RESUMO
BACKGROUND: The Rotterdam scale is one of the most commonly used radiological scales for evaluating and predicting outcomes in traumatic brain injury (TBI) cases. Given the evolving nature of TBI, our study is designed to compare the Rotterdam score of computed tomography (CT) findings upon admission (Rotterdam score I) with the score after 72 hours (Rotterdam score II) of treatment in the trauma intensive care unit (ICU). METHODS: A retrospective observational study was conducted on 54 patients who received intensive care treatment for isolated severe TBI over five years. We included severe TBI patients with no age restrictions who required admission to the ICU within 12 hours of the onset of trauma. An initial Rotterdam CT score was obtained via a CT head scan within four hours of the trauma, followed by a control CT head scan 72 hours after ICU admission. It was essential to have documentation on the clinical and laboratory treatment course and access to radiological CT diagnostics. Receiver operating characteristic (ROC) curves were employed in this study to evaluate the accuracy of diagnostic tests, such as the Rotterdam score. The ROC curves provided a graphical representation of the tests' diagnostic performance, which helped assess their effectiveness. RESULTS: There was a significant difference (p < 0.001) in the diagnostic scores of CT scans upon admission (Rotterdam score I) and control CT scans after 72 hours (Rotterdam score II) in the total sample. The Rotterdam score I was notably higher, 3.6 (±0.8), in patients requiring neurosurgical intervention compared to those who did not, 2.8 (±0.9), with significance (p = 0.003). The Rotterdam score I demonstrated a substantial predictive value for unfavorable outcomes (p = 0.048), as did the Rotterdam score II after the 72-hour mark (p = 0.006). CONCLUSION: The control Rotterdam score 72 hours after admission predicts mortality in isolated TBI patients more significantly than the Rotterdam score determined at the patient's admission to the intensive care unit.
RESUMO
Background: Differential diagnosis in radiology relies on the accurate identification of imaging patterns. The use of large language models (LLMs) in radiology holds promise, with many potential applications that may enhance the efficiency of radiologists' workflow. The study aimed to evaluate the efficacy of generative pre-trained transformer (GPT)-4, a LLM, in providing differential diagnoses in neuroradiology, comparing its performance with board-certified neuroradiologists. Methods: Sixty neuroradiology reports with variable diagnoses were inserted into GPT-4, which was tasked with generating a top-3 differential diagnosis for each case. The results were compared to the true diagnoses and to the differential diagnoses provided by three blinded neuroradiologists. Diagnostic accuracy and agreement between readers were assessed. Results: Of the 60 patients (mean age 47.8 years, 65% female), GPT-4 correctly included the diagnoses in its differentials in 61.7% (37/60) of cases, while the neuroradiologists' accuracy ranged from 63.3% (38/60) to 73.3% (44/60). Agreement between GPT-4 and the neuroradiologists, and among the neuroradiologists was fair to moderate [Cohen's kappa (kw) 0.34-0.44 and kw 0.39-0.54, respectively]. Conclusions: GPT-4 shows potential as a support tool for differential diagnosis in neuroradiology, though it was outperformed by human experts. Radiologists should remain mindful to the limitations of LLMs, while harboring their potential to enhance educational and clinical work.
RESUMO
First described by Virchow in the 19th century, intracranial arteriovenous malformations (AVMs) are complex, tangle-shaped vascular lesions with a number of associated neuroparenchymal, hemodynamic, and angio-architectural changes. However, the clinical description of extracranial AVMs dates back to the Ebers Papyrus (c. 1500 BC), with a still unknown definitive underlying etiology thus far. AVMs are rare lesions, with approximately 0.15% incidence and 0.001-0.5% prevalence, but of high importance as they tend to affect young patients who are frequently otherwise healthy. In the majority of cases, AVMs present as sudden intracranial hemorrhages that require immediate intervention, but incidentally-detected unruptured AVMs are only found in â¼15% of cases, leaving a confusing dilemma regarding the appropriate next step, particularly given the several therapeutic interventions available and clinical trials that were vulnerable to follow-up criticism. Herein, we present a case of an incidentally detected asymptomatic AVM in a 15-year-old boy via advanced imaging techniques that was initially misinterpreted as a post-traumatic subarachnoid hemorrhage on routine imaging studies. In providing a comprehensive overview of pathological classification schemes and the currently available diagnostic options for these silent dilemmatic AVMs, we highlight three management techniques: microsurgical resection, endovascular embolization, and stereotactic surgery, with the best option depends mostly on addressing lesion resection properly with minimal associated mortality and morbidity.
RESUMO
With increasing use of broad-spectrum antibiotics, advances in organ and stem-cell transplant therapy, and the continued diabetes mellitus II epidemic, as well as other risk factors, reports of fungal infections of the CNS have been increasing. The most lethal subset is the angioinvasive fungal infection. Aspergillus fumigatus, Mucor, and Fusarium tend to affect immunocompromised individuals depending on their risk factors. Exserohilum rostratum and Cladophialaphora species tend to infect immunocompetent individuals. Early diagnosis and treatment are imperative for improved outcomes and reduced morbidity and mortality. Clinical presentation is often nonspecific, while neuroimaging can be helpful for accurate diagnosis. CT of the head and/or the maxillofacial structures is the primary imaging modality. Once the infection begins to proliferate, areas of vasogenic and cytotoxic edema, with regional mass effect and shift of the midline structures may be seen. These findings, however, are often nonspecific and may also be seen in underlying neoplasm, inflammatory processes, and other intracranial infections. Characteristic findings on T1, T2, diffusion-weighted imaging (DWI), and gradient echo sequences (GRE) may help to further narrow the differential diagnoses. We present a review of neuroimaging findings that will aid the neuroradiologist in distinguishing intracranial angioinvasive fungal infections and lead to improved patient outcomes.
Assuntos
Infecções Fúngicas do Sistema Nervoso Central , Neuroimagem , Humanos , Neuroimagem/métodos , Infecções Fúngicas do Sistema Nervoso Central/diagnóstico por imagem , Tomografia Computadorizada por Raios X/métodos , Infecções Fúngicas Invasivas/diagnóstico por imagemRESUMO
BACKGROUND: Health systems are striving to improve delivery of mechanical thrombectomy (MT) for ischaemic stroke. With the move to 24/7 provision, we aimed to assess (1) the change in referral and procedural frequency and timing, (2) reasons referrals did not proceed to MT, and (3) nocturnal procedural efficacy and safety. METHODS: This was an observational study comparing 12-month data for an extended daytime service (2021/2022, hours, 0800-2000) to that for a 12-month period delivering 24/7 cover (2023-2024). Nocturnal and daytime outcomes (rate of recanalisation using modified TICI scoring), extent of postprocedural infarction (using ASPECTS grading), rate of early neurological improvement (using 24-h NIHSS change), 90-day mortality, and complicating symptomatic intracranial haemorrhage (SICH) in the latter period were compared. RESULTS: Both referrals (432 to 851) and procedural caseload (191 to 403) approximately doubled with the move to 24/7 cover; 36% of procedures occurred overnight (n = 145). The dominant reasons for referrals not proceeding to MT were a large core infarct (n = 144) or absence of a large vessel occlusion on baseline imaging (n = 140). There were no significant differences in successful recanalisation (TICI 2B/3: 85.5% vs 87.1%, P = .233), rates of postprocedural ASPECTS≥7 (74.9% vs 75.8%, P = .987), early neurological improvement (NIHSS reduction ≥30%: 43.4% vs 42.4%, P = .917), 90-day mortality (19.6% vs 18.6%, P = .896), or SICH (1.9% vs 4.1%, P = .214) obtained for daytime vs nighttime hours. CONCLUSION: 24/7 MT provision has resulted in a rapid rise in the number of patients who may benefit from MT. This service can be provided with an acceptable safety profile during nighttime hours in a high-volume comprehensive UK centre.
RESUMO
Wilson's disease (WD) is a genetic disorder of copper metabolism that is inherited as an autosomal recessive (AR) due to mutations in the ATP7B gene, which is involved in intracellular copper transport. Approximately 40% to 50% of the patients present with neurological symptoms as their first symptom. The most common neurological symptoms are dysarthria, gait abnormalities, ataxia, dystonia, tremor, parkinsonism, and drooling. This case report aims to present a diagnostically challenging case of WD presenting with neurological symptoms. The 38-year-old male patient was admitted with complaints of imbalance, gait disturbance, weakness in the legs, speech impairment, tremors in the hands, syncope, and drooling. The MRI primarily revealed FLAIR, T1, and T hyperintensities in the bilateral globus pallidus of the basal ganglias. At first, the patient was evaluated according to the Leipzig scoring and received one point from the serum ceruloplasmin level and two points from the neurological symptoms and was evaluated as "possible WD" with a total of three points. 24-hour urine copper was collected during and after the D-Penicillamine challenge. After the test, there was an increase of more than 5 times the upper limit. The Leipzig score was recalculated, and a diagnosis of WD was made with a score of five. Even cases without important diagnostic findings such as Kayser-Fleischer ring or high 24-hour urine copper should be evaluated according to the Leipzig score. It is vital to distinguish WD in patients with young-onset movement disorder and neurological symptoms.
RESUMO
Hemorrhagic Destruction of the Brain, Subependymal Calcification, and Congenital Cataracts (HDBSCC) is a rare syndrome caused by biallelic mutations in the JAM3 gene with significant intrafamilial variability in clinical presentation and brain imaging phenotypes. The clinical presentation of HDBSCC includes severe recurrent hemorrhages involving the brain parenchyma and the ventricles beginning in utero and continuing in infancy together with dense central cataracts present at birth. This comprehensive review documents reported cases on this unique condition and describes its genetic, neuroradiologic and ophthalmic features. It should be included in the differential diagnosis of children with congenital cataracts and neurodevelopmental abnormalities. Unique clinical, imaging findings and genetic testing can help the diagnosis.
RESUMO
Introduction: Acute disseminated encephalomyelitis (ADEM) is an acute autoimmune demyelinating disease of the central nervous system that typically follows an acute viral infection or post-vaccination. It is more common in children than in adults and is characterized by widespread demyelination of the white matter of the brain and spinal cord. ADEM typically presents as a monophasic illness. Multiphasic ADEM has been an emerging topic with case reports highlighting the importance of the need for further investigation and literature on this topic. Multiphasic ADEM in adults is a rare presentation with an atypical course that can cause a delay in diagnosis leading to disability in patients. Case Presentation: Multiple case reports have shown the presence of multiphasic ADEM without an established cause, emphasizing the importance of additional insight and data in this context. This is a unique case of multiphasic ADEM in a 59-year-old male with a temporal gap of 26 years with an idiopathic cause. He was suffering from many neurologic symptoms involving his upper and lower extremities and level of consciousness, and after he was diagnosed with ADEM, he was commenced on steroids and achieved a marked improvement in his clinical status. Conclusion: Rarely, ADEM can present with recurrence or multiple phases, which requires prompt diagnosis and treatment. These phases may or may not be preceded by an identifiable trigger. Most patients respond well to steroids. However, further studies are required to achieve a good understanding and guidelines when dealing with such cases.
RESUMO
Introduction: NF2-related schwannomatosis (NF2) is a rare inheritable autosomal dominant disorder characterized by multiple nervous system neoplasms. Case Presentation: In this report, we present a rare case of NF2 exhibiting the complete constellation of multiple inherited schwannomas, meningiomas, and ependymomas (MISME syndrome). Conclusion: This is the first documented case of simultaneous occurrence of all three tumors in a single patient in the Philippines, adding to the sparse evidence of MISME syndrome in world literature.
RESUMO
OBJECTIVE: Thoracic dorsal arachnoid webs are intradural membranes that may cause obstruction of CSF flow and spinal cord compression. While well-recognized, they are rare and there is a paucity of long-term data on their natural history and prognosis. We reviewed radiographic features, surgical indications, and pathologic specimens of patients diagnosed with focal thoracic dorsal arachnoid webs. METHODS: A radiology database and surgical case logs were queried for thoracic arachnoid webs at a single hospital system for a ten-year period. A retrospective chart review was performed on identified cases. RESULTS: We identified 127 patients with dorsal thoracic arachnoid webs. Arachnoid webs were radiographically classified into three morphologic types: Type 1 (54%) causing spinal cord deformity only, Type 2 (32%) producing cord deformity with myelomalacia, and Type 3 (14%) with cord deformity, myelomalacia, and syringomyelia. These arachnoid webs were commonly centered at the upper thoracic T4 segmental level. Forty-one cases (32%) required surgery, generally for thoracic myelopathy with gait instability (46%) and lower extremity numbness and pain (39%). In patients who underwent surgery, 79% experienced symptomatic improvement and 21% remained stable, after an average of 21 months follow-up evaluation. Surgical pathology revealed fibrous connective tissue (100%) with calcifications (26%) or inflammation (7%). CONCLUSIONS: The majority in a large series of patients with dorsal arachnoid webs did not undergo surgical intervention, but those with myelomalacia and syrinx experienced radiographic and clinical deterioration without surgery. Surgery to treat symptomatic arachnoid webs results in significant clinical improvement with low surgical morbidity.
RESUMO
AIMS AND OBJECTIVE: Advances in artificial intelligence (AI), particularly in large language models (LLMs) like ChatGPT (versions 3.5 and 4.0) and Google Gemini, are transforming healthcare. This study explores the performance of these AI models in solving diagnostic quizzes from "Neuroradiology: A Core Review" to evaluate their potential as diagnostic tools in radiology. MATERIALS AND METHODS: We assessed the accuracy of ChatGPT 3.5, ChatGPT 4.0, and Google Gemini using 262 multiple-choice questions covering brain, head and neck, spine, and non-interpretive skills. Each AI tool provided answers and explanations, which were compared to textbook answers. The analysis followed the STARD (Standards for Reporting of Diagnostic Accuracy Studies) guidelines, and accuracy was calculated for each AI tool and subgroup. RESULTS: ChatGPT 4.0 achieved the highest overall accuracy at 64.89%, outperforming ChatGPT 3.5 (62.60%) and Google Gemini (55.73%). ChatGPT 4.0 excelled in brain, head, and neck diagnostics, while Google Gemini performed best in head and neck but lagged in other areas. ChatGPT 3.5 showed consistent performance across all subgroups. CONCLUSION: This study found that advanced AI models, including ChatGPT 4.0 and Google Gemini, vary in diagnostic accuracy, with ChatGPT 4.0 leading at 64.89% overall. While these tools are promising in improving diagnostics and medical education, their effectiveness varies by area, and Google Gemini performs unevenly across different categories. The study underscores the need for ongoing improvements and broader evaluation to address ethical concerns and optimize AI use in patient care.
RESUMO
Acute manifestations of sickle cell disease (SCD) are numerous and multisystemic. Cerebral fat embolism (CFE) is a rare but serious complication of SCD caused by bone marrow necrosis (BMN) during vaso-occlusive crises (VOC). We present the case of a 41-year-old man with SCD who developed severe VOC and multi-organ dysfunction. He subsequently experienced neurological deterioration with decreased consciousness and diffuse encephalopathy on serial electroencephalograms. Bone marrow aspiration confirmed BMN. Brain MRI revealed extensive diffuse leukoencephalopathy, vasogenic and cytotoxic edema in the white matter, patchy edema in the cranial vault bone marrow on fat-suppressed FLAIR sequence (a finding consistent with the confirmed BMN), and multiple cerebral microbleeds on susceptibility-weighted imaging consistent with CFE. The management of acute neurological complications of SCD varies depending on the specific complication. Brain MRI plays a crucial role in the accurate diagnosis of these complications to guide appropriate treatment.
RESUMO
Decision making and patient selection for isolated occlusion of the M2 segment of the middle cerebral artery is challenging, with the efficacy of treatment for such distal occlusions still controversial in recent studies, as occlusion in such small and thin vessels tends to manifest with less severe symptoms initially and hold higher surgical risks. It is even less clear which endovascular technique is more effective for this type of occlusion, with the choice usually left to the radiologist's preference. We conducted a controlled prospective study of consecutive patients presenting to Novara Hospital with an acute M2 occlusion diagnosed at the AngioCt and, following a predetermined randomization, we divided them into two treatment arms with either stent retriever aspiration (SRa) or direct aspiration (DA). All patients were examined on admission, at discharge and after 3 months, and clinical data as well as mRS and NIHSS scale scores were recorded. Our primary aim was to evaluate the difference in recanalization rate in the two groups by comparing the angiographic eTICI obtained with the two techniques. We then investigated whether there were differences in clinical outcomes and complications. Our data confirm a good recanalization rate with an eTICI ≥ 2b in 76.19% of patients in the cohort. An overall good outcome was achieved in 57% of patients as ∆NIHSS and in 50.79% of patients considering mRS. We found no statistically significant difference in recanalization rate nor higher complication rate in either the SRa or DA group. Both techniques are safe, effective and can be considered equally.
RESUMO
INTRODUCTION: The establishment of diagnostic reference levels (DRLs) is challenge for interventional neuroradiology (INR) due to the complexity and variability of its procedures. OBJECTIVE: The main objective of this systematic review is to analyse and compare DRLs in fluoroscopy-guided procedures in INR. METHODS: An observational study reporting DRLs in INR procedures, specifically cerebral arteriography, cerebral aneurysm embolisation, cerebral thrombectomy, embolisation of arteriovenous malformations (AVM), arteriovenous fistulas (AVF), retinoblastoma embolisation, and spinal cord arteriography. Comprehensive literature searches for relevant studies published between 2017 and 2023 were conducted using the Scopus, PubMed, and Web of Science databases. RESULTS: A total of 303 articles were identified through an extensive literature search, with 159 removed due to duplication. The title and abstract of 144 studies were assessed and excluded if they did not meet the inclusion criteria. Thirty-one out of the 144 articles were selected for a thorough full-text screening. Twenty-one articles were included in the review after the complete text screening. CONCLUSION: The different conditions of patients undergoing INR procedures pose a barrier to the standardization of DRLs; nevertheless, they are extremely important for monitoring and optimising radiological practices.
RESUMO
This case report describes the potential utility of MR spectroscopy and MR perfusion imaging in a patient with central nervous system involvement of Erdheim-Chester disease (ECD). A 57-year-old male presented with a variety of neurological symptoms, and conventional MRI of the brain showed multiple supratentorial and infratentorial findings that generated a wide differential diagnosis. Advanced MRI sequences along with subsequent CT imaging of the abdomen and ultimately a renal biopsy helped narrow the differential and confirm the diagnosis of ECD. Case reports of ECD are sporadic, and the role of advanced neuroimaging in diagnosing this disease has not been fully elucidated.
RESUMO
IgG4-related disease (IgG4-RD) is a recently described multisystemic disorder with a spectrum of manifestations that continue to be described. Nonetheless, there are recognised distinct patterns of disease. Neurological involvement is rare, particularly in isolation, but IgG4-RD may present with orbital disease, hypophysitis or pachymeningitis. Typically, it is highly responsive to treatment. This review highlights neurological manifestations of IgG4-RD and emphasises the importance of a high index of clinical suspicion to facilitate investigation and appropriate management, avoiding irreversible tissue damage and neurological dysfunction. We present a treatment algorithm for suggested management of IgG4-RD affecting the nervous system.