined data from 1 large tertiary referral center. Future multicenter studies could assess differences in real-world management of BRVO among various institutions and practice settings.
Our study presents findings regarding BRVO management patterns and outcomes among 22 retina specialists at a tertiary referral center. It contributes to the existing literature and provides further evidence highlighting the clinical decisions—and associated outcomes—of BRVO management within a real-world setting. It raises the question of whether results from clinical trials can be extrapolated given such differences in baseline characteristics from a general BRVO population. Exploring and understanding real-world data allow the assessment and establishment of standards of BRVO treatment, taking into account the practical constraints of day-to-day clinical practice.
Footnotes
Ethical Approval: This research was conducted in accordance with the Declaration of Helsinki. It was approved by the Duke University Institutional Review Board (Pro00075701) and complied with HIPAA (the Health Insurance Portability and Accountability Act).
Statement of Informed Consent: The Duke University Institutional Review Board waived the need for informed consent because this study was retrospective in nature and all patient data were deidentified.
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
ORCID iDs: Terry Lee, BA 
https://orcid.org/0000-0002-3852-805X
Cason B. Robbins, BS
https://orcid.org/0000-0001-7909-510X
Sharon Fekrat, MD
https://orcid.org/0000-0003-4403-5996
References
- 1. Rogers S, McIntosh RL, Cheung N, et al. The prevalence of retinal vein occlusion: pooled data from population studies from the United States, Europe, Asia, and Australia. Ophthalmology. 2010;117(2):313–322. doi:10.1016/j.ophtha.2009.07.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Li J, Paulus YM, Shuai Y, Fang W, Liu Q, Yuan S. New developments in the classification, pathogenesis, risk factors, natural history, and treatment of branch retinal vein occlusion. J Ophthalmol. 2017;2017:4936924. doi:10.1155/2017/4936924 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Klein R, Moss SE, Meuer SM, Klein BEK. The 15-year cumulative incidence of retinal vein occlusion: the Beaver Dam Eye Study. Arch Ophthalmol. 2008;126(4):513–518. doi:10.1001/archopht.126.4.513 [DOI] [PubMed] [Google Scholar]
- 4. Zhou JQ, Xu L, Wang S, et al. The 10-year incidence and risk factors of retinal vein occlusion: the Beijing Eye Study. Ophthalmology. 2013;120(4):803–808. doi:10.1016/j.ophtha.2012.09.033 [DOI] [PubMed] [Google Scholar]
- 5. Jaulim A, Ahmed B, Khanam T, Chatziralli IP. Branch retinal vein occlusion: epidemiology, pathogenesis, risk factors, clinical features, diagnosis, and complications. An update of the literature. Retina. 2013;33(5):901–910. doi:10.1097/IAE.0b013e3182870c15 [DOI] [PubMed] [Google Scholar]
- 6. Nishinaka A, Inoue Y, Fuma S, et al. Pathophysiological role of VEGF on retinal edema and nonperfused areas in mouse eyes with retinal vein occlusion. Investig Ophthalmol Vis Sci. 2018;59(11):4701–4713. doi:10.1167/iovs.18-23994 [DOI] [PubMed] [Google Scholar]
- 7. Noma H, Minamoto A, Funatsu H, et al. Intravitreal levels of vascular endothelial growth factor and interleukin-6 are correlated with macular edema in branch retinal vein occlusion. Graefes Arch Clin Exp Ophthalmol. 2006;244(3):309–315. doi:10.1007/s00417-004-1087-4 [DOI] [PubMed] [Google Scholar]
- 8. Noma H, Funatsu H, Mimura T, Eguchi S, Shimada K. Role of soluble vascular endothelial growth factor receptor-2 in macular oedema with central retinal vein occlusion. Br J Ophthalmol. 2011;95(6):788–792. doi:10.1136/bjo.2010.192468 [DOI] [PubMed] [Google Scholar]
- 9. Flaxel CJ, Adelman RA, Bailey ST, et al. Retinal Vein Occlusions Preferred Practice Pattern. Ophthalmology. 2020;127(2):P288–P320. doi:10.1016/j.ophtha.2019.09.029 [DOI] [PubMed] [Google Scholar]
- 10. Campochiaro PA, Heier JS, Feiner L, et al. Ranibizumab for macular edema following branch retinal vein occlusion. Six-month primary end point results of a phase III study. Ophthalmology. 2010;117(6):1102–1112.e1. doi:10.1016/j.ophtha.2010.02.021 [DOI] [PubMed] [Google Scholar]
- 11. Tadayoni R, Waldstein SM, Boscia F, et al. Sustained benefits of ranibizumab with or without laser in branch retinal vein occlusion: 24-month results of the BRIGHTER study. Ophthalmology. 2017;124(12):1778–1787. doi:10.1016/j.ophtha.2017.06.027 [DOI] [PubMed] [Google Scholar]
- 12. Campochiaro PA, Hafiz G, Mir TA, et al. Scatter photocoagulation does not reduce macular edema or treatment burden in patients with retinal vein occlusion: the RELATE trial. Ophthalmology. 2015;122(7):1426–1437. doi:10.1016/j.ophtha.2015.04.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Clark WL, Boyer DS, Heier JS, et al. Intravitreal aflibercept for macular edema following branch retinal vein occlusion: 52-week results of the VIBRANT study. Ophthalmology. 2016;123(2):330–336. doi:10.1016/j.ophtha.2015.09.035 [DOI] [PubMed] [Google Scholar]
- 14. Campochiaro PA, Clark WL, Boyer DS, et al. Intravitreal aflibercept for macular edema following branch retinal vein occlusion: the 24-week results of the VIBRANT study. Ophthalmology. 2015;122(3):538–544. doi:10.1016/j.ophtha.2014.08.031 [DOI] [PubMed] [Google Scholar]
- 15. Narayanan R, Panchal B, Das T, et al. A randomised, double-masked, controlled study of the efficacy and safety of intravitreal bevacizumab versus ranibizumab in the treatment of macular oedema due to branch retinal vein occlusion: MARVEL report No. 1. Br J Ophthalmol. 2015;99(7):954–959. doi:10.1136/bjophthalmol-2014-306543 [DOI] [PubMed] [Google Scholar]
- 16. Shalchi Z, Mahroo O, Bunce C, Mitry D. Anti-vascular endothelial growth factor for macular oedema secondary to branch retinal vein occlusion. Cochrane Database Syst Rev. 2020;7(7):CD009510. doi:10.1002/14651858.CD009510.pub3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Jumper JM, Dugel PU, Chen S, Blinder KJ, Walt JG. Anti-VEGF treatment of macular edema associated with retinal vein occlusion: patterns of use and effectiveness in clinical practice (ECHO study report 2). Clin Ophthalmol. 2018;12:621–629. doi:10.2147/OPTH.S163859 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Wecker T, Ehlken C, Bühler A, et al. Five-year visual acuity outcomes and injection patterns in patients with pro-re-nata treatments for AMD, DME, RVO and myopic CNV. Br J Ophthalmol. 2017;101(3):353–359. doi:10.1136/bjophthalmol-2016-308668 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Vorum H, Olesen TK, Zinck J, Størling Hedegaard M. Real world evidence of use of anti-VEGF therapy in Denmark. Curr Med Res Opin. 2016;32(12):1943–1950. doi:10.1080/03007995.2016.1221803 [DOI] [PubMed] [Google Scholar]
- 20. Vaz-Pereira S, Marques IP, Matias J, Mira F, Ribeiro L, Flores R. Real-world outcomes of anti-VEGF treatment for retinal vein occlusion in Portugal. Eur J Ophthalmol. 2017;27(6):756–761. doi:10.5301/ejo.5000943 [DOI] [PubMed] [Google Scholar]
- 21. Lip PL, Cikatricis P, Sarmad A, et al. Efficacy and timing of adjunctive therapy in the anti-VEGF treatment regimen for macular oedema in retinal vein occlusion: 12-month real-world result. Eye. 2018;32(3):537–545. doi:10.1038/eye.2017.230 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Gale R, Gale R, Pikoula M, et al. UK EMR Users Group. Real world evidence on 5661 patients treated for macular oedema secondary to branch retinal vein occlusion with intravitreal anti-vascular endothelial growth factor, intravitreal dexamethasone or macular laser. Br J Ophthalmol. Published online June 12, 2020. doi:10.1136/bjophthalmol-2020-315836 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Powers JH, Thomas AS, Mir TA, et al. Impact and implication of fovea-involving intraretinal hemorrhage after acute branch retinal vein occlusion. Ophthalmol Retin. 2019;3(9):760–766. doi:10.1016/j.oret.2019.04.005 [DOI] [PubMed] [Google Scholar]
- 24. Stone TW, Hahn P, Raef S. ASRS 2018 preferences and trends membership survey. 2018. Accessed June 29, 2020. www.asrs.org/asrs-community/pat-survey
- 25. Smiddy WE. Economic considerations of macular edema therapies. Ophthalmology. 2011;118(9):1827–1833. doi:10.1016/j.ophtha.2010.12.034 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Suñer IJ, Margolis J, Ruiz K, Tran I, Lee P. Direct medical costs and resource use for treating central and branch retinal vein occlusion in commercially insured working-age and Medicare populations. Retina. 2014;34(11):2250–2258. doi:10.1097/IAE.0000000000000217 [DOI] [PubMed] [Google Scholar]
- 27. Brown DM, Campochiaro PA, Bhisitkul RB, et al. Sustained benefits from ranibizumab for macular edema following branch retinal vein occlusion: 12-month outcomes of a phase III study. Ophthalmology. 2011;118(8):1594–1602. doi:10.1016/j.ophtha.2011.02.022 [DOI] [PubMed] [Google Scholar]
- 28. Narayanan R, Stewart M, Chhablani J, et al. Baseline morphological characteristics as predictors of final visual acuity in patients with branch retinal vein occlusions: MARVEL report no. 3. Indian J Ophthalmol. 2018;66(9):1291–1294. doi:10.4103/ijo.IJO_342_18 [DOI] [PMC free article] [PubMed] [Google Scholar]