Analysis of risk factors for retinopathy of prematurity: a single-center retrospective cohort study

Introduction

Retinopathy of prematurity (ROP) remains a leading cause of preventable childhood blindness worldwide, particularly in countries with developing neonatal intensive care systems (1). Despite significant advances in neonatal care over the past decades, the incidence of ROP has not decreased proportionally, largely due to the improved survival rates of extremely premature infants (2). ROP incidence varies significantly by geographic region, with resource-limited areas often reporting higher rates than developed countries. For example, studies from rural tertiary hospitals in Thailand have demonstrated regional variations in ROP incidence, emphasizing the importance of understanding local risk factor profiles (3).

The pathogenesis of ROP is complex and multifactorial, involving two distinct phases of vascular development. The initial phase is characterized by delayed retinal vascular growth and partial regression of existing vessels, while the second phase involves pathological neovascularization (4). Multiple risk factors have been identified that may influence these processes, with prematurity and low birth weight being the most well-established (5).

Recent evidence suggests that beyond these traditional risk factors, various perinatal and postnatal factors may contribute to ROP development. These include prolonged oxygen therapy, mechanical ventilation, sepsis, bronchopulmonary dysplasia (BPD), and poor postnatal weight gain (6). The role of systemic inflammation, oxidative stress, and nutritional factors has also gained increasing attention in recent years (7). Furthermore, genetic predisposition and racial differences have been suggested to influence ROP susceptibility (8).

The timing and intensity of interventions in the neonatal intensive care unit (NICU) may significantly impact ROP development. Recent studies have highlighted the importance of careful oxygen management, optimal nutrition, and prevention of comorbidities in reducing ROP risk (9). However, the relative contribution of these various factors and their interactions remain incompletely understood, particularly in different healthcare settings and populations (10). Early identification of infants at high risk for ROP is crucial for timely screening and intervention. While current screening guidelines primarily rely on gestational age and birth weight, incorporating additional risk factors could potentially improve risk stratification (11). Moreover, understanding modifiable risk factors could inform preventive strategies and optimize NICU protocols (12).

The purpose of this study was to analyze the perinatal and postnatal risk factors associated with ROP development in premature infants in a single tertiary care center, with particular attention to potentially modifiable factors that could inform prevention strategies. This article is presented in accordance with the STROBE reporting checklist (available at https://tp.amegroups.com/article/view/10.21037/tp-2025-440/rc).

Methods

Study design and setting

This retrospective cohort study was conducted at the NICU of The Second Affiliated Hospital of Shantou University Medical College, a tertiary care center, from April 2020 to June 2023. Our facility is an 80-bed level III NICU providing comprehensive care for high-risk newborns, with approximately 1,600 admissions annually. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study protocol was approved by the Institutional Review Board of The Second Affiliated Hospital of Shantou University Medical College [IRB No. (2025-161)]. Informed consent was obtained from the legal guardians of all participating infants prior to their inclusion in the study.

Study population and sample size

The inclusion criteria were established based on current national guidelines for ROP screening. We included infants with gestational age ≤30 weeks or birth weight <1,500 g, as well as those with birth weight 1,500–2,000 g or gestational age >30 weeks who had an unstable clinical course, consistent with American Academy of Pediatrics screening recommendations (13). The initial sample size calculation, based on an estimated ROP incidence of 20% and a desired precision of 5%, indicated a minimum required sample size of 246 infants. We excluded infants who died before their first ROP screening, those with major congenital anomalies affecting systemic development, chromosomal abnormalities, and cases with incomplete medical records or missing key variables.

ROP screening protocol and diagnosis

Initial ROP screening was performed at 4–6 weeks postnatal age or at 31–32 weeks postmenstrual age, whichever came later. Follow-up examinations were scheduled according to International Classification of Retinopathy of Prematurity (ICROP) criteria: weekly for immature retina in zone I, every 1–2 weeks for zone II disease, and every 3–4 weeks for zone III vascularization until complete retinal vascularization or disease regression was documented.

Prior to examination, pupils were dilated using a standardized protocol of tropicamide 0.5% eye drops administered three times at 10-minute intervals. All examinations were performed by Chief Physician of Shantou International Ophthalmology Center with over 10 years of experience in ROP screening, using binocular indirect ophthalmoscopy with a 28-diopter lens and scleral depression when necessary. The diagnosis and staging of ROP followed the ICROP criteria, including detailed documentation of the zone, stage, extent of involvement, and presence or absence of plus disease.

Data collection process

Our data collection process employed a dual-reviewer system with standardized electronic forms developed through expert consensus. Two trained research nurses independently extracted data from electronic medical records, with a third senior neonatologist resolving any discrepancies. Regular quality control checks were performed on 10% of randomly selected records to ensure data accuracy. The electronic data collection form underwent pilot testing with 20 cases before full implementation, achieving an inter-rater reliability coefficient of 0.92.

Comprehensive variable assessment and measurement

All physiological parameters were measured using standardized equipment and protocols. Vital signs were recorded using continuous monitoring systems calibrated according to manufacturer specifications. Blood pressure measurements were obtained using non-invasive oscillometric devices, with readings documented hourly or more frequently as clinically indicated. Temperature regulation was maintained using servo-controlled incubators with continuous skin temperature monitoring.

The Neonatal Behavioral Neurological Assessment (NBNA) was performed according to standardized protocols, with a total score ranging from 0 to 40 points, where higher scores indicate better neurobehavioral status. Shock was defined as clinical signs of poor perfusion requiring vasopressor support, with or without elevated lactate levels, as documented by the attending neonatologist during the clinical course.

Respiratory support and monitoring

Respiratory support parameters were meticulously documented, including specific details of both non-invasive and invasive ventilation. Non-invasive respiratory support included nasal continuous positive airway pressure (nCPAP) and high-flow nasal cannula (HFNC), with detailed recording of pressure settings, flow rates, and fraction of inspired oxygen (FiO₂) requirements. For mechanically ventilated infants, we documented ventilator modes, peak inspiratory pressures, positive end-expiratory pressures, mean airway pressures, and tidal volumes. Oxygen saturation targets were maintained between 90–95% using standardized protocols, with continuous recording of oxygen saturation trends and documentation of deviation episodes. Oxygen saturation was continuously monitored and recorded hourly, with deviations below 85% or above 95% documented as significant events. Episodes of desaturation were categorized as mild (85–89%), moderate (80–84%), or severe (<80%), and their frequency, duration, and timing were analyzed in relation to clinical interventions.

Nutritional management protocol

The nutritional management followed a standardized protocol based on current international guidelines. Parenteral nutrition was initiated within the first 24 hours of life using a standardized solution containing amino acids, dextrose, and lipids. The composition and advancement of parenteral nutrition followed unit protocols, with regular monitoring of biochemical parameters. Enteral feeding was initiated according to The Second Affiliated Hospital of Shantou University Medical College’s protocol, with detailed documentation of feeding type, volume, and advancement rates. Human milk was fortified according to standardized protocols when feeding volume reached 50–100 mL/kg/day.

Laboratory monitoring and biochemical assessment

Blood sampling followed a predetermined schedule to minimize iatrogenic blood loss. Routine blood samples were collected at admission, then at 24, 48, and 72 hours of life, followed by twice weekly monitoring or more frequently as clinically indicated. Magnesium and phosphate measurements were included in the routine laboratory panel for all patients and were not selectively obtained. Complete blood counts, comprehensive metabolic panels, and blood gas analyses were performed using standardized laboratory equipment. Blood gas analyses were primarily performed using arterial samples when arterial access was available, or capillary samples in other cases, with the method clearly documented. Analyses were conducted at admission, then daily for the first 72 hours, and subsequently as clinically indicated, with particular attention to periods of respiratory deterioration or changes in ventilatory support. Specific thresholds were established for metabolic derangements: hypocalcemia was defined as serum calcium below 1.8 mmol/L, hypomagnesemia as serum magnesium below 0.6 mmol/L, and hypophosphatemia as serum phosphate below 0.85 mmol/L. Blood gas analyses were performed using point-of-care testing with regular quality control checks.

Statistical analysis

Statistical analyses were performed using SPSS version 25.0 (IBM Corp., Armonk, NY, USA). Continuous variables were expressed as means with standard deviations or medians with interquartile ranges, depending on their distribution. Categorical variables were presented as frequencies and percentages. The normality of continuous variables was assessed using the Shapiro-Wilk test. Comparisons between ROP and non-ROP groups were conducted using Student’s t-test or Mann-Whitney U test for continuous variables, and Chi-squared or Fisher’s exact test for categorical variables, as appropriate.

Variables with P<0.05 in bivariate analysis were included in the multivariate logistic regression model. Gestational age, birth weight, and sex were forced into the model as a priori confounders to control for confounding. Given the limited number of ROP events (n=17), we carefully considered the events-per-variable ratio to avoid overfitting. Model fit was evaluated using the Hosmer-Lemeshow test, and discrimination was assessed through the area under the receiver operating characteristic curve. Results were expressed as adjusted odds ratios with 95% confidence intervals, and P<0.05 was considered statistically significant.

Results

Demographic and basic clinical characteristics

Among the 346 preterm infants included in this study, 17 (4.91%) developed ROP and 329 (95.09%) did not. As shown in Table 1, there were no significant differences in sex distribution, birth weight, or gestational age between groups. Blood gas analysis showed no significant difference in partial pressure of oxygen in arterial blood (PaO₂) levels between groups (P=0.45). NBNA scores were significantly lower in the ROP group (25.35±2.45 vs. 26.81±2.76; P=0.03).

ROP characteristics and interventions

Among the 17 infants who developed ROP, the distribution by severity included 9 (52.9%) with stage 1, 5 (29.4%) with stage 2, and 3 (17.6%) with stage 3. Zone II was most commonly affected (76.5%), followed by zone III (17.6%) and zone I (5.9%). Plus disease was observed in 2 (11.8%) cases. Regarding interventions, 3 (17.6%) infants required laser photocoagulation therapy, while none needed anti-vascular endothelial growth factor (anti-VEGF) therapy or surgical intervention. Exploratory analysis of severe ROP (n=3) suggested possible associations with hypophosphatemia, prolonged mechanical ventilation, and shock episodes; however, given the extremely limited sample size, these findings are hypothesis-generating only and cannot support robust conclusions.

Maternal and pregnancy-related factors

Table 2 demonstrates that maternal and pregnancy-related characteristics showed no significant differences between groups.

Clinical interventions and respiratory support

Significant findings from Table 3 included longer duration of non-invasive ventilation in the ROP group (14.3±12.8 vs. 7.4±6.7 days; P=0.004), longer parenteral nutrition duration (35.0±17.10 vs. 24.60±15.19 days; P=0.006), longer time to achieve full enteral feeding (35.0±17.10 vs. 24.64±15.17 days; P=0.006), higher advanced antibiotic use (76.47% vs. 37.99%; P=0.003), and lower steroid use (47.06% vs. 82.07%; P=0.001).

Clinical complications and metabolic disorders

Significant findings from Table 4 included higher rates of BPD (64.71% vs. 21.88%; P<0.001), shock (76.47% vs. 35.26%; P=0.001), hypophosphatemia (29.41% vs. 4.86%; P<0.001), intraventricular hemorrhage (IVH) (41.18% vs. 17.02%; P=0.02), and apnea (41.18% vs. 17.33%; P=0.03) in the ROP group. Hypoxic-ischemic encephalopathy (HIE) showed a trend toward higher prevalence in the ROP group but was not statistically significant (70.59% vs. 55.02%; P=0.06).

Multivariate analysis

Variables with P<0.05 in bivariate analysis were included in the multivariate logistic regression model. Gestational age, birth weight, and sex were forced into the model as a priori confounders. The final model (Table 5) identified the following significant independent risk factors: duration of non-invasive ventilation [odds ratio (OR) =1.49; 95% confidence interval (CI): 1.2–1.8; P=0.002], hospital stay (OR =1.35; 95% CI: 1.2–1.5; P=0.005), hypophosphatemia (OR =2.29; 95% CI: 1.2–3.5; P<0.001), and apnea (OR =2.01; 95% CI: 1.2–2.1; P=0.03). Other variables showed non-significant trends.

Discussion

In this single-center retrospective cohort study, we identified several significant risk factors associated with ROP development in preterm infants. Most notably, our findings highlight the crucial role of metabolic derangements, particularly hypophosphatemia, alongside traditional risk factors such as prolonged respiratory support and complications like BPD. Additionally, we observed that the duration of parenteral nutrition and time to achieve full enteral feeding were significantly associated with ROP development, suggesting the importance of optimal nutritional management in ROP prevention (14).

The association between hypophosphatemia and ROP development observed in our study represents a novel finding that warrants further investigation. While phosphate homeostasis has been extensively studied in preterm infant metabolism, its specific role in retinal vascular development has received limited attention. Recent research has suggested that phosphate deficiency may influence angiogenesis through disruption of energy metabolism and cellular signaling pathways (15). Phosphate plays a crucial role in ATP production and cellular energy metabolism, and its deficiency might impair the normal development of retinal vessels through metabolic stress (16). Furthermore, recent studies have shown that phosphate deficiency can activate hypoxia-inducible factors, potentially contributing to pathological angiogenesis in the developing retina (17).

The significant relationship between hypomagnesemia and ROP adds to the growing body of evidence linking mineral homeostasis to retinal vascular development. Magnesium, as a crucial cofactor in numerous enzymatic processes and an important regulator of oxidative stress, may influence retinal vascularization through multiple pathways (18). Experimental studies have demonstrated that magnesium deficiency can enhance free radical formation and increase susceptibility to oxidative stress, which are key mechanisms in ROP pathogenesis (19). Additionally, magnesium’s role in stabilizing cell membranes and regulating vascular tone might explain its potential protective effect against ROP development (20).

The relationship between prolonged non-invasive ventilation and ROP development observed in our study merits particular attention. While the benefits of non-invasive respiratory support in reducing lung injury are well-established, our findings suggest that the duration of such support may influence ROP risk independently of oxygen concentration (21). Recent evidence indicates that fluctuations in oxygen delivery, rather than absolute oxygen levels alone, may be critical in ROP pathogenesis (22). The extended duration of non-invasive support might increase the likelihood of such fluctuations, particularly during routine care activities (23). The higher frequency of oxygen saturation fluctuations observed in our ROP group aligns with findings from Nobile et al. [2014], who demonstrated that implementing stringent oxygen saturation targets and minimizing fluctuations significantly reduced severe ROP incidence from 5.3% to 1% over an 8-year period (24). Their study emphasized that maintaining tight control of oxygen saturation between carefully defined intervals by well-trained staff was more effective than simply lowering oxygen targets.

The significant association between prolonged parenteral nutrition and ROP risk emphasizes the complex relationship between nutrition and retinal vascular development (25). Early aggressive nutrition has been shown to protect against ROP by promoting appropriate weight gain and reducing oxidative stress (26). Our findings regarding delayed achievement of full enteral feeding in the ROP group suggest that optimizing the transition from parenteral to enteral nutrition might represent a modifiable risk factor (27). Recent research has demonstrated that specific nutrients, including long-chain polyunsaturated fatty acids and antioxidants, may play protective roles in retinal development (28).

The higher incidence of BPD in our ROP group supports the concept of “oxygen radical disease of prematurity”, where multiple organs are affected by oxidative stress and inflammation (29). This association suggests common pathophysiological mechanisms underlying both conditions and emphasizes the importance of considering ROP prevention within a broader framework of reducing oxidative stress and inflammation in preterm infants (30). The significant relationship between shock episodes and ROP development indicates that systemic circulatory compromise may contribute to retinal vascular pathology through altered tissue perfusion and oxygenation (31).

The role of advanced antibiotic use in our ROP group raises important questions about the impact of inflammation and infection on retinal vascular development (32). Recent studies have suggested that systemic inflammation may modify the expression of angiogenic factors in the developing retina (33). Furthermore, the association between prolonged antibiotic exposure and altered gut microbiota might influence systemic inflammation and oxidative stress levels (34).

Limitations

The limitations of our study include its retrospective nature and single-center design, which may limit the generalizability of our findings. The relatively small number of ROP cases (n=17) in our cohort raises concerns about potential overfitting in the multivariate model, as evidenced by the events-per-variable ratio. This may lead to unstable estimates and overestimation of certain effects. Additionally, we were unable to assess certain potentially important variables, such as genetic factors and detailed oxygen saturation targeting practices. The frequency of laboratory testing, although following routine protocols, may have varied based on clinical status, potentially introducing detection bias for metabolic abnormalities. The direction and magnitude of this bias are difficult to quantify but could either overestimate or underestimate the true associations. Residual confounding from unmeasured variables cannot be excluded. Future prospective studies with larger sample sizes are needed to confirm our findings and better understand the mechanistic relationships between metabolic disorders and ROP development.

Conclusions

Our study identifies several potentially modifiable risk factors associated with ROP development in preterm infants, particularly highlighting the significance of metabolic derangements such as hypophosphatemia. The associations between prolonged respiratory support, delayed enteral feeding achievement, and increased ROP risk underscore the importance of optimizing both respiratory and nutritional management strategies. While our findings are limited by the study’s retrospective nature and small sample size, they suggest that careful attention to metabolic parameters, alongside established protocols for oxygen and nutritional management, may offer new opportunities for ROP prevention.

Acknowledgments

None.

Footnote

Reporting Checklist: The author has completed the STROBE reporting checklist. Available at https://tp.amegroups.com/article/view/10.21037/tp-2025-440/rc

Data Sharing Statement: Available at https://tp.amegroups.com/article/view/10.21037/tp-2025-440/dss

Peer Review File: Available at https://tp.amegroups.com/article/view/10.21037/tp-2025-440/prf

Funding: This study was funded by the Science and Technology of Shantou (Nos. 2017-183 and 2021-68).

Conflicts of Interest: The author has completed the ICMJE uniform disclosure form (available at https://tp.amegroups.com/article/view/10.21037/tp-2025-440/coif). The author has no conflicts of interest to declare.

Ethical Statement: The author is accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study protocol was approved by the Institutional Review Board of The Second Affiliated Hospital of Shantou University Medical College [IRB No. (2025-161)]. Informed consent was obtained from the legal guardians of all participating infants prior to their inclusion in the study.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Chiang MF, Quinn GE, Fielder AR, et al. International Classification of Retinopathy of Prematurity, Third Edition. Ophthalmology 2021;128:e51-68.
2. Dammann O, Hartnett ME, Stahl A. Retinopathy of prematurity. Dev Med Child Neurol 2023;65:625-31. [Crossref] [PubMed]
3. Ittarat M, Chansaengpetch S, Chansangpetch S. Incidence and Risk Factors for Retinopathy of Prematurity at a Rural Tertiary Hospital in Thailand. J Ophthalmic Vis Res 2023;18:81-7. [Crossref] [PubMed]
4. Daruich A, Bremond-Gignac D, Behar-Cohen F, et al. Retinopathy of prematurity: from prevention to treatment. Med Sci (Paris) 2020;36:900-7. [Crossref] [PubMed]
5. Sabri K, Ells AL, Lee EY, et al. Retinopathy of Prematurity: A Global Perspective and Recent Developments. Pediatrics 2022;150:e2021053924. [Crossref] [PubMed]
6. Bancalari A, Schade R. Update in the Treatment of Retinopathy of Prematurity. Am J Perinatol 2022;39:22-30. [Crossref] [PubMed]
7. Olevson C, Tufty G. Review of Retinopathy of Prematurity. S D Med 2023;76:372-5.
8. Kim ES, Calkins KL, Chu A. Retinopathy of Prematurity: The Role of Nutrition. Pediatr Ann 2023;52:e303-8. [Crossref] [PubMed]
9. Sen P, Wu WC, Chandra P, et al. Retinopathy of prematurity treatment: Asian perspectives. Eye (Lond) 2020;34:632-42. [Crossref] [PubMed]
10. Gilbert CE. Global perspectives of retinopathy of prematurity. Indian J Ophthalmol 2023;71:3431-3. [Crossref] [PubMed]
11. Dammann O, Rivera JC, Chemtob S. The prenatal phase of retinopathy of prematurity. Acta Paediatr 2021;110:2521-8. [Crossref] [PubMed]
12. Scruggs BA, Chan RVP, Kalpathy-Cramer J, et al. Artificial Intelligence in Retinopathy of Prematurity Diagnosis. Transl Vis Sci Technol 2020;9:5. [Crossref] [PubMed]
13. Fierson WM, et al. Screening Examination of Premature Infants for Retinopathy of Prematurity. Pediatrics 2018;142:e20183061. [Crossref] [PubMed]
14. Ahmed I, Hoyek S, Patel NA. Global Disparities in Retinopathy of Prematurity: A Literature Review. Semin Ophthalmol 2023;38:151-7. [Crossref] [PubMed]
15. Dai C, Xiao J, Wang C, et al. Neurovascular abnormalities in retinopathy of prematurity and emerging therapies. J Mol Med (Berl) 2022;100:817-28. [Crossref] [PubMed]
16. Fielder AR, Quinn GE, Paul Chan RV, et al. Retinopathy of prematurity classification updates: possible implications for treatment. J AAPOS 2022;26:109-12. [Crossref] [PubMed]
17. Martínez-Castellanos MA, Ortiz-Ramirez GY. Surgery for stage 5 retinopathy of prematurity. Curr Opin Ophthalmol 2021;32:482-8. [Crossref] [PubMed]
18. Li JQ, Kellner U, Lorenz B, et al. Retinopathy of Prematurity - Update on Classification, Screening, and Therapy. Klin Monbl Augenheilkd 2022;239:346-63. [Crossref] [PubMed]
19. Ge G, Zhang Y, Zhang M. Pregnancy-induced hypertension and retinopathy of prematurity: a meta-analysis. Acta Ophthalmol 2021;99:e1263-73. [Crossref] [PubMed]
20. Kvopka M, Chan W, Lake SR, et al. Fundus fluorescein angiography imaging of retinopathy of prematurity in infants: A review. Surv Ophthalmol 2023;68:849-60. [Crossref] [PubMed]
21. Sharma A, Shetty A, Reddy YCVG. Bevacizumab in Retinopathy of Prematurity : Concerns and adverse effects. Nepal J Ophthalmol 2020;12:298-307. [Crossref] [PubMed]
22. Chang E, Rao P. Adult retinopathy of prematurity: treatment implications, long term sequelae, and management. Curr Opin Ophthalmol 2021;32:489-93. [Crossref] [PubMed]
23. Kumawat D, Sachan A, Shah P, et al. Aggressive posterior retinopathy of prematurity: a review on current understanding. Eye (Lond) 2021;35:1140-58. [Crossref] [PubMed]
24. Nobile S, Gnocchini F, Pantanetti M, et al. The importance of oxygen control reaffirmed: experience of ROP reduction at a single tertiary care center. J Pediatr Ophthalmol Strabismus 2014;51:112-5. [Crossref] [PubMed]
25. Fu Z, Nilsson AK, Hellstrom A, et al. Retinopathy of prematurity: Metabolic risk factors. Elife 2022;11:e80550. [Crossref] [PubMed]
26. Godugu S, Vudugula SA, Neupane B, et al. Association between severe retinopathy of prematurity (ROP) and poor motor neurodevelopmental outcome. Indian J Ophthalmol 2023;71:2944-6. [Crossref] [PubMed]
27. Lei C, Duan J, Ge G, et al. Association between neonatal hyperglycemia and retinopathy of prematurity: a meta-analysis. Eur J Pediatr 2021;180:3433-42. [Crossref] [PubMed]
28. Fu Y, Lei C, Qibo R, et al. Insulin-like growth factor-1 and retinopathy of prematurity: A systemic review and meta-analysis. Surv Ophthalmol 2023;68:1153-65. [Crossref] [PubMed]
29. Tsai AS, Chou HD, Ling XC, et al. Assessment and management of retinopathy of prematurity in the era of anti-vascular endothelial growth factor (VEGF). Prog Retin Eye Res 2022;88:101018. [Crossref] [PubMed]
30. Raghuveer TS, Zackula R. Strategies to Prevent Severe Retinopathy of Prematurity: A 2020 Update and Meta-analysis. Neoreviews 2020;21:e249-63. [Crossref] [PubMed]
31. Hoyek S, Cruz NFSD, Patel NA, et al. Identification of novel biomarkers for retinopathy of prematurity in preterm infants by use of innovative technologies and artificial intelligence. Prog Retin Eye Res 2023;97:101208. [Crossref] [PubMed]
32. Barrero-Castillero A, Corwin BK, VanderVeen DK, et al. Workforce Shortage for Retinopathy of Prematurity Care and Emerging Role of Telehealth and Artificial Intelligence. Pediatr Clin North Am 2020;67:725-33. [Crossref] [PubMed]
33. Wood EH, Chang EY, Beck K, et al. 80 Years of vision: preventing blindness from retinopathy of prematurity. J Perinatol 2021;41:1216-24. [Crossref] [PubMed]
34. Cota F, Costa S, Giannantonio C, et al. Lutein supplementation and retinopathy of prematurity: a meta-analysis. J Matern Fetal Neonatal Med 2022;35:175-80. [Crossref] [PubMed]