The Paradigm Shifts in the Management of Retinal Vein Occlusion
Anti-VEGF is playing an increasing role in effective treatment of RVO.
John A. Wells, III, MD
New pharmacologic therapies have forever changed the management of macular edema due to retinal vein occlusions (Figure 1). The lessons learned from the landmark 1984 Branch Vein Occlusion Study (BVOS)1 and 1995 Central Vein Occlusion Study (CVOS)2,3 need to be reconsidered in the light of data from recent randomized clinical trials of various intravitreal steroids and anti-VEGF agents for the management of macular edema due to RVO.
Figure 1. Central retinal vein occlusion with macular edema in the left eye of a patient with hypertension.
While drug therapy for macular edema due to both branch and central retinal vein occlusions has clearly led to better vision and anatomic outcomes compared to the laser therapies of the previous era, the treatment burden and costs to physicians, patients and society are considerable. In this brief overview of changes in our management of RVO, I will consider how to apply new clinical trial data to the modern management of RVO, highlighting areas in which older management recommendations should no longer apply and tips from personal experience with the management of RVO.
The three decisions to consider are: (1) when to initiate intravitreal treatment, (2) with which drug, and (3) how often and for how long.
WHEN TO INITIATE TREATMENT
When determining whether to initiate therapy in RVO, it is important to consider the natural history vs the potential benefits and risks of therapy. As BRVO and CRVO are very different, we should consider each separately.
The true natural history of BRVO is poorly understood. As pointed out recently by Decroos and Fekrat, the natural history data set from the BVOS is limited to 35 untreated eyes followed for variable durations of perfused BRVO from three to 18 months prior to enrollment. No natural history from the onset of BRVO was obtained because study entry mandated at least three months' duration. However, though 37% of eyes assigned to observation rather than laser gained two or more lines of vision, the mean final vision was 20/70, and only 34% of eyes were >20/40, while 23% were <20>20>4
A recent meta-analysis of the literature for information on the natural history of BRVO suggested that 18% to 41% of eyes with macular edema due to BRVO resolve without treatment. However, vision better than 20/40 is uncommon without treatment, and vision may improve by only one to two lines over time.5
Recent clinical trials have also revealed limited natural history data. The SCORE-BRVO study compared intravitreal triamcinolone acetonide to grid laser that was deferred only if dense macular hemorrhage precluded laser, so no natural history information was obtained.6
In the BRAVO trial of intravitreal ranibizumab, laser was delayed for three months in the sham injection eyes, so there was three-month natural history data on 132 eyes in this treatment arm. Seventeen percent of these eyes gained three or more lines of vision at three months. Again, these eyes had variable durations of BRVO (0-16 months), and 13% had hemiretinal vein occlusions, so this was not a true natural history group. Also, while the sham injection group gained 7.3 letters at six months compared to baseline, 55% of these eyes had laser between three and six months.7
Also, in the trials for the Ozurdex (Allergan) dexamethasone implant, a subgroup analysis showed 20% of BRVO eyes in the sham group gained three or more lines of vision in six months.12
Thus, when counseling a patient with BRVO, it is difficult to give an accurate estimation of the chances of vision 0.improvement without treatment vs with treatment. Based on our limited knowledge, however, it would seem reasonable to conclude that the chances of spontaneous recovery of vision are somewhere between 20% and 40% and that the degree of vision recovery is small over a three-year period (less than two lines).
Conversely, with treatment, vision gains can be dramatic and large, particularly if vision loss is greater prior to initiation of therapy. In BRAVO, for example, eyes treated with ranibizumab monthly for six months gained 18 letters from baseline and improved by 11 letters, compared to the sham injection plus deferred laser group.7 Thus, intravitreal ranibizumab treatment in eyes with <20/40 vision seems to offer a better chance for vision recovery than observation.
As the SCORE-BRVO trial failed to identify a benefit of intravitreal steroid therapy compared to grid laser,6 I do not generally recommend intravitreal steroid therapy for BRVO. Also, the vision outcomes in BRAVO seemed superior to laser (which will be reviewed later), so intravitreal ranibizumab seems to offer the best vision results.
Guidelines to Follow
If intravitreal treatment provides better vision outcomes than observation or laser in BRVO, should you apply the BVOS guidelines for the application of grid laser to intravitreal treatment? The BVOS recommended that patients with perfused macular edema due to BRVO should be observed for at least three months before consideration of grid laser. It also did not recommend treatment for eyes that had better than 20/40 vision, as there was no evidence of benefit from laser unless vision was 20/40 or worse.1
These have been the standard recommendations for the management of macular edema due to BRVO for more than 25 years. But should they still apply? In the recent clinical trials of steroids and anti-VEGF agents, enrollment was generally restricted to eyes with vision of 20/40 to 20/400, but no three-month duration restriction applied. Therefore, data from these trials do not provide information about the course of vision in eyes presenting with better than 20/40 vision, but they do tell us something about eyes treated at less than three months.
In BRAVO, there was rapid resolution of macular edema after the initial ranibizumab injection, with a >250-µm reduction on OCT at day 7 and attendant improvement of 7.5 letters.7 Theoretically, immediate treatment could lead to prevention of macular tissue damage from chronic edema, more rapid resolution of intraretinal hemorrhage, and prevention of neovascular complications.
Supporting this theory, eyes in the sham injection plus deferred laser arm that began to receive ranibizumab after six months gained a mean of 12 letters vs a mean gain of 18 letters at one year in the eyes treated with ranibizumab from baseline. Once begun, the sham injection group improved from a seven-letter to a 12-letter mean gain after six months of as-needed ranibizumab treatment.9
Also in BRAVO, 65% of eyes had BRVO of less than three months' duration, and these eyes gained a mean of 19.9 letters vs 16.1 letters, compared to eyes with BRVO of more than three months' duration in the 0.5-mg dose group. Additionally, in the 0.5 mg dose group, 69.3% of eyes with less than three months' duration gained more than 15 letters, compared to 50% of eyes with more than three months' duration.7 These data suggest that treatment earlier in the disease course leads to better vision outcomes.
Regarding eyes with better than 20/40 vision at presentation, no clinical trial has addressed these eyes, so data are lacking. Practically, I think one needs to consider the level of baseline vision and the degree of visual dysfunction the patient is experiencing. If vision loss is mild, say 20/40 or better, and the patient is asymptomatic, then it would seem reasonable to follow the patient for a period of time to see whether spontaneous improvement occurs.
I typically follow such patients on a monthly basis initially, and I may extend the exam interval to three months if they show stability or improvement over the first three months of follow-up. Thus, in asymptomatic patients with good vision, I defer treatment (as no clinical trial has addressed this group of patients), but I follow them carefully and treat immediately if vision or edema worsens.
In CRVO, the decision to treat is relatively easier, as the natural history of CRVO is generally worse than that of BRVO. The CVOS showed in over three years of followup that eyes with 20/40 or better vision at baseline had a 65% chance of maintaining that vision at follow-up, but eyes with worse vision at baseline were more likely to have poor vision at the last follow-up.3
Similarly, a recent meta-analysis of the natural history data from the literature on CRVO found that untreated eyes generally have poor vision outcomes and that few eyes have >20/40 vision untreated.10 In the SCORE-CRVO study, only 7% of the observation group gained more than three lines of vision, while 48% lost more than three lines at two years.11 Additionally, the CRUISE trial of ranibizumab for macular edema due to CRVO showed only 17% threeline gainers in the sham control arm,8 and in the dexamethasone implant trial, only 12% of sham-treated CRVO eyes gained three or more lines.12,13
Given that intravitreal agents were the first treatments to show a benefit to vision in CRVO, I believe that treatment should be given without delay in eyes with worse than 20/40 vision but could be deferred with close followup in nonischemic eyes with mild CME and 20/40 or better vision.
Once the decision to treat has been made, the choice of drug must be made. The purpose of this article is not to review the results of the various clinical trials of triamcinolone (SCORE6,11), dexamethasone,12 pegaptanib sodium,14 ranibizumab (BRAVO,7,9 CRUISE8) or VEGF Trap-Eye (COPERNICUS,15 GALILEO16) in the treatment of RVO. These trials have all demonstrated the benefit of treatment over control, except for the SCORE-BRVO study, which showed no benefit of triamcinolone to grid laser.
It is important to remember that a valid comparison of vision results from these various clinical trials is impossible because of differences in enrollment criteria, baseline attributes of the patients enrolled, and different treatment and retreatment criteria, follow-up intervals and primary endpoints.
Bressler and Schachat suggested that a reasonable way to compare disparate trials is to look at similar outcomes, such as percentages of eyes losing or gaining three lines of vision at the clinically relevant endpoint for each trial. In their review, they noted that, in the CRVO trials, a greater percentage of patients in CRUISE showed improvement of three lines than in the SCORE or Ozurdex trials, and similarly, a smaller percentage of treated patients lost three or more lines of vision. They also suggested a concept of the number of eyes needed to treat to achieve one additional CRVO patient to gain three or more lines of vision, and they found ranibizumab to have a smaller number of eyes needed to treat for CRVO.17
These findings have led me to conclude that anti-VEGF agents should be the first choice for RVO treatment, as the vision results with ranibizumab, pegaptanib and VEGF Trap-Eye are highly significant, without the attendant side effects of cataract and glaucoma. In pseudophakic eyes in patients with no history of glaucoma, either triamcinolone or a dexamethasone implant could be considered if the goal is to minimize the treatment burden by reducing the number of injections and office visits required to control macular edema due to RVO.
While there was no demonstrable advantage of triamcinolone over laser in SCORE-BRVO, there was a considerable improvement in vision gained in BRAVO relative to the sham injection plus deferred laser group. However, BRAVO did not directly compare laser to ranibizumab given at baseline, so you cannot definitely state that the long-term outcomes of laser for BRVO are inferior to ranibizumab.17
Which Anti-VEGF Agent?
The question of which anti-VEGF agent to use for macular edema due to RVO is currently similar to the situation in which we found ourselves before the CATT trial results in AMD. Ranibizumab has received FDA approval for the treatment of macular edema due to both BRVO and CRVO, and as such, it is the only available FDA-approved therapy. Pegaptanib has been shown to have positive effects on vision and macular edema in CRVO but is not FDA-approved and has not seen widespread use off-label, possibly because of negative perceptions from its relative ineffectiveness in AMD and its cost relative to bevacizumab. VEGF Trap-Eye has shown excellent results but is still awaiting regulatory approval and is unavailable for clinical use.
Bevacizumab obviously has been shown in multiple case series to have a beneficial effect on macular edema and vision in both BRVO and CRVO, but it has not been subjected to a randomized clinical trial comparing it to laser in BRVO or to placebo in CRVO. It remains to be seen whether a comparative trial of ranibizumab and bevacizumab for macular edema in RVO will ever be conducted, but such a trial may be unlikely. Given the positive clinical experience of most retinal specialists worldwide and the substantial cost savings involved, bevacizumab is widely used as primary therapy for macular edema due to RVO.
However, we should not make the mistake of applying the CATT results to the management of RVO, as AMD and RVO are very different pathologies, and the effects of the two drugs may not be equal in RVO. In my practice, we typically use bevacizumab as first-line therapy due to cost constraints from a state sales tax on physician-administered drugs. If cost is not an issue, however, then I will generally opt for ranibizumab.
TREAT FOR HOW LONG?
Last but not least is to consider for how long treatment should be given. Again, in the BRAVO and CRUISE studies, treatment was given monthly for six months, so we find ourselves in a similar dilemma as following the MARINA and ANCHOR trials in AMD. Could as-needed treatment be as effective as monthly therapy?
In clinical practice, I typically give a monthly injection until the edema resolves or until two consecutive injections provide no further reduction in edema or improvement in vision. At that point, depending on patient preference, I will adopt either a treat and extend strategy or will stop treatment and examine monthly for recurrence of edema and loss of vision.
The recently published 12-month BRAVO data suggest that periodic retreatment with ranibizumab is required to maintain vision gains and to treat recurrent macular edema,9 so my expectation is that periodic anti-VEGF therapy will be required. A series of eyes with BRVO treated with pegaptanib showed a similar need for periodic retreatment to control recurrent macular edema over one year.18
The addition of grid laser to anti-VEGF therapy for macular edema due to BRVO or CRVO is also attractive but has not been validated by data from clinical trials. Indeed, in diabetic macular edema, the Diabetic Retinopathy Clinical Research Network has published data to suggest that the combination of laser and ranibizumab does not provide better vision results or reduce the number of ranibizumab injections needed, compared to ranibizumab alone.19
This result may not be directly applicable to macular edema from RVO, but my personal experience with combining laser with bevacizumab for macular edema due to BRVO has been disappointing. However, combining laser with triamcinolone has been shown to be equivalent to ranibizumab alone in a subset of pseudophakic eyes with DME, so this combination may also be effective in BRVO or CRVO.
In summary, the era of pharmacologic therapy has given the retinal specialist more effective therapies for macular edema due to RVO, but the wealth of data from these seminal treatment trials has raised many more questions regarding the optimal timing, choice of drug and duration of therapy for macular edema due to RVO. I have suggested some management guidelines based on our current knowledge, but future research will likely further refine our approach to eyes with RVO. RP
1. Branch Vein Occlusion Study Group. Argon laser photocoagulation for macular edema in branch vein occlusion. Am J Ophthalmol. 1984;98:271-282.
2. Central Vein Occlusion Study Group. Evaluation of grid pattern photocoagulation for macular edema in central vein occlusion. The Central Vein Occlusion Study Group M Report. Ophthalmology. 1995;102:1425-1433.
3. Central Vein Occlusion Study Group. Natural history and clinical management of central retinal vein occlusion. Arch Ophthalmol. 1997;115:486-491.
4. Decroos FC, Fekrat S. The natural history of retinal vein occlusion: what do we really know? Am J Ophthalmol. 2011;151:739-741.
5. Rogers SL, McIntosh RL, Lim L. Natural history of branch retinal vein occlusion: an evidence-based systematic review. Ophthalmology. 2010;117:1094-1101.
6. SCORE Study Research Group. A randomized trial comparing the safety and efficacy of intravitreal triamcinolone with standard care to treat vision loss associated with macular edema secondary to branch retinal vein occlusion: The Standard Care vs Corticosteroid for Retinal Vein Occlusion (SCORE) study report 6. Arch Ophthalmol. 2009;127:1101-1114.
7. Campochairo 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;117:1101-1112.
8. Brown DM, Campochiaro PA, Singh RP, et al. Ranibizumab for macular edema following central retinal vein occlusion: six-month primary end point results of a phase III study. Ophthalmology. 2010;117:1124-1133.
9. Brown DM, Campochiaro PA, Bhisitikul RB, et al. Sustained benefits from ranibizumab for macular edema followingbranch retinal vein occlusion: 12-month outcomes of a phase III study. Ophthalmology. 2011;118:1594-1602.
10. McIntosh RL, Rogers SL, Lim L, et al. Natural history of central retinal vein occlusion: an evidence-based systematic review. Ophthalmology. 2010; 117:1113-1123.
11. Ip MS, Scott IU, Van Veldhuisen PC, et al. A randomized trial comparing the safety and efficacy of intravitreal triamcinolone with observation to treat vision loss associated with macular edema secondary to central retinal vein occlusion: the Standard Care vs Corticosteroid for Retinal Vein Occlusion (SCORE) study report 5. Arch Ophthalmol. 2009;127:1101-1114.
12. Haller JA, Bandello F, Belfort R, et al. for the OZURDEX GENEVA Study Group. Randomized, sham-controlled trial of dexamethasone intravitreal implant in patients with macular edema due to retinal vein occlusion. Opthalmology. 2010;117:1134-1146.
13. Ehlers JP, Fekrat S. Retinal vein occlusion: beyond the acute event. Surv Ophthalmol. 2011;56:281-299.
14. Wroblewski JJ, Wells JA, Adamis AP, et al. Pegaptanib in Central Retinal Vein Occlusion Study Group. Pegaptanib sodium for macular edema secondary to central retinal vein occlusion. Arch Ophthalmol. 2009;127:374-380.
15. COPERNICUS trial. Clinicaltrials.gov Web site. http://clinicaltrials.gov/ct2/show/NCT01012973. Accessed October 31, 2011.
16. GALILEO trial. Clinicaltrials.gov Web site. http://clinicaltrials.gov/ct2/show/NCT00943072. Accessed October 31, 2011.
17. Bressler NM, Schachat AP. Management of macular edema from retinal vein occlusions: you can never have too many choices. Ophthalmology. 2010;117:1061-1063.
18. Wroblewski JJ, Wells JA, Gonzales CR. Pegaptanib sodium for macular edema secondary to branch retinal vein occlusion. Am J Ophthalmol. 2010;149:147-154.
19. Diabetic Retinopathy Clinical Research Network. Randomized trial evaluating ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema. Ophthalmology. 2010;117:1064-1077.e35.
|John A. Wells, III, MD, is a partner with the Palmetto Retina Center in Columbia, SC, and clinical associate professor of ophthalmology at the University of South Carolina in Columbia. Dr. Wells has received consulting fees, travel reimbursement and honoraria from Pfizer and OSI. Dr. Wells can be reached at (803) 931-0077.|