Predicting the treatment-requiring retinopathy of prematurity using birth weight, laboratory data and continuous peripheral capillary oxygen saturation monitoring
Related Experiment Video
Updated: May 6, 2026
Retinal Vascular Reactivity as Assessed by Optical Coherence Tomography Angiography
Published on: March 26, 2020
Taku Toyama1,2, Han Peng Zhou3, Gen Mihara3
1Department of Ophthalmology, Graduate School of Medicine and Faculty of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-8655, Japan. toyama.ophthalmol@gmail.com.
Purpose:
This single-center retrospective study aimed to identify independent predictors of retinopathy of prematurity (ROP) treatment by integrating continuous pulse oximetry (SpO₂) data with comprehensive laboratory parameters.
Study Design:
Single-center retrospective cohort study.
Methods:
We analyzed 244 neonates (gestational age < 34 weeks or birth weight < 1800 g) admitted between 2019 and 2023, excluding those with major genetic abnormalities. Logistic regression assessed associations between birth weight (BW), average SpO₂ at birth, hemoglobin (Hb), albumin (Alb), and selected inflammatory markers (e.g., neutrophil-to-lymphocyte ratio) with ROP treatment. Model discrimination was evaluated via receiver operating characteristic curves.
Results:
Of the 244 infants, 16.8% required treatment (laser therapy or anti-VEGF injections). Lower BW and lower Hb were significantly associated with treatment (P <0.05). In the final multivariable model, BW and Hb emerged as key predictors (AUC = 0.884; accuracy = 88%), while Alb and inflammatory indices were not independently significant. Predicted treatment probabilities decreased markedly as BW and Hb increased.
Conclusion:
BW and Hb were primary predictors of ROP treatment necessity in this cohort, whereas Alb and inflammatory markers did not show independent associations. Incorporating continuous SpO₂ monitoring provided valuable respiratory insights. Further prospective, multicenter studies are warranted to validate these findings and refine ROP risk stratification strategies.
