Pediatric Radiology最新文献

Anyone who has visited a radiology department in another continent, country, or even city has likely noticed subtle differences in how things are done compared to their own department. Many procedures become so ingrained in daily practice that we rarely stop to question whether they are truly supported by evidence. One such procedure is intussusception reduction. In this paper, we review the literature, examine the best reduction technique and the risks associated with general anaesthesia and sedation, and offer the Abdominal Task Force's opinion on this common but very variable procedure, based on the current available literature.

Pulled elbow, colloquially named nursemaid's elbow, is an injury that results from annular ligament displacement and interposition between the radial head and capitellum. There is a paucity of literature describing the sonographic findings of pulled elbow. Given that this condition is primarily diagnosed clinically, radiographs are typically only obtained when there is concern for fracture. Sonographic evaluation is most useful in cases when attempted reduction is unsuccessful and radiographs are negative for fracture. This review describes the essential pathoanatomy needed for accurate sonographic diagnosis, which has previously been described with cadaveric research and a small number of case reports.

Medication-induced lung disease is rare in children but can be associated with considerable morbidity and mortality and can alter treatment regimens for children with a variety of conditions. Medication-induced lung disease tends to occur in association with certain categories of medication and tends to manifest as one of several patterns of lung disease on CT. We review classes of medications associated with medication-induced lung disease, clinical diagnosis and management of the condition including the role of the radiologist, and CT patterns of disease to enable the radiologist to more fully contribute to multidisciplinary diagnosis and potentially be the first to recognize and suggest the possibility of this condition.

Background: Many children are born without prenatal determination of gestational age (GA). Postnatal determinations are limited. Fetal GA determination using transcerebellar diameter measurements is reliable for fetuses. We wanted to see if transcerebellar diameters obtained on neonatal head ultrasound exams could help GA determination in newborns and whether such measurements conformed to similar GA determinations in fetuses.

Objective: Our goal was to determine if neonatal GA can be estimated by measuring transcerebellar diameter via a transmastoid approach using fetal charts as the gold standard. If true, one could develop a neonatal chart for GA determination by transcerebellar diameter.

Materials and methods: Transmastoid views are a routine part of our neonatal intensive care unit neurosonograms. A retrospective analysis of transcerebellar diameters of neonates (1 day to 21 days old) born between 22 weeks and 40 weeks corrected GA was performed. Cases with congenital anomalies, intraventricular hemorrhage, or other neurosonographic abnormalities were excluded. Neonatal GA was determined by early antenatal crown rump lengths. We calculated transcerebellar mean and standard deviation for each prenatally determined GA week. GA was determined from fetal charts, both for subsets of neonates evaluated at less than or equal to (≤) 10 days of life and for those examined at ≤21 days of life. Statistical analysis using linear regression demonstrated no differences in GA determined by neonatal transcerebellar diameter compared to fetal charts (our gold standard).

Results: We evaluated 1,260 neurosonograms. Of these, 589 cases were excluded. A total of 671 exams were of neonates ≤21 days old; 530 of those were examined at ≤10 days of life. There were no significant differences between GA determined by fetal charts and our neonatal transcerebellar diameters, whether from the ≤21-day (P=0.15) or the younger ≤10-day group (P=0.87).

Conclusion: Neonatal GA estimation by transmastoid fontanelle measurements of cerebellar width appears as reliable as the accepted antenatal transcerebellar measurements of fetuses. Our proposed neonatal chart will hopefully aid reliable estimation of GA in neonates, improving patient care among neonates with unknown maternal last menstrual period.

Background: In vivo placental volume derived from magnetic resonance imaging (MRI) is a novel imaging tool to evaluate the placenta during pregnancy, as the placenta is difficult to access throughout gestation. There is a paucity of established standardized normative raw values and Z-scores for in vivo placental volume based on MRI.

Objective: To establish normative references for in vivo placental MRI-based volumes derived from a large cohort of healthy pregnant women carrying healthy fetuses throughout gestation.

Materials and methods: Healthy pregnant women with healthy singleton pregnancies greater than 16 weeks gestation were enrolled in a longitudinal, prospective observational study. In total, 313 placental MRIs were analyzed from 209 pregnant women. In-vivo placentas were manually segmented to derive volumes and Z-scores. Means, standard deviations, and percentiles for normative reference raw values were calculated using weekly gestational age (GA) bins. Placental volume Z-scores were calculated based on 2-week GA bins using means and standard deviations.

Results: Normative reference placental volumes from 209 subjects (313 scans) with median GA 31.43 [8.86] weeks are presented in weekly and bi-weekly GA bins. Using 2-week GA intervals, 95% of placental volume Z-scores were within ±2 standard deviations of the population mean.

Conclusion: This data provides established normative in vivo raw and Z-score values derived from placental MRI. The value of accessing the placenta in vivo through MRI has become increasingly recognized, as the importance of the placenta in fetal and postnatal health is now more widely known. Establishing normative reference values for the in vivo placenta throughout gestation benefits both the clinical and scientific communities.

This report presents the prenatal imaging findings and short-term clinical outcome of a rare case of fetal cerebral sinovenous thrombosis, along with a brief review of the literature. A 31-year-old pregnant Afghan woman at 34 weeks of gestation (gravida 3, para 0, abortion 2) presented to the hospital with complaints of decreased fetal movement. Fetal cranial ultrasonography revealed a thick-walled, avascular, hypoechoic, mass-like lesion (17×18×18 mm) extending from the left foramen of Monro to the superior aspect of the thalamus and displacing the body of the left lateral ventricle. Fetal cranial magnetic resonance imaging (MRI) suggested subacute hemorrhage and demonstrated hyperintensity in the left transverse venous sinus, indicative of thrombosis. Serial follow-up revealed stability of the lesion; however, at 36 weeks, an emergency cesarean section was performed due to fetal distress. Postnatal cranial MRI and magnetic resonance venography (MRV) confirmed left transverse venous sinus thrombosis with subacute hemorrhage. The newborn received enoxaparin therapy for 3 months, remained clinically stable, and had no neurological deficits at the 4-month follow-up. Fetal intracranial hemorrhage has a broad differential diagnosis, and accurate identification of the underlying etiology is critical for prognosis and management. Fetal cerebral sinovenous thrombosis should be considered in the differential diagnosis when mass-like hemorrhagic lesions are detected on prenatal imaging. While short-term outcomes may be favorable in the absence of additional cranial pathology, the long-term prognosis remains uncertain.

Neuroleukemiosis, leukemic infiltration of the peripheral nerves, is exceedingly rare. We report a child with neuroleukemiosis affecting multiple peripheral nerves, presenting 2 months after stem cell transplant for acute myeloid leukemia. We discuss this unusual entity and highlight the ultrasonographic and magnetic resonance imaging findings that facilitated biopsy confirmation. Recognition of this entity is critical as it may be the sole sign of primary disease or relapse.

Background: Most commercially available artificial intelligence (AI) tools in radiology are trained and approved for adult use, creating an access gap for pediatric patients. Intracranial hemorrhage (ICH) detection is a common adult AI application without pediatric FDA clearance.

Objective: To evaluate the performance of an FDA-cleared, adult-trained AI tool for ICH detection on non-contrast head CT (NCHCT) in pediatric patients aged 6-17 years.

Materials and methods: This retrospective, multi-institution study analyzed consecutive pediatric NCHCTs performed between January 2017 and November 2022 across 21 sites. Inclusion criteria were patient age 6-17 years and adequate imaging quality. Radiology reports were classified as ICH-positive or ICH-negative using a validated natural language processing (NLP) tool. The AI tool analyzed DICOM images independently. Discordant AI-NLP cases underwent blinded adjudication by three radiologists to establish ground truth. Performance metrics includingsensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) were calculated with Wilson 95% confidence intervals (CIs).

Results: The cohort included 1,996 NCHCTs (768 females, 1,223 males, 5 unknown). ICH prevalence was 8.6% (172/1,996). Compared with ground truth, AI achieved 94.2% sensitivity (162/172, 95% CI, 89.6-97.2%), 94.7% specificity (1,727/1,824, 95% CI, 93.6-95.7%), 94.6% accuracy (1,889/1,996, 95% CI, 93.6-95.6%), 62.5% PPV (162/259, 95% CI, 57.8-67.0%), and 99.4% NPV (1,727/1,737, 95% CI, 99.0-99.7%). AI correctly identified ICH in cases missed by radiologists, but false positives were common, most often due to streak artifact (21.6%) and misclassified anatomy (18.6%). Interrater agreement for ground truth adjudication was substantial (κ=0.683).

Conclusion: An adult-trained AI tool demonstrated high sensitivity, specificity, and accuracy for ICH detection in pediatric patients aged 6-17 years, comparable to its adult performance. Selective adaptation of adult-trained AI tools could expand access to AI-assisted triage for certain pediatric populations, potentially reducing delays in critical imaging interpretation. However, prospective validation is required before clinical deployment.

Chronic nonbacterial osteitis (CNO), including its multifocal form, chronic recurrent multifocal osteomyelitis (CRMO), is an autoinflammatory sterile bone disorder with protean imaging features across the appendicular and axial skeletons in children and adolescents. Magnetic resonance imaging (MRI)-particularly whole-body MRI (WBMRI)-is pivotal for diagnostic imaging, mapping total inflammatory burden, demonstrating multifocality, and monitoring treatment response. Radiographs are frequently normal early in the disease. Beyond the classic metaphyseal pattern, we demonstrate atypical presentations that often mimic other conditions, such as infection or malignancy: progressive collapsing vertebral disease with vertebra plana, hip arthritis due to involvement of intra-articular metaphyseal equivalents, myositis not related to areas of active osteitis, extensive periosteal new bone in long bones, pseudotumoral lesions of flat bones, and craniofacial involvement. This pictorial review provides a structured diagnostic approach to prompt consideration of CNO, facilitate differentiation from bacterial osteomyelitis and bone tumors in the differential diagnosis, and support early, targeted anti-inflammatory therapy.