Clinical prediction models for bronchopulmonary dysplasia: a systematic review and external validation study

Personalized lung care: Bronchopulmonary dysplasia risk prediction tool tailored for neonates born in resource-limited settings

PURPOSE

Predicting bronchopulmonary dysplasia (BPD) to assess the risk-benefit of therapy is necessary considering the side effects of medications. We developed and validated an instrument for predicting BPD and compared it with an instrument currently used for neonates born in a Brazilian hospital.

METHODS

This was a retrospective cohort study of patients born between 2016 and 2020 with a gestational age (GA) between 23 and 30 weeks. Predictive equations were elaborated using methods of component variable selection collected on the 14th day of life; 70% of the sample was randomly selected for the construction of risk prediction equations and the remaining 30% for their validation, application, and comparison with the National Institute of Child Health and Human Development (NICHD) instrument. The sensitivity, specificity, and predictive values of the equations were calculated.

RESULTS

The equation that used variables with p < 5% in Fisher's exact test presented the best results: specificity of 98% and positive predictive value of 93% and could be used for BPD prediction of all small-for-gestational-age (SGA) infants. The NICHD calculator applied to our population had a specificity of 93% and a positive predictive value of 75% and could not be applied to extremely SGA infants.

CONCLUSION

Our tool can predict the risk of BPD on the 14th day of life, has higher specificity and positive predictive value to our population than the NICHD instrument, and can be suitable for SGA infants. The results must be confirmed by applying it to other populations to validate our tool.

Adverse outcomes in subsequent pregnancies in women with history of recurrent spontaneous abortion: A meta-analysis

OBJECTIVE

This study aimed to investigate the association between a history of recurrent spontaneous abortion (RSA) and adverse outcomes in women with spontaneous conception.

METHODS

A search strategy from the inception to March 3, 2023 was run in PubMed, Embase, Cochrane Library, and Web of Science databases. The odds ratio (OR), and the 95% confidence interval (CI) or point estimation were used as the evaluation indexes. Each outcome measure tested was assessed for heterogeneity using the Cochran Q test. Sensitivity analyses were performed to test the credibility of the meta-analysis results.

RESULTS

Fifteen studies involving 1 475 389 pregnant women were included. A history of RSA was associated with gestational diabetes (OR: 2.21, 95% CI: 1.70-2.87, p < 0.001), preeclampsia (OR: 2.06, 95% CI: 1.49-2.86, p < 0.001), placenta previa (OR: 1.82, 95% CI: 1.09-3.02, p = 0.021), placental abruption (OR: 1.67, 95% CI: 1.36-2.06, p < 0.001), miscarriage (OR: 6.37, 95% CI: 3.83-10.57, p < 0.001), preterm birth (OR: 1.80, 95% CI: 1.36-2.37, p < 0.001), cesarean section (OR: 1.47, 95% CI: (1.13-1.91, p = 0.004), perinatal death (OR: 2.24, 95% CI: 1.39-3.60, p = 0.001), and neonatal intensive care unit admission (OR: 1.39, 95% CI: 1.01-1.92, p = 0.047). However, the associations of a history of RSA with gestational hypertension, small for gestational age, fetal anomalies, fetal growth restriction, and postpartum hemorrhage were not observed.

CONCLUSION

This meta-analysis indicates a history of RSA was associated with increased risks of several adverse outcomes in pregnant women with spontaneous conception.

A clinical informatics approach to bronchopulmonary dysplasia: current barriers and future possibilities

Bronchopulmonary dysplasia (BPD) is a complex, multifactorial lung disease affecting preterm neonates that can result in long-term pulmonary and non-pulmonary complications. Current therapies mainly focus on symptom management after the development of BPD, indicating a need for innovative approaches to predict and identify neonates who would benefit most from targeted or earlier interventions. Clinical informatics, a subfield of biomedical informatics, is transforming healthcare by integrating computational methods with patient data to improve patient outcomes. The application of clinical informatics to develop and enhance clinical therapies for BPD presents opportunities by leveraging electronic health record data, applying machine learning algorithms, and implementing clinical decision support systems. This review highlights the current barriers and the future potential of clinical informatics in identifying clinically relevant BPD phenotypes and developing clinical decision support tools to improve the management of extremely preterm neonates developing or with established BPD. However, the full potential of clinical informatics in advancing our understanding of BPD with the goal of improving patient outcomes cannot be achieved unless we address current challenges such as data collection, storage, privacy, and inherent data bias.

Glucocorticoid signature of preterm infants developing bronchopulmonary dysplasia

BACKGROUND

Systemic inflammation plays a key role in the development of bronchopulmonary dysplasia (BPD). Cortisol is known to dampen inflammation. However, adrenal function following preterm birth is characterized by insufficient cortisol levels for the degree of inflammation, and a relative abundancy of cortisol precursors. We investigated whether this pattern could contribute to the development of BPD in preterm infants born <30 weeks of gestation.

METHODS

Cortisol, cortisone, 17-OH progesterone (17-OHP) and 11-deoxycortisol were measured in serum obtained at postnatal days 1, 3, 7, 14 and 28, using liquid-chromatography-tandem-mass-spectrometry. The presence of BPD was ascertained at 36 weeks postmenstrual age.

RESULTS

Sixty-five infants were included for analysis, of whom 32 (49%) developed BPD. Preterm infants developing BPD, as compared to those without BPD, had higher levels of 17-OHP, 11-deoxycortisol and cortisone relative to cortisol in their first week of life, but not at birth or beyond day 7.

CONCLUSION

Preterm infants developing BPD had higher levels of cortisol precursors and cortisone relative to cortisol in their first week of life than infants without BPD. These findings suggest that BPD is preceded by an activated hypothalamus-pituitary-adrenal axis that could not meet the high cortisol demands, which may predispose to inflammation and BPD.

IMPACT

Relative adrenal insufficiency is common in the first weeks after preterm birth, resulting in insufficient cortisol production for the degree of inflammation and a relative abundance of cortisol precursors; Whether this pattern contributes to the development of bronchopulmonary dysplasia (BPD) is not fully elucidated, since most studies focused on cortisol levels; Preterm infants developing BPD had higher levels of cortisol precursors and cortisone relative to cortisol in the first week of life, suggestive of a hypothalamus-pituitary-adrenal-axis activation during BPD development which cannot meet the high cortisol demands in tissues; This glucocorticoid pattern is likely to dispose to inflammation and BPD.

Risk Calculator for Bronchopulmonary Dysplasia in Preterm Neonates: A Prospective Observational Study

OBJECTIVES

To determine the incidence, risk factors, and accuracy of a previously published risk estimator in predicting Bronchopulmonary dysplasia (BPD) in a very preterm cohort from India.

METHODS

A single-center prospective observational study was conducted in preterm neonates born at 23-30 wk of gestation with a birth weight of 501-1249 g. The incidence and risk factors of BPD were evaluated, and the accuracy of BPD prediction at six pre-specified time points using the National Institute of Child Health and Human Development (NICHD) BPD risk estimator was assessed by comparing the estimated risk with the observed rates.

RESULTS

A total of 310 neonates with mean gestation age of 28.7±1.5 wk and birth weight of 1023.6±171.4 g were enrolled in the study. The study cohort had 49.7% males and 32.3% neonates requiring resuscitation. Any BPD was observed in 54 (17.4%) neonates with severity being mild, moderate, and severe in 29 (53.7%), 17 (31.5%), and 8 (14.8%) neonates respectively. Hundred (32.3%) neonates died before discharge from the hospital. Sepsis, patent ductus arteriosus, retinopathy of prematurity, intraventricular hemorrhage, and blood transfusion were significant risk factors for the development of BPD. The calculator was accurate for the prediction of death or moderate-severe BPD on days 1 and 3 with the area under the curve of 0.82 and 0.77 respectively.

CONCLUSIONS

The NICHD BPD estimator helped to accurately predict moderate-severe BPD early in Indian preterm infants.

A predictive model for preterm infants with bronchopulmonary dysplasia based on ferroptosis-related lncRNAs

BACKGROUND

Bronchopulmonary dysplasia (BPD) is the most challenging chronic lung disease for prematurity, with difficulties in early identification. Given lncRNA emerging as a novel biomarker and the regulator of ferroptosis, this study aims to develop a BPD predictive model based on ferroptosis-related lncRNAs (FRLs).

METHODS

Using a rat model, we firstly explored mRNA levels of ferroptosis-related genes and ferrous iron accumulation in BPD rat lungs. Subsequently, a microarray dataset of umbilical cord tissue from 20 preterm infants with BPD and 34 preterm infants without BPD were downloaded from the Gene Expression Omnibus databases. Random forest and LASSO regression were conducted to identify diagnostic FRLs. Nomogram was used to construct a predictive BPD model based on the FRLs. Finally, umbilical cord blood lymphocytes of preterm infants born before 32 weeks gestational age and term infants were collected and determined the expression level of diagnostic FRLs by RT-qPCR.

RESULTS

Increased iron accumulation and several dysregulated ferroptosis-associated genes were found in BPD rat lung tissues, indicating that ferroptosis was participating in the development of BPD. By exploring the microarray dataset of preterm infants with BPD, 6 FRLs, namely LINC00348, POT1-AS1, LINC01103, TTTY8, PACRG-AS1, LINC00691, were determined as diagnostic FRLs for modeling. The area under the receiver operator characteristic curve of the model was 0.932, showing good discrimination of BPD. In accordance with our analysis of microarray dataset, the mRNA levels of FRLs were significantly upregulated in umbilical cord blood lymphocytes from preterm infants who had high risk of BPD.

CONCLUSION

The incorporation of FRLs into a predictive model offers a non-invasive approach to show promise in improving early detection and management of this challenging chronic lung disease in premature infant, enabling timely intervention and personalized treatment strategies.

Prediction Models for Bronchopulmonary Dysplasia in Preterm Infants: A Systematic Review and Meta-Analysis

OBJECTIVE

To review systematically and assess the accuracy of prediction models for bronchopulmonary dysplasia (BPD) at 36 weeks of postmenstrual age.

STUDY DESIGN

Searches were conducted in MEDLINE and EMBASE. Studies published between 1990 and 2022 were included if they developed or validated a prediction model for BPD or the combined outcome death/BPD at 36 weeks in the first 14 days of life in infants born preterm. Data were extracted independently by 2 authors following the Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies (ie, CHARMS) and PRISMA guidelines. Risk of bias was assessed using the Prediction model Risk Of Bias ASsessment Tool (ie, PROBAST).

RESULTS

Sixty-five studies were reviewed, including 158 development and 108 externally validated models. Median c-statistic of 0.84 (range 0.43-1.00) was reported at model development, and 0.77 (range 0.41-0.97) at external validation. All models were rated at high risk of bias, due to limitations in the analysis part. Meta-analysis of the validated models revealed increased c-statistics after the first week of life for both the BPD and death/BPD outcome.

CONCLUSIONS

Although BPD prediction models perform satisfactorily, they were all at high risk of bias. Methodologic improvement and complete reporting are needed before they can be considered for use in clinical practice. Future research should aim to validate and update existing models.

Cardiorespiratory Monitoring Data to Predict Respiratory Outcomes in Extremely Preterm Infants

Rationale: Immature control of breathing is associated with apnea, periodic breathing, intermittent hypoxemia, and bradycardia in extremely preterm infants. However, it is not clear if such events independently predict worse respiratory outcome. Objectives: To determine if analysis of cardiorespiratory monitoring data can predict unfavorable respiratory outcomes at 40 weeks postmenstrual age (PMA) and other outcomes, such as bronchopulmonary dysplasia at 36 weeks PMA. Methods: The Prematurity-related Ventilatory Control (Pre-Vent) study was an observational multicenter prospective cohort study including infants born at <29 weeks of gestation with continuous cardiorespiratory monitoring. The primary outcome was either "favorable" (alive and previously discharged or inpatient and off respiratory medications/O2/support at 40 wk PMA) or "unfavorable" (either deceased or inpatient/previously discharged on respiratory medications/O2/support at 40 wk PMA). Measurements and Main Results: A total of 717 infants were evaluated (median birth weight, 850 g; gestation, 26.4 wk), 53.7% of whom had a favorable outcome and 46.3% of whom had an unfavorable outcome. Physiologic data predicted unfavorable outcome, with accuracy improving with advancing age (area under the curve, 0.79 at Day 7, 0.85 at Day 28 and 32 wk PMA). The physiologic variable that contributed most to prediction was intermittent hypoxemia with oxygen saturation as measured by pulse oximetry <90%. Models with clinical data alone or combining physiologic and clinical data also had good accuracy, with areas under the curve of 0.84-0.85 at Days 7 and 14 and 0.86-0.88 at Day 28 and 32 weeks PMA. Intermittent hypoxemia with oxygen saturation as measured by pulse oximetry <80% was the major physiologic predictor of severe bronchopulmonary dysplasia and death or mechanical ventilation at 40 weeks PMA. Conclusions: Physiologic data are independently associated with unfavorable respiratory outcome in extremely preterm infants.

Early prediction of pulmonary outcomes in preterm infants using electrical impedance tomography

Introduction

Electrical impedance tomography (EIT) allows assessment of ventilation and aeration homogeneity which may be associated with respiratory outcomes in preterm infants.

Methods

This was a secondary analysis to a recent randomized controlled trial in very preterm infants in the delivery room (DR). The predictive value of various EIT parameters assessed 30 min after birth on important respiratory outcomes (early intubation <24 h after birth, oxygen dependency at 28 days after birth, and moderate/severe bronchopulmonary dysplasia; BPD) was assessed.

Results

Thirty-two infants were analyzed. A lower percentage of aerated lung volume [OR (95% CI) = 0.8 (0.66-0.98), p = 0.027] as well as a higher aeration homogeneity ratio (i.e., more aeration in the non-gravity-dependent lung) predicted the need for supplemental oxygen at 28 days after birth [9.58 (5.16-17.78), p = 0.0028]. Both variables together had a similar predictive value to a model using known clinical contributors. There was no association with intubation or BPD, where numbers were small.

Discussion

In very preterm infants, EIT markers of aeration at 30 min after birth accurately predicted the need for supplemental oxygen at 28 days after birth but not BPD. EIT-guided individualized optimization of respiratory support in the DR may be possible.

A prediction nomogram for moderate-to-severe bronchopulmonary dysplasia in preterm infants < 32 weeks of gestation: A multicenter retrospective study

Background

Moderate-to-severe bronchopulmonary dysplasia (msBPD) is a serious complication in preterm infants. We aimed to develop a dynamic nomogram for early prediction of msBPD using perinatal factors in preterm infants born at <32 weeks' gestation.

Methods

This multicenter retrospective study conducted at three hospitals in China between January 2017 and December 2021 included data on preterm infants with gestational age (GA) < 32 weeks. All infants were randomly divided into training and validation cohorts (3:1 ratio). Variables were selected by Lasso regression. Multivariate logistic regression was used to build a dynamic nomogram to predict msBPD. The discrimination was verified by receiver operating characteristic curves. Hosmer-Lemeshow test and decision curve analysis (DCA) were used for evaluating calibration and clinical applicability.

Results

A total of 2,067 preterm infants. GA, Apgar 5-min score, small for gestational age (SGA), early onset sepsis, and duration of invasive ventilation were predictors for msBPD by Lasso regression. The area under the curve was 0.894 (95% CI 0.869-0.919) and 0.893 (95% CI 0.855-0.931) in training and validation cohorts. The Hosmer-Lemeshow test calculated P value of 0.059 showing a good fit of the nomogram. The DCA demonstrated significantly clinical benefit of the model in both cohorts. A dynamic nomogram predicting msBPD by perinatal days within postnatal day 7 is available at https://sdxxbxzz.shinyapps.io/BPDpredict/.

Conclusion

We assessed the perinatal predictors of msBPD in preterm infants with GA < 32 weeks and built a dynamic nomogram for early risk prediction, providing clinicians a visual tool for early identification of msBPD.

Development and Validation of a Risk Scoring Tool for Bronchopulmonary Dysplasia in Preterm Infants Based on a Systematic Review and Meta-Analysis

Background: Bronchopulmonary dysplasia (BPD) is the most common serious pulmonary morbidity in preterm infants with high disability and mortality rates. Early identification and treatment of BPD is critical. Objective: This study aimed to develop and validate a risk scoring tool for early identification of preterm infants that are at high-risk for developing BPD. Methods: The derivation cohort was derived from a systematic review and meta-analysis of risk factors for BPD. The statistically significant risk factors with their corresponding odds ratios were utilized to construct a logistic regression risk prediction model. By scoring the weights of each risk factor, a risk scoring tool was established and the risk stratification was divided. External verification was carried out by a validation cohort from China. Results: Approximately 83,034 preterm infants with gestational age < 32 weeks and/or birth weight < 1500 g were screened in this meta-analysis, and the cumulative incidence of BPD was about 30.37%. The nine predictors of this model were Chorioamnionitis, Gestational age, Birth weight, Sex, Small for gestational age, 5 min Apgar score, Delivery room intubation, and Surfactant and Respiratory distress syndrome. Based on the weight of each risk factor, we translated it into a simple clinical scoring tool with a total score ranging from 0 to 64. External validation showed that the tool had good discrimination, the area under the curve was 0.907, and that the Hosmer-Lemeshow test showed a good fit (p = 0.3572). In addition, the results of the calibration curve and decision curve analysis suggested that the tool showed significant conformity and net benefit. When the optimal cut-off value was 25.5, the sensitivity and specificity were 0.897 and 0.873, respectively. The resulting risk scoring tool classified the population of preterm infants into low-risk, low-intermediate, high-intermediate, and high-risk groups. This BPD risk scoring tool is suitable for preterm infants with gestational age < 32 weeks and/or birth weight < 1500 g. Conclusions: An effective risk prediction scoring tool based on a systematic review and meta-analysis was developed and validated. This simple tool may play an important role in establishing a screening strategy for BPD in preterm infants and potentially guide early intervention.

Exhaled Volatile Organic Compounds for Early Prediction of Bronchopulmonary Dysplasia in Infants Born Preterm

OBJECTIVE(S)

To investigate the predictive performances of exhaled breath volatile organic compounds (VOCs) for development of bronchopulmonary dysplasia (BPD) in infants born preterm.

METHODS

Exhaled breath was collected from infants born <30 weeks' gestation at days 3 and 7 of life. Ion-fragments detected by gas-Chromatography-mass-spectrometry analysis were used to derive and internally validate a VOC prediction model for moderate or severe BPD at 36 weeks postmenstrual age. We tested the predictive performance of the National Institute of Child Health and Human Development (NICHD) clinical BPD prediction model with and without VOCs.

RESULTS

Breath samples were collected from 117 infants (mean gestation 26.8 [±1.5] weeks). Thirty-three percent of the infants developed moderate or severe BPD. The VOC model showed a c-statistic of 0.89 (95% confidence interval [CI] 0.80-0.97) and 0.92 (95% CI 0.84-0.99)) for the prediction of BPD at days 3 and 7, respectively. Adding the VOCs to the clinical prediction model in non-invasive supported infants resulted in significant improvement in discriminative power on both days (day 3: c-statistic 0.83 versus 0.92, p-value 0.04; day 7: c-statistic 0.82 versus 0.94, p-value 0.03).

CONCLUSIONS

This study showed that VOC profiles in exhaled breath of preterm infants on non-invasive support in the first week of life differ between those developing and not developing BPD. Adding VOCs to a clinical prediction model significantly improved its discriminative performance.

Comparison of lung ultrasound scores with clinical models for predicting bronchopulmonary dysplasia

Lung ultrasound scores (LUSs) have been demonstrated to accurately predict moderate-to-severe bronchopulmonary dysplasia (msBPD). This study attempted to explore the additional value of LUSs for predicting msBPD compared to clinical multivariate models in different gestational age (GA) groups. The study prospectively recruited preterm infants with GA < 32 weeks. Lung ultrasound was performed on days 3, 7, 14, and 21 after birth. A linear mixed-effects regression model was used to evaluate LUS evolution in infants born before and after 28 weeks. The receiver operating characteristic (ROC) procedure was used to analyze the reliability of LUS and clinical multivariable models for predicting msBPD. The optimal time to predict msBPD in all infants was 7 days with a cut-off point of 5 (area under the ROC (AUROC) curve: 0.78, 95% confidence interval (CI): 0.71-0.84). In infants with GA ≥ 28 weeks, LUSs provided a moderate diagnostic accuracy for all four time points (AUROC curve: 0.74-0.78), and the AUROC curve for the clinical multivariable model on day 14 was 0.91 (95% CI: 0.84-0.96), which was significantly higher than that of LUSs (AUROC curve: 0.77, 95% CI: 0.68-0.85, P < 0.05). In infants born at 23-27 weeks, LUSs showed a low diagnostic accuracy with higher cut-off points to predict msBPD, and the AUROC curve for GA to predict msBPD was 0.75 (95% CI: 0.59-0.85), providing diagnostic accuracy similar to that of LUSs.  Conclusion: The contribution of LUSs to predict msBPD in infants with different GAs remains controversial and requires further investigation. What is Known: • Lung ultrasound scores (LUSs) have been demonstrated to accurately predict moderate-to-severe bronchopulmonary dysplasia in infants with gestational age (GA)＜32 weeks. What is New: • The LUSs evolution differed between extremely preterm infants born before 28 weeks and preterm infants born at 28-32 weeks of gestation. • LUSs provided similar moderate predictive performance as GA-adjusted LUS and clinical multivariate models in infants born after 28 weeks, while LUSs seem to be less helpful in infants born before 28 weeks.

Validation of disease-specific biomarkers for the early detection of bronchopulmonary dysplasia

OBJECTIVE

To demonstrate and validate the improvement of current risk stratification for bronchopulmonary dysplasia (BPD) early after birth by plasma protein markers (sialic acid-binding Ig-like lectin 14 (SIGLEC-14), basal cell adhesion molecule (BCAM), angiopoietin-like 3 protein (ANGPTL-3)) in extremely premature infants.

METHODS AND RESULTS

Proteome screening in first-week-of-life plasma samples of n = 52 preterm infants <32 weeks gestational age (GA) on two proteomic platforms (SomaLogic^®, Olink-Proteomics^®) confirmed three biomarkers with significant predictive power: BCAM, SIGLEC-14, and ANGPTL-3. We demonstrate high sensitivity (0.92) and specificity (0.86) under consideration of GA, show the proteins' critical contribution to the predictive power of known clinical risk factors, e.g., birth weight and GA, and predicted the duration of mechanical ventilation, oxygen supplementation, as well as neonatal intensive care stay. We confirmed significant predictive power for BPD cases when switching to a clinically applicable method (enzyme-linked immunosorbent assay) in an independent sample set (n = 25, p < 0.001) and demonstrated disease specificity in different cohorts of neonatal and adult lung disease.

CONCLUSION

While successfully addressing typical challenges of clinical biomarker studies, we demonstrated the potential of BCAM, SIGLEC-14, and ANGPTL-3 to inform future clinical decision making in the preterm infant at risk for BPD.

TRIAL REGISTRATION

Deutsches Register Klinische Studien (DRKS) No. 00004600; https://www.drks.de .

IMPACT

The urgent need for biomarkers that enable early decision making and personalized monitoring strategies in preterm infants with BPD is challenged by targeted marker analyses, cohort size, and disease heterogeneity. We demonstrate the potential of the plasma proteins BCAM, SIGLEC-14, and ANGPTL-3 to identify infants with BPD early after birth while improving the predictive power of clinical variables, confirming the robustness toward proteome assays and proving disease specificity. Our comprehensive analysis enables a phase-III clinical trial that allows full implementation of the biomarkers into clinical routine to enable early risk stratification in preterms with BPD.

Magnetic resonance imaging-based scoring of the diseased lung in the preterm infant with bronchopulmonary dysplasia: UNiforme Scoring of the disEAsed Lung in BPD (UNSEAL BPD)

Neonatal chronic lung disease lacks standardized assessment of lung structural changes. We addressed this clinical need by the development of a novel scoring system [UNSEAL BPD (UNiforme Scoring of the disEAsed Lung in BPD)] using T2-weighted single-shot fast-spin-echo sequences from 3 T MRI in very premature infants with and without bronchopulmonary dysplasia (BPD). Quantification of interstitial and airway remodeling, emphysematous changes, and ventilation inhomogeneity was achieved by consensus scoring on a five-point Likert scale. We successfully identified moderate and severe disease by logistic regression [area under the curve (AUC), 0.89] complemented by classification tree analysis revealing gestational age-specific structural changes. We demonstrated substantial interreader reproducibility (weighted Cohen's κ 0.69) and disease specificity (AUC = 0.91). Our novel MRI score enables the standardized assessment of disease-characteristic structural changes in the preterm lung exhibiting significant potential as a quantifiable endpoint in early intervention clinical trials and long-term disease monitoring.

Bronchopulmonary dysplasia prediction models: a systematic review and meta-analysis with validation

Prediction models could identify infants at the greatest risk of bronchopulmonary dysplasia (BPD) and allow targeted preventative strategies. We performed a systematic review and meta-analysis with external validation of identified models. Studies using predictors available before day 14 of life to predict BPD in very preterm infants were included. Two reviewers assessed 7628 studies for eligibility. Meta-analysis of externally validated models was followed by validation using 62,864 very preterm infants in England and Wales. A total of 64 studies using 53 prediction models were included totalling 274,407 infants (range 32-156,587/study). In all, 35 (55%) studies predated 2010; 39 (61%) were single-centre studies. A total of 97% of studies had a high risk of bias, especially in the analysis domain. Following meta-analysis of 22 BPD and 11 BPD/death composite externally validated models, Laughon's day one model was the most promising in predicting BPD and death (C-statistic 0.76 (95% CI 0.70-0.81) and good calibration). Six models were externally validated in our cohort with C-statistics between 0.70 and 0.90 but with poor calibration. Few BPD prediction models were developed with contemporary populations, underwent external validation, or had calibration and impact analyses. Contemporary, validated, and dynamic prediction models are needed for targeted preventative strategies. IMPACT: This review aims to provide a comprehensive assessment of all BPD prediction models developed to address the uncertainty of which model is sufficiently valid and generalisable for use in clinical practice and research. Published BPD prediction models are mostly outdated, single centre and lack external validation. Laughon's 2011 model is the most promising but more robust models, using contemporary data with external validation are needed to support better treatments.

Development of a peripheral blood transcriptomic gene signature to predict bronchopulmonary dysplasia

Bronchopulmonary dysplasia (BPD) is the most common lung disease of extreme prematurity, yet mechanisms that associate with or identify neonates with increased susceptibility for BPD are largely unknown. Combining artificial intelligence with gene expression data is a novel approach that may assist in better understanding mechanisms underpinning chronic lung disease and in stratifying patients at greater risk for BPD. The objective of this study is to develop an early peripheral blood transcriptomic signature that can predict preterm neonates at risk for developing BPD. Secondary analysis of whole blood microarray data from 97 very low birth weight neonates on day of life 5 was performed. BPD was defined as positive pressure ventilation or oxygen requirement at 28 days of age. Participants were randomly assigned to a training (70%) and testing cohort (30%). Four gene-centric machine learning models were built, and their discriminatory abilities were compared with gestational age or birth weight. This study adheres to the transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD) statement. Neonates with BPD (n = 62 subjects) exhibited a lower median gestational age (26.0 wk vs. 30.0 wk, P < 0.01) and birth weight (800 g vs. 1,280 g, P < 0.01) compared with non-BPD neonates. From an initial pool (33,252 genes/patient), 4,523 genes exhibited a false discovery rate (FDR) <1%. The area under the receiver operating characteristic curve (AUC) for predicting BPD utilizing gestational age or birth weight was 87.8% and 87.2%, respectively. The machine learning models, using a combination of five genes, revealed AUCs ranging between 85.8% and 96.1%. Pathways integral to T cell development and differentiation were associated with BPD. A derived five-gene whole blood signature can accurately predict BPD in the first week of life.

Landscape analysis for a neonatal disease progression model of bronchopulmonary dysplasia: Leveraging clinical trial experience and real-world data

The 21^st Century Cures Act requires FDA to expand its use of real-world evidence (RWE) to support approval of previously approved drugs for new disease indications and post-marketing study requirements. To address this need in neonates, the FDA and the Critical Path Institute (C-Path) established the International Neonatal Consortium (INC) to advance regulatory science and expedite neonatal drug development. FDA recently provided funding for INC to generate RWE to support regulatory decision making in neonatal drug development. One study is focused on developing a validated definition of bronchopulmonary dysplasia (BPD) in neonates. BPD is difficult to diagnose with diverse disease trajectories and few viable treatment options. Despite intense research efforts, limited understanding of the underlying disease pathobiology and disease projection continues in the context of a computable phenotype. It will be important to determine if: 1) a large, multisource aggregation of real-world data (RWD) will allow identification of validated risk factors and surrogate endpoints for BPD, and 2) the inclusion of these simulations will identify risk factors and surrogate endpoints for studies to prevent or treat BPD and its related long-term complications. The overall goal is to develop qualified, fit-for-purpose disease progression models which facilitate credible trial simulations while quantitatively capturing mechanistic relationships relevant for disease progression and the development of future treatments. The extent to which neonatal RWD can inform these models is unknown and its appropriateness cannot be guaranteed. A component of this approach is the critical evaluation of the various RWD sources for context-of use (COU)-driven models. The present manuscript defines a landscape of the data including targeted literature searches and solicitation of neonatal RWD sources from international stakeholders; analysis plans to develop a family of models of BPD in neonates, leveraging previous clinical trial experience and real-world patient data is also described.

Application and potential of artificial intelligence in neonatal medicine

Neonatal care is becoming increasingly complex with large amounts of rich, routinely recorded physiological, diagnostic and outcome data. Artificial intelligence (AI) has the potential to harness this vast quantity and range of information and become a powerful tool to support clinical decision making, personalised care, precise prognostics, and enhance patient safety. Current AI approaches in neonatal medicine include tools for disease prediction and risk stratification, neurological diagnostic support and novel image recognition technologies. Key to the integration of AI in neonatal medicine is the understanding of its limitations and a standardised critical appraisal of AI tools. Barriers and challenges to this include the quality of datasets used, performance assessment, and appropriate external validation and clinical impact studies. Improving digital literacy amongst healthcare professionals and cross-disciplinary collaborations are needed to harness the full potential of AI to help take the next significant steps in improving neonatal outcomes for high-risk infants.

Developing a machine learning-based tool to extend the usability of the NICHD BPD Outcome Estimator to the Asian population

The NICHD BPD Outcome Estimator uses clinical and demographic data to stratify respiratory outcomes of extremely preterm infants by risk. However, the Estimator does not have an option in its pull-down menu for infants of Asian descent. We hypothesize that respiratory outcomes in extreme prematurity among various racial/ethnic groups are interconnected and therefore the Estimator can still be used to predict outcomes in infants of Asian descent. Our goal was to apply a machine learning approach to assess whether outcome prediction for infants of Asian descent is possible with information hidden in the prediction results using White, Black, and Hispanic racial/ethnic groups as surrogates. We used the three racial/ethnic options in the Estimator to obtain the probabilities of BPD outcomes for each severity category. We then combined the probability results and developed three respiratory outcome prediction models at various postmenstrual age (PMA) by a random forest algorithm. We showed satisfactory model performance, with receiver operating characteristics area under the curve of 0.934, 0.850, and 0.757 for respiratory outcomes at PMA 36, 37, and 40 weeks, respectively, in the testing data set. This study suggested an interrelationship among racial/ethnic groups for respiratory outcomes among extremely preterm infants and showed the feasibility of extending the use of the Estimator to the Asian population.

Machine learning for prediction of bronchopulmonary dysplasia-free survival among very preterm infants

Background

Bronchopulmonary dysplasia (BPD) is one of the most common and serious sequelae of prematurity. Prompt diagnosis using prediction tools is crucial for early intervention and prevention of further adverse effects. This study aims to develop a BPD-free survival prediction tool based on the concept of the developmental origin of BPD with machine learning.

Methods

Datasets comprising perinatal factors and early postnatal respiratory support were used for initial model development, followed by combining the two models into a final ensemble model using logistic regression. Simulation of clinical scenarios was performed.

Results

Data from 689 infants were included in the study. We randomly selected data from 80% of infants for model development and used the remaining 20% for validation. The performance of the final model was assessed by receiver operating characteristics which showed 0.921 (95% CI: 0.899–0.943) and 0.899 (95% CI: 0.848–0.949) for the training and the validation datasets, respectively. Simulation data suggests that extubating to CPAP is superior to NIPPV in BPD-free survival. Additionally, successful extubation may be defined as no reintubation for 9 days following initial extubation.

Conclusions

Machine learning-based BPD prediction based on perinatal features and respiratory data may have clinical applicability to promote early targeted intervention in high-risk infants.

Development of a Risk Score to Predict Sudden Infant Death Syndrome

Sudden Infant Death Syndrome (SIDS) is the third leading cause of death among infants younger than one year of age. Effective SIDS prediction models have yet to be developed. Hence, we developed a risk score for SIDS, testing contemporary factors including infant exposure to passive smoke, circumcision, and sleep position along with known risk factors based on 291 SIDS and 242 healthy control infants. The data were retrieved from death certificates, parent interviews, and medical records collected between 1989−1992, prior to the Back to Sleep Campaign. Multivariable logistic regression models were performed to develop a risk score model. Our finalized risk score model included: (i) breastfeeding duration (OR = 13.85, p < 0.001); (ii) family history of SIDS (OR = 4.31, p < 0.001); (iii) low birth weight (OR = 2.74, p = 0.003); (iv) exposure to passive smoking (OR = 2.64, p < 0.001); (v) maternal anemia during pregnancy (OR = 2.07, p = 0.03); and (vi) maternal age <25 years (OR = 1.77, p = 0.01). The area under the curve for the overall model was 0.79, and the sensitivity and specificity were 79% and 63%, respectively. Once this risk score is further validated it could ultimately help physicians identify the high risk infants and counsel parents about modifiable risk factors that are most predictive of SIDS.

Lung ultrasound score has better diagnostic ability than NT-proBNP to predict moderate-severe bronchopulmonary dysplasia

The N-terminal end of B-type natriuretic peptide (NT-proBNP) and lung ultrasound (LUS) score have been proven to be adequate early biomarkers of bronchopulmonary dysplasia (BPD) in preterm infants. Our aim was to study if the predictive capacity of each one is increased by analyzing them together. We included infants born before 32 weeks with NT-proBNP and LUS scores on the first day of life (DOL) and on the 3rd, 7th, and 14th DOL and compared the diagnostic ability for moderate-severe BPD (msBPD) of each biomarker and in combination. We also compared them with a multivariate model of msBPD using only clinical variables. The sample size was 133 patients, and twenty-seven (20%) developed msBPD. The LUS score on the 7th DOL had better performance than NT-proBNP at the same moment: area under the receiver operating characteristic curve (AUC) 0.83 (0.75-0.89) versus 0.66 (0.56-0.75), p = 0.003, without differences in the rest of the times studied. These values did not increase when using the combination of both. A multivariate regression model that included only clinical variables (birth weight and invasive mechanical ventilation (IMV) at the 7th DOL) predicted msBPD with the same AUC as after the addition of any of these biomarkers, neither together.

CONCLUSION

The LUS score is a better predictor of msBPD on the 7th DOL than NT-proBNP in preterm infants born before 32 weeks, although they have similar diagnostic accuracy on the 1st, 3rd, and 14th DOL. Neither of them, nor together, have a better AUC for msBPD than a clinical model with birthweight and the need for IMV at the 7th DOL.

WHAT IS KNOWN

• NT-proBNP and LUS score are early predictors of moderate-severe bronchopulmonary dysplasia (msBPD).

WHAT IS NEW

• The combination of both NT-proBNP and LUS score does not increase the predictive ability of each separately.

Postnatal Corticosteroids to Prevent or Treat Chronic Lung Disease Following Preterm Birth

The American Academy of Pediatrics continues to provide guidance on the use of postnatal corticosteroids to manage or prevent chronic lung disease following preterm birth (formerly referred to as bronchopulmonary dysplasia). Since the last revision of such guidance in 2010, several prospective randomized trials have been published. This revision provides a review of those studies as well as updated recommendations, which include the use of systemic low-dose corticosteroid in preterm neonates with or at high risk for chronic lung disease. High-dose dexamethasone (≥0.5 mg/kg per day) is not recommended. New evidence suggests that inhaled corticosteroids may confer benefit if provided with surfactant as a vehicle, but safety data are lacking. Evidence remains insufficient to make any recommendations regarding routine use of postnatal corticosteroids in preterm infants. Neonatologists and other hospital care providers must continue to use their clinical judgment in individual patients, balancing the potential adverse effects of corticosteroid treatment with those of chronic lung disease. The decision to use postnatal corticosteroids for this purpose should be made together with the infant's parents, and the care providers should document their discussions with parents in the patient's medical record.

Systemic Steroids in Preventing Bronchopulmonary Dysplasia (BPD): Neurodevelopmental Outcome According to the Risk of BPD in the EPICE Cohort

BACKGROUND

Postnatal steroids (PNS) have been used to prevent bronchopulmonary dysplasia (BPD) in preterm infants but have potential adverse effects on neurodevelopment. These effects might be modulated by their risk of BPD. We aimed to compare patients' neurodevelopment with PNS treatment according to their risk of BPD in a European cohort.

METHODS

We developed a prediction model for BPD to classify infants born between 24 + 0 and 29 + 6 weeks of gestation in three groups and compared patients' neurological outcome at two years of corrected age using the propensity score (PS) method.

RESULTS

Of 3662 neonates included in the analysis, 901 (24.6%) were diagnosed with BPD. Our prediction model for BPD had an area under the ROC curve of 0.82. In the group with the highest risk of developing BPD, PNS were associated with an increased risk of gross motor impairment: OR of 1.95 after IPTW adjustment (95% CI 1.18 to 3.24, p = 0.010). This difference existed regardless of the type of steroid used. However, there was an increased risk of cognitive anomalies for patients treated with dexa/betamethasone that was no longer observed with hydrocortisone.

CONCLUSIONS

This study suggests that PNS might be associated with an increased risk of gross motor impairment regardless of the group risk for BPD. Further randomised controlled trials exploring the use of PNS to prevent BPD should include a risk-based evaluation of neurodevelopmental outcomes. This observation still needs to be confirmed in a randomised controlled trial.

Lifelong Lung Sequelae of Prematurity

The clinical, functional, and structural pattern of chronic lung disease of prematurity has changed enormously in last years, mirroring a better perinatal management and an increasing lung immaturity with the survival of increasingly premature infants. Respiratory symptoms and lung function impairment related to prematurity seem to improve over time, but premature birth increases the likelihood of lung function impairment in late childhood, predisposing to chronic obstructive pulmonary disease (COPD). It is mandatory to identify those individuals born premature who are at risk for developing long-term lung disease through a better awareness of physicians, the use of standardized CT imaging scores, and a more comprehensive periodic lung function evaluation. The aim of this narrative review was to provide a systematic approach to lifelong respiratory symptoms, lung function impairment, and lung structural anomalies in order to better understand the specific role of prematurity on lung health.

Prediction Models for Bronchopulmonary Dysplasia in Preterm Infants: A Systematic Review

OBJECTIVE

To provide an overview and critical appraisal of prediction models for bronchopulmonary dysplasia (BPD) in preterm infants.

METHODS

We searched PubMed, Embase, and the Cochrane Library to identify relevant studies (up to November 2021). We included studies that reported prediction model development and/or validation of BPD in preterm infants born at ≤32 weeks and/or ≤1,500 g birth weight. We extracted the data independently based on the CHecklist for critical Appraisal and data extraction for systematic Reviews of prediction Modelling Studies (CHARMS). We assessed risk of bias and applicability independently using the Prediction model Risk Of Bias ASsessment Tool (PROBAST).

RESULTS

Twenty-one prediction models from 13 studies reporting on model development and 21 models from 10 studies reporting on external validation were included. Oxygen dependency at 36 weeks' postmenstrual age was the most frequently reported outcome in both development studies (71%) and validation studies (81%). The most frequently used predictors in the models were birth weight (67%), gestational age (62%), and sex (52%). Nearly all included studies had high risk of bias, most often due to inadequate analysis. Small sample sizes and insufficient event patients were common in both study types. Missing data were often not reported or were discarded. Most studies reported on the models' discrimination, while calibration was seldom assessed (development, 19%; validation, 10%). Internal validation was lacking in 69% of development studies.

CONCLUSION

The included studies had many methodological shortcomings. Future work should focus on following the recommended approaches for developing and validating BPD prediction models.

Respiratory morbidity in preterm infants predicted by natriuretic peptide (MR-proANP) and endothelin-1 (CT-proET-1)

BACKGROUND

Bronchopulmonary dysplasia (BPD) is a major complication in preterm infants <32 weeks. We aimed to assess whether plasma levels of mid-regional pro-atrial natriuretic peptide (MR-proANP) and C-terminal pro-endothelin-1 (CT-proET-1) predict respiratory morbidity.

METHODS

This was a prospective, two-center, observational cohort study. MR-proANP and CT-proET-1 were measured at day 7 (±2) of life. Associations with duration of supplemental oxygen and the composite outcome of moderate or severe BPD or death (BPD/death) were investigated.

RESULTS

Two hundred and twenty-nine infants <32 weeks were included (median gestational age [GA] 29.6 weeks [interquartile range 29.0-30.7], median birth weight 1150 g [IQR 840-1410]). MR-proANP and CT-proET-1 were associated with the duration of supplemental oxygen in univariable analysis (both p < 0.001) but not after adjusting for co-factors. Infants with BPD/death showed higher plasma levels of MR-proANP (623.50 pmol/L [IQR 458.50-881.38] vs. 308.35 pmol/L [IQR 216.72-538.10]; p < 0.001) and CT-proET-1 (255.40 pmol/L [IQR 202.60-311.15] vs. 198.30 pmol/L [IQR 154.70-297.95]; p = 0.015) compared to infants without BPD/death. Levels of both biomarkers were significantly associated with BPD/death in univariable models but not after adjusting for co-factors.

CONCLUSIONS

MR-proANP and CT-proET-1 are associated with the duration of supplemental oxygen and the composite outcome BPD/death, but their prognostic value does not complement that of clinical risk factors.

IMPACT

Plasma levels of MR-proANP and CT-proET-1, measured on day 7 of life (±2 days) are associated in univariable analyses with duration of supplemental oxygen and the combined outcome of BPD or death in VLGA infants. Associations between both biomarkers and respiratory morbidity do not persist in multivariable models, in particular when gestational age is included. MR-proANP and CT-proET-1 have limited additional value to predict respiratory morbidity in VLGA infants compared to clinical parameters.

Effect of human milk and other neonatal variables on lung function at three months corrected age

OBJECTIVE

To evaluate the impact of human milk and different neonatal variables on tidal breathing flow-volume loop (TBFVL) parameters within three months' corrected age (CA) in infants born ≤32 wks or weighing <1500 g.

METHODS

We retrospectively studied 121 infants with gestational age (GA) ≤ 32 weeks or birth weight (BW) <1500 gr who had lung function assessment within three months' CA by TBFVL analysis between June 2009 and April 2018. We investigated the impact of GA, gender, being Small for GA (SGA), sepsis, days of mechanical ventilation (MV) and human milk feeding (HMF) on later respiratory function, both in the entire group and according to BW ( ≤1000 g and >1000 g).

RESULTS

The mean(SD) z-score for tidal volume (Vt) and time to peak expiratory flow to expiratory time (tPTEF/tE) were respectively -4.3 (2.5) and -0.8 (2.0) for the overall population with no significant differences between infants <1000 g or ≥1000 g. The mean(SD) Vt standardized for body weight was 6.2(2.0) ml/kg. Being female was associated with better Vt/Kg, whereas longer MV or being born SGA were associated with worst tPTEF/tE. For infants with BW < 1000 gr, tPTEF/tE was positively associated with HMF.

CONCLUSION

An early TBFVL assessment within three months' CA already reveals lung function alteration in preterm infants. Being female is associated with better Vt/Kg, while longer duration of MV or being born SGA negatively affect tPTEF/tE. The positive association between HMF and better tPTEF/tE in infants with BW <1000 g has emerged, which deserves further investigation.

The challenge of accurately describing the epidemiology of bronchopulmonary dysplasia (BPD) based on the various current definitions of BPD

Despite marked improvements in the survival of extremely low birth weight preterm infants, bronchopulmonary dysplasia (BPD) remains a prevalent morbidity. BPD has evolved pathologically and epidemiologically but the definition has failed to keep up. The majority of the definitions of BPD still use the respiratory support provided to the infants at a single timepoint. The lack of a uniform definition of BPD presently reflects the changing BPD pathogenesis and phenotype and limits defining the epidemiology. To address the epidemiology of BPD, the definition should be clarified; even the newer definitions have not been validated entirely. The definition needs to be meaningful clinically and be predictive of long-term respiratory outcomes. We believe the definition should have a composite assessment like a score (quantitative measurement) and include the different phenotypes (qualitative measurements) so that optimally they can be applied to the different phases of BPD and at different timepoints. Furthermore, the definitions need to be easy to measure and assess so that generalizability is enhanced.

Prediction Model for Bronchopulmonary Dysplasia in Preterm Newborns

OBJECTIVE: To develop a multifactorial model that allows the prediction of bronchopulmonary dysplasia (BPD) in preterm newborns. MATERIALS AND METHODS: A single-center retrospective study of infants born below 32 + 0 weeks gestational age. We created a receiver operating characteristic curve to assess the multifactorial BPD risk and calculate the BPD risk accuracy using the area under the curve (AUC). BPD risk was categorized using a multifactorial predictive model based on the weight of the evidence. RESULTS: Of the 278 analyzed preterm newborns, 127 (46%) developed BPD. The significant risk factors for BPD in the multivariate analysis were gestational age, number of red blood cell concentrate transfusions, number of surfactant administrations, and hemodynamically significant patent ductus arteriosus. The combination of these factors determined the risk of developing BPD, with an AUC value of 0.932. A multifactorial predictive model based on these factors, weighted by their odds ratios, identified four categories of newborns with mean BPD risks of 9%, 59%, 82%, and 100%. CONCLUSION: A multifactorial model based on easily available clinical factors can predict BPD risk in preterm newborns and inform potential preventive measures.

The Intertemporal Role of Respiratory Support in Improving Neonatal Outcomes: A Narrative Review

Defining improvements in healthcare can be challenging due to the need to assess multiple outcomes and measures. In neonates, although progress in respiratory support has been a key factor in improving survival, the same degree of improvement has not been documented in certain outcomes, such as bronchopulmonary dysplasia. By exploring the evolution of neonatal respiratory care over the last 60 years, this review highlights not only the scientific advances that occurred with the application of invasive mechanical ventilation but also the weakness of the existing knowledge. The contributing role of non-invasive ventilation and less-invasive surfactant administration methods as well as of certain pharmacological therapies is also discussed. Moreover, we analyze the cost-benefit of neonatal care-respiratory support and present future challenges and perspectives.

Should Vitamin A Injections to Prevent Bronchopulmonary Dysplasia or Death Be Reserved for High-Risk Infants? Reanalysis of the National Institute of Child Health and Human Development Neonatal Research Network Randomized Trial

OBJECTIVE

To determine whether infants at higher risk of bronchopulmonary dysplasia (BPD) or death benefit more from vitamin A therapy than those at lower risk.

STUDY DESIGN

We conducted a post hoc reanalysis of a landmark phase III randomized controlled trial conducted from January 1996 to July 1997 at 14 university-affiliated neonatal intensive care units in the US. Data analysis was performed from October 2019 to October 2020. Infants born weighing 401-1000 g and receiving respiratory support at 24 hours of age were assigned to intramuscular vitamin A 5000 IU or sham procedure 3 times weekly for 4 weeks. The primary outcome was BPD, defined as use of supplemental oxygen, or death at 36 weeks postmenstrual age. An externally validated model for predicting BPD or death was used to estimate the risk of these outcomes for each infant.

RESULTS

As previously reported, 222 of 405 infants (54.8%) assigned vitamin A therapy and 248 of 402 infants (61.7%) in the control group developed BPD or died (relative risk [RR], 0.89 [95% CI, 0.80-0.99]; risk difference [RD], -6.9% [95% CI, -13.0 to -0.7]). The predicted individual risks of BPD or death ranged from 7.1% to 98.6% (median, 61.5%; mean, 60.9%). The effect of vitamin A therapy on BPD or death depended on infants' risk of the primary outcome (P = .03 for interaction): for example, a RR of 0.73 (RD, -14.5%) for infants with a 25% predicted risk and a RR of 0.96 (RD, -1.0%) for infants with a 75% risk. There was no difference in the decrease in vitamin A deficiency across risk groups.

CONCLUSIONS

Contrary to expectations, the effect of vitamin A therapy on BPD or death was greater for lower risk than higher risk infants.

TRIAL REGISTRATION

ClinicalTrials.gov NCT01203488.

Bronchopulmonary Dysplasia Predicted by Developing a Machine Learning Model of Genetic and Clinical Information

Background

An early and accurate evaluation of the risk of bronchopulmonary dysplasia (BPD) in premature infants is pivotal in implementing preventive strategies. The risk prediction models nowadays for BPD risk that included only clinical factors but without genetic factors are either too complex without practicability or provide poor-to-moderate discrimination. We aim to identify the role of genetic factors in BPD risk prediction early and accurately.

Methods

Exome sequencing was performed in a cohort of 245 premature infants (gestational age <32 weeks), with 131 BPD infants and 114 infants without BPD as controls. A gene burden test was performed to find risk genes with loss-of-function mutations or missense mutations over-represented in BPD and severe BPD (sBPD) patients, with risk gene sets (RGS) defined as BPD-RGS and sBPD-RGS, respectively. We then developed two predictive models for the risk of BPD and sBPD by integrating patient clinical and genetic features. The performance of the models was evaluated using the area under the receiver operating characteristic curve (AUROC).

Results

Thirty and 21 genes were included in BPD-RGS and sBPD-RGS, respectively. The predictive model for BPD, which combined the BPD-RGS and basic clinical risk factors, showed better discrimination than the model that was only based on basic clinical features (AUROC, 0.915 vs. AUROC, 0.814, P = 0.013, respectively) in the independent testing dataset. The same was observed in the predictive model for sBPD (AUROC, 0.907 vs. AUROC, 0.826; P = 0.016).

Conclusion

This study suggests that genetic information contributes to susceptibility to BPD. The predictive model in this study, which combined BPD-RGS with basic clinical risk factors, can thus accurately stratify BPD risk in premature infants.

Recent Advances in Bronchopulmonary Dysplasia

Bronchopulmonary dysplasia (BPD) is a form of chronic lung disease that occurs in preterm infants, usually those receiving substantial respiratory support with either mechanical ventilation or supplementation with oxygen. The pathogenesis of BPD is multifactorial, and the clinical phenotype is variable. BPD is associated with substantial mortality and short- and long-term morbidity. The incidence of BPD has remained stable or increased, as advances in neonatal care have led to improved survival of more extremely preterm infants. Extensive basic science, translational, and clinical research focusing on BPD has improved the current understanding of the factors that contribute to BPD pathogenesis. However, despite a better understanding of its pathophysiology, BPD continues to be challenging to prevent and manage adequately. The current review aims to provide a clinically useful synopsis of evidence on the prevention and management of BPD in preterm infants.

[Epidemiology and risk factors for bronchopulmonary dysplasia in prematures infants born at or less than 32 weeks of gestation]

OBJECTIVES

To describe risk factors of bronchopulmonary dysplasia in very preterm infants in the first weeks of life.

MATERIAL AND METHODS

Retrospective cohort study of preterm infants ≤ 32 weeks of gestational age and birth weight ≤ 1500 g. A multivariate logistic regression analysis was performed to identify independent risk factors for bronchopulmonary dysplasia in the first weeks of life.

RESULTS

A total of 202 newborns were included in the study (mean gestational age 29.5 ± 2.1 weeks), 61.4% never received invasive mechanical ventilation. The incidence of bronchopulmonary dysplasia was 28.7%, and 10.4% of the patients were diagnosed with moderate-severe bronchopulmonary dysplasia. Bronchopulmonary dysplasia was independently associated with gestational age at birth (p < 0.001; OR = 0.44 [95% CI = 0.30-0.65]), the need for mechanical ventilation on the first day of life (p = 0.001; OR = 8.13 [95% CI = 2.41-27.42]), nosocomial sepsis (p < 0.001; OR = 9.51 [95% CI = 2.99-30.28]) and FiO2 on day 14 (p < 0.001; OR = 1.39 [95% CI = 1.16-1.66]). Receiving mechanical ventilation at the first day of life (p = 0.008; OR = 5.39 [95% CI = 1.54-18.89]) and at the third day of life (p = 0.001; OR = 9.99 [95% CI = 2.47-40.44]) and nosocomial sepsis (p = 0.001; OR = 9.87 [95% CI = 2.58-37.80]) were independent risk factors for moderate-severe bronchopulmonary dysplasia.

CONCLUSIONS

Gestational age at birth, mechanical ventilation in the first days of life and nosocomial sepsis are early risk factors for bronchopulmonary dysplasia. The analysis of simple and objective clinical data, allows us to select a group of patients at high risk of bronchopulmonary dysplasia in whom it could be justified to act more aggressively, and shows areas for improvement to prevent its development or reduce its severity.

The Prediction of Bronchopulmonary Dysplasia in Very Low Birth Weight Infants through Clinical Indicators within 1 Hour of Delivery

BACKGROUND

Despite the advances in neonatology, the incidence of bronchopulmonary dysplasia (BPD) is increasing. It is important to prevent the development of BPD in the first place. The online BPD outcome estimator from National Institute of Children Health and Human Development and Neonatal Research Network is available. However, it is not applicable for Asians. Moreover, limits are set for birth weight and gestational weeks excluding those who may still have BPD. The aim of this study was to develop a prediction model for BPD using first hour perinatal and neonatal factors in Korean very low birth weight infants (VLBWIs).

METHODS

Data were collected for 8,022 VLBWIs with gestational age (GA) ≥ 22 weeks who were born between January 1, 2013 and December 31, 2016, and admitted to the neonatal intensive care units of the KNN. Multiple logistic regression models reanalyzed by stepwise selection with significant clinical indicators for BPD. PROC package was used to calculate the area under curve (AUC) and corresponding 95% confidence intervals. Moreover, it was used to search the best cut-off value. External validation was performed with the 2017 Korean neonatal network (KNN) data.

RESULTS

After all missing data were excluded, 4,600 VLBWIs were included in the training dataset of the prediction model. Predictability of presence of BPD was 90.8% and prediction P value cut off was 0.550. Five-minute Apgar score, birth weight, GA, sex, surfactant use were significant indicators. Predictability of severe BPD was 81.5% and prediction P value cut off was 0.160. Five-minute Apgar score, birth weight, maternal PIH, chronic maternal hypertension, GA, sex, respiratory distress syndrome, need of resuscitation at birth were significant indicators. After external validation, sensitivity and specificity did not change significantly.

CONCLUSION

From this study, high predictability was obtained using clinical parameters obtained within one hour of life. P value for prediction of each grade of BPD and equation for calculation was presented. It can be helpful for the early prediction of BPD in Korean VLBWI. This study will contribute to the prediction of BPD in Asians especially Korean VLBWIs, not currently included in the NICHD BPD online BPD predictor. In addition, the predictive power may be continuously increased with the cumulative data of KNN.

The Predictive Value of Lung Ultrasound Scores in Developing Bronchopulmonary Dysplasia: A Prospective Multicenter Diagnostic Accuracy Study

BACKGROUND

Different lung ultrasound (LUS) scanning protocols have been used, and the results in terms of diagnostic accuracy are heterogeneous.

RESEARCH QUESTIONS

What is the diagnostic accuracy of the LUS score to predict moderate to severe bronchopulmonary dysplasia (msBPD)? Does scanning of posterior lung fields improve the diagnostic accuracy?

STUDY DESIGN AND METHODS

This was a multicenter prospective, observational study in six centers. Two LUS aeration scores, one involving only anterolateral lung fields and the other adding the posterior fields were obtained at birth, on the third day of life (DOL), on the seventh DOL, on the 14th DOL, and on the 21st DOL. The diagnostic accuracy of both scores to predict msBPD was assessed at each time point.

RESULTS

Eight hundred thirty-two LUS examinations in 298 infants were included. Both LUS score using anterolateral and posterior fields and LUS score using only anterolateral fields showed a similar moderate diagnostic accuracy to predict msBPD on the third DOL (area under the receiver operating characteristic curve [AUC] 95% CI, 0.68-0.85 vs 0.68-0.85; P = .97), seventh DOL (AUC 95% CI, 0.74-0.85 vs 0.74-0.84; P = .26), and 21st DOL (AUC 95% CI, 0.72-0.86 vs 0.74-0.88; P = .17). The LUS score using anterolateral and posterior fields was slightly more accurate at 14th DOL (AUC 95% CI, 0.69-0.83 vs 0.66-0.80; P = .01). A cutoff of 8 points in the LUS score using only anterolateral fields on the seventh DOL provided a sensitivity, specificity, positive likelihood ratio, and negative likelihood ratio of 70%, 79%, 3.3, and 0.38, respectively, to predict msBPD. Adding gestational age (GA) and sex improved the discriminative value without significant differences compared with a predictive model based on multiple clinical variables: AUC 95% CI, 0.77-0.88 vs 0.80-0.91 (P = .52).

INTERPRETATION

The LUS score is able to predict msBPD from the third DOL with a moderate diagnostic accuracy. Scanning posterior lung fields slightly improved diagnostic accuracy only at the 14th DOL. Adding GA and sex improves the diagnostic accuracy of the LUS scores. The LUS score is useful to stratify BPD risk early after birth.

Isoprostanes as Biomarker for White Matter Injury in Extremely Preterm Infants

Background and Aim: Preterm white matter is vulnerable to lipid peroxidation-mediated injury. F2-isoprostanes (IPs), are a useful biomarker for lipid peroxidation. Aim was to assess the association between early peri-postnatal IPs, white matter injury (WMI) at term equivalent age (TEA), and neurodevelopmental outcome in preterm infants. Methods: Infants with a gestational age (GA) below 28 weeks who had an MRI at TEA were included. IPs were measured in cord blood (cb) at birth and on plasma (pl) between 24 and 48 h after birth. WMI was assessed using Woodward MRI scoring system. Multiple regression analyses were performed to assess the association between IPs with WMI and then with BSITD-III scores at 24 months corrected age (CA). Receiver operating characteristic (ROC) curve analysis was used to evaluate the predictive value of pl-IPs for the development of WMI. Results: Forty-four patients were included. cb-IPs were not correlated with WMI score at TEA, whereas higher pl-IPs and lower GA predicted higher WMI score (p = 0.037 and 0.006, respectively) after controlling for GA, FiO2 at sampling and severity of IVH. The area under the curve was 0.72 (CI 95% = 0.51-0.92). The pl-IPs levels plotted curve indicated that 31.8 pg/ml had the best predictive threshold with a sensitivity of 86% and a specificity of 60%, to discriminate newborns with any WMI from newborns without WMI. IPs were not associated with outcome at 24 months. Conclusion: Early measurement of pl-IPs may help discriminate patients showing abnormal WMI score at TEA, thus representing an early biomarker to identify newborns at risk for brain injury.

Long-Term Outcomes of Bronchopulmonary Dysplasia Under Two Different Diagnostic Criteria: A Retrospective Cohort Study at a Chinese Tertiary Center

Unlike other complications among very low birth weight infants (VLBW), the incidence of bronchopulmonary dysplasia (BPD) has not decreased substantially, partly because of the different definitions of BPD applied by different researchers. In this retrospective cohort study, we aimed to compare the 2018 revised definition and the 2001 consensus definition of BPD proposed by the National Institute of Child Health and Human Development (NICHD), as well as to identify which definition better predicts severe respiratory morbidities or death. We included 417 infants born at a gestational age <32 weeks and classified them as having BPD or without BPD based on the two definitions, with a final follow-up at 18-24 months. We performed between-group comparisons of death and respiratory outcomes. Statistical analyses were performed using descriptive statistics, comparative tests, and receiver operating characteristic curves. The mean ± standard deviation gestational age and birth weight of the 417 eligible infants were 29.1 ± 1.4 weeks and 1186.6 ± 197.8 g, respectively. Among the included infants, five and three infants died before and after 36 weeks of post-menstrual age (PMA), respectively, with 68 and 344 infants evaluated at discharge and 36 weeks' PMA, respectively. We diagnosed 163 (39.1%) and 70 (16.8%) infants with BPD according to the 2001 and 2018 NICHD definitions, respectively. The 2001 NICHD definition displayed a higher sensitivity (0.60 vs. 0.28), better negative predictive value (0.89 vs. 0.85), and larger area under the receiver operating characteristic curve (0.66 vs. 0.57), but a lower specificity (0.65 vs. 0.87) and worse positive predictive value (0.26 vs. 0.31), than the 2018 definition for serious respiratory morbidity or mortality at a corrected age of 18-24 months. Compared with the 2018 NICHD definition of BPD, the 2001 NICHD consensus definition may result in more cases of false-positive or unclassified severity. However, it may be a better indicator of severe respiratory morbidities or death during the first 18-24 months. Nevertheless, there is a need for future studies to assess the validity of the new diagnostic criteria.

Development and Validation of a Nomogram for Predicting Bronchopulmonary Dysplasia in Very-Low-Birth-Weight Infants

Background: Bronchopulmonary dysplasia is a common pulmonary disease in newborns and is one of the main causes of death. The aim of this study was to build a new simple-to-use nomogram to screen high-risk populations. Methods: In this single-center retrospective study performed from January 2017 to December 2020, we reviewed data on very-low-birth-weight infants whose gestational ages were below 32 weeks. LASSO regression was used to select variables for the risk model. Then, we used multivariable logistic regression to build the prediction model incorporating these selected features. Discrimination was assessed by the C-index, and and calibration of the model was assessed by and calibration curve and the Hosmer-Lemeshow test. Results: The LASSO regression identified gestational age, duration of ventilation and serum NT-proBNP in the 1st week as significant predictors of BPD. The nomogram-illustrated model showed good discrimination and calibration. The C-index was 0.853 (95% CI: 0.851-0.854) in the training set and 0.855 (95% CI: 0.77-0.94) in the validation set. The calibration curve and Hosmer-Lemeshow test results showed good calibration between the predictions of the nomogram and the actual observations. Conclusion: We demonstrated a simple-to-use nomogram for predicting BPD in the early stage. It may help clinicians recognize high-risk populations.

The association between clinical and biochemical characteristics of late-onset sepsis and bronchopulmonary dysplasia in preterm infants

Studies in preterm infants have shown an association between late-onset sepsis (LOS) and the development of bronchopulmonary dysplasia (BPD). It is unknown whether clinical or biochemical characteristics during sepsis modulate the risk for BPD. This single-center retrospective cohort study included all patients with a gestational age < 30 weeks, born between 2009 and 2015, in whom empiric antimicrobial treatment was initiated > 72 h after birth and continued for at least 5 days, independent on microbiological results. The association between clinical and biochemical characteristics of LOS and the development of BPD in survivors were assessed with multivariate logistic regression analysis adjusted for early-onset sepsis, small for gestational age, and gestational age. Of the 756 admitted infants, 256 infants (mean GA: 27.0 weeks; birthweight: 924 grams) had at least one LOS episode, of whom 79 (30.9%) developed BPD. Analyses showed that only the need for and duration of mechanical ventilation during LOS were independently associated with an increased risk for BPD (adjusted OR 2.62, 95% CI 1.38, 4.96, p value 0.003, and OR 1.004, 95% CI 1.00, 1.007, p value 0.045, respectively).Conclusion: During a LOS, the need for and duration of mechanical ventilation are independently associated with the risk of developing BPD in preterm infants. What is Known: • Premature infants diagnosed with a late-onset sepsis are at higher risk of developing bronchopulmonary dysplasia • This association is mainly shown in infants with a positive blood culture What is New: • This study investigates the clinical and biochemical characteristics of late-onset sepsis and the development of bronchopulmonary dysplasia • The need for mechanical ventilation and duration of mechanical ventilation during late-onset sepsis are associated with an increased risk of developing bronchopulmonary dysplasia.

Comparison of Multivariable Logistic Regression and Machine Learning Models for Predicting Bronchopulmonary Dysplasia or Death in Very Preterm Infants

Bronchopulmonary dysplasia (BPD) is the most prevalent and clinically significant complication of prematurity. Accurate identification of at-risk infants would enable ongoing intervention to improve outcomes. Although postnatal exposures are known to affect an infant's likelihood of developing BPD, most existing BPD prediction models do not allow risk to be evaluated at different time points, and/or are not suitable for use in ethno-diverse populations. A comprehensive approach to developing clinical prediction models avoids assumptions as to which method will yield the optimal results by testing multiple algorithms/models. We compared the performance of machine learning and logistic regression models in predicting BPD/death. Our main cohort included infants <33 weeks' gestational age (GA) admitted to a Canadian Neonatal Network site from 2016 to 2018 (n = 9,006) with all analyses repeated for the <29 weeks' GA subcohort (n = 4,246). Models were developed to predict, on days 1, 7, and 14 of admission to neonatal intensive care, the composite outcome of BPD/death prior to discharge. Ten-fold cross-validation and a 20% hold-out sample were used to measure area under the curve (AUC). Calibration intercepts and slopes were estimated by regressing the outcome on the log-odds of the predicted probabilities. The model AUCs ranged from 0.811 to 0.886. Model discrimination was lower in the <29 weeks' GA subcohort (AUCs 0.699-0.790). Several machine learning models had a suboptimal calibration intercept and/or slope (k-nearest neighbor, random forest, artificial neural network, stacking neural network ensemble). The top-performing algorithms will be used to develop multinomial models and an online risk estimator for predicting BPD severity and death that does not require information on ethnicity.

Bronchopulmonary dysplasia: a predictive scoring system for very low birth weight infants. A diagnostic accuracy study with prospective data collection

Our aim was to develop and validate a predictive risk score for bronchopulmonary dysplasia (BPD), according to two clinically used definitions: 1. Need for supplementary oxygen during ≥ 28 cumulative days, BPD28, 2. Need for supplementary oxygen at 36 weeks postmenstrual age (PMA), BPD36. Logistic regression was performed in a national cohort (infants born in Switzerland with a birth weight < 1501 g and/or between 23 0/7 and 31 6/7 weeks PMA in 2009 and 2010), to identify predictors of BPD. We built the score as the sum of predicting factors, weighted according to their ORs, and analysed its discriminative properties by calculating the area under the ROC (receiver operating characteristic) curves (AUCs). This score was then applied to the Swiss national cohort from the years 2014-2015 to perform external validation. The incidence of BPD28 was 21.6% in the derivation cohort (n = 1488) and 25.2% in the validation cohort (n = 2006). The corresponding numbers for BPD36 were 11.3% and 11.1%, respectively. We identified gestational age, birth weight, antenatal corticosteroids, surfactant administration, proven infection, patent ductus arteriosus and duration of mechanical ventilation as independent predictors of BPD28. The AUCs of the BPD risk scores in the derivation cohort were 0.90 and 0.89 for the BPD28 and BPD36 definitions, respectively. The corresponding AUCs in the validation cohort were 0.92 and 0.88, respectively.Conclusion: This score allows for predicting the risk of a very low birth weight infant to develop BPD early in life and may be a useful tool in clinical practice and neonatal research. What is Known: • Many studies have proposed scoring systems to predict bronchopulmonary dysplasia (BPD). • Such a risk prediction may be important to identify high-risk patients for counselling parents, research purposes and to identify candidates for specific treatment. What is New: • A predictive risk score for BPD was developed and validated in a large national multicentre cohort and its performance assessed by two indices of accuracy. • The developed scoring system allows to predict the risk of BPD development early but also at any day of life with high validity.