Neonatal hyperglycemia as an independent risk factor for retinopathy of prematurity: a multicenter retrospective cohort study with dose-response analysis
4 affiliations
- 1Neonatology Department, Yinchuan Maternal and Child Health Hospital Affiliated to Ningxia Medical University, Yinchuan, Ningxia Hui Autonomous Region, China
- 2Neonatology Department, People’s Hospital of Ningxia Hui Autonomous Region, Yinchuan, Ningxia Hui Autonomous Region, China
- 3Neonatology Department, General Hospital of Ningxia Medical University, Yinchuan, Ningxia Hui Autonomous Region, China
- 4Neonatology Department, Guyuan Municipal Maternal and Child Health Hospital, Guyuan, Ningxia Hui Autonomous Region, China
Abstract
BackgroundRetinopathy of prematurity (ROP) is a major cause of childhood visual impairment. Although neonatal hyperglycemia has drawn research interest, its independent effect on ROP remains controversial. This study explores the association between neonatal hyperglycemia and ROP among infants with gestational age <32 weeks.MethodsIn this multi-center retrospective cohort of 308 preterm infants, multivariable Cox proportional-hazards models adjusted for perinatal factors, maternal comorbidities and neonatal complications. Sensitivity analyses further accounted for invasive/non-invasive ventilation duration, total oxygen therapy time, peak inspired oxygen fraction and glucose infusion rate; Firth penalized regression corrected sparse-data bias. Additional sensitivity analyses examined the dose-response association with neonatal peak glucose levels.ResultsOur findings revealed that hyperglycemic infants had significantly lower birth weights and gestational ages compared to their non-hyperglycemic counterparts. After controlling for confounding variables, hyperglycemia was identified as an independent risk factor for ROP, with an adjusted hazard ratio (aHR) of 2.54 (95% CI: 1.45–4.44, P = 0.001). This association persisted after further adjustment for respiratory support variables and glucose infusion rate (aHR=2.46, 95% CI: 1.40–4.31) and in Firth penalized analysis (aHR=2.36, 95% CI: 1.36–4.11). Notably, a pronounced dose-response relationship was detected, with risks escalating markedly at peak glucose levels exceeding 8.33 mmol/L (150 mg/dL) (aHR=2.72) and intensifying further above 10 mmol/L (180 mg/dL) (aHR=3.64), indicating an approximately 50%–80% average risk increase per glucose tier across model specifications (all P-trend < 0.001), a gradient that was unchanged after further adjustment for neonatal complications, respiratory support, and glucose infusion rate. In exploratory analysis, severe ROP (stage ≥3, n = 13) was not significantly associated with hyperglycemia after penalized adjustment (OR = 1.45, 95% CI: 0.36–7.31).ConclusionThis study establishes neonatal hyperglycemia as a significant, dose-dependent predictor of ROP and associated multi-organ injury in preterm infants, independent of traditional risk factors. Clinically, this underscores the imperative for stringent glycemic monitoring, with a candidate vigilance threshold of 8.33 mmol/L (150 mg/dL), the clinical benefit of which requires prospective validation. Future prospective studies integrating continuous glucose monitoring and mechanistic investigations are essential to refine glycemic thresholds and optimize metabolic management strategies aimed at enhancing visual outcomes in this susceptible population.
Related
- Explainability-informed patient-disjoint evaluation of deep learning for plus disease in retinopathy of prematurity with an ophthalmologist reader study
- Predicting Retinopathy of Prematurity in Thailand: A Practical Model for Risk-Based Screening
- Neurodevelopmental Disorders in Infants Treated with Anti-VEGF Agents versus Laser Photocoagulation or Cryotherapy for Retinopathy of Prematurity
- Reduced serum Dickkopf-1 levels as a potential supportive biomarker for the development and severity of retinopathy of prematurity: a prospective observational clinical study
- Stress hyperglycemia ratio is associated with delirium in elderly cardiac surgery patients in the postoperative intensive care unit: a multicenter retrospective cohort study.
BibTeX
@article{wang2026neonatal,
title = {{Neonatal hyperglycemia as an independent risk factor for retinopathy of prematurity: a multicenter retrospective cohort study with dose-response analysis}},
author = {Yongming Wang and Xiaoqin Wang and Huijuan Yin and Fengzhi Xu and Wang Xu and Jianping He},
journal = {Frontiers in Pediatrics},
year = {2026},
month = oct,
doi = {10.3389/fped.2026.1975596},
url = {https://doi.org/10.3389/fped.2026.1975596},
note = {CC BY},
}
RIS
TY - JOUR
TI - Neonatal hyperglycemia as an independent risk factor for retinopathy of prematurity: a multicenter retrospective cohort study with dose-response analysis
AU - Yongming Wang
AU - Xiaoqin Wang
AU - Huijuan Yin
AU - Fengzhi Xu
AU - Wang Xu
AU - Jianping He
PY - 2026
DA - 2026/10/23
JO - Frontiers in Pediatrics
DO - 10.3389/fped.2026.1975596
UR - https://doi.org/10.3389/fped.2026.1975596
AB - BackgroundRetinopathy of prematurity (ROP) is a major cause of childhood visual impairment. Although neonatal hyperglycemia has drawn research interest, its independent effect on ROP remains controversial. This study explores the association between neonatal hyperglycemia and ROP among infants with gestational age <32 weeks.MethodsIn this multi-center retrospective cohort of 308 preterm infants, multivariable Cox proportional-hazards models adjusted for perinatal factors, maternal comorbidities and neonatal complications. Sensitivity analyses further accounted for invasive/non-invasive ventilation duration, total oxygen therapy time, peak inspired oxygen fraction and glucose infusion rate; Firth penalized regression corrected sparse-data bias. Additional sensitivity analyses examined the dose-response association with neonatal peak glucose levels.ResultsOur findings revealed that hyperglycemic infants had significantly lower birth weights and gestational ages compared to their non-hyperglycemic counterparts. After controlling for confounding variables, hyperglycemia was identified as an independent risk factor for ROP, with an adjusted hazard ratio (aHR) of 2.54 (95% CI: 1.45–4.44, P = 0.001). This association persisted after further adjustment for respiratory support variables and glucose infusion rate (aHR=2.46, 95% CI: 1.40–4.31) and in Firth penalized analysis (aHR=2.36, 95% CI: 1.36–4.11). Notably, a pronounced dose-response relationship was detected, with risks escalating markedly at peak glucose levels exceeding 8.33 mmol/L (150 mg/dL) (aHR=2.72) and intensifying further above 10 mmol/L (180 mg/dL) (aHR=3.64), indicating an approximately 50%–80% average risk increase per glucose tier across model specifications (all P-trend < 0.001), a gradient that was unchanged after further adjustment for neonatal complications, respiratory support, and glucose infusion rate. In exploratory analysis, severe ROP (stage ≥3, n = 13) was not significantly associated with hyperglycemia after penalized adjustment (OR = 1.45, 95% CI: 0.36–7.31).ConclusionThis study establishes neonatal hyperglycemia as a significant, dose-dependent predictor of ROP and associated multi-organ injury in preterm infants, independent of traditional risk factors. Clinically, this underscores the imperative for stringent glycemic monitoring, with a candidate vigilance threshold of 8.33 mmol/L (150 mg/dL), the clinical benefit of which requires prospective validation. Future prospective studies integrating continuous glucose monitoring and mechanistic investigations are essential to refine glycemic thresholds and optimize metabolic management strategies aimed at enhancing visual outcomes in this susceptible population.
SN - 2296-2360
ER -