AUTOMATED DETECTION OF VITRITIS USING ULTRAWIDE-FIELD FUNDUS PHOTOGRAPHS AND DEEP LEARNING.

BACKGROUND/PURPOSE: Evaluate the performance of a deep learning algorithm for the automated detection and grading of vitritis on ultrawide-field imaging. METHODS: Cross-sectional noninterventional study. Ultrawide-field fundus retinophotographs of uveitis patients were used. Vitreous haze was defined according to the six steps of the Standardization of Uveitis Nomenclature classification. The deep learning framework TensorFlow and the DenseNet121 convolutional neural network were used to perform the classification task. The best fitted model was tested in a validation study. RESULTS: One thousand one hundred eighty-one images were included. The performance of the model for the detection of vitritis was good with a sensitivity of 91%, a specificity of 89%, an accuracy of 0.90, and an area under the receiver operating characteristics curve of 0.97. When used on an external set of images, the accuracy for the detection of vitritis was 0.78. The accuracy to classify vitritis in one of the six Standardization of Uveitis Nomenclature grades was limited (0.61) but improved to 0.75 when the grades were grouped into three categories. When accepting an error of one grade, the accuracy for the six-class classification increased to 0.90, suggesting the need for a larger sample to improve the model performances. CONCLUSION: A new deep learning model based on ultrawide-field fundus imaging that produces an efficient tool for the detection of vitritis was described. The performance of the model for the grading into three categories of increasing vitritis severity was acceptable. The performance for the six-class grading of vitritis was limited but can probably be improved with a larger set of images.

Persistence of Juvenile Idiopathic Arthritis-Associated Uveitis in Adulthood: A Retrospective Study.

Background: Juvenile idiopathic arthritis (JIA) is a rheumatic condition of childhood that is frequently associated with anterior chronic uveitis. Evidence suggests that uveitis may persist up to adulthood in some cases, possibly causing severe visual impairment. Methods: We conducted a retrospective study on a series of patients aged 16 years or older with JIA-related active uveitis who were referred to the Uveitis Service of Sapienza University of Rome from 1990 to 2019 to evaluate the characteristics of ocular disease in patients with JIA-associated uveitis (JIA-U) who still exhibit uveitis in adulthood. Data on clinical features, treatment, complications and visual outcomes were collected. Results: Twenty adults (85% female; median age 23.4 ± 6.6 years, range 16−38 years) with ongoing uveitis (35 eyes) were identified. The median age at JIA onset was 6.15 ± 2.9 years (range 2−10), and uveitis onset was 8.7 ± 4.7 years (range 3−20). The patients were observed in a median follow-up of 16 ± 7.7 years (range 4−35). Fifty-seven percent of affected eyes (20 eyes) had good visual acuity (>0.4 logMAR), while eleven percent of affected eyes (4 eyes) were blind (≤20/200). Uveitis required topical steroids and mydriatic/cycloplegic in all cases. Orbital steroid injection was performed in 13 eyes. Systemic corticosteroids and biologic drugs were used in 14 patients. Conclusions: Although the visual prognosis of JIA-U has improved in recent years, persistent uveitis up to adulthood is still observed. Therefore, protracted follow-up of JIA-U patients is warranted because of the high burden of delayed visual complications.

Antiviral and immunomodulatory interferon-beta in high-risk COVID-19 patients: a structured summary of a study protocol for a randomised controlled trial.

OBJECTIVES: The primary objective of the study is to demonstrate the efficacy of low-dose IFN-ß in reducing the risk of SARS-CoV-2 recently infected elderly patients to progress towards severe COVID-19 versus control group within 28 days. Secondary objectives are: 1) To assess the reduction in Intensive Care Unit (ICU) admission in patients treated with IFN-ß versus control group within 28 days of randomization 2) To assess the reduction in number of deaths in IFN- ß compared to control group (day 28) 3) To evaluate the increase in proportion of participants returning to negative SARS-CoV-2 RT-PCR in IFN-ß -treated versus control group at Day 14 and Day 28 4) To assess the increase in SARS-CoV-2-specific binding antibody titers in IFN-ß compared to control group (day 28) 5) To assess the safety of IFN-ß -treated patients versus control group TRIAL DESIGN: Randomized, Open-Label, Controlled, Superiority Phase II Study. Patients, who satisfy all inclusion criteria and no exclusion criteria, will be randomly assigned to one of the two treatment groups in a ratio 2:1 (IFN-treated versus control patients). Randomization will be stratified by gender. Stratified randomization will balance the presence of male and female in both study arms. PARTICIPANTS: Male and female adults aged 65 years or older with newly diagnosed SARS-CoV-2 infection and mild COVID-19 symptoms are eligible for the study. The trial is being conducted in Rome. Participants will be either hospitalized or home isolated. A group of physicians belonging to the Special Unit for Regional Continued Care (USCAR), specifically trained for the study and under the supervision of the National Institute for Infectious Diseases "Lazzaro Spallanzani", will be responsible for the screening, enrolment, treatment and clinical monitoring of patients, thus acting as a bridge between clinical centers and territorial health management. Inclusion criteria are as follows: ≥ 65 years of age at time of enrolment; Laboratory-confirmed SARS-CoV-2 infection as determined by PCR, in any specimen < 72 hours prior to randomization; Subject (or legally authorized representative) provides written informed consent prior to initiation of any study procedures; Understands and agrees to comply with planned study procedures; Agrees to the collection of nasopharyngeal swabs and venous blood samples per protocol; Being symptomatic for less than 7 days before starting therapy; NEWS2 score ≤2. Exclusion criteria are as follows: Hospitalized patients with illness of any duration, and at least one of the following: Clinical assessment (evidence of rales/crackles on exam) and SpO2 ≤ 94% on room air at rest or after walking test, OR Acute respiratory failure requiring mechanical ventilation and/or supplemental oxygen; Patients currently using IFN-ß (e.g., multiple sclerosis patients); Patients undergoing chemotherapy or other immunosuppressive treatments; Patients with chronic kidney diseases; Known allergy or hypersensitivity to IFN (including asthma); Any autoimmune disease (resulting from patient anamnesis); Patients with signs of dementia or neurocognitive disorders; Patients with current severe depression and/or suicidal ideations; Being concurrently involved in another clinical trial; HIV infection (based on the anamnesis); Use of any antiretroviral medication; Impaired renal function (eGFR calculated by CKD-EPI Creatinine equation < 30 ml/min); Presence of other severe diseases impairing life expectancy (e.g. patients are not expected to survive 28 days given their pre-existing medical condition); Any physical or psychological impediment in a patient that could let the investigator to suspect his/her poor compliance; Lack or withdrawal of informed consent INTERVENTION AND COMPARATOR: Control arm: No specific antiviral treatment besides standard of care. Treatment arm: 11µg (3MIU) of IFN-ß1a will be injected subcutaneously at day 1, 3, 7, and 10 in addition to standard of care. The drug solution, contained in a pre-filled cartridge, will be injected by means of the RebiSmart® electronic injection device. Interferon ß1a (Rebif®, Merck KGaA, Darmstadt, Germany) is a disease-modifying drug used to treat relapsing forms of multiple sclerosis (MS). The dose selected for this study is expected to exploit the antiviral and immunomodulatory properties of the cytokine without causing relevant toxicity or inducing refractoriness phenomena sometimes observed after high-dose and/or chronic IFNß treatments. MAIN OUTCOMES: Primary endpoint of the study is the proportion of patients experiencing a disease progression, during at least 5 days, according to the National Early Warning Score (NEWS2). The NEWS2 score is a standardized approach aimed at promptly detecting signs of clinical deterioration in acutely ill patients and establishing the potential need for higher level of care. It is based on the evaluation of vital signs, including respiratory rate, oxygen saturation, temperature, blood pressure, pulse/heart rate, AVPU response. The resulting observations, compared to a normal range, are combined in a single composite "alarm" score. Any other clinical sign clearly indicating a disease worsening will be considered as disease progression. RANDOMIZATION: Sixty patients will be randomized 2:1 to receive IFN-ß1a plus the standard of care or the standard of care only. Eligible patients will be randomized (no later than 36 h after enrolment) by means of a computerized central randomization system. All patients will receive a unique patient identification number at enrolling visit when signing the informed consent and before any study procedure is performed. This number remains constant throughout the entire study. The randomization of patients will be closed when 60 patients have been enrolled. The randomization will be stratified by sex; for each stratum a sequence of treatments randomly permuted in blocks of variable length (3 or 6) will be generated. BLINDING (MASKING): This is an open-label study. After the randomization, patients will be notified whether they will be in the experimental arm or in the control arm. NUMBERS TO BE RANDOMISED (SAMPLE SIZE): The study plans to enrol 60 patients: 40 in the IFN-ß1a arm, 20 in the control arm, according to a 2:1 - treated: untreated ratio. TRIAL STATUS: Protocol Version: 3.0 Version Date: 18/03/2021 The study is open for recruitment since 16/04/2021.Recruitment is expected to l be completed before 15/08/2021. TRIAL REGISTRATION: EudraCT N°: 2020-003872-42, registration date: 19/10/2020. FULL PROTOCOL: The full protocol is attached as an additional file, accessible from the Trials website (Additional file 1). In the interest in expediting dissemination of this material, the familiar formatting has been eliminated; this Letter serves as a summary of the key elements of the full protocol."

Surgery for Idiopathic Epimacular Membrane: Morpho-Functional Outcomes Based on the Preoperative Macular Integrity of the Photoreceptoral Junction. A Prospective Pilot Study.

INTRODUCTION: This study aimed to evaluate whether the preoperative integrity of the inner segment (IS) and outer segment (OS) photoreceptoral junction may influence the postoperative visual acuity, the macular morphology [assessed by spectral domain optical coherence tomography (SD-OCT)], and macular function (evaluated by multifocal electroretinogram, mfERG) in patients with idiopathic epimacular membrane (EMM) followed up for 6 months. METHODS: In this observational prospective study, 18 patients with EMM (mean age 72.5 ± 6.87 years) were enrolled. They were divided into two groups according to the preoperative integrity of the SD-OCT IS/OS junction: the EMM-I group with an intact IS/OS junction (11 patients, mean age 72.75 ± 3.49 years, providing 11 eyes) and the EMM-D group with a disrupted IS/OS junction (7 patients, mean age 70.86 ± 10.79 years, providing 7 eyes). For each enrolled patient, visual acuity (VA), mfERG, and SD-OCT were assessed at baseline (preoperative) and after 1, 3, and 6 months of follow-up after surgical treatment for EMM (pars plana vitrectomy with EMM removal and internal limiting membrane peeling). RESULTS: During the whole follow-up, VA was significantly increased in EMM-I eyes and unmodified in EMM-D eyes. In both groups, mfERG responses were not significantly different and not related to VA differences. In EMM-I eyes a significant reduction of central retinal thickness (CRT) was observed; however, it was not correlated with VA changes. In EMM-D eyes CTR was not significantly reduced, whereas macular volume was significantly reduced. These changes were significantly related to the corresponding differences in VA. CONCLUSIONS: Our results suggest that the preoperative evaluation of the integrity of the IS/OS junction is relevant for postoperative outcomes. The recovery in VA was higher in EMM-I eyes than in EMM-D eyes. Postoperative recovery was not associated with morphology of the outer retina (photoreceptor and outer nuclear layer) and the function of preganglionic elements.