Acta Ophthalmologica最新文献

Purpose: To validate the performance of autonomous diabetic retinopathy (DR) grading by comparing a human grader and a self-developed deep-learning (DL) algorithm with gold-standard evaluation.

Methods: We included 500, 6-field retinal images graded by an expert ophthalmologist (gold standard) according to the International Clinical Diabetic Retinopathy Disease Severity Scale as represented with DR levels 0-4 (97, 100, 100, 103, 100, respectively). Weighted kappa was calculated to measure the DR classification agreement for (1) a certified human grader without, and (2) with assistance from a DL algorithm and (3) the DL operating autonomously. Using any DR (level 0 vs. 1-4) as a cutoff, we calculated sensitivity, specificity, as well as positive and negative predictive values (PPV and NPV). Finally, we assessed lesion discrepancies between Model 3 and the gold standard.

Results: As compared to the gold standard, weighted kappa for Models 1-3 was 0.88, 0.89 and 0.72, sensitivities were 95%, 94% and 78% and specificities were 82%, 84% and 81%. Extrapolating to a real-world DR prevalence of 23.8%, the PPV were 63%, 64% and 57% and the NPV were 98%, 98% and 92%. Discrepancies between the gold standard and Model 3 were mainly incorrect detection of artefacts (n = 49), missed microaneurysms (n = 26) and inconsistencies between the segmentation and classification (n = 51).

Conclusion: While the autonomous DL algorithm for DR classification only performed on par with a human grader for some measures in a high-risk population, extrapolations to a real-world population demonstrated an excellent 92% NPV, which could make it clinically feasible to use autonomously to identify non-DR patients.

Objectives: Patients with polypoidal choroidal vasculopathy (PCV) exhibit variability in response to anti-VEGF therapy. This study aimed to analyse the aqueous humour proteomic profiles of PCV patients and provide preliminary insights for the identification of biomarkers associated with anti-VEGF drug responsiveness.

Methods: PCV patients who were treatment-naïve or untreated for more than 3 months were prospectively recruited from two hospitals in Beijing and Tianjin. Based on the relative changes in central macular thickness (ΔCMT/baseline-CMT) before and after anti-VEGF treatment, the PCV patients were divided into a good response (GR) group (≤-25%) and a poor response (PR) group (>-25%). Aqueous humour proteomics was performed by the Data-independent Acquisition-Mass Spectrometry (DIA-MS) method, and differentially expressed proteins (DEPs) analysis between the different PCV groups and the control group was conducted. Key DEPs were selected for preliminary validation in the aqueous humour using the Luminex method retrospectively.

Results: A total of 31 PCV patients (31 eyes) were included, 13 in the GR group and 18 in the PR group. A total of 414 DEPs were identified, including 36 significantly upregulated proteins, such as G protein regulatory factor 10 (RGS10), podocin (PODN) and epidermal growth factor (EGF), and 32 downregulated proteins, including RAB11FIP4 (Rab11 family-interacting protein 4), α-synuclein (SNCA), haemoglobin subunit δ (HBD) and interleukin 6 (IL6). Compared to the cataract control group (10 eyes), 134 proteins were significantly upregulated, and 72 were downregulated. KEGG pathway enrichment analysis revealed that the GR and PR groups differ in terms of cell communication, and cell signal transduction. Protein-protein interaction analysis revealed interactions between EGF and various DEPs. Validation of aqueous humour proteins using the Luminex method revealed that changes in the levels of EGF were associated with the anti-VEGF treatment response in PCV patients.

Conclusions: PCV patients with good or poor anti-VEGF responses exhibit distinct aqueous humour proteomic profiles. Aqueous EGF may serve as a biomarker for the 'precise treatment' of PCV.

Central serous chorioretinopathy (CSC) is a prevalent exudative maculopathy. Understanding risk factors for CSC is important for disease prevention and to provide evidence-based advice to patients. In this study, we systematically reviewed the literature and performed meta-analysis on the association between tobacco consumption and CSC. We searched 12 literature databases on May 5, 2024, and identified 11 eligible studies of 27 595 patients with CSC and 105 354 control individuals. Studies were predominantly clinic-based case-control studies. We calculated a summary estimate of tobacco consumption as a risk factor for CSC at an odds ratio of 2.99 (95% CI: 1.82-4.93, p = 0.000017), which remained statistically significant in the sensitivity analyses. The exact mechanism by which tobacco consumption contribute to the pathophysiology of CSC remains unclear, although several potential hypotheses exist. However, tobacco consumption is a modifiable behaviour and tobacco cessation is an actionable advice with which patients with CSC themselves can play a large role in disease management. Further studies are warranted to understand the impact of tobacco cessation for risk modification and for the prognosis of patients who already have CSC.

Purpose: To examine the prevalence of self-reported cataract and cataract surgery, and the incidence of cataract surgery, in relation to smoking and use of the moist smokeless tobacco product snus.

Methods: In 2014/2015, individuals born in 1951 (n = 18 055) in the Västra Götaland County, Sweden, were invited to participate. Of these, 9743 (54%) accepted participation and 9316 (52%) remained after exclusion criteria were applied. Participants answered a questionnaire with items about eye conditions, smoking, snus, gender, education, asthma, chronic obstructive pulmonary disease, corticosteroid use, diabetes mellitus, weight and height. Prevalence ratios (PRs) for self-reported cataract and cataract surgery were calculated. The incidence of cataract surgery was assessed, and hazard ratios (HRs) were presented.

Results: Having ever smoked was associated with a higher prevalence ratio of self-reported cataract (PR 1.19, 95% confidence interval [CI] 1.04-1.35) and cataract surgery (PR 1.27, 95% CI 1.06-1.53), compared to those who had never been daily smokers. Currently, a smoker was associated with a higher HR of cataract surgery (HR 1.34, 95% CI 1.04-1.74), as well as having been a former smoker (HR 1.27, 95% CI 1.03-1.56). Total years of smoking were associated with an increased risk for cataract surgery (HR 1.05, 95% CI 1.02-1.08 for 5 years of smoking). Snus use was not associated with an increased prevalence of cataract or incidence of cataract surgery, except among women who were current snus users (HR for cataract surgery 2.04, 95% CI 1.16-3.60 n = 108).

Conclusion: Smoking is associated with a higher prevalence of cataracts, and a higher incidence of cataract surgery, indicating a dose-response relationship. However, there was no firm association between snus use and cataract.

Purpose: To determine the pattern, degree and prevalence of optic disc optical coherence tomography (OCT) alterations in adults born preterm with very low birth weight (VLBW; birth weight < 1500 g).

Methods: Optic disc OCT was assessed in 98 VLBW participants and 139 term-born controls from birth cohorts in Finland and Norway at the mean age of 36 years. The participants had not been treated for retinopathy of prematurity and had no diagnosed brain abnormality. OCT assessment included parameters for optic disc size, neural rim and peripapillary retinal nerve fibre layer thickness (pRNFLT), and for the foveal developmental stage. Background data, visual acuity, refractive error and intraocular pressure were recorded.

Results: In the VLBW group, optic disc neural rim and pRNFLT were significantly decreased, most frequently in the nasal and inferior sectors. In 40% (95% CI: 33-47) of the VLBW eyes, nerve fibre thickness of at least one optic disc sector was below the fifth percentile of the control group, including 9% (95% CI: 6-14) subgroup falling below the first percentile, that is, clinically below normal limits. VLBW participants with nerve fibre thickness below the fifth percentile did not differ by perinatal data or foveal developmental stage from the other VLBW participants. All participants had normal visual acuity.

Conclusion: A moderate decrease of the optic disc neural rim and pRNFLT is frequently seen in clinically healthy adults born preterm with VLBW. Awareness of the VLBW-related optic disc nerve fibre shallowness especially in the inferior and nasal sectors is important while evaluating acquired optic disc pathology in adulthood.

Purpose: To compare vitrectomy with and without internal limiting membrane (ILM) peeling in small idiopathic macular holes.

Methods: Retrospective multicentre study including consecutive eyes with ≤250 μm idiopathic macular hole treated with vitrectomy. The primary outcome was hole closure rate. Best-corrected visual acuity (BCVA) change, closure patterns on optical coherence tomography, rates of external limiting membrane (ELM) and ellipsoid zone (EZ) recovery, and rate of complications were also investigated.

Results: In total, 693 eyes were included. Hole closure rate was 98% in the peeling and 85% in the no-peeling group (p < 0.001). At 12 months, mean BCVA change was 0.38 ± 0.22 logMAR in the peeling and 0.45 ± 0.21 logMAR in the no-peeling group (p = 0.02); 66% versus 80% of eyes had a U-shaped morphology, respectively; EZ recovery rate was 75% and 93%, respectively (p = 0.02). In the no-peeling group, eyes with a vitreomacular traction (VMT) showed a 96% closure rate, comparable to the peeling group (p = 0.40). The incidence of adverse events was similar except for dissociated optic nerve fibre layer (55% in the peeling vs. 9% in the no-peeling group, p < 0.001).

Conclusions: In small idiopathic macular holes, ILM peeling provides a higher closure rate compared to no-peeling; however, if a VMT is present closure rates are comparable. In closed macular holes, the no-peeling technique provides advantages in terms of visual outcome and anatomical recovery.

Purpose: To analyse the impact on cost if faricimab is used as the first-line treatment for neovascular age-related macular degeneration (nAMD) compared to standard treatment with bevacizumab.

Methods: Retrospective registry study including real-world data from the Swedish Macula Registry between 2017 and 2022. The observed number of injections and visits for bevacizumab during the first two years of treatment was used (n = 437 patients). Number of faricimab injections was obtained from published clinical trial data and unit costs mostly from publicly available Swedish sources. The provider cost included medication and visit cost and societal cost included additionally patient travel cost. Costs are presented in 2023 EUR.

Results: The incremental societal cost of faricimab was 277 EUR per patient compared to bevacizumab in the base case. Medication cost was higher (1516 EUR) while visit cost (-1183 EUR) and patient travel cost (-56 EUR) were lower due to longer injection intervals. Faricimab was of similar cost as bevacizumab for patients residing far from the clinic. The faricimab injection interval and the number of bevacizumab injections were major drivers of uncertainty in the results.

Conclusion: Faricimab represents a cost-effective alternative to bevacizumab for patients with nAMD in Sweden. Its extended treatment interval is particularly beneficial for patients living far from clinics, and if the real-life faricimab injection interval extends beyond 12 weeks. Our findings emphasize faricimab's potential to free up healthcare staff to treat a larger patient population with existing clinic resources.

Purpose: To assess the risk for pseudophakic cystoid macular oedema (PCME) and posterior capsular opacification (PCO) associated with combined cataract surgery and trabeculectomy compared to cataract surgery alone.

Methods: Data analysis of subjects who underwent routine cataract surgery without and with concomitant trabeculectomy at the Department of Ophthalmology, Bristol Eye Hospital, the UK, between January 2008 and December 2017. Odds ratios (ORs) for PCME between the types of surgeries were calculated using univariate and multivariate regression analysis. Multivariate Cox regression controlling for age and gender was used to estimate the hazard ratio (HR) for neodymium-doped yttrium aluminium garnet (Nd:YAG) laser capsulotomies.

Results: This study included 56 973 cataract surgeries without and 288 with concomitant trabeculectomy (phaco-trab) with a mean follow-up time of 6.9 ± 4.2 years. Baseline variables (age and gender, diabetes, pseudoexfoliation, use of pupil expansion device and postoperative follow-up time) were comparable between the groups. Postoperative rates of PCME remained non-significant between the cataract surgery and phaco-trabe groups both in uni- and multi-variate analysis (OR 0.347, 95%CI 0.049-2.477, p = 0.291). Furthermore, in Cox regression analysis adjusted for the patients' age and gender, Nd:YAG laser capsulotomy rates remained non-significant between the cataract surgery and phaco-trabe groups (HR 1.250, 95%CI 0.883-1.769, p = 0.209).

Conclusions: In our large cohort study, combining trabeculectomy with cataract surgery did not predispose to an increased PCME or Nd:YAG laser capsulotomy rates.

Purpose: Recently, an 'hyperreflective ganglion cell layer band' (HGB) has been described on spectral-domain optical coherence tomography (SD-OCT) in a subset of patients with retinitis pigmentosa (RP). This study aims to validate and describe the frequency of HGB in a large cohort of Portuguese patients with RP.

Methods: This single-centre, cross-sectional cohort study included consecutive patients with a genetic diagnosis of RP. SD-OCT images were reviewed to identify the presence of the HGB and other retinal comorbidities. The HGB was defined as a continuous hyperreflective band within the thickness of the ganglion cell layer (GCL). We built mixed-effects regression models, accounting for inter-eye correlations, to investigate features predictive of visual acuity. Subsequently, a reduced model was fitted.

Results: A total of 398 eyes from 201 patients were included. HGB was identified in 69 (17.3%) eyes from 39 (19.4%) patients. Patients presenting with the HGB were significantly younger at diagnosis and at symptom onset. Median BCVA [ETDRS (IQR)] was 65 (29) letters in eyes with the HGB and 70 (21) letters in eyes without HGB (p < 0.001). In both the full and reduced mixed-effects models, the presence of HGB and macular hole (MH) was significantly associated with worse BCVA.

Conclusions: This study validates the recent description of HGB within the GCL in a subset of patients with RP. Eyes with HGB demonstrated significantly worse BCVA compared to those without HGB, suggesting that the presence of HGB may serve as an SD-OCT biomarker of worse visual prognosis in these patients.

Purpose: To examine the effectiveness and identify clinical response predictors of a short corticosteroid-based regimen consisting of topical preservative-free 0.1% dexamethasone (Monopex®, Théa Laboratories) in conjunction with artificial tears (AT) for dry eyes in a real-life clinical setting.

Methods: Patients were recruited from the Norwegian Dry Eye Clinic and were allowed to use ATs of their own choice in addition to the prescribed 14-day topical dexamethasone course. Ocular Surface Disease Index (OSDI), Dry Eye Questionnaire (DEQ-5), Schirmer test (ST), fluorescein tear film break-up time (FBUT), ocular surface staining (OSS), meibum expressibility (ME), meibum quality (MQ), number of expressible meibomian glands among the central eight glands in the lower lids (N_MG) and intraocular pressure (IOP) were measured at baseline and at 1-month follow-up. The average values of clinical parameters from both eyes were used for analyses. A paired t-test and a significance value of p < 0.05 were used for statistical analyses. Associations between sex, age, baseline values and the changes after the intervention (Δ) were explored using linear regression.

Results: A total of 167 patients (124 women, mean age 54 years ±17 (standard deviation)) were included. One month after initiation of intervention, OSDI and DEQ5 scores improved from 39.5 ± 22.1 to 31.4 ± 21.3 (p < 0.001) and from 12.6 ± 4.2 to 11.0 ± 4.6 (p < 0.001), respectively. OSS improved from 2.2 ± 1.4 to 1.8 ± 1.5 (p < 0.001), N_MG increased from 4.8 ± 2.2 to 5.1 ± 2.2 (p < 0.05), while IOP decreased from 12.9 ± 3.3 to 12.4 ± 3.5 mmHg (p < 0.05). Significant associations were found between the change in symptoms and objective measures of DED (Δ_OSDI, Δ_DEQ5, Δ_OSS, Δ_FBUT, ΔN_MG, Δ_MQ) and their respective baseline values (OSDI, DEQ5, OSS, FBUT, N_MG, MQ). The remaining tests did not show statistically significant changes.

Conclusion: Improvement in dry eye symptoms and signs were observed following a short course of topical, preservative-free 0.1% dexamethasone treatment in combination with AT. Individuals exhibiting more pronounced symptoms and signs witnessed the most profound improvements with the treatment regimen, suggesting that poor baseline parameters may serve as response predictors of the treatment regimen. While the real-life data presented herein are valuable, the conclusions are limited by the inherent biases of a non-controlled study.