Predictors of visual recovery in patients with macular edema secondary to central retinal vein occlusion after treatment with Conbercept.

BACKGROUND: The use of Spectral domain optical coherence tomography (SD-OCT) to evaluate the predictors of visual acuity-recovery in patients treated with conbercept for macular edema (ME) secondary to central retinal vein occlusion (CRVO) has rarely been seen. We collected 26 CRVO-ME patients with different OCT measures at 6 months follow-up to identify the factors that are most strongly correlated with the best-corrected visual acuity (BCVA) post-treatment in CRVO-ME patients treated with conbercept. PURPOSE: To evaluate the effectiveness of intravitreal conbercept injections for the treatment of CRVO-ME and to determine the major predictors of best-corrected visual acuity (BCVA) post-treatment. METHODS: A retrospective study methodology was used. Twenty-six eyes from 26 patients with CRVO-ME were enrolled in the study. After an initial intravitreal injection of conbercept (0.5 mg/0.05 ml), monthly injections for up to 6 months were given following a 1 + PRN (pro re nata) regimen. Data collected at monthly intervals included measurements of the logMAR BCVA, central subfield thickness (CST), macular volume (MV), photoreceptor layer thickness (PLT), outer nuclear layer thickness (ONLT), and the disrupted ellipsoid zone (DEZ). The correlation between BCVA, before and after injections, and each of CST, MV, PLT, ONLT, DEZ was analyzed. RESULTS: The logMAR BCVA in months 3 and 6 post-injection was significantly improved relative to the baseline. In this same period the CST, MV, PLT, ONLT and DEZ were also significantly improved relative to the baseline. There was a negative correlation between PLT and logMAR BCVA at months 3 and 6 after treatment (r = - 0.549, P < 0.001; r = - 0.087, P < 0.001). CONCLUSION: Intravitreal injection of conbercept is an effective treatment for CRVO-ME. With 6 months of follow-up, logMAR BCVA and CST, MV, PLT, ONLT, DEZ improved. PLT was negatively correlated with the visual function in CRVO-ME patients after conbercept treatment, which may be a predictor of vision recovery in patients with CRVO-ME.

Intra- and interobserver concordance of a new classification system for myopic maculopathy.

BACKGROUND: Myopic maculopathy (MM) is one of the major causes of visual impairment and irreversible blindness in eyes with pathologic myopia (PM). However, the classification of each type of lesion associated with MM has not been determined. Recently, a new MM classification system, known as the ATN grading and classification system, was proposed; it is based on the fundus photographs and optical coherence tomography (OCT) images and includes three variable components: atrophy (A), traction (T), and neovascularization (N). This study aimed to perform an independent evaluation of interobserver and intraobserver agreement for the recently developed ATN grading system for MM. METHODS: This was a retrospective study. Fundus photographs and OCT images of 125 patients (226 eyes) with various MMs were evaluated and classified using the ATN grading of the new MM classification system by four blinded and independent evaluators (2 attending ophthalmologists and 2 ophthalmic residents). All cases were randomly re-evaluated by the same observers after an interval of 6 weeks. The kappa coefficient (κ) and 95% confidence interval (CI) were used to determine the interobserver and intraobserver agreement. RESULTS: The interobserver agreement was substantial when considering the maculopathy type (A, T, and N). The weighted Fleiss κ values for each MM type (A, T, and N) were 0.651 (95% CI: 0.602-0.700), 0.734 (95% CI: 0.689-0.779), and 0.702 (95% CI: 0.649-0.755), respectively. The interobserver agreement when considering the subtypes was good or excellent, except for stages A1, A2, and N1, in which the weighted κ value was less than 0.6, with moderate agreement. The intraobserver agreement of types and subtypes was excellent, with κ > 0.8. No significant differences were observed between the attending ophthalmologists and residents for interobserver reliability or intraobserver reproducibility. CONCLUSIONS: The ATN classification allows an adequate agreement among ophthalmologists with different qualifications and by the same observer on separate occasions. Future prospective studies should further evaluate whether this classification can be better implemented in clinical decision-making and disease progression assessments.

Performance of Deep Transfer Learning for Detecting Abnormal Fundus Images.

PURPOSE: To develop and validate a deep transfer learning (DTL) algorithm for detecting abnormalities in fundus images from non-mydriatic fundus photography examinations. METHODS: A total of 1295 fundus images were collected to develop and validate a DTL algorithm for detecting abnormal fundus images. After removing 366 poor images, the DTL model was developed using 929 (370 normal and 559 abnormal) fundus images. Data preprocessing was performed to normalize the images. The inception-ResNet-v2 architecture was applied to achieve transfer learning. We tested our model using a subset of the publicly available Messidor dataset (using 366 images) and evaluated the testing performance of the DTL model for detecting abnormal fundus images. RESULTS: In the internal validation dataset (n = 273 images), the area under the curve (AUC), sensitivity, accuracy, and specificity of DTL for correctly classified fundus images were 0.997%, 97.41%, 97.07%, and 96.82%, respectively. For the test dataset (n = 273 images), the AUC, sensitivity, accuracy, and specificity of the DTL for correctly classifying fundus images were 0.926%, 88.17%, 87.18%, and 86.67%, respectively. CONCLUSION: DTL showed high sensitivity and specificity for detecting abnormal fundus-related diseases. Further research is necessary to improve this method and evaluate the applicability of DTL in community health-care centers.