The effect of circle of willis anatomy and scanning practices on outcomes for blunt cerebrovascular injuries.

BACKGROUND: Limited research has explored the effect of Circle of Willis (CoW) anatomy among blunt cerebrovascular injuries (BCVI) on outcomes. It remains unclear if current BCVI screening and scanning practices are sufficient in identification of concomitant COW anomalies and how they affect outcomes. METHODS: This retrospective cohort study included adult traumatic BCVIs at 17 level I-IV trauma centers (08/01/2017-07/31/2021). The objectives were to compare screening criteria, scanning practices, and outcomes among those with and without COW anomalies. RESULTS: Of 561 BCVIs, 65% were male and the median age was 48 y/o. 17% (n = 93) had a CoW anomaly. Compared to those with normal CoW anatomy, those with CoW anomalies had significantly higher rates of any strokes (10% vs. 4%, p = 0.04), ICHs (38% vs. 21%, p = 0.001), and clinically significant bleed (CSB) before antithrombotic initiation (14% vs. 3%, p < 0.0001), respectively. Compared to patients with a normal CoW, those with a CoW anomaly also had ischemic strokes more often after antithrombotic interruption (13% vs. 2%, p = 0.02).Patients with CoW anomalies were screened significantly more often because of some other head/neck indication not outlined in BCVI screening criteria than patients with normal CoW anatomy (27% vs. 18%, p = 0.04), respectively. Scans identifying CoW anomalies included both the head and neck significantly more often (53% vs. 29%, p = 0.0001) than scans identifying normal CoW anatomy, respectively. CONCLUSIONS: While previous studies suggested universal scanning for BCVI detection, this study found patients with BCVI and CoW anomalies had some other head/neck injury not identified as BCVI scanning criteria significantly more than patients with normal CoW which may suggest that BCVI screening across all patients with a head/neck injury may improve the simultaneous detection of CoW and BCVIs. When screening for BCVI, scans including both the head and neck are superior to a single region in detection of concomitant CoW anomalies. Worsened outcomes (strokes, ICH, and clinically significant bleeding before antithrombotic initiation) were observed for patients with CoW anomalies when compared to those with a normal CoW. Those with a CoW anomaly experienced strokes at a higher rate than patients with normal CoW anatomy specifically when antithrombotic therapy was interrupted. This emphasizes the need for stringent antithrombotic therapy regimens among patients with CoW anomalies and may suggest that patients CoW anomalies would benefit from more varying treatment, highlighting the need to include the CoW anatomy when scanning for BCVI. LEVEL OF EVIDENCE: Level III, Prognostic/Epidemiological.

A joinpoint analysis examining trends in firearm injuries at six us trauma centers from 2016 to 2022.

BACKGROUND: There is an epidemic of firearm injuries in the United States since the mid-2000s. Thus, we sought to examine whether hospitalization from firearm injuries have increased over time, and to examine temporal changes in patient demographics, firearm injury intent, and injury severity. METHODS: This was a multicenter, retrospective, observational cohort study of patients hospitalized with a traumatic injury to six US level I trauma centers between 1/1/2016 and 6/30/2022. ICD-10-CM cause codes were used to identify and describe firearm injuries. Temporal trends were compared for demographics (age, sex, race, insured status), intent (assault, unintentional, self-harm, legal intervention, and undetermined), and severity (death, ICU admission, severe injury (injury severity score ≥ 16), receipt of blood transfusion, mechanical ventilation, and hospital and ICU LOS (days). Temporal trends were examined over 13 six-month intervals (H1, January-June; H2, July-December) using joinpoint regression and reported as semi-annual percent change (SPC); significance was p < 0.05. RESULTS: Firearm injuries accounted for 2.6% (1908 of 72,474) of trauma hospitalizations. The rate of firearm injuries initially declined from 2016-H1 to 2018-H2 (SPC = - 4.0%, p = 0.002), followed by increased rates from 2018-H2 to 2020-H1 (SPC = 9.0%, p = 0.005), before stabilizing from 2020-H1 to 2022-H1 (0.5%, p = 0.73). NH black patients had the greatest hospitalization rate from firearm injuries (14.0%) and were the only group to demonstrate a temporal increase (SPC = 6.3%, p < 0.001). The proportion of uninsured patients increased (SPC = 2.3%, p = 0.02) but there were no temporal changes by age or sex. ICU admission rates declined (SPC = - 2.2%, p < 0.001), but ICU LOS increased (SPC = 2.8%, p = 0.04). There were no significant changes over time in rates of death (SPC = 0.3%), severe injury (SPC = 1.6%), blood transfusion (SPC = 0.6%), and mechanical ventilation (SPC = 0.6%). When examined by intent, self-harm injuries declined over time (SPC = - 4.1%, p < 0.001), assaults declined through 2019-H2 (SPC = - 5.6%, p = 0.01) before increasing through 2022-H1 (SPC = 6.5%, p = 0.01), while undetermined injuries increased through 2019-H1 (SPC = 24.1%, p = 0.01) then stabilized (SPC = - 4.5%, p = 0.39); there were no temporal changes in unintentional injuries or legal intervention. CONCLUSIONS: Hospitalizations from firearm injuries are increasing following a period of declines, driven by increases among NH Black patients. Trauma systems need to consider these changing trends to best address the needs of the injured population.

The effect of restrictive fluid management on outcomes among geriatric hip fractures: a retrospective cohort study at five level I trauma centers.

Restrictive fluid management (RFM) for hemodynamically unstable trauma patients has reduced mortality rates. The objective was to determine whether RFM benefits geriatric hip fracture patients, who are usually hemodynamically stable. Design: Retrospective propensity-matched study. Setting: Five Level I trauma centers (January 1, 2018-December 12, 2018). Patients: Geriatric patients (65 years or older) with hip fractures were included in this study. Patients with multiple injuries, nonoperative management, and preoperative blood products were excluded. Intervention: Patients were grouped by fluid volume (normal saline, lactated Ringer, dextrose, electrolytes, and medications) received preoperatively or ≤24 hours of arrival; patients with standard fluid management (SFM) received ≥150 mL and RFM <150 mL of fluids. Main Outcome Measurements: The primary outcomes were length of stay (LOS), delayed ambulation (>2 days postoperatively), and mortality. Paired Student t-tests, Wilcoxon paired rank sum tests, and McNemar tests were used; an α value of < 0.05 was considered statistically significant. Results: There were 523 patients (40% RFM, 60% SFM); after matching, there were 95 patients per arm. The matched patients were well-balanced, including no difference in time from arrival to surgery. RFM and SFM patients received a median of 80 mL and 1250 mL of preoperative fluids, respectively (P < 0.001). Postoperative fluid volumes were 1550 versus 2000 mL, respectively, (P = 0.73), and LOSs were similar between the two groups (5 versus 5 days, P = 0.83). Mortality and complications, including acute kidney injuries, were similar. Delayed ambulation rates were similar overall. When stratified by preinjury ambulation status, SFM was associated with delayed ambulation for patients not walking independently before injury (P = 0.01), but RFM was not (P = 0.09). Conclusions: RFM seems to be safe in terms of laboratory results, complications, and disposition. SFM may lead to delayed ambulation for patients who are not walking independently before injury.

Computer Vision and Machine Learning-Based Gait Pattern Recognition for Flat Fall Prediction.

BACKGROUND: Gait recognition has been applied in the prediction of the probability of elderly flat ground fall, functional evaluation during rehabilitation, and the training of patients with lower extremity motor dysfunction. Gait distinguishing between seemingly similar kinematic patterns associated with different pathological entities is a challenge for the clinician. How to realize automatic identification and judgment of abnormal gait is a significant challenge in clinical practice. The long-term goal of our study is to develop a gait recognition computer vision system using artificial intelligence (AI) and machine learning (ML) computing. This study aims to find an optimal ML algorithm using computer vision techniques and measure variables from lower limbs to classify gait patterns in healthy people. The purpose of this study is to determine the feasibility of computer vision and machine learning (ML) computing in discriminating different gait patterns associated with flat-ground falls. METHODS: We used the Kinect® Motion system to capture the spatiotemporal gait data from seven healthy subjects in three walking trials, including normal gait, pelvic-obliquity-gait, and knee-hyperextension-gait walking. Four different classification methods including convolutional neural network (CNN), support vector machine (SVM), K-nearest neighbors (KNN), and long short-term memory (LSTM) neural networks were used to automatically classify three gait patterns. Overall, 750 sets of data were collected, and the dataset was divided into 80% for algorithm training and 20% for evaluation. RESULTS: The SVM and KNN had a higher accuracy than CNN and LSTM. The SVM (94.9 ± 3.36%) had the highest accuracy in the classification of gait patterns, followed by KNN (94.0 ± 4.22%). The accuracy of CNN was 87.6 ± 7.50% and that of LSTM 83.6 ± 5.35%. CONCLUSIONS: This study revealed that the proposed AI machine learning (ML) techniques can be used to design gait biometric systems and machine vision for gait pattern recognition. Potentially, this method can be used to remotely evaluate elderly patients and help clinicians make decisions regarding disposition, follow-up, and treatment.