Indian Journal of Ophthalmology最新文献

Purpose: To evaluate various supervised machine learning (ML) statistical models to predict anatomical outcomes after macular hole (MH) surgery using preoperative optical coherence tomography (OCT) features.

Methods: This retrospective study analyzed OCT data from idiopathic MH eyes at baseline and at 1-month post-surgery. The dataset was split 80:20 between training and testing. XLSTAT® statistical software (Lumivero, USA) was used to train different ML models on 10°CT parameters: prefoveal posterior cortical vitreous status, epiretinal membrane, intraretinal cysts, foveal retinal pigment epithelium hyperreflectivity, MH basal diameter, MH area (MHA), hole-forming factor, MH index, tractional hole index, and diameter hole index. The most effective statistical model was identified and was further assessed for accuracy, sensitivity, and specificity on a separate testing dataset.

Results: Six ML statistical models were trained on 33,260°CT data points from 3326°CT images of 308 operated MH (300 patients) eyes. Following training and internal validation, the random forest (RF) model achieved the highest accuracy (0.92), precision (0.94), recall (0.97), and F-score (0.96), and lowest misclassification rate. RF model identified the MHA index as the best predictor of post-surgical anatomical success. Following external testing, the RF model confirmed the highest accuracy and lowest misclassification rate (8.8%).

Conclusion: ML-based statistical models can be used to predict MH status after surgery. The RF model was the most accurate ML model, and the MHA index was the best predictor of postoperative hole closure after surgery based on preoperative OCT parameters. These predictions may help with future surgical planning for MH patients.

Purpose: To explore the relationship between gut microbiome, gut mycobiome, and intraocular (aqueous humor) microbiome dysbiosis in people with type 2 diabetes (T2DM) and diabetic retinopathy (DR).

Design: Multiple case-control studies.

Methods: We evaluated three groups of people: healthy controls (HC), people with T2DM without retinopathy, and those with DR. The study samples included fecal matter (30-50 g) and aqueous humor (0.05-0.1 mL). After amplicon sequencing, we analyzed microbiome profiles (V3-V4 region of bacterial 16S rRNA gene) and mycobiome (ITS2 region of fungal rRNA gene). The main outcome measures were relative abundance, α and β diversity, and dysbiotic bacteria and fungi, analyzed based on the inferred functions of the taxa.

Results: We recruited 82 people for gut microbiome (30 HC, 24 DM, and 28 DR); 75 people for gut mycobiome (30 HC, 21 DM, and 24 DR); and 12 people for aqueous humor microbiome (4 each HC, DM, and DR) studies. Generally, there was an increased abundance of pro-inflammatory and pathogenic microorganisms and a decreased abundance of anti-inflammatory and probiotic microorganisms. The differences were higher between HC and DM/DR than between DM and DR. In aqueous humor, there was a wider separation in microbiome profiles of people with DR than their gut microbiome.

Conclusion: The gut and aqueous humor microbiota of people with diabetes and DR may differ from those without diabetes. Given these unique observations in individuals living in one region of India, further research involving people from different regions is required to identify indices for possible regional or global use.

Purpose: Prematurity has been known to trigger several cellular pathways, leading to the clinical occurrence of retinopathy of prematurity (ROP). This study compared the levels of a panel of serum cytokines in premature infants with and without ROP.

Methods: This is a prospective observational study. Premature infants at 36-38 weeks' gestational age were recruited, their clinical data recorded, and serum samples collected and assayed for 18 cytokines. Based on follow-up examinations, patients were divided into two groups: No ROP and ROP. The ROP group was further divided into two subgroups: non-vision-threatening ROP (non-VTROP), and vision-threatening ROP (VTROP).

Results: On univariate analysis, among the clinical parameters, gestation age, birth weight, duration of invasive ventilation, and duration of stay in neonatal intensive care unit (NICU) were found to be significant. The univariate analysis also showed an association between raised levels of VEGF-D and IL-8 in the VTROP group. Multiple logistic regression indicated that gestation age was a significant risk factor across all subgroups. Additionally, VEGF-D levels were found to be significantly associated with VTROP.

Conclusion: Higher VEGF-D levels are associated with an increased risk of severe ROP that requires treatment and could potentially be used as a biomarker.

This report describes a challenging case of refractory acute retinal necrosis (ARN) with peripapillary choroiditis as the initial presenting sign. Imaging studies confirmed multifocal choroidal lesions and noted the novel sign of vertical hyper-reflective strips (VHRS) in the outer nuclear layer. Initial treatment with acyclovir and valacyclovir failed, and involvement of the other eye suggested a resistant variant of varicella zoster virus. High-dose oral famciclovir therapy resulted in rapid resolution of ARN in both eyes. This case highlights the rare occurrence of choroiditis, the novel finding of VHRS, and the potential utility of high-dose oral famciclovir in treating refractory ARN.

Purpose: To evaluate changes in corneal backscattering after collagen cross-linking (CXL) for progressive keratoconus and compare its course with different techniques - standard epithelium-off CXL (SCXL), contact lens-assisted CXL (CACXL), and transepithelial CXL (TECXL).

Setting: Advanced Eye Center, Post Graduate Institute of Medical Education and Research, Chandigarh, India.

Design: Retrospective comparative study.

Methods: Ninety-four eyes (SCXL: 47, CACXL: 30, and TECXL: 17) were compared. Corneal haze was quantified using Scheimpflug tomography, pre- and post-CXL at 1, 3, 6, and 12 months.

Results: The baseline mean density score of the central anterior stromal layer was 16.14 ± 7.07, 15.85 ± 7.89 and 15.89 ± 7.21 in SCXL, CACXL, and TECXL groups, respectively ( P 0.93). After SCXL, the score increased to 28.83 and 31.34 at 1 and 3 months, respectively (both P < 0.001) and dropped at 6 months (28.66, P < 0.001) and 12 months (23.72, P 0.003). Post-CACXL, the mean densitometry peaked at 3 months (20.35, P 0.14) and returned toward baseline at 6 months (18.82, P 0.15). After TECXL, it increased slightly at 1 month (18.47, P 0.17), decreased at 3 months (14.88, P 0.7), and plateaued over 1 year. No correlation with visual acuity was seen.

Conclusion: Corneal haze increased significantly after SCXL, peaking at 3 months, declining over 6-12 months, and returning to baseline at 12 months. In contrast, post-TECXL and -CACXL, there was an insignificant increase in anterior corneal haze, which returned to baseline within 3-6 months.

Purpose: To investigate the 12-month outcomes of ziv-aflibercept for neovascular age-related macular degeneration (nAMD) in eyes previously treated with aflibercept.

Methods: Retrospective chart review of patients with nAMD previously treated with aflibercept for at least 12 months and subsequently transitioned to ziv-aflibercept between January 1, 2019, and December 31, 2022, for a period of at least 12 months. Participants were identified, and their clinical and imaging information was extracted from our electronic health records system. Data on best corrected visual acuity (BCVA), intraocular pressure, injection intervals, central retinal thickness (CRT), volume cube presence of subretinal fluid (SRF), and intraretinal fluid (IRF) were obtained. Main outcome measures included changes in BCVA, injection intervals, CRT, SRF, and IRF before and after 12 months of ziv-aflibercept treatment.

Results: Fifty-four eyes of 44 patients were included in the study. After 12 months of ziv-aflibercept treatment, BCVA decreased by 0.84 ETDRS letters (P = 0.424) compared to BCVA at the last visit prior to conversion from aflibercept. Injection intervals decreased by 1.18 weeks (P = 0.489). CRT significantly decreased by 15.66 µm (P = 0.005). SRF was present initially in 31.5% of eyes and decreased to 22.2% (P = 0.125). IRF was present initially in 42.6% of eyes and decreased to 35.2% (P = 0.219).

Conclusion: Ziv-aflibercept demonstrated effectiveness in maintaining treatment outcomes in nAMD eyes previously treated with aflibercept. The treatment was well-tolerated with no reported adverse events. Ziv-aflibercept may be a cost-effective alternative and a potential solution to the financial burden associated with conventional anti-VEGF agents.