Correlative Assessment of Machine Learning-Based Cobb Angle Measurements and Human-Based Measurements in Adolescent Idiopathic and Congenital Scoliosis.

Purpose: Scoliosis is a complex spine deformity with direct functional and cosmetic impacts on the individual. The reference standard for assessing scoliosis severity is the Cobb angle which is measured on radiographs by human specialists, carrying interobserver variability and inaccuracy of measurements. These limitations may result in lack of timely referral for management at a time the scoliotic deformity progression can be saved from surgery. We aimed to create a machine learning (ML) model for automatic calculation of Cobb angles on 3-foot standing spine radiographs of children and adolescents with clinical suspicion of scoliosis across 2 clinical scenarios (idiopathic, group 1 and congenital scoliosis, group 2). Methods: We retrospectively measured Cobb angles of 130 patients who had a 3-foot spine radiograph for scoliosis within a 10-year period for either idiopathic or congenital anomaly scoliosis. Cobb angles were measured both manually by radiologists and by an ML pipeline (segmentation-based approach-Augmented U-Net model with non-square kernels). Results: Our Augmented U-Net architecture achieved a Symmetric Mean Absolute Percentage Error (SMAPE) of 11.82% amongst a combined idiopathic and congenital scoliosis cohort. When stratifying for idiopathic and congenital scoliosis individually a SMAPE of 13.02% and 11.90% were achieved, respectively. Conclusion: The ML model used in this study is promising at providing automated Cobb angle measurement in both idiopathic scoliosis and congenital scoliosis. Nevertheless, larger studies are needed in the future to confirm the results of this study prior to translation of this ML algorithm into clinical practice.

Fever at Time of Leukemia Diagnosis in Children: Predictor of Bloodstream Infection or Catheter Removal?

PURPOSE: To assess the incidence of fever at diagnosis in children with leukemia and determine if fever at diagnosis is a predictor of bloodstream infection (BSI) or central venous access device (CVAD) removal for infection either within the first 30 days or between 30 and 90 days after CVAD insertion. MATERIALS AND METHODS: One hundred fifty-one patients with acute leukemia (July 1, 2018, to December 31, 2020) who underwent a CVAD insertion within 2 weeks of diagnosis were included. Patient data included demographic characteristics, fever at diagnosis, CVAD type, antibiotics before and/or on the day of CVAD insertion, BSI incidence, BSI rates per 1,000 catheter days, and need for catheter removal after CVAD insertion within 30 days and between 30 and 90 days. RESULTS: Patients with fever at diagnosis had a significantly higher incidence of BSI within the first 30 days after CVAD insertion (17/23) than that among patients without fever (6/23) (P = .046) at diagnosis. No statistically significant difference was observed in the incidence of BSI between 30 and 90 days after CVAD insertion between patients with fever (5/11) and those without fever at diagnosis (6/11) (P = .519). Fever at diagnosis was not a predictor of CVAD removal within 30 days (9 patients required CVAD removal; 7/9 had fever and 2/9 had no fever) (P = .181) or between 30 and 90 days (4 patients required CVAD removal; 1/4 had fever and 3/4 had no fever at diagnosis) (P = .343) after insertion. CONCLUSIONS: Fever at diagnosis in patients with leukemia is not a predictor of CVAD removal for infection.

The impact of high-frequency transducers on the sonographic measurements of the pyloric muscle thickness in infants.

BACKGROUND: Ultrasound is the modality of choice for the diagnosis of hypertrophic pyloric stenosis (HPS). The evolution of high-frequency transducers in ultrasound has led to inconsistent ways of measuring the pylorus. OBJECTIVE: To standardize the measurements and evaluate the appearance of the normal and hypertrophied pylorus with high-frequency transducers. MATERIALS AND METHODS: We retrospectively analyzed abdominal ultrasounds of infants with suspected HPS from January 2019-December 2020. We classified the layers of the pylorus while assessing the stratified appearance. Two pediatric radiologists measured the muscle thickness of the pylorus independently by two methods for interrater agreement. Measurement (a) includes the muscularis propria and muscularis mucosa. Measurement (b) includes only the muscularis propria. We also evaluated the echogenicity of the muscularis propria. The interrater agreement, mean, range of the muscle thickness, and the diagnostic accuracy of the two sets of measurements were calculated. RESULTS: We included 300 infants (114 F:186 M), 59 with HPS and 241 normal cases. There was a strong agreement between the readers assessed in the first 100 cases, and ICC was 0.99 (95% CI, 0.98-0.99). Measurement (a), median thickness is 2.4 mm in normal cases and 4.8 mm in HPS. Measurement (b), median thickness is 1.4 mm in normal cases and 4.0 mm in HPS. Measurement (a) has an accuracy of 89.7% (95% CI, 85.7-92.8%) with 98.3% sensitivity and 87.6% specificity. Measurement (b) has an accuracy of 98.0% (95% CI, 95.7-99.3%) with 89.8% sensitivity and 100.0% specificity. The pylorus stratification is preserved in all normal cases and 31/59 (52.5%) cases of HPS. There was complete/partial loss of stratification in 28/59 (47.5%) cases of HPS. In all HPS cases, the muscularis propria was echogenic. CONCLUSION: Measuring the muscularis propria solely has a better diagnostic accuracy, decreasing the overlap of negative and positive cases. The loss of pyloric wall stratification and echogenic muscularis propria is only seen in HPS.

Characterization and quantification of fluid in the abdomen by ultrasound and magnetic resonance imaging in children with clinical suspicion of appendicitis.

PURPOSE: While regarded as a secondary sign of pediatric appendicitis, the frequency of physiologic intra-abdominal fluid in children with suspected but absent appendicitis is unknown. Ex vivo: to assess the validity of US/MRI measurements of free fluid. In vivo: in suspected pediatric appendicitis, to assess the amount of abdominal fluid by US and MRI, determine performance characteristics of US in fluid detection and identify fluid volume ranges in confirmed appendicitis. METHODS: Ex vivo: criterion validity of US and MRI for fluid volume measurements was tested using tissue-mimicking phantoms filled with different volumes of distilled water. In vivo: all participants from a previous prospective study of suspected appendicitis were evaluated by US; MRI was performed after equivocal USs. Qualitative and quantitative analyses of abdominal fluid and correlation of fluid presence with appendicitis were performed. RESULTS: Ex vivo: no difference was found between phantom-fluid amount and measured volume using the formula for volume of an ellipsoid for US (P=0.19) or MRI (P=0.08). In vivo: intra-abdominal fluid was present in 212/591 (35.9%) patients; 75/212 patients with fluid (35.4%) had appendicitis, 60 (28.3%) had alternate diagnoses, and 77 (36.3%) had physiologic fluid. Sensitivity and specificity of US for fluid detection were 84% (95% CI 71-93) and 65% (95% CI 52-77), respectively. In children with versus without appendicitis, the respective ranges of fluid volume were 0.7-1148.8 ml and 0.8-318 ml. CONCLUSION: The volume of an ellipsoid formula is a valid method for quantifying intra-abdominal fluid. The sole presence of intra-abdominal fluid on US does not support the diagnosis of pediatric appendicitis.

Analyzing Barriers and Enablers for the Acceptance of Artificial Intelligence Innovations into Radiology Practice: A Scoping Review.

OBJECTIVES: This scoping review was conducted to determine the barriers and enablers associated with the acceptance of artificial intelligence/machine learning (AI/ML)-enabled innovations into radiology practice from a physician's perspective. METHODS: A systematic search was performed using Ovid Medline and Embase. Keywords were used to generate refined queries with the inclusion of computer-aided diagnosis, artificial intelligence, and barriers and enablers. Three reviewers assessed the articles, with a fourth reviewer used for disagreements. The risk of bias was mitigated by including both quantitative and qualitative studies. RESULTS: An electronic search from January 2000 to 2023 identified 513 studies. Twelve articles were found to fulfill the inclusion criteria: qualitative studies (n = 4), survey studies (n = 7), and randomized controlled trials (RCT) (n = 1). Among the most common barriers to AI implementation into radiology practice were radiologists' lack of acceptance and trust in AI innovations; a lack of awareness, knowledge, and familiarity with the technology; and perceived threat to the professional autonomy of radiologists. The most important identified AI implementation enablers were high expectations of AI's potential added value; the potential to decrease errors in diagnosis; the potential to increase efficiency when reaching a diagnosis; and the potential to improve the quality of patient care. CONCLUSIONS: This scoping review found that few studies have been designed specifically to identify barriers and enablers to the acceptance of AI in radiology practice. The majority of studies have assessed the perception of AI replacing radiologists, rather than other barriers or enablers in the adoption of AI. To comprehensively evaluate the potential advantages and disadvantages of integrating AI innovations into radiology practice, gathering more robust research evidence on stakeholder perspectives and attitudes is essential.

Clinical and subclinical microemboli following neuroangiography in children.

BACKGROUND: To assess the frequency, imaging appearances, and risk factors of brain microemboli following pediatric neuroangiography, as assessed by early diffusion-weighted MRI imaging (DWI). METHODS: This single-center, retrospective analysis investigated early DWI post-pediatric neuroangiography. Patients aged 0-18 years who had diagnostic neuroangiography and DWI within a week postprocedure were included. Data on clinical and procedural parameters and MRI findings were recorded. Univariate and multivariate analyses were performed on the following risk factors: age, weight, vasculopathy, antiplatelet drug use, access type, intraprocedural heparin, procedure duration, neck arteries catheterized, and angiographic runs. A p-value<0.05 indicated statistical significance. RESULTS: Eighty-two children were included (40.2% female), mean age 10.1±4.5 years (range: 7 months-17 years). There were no intraprocedural thromboembolic complications recognized. DWI positivity was seen following 3.6% (3/82) procedures: two with transient symptoms, and one instance of silent microemboli. There were no territorial infarcts or clinical stroke. Children with underlying vasculopathy had a higher risk of microemboli from angiography than children without vasculopathy (OR 31.6, p=0.02), and the OR of microemboli following transradial angiography was 79.1 (p=0.005) as compared with transfemoral angiography. Univariate and multivariate analysis showed a significant association between microemboli and number of angiographic runs (p=0.004). Follow-up MRI in all three patients showed no residual abnormal signal. CONCLUSIONS: Cerebral microemboli are unusual following uncomplicated neuroangiography in children. However, in the presence of underlying vasculopathy and with transradial technique, the incidence approaches that reported in the adult literature. An increased association with the number of angiographic runs is an important and controllable factor.

Increased rate of significant findings on brain MRI during the early stage of the COVID-19 pandemic.

OBJECTIVES: To assess the effect of the COVID-19 pandemic on the proportion of abnormal paediatric neuroimaging findings as a surrogate marker for potential underutilisation. METHODS: Consecutive paediatric brain MRIs performed between March 27th and June 19th 2019 (Tbaseline) and March 23rd and June 1st 2020 (Tpandemic) were reviewed and classified according to presence or absence and type of imaging abnormality, and graded regarding severity on a 5-point Likert scale, where grade 4 was defined as abnormal finding requiring non-urgent intervention and grade 5 was defined as acute illness prompting urgent medical intervention. Non-parametric statistical testing was used to assess for significant differences between Tpandemic vs. Tbaseline. RESULTS: Fewer paediatric MRI brains were performed during Tpandemic compared to Tbaseline (12.2 vs 14.7 examinations/day). No significant difference was found between the two time periods regarding sex and age (Tbaseline: 557 females (44.63%), 7.95 ± 5.49 years, Tpandemic: 385 females (44.61%), 7.64 ± 6.11 years; p = 1 and p = .079, respectively). MRI brain examinations during Tpandemic had a higher likelihood of being abnormal, 41.25% vs. 25.32% (p<.0001). Vascular abnormalities were more frequent during Tpandemic (11.01% vs 8.01%, p = .02), congenital malformations were less common (8.34% vs 12.34%, p = .004). Severity of MRI brain examinations was significantly different when comparing group 4 and group 5 individually and combined between Tbaseline and Tpandemic (p = .0018, p < .0001, and p <.0001, respectively). CONCLUSIONS: The rate of abnormality and severity found on paediatric brain MRI was significantly higher during the early phase of the pandemic, likely due to underutilisation.

Association of new sonographic features with outcome in neonates with necrotizing enterocolitis.

BACKGROUND: We have recently noted some sonographic features in necrotizing enterocolitis that have received little or no attention in the current literature. These include thickening of the mesentery, hyperechogenicity of intraluminal intestinal contents, abnormalities of the abdominal wall, and poor definition of the intestinal wall. It has been our impression that the above four sonographic findings are generally seen in neonates with more severe necrotizing enterocolitis and may be useful in predicting outcome. OBJECTIVES: The aim of this study is, firstly, to review a large series of neonates, known to have clinical NEC, to document how frequently the above four sonographic features occur in neonates with necrotizing enterocolitis and, secondly, to determine whether they are predictive of outcome. MATERIALS AND METHODS: We retrospectively analyzed the clinical, radiographic, sonographic, and surgical findings in neonates with necrotizing enterocolitis between 2018 and 2021. The neonates were categorized into two groups based on outcome. Group A included neonates with a favorable outcome defined as successful medical treatment with no surgical intervention. Group B included neonates with an unfavorable outcome defined as failed medical treatment requiring surgery (for acute complications or late strictures) or death because of necrotizing enterocolitis. The sonographic examinations were reviewed with attention to the features of mesenteric thickening, hyperechogenicity of intraluminal intestinal contents, abnormalities of the abdominal wall, and poor definition of the intestinal wall. We then determined the association of these four findings with the two groups. RESULTS: We included 102 neonates with clinical necrotizing enterocolitis: 45 in group A and 57 in group B. Neonates in group B were born at a significantly earlier gestational age (median 25 weeks, range 22-38 weeks) and had a significantly lower birth weight (median 715.5 g, range 404-3120 g) than those in group A (median age 32 weeks, range 22-39 weeks, p = 0.003; median weight 1190 g, range 480-4500 g, p = 0.002). The four sonographic features were present in both study groups but with different frequency. More importantly, all four were statistically significantly more frequently present in neonates in group B compared to group A: (i) mesenteric thickening, A = 31 (69%), B = 52 (91%), p = 0.007; (ii) hyperechogenicity of intestinal contents, A = 16 (36%), B = 41 (72%), p = 0.0005; (iii) abnormalities of the abdominal wall, A = 11 (24%), B = 35 (61%), p = 0.0004; and (iv) poor definition of the intestinal wall, A = 7 (16%), B = 25 (44%), p = 0.005. Furthermore, the proportion of neonates with more than two signs was greater in group B compared to group A (Z test, p < 0.0001, 95% CI = 0.22-0.61). CONCLUSION: The four new sonographic features described were found to occur statistically significantly more frequently in those neonates with an unfavorable outcome (group B) than in those with a favorable outcome (group A). The presence or absence of these signs should be included in the sonographic report to convey the radiologists concern regarding the severity of the disease in every neonate, suspected or known to have necrotizing enterocolitis, as the findings may impact further medical or surgical management.

Imaging Evaluation for Thoracic Spine Fractures in Pediatric Trauma Patients: A Single-Center Experience at an Academic Children's Hospital.

BACKGROUND: Imaging workup for evaluating thoracic spine fracture (TSF) in pediatric blunt trauma is variable. PURPOSE: The aim of the study was to determine the number of TSFs missed by radiography and identified on computed tomography (CT) or magnetic resonance imaging (MRI) that required intervention or resulted in a change in management. METHODS: A retrospective review of children with TSFs was performed. Diagnostic images and reports for these patients were reviewed. Data regarding demographics, clinical presentation, management, and outcomes were extracted from institutional electronic medical records. Use of radiographs, CT, and MRI for evaluation of TSF was quantified. Incidence of TSFs was calculated and stratified by mechanism. The number of TSFs and complicating factors missed on radiography but identified on subsequent CT or MRI were quantified. RESULTS: Three thousand two hundred sixty-five trauma patients 18 years or younger were reviewed. Of these, 3.3% (90/3265) had TSFs (36 females, 54 males; mean age, 10.80 ± 4.4 years). The most common mechanism of injury was fall (43% [39/90]) followed by motor vehicle collisions (30% [27/90]). The most common fracture was simple compression fracture 64%, which occurred most frequently in the mid thoracic spine, followed by transverse process fractures 19% and spinous process fractures 7%. Almost half of all TSFs diagnosed on CT and/or MRI were missed on initial radiographs. While all fractures that required operative management were identified on radiographs, 13 of the 19 fractures that required nonoperative intervention were missed. CONCLUSIONS: Approximately 50% of TSFs diagnosed on CT or MRI were not identified on preceding radiographs. This is similar to studies in adult populations that show poor sensitivity of radiographs.

Myocardial late gadolinium enhancement using delayed 3D IR-FLASH in the pediatric population: feasibility and diagnostic performance compared to single-shot PSIR-bSSFP.

BACKGROUND: This study compares three-dimensional (3D) high-resolution (HR) late gadolinium enhancement (LGE; 3D HR-LGE) imaging using a respiratory navigated, electrocardiographically-gated inversion recovery gradient echo sequence with conventional LGE imaging using a single-shot phase-sensitive inversion recovery (PSIR) balanced steady-state free precession (bSSFP; PSIR-bSSFP) sequence for routine clinical use in the pediatric population. METHODS: Pediatric patients (0-18 years) who underwent clinical cardiovascular magnetic resonance (CMR) with both 3D HR-LGE and single-shot PSIR-bSSFP LGE between January 2018 and June 2020 were included. Image quality (0-4) and detection of LGE in the left ventricle (LV) (per 17 segments), in the right ventricle (RV) (per 3 segments), as endocardial fibroelastosis (EFE), at the hinge points, and at the papillary muscles was analyzed by two blinded readers for each sequence. Ratios of the mean signal intensity of LGE to normal myocardium (LGE:Myo) and to LV blood pool (LGE:Blood) were recorded. Data is presented as median (1st-3rd quartiles). Wilcoxon signed rank test and chi-square analyses were used as appropriate. Inter-rater agreement was analyzed using weighted κ-statistics. RESULTS: 102 patients were included with median age at CMR of 8 (1-13) years-old and 44% of exams performed under general anesthesia. LGE was detected in 55% of cases. 3D HR LGE compared to single-shot PSIR-bSSFP had longer scan time [4:30 (3:35-5:34) vs 1:11 (0:47-1:32) minutes, p < 0.001], higher image quality ratings [3 (3-4) vs 2 (2-3), p < 0.001], higher LGE:Myo [23.7 (16.9-31.2) vs 5.0 (2.9-9.0), p < 0.001], detected more segments of LGE in both the LV [4 (2-8) vs 3 (1-7), p = 0.045] and RV [1 (1-1) vs 1 (0-1), p < 0.001], and also detected more cases of LGE with 13/56 (23%) of patients with LGE only detectable by 3D HR LGE (p < 0.001). 3D HR LGE specifically detected a greater proportion of RV LGE (27/27 vs 17/27, p < 0.001), EFE (11/11 vs 5/11, p = 0.004), and papillary muscle LGE (14/15 vs 4/15, p < 0.001). Inter-rater agreement for the recorded variables ranged from 0.42 to 1.00. CONCLUSIONS: 3D HR LGE achieves greater image quality and detects more LGE than conventional single-shot PSIR-bSSFP LGE imaging, and should be considered an alternative to conventional LGE sequences for routine clinical use in the pediatric population.

Findings in percutaneous trans-hepatic cholecysto-cholangiography in neonates and infants presenting with conjugated hyperbilirubinemia: emphasis on differential diagnosis and cholangiographic patterns.

BACKGROUND: Biliary atresia (BA) is one of the causes of conjugated hyperbilirubinemia in infants which if untreated leads to end-stage liver disease and death. Percutaneous Trans-hepatic Cholecysto-Cholangiography (PTCC) is a minimally invasive study which can be utilized in the diagnostic work-up of these patients. This study's purpose is to describe the experience with PTCC in neonates, the imaging findings encountered, and the abnormal patterns which warrant further investigation. METHODS: A 16-year single-center retrospective study of patients with persistent neonatal cholestasis (suspected BA) undergoing PTCC. Patient demographics, laboratory values, PTCC images, pathology and surgical reports were reviewed. RESULTS: 73 patients underwent PTCC (68% male, mean age 8.7 weeks, mean weight 4.0 Kg). The majority of studies were normal (55%). Abnormal patterns were identified in 33 cases, 79% were diagnosed with BA and 12% with Alagille syndrome. Non-opacification of the common hepatic duct with a narrowed common bile duct (42%) and isolated small gallbladder (38%) were the most common patterns in BA. CONCLUSION: PTCC is a minimally invasive study in the diagnostic work-up of infants presenting with conjugated hyperbilirubinemia (suspected BA). Further invasive investigations or surgery can be avoided when results are normal.

Radiomic Features Based on MRI Predict Progression-Free Survival in Pediatric Diffuse Midline Glioma/Diffuse Intrinsic Pontine Glioma.

Purpose: Biopsy-based assessment of H3 K27 M status helps in predicting survival, but biopsy is usually limited to unusual presentations and clinical trials. We aimed to evaluate whether radiomics can serve as prognostic marker to stratify diffuse intrinsic pontine glioma (DIPG) subsets. Methods: In this retrospective study, diagnostic brain MRIs of children with DIPG were analyzed. Radiomic features were extracted from tumor segmentations and data were split into training/testing sets (80:20). A conditional survival forest model was applied to predict progression-free survival (PFS) using training data. The trained model was validated on the test data, and concordances were calculated for PFS. Experiments were repeated 100 times using randomized versions of the respective percentage of the training/test data. Results: A total of 89 patients were identified (48 females, 53.9%). Median age at time of diagnosis was 6.64 years (range: 1-16.9 years) and median PFS was 8 months (range: 1-84 months). Molecular data were available for 26 patients (29.2%) (1 wild type, 3 K27M-H3.1, 22 K27M-H3.3). Radiomic features of FLAIR and nonenhanced T1-weighted sequences were predictive of PFS. The best FLAIR radiomics model yielded a concordance of .87 [95% CI: .86-.88] at 4 months PFS. The best T1-weighted radiomics model yielded a concordance of .82 [95% CI: .8-.84] at 4 months PFS. The best combined FLAIR + T1-weighted radiomics model yielded a concordance of .74 [95% CI: .71-.77] at 3 months PFS. The predominant predictive radiomic feature matrix was gray-level size-zone. Conclusion: MRI-based radiomics may predict progression-free survival in pediatric diffuse midline glioma/diffuse intrinsic pontine glioma.

Data governance functions to support responsible data stewardship in pediatric radiology research studies using artificial intelligence.

The integration of human and machine intelligence promises to profoundly change the practice of medicine. The rapidly increasing adoption of artificial intelligence (AI) solutions highlights its potential to streamline physician work and optimize clinical decision-making, also in the field of pediatric radiology. Large imaging databases are necessary for training, validating and testing these algorithms. To better promote data accessibility in multi-institutional AI-enabled radiologic research, these databases centralize the large volumes of data required to effect accurate models and outcome predictions. However, such undertakings must consider the sensitivity of patient information and therefore utilize requisite data governance measures to safeguard data privacy and security, to recognize and mitigate the effects of bias and to promote ethical use. In this article we define data stewardship and data governance, review their key considerations and applicability to radiologic research in the pediatric context, and consider the associated best practices along with the ramifications of poorly executed data governance. We summarize several adaptable data governance frameworks and describe strategies for their implementation in the form of distributed and centralized approaches to data management.

Exploring the impact of anterior chest wall scars from implantable venous ports in adolescent survivors of cancer.

BACKGROUND: In children with cancer, port-a-caths (ports) are commonly placed in the right anterior chest wall, leaving a visible scar when removed. The psychological impact of port scars on survivors is unknown. It is unclear whether alternative sites should be considered. We assessed the impact of port scars on pediatric cancer survivors to determine whether a change in location is indicated. METHODS: We performed a cross-sectional single-center study of pediatric cancer survivors aged 13-18 years. A questionnaire explored participants' perceptions of their port scars. Four additional validated tools were used: Fitzpatrick scale, Patient and Observer Scar Assessment Scale (POSAS), Children's Dermatology Life Quality Index, and a Distress Thermometer. RESULTS: Among 100 participants (median age 15.8 years [13-18], median duration since treatment 8 years [1.5-14.8]), 75 'never/occasionally' thought about their port scars, 85 were not bothered by its location and 87 would not have preferred another site. Eleven participants were highly impacted by their scars: six thought about their scar 'everyday/all the time', four were highly bothered by its location, and nine would have preferred a different location. There was an association between the desire for different scar location and how much the location bothered participants (p < 0.0001), female sex (p = 0.03) and Patient POSAS score (p = 0.04). CONCLUSION: A port scar on the anterior chest wall was not a major concern for the majority of this cohort. A minority of participants were highly impacted by the scar and its location. Advance identification of those likely to be impacted by their scars may not be possible.

Rotational Vertebrobasilar Insufficiency: Is There a Physiological Spectrum? Phase-Contrast Magnetic Resonance Imaging Quantification in Healthy Volunteers.

BACKGROUND: Some cases of cerebral ischemia have been attributed to dynamic flow limitation in neck vessels. It however remains unknown whether this represents the extreme end of a physiological response. METHODS: Eighteen healthy volunteers were recruited to this prospective study. Cervical blood flow (ml/min/m2) was assessed using phase-contrast MRI, and cerebral perfusion ratios were assessed using arterial spin labeling perfusion at neutral position, predefined head rotations, as well as flexion and extension. Inter-reader agreements were assessed using intraclass correlation coefficient. RESULTS: The mean age was 38.6 ± 10.8 (range = 22-56) years, for five male participants and 13 females. The means for height and weight were 168 cm and 73.2 kg, respectively. There were no significant differences in individual arterial blood flow with change in head position (P > 0.05). Similarly, the repeated-measures analysis of variance test demonstrated no significant difference in perfusion ratios in relation to head position movement (P > 0.05). Inter-reader agreement was excellent (intraclass correlation coefficient = 0.97). CONCLUSIONS: There is neither significant change in either individual cervical arterial blood flow nor cerebral perfusion within the normal physiological/anatomical range of motion in healthy individuals. It is therefore reasonable to conclude that any such hemodynamic change identified in a patient with ischemic stroke be considered causative.

Neurovascular Manifestations in Pediatric Patients With Hereditary Haemorrhagic Telangiectasia.

BACKGROUND: Hereditary hemorrhagic telangiectasia (HHT) is a multiorgan vascular dysplasia with limited data regarding its neurovascular manifestations and genotype-phenotype correlation in children. The objective of this study was to describe the neurovascular findings in a large cohort of children with HHT and correlate between phenotype and genotype. METHODS: This retrospective study was conducted on 221 children (<18 years) with a definite or possible diagnosis of HHT based on Curacao criteria, or with positive genetics for the mutated genes of ENG, ACVRL-1, and SMAD-4, who also underwent brain MRI and/or conventional angiography. Demographic and clinical information, imaging findings, and follow up information were gathered. RESULTS: Two hundred twenty-one children with HHT (70.6% genetically confirmed, and 99.5% positive family history) were included, with a median age of 7 years (interquartile range: 3 to 11 years) and 58.8% male predominance. Neurovascular lesions were found in 64 of 221 (28.9%), with 3.1% prevalence of intracranial hemorrhage. The most commonly observed vascular malformations were developmental venous anomalies (48.5%) and brain arteriovenous malformations (AVMs) (31.2%), followed by capillary malformations (14.1%). Multiple AVMs were seen in 10.0% of the cohort. We found no instances of de novo AVM (1281.8 patient-years).A significantly higher proportion of patients with ENG mutations (19.7%) had brain AVM than those with ACVRL-1 (4.9%) and SMAD-4 (0%) mutations (P < 0.01). There was no significant difference in the hemorrhagic risk of shunting lesions associated with ENG (35.3%) or ACVRL-1 (33.3%) positivity (P = 0.9). CONCLUSIONS: We describe the neurovascular imaging and genetic findings from a large pediatric cohort of HHT, to enhance clinical awareness and guide management of patients with HHT.

Semiquantitative characterization of dynamic magnetic resonance perfusion of the liver in pediatric Fontan patients.

BACKGROUND: Dynamic contrast-enhanced magnetic resonance imaging (MRI) of the liver in pediatric Fontan patients often shows peripheral reticular areas of hypoenhancement, which has not been studied in detail. OBJECTIVE: To semiquantitatively score the hepatic MR perfusion abnormality seen in pediatric Fontan patients, and to correlate the perfusion abnormality with functional clinical and hemodynamic parameters. MATERIALS AND METHODS: All children (< 18 years old) after Fontan palliation with combined clinical cardiac and liver MRI performed between May 2017 and April 2019 were considered for inclusion. A semiquantitative perfusion score was used to assess the severity of the hepatic reticular pattern seen on dynamic contrast-enhanced liver imaging. The liver was divided into four sections: right posterior, right anterior, left medial and left lateral. Each liver section was assigned a score from 0 to 4 depending on the amount of abnormal reticular hypoenhancement. Scoring was assigned for each section of the liver across eight successive dynamic contrast-enhanced modified spoiled gradient echo runs. Scores were correlated with clinical and hemodynamic parameters. RESULTS: All Fontan children showed hepatic reticular hypoenhancement by MRI, most severe in the early portal venous phase with a median maximum total perfusion abnormality score of 12 (range: 9-14). All perfusion abnormalities progressively resolved during the hepatic venous phase. Perfusion abnormality scores were greatest in the right compared to left hepatic lobes (7 range: [6-8] vs. 5 [range: 3-6], P < 0.01). The maximum left hepatic lobe perfusion abnormality scores were greatest in children with versus without imaging signs of portal hypertension (8 [range: 7-8] vs. 4 [range: 3-5], P < 0.01). High unconjugated bilirubin and low platelets correlated with greater perfusion abnormality (R = 0.450, P = 0.024, and R = - 0.458, P < 0.01, respectively). Age at MRI, time from Fontan, focal liver lesions and cardiac MRI hemodynamic parameters did not show significant correlations with the severity of the liver perfusion abnormality. CONCLUSION: All Fontan children have hepatic reticular hypoenhancement abnormalities seen with MRI that are most severe in the right hepatic lobe and universally show gradual resolution through the hepatic venous phase. Perfusion abnormality in the left hepatic lobe is worse in children with portal hypertension.

Clinical and Angioarchitectural Features of Hemorrhagic Brain Arterio-Venous Malformations in Adults and Children: Contrasts and Implications on Outcome.

BACKGROUND: Hemorrhage from brain arteriovenous malformations (bAVMs) is estimated at 3% per annum. Features influencing risk of hemorrhage include perforator/posterior circulation supply, associated aneurysms, and deep drainage. Children are more likely to present with bAVM bleeds. OBJECTIVE: To analyze differences in bAVM angioarchitecture between children and adults and describe predictors of poor outcome. METHODS: Data were collected from adult and pediatric tertiary referral hospitals. Demographic data, bleed location, treatment, and follow-up modified Rankin Scale (mRS) were collected. Angioarchitectural assessment included aneurysm presence, nidus morphology, perinidal angiogenesis, intranidal shunting, steal phenomenon, venous ectasia, venous stenosis, venous reflux, and pseudophlebitic pattern. Regression analyses conducted to determine predictors of mRS > 2. RESULTS: A total of 270 adult and 135 pediatric ruptured bAVMs were assessed. Median age was 42 (adults) and 10.9 (children) yr. Intranidal aneurysms were more frequent in children (P = .012), whereas prenidal aneurysms were more common in adults (P < .01). Children demonstrated more perinidal angiogenesis (P = .04), whereas steal phenomenon was commoner in adults (P < .01). Venous ectasia (P < .01), reflux (P < .01), and pseudophlebitic pattern (P = .012) were more frequent in adults. Children had better outcome (mRS score ≤ 2) (P < .01). Older age (odds ratio [OR] = 1.02), eloquent location (OR = 2.5), multicompartmental hemorrhage (OR = 1.98), venous reflux (OR = 2.5), diffuse nidus (OR = 1.83), pseudophlebitic pattern (OR = 1.96), intranidal shunts (OR = 2), and no treatment (OR = 3.68) were significant predictors of mRS > 2. CONCLUSION: Children are more likely to have intranidal aneurysms and perinidal angiogenesis, whereas adults have more prenidal aneurysms, venous ectasia, corticovenous reflux, and pseudophlebitic pattern. Eloquent location, diffuse nidus, intranidal shunts, venous reflux, and pseudophlebitic pattern predict poorer outcome.

Usefulness of TI-scout images in the assessment of late gadolinium enhancement in children.

BACKGROUND: Cardiovascular magnetic resonance (CMR) late gadolinium enhancement (LGE) requires identification of the normal myocardial nulling time using inversion time (TI)-scout imaging sequence. Although TI-scout images are not primarily used for myocardial assessment, they provide information regarding different signal recovery patterns of normal and abnormal myocardium facilitating identification of LGE in instances where standard LGE images alone are not diagnostic. We aimed to assess the diagnostic performance of TI-scout as compared to that of standard LGE images. METHODS: CMR studies with LGE imaging in 519 patients (345 males, 1-17 years) were reviewed to assess the diagnostic performance of LGE imaging in terms of the location of LGE and the pathologic entities. The diagnostic performance of the TI-scout and standard LGE imaging was classified into four categories: (1) equally diagnostic, (2) TI-scout superior to standard LGE, (3) standard LGE superior to TI-scout, and (4) complementary, by the consensus of the two observers. RESULTS: The study cohort consisted of 440 patients with negative LGE and 79 with evidence for LGE. For a negative diagnosis of LGE, TI-scout and standard LGE images were equally diagnostic in 75% of the cases and were complementary in 12%. For patients with LGE, TI-scout images were superior to standard LGE images in 52% of the cases and were complementary in 19%. The diagnostic performance of TI-scout images was superior to that of standard LGE images in all locations. TI-scout images were superior to standard LGE images in 11 of 12 (92%) cases with LGE involving the papillary muscles, in 7 /12 (58%) cases with subendocardial LGE, and in 4/7 (57%) cases with transmural LGE. TI-scout images were particularly useful assessing the presence and extent of LGE in hypertrophic cardiomyopathy (HCM). TI-scout was superior to standard LGE in 6/10 (60%) and was complementary in 3/10 (30%) of the positive cases with HCM. CONCLUSIONS: TI-scout images enhance the diagnostic performance of LGE imaging in children.

Use of radial access sheaths for transfemoral neuroendovascular procedures in children.

PURPOSE: Pediatric neuroendovascular procedures require special considerations. Given small vessel sizes, risk for arterial injury must be weighed against use of larger devices, with the diameter of the sheath being a known association with arterial complications. We recently transitioned to using thin-walled radial sheaths for transfemoral angiography in children, given their lower profile. Here, we report on these sheaths' technical success and complications, comparing against a historical cohort where regular vascular sheaths were employed. METHODS: We retrospectively recorded patient and procedural data from 168 consecutive procedures from September 2017 to January 2019 when radial-specific sheaths were exclusively used at our tertiary pediatric hospital. These results were compared to data from September 2015 to January 2017, when regular vascular sheaths were exclusively used in 152 consecutive procedures. Statistical analysis was performed using unpaired t test or chi-square test, with p < 0.05 considered statistically significant. RESULTS: Patient characteristics (age, sex, weight) were not statistically different between the case and control group. No significant differences were found in the procedural data with the exception of heparin use which was higher in the radial-sheath cohort. There was a decrease in the rate of complications in the case group (1.2%) versus control group (2.6%), though not statistically different. CONCLUSION: Radial-specific sheaths have numerous beneficial characteristics that make them well-suited to the pediatric population. In our study, we show that radial sheaths are equally effective and safe perioperatively. Follow-up research may show if radial sheaths used transfemorally decrease long-term complications such as limb-length discrepancy and mortality.