Journal Club

Recent Noteworthy Studies to Stimulate Discussion and Debate

■ Benefits and costs. Everyone knows that ranibizumab costs more than bevacizumab, and the vast majority of the people receiving either drug in the United States are Medicare recipients. But just how much more is Medicare paying for ranibizumab, given retinal physicians' ability to administer either? A team from the Bascom Palmer Eye Institute (BPEI) sought to answer that question in the May 2011 issue of the American Journal of Ophthalmology.

The BPEI team, which included Philip Rosenfeld, MD, whose studies of bevacizumab for wet AMD in the mid-2000s paved the way for its regular use, conducted a retrospective review of over 200,000 Medicare Part B claims for wet AMD, limiting their analysis to a single year, 2008.

Despite 64.4% of the 222,866 beneficiaries' claims studied having been filed for bevacizumab, expenditures for ranibizumab constituted a larger amount spent by Medicare than bevacizumab ($536 million vs $20 million). The other chief outcome measurement, number of injections administered on a state-by-state basis, found that in 39 out of 50 states, rates of injections for bevacizumab were higher.

This is the first study on reimbursement data for competing anti-VEGF drugs since the ASRS's PAT survey addressed the issue. The article was in press when the CATT trial results were announced, so the BPEI could not augment their cost data with safety or efficacy data.

■ Can VEGF blockade hurt the retina? Anti-VEGF drugs for wet AMD have been in use for only six years, so we still don't know the long-term ocular effects of blockad ing VEGF. In the May 2011 Archives of Ophthalmology, a study team from Washington University in St. Louis examined this issue using pattern electroretinography.

Seventeen eyes of 17 treatment-naïve patients were prospectively enrolled in treatment with ranibizumab, which was administered monthly for six months. The authors performed electroretinograph imaging at baseline and at one, three and six months after the first treatment.

At the six-month mark, there were no decreases in P50 and N95 amplitudes in the eyes, and the trial had a mean improvement in visual acuity, typical for ranibizumab, from 20/85 at baseline to 20/55 at month six.

Although this study was limited by its small size and lack of a control group, the authors nevertheless are confident in stating that ranibizumab used over a six-month period does not damage the retina or retinal ganglion cells, at least as evaluated by electroretinography.

■ Small-gauge vitrectomy study. To determine whether 23-gauge vitrectomy has a favorable rate of postoperative retinal detachment when compared to standard 20-g vitrectomy, a surgical team in France conducted a retrospective study of nearly 700 eyes. They report their findings in the May 2011 issue of Retina.

To compile the cases, 349 eyes operated on from June 2007 to December 2008 with 23-g vitrectomy were compared to 346 eyes operated on between October 2003 and September 2007 with 20-g vitrectomy. Six-month follow-up data were compared between the groups, as were longer-term follow-up data as available. Furthermore, rates of retinal detachment relative to the retina disorder being treated surgically were analyzed.

At six months, the cumulative probability of retinal detachment following 23-g vitrectomy was 1.1%, as compared to 3.5% for 20-g vitrectomy. At the median-length follow-up of 14 months, the rate in the 20-g rose to 4.9%, while the rate in the 23-g group remained the same. Furthermore, detachments were observed most commonly following macular hole repair (7.3%), followed by vitreo-macular tractions surgery (4.3%) and epiretinal membrane surgery (2.3%).

The study authors offer six possible explanations for why 23-g had a demonstrably lower rate of postop retinal detachment. These explanations include the use of trocars and cannulas in 23-g vitrectomy and smaller-sized sclerotomies. The authors are also careful to note that, due to their study's limited follow-up, the true rates of postvitrectomy retinal detachment may be higher and the better results with 23-g victrectomy not borne out over time.

■ Sirolimus for DME. Retinal physicians at the National Eye Institute report online in Graefe's Archive of Clinical and Experimental Ophthalmology the one-year results from a trial of sirolimus in diabetic macular edema. In this phase 1/2 study, sirolimus, which has already undergone extensive testing in AMD, appeared safe in eyes with DME.

The NEI team gave five patients with DME involving the fovea a 20-µL subconjunctival injection of sirolimus and then every two months, unless the edema and/or sub-retinal leakage resolved, as measured with optical coherence tomography and fluorescein angiography, respectively.

While, as noted, sirolimus injections appeared safe and resulted in no serious adverse events, only redness and swelling at the injection site, they also did not result in any significant improvement in visual acuity or, indeed, any consistent treatment effect at all. This trial adds to the growing amount of data suggesting that sirolimus may not be effective in treating macular disease. However, the authors stress multiple investigations of immunosuppressive drugs in macular disease are under way and it remains a promising avenue of research. RP