V for DME

Diabetic macular edema (DME) is a major cause of visual impairment. It is currently treated with anti-VEGF therapy – but it might not stay that way for much longer. As private insurers and national health care systems struggle to manage the escalating costs associated with monthly anti-VEGF therapy, it is obvious that a long-lasting, reasonably priced treatment for DME is needed – and I think vitrectomy could be the answer.

Vitrectomy removes traction, eliminates vasoproliferative factors, and improves oxygenation. But despite several retrospective and prospective studies showing that vitrectomy significantly decreases DME, it is not yet supported by good clinical evidence. Many of the existing studies lack consistent enrollment criteria, control groups, and standardized measurements of visual acuity (VA), as well as the use of more advanced OCT technology.

The DRCR.net study is one such example; it showed that vitrectomy effectively decreases DME, but only eyes with vitreomacular traction experienced an average of 3 letters improvement in VA (1). The trial, however, only enrolled “eyes that in the estimation of the investigator would not benefit from any other therapy.” In other words, the use of vitrectomy was a last resource. Another study by the DRCR.net included eyes without vitreomacular traction, and found that average central subfield thickness improved from 412 µm to 278 µm at 6 months but median VA remained unchanged at 20/80 (2). Greater improvements in VA occurred in eyes with worse initial VA and epiretinal membranes. Older trials, including those performed by the DRCR.net, have used time domain (TD)-OCT to evaluate the macula; however, though TD-OCT is proficient at accurately and reproducibly measuring macular thickness, it is incapable of evaluating outer retinal morphology.

But things are looking up. More recent trials have attempted to correlate the integrity of the external limiting membrane (ELM) and IS/OS lines, as visualized by spectral domain (SD)-OCT, with visual improvement after vitrectomy. Eyes with significant pre-vitrectomy defects in the ELM and/or IS/OS often experience favorable resolution of macular edema but do not actually achieve improved VA. In a retrospective analysis of eyes that underwent vitrectomy for DME, those with intact IS/OS lines improved by an average of 13.3 letters whereas those with IS/OS defects improved by only 3.9 letters (3). Therefore, outer segment findings on SD-OCT may be used to select patients who would be expected to do well visually after vitrectomy. In another study by Adelman et al., retina specialists from 29 countries provided clinical information on 2,603 patients with macular edema, including 870 patients with DME (4). They found that treatment with vitrectomy and ILM peeling alone resulted in better visual improvement when compared with other therapies.

In my view, vitrectomy for DME has not been adequately studied in eyes that still have the potential for visual improvement. To truly understand the potential of vitrectomy as a therapy for DME, a multi-center, prospective trial on preoperative SD-OCT is needed. If vitrectomy were shown to be as effective as ranibizumab or aflibercept therapy, it would offer economic advantages too, as expenditure associated with vitrectomy management of DME are generally “front loaded” – meaning two-year costs may be only 1/10 of those incurred by anti-VEGF therapy. But we can only enjoy these benefits with the appropriate clinical research – and I look forward to seeing what might be revealed in the next few years.