Diabetic Macular Edema

Diabetic Macular Edema Dear Editor: I read with great interest the article by the Diabetic Retinopathy Clinical Research network (DRCR) in the September 2008 issue on the subject of diabetic macular edema (DME).1 The DRCR conducted a multicenter randomized clinical trial on the efficacy of intravitreal 1-mg and 4-mg triamcinolone in comparison with focal/grid photocoagulation for the treatment of DME. Their primary outcome was visual acuity. The network concluded that after 2 years, laser was more effective and safer than 1-mg or 4-mg intravitreal triamcinolone. I have some questions regarding this conclusion. Nearly half (147/330; 45%) of laser-treated eyes had their fellow eye randomized to either 1-mg or 4-mg intravitreal triamcinolone (IVK). While no rational mechanism can explain a contralateral effect of laser photocoagulation, IVK can be detected in the blood2 and bevacizumab has been detected in the fellow eye after injection.3 Given triamcinolone’s long half-life and repeated application in this study, a small clinical effect on the fellow eye cannot be excluded. Fortunately, the DRCR should be able to clarify this concern by comparing the visual and optical coherence tomography (OCT) thickness outcomes of the 75 subjects in whom 1 eye was treated with laser and the fellow eye with 4-mg IVK to the remaining 183 eyes treated with laser in 1 eye only. I look forward to their findings. Almost half of eyes injected with IVK develop visually significant posterior subcapsular cataract (PSC) by 1 year and even subtle PSC cataract can impact vision due to its central location and proximity to the nodal point.4 Fifty-one percent of phakic eyes in the 4-mg IVK group underwent cataract surgery. Despite this, the DRCR contends that corticosteroid-induced cataract changes were unlikely to account for the visual acuity results of IVK eyes because of greater OCT thickening and lack of visual improvement in pseudophakic eyes at baseline. However, mean central subfield thickness was less in the 4-mg IVK group than lasertreated eyes for the first year, and its increase thereafter may have coincided with cataract surgery. We have previously reported that a history of DME conveys a substantial risk of macular thickening after surgery.5 In Figure 1 (available at http://aaojournal.org), 5 eyes with resolved DME and a mean (normal) preoperative center point thickness of 169⫾ 36 ␮m experienced an increase of 156 ␮m to 325⫾148 ␮m at 1 month, which was significantly greater than the remaining 41 diabetic eyes without prior DME (P ⫽ 0.02). Furthermore, in Figure 2 (available at http://aaojournal.org) increased thickening directly translated into decreased visual acuity. Finally, although the DRCR has already established that OCT thickness in eyes with DME may not be a reliable substitute for visual acuity, this relationship was not analyzed in the setting of cataract surgery.6 Despite general acceptance that diabetic eyes with DME at the time of cataract surgery have poorer visual outcomes,7

1228

there is no information provided on the status of the macula before cataract surgery and to what degree, if any, preoperative management was coordinated and instituted. Such information, if available, would be both relevant and informative. Although the DRCR did not observe a visual benefit of pseudophakic (at baseline) IVK eyes over pseudophakic laser treated eyes, their sample size was small. Moreover, pseudophakia is not a random occurrence, and thus the possibility of selection bias exists. In particular, cataract surgery is a known risk factor for DME and progression of diabetic retinopathy and therefore, pseudophakic eyes enrolled in this study may have had different risk characteristics than phakic eyes. Consequently, direct extrapolation of visual results from pseudophakic to phakic eyes may not be valid. In conclusion, cataract formation and subsequent surgery instead of lack of efficacy may explain the decreased visual acuity and increase macular thickness observed in IVK eyes when compared with laser-treated eyes. One may argue, however, that such a distinction is ultimately immaterial since cataract formation is a natural consequence of IVK treatment. In response, I would counter that such a distinction is in fact essential since it would emphasize the importance of perioperative management in these high-risk eyes and motivate investigation into alternative anti-inflammatory agents which do not cause cataracts. STEPHEN JAE KIM, MD Nashville, Tennessee References 1. Diabetic Retinopathy Clinical Research Network. A randomized trial comparing intravitreal triamcinolone acetonide and focal/grid photocoagulation for diabetic macular edema. Ophthalmology 2008;115:1447–59. 2. Degenring RF, Jonas JB. Serum levels of triamcinolone acetonide after intravitreal injection. Am J Ophthalmol 2004; 137:1142–3. 3. Bakri SJ, Snyder MR, Reid JM, et al. Pharmacokinetics of intravitreal bevacizumab (Avastin). Ophthalmology 2007;114: 855–9. 4. Thompson JT. Cataract formation and other complications of intravitreal triamcinolone for macular edema. Am J Ophthalmol 2006;141:629 –37. 5. Kim SJ, Equi R, Bressler NM. Analysis of macular edema after cataract surgery in patients with diabetes using optical coherence tomography. Ophthalmology 2007;114:881–9. 6. Diabetic Retinopathy Clinical Research Network. Relationship between optical coherence tomography-measured central retinal thickness and visual acuity in diabetic macular edema. Ophthalmology 2007;114:525–36. 7. Dowler JG, Sehmi KS, Hykin PG, Hamilton AM. The natural history of macular edema after cataract surgery in diabetes. Ophthalmology 1999;106:663– 8.

Letters to the Editor

Figure 1. Mean center point thickness (CPT) in microns of diabetic eyes with (5 eyes) and without (41 eyes) a prior history of clinically significant macular edema (CSME) preoperatively and at 1 and 3 months after cataract surgery. The difference in thickness at 1 month was statistically different (P ⫽ 0.02). CPT in eyes with treated CSME remained elevated at 3 months (253 ⫾ 105 ␮m).

Figure 2. Mean change in vision (logMar units) from preoperative baseline at 1 and 3 months of diabetic eyes with a history of resolved clinically significant macular edema (CSME) (5 eyes) and eyes without a history of CSME (41 eyes). The difference in visual outcome at 1 month was statistically different (P ⫽ 0.01).

1228.e1