Christopher J. Brady

Clinical management of proliferative vitreoretinopathy: an update.

Background: Proliferative vitreoretinopathy (PVR) remains the most significant obstacle to successful retinal reattachment surgery. Preclinical studies continue to add insights into the complex molecular events leading to PVR development, helping to identify new targets for potential prophylactic or therapeutic agents. This article reviews the recent evidence supporting surgical and medical treatments for PVR. Methods: PUBMED was used for literature search. Clinical studies regarding surgical management of PVR from January 1, 2000 to August 1, 2014 were included. Clinical studies regarding medical management of PVR from January 1, 2000 to August 1, 2014 were included if the design of study was a randomized controlled trial. Results: Many recent studies have evaluated surgical and medical strategies for the treatment and prevention of PVR. Newer vitreoretinal surgery technology (23- and 25-gauge vitrectomy) and tamponade agents (heavy silicone oils) have been studied. Medical therapies evaluated include antiinflammatory agents, low molecular weight heparin, 5-fluorouracil, 13-cis-retinoic acid, and daunorubicin, amongst others. Conclusion: Surgical management with pars plana vitrectomy, with or without scleral buckle or inferior retinectomy, remains an effective treatment for PVR-related detachments. Consensus regarding a preferred surgical strategy remains controversial. Many medical therapies have been studied but fail to demonstrate a statistically significant benefit in clinical trials. Further studies to clarify the efficacy of available and novel treatment options are warranted.

Rapid Grading of Fundus Photographs for Diabetic Retinopathy Using Crowdsourcing

Background Screening for diabetic retinopathy is both effective and cost-effective, but rates of screening compliance remain suboptimal. As screening improves, new methods to deal with screening data may help reduce the human resource needs. Crowdsourcing has been used in many contexts to harness distributed human intelligence for the completion of small tasks including image categorization. Objective Our goal was to develop and validate a novel method for fundus photograph grading. Methods An interface for fundus photo classification was developed for the Amazon Mechanical Turk crowdsourcing platform. We posted 19 expert-graded images for grading by Turkers, with 10 repetitions per photo for an initial proof-of-concept (Phase I). Turkers were paid US

A randomized clinical trial to evaluate ready-made spectacles in an adult population in India

0.10 per image. In Phase II, one prototypical image from each of the four grading categories received 500 unique Turker interpretations. Fifty draws of 1-50 Turkers were then used to estimate the variance in accuracy derived from randomly drawn samples of increasing crowd size to determine the minimum number of Turkers needed to produce valid results. In Phase III, the interface was modified to attempt to improve Turker grading. Results Across 230 grading instances in the normal versus abnormal arm of Phase I, 187 images (81.3%) were correctly classified by Turkers. Average time to grade each image was 25 seconds, including time to review training images. With the addition of grading categories, time to grade each image increased and percentage of images graded correctly decreased. In Phase II, area under the curve (AUC) of the receiver-operator characteristic (ROC) indicated that sensitivity and specificity were maximized after 7 graders for ratings of normal versus abnormal (AUC=0.98) but was significantly reduced (AUC=0.63) when Turkers were asked to specify the level of severity. With improvements to the interface in Phase III, correctly classified images by the mean Turker grade in four-category grading increased to a maximum of 52.6% (10/19 images) from 26.3% (5/19 images). Throughout all trials, 100% sensitivity for normal versus abnormal was maintained. Conclusions With minimal training, the Amazon Mechanical Turk workforce can rapidly and correctly categorize fundus photos of diabetic patients as normal or abnormal, though further refinement of the methodology is needed to improve Turker ratings of the degree of retinopathy. Images were interpreted for a total cost of US

Changes in Retinal Nonperfusion Associated with Suppression of Vascular Endothelial Growth Factor in Retinal Vein Occlusion

1.10 per eye. Crowdsourcing may offer a novel and inexpensive means to reduce the skilled grader burden and increase screening for diabetic retinopathy.

Visual function after correction of distance refractive error with ready-made and custom spectacles: A randomized clinical trial

BACKGROUND Ready-made spectacles (RMS) have advantages; however, visual performance and satisfaction has not been evaluated. METHODS A 1-month, double-masked, randomized clinical trial comparing planned continued use and visual performance of RMS to Custom Spectacles (CS) in adults aged 18-45 years with > or =1 diopter (D) of uncorrected refractive error (URE). RESULTS A total of 373 of 400 participants (93%) completed; mean age was 30 +/- 9 years, and 58% were female. Average URE was 2.21 +/- 1.31D and habitual vision was 0.58 +/- 0.21 logMAR (logarithm of Minimum Angle of Resolution, 20/63(+1) Snellen acuity). Ten participants with habitual vision better than 20/40 were excluded (3%). A lower proportion in the RMS group intended to continue to wear the study spectacles after 1 month (165/183, 90% vs 174/180, 97%, P = 0.02). Spectacle vision in the eye with lower URE was 0.08 +/- 0.15 vs 0.02 +/- 0.08, P < 0.0001 and higher URE was 0.12 +/- 0.18 vs 0.02 +/- 0.08, P < 0.0001 (logMAR) for RMS and CS. Subgroup analyses excluding participants with astigmatism > or =2.00 D and anisometropia > or =1.00 D (74/363, 20%) found no difference in planned continued use (139/143, 97% vs 141/146, 97%, P = 1.0) for RMS vs CS. CONCLUSIONS While vision is slightly better with CS, 90% of an adult population with URE planned to continue to use their RMS at 1 month. Furthermore, if those without high astigmatism or anisometropia are excluded, virtually all are satisfied with RMS and there is no difference when compared with CS. The findings of this study support the use of RMS for the delivery of refractive services in settings where there is a high level of need, limited resources and low access to refractive services.

Tele-ophthalmology Screening for Proliferative Diabetic Retinopathy in Urban Primary Care Offices: An Economic Analysis

PURPOSE To assess changes in retinal nonperfusion (RNP) in patients with retinal vein occlusion (RVO) treated with ranibizumab. DESIGN Secondary outcome measure in randomized double-masked controlled clinical trial. PARTICIPANTS Thirty-nine patients with central RVO (CRVO) and 42 with branch RVO (BRVO). METHODS Subjects were randomized to 0.5 or 2.0 mg ranibizumab every month for 6 months and then were re-randomized to pro re nata (PRN) groups receiving either ranibizumab plus scatter laser photocoagulation or ranibizumab alone for an additional 30 months. MAIN OUTCOME MEASURES Comparison of percentage of patients with increased or decreased area of RNP in patients with RVO treated with 0.5 versus 2.0 mg ranibizumab, during monthly injections versus ranibizumab PRN, and in patients treated with ranibizumab PRN versus ranibizumab PRN plus laser. RESULTS In RVO patients given monthly injections of 0.5 or 2.0 mg ranibizumab for 6 months, there was no significant difference in the percentage who showed reduction or increase in the area of RNP. However, regardless of dose, during the 6-month period of monthly injections, a higher percentage of patients showed a reduction in area of RNP and a lower percentage showed an increase in area of RNP compared with subsequent periods of ranibizumab PRN treatment. After the 6-month period of monthly injections, BRVO patients, but not CRVO patients, randomized to ranibizumab PRN plus laser showed significantly less progression of RNP compared with patients treated with ranibizumab PRN. CONCLUSIONS Regardless of dose (0.5 or 2.0 mg), monthly ranibizumab injections promote improvement and reduce progression of RNP compared with PRN injections. The addition of scatter photocoagulation to ranibizumab PRN may reduce progression of RNP in patients with BRVO, but a statistically significant reduction was not seen in patients with CRVO.

Microbial spectrum and outcomes of endophthalmitis after intravitreal injection versus pars plana vitrectomy

PURPOSE To evaluate patient-reported outcome measures with the use of ready-made spectacles (RMS) and custom spectacles (CS) in an adult population in India with uncorrected refractive error (URE). DESIGN Prospective, double-masked, randomized trial with 1-month follow-up. PARTICIPANTS A total of 363 adults aged 18 to 45 years with ≥1 diopter (D) of URE (RMS, n = 183; CS, n = 180). INTERVENTION All participants received complete refraction and were randomized to receive CS (full sphero-cylindrical correction) or RMS based on the spherical equivalent for the eye with lower refractive error but limited to the powers in the RMS inventory. MAIN OUTCOME MEASURES Visual function and quality of life (VFQoL) instrument and participant satisfaction. RESULTS Rasch scores for VFQoL increased from 1.14 to 4.37 logits in the RMS group and from 1.11 to 4.72 logits in the CS group: respective mean changes of 3.23 (95% confidence interval [CI], 2.90-3.56) vs. 3.61 (95% CI, 3.34-3.88), respectively. Mean patient satisfaction also increased by 1.83 points (95% CI, 1.60-2.06) on a 5-point Likert scale in the RMS group and by 2.04 points (95% CI, 1.83-2.24) in the CS group. In bivariate analyses, CS was not associated with increased VFQoL or patient satisfaction compared with the RMS group. In the full multivariable linear regression, the CS group had greater improvement when compared with those receiving RMS (+0.45 logits; 95% CI, 0.02-0.88), and subjects with astigmatism >2.00 D had significantly less improvement (-0.99 logits; 95% CI, -1.68 to -0.30) after controlling for demographic and vision-related characteristics. In multivariable analysis, increased change in patient satisfaction was related to demographic and optical characteristics, but not spectacle group. CONCLUSIONS Ready-made spectacles produce large but slightly smaller improvements in VFQoL and similar satisfaction with vision at 1-month follow-up when compared with CS. Ready-made spectacles are suitable for the majority of individuals with URE in our study population, although those with high degrees of astigmatism may benefit from a trial of CS. This study provides further evidence for the use of RMS in settings where CS are unavailable or unaffordable, or refractive services are inaccessible to those in need.

Quality and diagnostic utility of mydriatic smartphone photography: The smartphone ophthalmoscopy reliability trial

BACKGROUND AND OBJECTIVE To determine whether tele-ophthalmology screening for proliferative diabetic retinopathy (PDR) can be cost-saving. PATIENTS AND METHODS Adults with diabetes presenting for routine medical care underwent non-mydriatic fundus photography with remote grading. Direct medical costs were estimated using the Medicare fee schedule in the base case, with Medicaid and commercial insurance rates used for low and high values, respectively. One-way and probabilistic sensitivity analyses were performed. RESULTS Of 99 participants, at least mild retinopathy was found in 24 (24.2%). Urgent consultation was recommended for eight participants (8.1%) for possible vision-threatening diabetic retinopathy, including two participants (three eyes) with PDR. In the base case, screening saved

Crowdsourcing: an overview and applications to ophthalmology

36 per patient. A Monte Carlo simulation indicated that screening saved a median of

Crowdsourcing to Evaluate Fundus Photographs for the Presence of Glaucoma

48 per patient. CONCLUSION A substantial burden of diabetic retinopathy was identified, most of which was undiagnosed. In a closed system, tele-ophthalmology screening for PDR is likely to be cost-saving across the range of scenarios explored.