Search

Diagnostic accuracy, a measure of diagnostic tests for correctly identifying patients with or without a target disease, plays an important role in evidence-based medicine. Diagnostic accuracy of a new test ideally should be evaluated by comparing to a gold standard; however, in many medical applications it may be invasive, costly, or even unethical to obtain a gold standard for particular diseases. When the accuracy of a new candidate test under evaluation is assessed by comparison to an imperfect reference test, bias is expected to occur and result in either overestimates or underestimates of its true accuracy. In addition, diagnostic test studies often involve repeated measurements of the same patient, such as the paired eyes or multiple teeth, and generally lead to correlated and clustered data. Using the conventional statistical methods to estimate diagnostic accuracy can be biased by ignoring the within-cluster correlations. Despite numerous statistical approaches have been proposed to tackle this problem, the methodology to deal with correlated and clustered data in the absence of a gold standard is limited. In this article, we propose a method based on the composite likelihood function to derive simple and intuitive closed-form solutions for estimates of diagnostic accuracy, in terms of sensitivity and specificity. Through simulation studies, we illustrate the relative advantages of the proposed method over the existing methods that simply treat an imperfect reference test as a gold standard in correlated and clustered data. Compared with the existing methods, the proposed method can reduce not only substantial bias, but also the computational burden. Moreover, to demonstrate the utility of this approach, we apply the proposed method to the study of National-Eye-Institute-funded Telemedicine Approaches to Evaluating of Acute-Phase Retinopathy of Prematurity (e-ROP), for estimating accuracies of both the ophthalmologist examination and the image evaluation.

PURPOSE To evaluate associations between visual function and the level of uncorrected hyperopia in 4- and 5-year-old children without strabismus or amblyopia. METHODS Children with spherical equivalent (SE) cycloplegic refractive error of -0.75 to +6.00 on eligibility testing for the Vision in Preschoolers-Hyperopia in Preschoolers (VIP-HIP) study were included. Children were grouped as emmetropic (

OBJECTIVE: This study evaluates the 24 month follow up for the NICHD Neonatal Research Network (NRN) Inositol for Retinopathy Trial. STUDY DESIGN: Bayley Scales of Infants Development-III and a standardized neurosensory examination were performed in infants enrolled in the main trial. Moderate/severe NDI was defined as BSID-III Cognitive or Motor composite score

Background Retinopathy of prematurity (ROP) is a widely recognized cause of blindness after preterm birth. The incidence of ROP is rising especially in low- and middle-income countries (LMIC) because of improved neonatal care and increased survival of very premature neonates. To date, there is no data on incidence of ROP in Botswana. Objective The purpose of this study was to provide initial data and determine ROP-associated risk factors from a single neonatal care center on the incidence of ROP in Gaborone, Botswana. Methods A prospective observational study was conducted at Princess Marina Hospital (PMH) in Gaborone, Botswana. Premature neonates with birth weights (BW) of

BACKGROUND AND OBJECTIVE: To compare morphologic and functional status at age 4 years for patients treated in one eye with laser photocoagulation and the other eye with intravitreal bevacizumab (IVB) injection for Type 1 retinopathy of prematurity (ROP). PATIENTS AND METHODS: In this single-center, randomized, controlled trial, best-corrected visual acuity (BCVA) in logMAR was obtained along with spherical equivalent refraction (SER), fluorescein angiography (FA), optical coherent tomography (OCT), and OCT angiography (OCTA). RESULTS: Eighteen babies (36 eyes) were selected for this study. BCVA and SER were similar in the two groups, but six patients had anisometropia of 4 diopters or more. IVB-treated eyes tended to have thinner foveal thickness than laser-treated eyes (mean difference: −5.33 pixels; 95% confidence interval, −9.62 to −1.05). CONCLUSION: Although the differences found here are minimal between the IVB-treated and laser-treated groups, further long-term evaluation of not only FA, but also OCT and OCTA, are needed in larger studies. [ Ophthalmic Surg Lasers Imaging Retina . 2020;51:180–186.]

BACKGROUND AND OBJECTIVES: Retinopathy of prematurity (ROP) is a leading cause of childhood blindness. Screening and treatment reduces this risk, but requires multiple examinations of babies, most of whom will not develop severe disease. Previous work has suggested that artificial intelligence (AI) may be able to detect incident severe (treatment-requiring [TR]) ROP prior to clinical diagnosis. We aimed to build a risk model that combined AI with clinical demographics to reduce the number of examinations without missing cases of TR-ROP. METHODS: Infants undergoing routine ROP screening examinations (1579 total eyes, 190 with TR-ROP) were recruited from eight North American study centers. A vascular severity score (VSS) was derived from retinal fundus images obtained at 32–33 weeks postmenstrual age. Seven ElasticNet logistic regression models were trained on all combinations of birthweight (BW), gestational age (GA), and VSS. Area under the precision-recall curve (AUPR) identified the highest performing model. RESULTS: The GA + VSS model had the highest performance (mean ± SD AUPR: 0.35 ± 0.11). On two different test datasets (n = 444 subjects and n = 132 subjects), sensitivity was 100% (positive predictive value: 28.1% and 22.6%) and specificity was 48.9% and 80.8% (negative predictive value: 100.0%). CONCLUSIONS: Using a single exam, this model identified all infants who developed TR-ROP, on average more than one month prior to diagnosis, with moderate to high specificity. This approach could lead to earlier identification of incident severe ROP, reducing late diagnosis and treatment, while simultaneously reducing the number of ROP exams and unnecessary physiological stress for low-risk babies.

Purpose To describe characteristics and predictors of plus disease, and the accuracy of image grading for plus disease in the e-ROP Study. Design Secondary analyses of data from 13 North American centers. Participants Premature infants with birth weight (BW) Methods Infants underwent regularly scheduled diagnostic examinations by ophthalmologists and digital imaging by trained imagers using a wide-field digital camera. Two masked nonphysician trained readers independently evaluated images for posterior pole abnormality (normal, preplus, plus), with discrepancies adjudicated by a reading supervisor. Logistic regression models were used to determine predictors for plus disease. The sensitivity and specificity of image grading for plus disease were calculated using the clinical examination finding as reference standard. Main Outcome Measures Odds ratios (OR), sensitivity, and specificity. Results Among 1239 infants (mean BW 864 g, mean gestational age [GA] 27 weeks), 129 infants (10%) (226 eyes, 75% bilateral) had plus disease from clinical examination. When plus disease was first diagnosed in clinical examination at median postmenstrual age (PMA) of 36 weeks (range: 32–43 weeks), 94% to 96% of plus occurred in the superior or inferior temporal quadrant. Significant predictors for plus disease from multivariate analysis were as follows: GA (OR = 3.2 for ≤24 vs. ≥28 weeks, P = 0.004), race (OR = 2.4 for white vs. black, P = 0.01), respiratory support (OR = 7.1, P = 0.006), weight gain (OR = 1.5 for weight gain ≤12 vs. >18 g/day, P = 0.03), and image findings at the first image session, including presence of preplus/plus disease (OR = 2.7, P = 0.003), ROP stage (OR = 4.2 for stage 3 ROP vs. no ROP, P = 0.006), and blot hemorrhage (OR = 3.1, P = 0.003). These features predicted plus disease with an area under the receiver operating characteristic curve of 0.89 (95% confidence interval [CI]: 0.85–0.92). The image grading using preplus as the cut point had sensitivity of 94% (95% CI: 90%–97%) and specificity of 81% (95% CI: 79%–82%) for detecting plus disease in an eye. Conclusions Among e-ROP infants, plus disease developed in 10% of infants at a median PMA of 37 weeks, with the majority being bilateral and mostly in the superior or inferior temporal quadrant. GA, race, respiratory support, postnatal weight gain, image findings of the posterior pole, and ROP predict development of plus disease. Nonphysician image grading can detect almost all plus disease with good specificity.

To determine the symmetry of retinopathy of prematurity (ROP) between fellow eyes in a broad-risk cohort.A retrospective cohort study, the Postnatal Growth and ROP (G-ROP) Study, of 7483 infants undergoing ROP examinations conducted at 29 hospitals in the United States and Canada from 2006 to 2012. The main outcomes were the symmetry for the highest stage and the most severe type (1, 2, not 1 or 2, no ROP) of ROP and disease course of the fellow eye when only one eye developed type 1.93% of infants had eyes symmetric for the highest stage and 94% for type. Among 459 infants who developed type 1, 379 (82.6%) did so in both eyes simultaneously and were treated bilaterally; 44 (10%) were treated for type 1 in one eye and type 2 in the fellow eye; and 36 (8%) were treated unilaterally initially, of which 6 fellow eyes developed type 1 and were treated (4 within 2 weeks, all within 4 weeks); 5 developed type 2 and regressed; and 25 developed ROP less than type 1 or 2, which was treated in 13 cases and regressed spontaneously in 12 cases.ROP was highly symmetric between eyes with respect to the presence and severity of disease in a large, broad-risk cohort representative of infants undergoing ROP screening. When type 1 develops in one eye and type 2 in the fellow eye, the risk of progression to type 1 in the fellow eye appears very low if it has not occurred within 4 weeks.

Purpose Outbreaks of adenovirus in neonatal intensive care units (NICUs) can lead to widespread transmission and serious adverse outcomes. We describe the investigation, response, and successful containment of an adenovirus outbreak in a NICU associated with contaminated handheld ophthalmologic equipment used during retinopathy of prematurity (ROP) screening. Design Epidemiologic outbreak investigation. Participants A total of 23 hospitalized neonates, as well as NICU staff and parents of affected infants. Main Outcome Measures Routine surveillance identified an adenovirus outbreak in a level IV NICU in August 2016. Epidemiologic investigation followed, including chart review, staff interviews, and observations. Cases were defined as hospital-acquired adenovirus identified from any clinical specimen (NICU patient or employee) or compatible illness in a family member. Real-time polymerase chain reaction (PCR) and partial- and whole-genome sequencing assays were used for testing of clinical and environmental specimens. Results We identified 23 primary neonatal cases and 9 secondary cases (6 employees and 3 parents). All neonatal case-patients had respiratory symptoms. Of these, 5 developed pneumonia and 12 required increased respiratory support. Less than half (48%) had ocular symptoms. All neonatal case-patients (100%) had undergone a recent ophthalmologic examination, and 54% of neonates undergoing examinations developed adenovirus infection. All affected employees and parents had direct contact with infected neonates. Observations revealed inconsistent disinfection of bedside ophthalmologic equipment and limited glove use. Sampling of 2 handheld lenses and 2 indirect ophthalmoscopes revealed adenovirus serotype 3 DNA on each device. Sequence analysis of 16 neonatal cases, 2 employees, and 2 lenses showed that cases and equipment shared 100% identity across the entire adenovirus genome. Infection control interventions included strict hand hygiene, including glove use; isolation precautions; enhanced cleaning of lenses and ophthalmoscopes between all examinations; and staff furlough. We identified no cases of secondary transmission among neonates. Conclusions Adenovirus outbreaks can result from use of contaminated ophthalmologic equipment. Even equipment that does not directly contact patients can facilitate indirect transmission. Patient-to-patient transmission can be prevented with strict infection control measures and equipment cleaning. Ophthalmologists performing inpatient examinations should take measures to avoid adenoviral spread from contaminated handheld equipment.

To determine if visual acuity (VA) outcomes are comparable using the amblyopia treatment study HOTV protocol (ATS-HOTV) and electronic Early Treatment of Diabetic Retinopathy Study (E-ETDRS) protocol in children with optic pathway gliomas (OPGs).Children enrolled in a prospective study of OPGs were eligible if they completed both the ATS-HOTV and E-ETDRS during the same visit. The contribution of age, testing order, having neurofibromatosis type 1, visual field loss, and circumpapillary retinal nerve fiber layer thickness to VA difference were assessed using generalized estimating equations to account for the intereye correlation.Forty-eight children (median age, 10.3 years; range, 5.2-17.1 years; 49% female) met inclusion criteria and contributed 93 study eyes at their initial visit. Eleven patients (22 eyes) had more than one study visit, permitting longitudinal evaluation. ATS-HOTV measures of VA were higher than E-ETDRS at the initial (0.13 ± 0.36 vs. 0.23 ± 0.39 logarithm of the minimum angle of resolution [logMAR], P0.001) and all visits (0.13 ± 0.34 vs. 0.21 ± 0.36 logMAR, P0.001). VA remained significantly higher with ATS-HOTV regardless of test order, but the mean difference between tests was most profound when tested with ATS-HOTV first compared to E-ETDRS first (P0.001).VA results differ significantly between the ATS-HOTV and E-ETDRS testing methods in children with OPGs. Given the wide range of ages and testing ability of children, one VA testing method should be used throughout longitudinal OPG clinical trials.It is imperative that age-appropriate VA testing methods are standardized across all pediatric OPG clinical trials.

To report findings of a telemedicine retinopathy of prematurity (ROP) screening program in six neonatal units in rural areas of Guatemala, using a portable, noncontact, 40° field digital fundus camera (Pictor Plus) operated by trained technicians.National ROP Program Guidelines screening criteria were used: gestational age36 weeks and/or birth weight (BW)2000 g, or GA36 weeks but BW ≥2000 g, with qualifying medical history. Retinal images were obtained by two technicians and graded by ophthalmologists experienced in ROP. Infants with signs of pre-plus or plus disease in one or both eyes were referred for clinical examination. Screening was stopped when retinal vessels in anterior zone II were normal on two successive evaluations or the infant had reached 45 week's postmenstrual age.A total of 418 of 1,890 eligible infants (22.1%) were screened. Mean GA was 33.9 ± 2.2 weeks (range, 27-36), and mean BW 1728.3 ± 379.3 g (range, 840-2830 g). Thirty-three infants (8.6%) developed plus or pre-plus disease, and 19 (58%) underwent ophthalmologic examination. Fifteen infants were confirmed with type 1 ROP, and 14 were treated. Mean GA of treated infants treated was 33.6 ± 3.0 weeks (range, 32-34.9), and mean BW was 1,646 ± 245.8 g (range, 1100-1774.1 g).Imaging with a noncontact fundus camera can facilitate detection of treatable ROP in countries with limited resources. Strengthening the health systems, including motivation and continued training of neonatal intensive care personnel is essential to improve and maintain program effectiveness. Reasons for, and interventions to address the low uptake of screening need to be explored to extend coverage of ROP screening to district hospitals in Guatemala.

To evaluate the effect of changes in institutional peripheral oxygen saturation (SpOThis study was a secondary analysis of data from 21 North American hospitals during the periods 2006-2012 and 2015-2017. Hospitals were categorized based on whether or not SpOA total of 8,142 infants underwent ROP examinations at 21 hospitals during the two study periods: 5,716 in 2006-2012 (mean BW, 1109 ± 369 g; GA, 28.0 ± 2.6; 11.6% severe ROP); 2,426 in 2015-2017 (mean BW, 1086 ± 347 g; GA, 28.0 ± 2.8; 12.8% severe ROP). Fourteen hospitals increased SpOIncreases in institutional SpO

To evaluate the sensitivity and specificity of single, independent, nonphysician trained reader (TR) gradings in the Telemedicine Approaches to Evaluating Acute-phase Retinopathy of Prematurity (e-ROP) study.Secondary analyses of image grading results from 1,235 infants of birth weights1251 g. Two of three TRs independently graded image sets; discrepancies were adjudicated by the reading center director (an ophthalmologist) to reach final grading. Sensitivity and specificity of each TR grading and final grading was calculated by comparing gradings to clinical examination results.Of 7,808 double graded image sets, TR1 graded 5,165; TR2, 3,787; and TR3, 6,664. Compared to final grading for referral warranted retinopathy of prematurity (RW-ROP), two TRs had relatively lower sensitivity (TR1, 75% vs 79% [P = 0.03]; TR2, 73% vs 77% [P = 0.02]) and specificity (TR1, 80% vs 83% [P 0.001]; TR2, 82% vs 83% [P = 0.09]). TR3 had similar sensitivity (83% vs 83% [P = 0.78]) and specificity (83% vs 84% [P = 0.02]). Compared to final grading, TR1 had lower sensitivity for zone I ROP (47% vs 56% [P = 0.04]) and stage ≥3 ROP (71% vs 77% [P = 0.002]); TR2 had lower sensitivity for stage ≥3 ROP (69% vs 77% [P 0.001]) and lower specificity for all three components (P 0.001); TR3 had lower sensitivity for detecting plus disease (23% vs 35% [P 0.001]) and similar sensitivity for zone I ROP and stage ≥3 ROP.There is a small but significant decrease in the sensitivity and specificity for RW-ROP when single-reader grading is compared to double adjudicated grading.

SIGNIFICANCEAmong 4- and 5-year-old children, deficits in measures of attention, visual-motor integration (VMI) and visual perception (VP) are associated with moderate, uncorrected hyperopia (3 to 6 diopters [D]) accompanied by reduced near visual function (near visual acuity worse than 20/40 or ste

Purpose To determine the symmetry on retinal image grading of fellow eyes for retinopathy of prematurity (ROP) features (stage, zone, and plus disease) and severity, to provide the basis for the within-subject comparison for ROP trials. Design Secondary analyses of data from the Telemedicine Approaches to Evaluating of Acute-Phase ROP (e-ROP) Study. Subjects Infants with birth weight less than 1251 g. Methods Infants underwent serial retinal imaging sessions in both eyes by certified imagers starting at 32 weeks' postmenstrual age. Two trained nonphysician readers graded each eye independently for ROP features in a 5-retinal-image set from each session. Discrepancies were adjudicated by a reading center supervisor. Readers were masked to all eye examination results, previous gradings of both eyes, current grading of the fellow eye, and demographic data. Main Outcome Measures The intereye agreement assessed using percentage exact agreement and weighted kappa ( K w ) for stage, zone, plus, referral-warranted ROP (RW-ROP; defined as the presence of stage 3 or above, plus disease, or zone I ROP), and severity of ROP. Results Among 3918 imaging sessions in 1235 infants, the percent agreement ( K w ) between paired eyes was 75.3% (0.65) for stage of ROP, 82.3% (0.68) for zone of ROP, 78.7% (0.51) for plus disease, 84.7% (0.56) for RW-ROP, and 72.7% (0.63) for severity of ROP. Similar intereye agreements were found when considering ROP features at the first imaging session, at the last imaging session, and at any imaging session. Based on image evaluations from all sessions, 412 infants (33.4%) had ROP stage 3 or above, 148 (12.0%) had zone I ROP, 70 (5.7%) had plus disease, and 419 (33.9%) had RW-ROP in 1 or both eyes; symmetrical findings were present in 71.4% for ROP stage 3 or above, 56.8% for zone I ROP, 50.0% for plus disease, and 73.7% for RW-ROP. Conclusions Image evaluations by masked, trained readers showed good intereye agreement in ROP characteristics, consistent with the high intereye agreement in ROP from clinical examinations by ophthalmologists in other studies. These data confirm that acute ROP is typically symmetrical and supports within-subject comparison in ROP trials.

Purpose To determine the accuracy of identifying referral-warranted retinopathy of prematurity (RW-ROP, defined as any zone I ROP, stage 3 or worse, or plus disease) from retinal image sets using three grading protocols: a single optic disk–centered image, a set of 3 horizontal images, and a 5-image set. Methods In this secondary analysis of images from the e-ROP study, a weighted sample of 250 image sets from 250 infants (125 with RW-ROP and 125 without RW-ROP) was randomly selected. The sensitivities and specificities for detecting RW-ROP and its components from a single disk center image, along with nasal and temporal retinal images, were calculated and compared with the e-ROP grading of RW-ROP of all 5 retinal images (disk center and nasal, temporal, superior, and inferior retinal images). Results RW-ROP was identified with a sensitivity of 11.2% (95% CI, 6.79%-17.9%) using a single disk center image, with a sensitivity of 70.4% (95% CI, 61.9%-77.9%) using 3 horizontal images, and a statistically higher sensitivity of 82.4% (95% CI, 75.0%-89.0%) using all 5 images ( P = 0.002). The specificities were 100%, 86.4%, and 90.4%, respectively. For grading using 3 horizontal images, sensitivity was 14.3% for plus disease, 25% for zone I ROP, and 71.2% for stage 3 or worse compared to 40.8%, 50%, and 79.8% for grading using 5-image sets, respectively. Conclusions Both a single, disk-centered, posterior pole image and 3 horizontal images were less effective than a 5-image set in determining the presence of RW-ROP on qualitative grading by trained readers.

Purpose To describe the clinical characteristics of intraocular hemorrhages (IOHs) in infants in the Telemedicine Approaches to Evaluating Acute-Phase Retinopathy of Prematurity (e-ROP) Study and to evaluate their potential use for prediction of disease severity. Design Secondary data analysis from a prospective study. Participants Preterm infants with birth weight (BW) ≤1250 g. Methods Infants underwent serial digital retinal imaging in both eyes starting at 32 weeks' postmenstrual age. Nonphysician trained readers (TRs) evaluated all image sets from eyes that ever had IOHs documented on image evaluation or eye examination for the presence, location, type, area, and relation of the IOH to the junction between vascularized and avascular retina. Associations of IOH with demographic and neonatal factors, and with the presence and severity of retinopathy of prematurity (ROP) were investigated by univariate and multivariate analyses. Sensitivity and specificity of the telemedicine system for detecting referral-warranted ROP (RW-ROP) were calculated with and without incorporating hemorrhage into the standardized grading protocol. Main Outcome Measures Retinal and vitreous hemorrhage. Results Among 1239 infants (mean [standard deviation] BW = 864 [212] g; gestational age [GA] = 27 [2.2] weeks) who underwent an average of 3.2 imaging sessions, 22% had an IOH in an eye on at least 1 of the e-ROP visits. Classification of IOH was preretinal (57%), blot (57%), dot (38%), flame-shaped (16%), and vitreous (8%); most IOHs were unilateral (70%). The IOH resolved in 35% of eyes by the next imaging session and in the majority (76%) of cases by 8 weeks after initial detection. Presence of IOH was inversely associated with BW and GA and significantly associated ( P Conclusions Approximately 1 in 5 preterm infants examined had IOHs, generally unilateral. The presence of hemorrhage was directly correlated with both presence and severity of ROP and inversely correlated with BW and GA, although including hemorrhage in the grading algorithm only minimally improved the sensitivity of the telemedicine system to detect RW-ROP.

Until advances in retinal imaging over recent decades, detection of retinopathy of prematurity requiring treatment depended largely on indirect ophthalmoscopy by ophthalmologists with experience in examination of low birth weight, often fragile infants. Recent studies have added to our understanding of the pros and cons of using this technology to provide timely care for the increasing number of infants at risk for ROP blindness worldwide.

Key Points Question Can deep learning algorithms achieve a performance comparable with that of ophthalmologists on multidimensional identification of retinopathy of prematurity (ROP) using wide-field retinal images? Findings In this diagnostic study of 14 108 eyes of 8652 preterm infants, a deep learning–based ROP screening platform could identify retinal images using 5 classifiers, including image quality, stages of ROP, intraocular hemorrhage, preplus/plus disease, and posterior retina. The platform achieved an area under the curve of 0.983 to 0.998, and the referral system achieved an area under the curve of 0.9901 to 0.9956; the platform achieved a Cohen κ of 0.86 to 0.98 compared with 0.93 to 0.98 by the ROP experts. Meaning Results suggest that a deep learning platform could identify and classify multidimensional ROP pathological lesions in retinal images with high accuracy and could be suitable for routine ROP screening in general and children’s hospitals., This diagnostic study uses data from the eye examinations of preterm infants to assess whether a deep learning algorithm can perform at a level comparable with that of ophthalmologists in identifying retinopathy of prematurity on wide-field retinal images., Importance A retinopathy of prematurity (ROP) diagnosis currently relies on indirect ophthalmoscopy assessed by experienced ophthalmologists. A deep learning algorithm based on retinal images may facilitate early detection and timely treatment of ROP to improve visual outcomes. Objective To develop a retinal image–based, multidimensional, automated, deep learning platform for ROP screening and validate its performance accuracy. Design, Setting, and Participants A total of 14 108 eyes of 8652 preterm infants who received ROP screening from 4 centers from November 4, 2010, to November 14, 2019, were included, and a total of 52 249 retinal images were randomly split into training, validation, and test sets. Four main dimensional independent classifiers were developed, including image quality, any stage of ROP, intraocular hemorrhage, and preplus/plus disease. Referral-warranted ROP was automatically generated by integrating the results of 4 classifiers at the image, eye, and patient levels. DeepSHAP, a method based on DeepLIFT and Shapley values (solution concepts in cooperative game theory), was adopted as the heat map technology to explain the predictions. The performance of the platform was further validated as compared with that of the experienced ROP experts. Data were analyzed from February 12, 2020, to June 24, 2020. Exposure A deep learning algorithm. Main Outcomes and Measures The performance of each classifier included true negative, false positive, false negative, true positive, F1 score, sensitivity, specificity, receiver operating characteristic, area under curve (AUC), and Cohen unweighted κ. Results A total of 14 108 eyes of 8652 preterm infants (mean [SD] gestational age, 32.9 [3.1] weeks; 4818 boys [60.4%] of 7973 with known sex) received ROP screening. The performance of all classifiers achieved an F1 score of 0.718 to 0.981, a sensitivity of 0.918 to 0.982, a specificity of 0.949 to 0.992, and an AUC of 0.983 to 0.998, whereas that of the referral system achieved an F1 score of 0.898 to 0.956, a sensitivity of 0.981 to 0.986, a specificity of 0.939 to 0.974, and an AUC of 0.9901 to 0.9956. Fine-grained and class-discriminative heat maps were generated by DeepSHAP in real time. The platform achieved a Cohen unweighted κ of 0.86 to 0.98 compared with a Cohen κ of 0.93 to 0.98 by the ROP experts. Conclusions and Relevance In this diagnostic study, an automated ROP screening platform was able to identify and classify multidimensional pathologic lesions in the retinal images. This platform may be able to assist routine ROP screening in general and children hospitals.

Multicenter studies addressing screening and intervention for retinopathy of prematurity (ROP) inform the management guidelines jointly recommended by American pediatric and ophthalmology academies. Current research focuses on improvements in the identification of ROP and in the treatment of high-risk disease. The development of digital image technology may address various challenges in the diagnosis of ROP, including availability of pediatric ophthalmologic expertise, interobserver variation in diagnosis, and inherent limitations in visual diagnosis. Improved clinical prediction models based on nonophthalmologic data may complement examination-based ROP diagnosis. Alternatives to retinal ablation therapy are being studied to decrease the associated morbidities of such therapy.

Telemedicine for the detection of retinopathy of prematurity (ROP) is becoming increasingly common; however, obtaining the required multiple retinal images from an infant can be challenging. This secondary analysis from the Telemedicine Approaches to Evaluating Acute-Phase Retinopathy of Prematurity (e-ROP) study evaluated the detection of referral-warranted ROP (RW-ROP) by trained readers when a full set of 5 retinal images could not be obtained.A total of 7,905 image sets from 1,257 infants in the study were evaluated. Retinal location of images and image quality were recorded. Sensitivity and specificity of RW-ROP detection by trained readers were calculated by comparing findings in incomplete image sets to the findings on standard eye examination.The majority of image sets contained all 5 retinal images (92.8%). The disk center view was the image most likely to be present and to be of acceptable image quality (96.8%). The nasal retina was the most difficult to obtain with acceptable image quality (83.4%). Sensitivity of detection of RW-ROP was 82.1% when 5 retinal images of acceptable quality were submitted for grading, 67.2% when 4 acceptable images were submitted, and 66.7% for 3 or fewer acceptable images (P = 0.02), with corresponding specificity of 82.2%, 89.0%, and 81.7% respectively (P0.0001). When images of any quality were evaluated, sensitivity was not increased (P = 0.74).The likelihood of detecting RW-ROP by telemedicine screening is decreased when a full set of retinal images is not obtained.

Purpose To develop a risk score for predicting treatment-requiring retinopathy of prematurity (TR-ROP) in the Telemedicine Approaches to Evaluating Acute-Phase Retinopathy of Prematurity (e-ROP) study. Design Second analyses of an observational cohort study. Participants Infants with birth weight (BW) Methods Nonphysician trained readers evaluated wide-field retinal image sets for characteristics of ROP, pre-plus/plus disease, and retinal hemorrhage. Risk score points for predicting TR-ROP were derived from the regression coefficients of significant predictors in a multivariate logistic regression model. Main Outcome Measures TR-ROP. Results Eighty-five of 771 infants (11.0%) developed TR-ROP. In a multivariate model, significant predictors for TR-ROP were gestational age (GA) (odds ratio [OR], 5.7; 95% confidence interval [CI], 1.7–18.9 for ≤25 vs. ≥28 weeks), need for respiratory support (OR, 7.0; 95% CI, 1.3–37.1 for high-frequency oscillatory ventilation vs. no respiratory support), slow weight gain (OR, 2.4; 95% CI, 1.2–4.6 for weight gain ≤12 g/day vs. >15 g/day), and image findings at the first image session including number of quadrants with pre-plus (OR, 3.8; 95% CI, 1.5–9.7 for 4 pre-plus quadrants vs. no pre-plus), stage and zone of ROP (OR, 4.7; 95% CI, 2.1–11.8 for stage 1–2 zone I, OR, 5.9; 95% CI, 2.1–16.6 for stage 3 zone I vs. no ROP), and presence of blot hemorrhage (OR, 3.1; 95% CI, 1.4–6.7). Image findings predicted TR-ROP better than GA (area under receiver operating characteristic curve [AUC] = 0.82 vs. 0.75, P = 0.03). The risk of TR-ROP steadily increased with higher risk score and predicted TR-ROP well (AUC = 0.88; 95% CI, 0.85–0.92). Risk score ≥3 points for predicting TR-ROP had a sensitivity of 98.8%, specificity of 40.1%, and positive and negative predictive values of 17.0% and 99.6%, respectively. Conclusions Image characteristics at 34 PMA weeks or earlier independently predict TR-ROP. If externally validated in other infants, risk score, calculated from image findings, GA, weight gain, and respiratory support, enables early identification of infants in need of increased surveillance for TR-ROP.

Purpose To compare infant and retinopathy of prematurity (ROP) characteristics from 3 clinical studies conducted over a 27-year period in the United States. Design Secondary analysis of results of 3 clinical studies. Participants Infants with birth weight (BW) Methods Analysis of data from the Cryotherapy for Retinopathy of Prematurity (CRYO-ROP) and Early Treatment for Retinopathy of Prematurity (ETROP) trials and the primary data from the Telemedicine Approaches for the Evaluation of Acute-Phase Retinopathy of Prematurity (e-ROP) study. Main Outcome Measures Infant characteristics and onset, severity, and time course of ROP. Results Across the 3 studies, mean (standard deviation) BW and mean gestational age (GA) decreased over time from CRYO-ROP (954 g [185 g], 27.9 weeks [2.2 weeks]) to ETROP (907 g [205 g], 27.4 weeks [2.2 weeks]) to e-ROP (864 g [212 g], 27.0 weeks [2.2 weeks]), with an increase in the percentage of infants enrolled weighing P P = 0.003). Moderately severe ROP (defined as prethreshold or referral warranted) varied (17.8% CRYO, 12.3% ETROP, 19.4% e-ROP; P Conclusions The BW and GA of infants enrolled in ROP studies in the United States have decreased over the past 27 years, whereas ROP prevalence and onset of disease are stable.

Purpose To compare early literacy of 4- and 5-year-old uncorrected hyperopic children with that of emmetropic children. Design Cross-sectional. Participants Children attending preschool or kindergarten who had not previously worn refractive correction. Methods Cycloplegic refraction was used to identify hyperopia (≥3.0 to ≤6.0 diopters [D] in most hyperopic meridian of at least 1 eye, astigmatism ≤1.5 D, anisometropia ≤1.0 D) or emmetropia (hyperopia ≤1.0 D; astigmatism, anisometropia, and myopia Main Outcome Measures Trained examiners administered the Test of Preschool Early Literacy (TOPEL), composed of Print Knowledge, Definitional Vocabulary, and Phonological Awareness subtests. Results A total of 492 children (244 hyperopes and 248 emmetropes) participated (mean age, 58 months; mean ± standard deviation of the most hyperopic meridian, +3.78±0.81 D in hyperopes and +0.51±0.48 D in emmetropes). After adjustment for age, race/ethnicity, and parent/caregiver's education, the mean difference between hyperopes and emmetropes was −4.3 ( P = 0.01) for TOPEL overall, −2.4 ( P = 0.007) for Print Knowledge, −1.6 ( P = 0.07) for Definitional Vocabulary, and −0.3 ( P = 0.39) for Phonological Awareness. Greater deficits in TOPEL scores were observed in hyperopic children with ≥4.0 D than in emmetropes (−6.8, P = 0.01 for total score; −4.0, P = 0.003 for Print Knowledge). The largest deficits in TOPEL scores were observed in hyperopic children with binocular near VA of 20/40 or worse (−8.5, P = 0.002 for total score; −4.5, P = 0.001 for Print Knowledge; −3.1, P = 0.04 for Definitional Vocabulary) or near stereoacuity of 240 seconds of arc or worse (−8.6, P P Conclusions Uncorrected hyperopia ≥4.0 D or hyperopia ≥3.0 to ≤6.0 D associated with reduced binocular near VA (20/40 or worse) or reduced near stereoacuity (240 seconds of arc or worse) in 4- and 5-year-old children enrolled in preschool or kindergarten is associated with significantly worse performance on a test of early literacy.

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows: To compare treatment with insulin‐like growth factor‐1 to no treatment or placebo for the prevention of retinopathy of prematurity or treatment of early retinopathy of prematurity.

Importance The current guidelines for retinopathy of prematurity (ROP) detection programs in the United States include a range of birth weights (BWs) and gestational ages and likely require examinations of many premature infants who are at low risk for developing serious retinopathy. Objective To determine the incidence, onset, and early course of ROP in what to our knowledge is the largest cohort to date that is representative of infants who are undergoing ROP screening. Design, Setting, and Participants This secondary analysis of data from the Postnatal Growth and Retinopathy of Prematurity (G-ROP) retrospective cohort study was conducted in 29 hospitals in the United States and Canada between January 2006 and December 2011 and included 7483 infants who underwent serial ROP examinations. Main Outcomes and Measures Most severe ROP in either eye, classified as no ROP, mild ROP, type 2 ROP, or type 1 ROP (per Early Treatment for ROP Study criteria). Onset at postmenstrual age for zone I disease and stage of ROP, plus disease, and treatment. Results This study included 7483 infants with a mean (SD) BW of 1099 (259) g and a mean (SD) gestational age of 28 (3) weeks who underwent ROP examinations. Of these, 3224 infants (43.1%) developed ROP, 459 (6.1%) developed type 1 and 472 (6.3%) type 2 ROP, 514 (6.9%) underwent treatment in 1 or both eyes, and 147 (2%) had zone I disease. Additionally, 98.1% of type 1 or 2 ROP cases occurred in infants with a BW of less than 1251 g. Only about half of the eyes (49.4%) were vascularized into zone III by 37 weeks postmenstrual age. Conclusions and Relevance These findings add to our knowledge of ROP screening as they include all eligible infants, not just high-risk infants as in previous studies. More than 40% of at-risk premature infants develop some stage of ROP, and most retinopathy regresses without treatment. However, approximately 12.5% develop severe ROP, which occurs almost exclusively among infants with a BW of less than 1251 g.

Retinopathy of prematurity (ROP) is a largely avoidable cause of blindness in children worldwide, requiring high-quality neonatal care, early detection and treatment. In middle-income countries throughout Latin America, Eastern Europe and South Asia, there has been a rise in ROP blindness due to a combination of increased survival of preterm infants, resource-scarce medical environments and lack of policies, training and human resources. However, Argentina is an example of country where rates of ROP blindness have declined and ROP programmes have been successfully and effectively embedded within the health and legal system. The purpose of this study is to describe the activities and stakeholders, including Ministry of Health (MoH) and UNICEF, involved in the process, from recognition of an epidemic of ROP blindness to the development of national guidelines, policies and legislation for control. Using a retrospective mixed methods case study design, data on rates of severe ROP was collected from 13 neonatal intensive care units from 1999 to 2012, and on the proportion of children blind from ROP in nine blind schools in seven provinces. Legislative document review, focus group discussions and key informant interviews were conducted with neonatologists, ophthalmologists, neonatal nurses, parents, MoH officials, clinical societies, legislators and UNICEF officials in seven provinces. Results are presented combining the stages heuristic policy framework and Shiffman including: agenda setting, policy formulation, implementation and evaluation. By 2012, ROP had declined as a cause of blindness in children in schools for the blind as had rates of severe ROP needing treatment in the NICUs visited. Multiple factors played a role in reducing blindness from ROP in Argentina and successfully coordinating its control including national advocacy, leadership, legislation and international collaboration. Lessons learned in Argentina can potentially be scaled to other LMICs in Latin America and beyond with further context-specific research.

BACKGROUND: An association between thrombocytopenia and retinopathy of prematurity (ROP) has been suggested but not been studied longitudinally. We sought to identify the time period in postnatal development during which thrombocytopenia and the subsequent development of severe ROP are associated. METHODS: This was a retrospective case–control study of 100 subjects who received laser photocoagulation for type 1 ROP between 2005 and 2009 and 100 controls with no ROP or only stage 1 ROP. The proportions of infants with thrombocytopenia, defined as a serum platelet level of

Supplemental digital content is available in the text., Purpose This article provides recommendations for screening children aged 36 to younger than 72 months for eye and visual system disorders. The recommendations were developed by the National Expert Panel to the National Center for Children’s Vision and Eye Health, sponsored by Prevent Blindness, and funded by the Maternal and Child Health Bureau of the Health Resources and Services Administration, United States Department of Health and Human Services. The recommendations describe both best and acceptable practice standards. Targeted vision disorders for screening are primarily amblyopia, strabismus, significant refractive error, and associated risk factors. The recommended screening tests are intended for use by lay screeners, nurses, and other personnel who screen children in educational, community, public health, or primary health care settings. Characteristics of children who should be examined by an optometrist or ophthalmologist rather than undergo vision screening are also described. Results There are two current best practice vision screening methods for children aged 36 to younger than 72 months: (1) monocular visual acuity testing using single HOTV letters or LEA Symbols surrounded by crowding bars at a 5-ft (1.5 m) test distance, with the child responding by either matching or naming, or (2) instrument-based testing using the Retinomax autorefractor or the SureSight Vision Screener with the Vision in Preschoolers Study data software installed (version 2.24 or 2.25 set to minus cylinder form). Using the Plusoptix Photoscreener is acceptable practice, as is adding stereoacuity testing using the PASS (Preschool Assessment of Stereopsis with a Smile) stereotest as a supplemental procedure to visual acuity testing or autorefraction. Conclusions The National Expert Panel recommends that children aged 36 to younger than 72 months be screened annually (best practice) or at least once (accepted minimum standard) using one of the best practice approaches. Technological updates will be maintained at http://nationalcenter.preventblindness.org.

Graham E Quinn Division of Ophthalmology, The Children’s Hospital of Philadelphia, Wood Center, Philadelphia, PA, USA Abstract: Blindness due to retinopathy of prematurity (ROP) is an increasing problem worldwide as improved levels of neonatal care are provided in countries with developing neonatal intensive care units. The occurrence of ROP blindness varies dramatically with the socioeconomic development of a country. In regions with high levels of neonatal care and adequate resources, ROP blindness is largely restricted to premature infants with very low birth weight and low gestational age while in middle- and low-income countries with regional variation in technology and capacity, limited health resources may well limit the care of the premature newborn. Keywords: ROP, international, blindness

Importance The first Postnatal Growth and Retinopathy of Prematurity Study (G-ROP-1) developed new screening criteria with 100% sensitivity for type 1 retinopathy of prematurity (ROP) and 30% reduction of infants requiring examinations in a retrospective development cohort of 7483 infants from 29 North American hospitals in 2006-2012. Infants meeting 1 or more of the following criteria undergo examinations: gestational age less than 28 weeks or birth weight less than 1051 g; weight gain less than 120 g during age 10 to 19 days, weight gain less than 180 g during age 20 to 29 days, or weight gain less than 170 g during age 30 to 39 days; or hydrocephalus. Objective To evaluate the generalizability of the G-ROP screening criteria in a new cohort of at-risk infants. Design, Setting, and Participants This prospective validation cohort study (G-ROP-2) was conducted at 41 hospitals in the United States and Canada (25 G-ROP-1 hospitals and 16 new hospitals) from September 8, 2015, to June 13, 2017, among 3981 premature infants at risk for ROP and with known ROP outcomes. Main Outcomes and Measures Sensitivity for Early Treatment for Retinopathy of Prematurity Study type 1 ROP and potential reduction in infants receiving examinations. Results Among the 3981 infants in the study (1878 girls and 2103 boys; median gestational age, 28 weeks [range, 22-35 weeks]; median birth weight, 1072 g [range, 350-4080 g]; 1966 white; 942 black; 321 Latino; 120 Asian; 22 Native Hawaian or Pacific Islander; and 25 American Indian or Alaskan Native), the G-ROP criteria correctly predicted 219 of 219 cases of type 1 ROP (sensitivity, 100%; 95% CI, 98.3%-100%), while reducing the number of infants undergoing examinations by 35.6% (n = 1418). In a combined G-ROP-1 and G-ROP-2 cohort of 11 463 infants, the G-ROP criteria predicted 677 of 677 cases of type 1 ROP (sensitivity, 100%; 95% CI, 99.4%-100%), reducing the number of infants receiving examinations by 32.5% (n = 3730), while current criteria (birth weight Conclusions and Relevance This study found that the G-ROP screening criteria were generalizable on validation and, if used clinically in the United States and Canada, could reduce the number of infants receiving examinations. The large G-ROP cohorts provide evidence-based screening criteria that have higher sensitivity and higher specificity (fewer infants receiving examinations) for type 1 ROP than currently recommended guidelines.

Purpose To compare the structural outcome at 9 months of eyes treated with intravitreal injection of bevacizumab with fellow eyes treated with conventional laser photoablation in zone I type 1 retinopathy of prematurity (ROP). Design Single randomized controlled trial. Participants All inborn babies with type 1 zone I ROP at a single institution were included in the study. One eye was randomized to receive an intravitreal injection of 0.5 mg bevacizumab; the fellow eye received conventional laser photoablation. Methods Digital fundus photographs and fluorescein angiography (FA) using the RetCam (Clarity Medical Systems Inc., Pleasanton, CA) were performed before treatment and 9 months after treatment. Main Outcome Measures Presence of retinal and choroidal abnormalities on FA at 9 months. Results Thirteen infants were enrolled; 1 died 3 months after birth. One laser-treated eye progressed to stage 5 retinal detachment. The remaining 23 eyes had favorable structural results at the 9-month follow-up and provided FA results. At 9 months of age, all eyes treated with a bevacizumab injection were noted to have abnormalities at the periphery (large avascular area, abnormal branching, shunt) or the posterior pole (hyperfluorescent lesion, absence of foveal avascular zone). These posterior and peripheral lesions were not observed in the majority of the lasered eyes. Conclusions This study documents significant vascular and macular abnormalities of eyes in the bevacizumab group. Long-lasting implications of these abnormalities for visual function of the child need to be studied.

PURPOSE Detecting sight-threatening retinopathy of prematurity (ROP) relies on a diagnostic examination (DE) performed by an experienced ophthalmologist. An alternative may be a telemedicine system where retinal images of at-risk infants are graded by readers to determine features of ROP indicating the need for a DE. METHODS The multicenter Telemedicine Approaches to Evaluating Acute-phase ROP (e-ROP) Study is a cohort study of 2000 infants with birth weights

To determine demographic and refractive risk factors for astigmatism in the Vision in Preschoolers Study.Three- to 5-year-old Head Start preschoolers (N = 4040) from five clinical centers underwent comprehensive eye examinations by study-certified optometrists and ophthalmologists, including monocular visual acuity testing, cover testing, and cycloplegic retinoscopy. Astigmatism was defined as the presence of greater than or equal to +1.5 diopters (D) cylinder in either eye, measured with cycloplegic refraction. The associations of risk factors with astigmatism were evaluated using the odds ratio (OR) and its 95% confidence interval (CI) from logistic regression models.Among 4040 Vision in Preschoolers Study participants overrepresenting children with vision disorders, 687 (17%) had astigmatism, and most (83.8%) had with-the-rule astigmatism. In multivariate analyses, African American (OR, 1.65; 95% CI, 1.22 to 2.24), Hispanic (OR, 2.25; 95% CI, 1.62 to 3.12), and Asian (OR, 1.76; 95% CI, 1.06 to 2.93) children were more likely to have astigmatism than non-Hispanic white children, whereas American Indian children were less likely to have astigmatism than Hispanic, African American, and Asian children (p0.0001). Refractive error was associated with astigmatism in a nonlinear manner, with an OR of 4.50 (95% CI, 3.00 to 6.76) for myopia (≤-1.0 D in spherical equivalent) and 1.55 (95% CI, 1.29 to 1.86) for hyperopia (≥+2.0 D) when compared with children without refractive error (-1.0 D,+2.0 D). There was a trend of an increasing percentage of astigmatism among older children (linear trend p = 0.06). The analysis for risk factors of with-the-rule astigmatism provided similar results.Among Head Start preschoolers, Hispanic, African American, and Asian race as well as myopic and hyperopic refractive error were associated with an increased risk of astigmatism, consistent with findings from the population-based Multi-ethnic Pediatric Eye Disease Study and the Baltimore Pediatric Eye Disease Study. American Indian children had lower risk of astigmatism.

Purpose—To determine the inter-tester agreement of refractive error measurements between lay and nurse screeners using the Retinomax Autorefractor (Retinomax) and the SureSight Vision Screener (SureSight). Methods—Trained lay and nurse screeners measured refractive error in 1452 preschoolers (3- to 5-years old) using the Retinomax and the SureSight in a random order for screeners and instruments. Inter-tester agreement between lay and nurse screeners was assessed for sphere, cylinder and spherical equivalent (SE) using the mean difference and the 95% limits of agreement. The mean inter-tester difference (lay minus nurse) was compared between groups defined based on child’s age, cycloplegic refractive error, and the reading’s confidence number using analysis of variance. The limits of agreement were compared between groups using the Brown-Forsythe test. Inter-eye correlation was accounted for in all analyses. Results—The mean inter-tester differences (95% limits of agreement) were −0.04 (−1.63, 1.54) Diopter (D) sphere, 0.00 (−0.52, 0.51) D cylinder, and −0.04 (1.65, 1.56) D SE for the Retinomax; and 0.05 (−1.48, 1.58) D sphere, 0.01 (−0.58, 0.60) D cylinder, and 0.06 (−1.45, 1.57) D SE for the SureSight. For either instrument, the mean inter-tester differences in sphere and SE did not differ by the child’s age, cycloplegic refractive error, or the reading’s confidence number. However, for both instruments, the limits of agreement were wider when eyes had significant refractive error or the reading’s confidence number was below the manufacturer’s recommended value. Conclusions—Among Head Start preschool children, trained lay and nurse screeners agree well in measuring refractive error using the Retinomax or the SureSight. Both instruments had similar inter-tester agreement in refractive error measurements independent of the child’s age. Significant refractive error and a reading with low confidence number were associated with worse inter-tester agreement.