Annamaria Sbordone

Objectives: Non-invasive ventilation (NIV) is a widely used treatment for neonatal respiratory distress syndrome (RDS). Data on diaphragm contractility and thickness during NIV is scarce. We aimed to describe changes in diaphragm thickness/contractility during NIV and to explore associations with NIV discontinuation failure. Methods: This is a single-center prospective study. Diaphragmatic ultrasound was performed weekly during NIV, then within 7 days from NIV discontinuation. Diaphragm thickness was measured at end-inspiration (DTI) and end-expiration (DTE). Diaphragm thickening fraction (DTF) was calculated as (DTI-DTE/DTE). The clinical characteristics of patients and NIV discontinuation failure were recorded. Univariate analysis, logistic regression and linear regression were performed to describe diaphragm features during NIV and associations with NIV discontinuation failure. Results: We studied 17 NIV cycles (median duration 21 days). Median DTE increased from 0.12 cm (SD 0.05) at the start of NIV to 0.15 cm (SD 0.04) at NIV discontinuation. The mean DTF decreased from 32.8 (SD 16.8) at the start of NIV to 25.6 (SD 8.9) at NIV discontinuation. NIV discontinuation failure occurred in 23.5% of infants and was associated with higher DTI and DTE at the start of NIV and with a more pronounced decrease in DTI and DTE over the NIV cycle, compared to infants with NIV discontinuation success. There were no differences in neonatal outcomes between the infants with NIV discontinuation failure vs. success. We did not find any significant predictors of NIV failure. Conclusions: Diaphragm thickness increased, whereas DTF decreased over time on NIV in preterm infants with RDS. NIV duration was not associated with changes in diaphragm trophism. NIV discontinuation failure was associated with thicker diaphragm at the start of NIV, as well as with a reduction in diaphragm trophism over the NIV cycle.

Background: Respiratory problems are frequent in newborns, and are mainly studied with chest X-rays, whereas CT scans are usually needed for the evaluation of rare malformations and diseases. Lung ultrasound (LUS] has been proposed as an alternative method of diagnosing a variety of respiratory conditions. In recent years, there has been a rapid increase in LUS studies, thanks to the ability of LUS to rapidly exclude complications and significantly reduce radiation exposure in this fragile population. We aimed to summarize the current knowledge about LUS. Methods: A literature search was conducted on the Medline and Cochrane databases using appropriate terms. The inclusion criteria were: English language and human species. Exclusion criteria were: non-English language, animal species, case reports, case series, non-systematic reviews, and editorials. Results: The search returned 360 results. No Cochrane reviews were found. Titles and abstracts were screened, and 37 were finally considered. Studies concerning the use of lung ultrasound for the following conditions were presented: neonatal respiratory distress syndrome, transient tachypnea of the newborn, pneumothorax, pulmonary hemorrhage, pneumonia, bronchopulmonary dysplasia, and prediction of extubation success. Conclusions: We discussed the utility of LUS for the diagnosis and treatment of neonatal diseases according to the most recent literature.

Background: Tactile stimulation manoeuvres stimulate spontaneous breathing in preterm newborns. The aim of this study is to evaluate the effect of early respiratory physiotherapy on the need for mechanical ventilation during the first week of life in preterm infants with respiratory failure. Methods: This is a monocentric, randomised controlled trial. Preterm infants (gestational age ≤ 30 weeks) not intubated in the delivery room and requiring non-invasive respiratory support at birth were eligible for the study. The intervention group received early respiratory physiotherapy, while the control group received only a daily physiotherapy program (i.e., modifying the infant’s posture in accordance with the patient’s needs). Results: between October 2019 and March 2021, 133 preterm infants were studied, 68 infants in the study group and 65 in routine care. The study group showed a reduction in the need for mechanical ventilation (not statistically significant) and a statistically significant reduction in hemodynamically significant patent ductus arteriosus with respect to the control group (19/68 (28%) vs. 35/65 (54%), respectively, p = 0.03). Conclusions: early respiratory physiotherapy in preterm infants requiring non-invasive respiratory support at birth is safe and has proven to be protective against haemodynamically significant PDA.

Intrahepatic cholestasis of pregnancy (ICP) complicates among 0.2–2% of pregnancies and has been associated with adverse perinatal outcomes, including sudden stillbirth, meconium strained fluid, preterm birth, perinatal asphyxia, and transient tachypnea of the newborn. The diagnosis of “bile acids pneumonia” was previously proposed and a causative role of bile acids (BA) was supposed with a possible mechanism of action including surfactant dysfunction, inflammation, and chemical pneumonia. In the last few years, the role of lung ultrasound (LUS) in the diagnosis and management of neonatal respiratory distress syndrome has grown, and LUS scores have been introduced in the literature, as an effective predictor of the need for surfactant treatment among neonates with respiratory distress syndrome. We present four cases of infants born from pregnancies complicated by ICP, who developed respiratory distress syndrome early after birth. Lung ultrasound showed the same pattern for all infants, corresponding to a homogeneous alveolar–interstitial syndrome characterized by a diffuse coalescing B-line pattern (white lung). All infants evaluated require non-invasive respiratory support and in three cases surfactant administration, despite the near-term gestational age, with rapid improvement of respiratory disease and a good clinical outcome.

Background: The present study was designed to assess the feasibility and reliability of a Continuous Glucose Monitoring System (CGMS) in a population of asphyxiated neonates during therapeutic hypothermia. Methods: This non-randomized feasibility study was conducted in the Neonatal Intensive Care Unit (NICU) facilities of Fondazione Policlinico A. Gemelli IRCSS. Infants matching the criteria for hypothermic treatment were included in this study and were connected to the CGMS (Medtronic, Northridge, CA, USA) within the first 12 h of life. Hypoglycemia was defined as a glucose value ≤ 47 mg/dL, and hyperglycemia was defined as a glucose value ≥ 180 mg/dL. Data obtained via the CGMS were compared with those obtained via a point-of-care blood glucometer (GTX). Results: The two measuring techniques were compared using the Modified Clarke Error Grid (MCEG). Sixteen infants were enrolled. The sensor had an average (standard deviation) duration of 93 (38) h. We collected 119 pairs of glycemia values (CGMVs) from the CGMS vs. GTX measurements. The CGMS detected twenty-five episodes of hypoglycemia and three episodes of hyperglycemia. All the CGMVs indicating hyperglycemia matched with the blood sample taken via the point-of-care glucometer. Conclusions: The use of a CGMS would be useful as it could detect more episodes of disglycemia than standard care. Our data show poor results in terms of the accuracy of the CGMS in this particular setting.

Background: Artificial intelligence (AI) is a promising field in the neonatal field. We focused on lung ultrasound (LU), a useful tool for the neonatologist. Our aim was to train a neural network to create a model able to interpret LU.Methods: Our multicentric, prospective study included newborns with gestational age (GA) ≥ 33 + 0 weeks with early tachypnea/dyspnea/oxygen requirements. For each baby, three LU were performed: within 3 h of life (T0), at 4–6 h of life (T1), and in the absence of respiratory support (T2). Each scan was processed to extract the region of interest used to train a neural network to classify it according to the LU score (LUS). We assessed sensitivity, specificity, positive and negative predictive value of the AI model's scores in predicting the need for respiratory assistance with nasal continuous positive airway pressure and for surfactant, compared to an already studied and established LUS.Results: We enrolled 62 newborns (GA = 36 ± 2 weeks). In the prediction of the need for CPAP, we found a cutoff of 6 (at T0) and 5 (at T1) for both the neonatal lung ultrasound score (nLUS) and AI score (AUROC 0.88 for T0 AI model, 0.80 for T1 AI model). For the outcome “need for surfactant therapy”, results in terms of area under receiver operator characteristic (AUROC) are 0.84 for T0 AI model and 0.89 for T1 AI model. In the prediction of surfactant therapy, we found a cutoff of 9 for both scores at T0, at T1 the nLUS cutoff was 6, while the AI's one was 5. Classification accuracy was good both at the image and class levels.Conclusions: This is, to our knowledge, the first attempt to use an AI model to interpret early neonatal LUS and can be extremely useful for neonatologists in the clinical setting.

The development of artificial intelligence methods has impacted therapeutics, personalized diagnostics, drug discovery, and medical imaging. Although, in many situations, AI clinical decision-support tools may seem superior to rule-based tools, their use may result in additional challenges. Examples include the paucity of large datasets and the presence of unbalanced data (i.e., due to the low occurrence of adverse outcomes), as often seen in neonatal medicine. The most recent and impactful applications of AI in neonatal medicine are discussed in this review, highlighting future research directions relating to the neonatal population. Current AI applications tested in neonatology include tools for vital signs monitoring, disease prediction (respiratory distress syndrome, bronchopulmonary dysplasia, apnea of prematurity) and risk stratification (retinopathy of prematurity, intestinal perforation, jaundice), neurological diagnostic and prognostic support (electroencephalograms, sleep stage classification, neuroimaging), and novel image recognition technologies, which are particularly useful for prompt recognition of infections. To have these kinds of tools helping neonatologists in daily clinical practice could be something extremely revolutionary in the next future. On the other hand, it is important to recognize the limitations of AI to ensure the proper use of this technology.

MAS is a common cause of neonatal respiratory distress in term and post-term neonates. Meconium staining of the amniotic fluid occurs in about 10–13% of normal pregnancies, and about 4% of these infants develop respiratory distress. In the past, MAS was diagnosed mainly on the basis of history, clinical symptoms, and chest radiography. Several authors have addressed the ultrasonographic assessment of the most common respiratory patterns in neonates. In particular, MAS is characterised by a heterogeneous alveolointerstitial syndrome, subpleural abnormalities with multiple lung consolidations, characterised by a hepatisation aspect. We present six cases of infants with a clinical history of meconium-stained fluid who presented with respiratory distress at birth. Lung ultrasound allowed the diagnosis of MAS in all the studied cases, despite the mild clinical picture. All children had the same ultrasound pattern with diffuse and coalescing B-lines, pleural line anomalies, air bronchograms, and subpleural consolidations with irregular shapes. These patterns were distributed in different areas of the lungs. These signs are specific enough to distinguish between MAS and other causes of neonatal respiratory distress, allowing the clinician to optimise therapeutic management.

Background Artificial intelligence (AI) is a promising field in the neonatal field.   We focused on lung ultrasound (LUS), a useful tool for the neonatologist. Our aim was to train a neural network to create a model able to interpret LUS. Methods Our multicentric, prospective study included newborns with gestational age (GA) ≥ 33+0 weeks with early tachypnea/dyspnea/oxygen requirements. For each baby, three LUS were performed: within 3 hours of life (T0), at 4–6 hours of life (T1) and in the absence of respiratory support (T2). Each scan was processed to extract ROI used to train a neural network to classify it according to the LUS score. We assessed sensitivity, specificity, positive and negative predictive value of the AI model’s scores in predicting the need for respiratory assistance with nasal Continuous Positive Airway Pressure (nCPAP) and for surfactant, compared to the “classical” scores. Results We enrolled 62 newborns (GA=36±2 weeks). In the prediction of the need for CPAP, we found a cut-off of 6 (at T0) and 5 (at T1) for both the classical nLUS and AI score. In the prediction of surfactant therapy we found a cut-off of 9 for both scores at T0, at T1 the nLUS cut-off was 6, while the AI’s one was 5. Classification accuracy was good both at the image and classes level. Conclusions This is, to our knowledge, the first attempt to use an AI model to interpret early neonatal LUS and can be extremely useful for neonatologist in the clinical setting.