Complete blood count parameters as biomarkers of retinopathy of prematurity: a Portuguese multicenter study

Prediction of retinopathy of prematurity development and treatment need with machine learning models

BACKGROUND

To evaluate the effectiveness of machine learning (ML) models in predicting the occurrence of retinopathy of prematurity (ROP) and treatment need.

METHODS

Four ML models were created using 49 parameters known within the first 24 h post-birth and obtained during the initial screening examination, encompassing demographic, maternal, clinical, and neonatal intensive care unit-related data. The models' performances were assessed using five machine learning (ML) classifier algorithms: logistic regression (LR), decision tree (DT), support vector machine (SVM), random forest (RF), and extreme gradient boosting (XGBoost). Performance metrics were calculated, and the top ten parameters with the highest predictive value were identified.

RESULTS

In the cohort of 355 preterm infants, Model I, predicting ROP development using birth data, achieved a balanced accuracy of 80%, with gestational age (GA), birth weight (BW) and mean corpuscular volume (MCV) as the top predictive parameters. Model II, predicting treatment-requiring ROP using birth data, exhibited a balanced accuracy of 81%. Key predictive parameters included low GA, BW, 1-minute and 5-minute APGAR scores, and low erythrocyte counts. For Model III, predicting ROP using the first screening examination data, and Model IV, predicting treatment-requiring ROP using the same data, the accuracy values were 80% and 66%, respectively, with BW, daily weight gain, total O2 support duration, and platelet/lymphocyte ratio emerged as the most significant predictive parameters in both models.

CONCLUSION

This study demonstrates the potential of ML models to predict ROP development and treatment need. Incorporating clinical and intensive care-related parameters can enhance ROP screening and clinical decision-making.

Influence of Functional Variations in Genes of Neurotrophins and Neurotransmitter Systems on the Development of Retinopathy of Prematurity

Retinal neurodevelopment, vascularization, homeostasis, and stress response are influenced by factors such as nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), tyrosine hydroxylase (TH), and erythropoietin (EPO). As retinopathy of prematurity (ROP) is a neurovascular retinal disease, this study analyzed the contributions of NGF (rs6330), BDNF (rs7934165), TH (rs10770141), and EPO (rs507392) genetic functional polymorphisms to the modulation of hematological and biochemical parameters of the first week of life and their association with ROP development. A multicenter cohort of 396 preterm infants (gestational age < 32 weeks or birth weight < 1500 g) was genotyped using MicroChip DNA and iPlex MassARRAY® platform. Multivariate regression followed univariate assessment of ROP risk factors. NGF (GG) genotype was associated with a higher ROP risk (OR = 1.79), which increased further (OR = 2.38) when epistatic interactions with TH (allele C) and BDNF (allele G) were present. Significant circulating biomarker differences, including bilirubin, erythrocytes, monocytes, neutrophils, lymphocytes, and platelet markers, were found between ROP and non-ROP groups, with variations depending on the polymorphism. These findings suggest that NGF (rs6330) and its interactions with related genes contribute to ROP risk, providing valuable insights into the genetic and biological mechanisms underlying the disease and identifying potential predictive biomarkers.

Platelet count and retinopathy of prematurity, a systematic review and meta-analysis

OBJECTIVE

This study aims to investigate the relationship between platelet count (PLT) and retinopathy of prematurity (ROP), with the goal of identifying a straightforward screening method for the early detection of ROP.

METHODS

A systematic search was conducted in PubMed, EMBASE, Web of Science, Cochrane Library, and ClinicalTrials.gov from January 2005 to 26 September 2023. Subsequently, a subgroup analysis was performed.

RESULTS

This meta-analysis included ten studies involving 1762 neonates, with 747 cases of ROP identified. The pooled analysis indicated a significant decrease in mean platelet count among the ROP group compared to the non-ROP group (mean difference: -18.65, 95% confidence interval: -22.80 to -14.50, P < .00001). Subgroup analysis indicated that the heterogeneity in the results can be attributed to factors such as the location of the studies, gestational age (GA), and publication time.

CONCLUSION

Decreased PLT counts may be associated with the occurrence of ROP. The PLT count should be standardized according to the detection location, detection time and GA.

Fetal Hemoglobin as a Predictive Biomarker for Retinopathy of Prematurity: A Prospective Multicenter Cohort Study in Portugal

Background/Objectives: Retinopathy of prematurity (ROP) is a leading cause of vision impairment in preterm infants, with its pathogenesis linked to oxygen exposure. Red blood cell (RBC) transfusions, commonly performed in neonatal intensive care units (NICUs), reduce fetal hemoglobin (HbF) fraction, altering oxygen dynamics and potentially contributing to ROP. We aimed to investigate the relationship between RBC transfusions, HbF percentage, and ROP, evaluating HbF as a potential predictive biomarker. Methods: A multicenter, prospective study was conducted across eight Portuguese NICUs, involving infants born at <32 weeks gestational age (GA) or <1500 g. ROP staging followed the International Classification of ROP (ICROP2). Clinical data were collected during hospitalization, and HbF fractions were measured from blood samples in the first four weeks of life using standardized methods. Infants were stratified by ROP presence and treatment requirement. Statistical analysis was performed using SPSS 28.0, with p < 0.05. Results: Eighty-two infants (mean GA: 28.1 ± 2.1 weeks, birth weight: 1055.8 ± 258.3 g) were included. Among them, 29 (35.4%) presented ROP and 4 (4.9%) required treatment. Infants with ROP had more RBC transfusions and lower HbF percentages than those without ROP (p < 0.05). Lower HbF was associated with more RBC transfusions (p < 0.001). Kaplan-Meier survival curves showed a higher ROP risk in infants with reduced HbF (p < 0.05). Conclusions: Low HbF percentage in the first four weeks of life may increase ROP risk in preterm infants. HbF could serve as a biomarker for ROP prediction. Interventions preserving HbF may reduce ROP risk. Further studies are needed to validate HbF as a biomarker and refine prevention strategies.

Red blood cell parameters as biomarkers of retinopathy of prematurity in preterm infants born before 30 weeks of gestation

Retinopathy of prematurity (ROP) is a major cause of preventable blindness in preterm infants. The association between red blood cell (RBC) parameters and the development of ROP remains unclear. The objectives of the present study were to evaluate the association between RBC parameters and ROP treatment. This single-center, retrospective cohort study included preterm infants born at < 30 weeks of gestation. Data pertaining to RBC parameters and ROP treatment were obtained from the medical records. A receiver operating characteristic (ROC) analysis was performed to determine the cut-off values of the RBC parameters according to the need for ROP treatment. Multiple logistic regression analyses assessed the association between ROP treatment and RBC parameters at birth and on day of life (DOL) 28. We included 202 infants, and 44.1% were treated for ROP. After adjusting for confounders, we observed associations between ROP treatment and mean corpuscular volume (MCV) values > 117.3 fL at birth (adjusted odds ratio [aOR]:2.3; 95% confidence intervals CI 1.0-5.3). Additionally, on DOL 28, hemoglobin (Hb) values < 9.9 g/dL (aOR:3.0; 95% CI 1.4-6.7), hematocrit (Hct) values < 31.0% (aOR:2.7; 95% CI 1.3-5.6), and red cell distribution width (RDW) values > 18.5% (aOR:2.6; 95% CI 1.1-6.2) were associated with ROP treatment. In conclusion, our study indicated that infants born at < 30 weeks of gestation with an MCV > 117.3 fL at birth, along with Hb < 9.9 g/dL, Hct < 31.0%, and RDW > 18.5% on DOL 28, had an increased risk of requiring ROP, warranting treatment.

The relationship between inflammatory markers and retinopathy of prematurity in extremely premature infants

PURPOSE

This study aimed to evaluate the relationships among blood parameters, clinical factors, and retinopathy of prematurity (ROP) in extremely premature (EP) infants.

METHODS

This retrospective study included 153 EP infants who were categorized into two groups based on the presence of inflammatory diseases such as necrotizing enterocolitis, neonatal sepsis, bronchopulmonary dysplasia, severe intraventricular hemorrhage, preeclampsia, and premature rupture of membranes. Complete blood count parameters, neutrophil-lymphocyte ratio (NLR), lymphocyte-monocyte ratio, platelet-lymphocyte ratio (PLR), systemic inflammatory index, and platelet mass index were recorded during the first week and first month after birth. The study analyzed the impact of parameters obtained through blood tests during the first week and first month on the development of ROP and the requirement for treatment.

RESULTS

In this study, 96 infants were diagnosed with inflammatory diseases. After multivariate regression analyses, the duration of mechanical ventilation (p = 0.010) was found to be the only factor that led to ROP development. Moreover, lower gestational age (GA) (p = 0.006), higher NLR (p = 0.026), and lower PLR (p = 0.019) were observed in infants requiring treatment compared to infants with spontaneous resolution of ROP in this group. 57 infants did not have inflammatory diseases. Although the duration of mechanical ventilation (p = 0.041) and low levels of platelets (PLT) (p = 0.046) measured in the first month postnatally were significantly found to be associated with ROP developement, no parameter affecting the required treatment could be determined.

CONCLUSION

EP infants with longer mechanical ventilation durations and lower PLT counts are vulnerable to ROP development. GA, PLR, and NLR are predictive factors for treatment.

Retinopathy of Prematurity in Eight Portuguese Neonatal Intensive Care Units: Incidence, Risk Factors, and Progression-A Prospective Multicenter Study

BACKGROUND/OBJECTIVES

Retinopathy of prematurity (ROP) is a retinal neovascular disease affecting preterm infants. Identifying risk factors for its development and progression is critical for effective screening and prevention. This study aimed to analyze the incidence of ROP and identify key risk factors for its development and progression.

METHODS

We conducted a prospective, observational cohort study on 455 neonates (gestational age [GA] < 32 weeks or birth weight < 1500 g) across eight Portuguese NICUs.

RESULTS

ROP incidence was 37.8%, with 4.6% requiring treatment. Multivariate analysis identified low GA and the number of red blood cell (RBC) transfusions as significant factors for ROP development and progression. After adjusting for these variables, platelet transfusions, high maximum fraction of inspired oxygen (FiO2) in the second week, and surfactant use remained significantly associated with ROP development, while early and late sepsis, maternal chronic hypertension, and delayed enteral nutrition were associated with progression to ROP requiring treatment.

CONCLUSIONS

These findings underscore the importance of addressing low GAs and adult RBC transfusions in ROP risk management and suggest that maximum FiO2, platelet transfusions, and sepsis also play crucial roles. Larger studies are needed to validate these results and explore preventive interventions, particularly regarding the impact of multiple adult RBC transfusions on fetal hemoglobin percentages.

Progress in the study of association between hematological indicators and retinopathy of prematurity (Review)

Retinopathy of prematurity (ROP) is a retinopathy caused by abnormal proliferation of blood vessels in premature infants. It can lead to retinal detachment and, in severe cases, blindness, rendering ROP a critical condition. Advances in neonatal medicine have improved survival rates of low birth weight and low gestational age infants. However, this progress has also led to a rise in incidence of ROP. Currently, premature birth, low birth weight and high postpartum oxygen levels are independent risk factors for ROP. Other factors include mode of delivery, multiple births, anemia, blood transfusion, maternal pregnancy factors, neonatal bronchopulmonary dysplasia, use of surfactants, arterial ductus arteriosus and necrotizing enterocolitis. Laboratory indicators in premature infants such as platelet count, levels of blood glucose, inflammatory cells, lipid and hemoglobin and blood transfusion may also be associated with ROP. However, the etiology and pathogenesis of ROP are not fully understood. A number of factors may influence the onset and progression of ROP, including decreased platelet counts, decreased hemoglobin levels, increased white blood cell counts, increased blood glucose levels, and disorders of lipid metabolism. The present study reviewed the effects of platelet count, hemoglobin, blood glucose, inflammatory cells and factors, blood lipids, and plasma metabolic pathways on ROP.

Platelet Counts and Risk of Severe Retinopathy of Prematurity: A Bayesian Model-Averaged Meta-Analysis

BACKGROUND

We aimed to conduct a systematic review and Bayesian model-averaged meta-analysis (BMA) on the association between platelet counts and severe retinopathy of prematurity (ROP).

METHODS

We searched for studies reporting on platelet counts (continuous variable) or thrombocytopenia (categorical variable) and severe ROP or aggressive posterior ROP (APROP). The timing of platelet counts was divided into Phase 1 (<2 weeks) and Phase 2 (around ROP treatment). BMA was used to calculate Bayes factors (BFs). The BF10 is the ratio of the probability of the data under the alternative hypothesis (H1) over the probability of the data under the null hypothesis (H0).

RESULTS

We included 21 studies. BMA showed an association between low platelet counts and severe ROP. The evidence was strong (BF10 = 13.5, 7 studies) for phase 1 and very strong (BF10 = 51.0, 9 studies) for phase 2. Thrombocytopenia (<100 × 109/L) in phase 2 was associated with severe ROP (BF10 = 28.2, 4 studies). Following adjustment for publication bias, only the association of severe ROP with thrombocytopenia remained with moderate evidence in favor of H1 (BF10 = 4.30).

CONCLUSIONS

Thrombocytopenia is associated with severe ROP. However, the evidence for this association was tempered when results were adjusted for publication bias.

Genetic Modulation of the Erythrocyte Phenotype Associated with Retinopathy of Prematurity-A Multicenter Portuguese Cohort Study

The development of retinopathy of prematurity (ROP) may be influenced by anemia or a low fetal/adult hemoglobin ratio. We aimed to analyze the association between DNA methyltransferase 3 β (DNMT3B) (rs2424913), methylenetetrahydrofolate reductase (MTHFR) (rs1801133), and lysine-specific histone demethylase 1A (KDM1A) (rs7548692) polymorphisms, erythrocyte parameters during the first week of life, and ROP. In total, 396 infants (gestational age < 32 weeks or birth weight < 1500 g) were evaluated clinically and hematologically. Genotyping was performed using a MicroChip DNA on a platform employing iPlex MassARRAY^®. Multivariate regression was performed after determining risk factors for ROP using univariate regression. In the group of infants who developed ROP red blood cell distribution width (RDW), erythroblasts, and mean corpuscular volume (MCV) were higher, while mean hemoglobin and mean corpuscular hemoglobin concentration (MCHC) were lower; higher RDW was associated with KDM1A (AA), MTHFR (CC and CC + TT), KDM1A (AA) + MTHFR (CC), and KDM1A (AA) + DNMT3B (allele C); KDM1A (AA) + MTHFR (CC) were associated with higher RDW, erythroblasts, MCV, and mean corpuscular hemoglobin (MCH); higher MCV and MCH were also associated with KDM1A (AA) + MTHFR (CC) + DNMT3B (allele C). We concluded that the polymorphisms studied may influence susceptibility to ROP by modulating erythropoiesis and gene expression of the fetal/adult hemoglobin ratio.