Reena Chopra

Use of Corneal Confocal Microscopy to Evaluate Small Nerve Fibers in Patients With Human Immunodeficiency Virus

Importance Objective quantification of small fiber neuropathy in patients with human immunodeficiency virus (HIV)–associated sensory neuropathy (HIV-SN) is difficult but needed for diagnosis and monitoring. In vivo corneal confocal microscopy (IVCCM) can quantify small fiber damage. Objective To establish whether IVCCM can identify an abnormality in corneal nerve fibers and Langerhans cells in patients with and without HIV-SN. Design, Setting, and Participants This prospective, cross-sectional cohort study was conducted between July 24, 2015, and September 17, 2015. Twenty patients who were HIV positive were recruited from adult outpatient clinics at Chelsea and Westminster Hospital NHS Foundation Trust in England. These patients underwent IVCCM at Moorfields Eye Hospital NHS Foundation Trust in London, England, and the IVCCM images were analyzed at Weill Cornell Medicine–Qatar in Ar-Rayyan, Qatar. Patients were given a structured clinical examination and completed validated symptom questionnaires and the Clinical HIV-Associated Neuropathy Tool. Results from patients with HIV were compared with the results of the age- and sex-matched healthy control participants (n = 20). All participants were classified into 3 groups: controls, patients with HIV but without SN, and patients with HIV-SN. Main Outcomes and Measures Comparison of corneal nerve fiber density, corneal nerve branch density, corneal nerve fiber length, corneal nerve fiber tortuosity, and corneal Langerhans cell density between healthy controls and patients with HIV with and without SN. Results All 40 participants were male, and most (≥70%) self-identified as white. Of the 20 patients with HIV, 14 (70%) had HIV-SN. This group was older (mean [SD] age, 57.7 [7.75] years) than the group without HIV-SN (mean [SD] age, 42.3 [7.26] years) and the controls (mean [SD] age, 53.8 [10.5] years). Corneal nerve fiber density was reduced in patients with HIV compared with the controls (26.7/mm2 vs 38.6/mm2; median difference, −10.37; 95.09% CI, −14.27 to −6.25; P < .001) and in patients with HIV-SN compared with those without (25.8/mm2 vs 30.7/mm2; median difference, −4.53; 95.92% CI, −8.85 to −0.26; P = .03). Corneal nerve branch density and corneal nerve fiber length were reduced in patients with HIV, but no differences were identified between those with neuropathy and without neuropathy (corneal nerve branch density: 95.83/mm2 for the controls vs 72.37/mm2 for patients with HIV; median difference, −24.53; 95.32% CI, −50.62 to −3.13; P = .01; and corneal nerve fiber length: 28.4 mm/mm2 for the controls vs 21.9 mm/mm2 for patients with HIV; median difference, −5.24; 95.09% CI, −8.83 to −1.38; P = .001). Tortuosity coefficient was increased in patients with HIV compared with controls (16.44 vs 13.95; median difference, 2.34; 95.09% CI, 0.31 to 4.65; P = .03) and in those with HIV-SN compared with those without (17.84 vs 14.18; median difference, 4.32; 95.92% CI, 0.68-9.23; P = .01). No differences were identified in corneal Langerhans cell density (19.84 cells/mm2 for the controls vs 41.43 cells/mm2 for patients with HIV; median difference, 9.38; 95% CI, −12.51 to 26.34; P = .53). Conclusions and Relevance In vivo corneal confocal microscopy could be used in the assessment of HIV-SN, but larger studies are required to confirm this finding.

Clinically applicable deep learning for diagnosis and referral in retinal disease

The volume and complexity of diagnostic imaging is increasing at a pace faster than the availability of human expertise to interpret it. Artificial intelligence has shown great promise in classifying two-dimensional photographs of some common diseases and typically relies on databases of millions of annotated images. Until now, the challenge of reaching the performance of expert clinicians in a real-world clinical pathway with three-dimensional diagnostic scans has remained unsolved. Here, we apply a novel deep learning architecture to a clinically heterogeneous set of three-dimensional optical coherence tomography scans from patients referred to a major eye hospital. We demonstrate performance in making a referral recommendation that reaches or exceeds that of experts on a range of sight-threatening retinal diseases after training on only 14,884 scans. Moreover, we demonstrate that the tissue segmentations produced by our architecture act as a device-independent representation; referral accuracy is maintained when using tissue segmentations from a different type of device. Our work removes previous barriers to wider clinical use without prohibitive training data requirements across multiple pathologies in a real-world setting.A novel deep learning architecture performs device-independent tissue segmentation of clinical 3D retinal images followed by separate diagnostic classification that meets or exceeds human expert clinical diagnoses of retinal disease.

One and Two Year Visual Outcomes from the Moorfields AMD Database - an Open Science Resource for the Study of Neovascular Age-related Macular Degeneration

Objectives To analyse treatment outcomes and share clinical data from a large, single-center, well-curated database (8174 eyes / 6664 patients with 120,756 single entries) of patients with neovascular age related macular degeneration (AMD) treated with anti-vascular endothelial growth factor (VEGF). By making our depersonalised raw data openly available, we aim to stimulate further research in AMD, as well as setting a precedent for future work in this area. Setting Retrospective, comparative, non-randomised electronic medical record (EMR) database cohort study of the UK Moorfields AMD database with data extracted between 2008 and 2018. Participants 3357 eyes/patients (61% female). Extraction criteria were ≥ 1 ranibizumab or aflibercept injection, entry of “AMD” in the diagnosis field of the EMR, and a minimum of one year of follow-up. Exclusion criteria were unknown date of first injection and treatment outside of routine clinical care at Moorfields before the first recorded injection in the database. Main outcome measures Primary outcome measure was change in VA at one and two years from baseline as measured in Early Treatment Diabetic Retinopathy Study (ETDRS) letters. Secondary outcomes were the number of injections and predictive factors for VA gain. Results Mean VA gain at one-year and two years were +5.5±0.5 and +4.9±0.68 letters respectively. Fifty-four percent of eyes gained ≥5 letters at two years, 63% had stable VA (±≤14 letters), forty-four percent of eyes maintained good VA (≥70 letters). Patients received a mean of 7.7±0.06 injections during year one and 13.0±0.2 injections over two years. Younger age, lower baseline VA, and more injections were associated with higher VA gain at two years. Conclusion This study benchmarks high quality EMR study results of real life AMD treatment and promotes open science in clinical AMD research by making the underlying data publicly available. Strengths and limitations of this study - Large sample size, retrospective, single centre, electronic medical record database study - High quality real life data - Open science approach with sharing of depersonalised raw data

Use of a Binocular Optical Coherence Tomography System to Evaluate Strabismus in Primary Position

Importance Current clinical methods for assessing strabismus can be prone to error. Binocular optical coherence tomography (OCT) has the potential to assess and quantify strabismus objectively and in an automated manner. Objective To evaluate the use of a binocular OCT prototype to assess the presence and size of strabismus. Design, Setting, and Participants Fifteen participants with strabismus were recruited in 2016 as part of the EASE study from Moorfields Eye Hospital National Health Service Foundation Trust, London, England, and 15 healthy volunteers underwent automated anterior segment imaging using the binocular OCT prototype. All participants had an orthoptic assessment, including alternating prism cover test (APCT), before undergoing imaging. Simultaneously acquired pairs of OCT images, captured with 1 eye fixating, were analyzed using ImageJ (National Institutes of Health) to assess the presence and angle of strabismus. Main Outcomes and Measures The direction and size of strabismus measured using binocular OCT was compared with that found using APCT. Results The median age for participants with strabismus was 55 years (interquartile range [IQR], 33-66.5 years) and for the healthy group, 50 years (IQR, 41-59 years); 15 participants (50%) were women, and 25 participants (83.3%) were white. The median magnitude of horizontal deviation was 20[INCREMENT] (IQR, 13-35[INCREMENT]) and for vertical deviation, 3[INCREMENT] (IQR, 0-5[INCREMENT]). Binocular OCT imaging correctly revealed the type and direction of the deviation in all 15 participants with strabismus, including horizontal and vertical deviations. The APCT and OCT measurements were strongly correlated for the horizontal (Pearson r = 0.85; 95% CI, 0.60-0.95; P < .001) and vertical (r = 0.89; 95% CI, 0.69-0.96; P < .001) deviations. In the healthy cohort, 9 of 15 participants (60%) had a latent horizontal deviation on APCT results (median magnitude 2[INCREMENT], range 2-4[INCREMENT]). Six (40%) had orthophoria. Horizontal deviations were observed on OCT imaging results in 12 of the 15 participants (80%), and a vertical deviation was visible in 1 participant (6.7%). Conclusions and Relevance These findings suggest that binocular anterior segment OCT imaging can provide clinicians with a precise measurement of strabismus. The prototype can potentially incorporate several binocular vision tests that will provide quantitative data for the assessment, diagnosis, and monitoring of ocular misalignments.