Predictive algorithms for early detection of retinopathy of prematurity

Neonatal Sepsis Episodes and Retinopathy of Prematurity in Very Preterm Infants

Importance

Retinopathy of prematurity (ROP) is a major morbidity of preterm infants causing visual impairment, including blindness, for which timely treatment is vital and prevention is key. Increasing evidence suggests that exposure to neonatal sepsis contributes to ROP development.

Objective

To investigate the association between neonatal sepsis and ROP in 2 large-scale cohorts of preterm infants born at less than 29 weeks' gestation.

Design, Setting, and Participants

This retrospective cohort study was conducted using data from the German Neonatal Network (GNN) and Norwegian Neonatal Network (NNN). The GNN involves 68 and the NNN includes 21 level III neonatal intensive care units. Participants were infants born at a gestation of 22 weeks and 0 days to 28 weeks and 6 days and enrolled in the GNN between January 1, 2009, and December 31, 2022, and NNN between January 1, 2009, and December 31, 2018. Data were analyzed from February through September 2023.

Exposure

Single or multiple episodes of culture-proven sepsis.

Main Outcomes and Measures

Any ROP and treatment-warranted ROP.

Results

Among 12 794 infants in the GNN (6043 female [47.2%] and 6751 male [52.8%]; mean [SD] gestational age, 26.4 [1.5] weeks) and 1844 infants in the NNN (866 female [47.0%] and 978 male [53.0%]; mean [SD] gestational age, 25.6 [1.5] weeks), the mean (SD) birth weight was 848 (229) g and 807 (215) g, respectively. Any ROP was present in 6370 infants (49.8%) in GNN and 620 infants (33.6%) in NNN, and treatment-warranted ROP was present in 840 infants (6.6%) in GNN and 140 infants (7.6%) in NNN. In both cohorts, there were increasing rates of treatment-warranted ROP with each sepsis episode (no sepsis: 572 of 10 658 infants [5.4%] in GNN and 85 of 1492 infants (5.7%) in NNN; 1 episode: 190 of 1738 infants in GNN [10.9%] and 29 of 293 infants [9.9%] in NNN; 2 episodes: 53 of 314 infants in GNN [16.9%] and 13 of 49 infants [26.5%] in NNN; 3 episodes: 25 of 84 infants [29.8%] in GNN and 3 of 10 infants [30.0%] in NNN). After adjusting for multiple confounders in the GNN dataset, the number of sepsis episodes was associated with ROP and treatment-warranted ROP compared with 0 episodes (1 episode: adjusted odds ratio [aOR], 1.44 [95% CI, 1.27-1.63]; P < .001 and OR, 1.60 [95% CI, 1.31-1.96]; P < .001, respectively; 2 episodes: OR, 1.81 [95% CI, 1.35-2.42]; P < .001 and OR, 2.38 [95% CI, 1.68-3.37]; P < .001, respectively; 3 episodes: OR, 4.39 [95% CI, 2.19-8.78]; P < .001 and OR, 3.88 [95% CI, 2.29-6.55]; P < .001, respectively). These associations were confirmed for any ROP by propensity score matching (for example, the aOR with propensity score matching was 1.76 [95% CI, 1.54-2.02]; P < .001 for 1 episode vs 0 episodes and 1.58 [95% CI, 1.12-2.22]; P = .007 for 3 episodes vs 0 or 1 episode). In the NNN dataset, surgical NEC was associated with treatment-warranted ROP (multivariable analysis: aOR, 3.37 [95% CI, 1.78-6.37]; P < .001).

Conclusions and Relevance

This study found that in the large-scale GNN cohort, recurrent culture-proven sepsis was associated with ROP and treatment-warranted ROP in infants born at less than 29 weeks.

Image-free recognition of moderate ROP from mild with machine learning algorithm on plasma Raman spectrum

The retinopathy of prematurity (ROP) can cause serious clinical consequences and, fortunately, it is remediable while the time window for treatment is relatively narrow. Therefore, it is urgent to screen all premature infants and diagnose ROP degree timely, which has become a large workload for pediatric ophthalmologists. We developed a retinal image-free procedure using small amount of blood samples based on the plasma Raman spectrum with the machine learning model to automatically classify ROP cases before medical intervention was performed. Statistical differences in infrared Raman spectra of plasma samples were found among the control, mild (ZIIIS1), moderate (ZIIIS2 & ZIIS1), and advanced (ZIIS2) ROP groups. With the different wave points of Raman spectra as the inputs, the outputs of our support vector machine showed that the area under the curves in the receiver operating characteristic (AUC) were 0.763 for the pair comparisons of the control with the mild groups, 0.821 between moderate and advanced groups (ZIIS2), while more than 90% in comparisons of the other four pairs: control vs. moderate (0.981), control vs. advanced (0.963), mild vs. moderate (0.936), and mild vs. advanced (0.953), respectively. Our study could advance principally the ROP diagnosis in two dimensions: the moderate ROPs have been classified remarkably from the mild ones, which leaves more time for the medical treatments, and the procedure of Raman spectrum with a machine learning model based on blood samples can be conveniently promoted to those hospitals lacking of the pediatric ophthalmologists with experience in reading retinal images.

Usefulness of Children's Hospital of Philadelphia ROP (CHOP ROP) model in the prediction of type 1 ROP

Purpose

Children's Hospital of Philadelphia retinopathy of prematurity (CHOP ROP) model can be used to predict ROP, a leading cause of childhood blindness, using risk factors such as postnatal weight gain, birth weight (BW), and gestation age (GA). The purpose of this study was to determine the usefulness of the CHOP ROP for the prediction of treatable ROP.

Methods

This was a prospective observational study. Babies <34 weeks of GA, BW <2000 grams, and GA 34-36 weeks with risk factors such as respiratory distress syndrome (RDS) were included; ROP screening, follow-up, and treatment were performed based on national guidelines. The average daily postnatal weight gain was measured, and the CHOP nomogram was plotted. Babies were categorized as high risk or low risk based on the "CHOP" alarm. The sensitivity and specificity of the CHOP ROP for the detection of treatable ROP were determined. In case of poor sensitivity, a new cutoff alarm level was planned using logistic regression analysis.

Results

Of 62 screened infants, 23 infants did not fulfill the criteria of the CHOP algorithm and were excluded. Thus, in the study on 39 infants, the predictive model with an alarm level of 0.014 had 100% specificity and 20% sensitivity. With the "new" alarm level (cutoff) of 0.0003, the CHOP nomogram could detect all the infants who developed treatable ROP, that is, sensitivity increased to 100% but specificity decreased to 10.5%.

Conclusion

The CHOP ROP model with a cutoff point (0.014) performed poorly in predicting severe ROP in the study. Thus, there is a need to develop inclusive and more sensitive tailor-made algorithms.

Comparison of Weight-Gain-Based Prediction Models for Retinopathy of Prematurity in an Australian Population

Purpose

Four weight-gain-based algorithms are compared for the prediction of type 1 ROP in an Australian cohort: the weight, insulin-like growth factor, neonatal retinopathy of prematurity (WINROP) algorithm, the Children's Hospital of Philadelphia Retinopathy of Prematurity (CHOPROP), the Colorado Retinopathy of Prematurity (CO-ROP) algorithm, and the postnatal growth, retinopathy of prematurity (G-ROP) algorithm.

Methods

A four-year retrospective cohort analysis of infants screened for ROP in a tertiary neonatal intensive care unit in Brisbane, Australia. The main outcome measures were sensitivities, specificities, and positive and negative predictive values.

Results

531 infants were included (mean gestational age 28 + 3). 24 infants (4.5%) developed type 1 ROP. The sensitivities, specificities, and negative predictive values, respectively, for type 1 ROP (95% confidence intervals) were for WINROP 83.3% (61.1-93.3%), 52.3% (47.8-56.7%), and 98.4% (96.1-99.4%); for CHOPROP 100% (86.2-100%), 46.0% (41.7-50,3%), and 100% (98.4-100%); for CO-ROP 100% (86.2-100%), 32.0% (28.0%-36.1%), and 100% (98.3-100%); and for G-ROP 100% (86.2-100%), 28.2% (24.5-32.3%), and 100% (97.4-100%). Of the five infants with persistent nontype 1 ROP that underwent treatment, only CO-ROP was able to successfully identify all.

Conclusions

CHOPROP, CO-ROP, and G-ROP performed well in this Australian population. CHOPROP, CO-ROP, and G-ROP would reduce the number of infants requiring examinations by 43.9%, 30.5%, and 26.9%, respectively, compared to current ROP screening guidelines. Weight-gain-based algorithms would be a useful adjunct to the current ROP screening.

Validation of three weight gain-based algorithms as a screening tool to detect retinopathy of prematurity: A multicenter study

Purpose

Screening guidelines for retinopathy of prematurity (ROP) are updated frequently to help clinicians identify infants at risk of type 1 ROP. This study aims to evaluate the accuracy of three different predictive algorithms-WINROP, ROPScore, and CO-ROP-in detecting ROP in preterm infants in a developing country.

Methods

This retrospective study was conducted on 386 preterm infants from two centers between 2015 and 2021. Neonates with gestational age ≤30 weeks and/or birth weight ≤1500 g who underwent ROP screening were included.

Results

One hundred twenty-three neonates (31.9%) developed ROP. The sensitivity to identify type 1 ROP was as follows: WINROP, 100%; ROPScore, 100%; and CO-ROP, 92.3%. The specificity was 28% for WINROP, 1.4% for ROPScore, and 19.3% for CO-ROP. CO-ROP missed two neonates with type 1 ROP. WINROP provided the best performance for type 1 ROP with an area under the curve score at 0.61.

Conclusion

The sensitivity was at 100% for WINROP and ROPScore for type 1 ROP; however, specificity was quite low for both algorithms. Highly specific algorithms tailored to our population may serve as a useful adjunctive tool to detect preterm infants at risk of sight-threatening ROP.

Latest Trends in Retinopathy of Prematurity: Research on Risk Factors, Diagnostic Methods and Therapies

Retinopathy of prematurity (ROP) is a vasoproliferative disorder with an imminent risk of blindness, in cases where early diagnosis and treatment are not performed. The doctors' constant motivation to give these fragile beings a chance at life with optimal visual acuity has never stopped, since Terry first described this condition. Thus, throughout time, several specific advancements have been made in the management of ROP. Apart from the most known risk factors, this narrative review brings to light the latest research about new potential risk factors, such as: proteinuria, insulin-like growth factor 1 (IGF-1) and blood transfusions. Digital imaging has revolutionized the management of retinal pathologies, and it is more and more used in identifying and staging ROP, particularly in the disadvantaged regions by the means of telescreening. Moreover, optical coherence tomography (OCT) and automated diagnostic tools based on deep learning offer new perspectives on the ROP diagnosis. The new therapeutical trend based on the use of anti-VEGF agents is increasingly used in the treatment of ROP patients, and recent research sustains the theory according to which these agents do not interfere with the neurodevelopment of premature babies.

Retrospective validation of G-ROP, CO-ROP, Alex-ROP, and ROPscore predictive algorithms in two Chinese medical centers

Purpose

To evaluate the sensitivity and specificity of four predictive algorithms (G-ROP, CO-ROP, Alex-ROP, and ROPscore) for retinopathy of prematurity and compare their performances in the Chinese population.

Methods

A retrospective study was conducted at two medical centers in China of infants born at Women's Hospital School of Medicine Zhejiang University and Yiwu Maternal and Child Health Hospital. A total of 1,634 infants who met the criteria and who were GA < 32 weeks or BW < 2,000 g according to Chinese guidelines for ROP screening were included. The ROP group was further grouped into severe ROP and mild ROP. The sensitivity and specificity of G-ROP, two simplified G-ROPs, CO-ROP, Alex-ROP, and ROPscore were analyzed.

Results

Severe ROP and any ROP were identified in 25 and 399 of 1,634 infants, respectively. According to the criteria of different models, 844, 1,122, 1,122, and 587 infants were eligible in the G-ROP, CO-ROP, Alex-ROP, and ROPscore, respectively. G-ROP had 96.0% sensitivity and 35.0% specificity for severe ROP. For two simplified G-ROPs (180 g and 200 g models), similar sensitivity was showed with original G-ROP and they had specificity of 21.8% and 14.0%, respectively. The sensitivity and specificity of Co-ROP were 96% and 64.3% for severe ROP, while Alex-ROP only had sensitivity of 56.0% and specificity of 61.4% for severe ROP. ROPscore had a sensitivity of 91.3% and a specificity of 62.4% for severe ROP. In 546 infants who met all 4 models' inclusion criteria and included 23 infants with severe ROP, the validation outcomes showed the sensitivity of G-ROP, ROPscore, CO-ROP, and Alex-ROP for severe ROP was 95.6%, 91.3%, 100%, and 56.0%, and their specificity was 38.0%, 60.8%, 39.9%, and 52.9%, respectively.

Conclusion

G-ROP, ROPscore, and CO-ROP had high sensitivity for severe ROP in the Chinese population, but both the sensitivity and specificity of Alex-ROP were low. CO-ROP (not high-grade CO-ROP) provided the best performance for severe ROP in a fair comparison. For further application, ROP screening models need to be adjusted by local populations.

Daily Oxygen Supplementation and Risk of Retinopathy of Prematurity

PURPOSE

Excessive oxygen supplementation increases risk of retinopathy of prematurity (ROP). While numerous oxygen parameters could be considered when predicting ROP (saturation targets, actual saturation, fraction of inspired oxygen, etc.), complicated measures are impractical as screening criteria. We sought to develop a simple, clinically useful measure of daily oxygen supplementation during ages 0-28 days to improve prediction of ROP.

METHODS

Secondary analysis of two Postnatal Growth and ROP (G-ROP) Study cohorts (G-ROP-1 and G-ROP-2) at 45 hospitals. Infants with a known ROP outcome and complete oxygen data were included. Associations between severe ROP and days on supplemental oxygen (FiO2 > 21%), during ages 0-28 days (DSO28) were assessed, controlling for birth weight (BW) and gestational age (GA). New screening criteria incorporating DSO were developed and compared to current guidelines.

RESULTS

Among 8,949 studied infants, 459 (5.1%) developed type 1 ROP. DSO28 was associated with severe ROP (adjusted-OR 1.05 per day supplemental oxygen, 95%CI 1.03-1.07, p < .0001). The following criteria had 100% sensitivity for type 1 ROP and higher specificity than current guidelines: new BW/GA criteria with DSO (BW<901 g, GA<26 weeks, or DSO >3), 23.4% fewer infants examined; modified G-ROP criteria including DSO, 29.0% fewer infants; original G-ROP criteria, 31.8% fewer infants.

CONCLUSION

In high-level neonatal-care settings, incorporating DSO (a simple measure of oxygen supplementation) into screening criteria improves sensitivity and specificity for type 1 ROP over current BW-GA criteria, but does not perform as well as the validated G-ROP criteria.

Early and late onset sepsis and retinopathy of prematurity in a cohort of preterm infants

This study investigates the impact of antenatal and postnatal infection or inflammation on the onset and progression of Retinopathy of Prematurity (ROP). We retrospectively collected clinical and demographic data of preterm infants with birth weight ≤ 1500 g or gestational age < 30 weeks admitted to the neonatal intensive care unit of Verona from 2015 to 2019. Uni- and multivariable analysis was performed to evaluate the potential effect of selected variables on the occurrence of any stage ROP and its progression to severe ROP, defined as ROP requiring treatment. Two hundred and eighty neonates were enrolled and 60 of them developed ROP (21.4%). Oxygen need for 28 days and late-onset sepsis (LOS) increased the risk of any grade ROP after adjusting for birth weight and gestational age (OR 6.35, 95% CI 2.14–18.85 and OR 2.49, 95% CI 1.04–5.94, respectively). Days of mechanical ventilation and of non-invasive ventilation increased the risk of progression to severe ROP after adjusting for birth weight and gestational age (OR 1.08, CI 1.02–1.14 and OR 1.06, CI 1.01–1.11, respectively). Exposure to infection with production of inflammatory mediators may contribute to increase the risk of ROP occurrence in very preterm neonates.

The Use of Postnatal Weight Gain Algorithms to Predict Severe or Type 1 Retinopathy of Prematurity: A Systematic Review and Meta-analysis

IMPORTANCE

The currently recommended method for screening for retinopathy of prematurity (ROP) is binocular indirect ophthalmoscopy, which requires frequent eye examinations entailing a heavy clinical workload. Weight gain-based algorithms have the potential to minimize the need for binocular indirect ophthalmoscopy and have been evaluated in different setups with variable results to predict type 1 or severe ROP.

OBJECTIVE

To synthesize evidence regarding the ability of postnatal weight gain-based algorithms to predict type 1 or severe ROP.

DATA SOURCES

PubMed, MEDLINE, Embase, and the Cochrane Library databases were searched to identify studies published between January 2000 and August 2021.

STUDY SELECTION

Prospective and retrospective studies evaluating the ability of these algorithms to predict type 1 or severe ROP were included.

DATA EXTRACTION AND SYNTHESIS

Two reviewers independently extracted data. This meta-analysis was performed according to the Cochrane guidelines and reported according to the Preferred Reporting Items for Systematic Reviews and Meta-analysis of Diagnostic Test Accuracy Studies (PRISMA-DTA) guidelines.

MAIN OUTCOMES AND MEASURES

Ability of algorithms to predict type 1 or sever ROP was measured using statistical indices (pooled sensitivity, specificity, and summary area under the receiver operating characteristic curves, as well as pooled negative likelihood ratios and positive likelihood ratios and diagnostic odds ratios).

RESULTS

A total of 61 studies (>37 000 infants) were included in the meta-analysis. The pooled estimates for sensitivity and specificity, respectively, were 0.89 (95% CI, 0.85-0.92) and 0.57 (95% CI, 0.51-0.63) for WINROP (Weight, IGF-1 [insulinlike growth factor 1], Neonatal, ROP), 1.00 (95% CI, 0.88-1.00) and 0.60 (95% CI, 0.15-0.93) for G-ROP (Postnatal Growth and ROP), 0.95 (95% CI, 0.71-0.99) and 0.52 (95% CI, 0.36-0.68) for CHOP ROP (Children's Hospital of Philadelphia ROP), 0.99 (95% CI, 0.73-1.00) and 0.49 (95% CI, 0.03-0.74) for ROPScore, 0.98 (95% CI, 0.94-0.99) and 0.35 (95% CI, 0.22-0.51) for CO-ROP (Colorado ROP). The original PINT (Premature Infants in Need of Transfusion) ROP study reported a sensitivity of 0.98 (95% CI, 0.91-0.99) and a specificity of 0.36 (95% CI, 0.30-0.42). The pooled negative likelihood ratios were 0.19 (95% CI, 0.13-0.27) for WINROP, 0.0 (95% CI, 0.00-0.32) for G-ROP, 0.10 (95% CI, 0.02-0.53) for CHOP ROP, 0.03 (95% CI, 0.00-0.77) for ROPScore, and 0.07 (95% CI, 0.03-0.16) for CO-ROP. The pooled positive likelihood ratios were 2.1 (95% CI, 1.8-2.4) for WINROP, 2.5 (95% CI, 0.7-9.1) for G-ROP, 2.0 (95% CI, 1.5-2.6) for CHOP ROP, 1.9 (95% CI, 1.1-3.3) for ROPScore, and 1.5 (95% CI, 1.2-1.9) for CO-ROP.

CONCLUSIONS AND RELEVANCE

This study suggests that weight gain-based algorithms have adequate sensitivity and negative likelihood ratios to provide reasonable certainty in ruling out type 1 ROP or severe ROP. Given the implications of missing even a single case of severe ROP, algorithms with very high sensitivity (close to 100%) and low negative likelihood ratios (close to zero) need to be chosen to safely reduce the number of unnecessary examinations in infants at lower risk of severe ROP.

Neonatal Fundus Image Registration and Mosaic Using Improved Speeded Up Robust Features Based on Shannon Entropy

Fundus examination of the newborn is quite important, which needs to be done timely so as to avoid irreversible blindness. Ophthalmologists have to review at least five images of each eye during one examination, which is a time-consuming task. To improve the diagnosis efficiency, this paper proposed a stable and robust fundus image mosaic method based on improved Speeded Up Robust Features (SURF) with Shannon entropy and make real assessment when ophthalmologists used it clinically. Our method is characterized by avoiding the useless detection and extraction of the feature points in the non-overlapping region of the paired images during registration process. The experiments showed that the proposed method successfully registered 90.91% of 110 different field of view (FOV) image pairs from 22 eyes of 13 screening newborns and acquired 93.51% normalized correlation coefficient and 1.2557 normalized mutual information. Also, the total fusion success rate reached 86.36% and a subjective visual assessment approach was adopted to measure the fusion performance by three experts, which obtained 84.85% acceptance rate. The performance of our proposed method demonstrated its accuracy and effectiveness in the clinical application, which can help ophthalmologists a lot during their diagnosis.

Diagnostic Accuracy of WINROP, CHOP-ROP and ROPScore in Detecting Type 1 Retinopathy of Prematurity

Background

Algorithms for predicting retinopathy of prematurity (ROP) requiring treatment need to be validated in Indian settings to determine if the burden of screening can be reduced without compromising the sensitivity of existing gestation and weight-based cut offs.

Objective

To evaluate the performance of the available algorithms namely, WINROP (Weight, Insulin-like growth factor I, Neonatal ROP), CHOP-ROP (children’s Hospital of Philadelphia ROP) and ROPScore in predicting type 1 ROP and time from alarm to treatment by each algorithm.

Study design

Ambispective observational.

Setting

Tertiary care neonatal intensive care unit in India.

Participants

Neonates less than 32 weeks or less than 1500 g born between July, 2013 to June, 2019 (N=578), who underwent ROP screening.

Primary outcome

Sensitivity, specificity and time from alarm to treatment by each algorithm.

Results

The sensitivity and specificity of WINROP was 85% and 36%, for CHOP-ROP it was 54% and 71%, and for ROPScore it was 73% and 67%, respectively in detecting type 1 ROP. A total of 50/51 (98%) of neonates with type 1 ROP underwent treatment at median gestation of 9 weeks and median time from alarm to treatment by WINROP, CHOP-ROP and ROPScore was 7, 7 and 3 weeks, respectively.

Conclusion

WINROP, CHOP-ROP and ROPScore were not sensitive enough to replace the gestational age, weight and risk factor-based screening criteria for type 1 ROP.

Use of an online screening algorithm - Weight, Insulin-derived growth factor 1, Neonatal Retinopathy of Prematurity (WINROP) for predicting retinopathy of prematurity in Indian preterm babies

Purpose:

Inopathy of prematurity (WINROP) Weight, insulin-derived growth factor 1, neonatal ROP algorithm is an online tool that has been validated as a predictor of retinopathy of prematurity (ROP) in various countries. The current study was designed to evaluate the predictive ability of WINROP algorithm (http://winrop.com) using postnatal weight gain in detecting Type 1 ROP in Indian babies.

Methods:

Prospective single centre observational study of 153 consecutive preterm babies who were eligible for screening for ROP as per the standard guidelines. Sixteen babies were excluded from the study because of various reasons. Thirty-five babies had gestational age ≥32 weeks and were ineligible for WINROP algorithm. Online WINROP algorithm was used for 102 babies with gestation at birth less than 32 weeks. The alarms triggered by WINROP were documented.

Results:

Laser treatment was done in 30 babies who developed Type 1 ROP. Of these, WINROP alarm was signaled in 24 babies and 6 babies developed ROP without any WINROP alarm. These babies had associated comorbidities like respiratory distress syndrome, patent ductus arteriosus, bacterial sepsis, and ventilatory support. WINROP alarm was significantly associated with Type 1 ROP (P < 0.001). The sensitivity of WINROP was 80% and specificity was 80.6% with a positive predictive value of 63.2% and negative predictive value of 90.6% in detecting Type 1 ROP. In the present study, no baby who was ineligible for WINROP developed Type 1 ROP.

Conclusion:

WINROP provides a novel online monitoring screening tool for identifying babies at risk of developing Type 1 ROP. In our cohort, none of the babies whose period of gestation was more than or equal to 32 weeks developed sight threatening Type 1 ROP. WINROP algorithm may also be useful in Indian population.

Retrospective comparison between growth and retinopathy of prematurity model versus WINROP model

OBJECTIVE

To compare the weight and insulin-like growth factor-1 in neonatal retinopathy (WINROP) to the growth and retinopathy of prematurity (G-ROP) model in a Portuguese cohort.

DESIGN

Retrospective case series.

METHODS

Clinical records of consecutive infants who underwent retinopathy of prematurity (ROP) screening from April 2012 to May 2019 were retrospectively reviewed. Both WINROP and G-ROP models were accessed for sensitivity and specificity for type 1 ROP. A separate analysis of both algorithms was performed in infants with gestational age (GA) <30 weeks.

RESULTS

Of the 375 infants included in the study, 313 were eligible for G-ROP analysis and 311 for WINROP. In the G-ROP group, 22 infants developed type 1 ROP (sensitivity 90.91%, 95% confidence interval [CI] 70.84%-98.98%). In the WINROP group, 23 infants needed treatment (sensitivity of 86.96%, 95% CI 66.41%-97.22%). Both models reached 100% sensitivity for type 1 ROP if restricted to GA <30 weeks.

CONCLUSIONS

Both models were easy to use and had similar sensitivities. If restricted to GA <30 weeks, both models detected all type 1 ROP.

Using ROPScore and CHOP ROP for early prediction of retinopathy of prematurity in a Chinese population

PURPOSE

Retinopathy of prematurity (ROP) is a disease that causes vision loss, vision impairment, and blindness, most frequently manifesting among preterm infants. ROPScore and CHOP ROP (Children's Hospital of Philadelphia ROP) are similar scoring models to predict ROP using risk factors such as postnatal weight gain, birth weight (BW), and gestation age (GA). The purpose of this study was to compare the accuracy and difference between using ROPScore and CHOP ROP for the early prediction of ROP.

METHODS

A retrospective study was conducted from January 2009 to December 2019 in China. Patients eligible for enrollment included infants admitted to NICU at ≤32 weeks GA or those with ≤1500 g BW. The sensitivity and specificity of ROPScore and CHOP ROP were analyzed, as well as its suitability as an independent predictor of ROP.

RESULTS

Severe ROP was found in 5.0% of preterm infants. The sensitivity and specificity of the ROPScore test at any stage of ROP was 55.8 and 77.8%, respectively. For severe ROP, the sensitivity and specificity was 50 and 87.0%, respectively. The area under the receiver operating characteristic curve for the ROPScore for predicting severe ROP was 0.76. This value was significantly higher than the values for birth weight (0.60), gestational age (0.73), and duration of ventilation (0.63), when each was category measured separately. For the CHOP ROP, it correctly predicted infants who developed type 1 ROP (sensitivity, 100%, specificity, 21.4%).

CONCLUSIONS

The CHOP ROP model predicted infants who developed type 1 ROP at a sensitivity of 100% whereas ROPScore had a sensitivity of 55.8%. Therefore, the CHOP ROP model is more suitable for Chinese populations than the ROPScore test.

CLINICAL REGISTRATION NUMBER AND STROBE GUIDELINES

This article was a retrospective cohort study and reported the results of the ROPScore and CHOP ROP algorithms. No results pertaining to interventions on human participants were reported. Thus, registration was not required and this study followed STROBE guidelines.

ROP Screening Tool Assessment and Validation in a Third-Level Hospital in Argentina: A Pilot Study

PURPOSE

To evaluate whether a mathematical tool that predicts severe retinopathy of prematurity (ROP) using clinical parameters at 6 weeks of life (ROPScore calculator smartphone application; PABEX Corporation) can be useful to predict severe ROP in a population of premature infants in Argentina.

METHODS

In this retrospective study, data from the clinical records of all premature infants examined between 2012 and 2018 in the ophthalmology department of a public third-level hospital in Córdoba, Argentina, were obtained. ROPScore screening was applied using a Microsoft Excel spreadsheet (Microsoft Corporation). The sensitivity, specificity, and positive (PPV) and negative (NPV) predictive values of the algorithm were analyzed.

RESULTS

Between 2012 and 2018, a total of 2,894 pre-term infants were examined and 411 met the inclusion criteria, of whom 34% (n = 139) presented some form of ROP and 6% (n = 25) developed severe forms that required treatment. The sensitivity of the algorithm for any ROP and severe ROP was 100%. The PPV and NPV were 35.64% and 100%, respectively, for any ROP and 9.88% and 100% for severe ROP.

CONCLUSIONS

One-time only calculation of the ROPScore algorithm could identify severe cases after validation, reducing the number of screened infants by 38% in infants with a birth weight of 1,500 g or less or a gestational age of 32 weeks or younger. [J Pediatr Ophthalmol Strabismus. 2021;58(1):55-61.].

Prediction of severe retinopathy of prematurity using the weight gain, insulin-like growth factor 1, and neonatal retinopathy of prematurity algorithm in a Japanese population of preterm infants

PURPOSE

To retrospectively investigate the sensitivity and specificity of weight gain, insulin-like growth factor 1, and neonatal retinopathy of prematurity (WINROP) algorithm for the prediction of severe retinopathy of prematurity (ROP) in a Japanese population of preterm infants. The WINROP algorithm is a tool based on postnatal weight gain.

STUDY DESIGN

Retrospective cohort study.

METHODS

The medical records of preterm infants born between January 2011 and March 2017 were retrospectively reviewed. Infants born after 33 weeks of gestation were excluded based on the indications of the WINROP algorithm. Postnatal weight was recorded weekly on the WINROP system until postmenstrual week 36. The sensitivity and specificity of the WINROP algorithm were analyzed.

RESULTS

In total, 278 infants were included in this study. Based on the WINROP algorithm 110 of these infants were predicted to be at low risk for developing severe ROP and 105 did not develop severe ROP. Based on the WINROP algorithm 168 infants were predicted to be at high risk for developing severe ROP　and 27 developed severe ROP. Thus, the sensitivity of the WINROP algorithm was 84.4% and the specificity 42.7%.

CONCLUSION

The WINROP algorithm could be used for preterm infants (gestational age of <28 weeks) without a complicated hospital course. Modification of the algorithm will improve its sensitivity and specificity for the Japanese population.

Development of Modified Screening Criteria for Retinopathy of Prematurity: Primary Results From the Postnatal Growth and Retinopathy of Prematurity Study

Importance

Current retinopathy of prematurity (ROP) guidelines, which are based on studies of high-risk infants and expert opinion, have low specificity for disease requiring treatment. Postnatal weight gain-based models improve specificity but have been limited by complexity and small development cohorts, which results in model overfitting and resultant decreased sensitivity in validation studies.

Objective

To develop a birth weight (BW), gestational age (GA), and weight gain (WG) prediction model using data from a broad-risk cohort of premature infants.

Design, Setting, and Participants

The Postnatal Growth and ROP Study was a retrospective multicenter cohort study conducted in 29 hospitals in the United States and Canada from 2006 to 2012 that included 7483 premature infants at risk for ROP with a known ROP outcome. A hybrid modeling approach was used that combined BW/GA criteria, weight comparison with expected growth from infants without ROP, multiple growth-interval assessments, consideration of nonphysiological WG, and user-friendly screening criteria. Numerous BW/GA levels, postnatal age periods, time intervals, and WG percentile thresholds were evaluated to identify the most robust parameters.

Main Outcome and Measures

Sensitivity for Early Treatment of ROP Study type 1 ROP and potential reduction in infants who require examinations.

Results

Of 7483 infants, the median (SD) BW was 1099 (359) g, the median GA was 28 weeks (range, 22-35), 3575 (47.8%) were female, 3615 (48.4%) were white, 2310 (30.9%) were black, 233 (3.1%) were Asian, 93 (1.2%) were Pacific Islander, and 40 (0.5%) were American Indian/Alaskan Native. Infants who met any of 6 criteria would undergo examinations: (1) a GA of younger than 28 weeks; (2) a BW of less than 1051 g; a WG of less than 120 g, 180 g, or 170 g during ages 10 to 19, 20 to 29, or 30 to 39 days, respectively; or hydrocephalus. These criteria predicted 459 of 459 (100%) type 1 (sensitivity, 100%; 95% CI, 99.2%-100%), 524 of 524 (100%) treated, and 466 of 472 (98.7%) type 2 cases while reducing the number of infants who required examinations by 2269 (30.3%).

Conclusions and Relevance

This cohort study, broadly representative of infants who are undergoing ROP examinations, provides evidence-based screening criteria. With validation, the Postnatal Growth and ROP Study criteria could be incorporated into ROP screening guidelines to reduce the number of infants who require examinations in North America.

Postnatal weight gain and retinopathy of prematurity

Infants meeting retinopathy of prematurity (ROP) screening guidelines based on birth weight and gestational age undergo serial examinations by ophthalmologists for detection and treatment. However, less than 10% require treatment, and less than half develop ROP. Slow postnatal weight gain is highly predictive of ROP, and investigators have incorporated weight gain measures to develop more specific criteria for ROP screening. Such clinical prediction model use involves a large development study, validation studies specific to the target populations, and ongoing impact surveillance, with adjustment as necessary. Of the many weight gain inclusive prediction models intended to improve the precision of ROP screening, the Postnatal Growth and ROP (G-ROP) modified screening criteria were developed using the largest dataset and may provide the most robust model for clinical use. A recently completed G-ROP validation study will evaluate the generalizability of these modified criteria prior to clinical use.

Levels and effectiveness of oral retinol supplementation in VLBW preterm infants

Retinol palmitate oral administration is convenient, but it is difficult to assess/monitor its nutritional status in preterm infants and literature is controversial about the administration route and the effectiveness of vitamin A supplementation. We primarily evaluated retinol plasma levels to assess the vitamin A nutritional status in preterm infants (<1500 g; 32 weeks) after 28 days of oral supplementation (3000 IU/kg/day, retinol palmitate drops), in addition to vitamin A standard amount as suggested by European Society of Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) guidelines. We then observed the rate of typical preterm pathologies in the supplemented group (31 newborns) and in 10 matching preterm infants, hospitalized in neonatal intensive care unit (NICU) in the same period, who received neither vitamin A supplementation nor parents allowed plasma sampling. Oral integration resulted in constant retinol plasma concentration around the desired level of 200 ng/mL, but without statistical increase during the study period. Due to the complexity of vitamin A metabolism and the immaturity of preterm infant’s organs, retinol supplementation may had first saturated other needy tissues; therefore, plasmatic measures may not be consistent with improved global vitamin A body distribution. Therefore, achieving a constant retinol concentration is a valuable result and supportive for oral administration: decreasing levels, even after parenteral/enteral supplementation, were reported in the literature. In spite of favourable trend and no adverse events, we did not report statistical difference in co-morbidities. This investigation confirms the necessity to perform further trials in preterm newborns, to find an index reflecting the complex nutritional retinol status after oral administration of vitamin A, highlighting its effectiveness/tolerability in correlated preterm infant’s pathologies.

Chorioamnionitis as a risk factor for retinopathy of prematurity: An updated systematic review and meta-analysis

The role of chorioamnionitis (CA) in the development of retinopathy of prematurity (ROP) is difficult to establish, because CA-exposed and CA-unexposed infants frequently present different baseline characteristics. We performed an updated systematic review and meta-analysis of studies reporting on the association between CA and ROP. We searched PubMed and EMBASE for relevant articles. Studies were included if they examined preterm or very low birth weight (VLBW, <1500g) infants and reported primary data that could be used to measure the association between exposure to CA and the presence of ROP. Of 748 potentially relevant studies, 50 studies met the inclusion criteria (38,986 infants, 9,258 CA cases). Meta-analysis showed a significant positive association between CA and any stage ROP (odds ratio [OR] 1.39, 95% confidence interval [CI] 1.11 to 1.74). CA was also associated with severe (stage ≥3) ROP (OR 1.63, 95% CI 1.41 to 1.89). Exposure to funisitis was associated with a higher risk of ROP than exposure to CA in the absence of funisitis. Additional meta-analyses showed that infants exposed to CA had lower gestational age (GA) and lower birth weight (BW). Meta-regression showed that lower GA and BW in the CA-exposed group was significantly associated with a higher risk of ROP. Meta-analyses of studies with data adjusted for confounders could not find a significant association between CA and ROP. In conclusion, our study confirms that CA is a risk factor for developing ROP. However, part of the effects of CA on the pathogenesis of ROP may be mediated by the role of CA as an etiological factor for very preterm birth.

The ROPScore as a Screening Algorithm for Predicting Retinopathy of Prematurity in a Brazilian Population

OBJECTIVES

To evaluate the accuracy of the ROPScore algorithm as a predictor of retinopathy of prematurity (ROP).

METHODS

A prospective cohort of 220 preterm infants with a birth weight ≤1500 g and/or gestational age ≤32 weeks was included. The ROPScore was determined in the sixth week of life in 181 infants who then survived until a corrected gestational age of 45 weeks. The sensitivity, specificity, and positive (PPV) and negative predictive values (NPV) of the algorithm were analyzed.

RESULTS

ROP was found in 17.6% of the preterm infants. The sensitivity of this test for any stage of ROP was 87.5%, while that for severe ROP was 95.4% (21/22 cases). The PPV and NPV were 59.6% and 97%, respectively, for any stage of ROP and 44.7% and 99.25%, respectively, for severe ROP. The ROPScore could therefore hypothetically reduce the number of ophthalmologic examinations required to detect ROP by 71.8%.

CONCLUSION

The ROPScore is a useful screening tool for ROP and may optimize examinations and especially the identification of severe ROP.

Validation of the Colorado Retinopathy of Prematurity Screening Model

Importance

The Colorado Retinopathy of Prematurity (CO-ROP) model uses birth weight, gestational age, and weight gain at the first month of life (WG-28) to predict risk of severe retinopathy of prematurity (ROP). In previous validation studies, the model performed very well, predicting virtually all cases of severe ROP and potentially reducing the number of infants who need ROP examinations, warranting validation in a larger, more diverse population.

Objective

To validate the performance of the CO-ROP model in a large multicenter cohort.

Design, Setting, Participants

This study is a secondary analysis of data from the Postnatal Growth and Retinopathy of Prematurity (G-ROP) Study, a retrospective multicenter cohort study conducted in 29 hospitals in the United States and Canada between January 2006 and June 2012 of 6351 premature infants who received ROP examinations.

Main Outcomes and Measures

Sensitivity and specificity for severe (early treatment of ROP [ETROP] type 1 or 2) ROP, and reduction in infants receiving examinations. The CO-ROP model was applied to the infants in the G-ROP data set with all 3 data points (infants would have received examinations if they met all 3 criteria: birth weight, <1501 g; gestational age, <30 weeks; and WG-28, <650 g). Infants missing WG-28 information were included in a secondary analysis in which WG-28 was considered fewer than 650 g.

Results

Of 7438 infants in the G-ROP study, 3575 (48.1%) were girls, and maternal race/ethnicity was 2310 (31.1%) African American, 3615 (48.6%) white, 233 (3.1%) Asian, 40 (0.52%) American Indian/Alaskan Native, and 93 (1.3%) Pacific Islander. In the study cohort, 747 infants (11.8%) had type 1 or 2 ROP, 2068 (32.6%) had lower-grade ROP, and 3536 (55.6%) had no ROP. The CO-ROP model had a sensitivity of 96.9% (95% CI, 95.4%-97.9%) and a specificity of 40.9% (95% CI, 39.3%-42.5%). It missed 23 (3.1%) infants who developed severe ROP. The CO-ROP model would have reduced the number of infants who received examinations by 26.1% (95% CI, 25.0%-27.2%).

Conclusions and Relevance

The CO-ROP model demonstrated high but not 100% sensitivity for severe ROP and missed infants who might require treatment in this large validation cohort. The model requires all 3 criteria to be met to signal a need for examinations, but some infants with a birth weight or gestational age above the thresholds developed severe ROP. Most of these infants who were not detected by the CO-ROP model had obvious deviation in expected weight trajectories or nonphysiologic weight gain. These findings suggest that the CO-ROP model needs to be revised before considering implementation into clinical practice.

Validation of the Children's Hospital of Philadelphia Retinopathy of Prematurity (CHOP ROP) Model

Importance

The Children's Hospital of Philadelphia Retinopathy of Prematurity (CHOP ROP) model uses birth weight (BW), gestational age at birth (GA), and weight gain rate to predict the risk of severe retinopathy of prematurity (ROP). In a model development study, it predicted all infants requiring treatment, while greatly reducing the number of examinations compared with current screening guidelines.

Objective

To validate the CHOP ROP model in a multicenter cohort that is large enough to obtain a precise estimate of the model's sensitivity for treatment-requiring ROP.

Design, Setting, and Participants

This investigation was a secondary analysis of data from the Postnatal Growth and Retinopathy of Prematurity (G-ROP) Study. The setting was 30 hospitals in the United States and Canada between January 1, 2006, and June 30, 2012. The dates of analysis were September 28 to October 5, 2015. Participants were premature infants at risk for ROP with a known ROP outcome.

Main Outcomes and Measures

Sensitivity for Early Treatment of Retinopathy of Prematurity type 1 ROP and potential reduction in the number of infants requiring examinations. In the primary analysis, the CHOP ROP model was applied weekly to predict the risk of ROP. If the risk was above a cut-point level (high risk), examinations were indicated, while low-risk infants received no examinations. In a secondary analysis, low-risk infants received fewer examinations rather than no examinations.

Results

Participants included 7483 premature infants at risk for ROP with a known ROP outcome. Their median BW was 1070 g (range, 310-3000 g), and their median GA was 28 weeks (range, 22-35 weeks). Among them, 3575 (47.8%) were female, and their race/ethnicity was 3615 white (48.3%), 2310 black (30.9%), 233 Asian (3.1%), 93 Pacific Islander (1.2%), and 40 American Indian/Alaskan native (0.5%). The original CHOP ROP model correctly predicted 452 of 459 infants who developed type 1 ROP (sensitivity, 98.5%; 95% CI, 96.9%-99.3%), reducing the number of infants requiring examinations by 34.3% if only high-risk infants received examinations. Lowering the cut point to capture all type 1 ROP cases (sensitivity, 100%; 95% CI, 99.2%-100%) resulted in only 6.8% of infants not requiring examinations. However, if low-risk infants were examined at 37 weeks' postmenstrual age and followed up only if ROP was present at that examination, all type 1 ROP cases would be captured, and the number of examinations performed among infants with GA exceeding 27 weeks would be reduced by 28.4%.

Conclusion and Relevance

The CHOP ROP model demonstrated high but not 100% sensitivity and may be better used to reduce examination frequency. The model might be used reliably to guide a modified ROP screening schedule and decrease the number of examinations performed.

Late-onset Circulatory Collapse and Continuous Positive Airway Pressure are Useful Predictors of Treatment-requiring Retinopathy of Prematurity: A 9-year Retrospective Analysis

Visual loss caused by retinopathy of prematurity (ROP) will be prevented if treatment-requiring ROP (TR-ROP) can be predicted. In this retrospective study including 418 infants with ≤32 weeks of gestational age (GA) and/or ≤1500 grams of birthweight, we attempted to identify useful predictors. We also examined the efficiency of significant predictors compared with existing predictive models, ROPScore and CHOP model. Multivariable logistic regression analyses supported the following factors were useful for predicting TR-ROP from all infants and infants with any ROP: GA (odds ratio [OR], 0.47 and 0.48), history of late-onset circulatory collapse (LCC) (OR, 2.76 and 2.44) and use of continuous positive airway pressure (CPAP) at 35 weeks of postmenstrual age (OR, 3.78 and 4.50). The comparison of areas under receiver operating characteristic curves indicated the combination of LCC, CPAP and ROPScore was better than ROPScore to predict TR-ROP from all infants and infants with any ROP (P = 0.007 and 0.02) and the combination of LCC, CPAP and CHOP model was also better than CHOP model to predict TR-ROP from all infants and infants with any ROP (P = 0.01 and 0.02). Our results suggested infants with a history of LCC and a long CPAP support have a high incidence of TR-ROP.

Validation of WINROP for detecting retinopathy of prematurity in a North American cohort of preterm infants

BACKGROUND

WINROP (weight, insulin-like growth factor 1, neonatal, retinopathy of prematurity) is a web-based retinopathy of prematurity (ROP) risk algorithm that uses postnatal weight gain as a surrogate of insulin-like growth factor-1 (IGF-1) to predict the risk of severe ROP in premature infants. The purpose of this study was to validate the web-based algorithm WINROP in detecting severe (type 1 or type 2) ROP in a North American cohort of infants.

METHODS

The records of consecutive infants who underwent ROP examinations between 2008 and 2011 were reviewed retrospectively. Infants were classified into categories of "alarm" (at risk for developing severe ROP) and "no alarm" (minimal risk for severe ROP).

RESULTS

A total of 483 were included. Alarm occurred in 241 neonates (50%), with the median time from birth to alarm of 2 weeks. WINROP had a sensitivity of 81.8% (95% CI, 67.3%-91.8%) and specificity of 53.3% (95% CI, 48.5%-58.0%) for identifying infants with severe ROP. Eight of the 44 infants with severe ROP were not detected (5 with type 1 and 3 with type 2). Of these 8 infants, 7 (88%) had birth weight in excess of the 70^th pecentile. With additional weight data entry, sensitivity of WINROP rose to 88.6%.

CONCLUSIONS

Very preterm infants (gestational age of ≤27 weeks) with relatively high birth weight for gestational age may not be detected by WINROP as high risk for developing severe ROP.

Screening Tool for Early Postnatal Prediction of Retinopathy of Prematurity in Preterm Newborns (STEP-ROP)

BACKGROUND

Retinopathy of prematurity (ROP) is a disorder of the preterm newborn characterized by neurovascular disruption in the immature retina that may cause visual impairment and blindness.

OBJECTIVE

To develop a clinical screening tool for early postnatal prediction of ROP in preterm newborns based on risk information available within the first 48 h of postnatal life.

METHODS

Using data submitted to the Vermont Oxford Network (VON) between 1995 and 2015, we created logistic regression models based on infants born <28 completed weeks gestational age. We developed a model with 60% of the data and identified birth weight, gestational age, respiratory distress syndrome, non-Hispanic ethnicity, and multiple gestation as predictors of ROP. We tested the model in the remaining 40%, performed tenfold cross-validation, and tested the score in ELGAN study data.

RESULTS

Of the 1,052 newborns in the VON database, 627 recorded an ROP status. Forty percent had no ROP, 40% had mild ROP (stages 1 and 2), and 20% had severe ROP (stages 3-5). We created a weighted score to predict any ROP based on the multivariable regression model. A cutoff score of 5 had the best sensitivity (95%, 95% CI 93-97), while maintaining a strong positive predictive value (63%, 95% CI 57-68). When applied to the ELGAN data, sensitivity was lower (72%, 95% CI 69-75), but PPV was higher (80%, 95% CI 77-83).

CONCLUSIONS

STEP-ROP is a promising screening tool. It is easy to calculate, does not rely on extensive postnatal data collection, and can be calculated early after birth. Early ROP screening may help physicians limit patient exposure to additional risk factors, and may be useful for risk stratification in clinical trials aimed at reducing ROP.