More than 25 years ago, the Branch Vein Occlusion Study (BVOS) reported the usefulness of grid pattern laser photocoagulation for perfused macular edema in eyes with branch retinal vein occlusion (BRVO) when compared with its natural history. The Central Retinal Vein Occlusion Study reported the natural history of central retinal vein occlusion (CRVO) approximately 1 decade later. These trials were considered the gold standard for natural history information for many years, despite their limitations, and still are quoted frequently. The BVOS reported that visual acuity improved a mean of 0.23 lines over 3 years in untreated eyes. However, the BVOS natural history data involved only a small number of subjects, with 35 untreated eyes at 3 years, and different BRVO durations from 3 to 18 months were aggregated and then natural history determined. Moreover, the BVOS did not provide any natural history data from onset of BRVO until 3 months because study enrollment required at least a 3-month duration of BRVO. Because only patients with BRVOs with perfused edema without foveal hemorrhage were enrolled, no natural history information about eyes with ischemic edema or foveal hemorrhage can be ascertained from the BVOS.

The Central Retinal Vein Occlusion Study provided some useful natural history information on a smaller subset of 187 eyes with CRVO duration of 1 month or less; however, additional natural history information was determined from the larger cohort of eyes with CRVO of variable durations lumped together. For example, the Central Retinal Vein Occlusion Study stated that 81 of 547 eyes with perfused CRVO became ischemic during the first 4 months of study follow-up, not during the first 4 months after CRVO onset. Numerous unanswered natural history questions remained for eyes with BRVO and CRVO after completion of these landmark trials.

In recent years, however, other multicenter clinical trials for both BRVO and CRVO have evaluated a variety of novel interventions—intravitreal vascular endothelial growth factor inhibition, intravitreal steroid injection, and sustained drug delivery. The expansion of these therapeutic interventions emphasizes the need to understand better the natural history of these retinal vascular diseases. However, some of the same aforementioned limitations also limit the natural history information that can be extracted from the more recent multicenter clinical trials.

Because the BVOS determined grid pattern laser photocoagulation to be the standard of care for BRVO eyes with perfused edema, no natural history information for such eyes can be ascertained from the Standard of Care vs. Corticosteroid for Retinal Vein Occlusion (SCORE) study, because all eyes in the standard of care group received grid pattern laser either at study entry or when there was sufficient reabsorption of intraretinal hemorrhage to permit laser administration. However, Genentech’s Branch RetinAl Vein Occlusion (BRAVO) study evaluating intravitreal ranibizumab for BRVO did not allow for grid pattern laser in the sham group until at least 3 months after study entry. Based on this, some may argue that the natural history information from onset of BRVO to 3 months’ duration can be ascertained from BRAVO’s sham group, and as such, approximately 17% of 132 eyes in the sham group demonstrated a visual gain of 3 lines or more 3 months after study entry. However, 13% in the sham group had a hemiretinal vein occlusion; the natural history of hemiretinal vein occlusion indeed may be different from that of BRVO. Also, eyes with different BRVO durations (range, 0 to 16 months) were lumped together at study entry and at best can be used to estimate natural history for the first 3 months after study entry, and not for the first 3 months after BRVO onset. The natural history of eyes with a BRVO duration of 3 months is likely different from the natural history of eyes with a BRVO duration of 19 months.

In the OZURDEX GENEVA (Global Evaluation of implaNtable DexamEthasone in retinal Vein occlusion with macular edemA) study evaluating the dexamethasone intravitreal implant for retinal vein occlusion (RVO), the sham group comprised 426 eyes with both BRVO and CRVO. Of all 426 eyes, 9% had laser before study entry, so it is not a true sham group from which to determine natural history. Moreover, approximately 85% had the RVO for 3 months or more before study entry, whereas approximately 33% had the RVO for 6 months or longer before study entry. Again, eyes with different RVO durations were grouped together to determine the natural history after study entry, but not after RVO onset. It would be more useful to know what the natural history is after RVO onset or some other consistent interval. Familiarity with the natural course at its various stages would enable improved patient counseling regarding prognosis and expectation.

Although a post hoc subgroup analysis was performed based on the duration of macular edema at baseline in the OZURDEX study, this analysis was performed on both BRVO and CRVO together. These are very different retinal vascular diseases, and separate analysis for the duration of macular edema from baseline would yield more meaningful information. Subgroup analysis to separately evaluate the visual outcomes in the sham group of solely the BRVO eyes or solely the CRVO eyes was performed. Within these subgroups, different disease durations were mixed together, and 20% of BRVO eyes in the sham group gained 3 lines at month 6, whereas 12% of CRVO eyes in the sham group gained 3 lines at month 6. Eyes with visual acuities of 20/200 were entered into this study, and some of the BRVO eyes with 20/200 acuities likely had ischemic edema, and thus may have a very different natural history than those eyes with perfused edema. The perfusion status of the macular edema for the BRVO subgroup was not evaluated, yet it is likely an influential factor in natural history. Such differentiation is necessary in a well-designed natural history study for eyes with BRVO.

For eyes with CRVO, the SCORE study reported a 15-letter or more loss in approximately 44% of 73 eyes at month 12. However, the duration of macular edema at study entry varied from less than 3 months in 33% to more than 6 months in almost 20%. All eyes with CRVO were grouped together, despite this wide baseline disease duration. Letters lost at 12 months were reported from the aggregate group of CRVO patients. Because some eyes received an intravitreal triamcinolone injection before study entry and 9% of eyes in this observation group received an intravitreal triamcinolone injection during the first 12 months, extrapolating natural history from the control group is limited. Of the initial 88 patients who started the SCORE study in the observation group, only 46 remained at 24 months, and 25 remained at 36 months. Note that CRVO eyes with 20/400 vision were included in SCORE, and those with an afferent pupillary defect were not excluded, so it is likely that some eyes with ischemic CRVO are included in the SCORE observation group’s natural history.

Genentech’s Central Retinal vein occlUsIon (CRUISE) study, which evaluated intravitreal ranibizumab for eyes with CRVO, had a sham group in which 17% of 130 eyes gained 3 lines or more 6 months after study entry. The CRUISE study may have more stringently selected for perfused CRVO by excluding cases with a brisk afferent pupillary defect, so one could extrapolate to say perhaps this is the natural history of perfused CRVO. Likewise, one cannot conclude much about the natural history of ischemic CRVO from the CRUISE data set. Subgroup analysis of CRVO eyes with a duration of 3 months or fewer demonstrated a mean gain of 1.1 lines at month 6 compared with baseline, whereas eyes with a duration 3 months or more demonstrated a mean gain of 0.4 lines. This provides some approximate natural history data to better guide our patients with perfused CRVO better.

The natural history of RVO has never been prospectively studied specifically in a rigorous scientific manner, but instead has been a byproduct of clinical trials evaluating various therapeutic interventions. So although many data are available, as a profession, we still lack comprehensive knowledge about the natural history of RVO. We cannot tell the patient with a CRVO of 5 months’ duration and visual acuity of 20/60 what the likelihood is that their vision can improve without treatment over the ensuing 6 months and rigorously substantiate our answer with trial data. Nevertheless, for persons with RVO, an abundance of recent treatment options offers hope for improved vision and reduced complications. The effectiveness of these therapies, imperfect although they may be, makes it improbable that we will ever have a comprehensive understanding of the natural history of RVO. A lesson to be learned from all of this, however, is to conduct natural history studies now on diseases that currently do not have effective treatments, so that when effective therapies become available, we will have robust natural history data from which to compare and thus advise our patients.