Eyeworld

EW RESIDENTS 70 February 2017 by Ashlie Bernhisel, MD, Christopher Conrady, MD, PhD, Tara Hahn, MD, Sravanthi Vegunta, MD, and Jeff Pettey, MD, John A. Moran Eye Center, University of Utah School of Medicine single-surgeon study to determine a standardized method of diagnosing csCME in postoperative pseudopha- kic diabetic patients. Patients with a history of intraocular surgery, retinal or choroidal disease, uveitis, prolif- erative retinopathy, laser photoco- agulation 3 months prior to surgery, and DME or abnormal retinal thick- ening on preoperative OCT 1 month prior to surgery were excluded. All cataracts were removed by routine phacoemulsification. Risk factor data was collected: insulin dependence, DM duration, HgbA1C, and cumu- lative dissipated energy and ultra- sound time during phacoemulsifica- tion. Patients followed up at 1 week, 1 month, 3 months, and 6 months, at which times OCT for central FT, microperimetry for macular sensitiv- ity (MS), and BCVA were measured. Postoperatively all patients followed the same treatment regimen of 1% prednisolone acetate, nonsteroi- dal anti-inflammatory drug, and antibiotic. Statistical analysis was performed using either two-sided student t-test or ANOVA. Multiple group comparisons were corrected by Bonferroni post hoc tests, and correlation analysis was performed using distance matrices. These analyses were presented in graph- ic form as heat maps. Risk factors were evaluated using multiple linear regression and Spearman correlation coefficient. Results There were 1,002 participants (511 males, 491 females) included in the final analysis. BCVA and MS improved after cataract surgery; however, FT increased, peaking at 1 month postoperatively on average. Baseline characteristics associated with increased FT included worse di- abetic retinopathy, longer duration of DM, insulin dependence, higher HgbA1C, more severe nuclear sclero- sis, and longer ultrasound time. A subanalysis was then per- formed and found that MS had a better correlation with FT than BCVA postoperatively. Patients with 0% to 39% increase in FT showed minimal decrease in MS and were categorized as subclinical macu- lar edema (ME). Those with ≥40% change had at least a 20% decrease A ccording to the American Academy of Ophthalmol- ogy, more than 2 million cataract surgeries are performed each year in the U.S., and clinically significant cystoid macular edema (csCME) is the most common visual impair- ment following surgery. csCME has historically been diagnosed by identifying loss of the foveal reflex on slit lamp biomicroscopy and petaloid leakage on fluorescein an- giography (FA). Recent advances in imaging have led to the widespread use of optical coherence tomogra- phy (OCT) in lieu of FAs despite an unclear understanding of clinically significant change in foveal thick- ness (FT). Diabetes mellitus (DM) is known to increase the risk of devel- opment and rate of progression of cataracts. Patients with a history of diabetic macular edema (DME) or non-central involving CME are at a higher risk of developing csCME following cataract surgery. 1 Conse- quently, many diabetics are requir- ing cataract surgery and are at high risk of developing csCME. Identifica- tion of high-risk individuals to guide therapeutic intervention is needed. Methods Yang et al. published a retrospec- tive, nonrandomized, single-center, in MS and were defined as csCME (3.2% incidence). Discussion Yang et al. investigated a topic that is pertinent to a large subpopula- tion of patients and both anterior and posterior segment surgeons, and their study has many notable strengths. Sample size was adequate (n=1,002) for a retrospective study evaluating a surgical complica- tion. The authors had well-defined methods for grading cataract density using the Lens Opacities Classifica- tion System III and grading diabetic retinopathy using fundus photos. Grading systems allow for accu- rate and reproducible stratification of risk factors in clinical practice. Additionally, Yang et al. conducted a single-surgeon study, which largely eliminates variability in surgeon technique and skill. As noted by the authors, this study does have important limita- tions. To control risk of confound- ing bias, patients with any history of DME or CME should have been excluded. Diagnosing CME in dia- betics with OCT alone is controver- sial. Studies have shown that OCT cannot reliably distinguish between DME and/or CME, and FA continues to be the gold standard for diagnosis and subtyping of ME. 2–4 The authors argued that the temporal relation- ship between the development of ME after cataract surgery was suggestive of CME rather than DME; however, they risk bias from mis- classification of outcome. While we question whether the ME described was DME versus CME, the two clin- ical entities may be difficult, if not impossible, to distinguish due to a presumably similar pathophysiology. Extracapsular cataract extraction, and specifically phacoemulsifica- tion, is a well-known risk factor for worsening DME, but it is still unclear whether the cause is surgery or simply natural progression of the disease. 4–6 Another factor that could have affected classification and progression of ME is therapy. The authors did not include information regarding interventions or treat- ments the ME patients underwent over the course of 6 months. Thera- py could have affected change in FT and other measured risk factors. Both OCT and microperimetry were employed in unconventional ways in this study. When using OCT to evaluate CME, qualitative find- ings such as paracentral thickening, homogenous versus cystoid changes, and subfoveal macular detachments are considered. Yang et al. evaluat- ed only quantitative change in FT, which may have limited the value of OCT in CME. Although at this time microperimetry is not widely used in clinical practice, the authors developed a threshold MS value for diagnosis of CME, which can be Review of "The risk factors and diagnosis of pseudophakic cystoid macular edema after cataract surgery in diabetic patients" Jeff Pettey, MD, residency program director, John A. Moran Eye Center, University of Utah School of Medicine, Salt Lake City How much of a risk factor is diabetes for CME after uncom- plicated cataract surgery? I asked the Utah residents to review this single-surgeon study of 1,000 patients with no preoperative macular edema who were followed for 6 months. –David F. Chang, MD, EyeWorld journal club editor From left: Christopher Conrady, MD, PhD, Tara Hahn, MD, Sravanthi Vegunta, MD, and Ashlie Bernhisel, MD Source: University of Utah School of Medicine continued on page 72

• Cover