Related Experiment Video
Updated: Jul 3, 2026
An Automated Method for Assessing Visual Acuity in Infants and Toddlers Using an Eye-Tracking System
Published on: March 17, 2023
Optimizing retinopathy of prematurity screening in China using a single objective criterion: a 10-year retrospective
Zhenglin Li1, Suo Guo1, Haojue Xu1
1Department of Ophthalmology, Sichuan Provincial People's Hospital, School of Medicine, University of Electronic Science and Technology of China, Chengdu, China.
Insights
A birth weight criterion of less than 2000g or gestational age less than 32 weeks effectively screens for retinopathy of prematurity (ROP). This objective screening method identifies most ROP cases, reducing the need for subjective assessments in high-risk infants.
Area of Science:
- Ophthalmology
- Neonatology
- Public Health
Background:
- Retinopathy of prematurity (ROP) is a leading cause of blindness in premature infants.
- Current screening guidelines often rely on subjective clinical judgment, potentially leading to over-screening.
- Objective criteria are needed to optimize ROP screening efficiency.
Purpose of the Study:
- To evaluate the efficacy of a single objective screening criterion (birth weight < 2000g or gestational age < 32 weeks) for ROP detection.
- To provide evidence for reducing over-screening in ROP detection.
- To assess the clinical value of subjective screening criteria in tertiary neonatal care.
Main Methods:
- Retrospective study of 1,453 premature infants screened for ROP (2014-2024).
- Infants grouped by objective criteria (BW < 2000g or GA < 32 weeks) versus subjective criteria (BW ≥ 2000g and GA ≥ 32 weeks).
- Fundus examinations performed via indirect ophthalmoscopy or RetCam III; ROP diagnosis followed Chinese Guidelines.
Main Results:
- The objective criterion group identified 98.2% of ROP cases and 99.2% of severe ROP cases.
- ROP and severe ROP prevalence were significantly higher in the objective group (P < 0.001).
- Lower gestational age and lower birth weight were independent risk factors for severe ROP.
Conclusions:
- A single objective criterion (BW < 2000g or GA < 32 weeks) effectively identifies high-risk infants for ROP.
- Subjective extended screening offers minimal clinical value in tertiary neonatal care.
- Cautious omission of subjective high-risk clauses is recommended in advanced tertiary hospitals.
Objective:
To evaluate the efficacy of a single objective screening criterion-birth weight (BW) < 2000g or gestational age (GA) < 32 weeks-for retinopathy of prematurity (ROP) detection, and to provide evidence for reducing over-screening caused by subjective clinical judgment.
Methods:
This retrospective study enrolled 1,453 premature infants who underwent ROP screening at Sichuan Provincial People's Hospital (2014-2024). Fundus examinations were performed using binocular indirect ophthalmoscopy or RetCam III. A two-tier grouping strategy was adopted: objective criterion group (BW < 2000g or GA < 32 weeks) vs. subjective group (BW ≥ 2000g and GA ≥ 32 weeks). The objective group was further subdivided into three subgroups. ROP and severe ROP were diagnosed following Chinese Guidelines for the Screening of Retinopathy of Prematurity (2014), with severe ROP defined as type 1 ROP requiring clinical treatment. The study population was stratified into inborn infants and referred infants for sensitivity analysis. Multivariable logistic regression identified independent risk factors for severe ROP.
Results:
Among 1,453 infants (618 inborn, 836 referred), 705 ROP cases (48.5%) and 370 severe ROP cases (25.5%) were identified. The objective group covered 98.2% of ROP cases and 99.2% of severe ROP cases. ROP and severe ROP prevalence were significantly higher in the objective group than in the subjective group (both P < 0.001). Multivariable regression identified lower GA (OR = 1.42 per week decrease) and lower BW (OR = 1.21 per 100 g decrease) as independent risk factors for severe ROP (both P < 0.001); sex, case source, and birth year were not significant. All three severe ROP cases in the subjective group were diagnosed in or before 2017, with no new cases reported after 2017.
Conclusion:
The single objective criterion of BW < 2000g or GA < 32 weeks can effectively identify the high-risk population for ROP. Subjective extended screening has minimal clinical value in tertiary neonatal care institutions; it is recommended that the subjective high-risk clause may be cautiously omitted in clinical practice within tertiary hospitals with advanced neonatal care capabilities.
