Neuroimaging of Germinal Matrix and Intraventricular Hemorrhage in Premature Infants

Ultrasound-Guided External Ventricular Drain Insertion After Decompressive Craniectomy

BACKGROUND

Ultrasound guidance offers real-time visualization of patient-specific anatomy during external ventricular drain (EVD) insertion. A craniectomy defect provides a sonolucent window, enabling the use of a large, low-frequency probe with deep penetration and a wide field of view. While specialized burr-hole probes exist, use of a curvilinear probe through a craniectomy defect for bedside EVD placement has not been previously described.

METHODS

Using a curvilinear probe, we performed ultrasound-guided bedside insertion of a left frontal EVD through a hemicraniectomy flap.

RESULTS

Bedside ultrasound enabled visualization of the entire supratentorial ventricular system. Drain insertion was successfully performed, with immediate sonographic visualization of the catheter tip in the left frontal horn. Placement was confirmed with a computed tomography scan.

CONCLUSIONS

Bedside ultrasound-guided EVD insertion in post-craniectomy patients can be a valuable method for safely accessing the ventricle in the face of abnormal or distorted anatomy.

Noninvasive optical monitoring of cerebral hemodynamics in a preclinical model of neonatal intraventricular hemorrhage

Intraventricular hemorrhage (IVH) is a common complication in premature infants and is associated with white matter injury and long-term neurodevelopmental disabilities. Standard diagnostic tools such as cranial ultrasound and MRI are widely used in both preclinical drug development and clinical practice to detect IVH. However, these methods are limited to endpoint assessments of blood accumulation and do not capture real-time changes in germinal matrix blood flow leading to IVH. This limitation could potentially result in missed opportunities to advance drug candidates that may have protective effects against IVH. In this pilot study, we aimed to develop a noninvasive optical approach using diffuse correlation spectroscopy (DCS) to monitor real-time hemodynamic changes associated with hemorrhagic events and pre-hemorrhagic blood flow in a preclinical rabbit model of IVH. DCS measurements were conducted during the experimental induction of IVH, and results were compared with ultrasound and histological analysis to validate findings. Significant changes in hemodynamics were detected in all animals subjected to IVH-inducing procedures, including those that did not show clear positive results on ultrasound 18 h later. The study revealed progressively elevated coefficients of variation in blood flow, largely driven by temporal fluctuations in the <0.25 Hz range. Our findings suggest that real-time optical monitoring with DCS can provide critical insights heralding pathological blood flow changes, offering a more sensitive and informative tool for evaluating potential therapeutics that may help avert the progression to IVH.

Elimination of Candida Sepsis and Reducing Several Morbidities in a Tertiary NICU in Greece After Changing Antibiotic, Ventilation, and Nutrition Protocols

BACKGROUND/OBJECTIVES

In recent years, strategies for improving outcomes in preterm neonates have been implemented in various aspects of neonatal care. This study aims to determine the prevalence, microbiology, and outcomes of late-onset sepsis (LOS) and the incidence of other morbidities in very preterm neonates following the implementation of specific infection control, enteral feeding, and ventilation strategies.

METHODS

This study retrospectively compared the morbidity and mortality of preterm neonates with a 23-32 weeks gestational age over two periods, period A (2010-2014),and period B (2018-2022). A series of changes were introduced between these periods, including restrictive use of antibiotics, aggressive enteral feeding, and wider use of non-invasive ventilation modalities.

RESULTS

A total of 310 neonates were included: 163 in period A and 147 in period B. The mean duration of antibiotic treatment was reduced from 4 ± 2 to 2 ± 1 days and from 5 ± 2 to 3 ± 1 days for suspected early-onset sepsis and LOS, respectively, and from 11.2 ± 4 to 16 ± 4 days for confirmed LOS between the two study periods. The incidence of LOS was 24% and 18%, while, for multiple LOS episodes, it was 26% and 11% in periods A and B, respectively. Total parenteral nutrition (TPN) duration and gestational age were independent predictors of LOS in both periods. The rate of Candida infections declined from 9.2% to 0.7%. The full enteral nutrition in period B was achieved after a median of 7.5 days compared with 10 days (p = 0.001), resulting in fewer days of TPN (p = 0.008). Episodes of feeding intolerance and necrotizing enterocolitis I (NEC I) were significantly reduced (p < 0.001). Incidence of intraventricular hemorrhage were significantly decreased.

CONCLUSIONS

After changing antibiotic, ventilation, and nutrition protocols, Candida infections were almost completely eliminated. The incidence of LOS and multiple LOS episodes decreased. Early full enteral nutrition was achieved without adverse effects, and fewer episodes of food intolerance were observed. Candida elimination appears feasible when antibiotic stewardship is implemented in conjunction with other interventions in an NICU.

Neonatal cerebral ultrasound: anatomical variants and age-related diseases

Cerebral ultrasound is a non-invasive imaging technique widely used for the assessment of brain anatomy and diseases in neonates and infants. Indeed, it allows a precise characterization of common variants such as cavum septum pellucidum or diseases like intraventricular hemorrhage. The aim of this pictorial review is to provide a comprehensive overview of the main ultrasound features of the most common cerebral anatomical variants and disorders detectable by cerebral ultrasound using an age-related approach which could support non-subspecialized radiologists.

Adequacy of an In-Neonatal Intensive Care Unit 1T Magnetic Resonance Imaging Compared With 3T Magnetic Resonance Imaging for Clinical Management

BACKGROUND

This retrospective study aims to assess the added diagnostic utility and clinical value of a 3-Tesla neonatal brain magnetic resonance imaging after obtaining a 1-Tesla magnetic resonance imaging within the neonatal intensive care unit.

METHODS

A cohort of 34 infants had an initial 1-Tesla magnetic resonance imaging and repeat imaging within 14 days in a 3-Tesla scanner. All infants were admitted to the level III neonatal intensive care unit at Brigham and Women's Hospital, and all images were interpreted by pediatric neuroradiologists.

RESULTS

For 31 infants (91%), the 3-Tesla magnetic resonance imaging showed similar or expected evolution of known findings found on 1 Tesla. For infants with change between the 1-Tesla and 3-Tesla imaging results, there was no clinical impact.

CONCLUSION

Images from 1-Tesla magnetic resonance imaging were sufficient for characterizing a wide range of neonatal brain injuries and abnormalities and repeated 3-Tesla magnetic resonance imaging did not yield further clinical benefit.

Diagnostic Accuracy of Cranial Ultrasound for the Detection of Intracranial Hemorrhage in Preterm Neonates Using Magnetic Resonance Imaging As the Gold Standard

Background and objective Intracranial hemorrhage (ICH) and white-matter damage are the main brain injuries in preterm infants. Magnetic resonance imaging (MRI) is the best way to examine cerebral bleeding. The evidence on cranial ultrasound diagnostic accuracy in neonates is limited in Pakistani publications, which show variability in evidence, necessitating the present study. The aim is to find out the diagnostic accuracy of cranial ultrasound for detecting ICH in preterm neonates, taking MRI as the gold standard. Methodology A cross-sectional study was carried out in the Department of Radiology, The Children's Hospital, Lahore, Pakistan, from June 19, 2018 to December 18, 2018. The study involved 103 preterm neonates of both genders aged between 1 and 27 days and presenting with suspicion of ICH. The cranial ultrasound with MRI (gold standard) was used to diagnose ICH, and the results of the cranial ultrasound were assessed accordingly. The diagnostic performance of ultrasound was estimated using parameters, e.g., sensitivity, specificity, accuracy, and positive and negative predictive values. The p < 0.05 was taken as statistically significant. Results The neonates had a mean age of 9.9 ± 6.6 days. There were 53 (51.5%) male and 50 (48.5%) female neonates. The cranial ultrasound diagnosed ICH in 39 (37.9%) neonates, which is confirmed by MRI as the gold standard. Cranial ultrasound had 100.0% sensitivity, 90.62% specificity, 94.17% accuracy, 86.67% positive predictive value, and 100.0% negative predictive value for detecting ICH in preterm neonates. Similar diagnostic performance was seen across age and gender groups (p > 0.05). Conclusion Cranial ultrasound was 100% sensitive, 90.6% specific, and 94.2% accurate in identifying cerebral bleeding in preterm neonates. The noninvasive, ionizing radiation-free nature and availability at the bedside of cranial ultrasound make it a promising future modality.

Imaging of Hemorrhagic Stroke in Children

Hemorrhagic stroke (HS) is an important cause of neurologic morbidity and mortality in children and is more common than ischemic stroke between the ages of 1 and 14 years, a notable contradistinction relative to adult stroke epidemiology. Rapid neuroimaging is of the utmost importance in making the diagnosis of HS, identifying a likely etiology, and directing acute care. Computed tomography and MR imaging with flow-sensitive MR imaging and other noninvasive vascular imaging studies play a primary role in the initial diagnostic evaluation. Catheter-directed digital subtraction angiography is critical for definitive diagnosis and treatment planning.

Deep Learning Approaches for the Assessment of Germinal Matrix Hemorrhage Using Neonatal Head Ultrasound

Germinal matrix hemorrhage (GMH) is a critical condition affecting premature infants, commonly diagnosed through cranial ultrasound imaging. This study presents an advanced deep learning approach for automated GMH grading using the YOLOv8 model. By analyzing a dataset of 586 infants, we classified ultrasound images into five distinct categories: Normal, Grade 1, Grade 2, Grade 3, and Grade 4. Utilizing transfer learning and data augmentation techniques, the YOLOv8 model achieved exceptional performance, with a mean average precision (mAP50) of 0.979 and a mAP50-95 of 0.724. These results indicate that the YOLOv8 model can significantly enhance the accuracy and efficiency of GMH diagnosis, providing a valuable tool to support radiologists in clinical settings.

Noninvasive optical monitoring of cerebral hemodynamics in a preclinical model of neonatal intraventricular hemorrhage

Intraventricular hemorrhage (IVH) is a common complication in premature infants and is associated with white matter injury and long-term neurodevelopmental disabilities. Standard diagnostic tools such as cranial ultrasound and MRI are widely used in both preclinical drug development and clinical practice to detect IVH. However, these methods only provide endpoint assessments of blood accumulation and lack real-time information about dynamic changes in ventricular blood flow. This limitation could potentially result in missed opportunities to advance drug candidates that may have protective effects against IVH. In this pilot study, we aimed to develop a noninvasive optical approach using diffuse correlation spectroscopy (DCS) to monitor real-time hemodynamic changes associated with hemorrhagic and sub-hemorrhagic events in a preclinical rabbit model of IVH. DCS measurements were conducted during the experimental induction of IVH, and results were compared with ultrasound and histological analysis to validate findings. Significant changes in hemodynamics were detected in all animals subjected to IVH-inducing procedures, including those that did not show clear positive results on ultrasound. The study revealed progressively elevated coefficients of variation in blood flow, particularly driven by increased oscillations within the 0.05-0.1 Hz frequency band. These hemodynamic changes were more pronounced in animals that developed IVH, as confirmed by ultrasound. Our findings suggest that real-time optical monitoring with DCS can provide critical insights into pathological blood flow changes, offering a more sensitive and informative tool for evaluating potential therapeutics in the context of IVH.