ObjectiveNeurological deterioration after mild traumatic brain injury (TBI) has been recognized as a poor prognostic factor. Early detection of neurological deterioration would allow appropriate monitoring and timely therapeutic interventions to improve patient outcomes. In this study, we developed a machine learning model to predict the occurrence of neurological deterioration after mild TBI using information obtained on admission.MethodsThis was a retrospective cohort study of data from the Think FAST registry, a multicenter prospective observational study of elderly TBI patients in Japan. Patients with an admission Glasgow Coma Scale (GCS) score of 12 or below or who underwent surgical treatment immediately upon admission were excluded. Neurological deterioration was defined as a decrease of 2 or more points from a GCS score of 13 or more within 24 h of hospital admission. The model predictive accuracy was judged with the area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC), and the Youden index was used to determine the cutoff value.ResultsA total of 421 of 721 patients registered in the Think FAST registry between December 2019 and May 2021 were included in our study, among whom 25 demonstrated neurological deterioration. Among several machine learning algorithms, eXtreme Gradient Boosting (XGBoost) demonstrated the highest predictive accuracy in cross-validation, with an AUROC of 0.81 (±0.07) and an AUPRC of 0.33 (±0.08). Through SHapley Additive exPlanations (SHAP) analysis, five important features (D-dimer, fibrinogen, acute subdural hematoma thickness, cerebral contusion size, and systolic blood pressure) were identified and used to construct a better performing model (cross-validation AUROC of 0.84 and AUPRC of 0.34; testing data AUROC of 0.77 and AUPRC of 0.19). At the cutoff value from the Youden index, the model showed a sensitivity, specificity, and positive predictive value of 60, 96, and 38%, respectively. When neurosurgeons attempted to predict neurological deterioration using the same testing data, their values were 20, 94, and 19%, respectively.ConclusionIn this study, our predictive model showed an acceptable performance in detecting neurological deterioration after mild TBI. Further validation through prospective studies is necessary to confirm these results.

OBJECTIVE: To evaluate the effectiveness of a "telestration" system in which the mentor annotates the view of the surgical field, for endoscopic transsphenoidal surgery (ETS). METHODS: The use of telestration was evaluated for sellar floor-opening during ETS and for a task performed using ETS simulation training. During ETS, the mentor outlined the opening area of the sella turcica on the monitor and then the trainee surgeon opened the sella, either with the telestration displayed (telestration (+) + ) group, n = 8) or without (telestration (-)- ) group, n = 7). In the task using an ETS training model, 18 subjects were asked to touch the indicated targets with the forceps, once with the instructions given via telestration and once with verbal instructions only. RESULTS: During ETS, the telestration (+) + ) group had a significantly higher concordance rate between the planned bone window and actual bone window than the telestration (-)- ) group (92.97 +/- 4.16% vs. 77.57 +/- 10.51%, P = 0.014). In the ETS model, the time required to finish the task was significantly less with telestration than with verbal instructions alone (P P = 0.002). None of the subjects had errors when telestration was used, while subjects made an average of 0.33 +/- 0.59 errors and had to re-listen to the instructions 0.27 +/- 0.46 times when only verbal instructions were given. CONCLUSIONS: The use of the telestration system during ETS facilitated the communication of the mentor's intentions to the trainee surgeon and contributed to safer, more accurate surgery. The system was also thought to be useful in reducing operative time.

IMPORTANCE An adequate system for triaging patients with head trauma in prehospital settings and choosing optimal medical institutions is essential for improving the prognosis of these patients. To our knowledge, there has been no established way to stratify these patients based on their head trauma severity that can be used by ambulance crews at an injury site. OBJECTIVES To develop a prehospital triage system to stratify patients with head trauma according to trauma severity by using several machine learning techniques and to evaluate the predictive accuracy of these techniques. DESIGN, SETTING, AND PARTICIPANTS This single-center retrospective cohort study was conducted by reviewing the electronic medical records of consecutive patients who were transported to Tokyo Medical and Dental University Hospital in Japan from April 1.2018, to March 31, 2021. Patients younger than 16 years with cardiopulmonary arrest on arrival or with a significant amount of missing data were excluded. MAIN OUTCOMES AND MEASURES Machine learning-based prediction models to detect the presence of traumatic intracranial hemorrhage were constructed. The predictive accuracy of the models was evaluated with the area under the receiver operating curve (ROC-AUC), area under the precision recall curve (PR-AUC), sensitivity, specificity, and other representative statistics. RESULTS A total of 2123 patients (1527 male patients [71.9%]; mean [SD] age, 57.6 [19.8] years) with head trauma were enrolled in this study. Traumatic intracranial hemorrhage was detected in 258 patients (12.2%). Among several machine learning algorithms, extreme gradient boosting (XGBoost) achieved the mean (SD) highest ROC-AUC (0.78 [0.02]) and PR-AUC (0.46 [0.01]) in cross-validation studies. In the testing set, the ROC-AUC was 0.80, the sensitivity was 74.0% (95% CI, 59.7%-85.4%), and the specificity was 74.9% (95% CI, 70.2%-79.3%). The prediction model using the National Institute for Health and Care Excellence (NICE) guidelines, which was calculated after consultation with physicians, had a sensitivity of 72.0% (95% CI, 57.5%-83.8%) and a specificity of 73.3% (95% CI, 68.7%-77.7%). The McNemar test revealed no statistically significant differences between the XGBoost algorithm and the NICE guidelines for sensitivity or specificity (P = .80 and P = .55, respectively). CONCLUSIONS AND RELEVANCE In this cohort study, the prediction model achieved a comparatively accurate performance in detecting-traumatic intracranial hemorrhage using only the simple pretransportation information from the patient. Further validation with a prospective multicenter data set is needed.

Cerebrospinal fluid (CSF) leakage is a major complication following endoscopic endonasal skull base surgery. Various skull base reconstruction methods are available, and the use of a vascularized nasoseptal flap (NSF) in skull base reconstruction has greatly contributed to a decrease in the CSF leak rate. A balloon catheter such as a sinus balloon or a Foley catheter is often used to support an NSF; however, in cases wherein nasal and/or paranasal structures supporting the balloon are lacking following the surgery, the NSF is not properly fixed and postoperative CSF leak may occur. Here we introduce a new technique of using multiple-balloon catheters to fix an NSF in such cases and provide the results of our analysis of the new technique's efficacy. Eight patients who underwent endonasal endoscopic surgery for the following cases were included: olfactory neuroblastoma (n = 6), recurrent craniofacial meningioma (n = 1), and recurrent chordoma (n = 1). After tumor resection, multilayered reconstruction with vascularized NSF was performed. Given that the Foley catheter was not stable to fix the flap in each case, we used an additional nasal catheter to support the Foley catheter. No complications such as postoperative CSF leak and necrosis of the vascularized flap were observed. These results suggest that the multiple-balloon catheter technique is a useful method for fixing the NSF to the skull base even when nasal cavity structures are missing due to surgical removal.

It is unclear whether the visual assessment of noninvasive arterial spin labeling magnetic resonance imaging (ASL) can identify instances of hemodynamic compromise including an elevated oxygen extraction fraction (OEF) measured by O-15-gas positron emission tomography (PET). Here we evaluated the relationship between a four-point visual assessment system referred to as 'ASL scores' using ASL with two postlabeling delays (PLDs: 1525 ms and 2525 ms) and some quantitative hemodynamic parameters measured by PET. We retrospectively evaluated the cases of 18 Japanese patients with moyamoya disease who underwent ASL and PET. We compared the patients' regional ASL scores on two ASL images to the regional values of PET parameters, and we observed a significant trend in accord with the presumed clinical severity among all PET parameters and ASL scores (p < .003). The ASL score of the long PLD (2525 ms) showed the highest specificity (98.5%) for elevated OEF. Our results suggest that hemodynamic impairment (including elevated OEF) in patients with moyamoya disease may be grossly assessed by a visual assessment of noninvasive ASL images, which can be easily obtained in clinical settings. (C) 2019 Elsevier Ltd. All rights reserved.

Background: Aneurysms less than 5 mm in diameter (small aneurysms) are generally believed to have a low rupture rate. In our clinical experience, they are often found ruptured, causing subarachnoid hemorrhage (SAH). Herein, we report our investigation on the characteristics, severity, and prognosis of cases of ruptured small aneurysm.Patients and Methods: We reviewed the data of 158 consecutive patients with aneurysmal SAH (except for those with dissecting aneurysms) treated in our hospital between 2009 and 2014. The maximum size, configuration, location, and distribution of aneurysms were examined using computed tomography angiography or digital subtraction angiography. We chose the following 7 risk factors for rupture to examine 53 cases of small ruptured aneurysms: age (< 50 years old), aneurysm with daughter sac, past medical history of hypertension, multiple aneurysms, family history of SAH, smoking habits, and excessive alcohol intake. In addition, we compared these 53 cases to the rest of the group with regard to symptoms (Hunt and Hess grade [H&H]), amount of hemorrhage (modified Fisher grade [mFisher]), prognosis (modified Rankin Scale [mRS]), and number of risk factors.Results: Of the 158 cases of ruptured aneurysms, 49 involved the anterior communicating artery (Acom), 41 involved the middle cerebral artery (MCA), 29 involved the internal carotid-posterior communicating artery (IP-PC), and 39 involved other locations. The average maximum diameter was 6.6 mm (range: 1.2-30 mm). Among the 53 small aneurysm cases (34% of the total), 33% were located in the Acom, 28% in the IC-PC, and 17% in the MCA. In low sites with infrequent aneurysms such as the internal carotid-anterior choroidal artery (IC-ACh) or the basilar top artery (BA top), small aneurysms accounted for more than 50%. After analysis of the risk factors for the 53 small aneurysm cases, 4 patients (8%) had no risk factors. There was no statistically significant difference between H&H grade, mFisher grade, mRS, and the number of risk factors between the groups with small aneurysms and aneurysms 5 mm or larger.Conclusion: Cases with ruptured small aneurysms (<5 mm in diameter) accounted for approximately one-third of aneurysmal SAH cases in our institution. Many ruptured small aneurysms were located in the Acom or IC-PC. Four patients (8%) with ruptured small aneurysms had no risk factors. Further study is needed to identify the characteristics of ruptured small aneurysms.