Neonatology最新文献

Introduction: Quantitative lung ultrasound (LUS) predicts bronchopulmonary dysplasia (BPD), but variability in BPD definitions raises concerns about its predictive consistency. We hypothesized that predictive accuracy of LUS would remain stable regardless of the definition applied.

Methods: In this prospective, multicenter cohort study, preterm infants ≤30 weeks of gestation underwent extended LUS (eLUS, adj-eLUS) aeration score at days 10, 21, and 28. BPD was assessed at 36 weeks of postmenstrual age using Jobe and Bancalari (2001), NICHD (2018), and Jensen (2019) definitions. Receiver operating characteristic (ROC) analysis compared predictive performance (areas under ROC curve [AUC]) across definitions.

Results: Among 337 infants (mean gestational age: 27 weeks, mean birth weight: 941 g), BPD incidence ranged from 22.8 to 25.8% depending on definition. AUCs for BPD prediction ranged between 0.732 and 0.832. The mean difference (ΔAUC) between definitions was minimal (≈0.02, 95% confidence interval: 0.01-0.03) and nonsignificant at all time points.

Conclusions: Quantitative LUS reliably predicts BPD regardless of its definition, and this support its use in early respiratory care and monitoring.

Introduction: Postnatal weight loss in infants is physiological, but excessive loss predisposes for dehydration and other morbidities. Existing nomograms, primarily developed in temperate climates, may not apply to arid regions as environmental conditions influence weight loss patterns. This study aimed to develop hour-specific percentile nomograms for postnatal weight loss in exclusively breastfed, healthy infants (≥36 weeks) from an arid region, facilitating early identification of those at risk of excessive loss.

Methods: A prospective cohort study was conducted between November 2021 and February 2023 at a tertiary center in Western India. Exclusively breastfed infants ≥36 weeks without major morbidities were enrolled and weighed twice daily until 100 h after birth or discharge. Infants with abnormal clinical/biochemical findings or requiring any milk supplementation were censored. Quantile regression was used to generate percentile curves for weight loss.

Results: Out of 2,458 enrolled infants, 29.6% got censored, so 1,730 (1,134 vaginal, 596 cesarean) were included in the final analysis, contributing to 10,346 weight measurements. Median weight loss was 7.4% for vaginal and 8.7% for cesarean births; ≥10% loss occurred in 9.4% and 22.4% of infants, respectively. Distinct patterns and nadirs were observed based on mode of delivery. Compared to existing nomograms, our data showed greater early weight loss but quicker recovery after 48 h.

Conclusions: These nomograms provide region-specific reference standards for monitoring postnatal weight loss among exclusively breastfed infants in a semiarid to arid region, offering a basis for further validation in other arid settings globally.

Introduction: Spontaneous ductal closure is common in preterm populations; however, a subset of infants develops a hemodynamically significant PDA (hsPDA), which has been associated with adverse outcomes. The objective was to develop and internally validate a predictive model for hsPDA in preterm infants using a machine learning approach.

Methods: A prospective cohort study including infants born at <33 weeks of gestation. B-type natriuretic peptide (BNP) levels within the first 120 h, gestational age, birth weight, and surfactant use were used to train a random forest classifier. The outcome was hsPDA diagnosed by standardized echocardiography. Model performance was assessed using stratified 5-fold cross-validation.

Results: Sixty-seven infants were included; 46.3% had hsPDA. The random forest model achieved an area under the receiver operating characteristic curve (AUC) of 0.86, outperforming logistic regression using BNP alone (AUC 0.82). BNP was the strongest predictor (48% importance), followed by gestational age, birth weight, and surfactant use.

Conclusion: A machine learning-based model combining BNP with clinical variables showed high accuracy in predicting hsPDA. The accompanying calculator may assist clinicians in early risk stratification, though external validation is required before clinical implementation.

Introduction: Monitoring cerebral oxygenation during immediate foetal-to-neonatal transition may provide additional information in preterm neonates. Cerebral fractional tissue oxygen extraction (cFTOE), derived from cerebral oxygen saturation (crSO₂) and arterial oxygen saturation (SpO₂), reflects the relative extraction of oxygen from the arterial to the tissue compartment, providing information about the balance between oxygen delivery and oxygen consumption. We aimed to describe centiles of cFTOE during the first 15 min after birth in extremely and very preterm neonates.

Methods: This is a secondary outcome parameter analysis of the multicentre randomised-controlled COSGOD III trial. Neonates <32 weeks of gestational age included in the near-infrared spectroscopy-open group of the COSGOD III trial with favourable outcome, defined as survival without cerebral injury at term-equivalent age, were assigned for this analysis. CFTOE was calculated for every minute in each included neonate: cFTOE = (SpO₂-crSO₂)/SpO₂. CrSO2 was measured with the INVOS 5100 monitor with the neonatal sensor. Centiles of cFTOE (10th to 90th) from minute 2 to 15 after birth were described.

Results: A total of 199 preterm neonates with a median (interquartile range) gestational age of 29.7 (27.7-30.9) weeks and a weight of 1,200 (925-1,460) g were analysed. The 50th centile of cFTOE at minute 2, 5, 10, and 15 was 0.492, 0.296, 0.177, and 0.151, respectively.

Conclusion: This study provides centile for cFTOE for extremely and very preterm neonates with favourable outcome independent of interventions during postnatal stabilisation period. These centile charts may assist in interpreting cerebral oxygenation patterns.

Introduction: Neonatal gastrointestinal perforation is a life-threatening condition that requires timely and accurate diagnosis. However, interpreting abdominal radiographs in this population is often challenging. In this study, we aimed to develop a deep convolutional neural network (DCNN) model to segment pneumoperitoneum on neonatal abdominal radiographs and to evaluate its potential to assist in detecting neonatal gastrointestinal perforation.

Methods: This multicenter retrospective study included 1,187 abdominal radiographs (181 perforation and 1,006 control images) from neonates with gastrointestinal perforation and controls. Pneumoperitoneum regions were annotated by experienced clinicians. The dataset was randomly divided into training (n = 830), validation (n = 118), and test (n = 239) sets. A DeepLabV3+ model with ResNet50 backbone was fine-tuned for pixel-level segmentation. A single pixel-based threshold, derived from ROC analysis, was used to classify gastrointestinal perforation, with diagnostic performance subsequently compared to that of clinicians.

Results: The DCNN model achieved a median Dice similarity coefficient of 0.81 on the test dataset, indicating strong overlap between predicted and actual pneumoperitoneum regions. Furthermore, segmentation performance was positively correlated with pneumoperitoneum volume (Spearman ρ = 0.83, p < 0.001). Classification using the pixel-based cut-off demonstrated excellent diagnostic accuracy (AUC, 0.999; sensitivity, 100%; specificity, 98.5%), comparable to experienced clinicians.

Conclusion: The DCNN model demonstrated robust segmentation and classification performance, highlighting its potential as a clinical decision support tool for early detection of gastrointestinal perforation in neonates. Future studies should validate the model's generalizability and assess its integration into clinical practice.

Introduction: The use of reliable, validated, and multidimensional tools for pain evaluation has been recommended to manage neonatal pain. However, these tools have limited use in Japan due to their complexity, which involves numerous evaluation and observation items, making thorough observation challenging.

Methods: We developed a new method based on the Premature Infant Pain Profile-Revised (PIPP-R), a multidimensional evaluation tool that includes physiological and behavioral indicators, to simplify the evaluation of facial expressions. Instead of assessing three facial expressions individually, we evaluate them in four categories. We also created a system that automatically records changes in vital signs and calculates scores. In this study, we determined if the facial expression score aligns with the conventional PIPP-R-based facial expression score. The scoring methods were categorized into three types: (1) a real-time new method, where facial expressions were evaluated concurrently with the puncture procedure using a new method; (2) an individual assessment method using recorded videos from the puncture sessions with facial expressions scored later using the PIPP-R; and (3) a new video-based method, in which facial expressions were evaluated using the new method while watching recorded videos.

Results: The study included 63 healthy neonates (born at ≥37 weeks' gestation) delivered at our hospital. The degree of agreement among the methods and the inter-rater agreement showed high levels of consistency.

Conclusion: The new facial expression assessment method based on the PIPP-R demonstrated equivalence to conventional scoring in full-term neonates. Further validation, particularly in preterm infants and diverse clinical settings, is needed.

Introduction: The primary aim of this study was to evaluate the total seizure burden (TSB) and maximum hourly seizure burden (MSB) before and after the administration of lidocaine (LDC) as add-on therapy in neonates with amplitude-integrated electroencephalography (aEEG)-confirmed seizures. Secondary aims were documenting the need for additional ASM at 4, 12, and 24 h after LDC and the rate of seizure freedom for at least 24, 48, and 72 h after LDC.

Methods: This single-center, retrospective cohort study included neonates with persistent seizures after phenobarbital who received add-on LDC therapy. Neonates were monitored with a continuous 2-channel aEEG. The TSB and MSB were calculated using raw EEG data collected 4 h before and after the administration of LDC.

Results: Sixty-one neonates were included. Seizure etiology consisted of hypoxic-ischemic encephalopathy (n = 24), hemorrhagic or ischemic stroke (n = 16), central nervous system infection (n = 7), genetic (n = 8), metabolic disorders (n = 4), and unknown etiology (n = 2). After LDC administration, median TSB decreased significantly from 31 (interquartile range [IQR] 16-68) to 0 min (IQR 0-0, p < 0.01) and MSB from 10 (IQR 6-41) to 0 min/h (IQR 0-0, p < 0.01). The need for additional ASM was 3% (2/61) within 4 h and 41% (25/61) within 24 h. Seizure freedom after LDC was achieved in 71% (42/59) for at least 24 h and 52% (29/56) for 72 h.

Discussion: LDC significantly reduced TSB and MSB in neonates with (a)EEG-confirmed seizures, achieving 100% seizure reduction in the majority, with half of the neonates remaining seizure-free for at least 72 h.

Introduction: The PREMOD2 trial, comparing cord milking (CM) and deferred cord clamping (DCC) in preterm infants, was stopped for increased severe intraventricular hemorrhage (sIVH) with CM. Six of 9 centers had approval for waiver of antenatal consent.

Methods: We examined the relationship of enrollment procedures with characteristics and outcomes of trial-enrolled patients.

Results: A total of 474 infants were enrolled. Participants enrolled at sites with waiver of consent (N = 375, 79% of participants) were less likely exposed to antenatal steroids and magnesium. The overall effect of CM on sIVH was independently observed in sites with a waiver but not observed in sites without a waiver. However, the effects of CM observed between sites based on availability of waiver were not different. Chorioamnionitis exposure also modified the risk of sIVH from CM vs. DCC.

Conclusions: Trial-enrolled infants differed between hospitals with and without access to initial waiver, including in exposure to chorioamnionitis. These observations may be helpful to designing future studies.

Introduction: Retinopathy of prematurity (ROP) is a leading cause of childhood blindness. However, current screening guidelines may be overly broad, necessitating better models to detect high-risk infants.

Methods: From a multicenter cohort of 103,701 infants (3,301 [3.2%] treated for ROP) discharged from 298 neonatal intensive care units from 2006 to 2017 with birth weight ≤1,500 grams or gestational age ≤30 weeks, we used clinically relevant variables to develop machine learning (ML) models at 2-week intervals from postnatal day 14 to 98 to stratify infants by ROP treatment timing. We assessed model performance by concordance index, area under the receiver operating characteristic curve (AUROC), and average precision (AP), validated performance in a cohort of 25,105 infants across 231 sites from 2018 to 2020, and compared model performance to a logistic regression (LR) model.

Results: In the validation cohort, the day 28 ML model outperformed the LR model by AUROC (0.916 [0.905-0.926] vs. 0.903 [0.892-0.914]; p < 0.001) and AP (0.190 [0.167-0.217] vs. 0.160 [0.140-0.183]; p < 0.001). Using the ML model at a 100% sensitivity threshold would have negative predictive value of >99.9% and could reduce the number of infants needing screening by 14% compared to current guidelines.

Conclusion: ML models can effectively predict the need for ROP treatment and stratify infants by risk, potentially reducing unneeded screening. Future work is needed to translate model-based ROP predictions to the clinical setting.

Introduction: High-dose ambroxol is an effective pharmacological chaperone therapy for the systemic and neurological symptoms of Gaucher disease (GD). However, no clinical evidence of perinatal-onset GD has been documented.

Case presentation: The patient had perinatal-onset neuronopathic GD (PnGD) and received high-dose ambroxol, beginning at 10 days of life after a newborn screening report. There was a transient hematological response after combined ambroxol and enzyme replacement therapy; however, laryngospasm, epileptic seizures, liver dysfunction, and heart failure progressed. The patient died 95 days after birth. Genetic testing revealed a homozygous L483R variant in GBA1. A literature review of 56 patients with nGD confirmed poor survival outcomes for patients with PnGD.

Conclusion: Ambroxol therapy may be insufficient to improve the prognosis of patients with PnGD, underscoring the limitations of early intervention in newborn-screened patients with GD. Therefore, pre-emptive therapeutic strategies are required to rescue and cure neonates with PnGD.