AJNR. American journal of neuroradiology最新文献

Background and purpose: Small intracerebral hemorrhage, defined as baseline noncontrast CT (NCCT) hematoma volume <30 mL, is often considered lower risk for early expansion. We tested whether the edema-to-hematoma ratio (EHR) on baseline noncontrast CT (NCCT) independently predicts expansion and whether combining EHR with hematoma volume (HV) enables practical bedside risk stratification.

Materials and methods: We retrospectively analyzed 219 conservatively managed patients with small-volume intracerebral hemorrhage. Baseline NCCT underwent AI-assisted 3D volumetry to quantify HV and perihematomal edema and to compute EHR. The primary outcome was binary early hematoma expansion on the first follow-up CT. Multivariable logistic regression included HV, EHR, age, deep location, intraventricular hemorrhage, and anticoagulation status. Discrimination and calibration were assessed with five-fold cross-validation. Youden-index cut points were used to derive a simple bedside rule combining HV and EHR.

Results: Early expansion occurred in 20.1% (44/219). Lower EHR and larger HV independently predicted expansion. Per 0.10 increase in EHR, adjusted odds decreased by 23% (aOR, 0.77; 95% CI, 0.66-0.88); per additional 5 mL in HV, odds increased by 37% (aOR, 1.37; 95% CI, 1.09-1.73). A bedside rule (HV ≥9.2 mL plus EHR ≤0.683) identified a high-risk phenotype with a 39.3% expansion rate versus 4.4% in the low-risk reference. The multivariable model outperformed either predictor alone (AUC, 0.748 vs 0.700 and 0.672) with acceptable calibration (intercept, -0.17; slope, 0.92; Brier score, 0.141).

Conclusions: A lower baseline EHR independently predicts early expansion in small-volume intracerebral hemorrhage. Pairing EHR with HV yields a practical bedside rule with a large risk gradient and improves discrimination beyond volume alone, supporting targeted monitoring and timely repeat imaging.

Background and purpose: Medical errors and complications (MEAC) in neurointervention have significant implications for patient safety and healthcare costs, with 50% of patient harm due to medical errors being preventable. Despite this, literature on MEAC in the outpatient setting is scarce.

Methods: We conducted a systematic review of PubMed/Medline in adherence to PRISMA guidelines and identified studies on neurointerventional complications published from 2004 to 2024. Eligible studies were original research articles that reported on medical errors or complications, stratified by care phase: pre-hospital, in-hospital non-procedural, procedural, and post-discharge. Data extraction included demographic characteristics, time points, types of errors and complications, follow-up methods, and reporting practices. Descriptive statistics and unadjusted logistic regression were used to analyze the data.

Results: From 174 studies involving 60,812 patients, we observed a total complication rate of 17.2%. Most studies reported on in-hospital procedural complications in their data (83.3%), while only 1.7% and 31.6% addressed pre-hospital and post-discharge complications, respectively. Systematic patient follow-up significantly increased the likelihood of identifying post-discharge complications (OR 2.16, 95% CI 1.06-4.40). Severe adverse events dominated post-discharge reports, highlighting a possible gap in documenting minor complications and medical errors, particularly in outpatient settings.

Conclusion: Neurointerventional literature primarily focuses on in-hospital, procedure-related events, while only few studies on pre-hospital and post-discharge errors and complications are available. Standardized tools, such as patient-reported outcome measures, and rigorous follow-up protocols are critical for comprehensive complication reporting. Establishing quality standards for complication documentation would facilitate improved patient care and enable systematic comparisons across studies.

Head and neck cancer is one of the most common malignancies in the United States. In the last decades, advances in treatment have greatly improved survival. However, as patients are living longer, late effects of treatment have been increasingly recognized. Carotid artery stenosis secondary to prior neck irradiation predisposes survivors to an increased risk of cerebrovascular events, cranial nerve deficits, and cardiovascular morbidity. Understanding the pathophysiology, imaging findings, and clinical course of radiation-induced steno-occlusive disease of the cervical carotid artery is crucial for radiologists to understand and accurately interpret imaging examinations of such patients. In addition, this review outlines the epidemiologic risk factors, clinical and radiologic features, and recommended treatment modalities for patients with prior irradiation of the head and neck. By consolidating the current evidence, we aim to underscore the clinical significance of this complication and highlight the importance of long-term surveillance and tailored management of RICS among head and neck cancer survivors.

Background and purpose: Chronic hypoperfusion is considered a central mechanism in cerebral small vessel disease (CSVD). However, the results of studies on CBF vary across the imaging method, with phase-contrast MRI (PC-MRI) reflecting large vessel blood flow and arterial spin labeling (ASL) capturing tissue perfusion. We aimed to investigate the relationship between CSVD and CBF as measured by these two methods.

Materials and methods: 94 subjects from community and neurology clinics were enrolled. Global CBF was measured using 4D flow by summing the blood flow in the bilateral internal carotid arteries and the basilar artery, then normalizing by brain volume and density (ntCBF_PC). Mean CBF was measured in the whole brain using multi-delay ASL (wbCBF_ASL). CSVD imaging markers, including white matter hyperintensities (WMH), lacunes, perivascular spaces (PVS), and microbleeds, were assessed in each subject, and a CSVD summary score was calculated to reflect the overall CSVD burden. Free water (FW) and tissue fractional anisotropy (tFA) were used to evaluate the microstructural integrity of white matter. Cognitive performance was assessed using neuropsychological tests. Associations between the two CBF measures, all CSVD imaging markers, and cognition were assessed while controlling for potential confounding variables.

Results: Lower ntCBF_PC (OR, 0.933; 95% CI: 0.884, 0.984; p = 0.01) and wbCBF_ASL (OR, 0.914; 95% CI: 0.842, 0.992; p = 0.03) were associated with higher BG-PVS scores. Lower ntCBF_PC (OR, 0.798; 95% CI: 0.683, 0.932; p = 0.004) and wbCBF_ASL (OR, 0.767; 95% CI: 0.636, 0.926; p = 0.006) were associated with the presence of lacune. Only wbCBF_ASL correlated with white matter injury: higher PWMH Fazekas score (OR, 0.868; 95% CI: 0.791, 0.951; p = 0.003), larger WMH volume (β, -0.252, p = 0.02), higher FW (β, -0.314, p = 0.001), and lower tFA (β, 0.254, p = 0.02). No significant associations existed between ntCBF_PC and WMH or diffusion imaging markers. Only ntCBF_PC was positively associated with MMSE (β, 0.210, p = 0.04).

Conclusions: CBF measured by ASL demonstrates stronger CSVD associations than those derived from PC-MRI, suggesting that microcirculation impairment, rather than global blood flow, contributes more significantly to CSVD.

Background and purpose: Cerebrovascular reactivity (CVR) is commonly used to estimate hemodynamic impairment. Conventional use is best suited to unilateral vascular disease, such that CVR can be normalized to reference values from the contralateral hemisphere or to posterior circulation territories; however, major confounds have been identified that leave implementation difficult in more common cases of bilateral disease, even despite common cerebellar normalization. Recently, we reported data-driven identification of candidate healthy voxel signatures learned from contemporaneous imaging data. Here, we introduce an entirely inline, automated approach exploiting the dynamics of resting-state BOLD (rs-BOLD) signal from the BOLD baseline, hypothesizing prediction to within ten percent error relative to ground truth healthy-voxel CVR values.

Materials and methods: 22 subjects with strictly unilateral intracranial steno-occlusive disease (SOD) underwent 28 CVR studies under pharmacologic provocation using acetazolamide with BOLD-MRI (ACZ-BOLD). Separate affected and unaffected hemispheric masks were segmented to train machine learning models to learn signatures of the unaffected hemisphere using the rs-BOLD baseline, as well as anatomic and vascular parameters. Twenty additional healthy subjects from the Human Connectome Project supplemented training, wherein all voxels were classified normals. 32 distinct time-delays were computed voxelwise, with 32 maximum correlation values constrained to each of 32 paired arterial territories. Performance in prediction of ground-truth reference CVR was computed and compared.

Results: The ensembled model achieved AUC of 0.81 in predicting candidate unaffected voxels, demonstrating strong performance in estimation of normal-hemisphere CVR (median absolute percent error [95%CIs] 7.28[3.48-10.34] and 5.61[2.90-9.86] to predict median and mean reference CVR), exhibiting significant improvements over naïve whole-brain voxel selection (P=0.005 and P=0.048, respectively) or conventional cerebellar normalization (26.4[10.1-40.3] median and 27.6[23.7-33.2] mean). In nine bilateral cases assessed to illustrate potential use, the proportion of candidate voxels and corresponding volumes predicted by the ensembled model was significantly lower than in most healthy hemispheres, but yielded subjectively improved delineation of putatively abnormal regions.

Conclusions: We demonstrate feasibility of learning unaffected reference voxel CVR signatures for BOLD-CVR MRI. The approach facilitates extension of brain CVR beyond existing constraints in subjects with bilateral disease.

Background: There is no consensus regarding the optimal imaging modality for assessing aneurysms treated with flow-diverters (FD).

Objectives: To evaluate the image quality of intravenous VasoCT (IV-VasoCT) in the follow-up of aneurysms treated with FD and to explore its potential for depicting braid stent deformation and vessel lumen modifications.

Material and methods: We retrospectively enrolled patients who underwent FD placement for the treatment of aneurysms, with post-treatment IV-VasoCT at 1-3 months and DSA at least 6 months. IV-VasoCT image quality assessment evaluated: vessel opacification, vessel border, stent radiopacity, stent/vessel contrast, movement artefacts. Braid stent and parent vessel findings were compared between IV-VasoCT and DSA in terms of: neointimal lining, stent foreshortening, fish-mouthing, bumping, braid collapse.

Results: 26 aneurysms were evaluated in 24 patients (F/M: 15/9), with a median age of 54.5 years (IQR: 47.0-59.0 years). The in-consensus image quality reading evaluated: vessel opacification as optimal in 19 cases (73.1%), vessel borders as well-defined in 17 (65.4%); stent radiopacity as well-defined in 25 (96.2%), stent/vessel contrast as optimal in 20 (76.9%). Compared to DSA, IV-VasoCT diagnosed 2/3 (66%) cases of neointimal lining, 2/2 (100%) of fish-mouth, 1/2 (50%) of foreshortening, 3/6 (50%) of bumping, 2/2 (100%) of braid collapse.

Conclusion: IV-VasoCT is feasible for the non-invasive follow-up of aneurysm treated with FD. It may depict modifications in braid stent and vessel neointimal lining, but the presence of coils may reduce its image quality because of artefacts.

Background and purpose: Spinal CSF-venous fistulas are an important cause of spontaneous intracranial hypotension (SIH). Most of these fistulas are associated with a meningeal diverticulum, but they arise from the largest diverticulum in only about one-third of cases. De novo recurrent CSF-venous fistulas are defined as a CSF-venous fistula at a different spinal level (or side) following treatment of the initial CSF-venous fistula. Such de novo recurrent fistulas offer a unique opportunity to study the growth of diverticula prior to the development of a fistula.

Methods: In this single-center retrospective observational study, data elements indicating a de novo recurrent fistula were extracted from a prospectively maintained data base. Using this registry, we identified a consecutive group of patients with de novo recurrent fistulas.

Results: The study population consisted of nine women and six men (mean age: 57.9 years) with SIH and a de novo recurrent fistula. All patients had multiple diverticula, ranging from three to 35 diverticula per patient (total: 215 diverticula). The initial fistula was treated with microsurgical clip ligation in all patients. The mean interval between initial fistula treatment and de novo recurrent fistula formation was two years and four months (range, five months to six years and six months). A change in diverticular size (mean, 2 mm; range: 1 to 3 mm) associated with the de novo recurrent fistula was observed in five (33.3%) of the 15 patients. Diverticular size had increased in three patients and decreased in two patients. No change in size was observed in the other 195 meningeal diverticula.

Conclusions: We found that in one-third of patients a change in diverticular size was observed corresponding to the site of the de novo recurrent fistula. An increase in diverticular size could be explained by increased diverticular wall tension (law of LaPlace) and attenuation of the diverticular wall resulting in a higher likelihood of developing a CSF-venous fistula and a decrease in size could be explained by decompression of the diverticulum through the fistula. A change in size of meningeal diverticula may help guide the invasive myelographic studies necessary to locate CSF-venous fistulas.

Bevacizumab, a monoclonal antibody targeting vascular endothelial growth factor A, is emerging as a promising therapy for high-grade meningioma, an aggressive central nervous system tumor with limited systemic treatment options. While clinical stabilization has been observed in prior trials, radiographic response on conventional MRI is often minimal, highlighting the need for alternative imaging biomarkers to monitor treatment effect. [68Ga]-DOTATATE is a somatostatin analog that binds with high affinity to somatostatin receptor 2, which is overexpressed in meningiomas. In this case series, we present three patients with WHO grade 3 meningioma who underwent baseline and follow-up [68Ga]-DOTATATE PET imaging in the setting of bevacizumab therapy. All patients had biopsy-confirmed high-grade disease and demonstrated multifocal [68Ga]-DOTATATE-avid lesions at baseline. Across 12 lesions, 67% showed a reduction in SUV_max following therapy. In contrast, matched control patients who underwent serial [68Ga]-DOTATATE PET imaging without bevacizumab or other systemic therapy demonstrated no significant change in SUV_max Linear mixed-effects modeling confirmed a significant treatment effect, with a 7.2-unit greater reduction in SUV_max in the bevacizumab-treated group than in controls (p = 0.01). Our findings raise the possibility that [68Ga]-DOTATATE PET may capture early biologic responses to antiangiogenic therapy not reflected by anatomic imaging alone, potentially reflecting changes in somatostatin receptor expression, tumor perfusion, or metabolic activity. This series suggests that [68Ga]-DOTATATE PET may serve as a noninvasive, biologically informative adjunct to conventional imaging and supports further investigation of its role as a molecular imaging biomarker of bevacizumab response in high-grade meningioma.

Gadopiclenol (Vueway) has a substantially higher T1 relaxivity than other gadolinium-based contrast agents, including gadobenate dimeglumine (MultiHance), allowing a 50% dose reduction with noninferiority in detection of enhancing lesions on T1-weighted imaging. However, such reduced dosing may have unintended consequences for DSC perfusion imaging. In this retrospective analysis of 30 patients imaged using package-insert dosing of both MultiHance and Vueway (on separate occasions), DSC signal transients were compared between these agents. There was a statistically significant reduction in % signal change and area under the contrast concentration-time curve for Vueway relative to MultiHance. When the signal was normalized for contrast dose, this difference was statistically eliminated. The results demonstrate that package-insert dosing of Vueway produces substantially weaker signal transients for DSC perfusion imaging, in proportion to the one-half gadolinium dose regimen. Potential impact on perfusion metrics, with concerns including SNR and leakage correction, warrants further study.

Background: The adult hippocampal head (HH) is characterized by sulci and digitations visible both histologically and on MRI. The aim of the study was to describe the emergence and progression of HH digitations in the fetal brain across different gestational ages (GA) using MRI.

Materials and methods: A retrospective assessment was performed on 383 fetal brain MRIs (19-39 weeks GA) acquired on 1.5T (n=351) and 3T (n=32) scanners. Imaging included ssFSE T2, T1-weighted sequences, and axial DWI. The fetal HH was identified on coronal T2 images and HH morphological variants were classified using an established adult classification system as: Class 0 (no sulci, one digitation), Class 1 (one sulcus, two digitations), and Class 2 (two sulci, three digitations). MRIs with brain anomalies identified prenatally or postnatally, abnormal sulcation, or severe artifacts were excluded. Variant frequencies in both hemispheres were grouped by GA. Correspondence Analysis and Cumulative Link Mixed Models were used to assess the association between GA and class distribution.

Results: 271 fetuses were included, grouped into eight GA categories. Overall frequencies were 33.6% for Class 0, 48.9% for Class 1, and 17.5% for Class 2. With increasing GA, class 0 frequency decreased, while Classes 1 and 2 increased. Between 23- and 25-weeks GA, detection of more digitated HH variants rose sharply from 1% to 64%. Follow-up in 30 fetuses showed an increase in HH digitations, with no observed reductions. Regression analysis revealed greater digitation complexity in the right hemisphere.

Conclusion: Digitations in the fetal HH begin to appear around 24 weeks GA, reaching frequencies similar to those observed in adults by the later stages of gestation. The right hemisphere shows higher complexity, possibly reflecting faster growth. These findings contribute to the understanding of hippocampal development and may serve as a biomarker of fetal brain maturation.