ABSTRACT
BACKGROUND: Diabetic retinopathy is the most common complication of diabetes mellitus and is one of the leading causes of vision impairment globally, which is also relevant for the Russian Federation. OBJECTIVE: To evaluate the diagnostic efficiency of a convolutional neural network trained for the detection of diabetic retinopathy and estimation of its severity in fundus images of the Russian population. METHODS: In this cross-sectional multicenter study, the training data set was obtained from an open source and relabeled by a group of independent retina specialists; the sample size was 60,000 eyes. The test sample was recruited prospectively, 1186 fundus photographs of 593 patients were collected. The reference standard was the result of independent grading of the diabetic retinopathy stage by ophthalmologists. RESULTS: Sensitivity and specificity were 95.0% (95% CI; 90.8-96.4) and 96.8% (95% CI; 95.5- 99.0), respectively; positive predictive value - 98.8% (95% CI; 97.6-99.2); negative predictive value - 87.1% (95% CI, 83.4-96.5); accuracy - 95.9% (95% CI; 93.3-97.1); Kappa score - 0.887 (95% CI; 0.839-0.946); F1score - 0.909 (95% CI; 0.870-0.957); area under the ROC-curve - 95.9% (95% CI; 93.3-97.1). There was no statistically significant difference in diagnostic accuracy between the group with isolated diabetic retinopathy and those with hypertensive retinopathy as a concomitant diagnosis. CONCLUSION: The method for diagnosing DR presented in this article has shown its high accuracy, which is consistent with the existing world analogues, however, this method should prove its clinical efficiency in large multicenter multinational controlled randomized studies, in which the reference diagnostic method would be unified and less subjective than an ophthalmologist.
ABSTRACT
BACKGROUND: Since the efficacy of ranibizumab (RBZ), bevacizumab (BVZ) and aflibercept (AFB) is comparable in neovascular age-related macular degeneration (AMD), we conducted a systematic review and meta-analysis to evaluate the long-term safety profiles of these agents, including ocular safety. METHODS: Systematic review identifying randomized controlled trials (RCTs) comparing RBZ, BVZ and AFB directly published before March 2019. Serious ocular adverse events (SOAE) of special interest were endophthalmitis, pseudo-endophthalmitis, retinal pigment epithelium tear and newly identified macular atrophy. RESULTS: Thirteen RCTs selected for meta-analysis (4952 patients, 8723 people-years follow-up): 10 compared RBZ vs. BVZ and three RBZ vs. AFB. There were no significant differences in almost all adverse events (systemic and ocular) between BVZ, RBZ and AFB in up to two years' follow-up. Macular atrophy was reported heterogeneously and not reported as SOAE in most trials. CONCLUSIONS: Direct comparison of RBZ, BVZ and AFB safety profiles in the RCT network meta-analytical setting have not revealed a consistent benefit of these three commonly used anti-vascular endothelial growth factor (anti-VEGF) agents in AMD. Network model ranking highlighted potential benefits of RBZ in terms of a systemic safety profile; however, this appears a hypothesis rather than a conclusion. Newly identified macular atrophy is underestimated in RCTs-future real-world data should be focused on SOAE.
ABSTRACT
AIM: To study antibiotic resistance patterns and susceptibility to eye antiseptic picloxydine of conjunctival flora in patients undergoing intravitreal injections (IVIs). METHODS: Conjunctival swabs were taken in 4 groups of patients, 20 patients in each group (n=80): without IVIs and ophthalmic operations in history (group N1; control group); with the first IVI and antibiotic eye drops Tobrex applied 3d before IVI and 5d after it (group N2); with 20 or more IVIs and repeated courses of antibiotic eye drops (group N3); with the first IVI and antiseptic eye drops Vitabact (picloxydine) applied 3d before IVI and 5d after it (group N4). In groups N2 and N4 swabs were taken at baseline and after the treatment. Efficacy of picloxydine in inhibition of growth of conjunctival isolates susceptible and resistant to antibiotic was studied in vitro. Minimal inhibition concentrations (MIC) were determined with microdilution test. RESULTS: Two of the three patients who had to undergo the IVI procedure showed conjunctiva bacterial contamination. Along with few Staphylococcus aureus and Gram-negative isolates susceptible to most antibiotics, the majority (71%-77%) of causative agents were coagulase-negative Staphylococci (CoNS), 40%-50% of which were multidrug resistant (MDR). Eye disinfection in the operating room and peri-injection courses of Tobrex or Vitabact resulted in total elimination of isolates found at baseline. However, in 10% and 20% of patients, respectively, recolonization of the conjunctiva with differing strains occurred. In patients with repeated IVI and Tobrex/Maxitrol treatment, the conjunctival flora showed high resistance rates: 90% of CoNS were MDR. In the in vitro study, picloxydine showed bactericidal effect against Staphylococci isolates both antibiotic resistant and susceptible with MIC≥13.56 µg/mL. Incubation of bacteria for 15min in Vitabact eye drops, commercially available form of picloxydine, 434 µg/mL, showed total loss of colony forming units of all tested isolates including Pseudomonas aeruginosa. CONCLUSION: The confirmed efficacy of eye antiseptic picloxydine against conjunctival bacterial isolates and the presence of its commercial form, 0.05% eye drops, convenient for use by patients before and after injection, make this eye antiseptic promising for prophylaxis of IVI-associated infectious complications.
