A Computer-Assisted System for Early Mortality Risk Prediction in Patients with Traumatic Brain Injury Using Artificial Intelligence Algorithms in Emergency Room Triage.

Traumatic brain injury (TBI) remains a critical public health challenge. Although studies have found several prognostic factors for TBI, a useful early predictive tool for mortality has yet to be developed in the triage of the emergency room. This study aimed to use machine learning algorithms of artificial intelligence (AI) to develop predictive models for TBI patients in the emergency room triage. We retrospectively enrolled 18,249 adult TBI patients in the electronic medical records of three hospitals of Chi Mei Medical Group from January 2010 to December 2019, and undertook the 12 potentially predictive feature variables for predicting mortality during hospitalization. Six machine learning algorithms including logistical regression (LR) random forest (RF), support vector machines (SVM), LightGBM, XGBoost, and multilayer perceptron (MLP) were used to build the predictive model. The results showed that all six predictive models had high AUC from 0.851 to 0.925. Among these models, the LR-based model was the best model for mortality risk prediction with the highest AUC of 0.925; thus, we integrated the best model into the existed hospital information system for assisting clinical decision-making. These results revealed that the LR-based model was the best model to predict the mortality risk in patients with TBI in the emergency room. Since the developed prediction system can easily obtain the 12 feature variables during the initial triage, it can provide quick and early mortality prediction to clinicians for guiding deciding further treatment as well as helping explain the patient's condition to family members.

Spontaneous Spinal Epidural Hematoma After Normal Spontaneous Delivery with Epidural Analgesia: Case Report and Literature Review.

BACKGROUND: Pregnancy is a known risk factor for spontaneous spinal epidural hematoma. During cesarean section or vaginal delivery, the unstable hemodynamic status that may occur owing to fluctuation of intra-abdominal pressure increases the possibility of spontaneous spinal epidural hematoma. During labor and the postpartum period, neurologic symptoms may be masked by labor pain or anesthesia block, which makes early diagnosis difficult, especially in the obstetric clinic without a neurologist or neurosurgeon. CASE DESCRIPTION: A 28-year-old woman who had a normal spontaneous delivery under epidural anesthesia developed bilateral lower limb flaccid paralysis and loss of sensation 12.5 hours after delivery. Magnetic resonance imaging showed a 5.2 × 0.9 × 2 cm spinal epidural hematoma with severe spinal cord stenosis at the T2-T5 level with no evidence of a vascular anomaly. After emergent evacuation of the spinal epidural hematoma, lower limb muscle power improved from 0/5 to 1/5, and sensation gradually returned to bilateral lower limbs 22 days postoperatively. Deep vein thrombosis developed at 35 days postoperatively, and an inferior vena cava filter was implanted with urokinase infusion for thrombolytic therapy. She was discharged on day 52 after admission, and lower limb muscle power returned to normal after 3 months. CONCLUSIONS: Clinicians should observe postpartum women for signs of myelopathy or back tenderness and closely monitor neurologic function until anesthesia has run its course. A prompt diagnosis can enable prompt intervention.

In-Hospital Mortality After Spinal Surgery in Hemodialysis Patients: An 11-Year Population-Based Study.

BACKGROUND: Patients with end-stage renal disease (ESRD) are at an increased risk of surgical mortality. We aimed to investigate the factors associated with in-hospital mortality in patients with ESRD who underwent spinal surgery, which remains to be determined. MATERIAL AND METHODS: An age- and sex-matched cohort study was conducted using the Taiwan Longitudinal Health Insurance Database between January 2000 and December 2012. Kaplan-Meier curves were plotted with log-rank test to compare the differences between these 2 groups. The Cox proportional hazard model was used to estimate the hazard ratio of in-hospital mortality adjusted with potential confounding. RESULTS: In total, 4109 participants with pre-existing ESRD and 8218 patients without ESRD were included. The in-hospital mortality in ESRD (10.17%) was greater than without ESRD (1.39%). Spinal surgery patients with pre-existing ESRD had a 6.78-fold increase in-hospital mortality risk compared with those without ESRD. Spinal surgery patients with ESRD of any age, male or female, and comorbidities experienced a greater incidence of hospital mortality. In patients with ESRD, operations on spinal cords and spinal canal structures had the greatest hospital mortality (14.87%) compared with spinal fusion (3.46%) or excision or destruction of intervertebral disc (3.01%). Kaplan-Meier survival curves showed that patients with ESRD experienced greater hospital mortality than patients without ESRD in all 3 spinal surgery methods (log rank P < 0.0001). CONCLUSIONS: Spinal surgery patients with ESRD have greater in-hospital mortality than patients without ESRD. Age, sex, history of comorbidities, and types of surgical methods were associated with greater in-hospital mortality among patients with ESRD.

Traumatic Brain Injury Increases the Risk of Major Adverse Cardiovascular and Cerebrovascular Events: A 13-Year, Population-Based Study.

BACKGROUND: Previous studies have indicated traumatic brain injury (TBI) as a risk factor for stroke and myocardial injury. Whether TBI increases new onset of major adverse cardiovascular and cerebrovascular events (MACCE) is not well established. METHODS: Patients with a diagnosis of TBI from 2000 to 2012 were 1:2 age-, sex-, and age-adjusted comorbidities matched with normal population cohorts. The MACCE, which included coronary artery disease, heart failure and arrhythmia, ischemic and hemorrhagic stroke, and death, was defined as one inpatient admission with MACCE diagnosis. The maximum follow-up duration to MACCE after the initial TBI diagnosis was 5 years. The baseline comorbidities before TBI, including hypertension, diabetes mellitus, renal disease, and liver disease, also were considered to estimate the risk of MACCE. RESULTS: In total, 16,211 patients with TBI and 32,422 people from the control group were enrolled in the current study. Our results showed that patients with TBI had a 2.77-fold risk of MACCE, 1.72-fold risk of cardiovascular disease, 2.10-fold risk of ischemic stroke, 6.02-fold risk of hemorrhagic stroke, and 3.13-fold risk of mortality compared with the control group (all P < 0.0001) after adjusting the confounding factors. In addition, the trend of cumulated incidence risk among MACCE, cardiovascular disease, ischemic and hemorrhagic stroke, and mortality presented the greatest incidence within the first year after diagnosis and persisted during the 5 years of follow-up. CONCLUSIONS: Our results showed that patients with TBI have a significantly greater risk of MACCE than the control group. We hope this information will remind critical-care physicians and neurosurgeons to keep in mind the long-term effects of TBI on MACCE.

The Risk of Traumatic Brain Injury Occurring Among Patients with Parkinson Disease: A 14-Year Population-Based Study.

BACKGROUND: Previous studies have implicated traumatic brain injury (TBI) as a risk factor for Parkinson disease (PD). However, the incidence risk of new-onset TBI among patients with PD is not well established. This study investigated the contribution of PD to new-onset TBI associations. METHODS: The study selected 6076 patients with PD and using 1:2 propensity score matching 12,152 general population cohorts in a longitudinal population database in Taiwan. The Cox proportional hazard regression model was used to estimate the hazard ratio of TBI adjusted with the potential confounding factors. RESULTS: The incidence of TBI in patients with PD (2.57%) and compared cohorts (1.81%) was significantly different (P = 0.0007). Patients with PD had a higher risk of TBI (hazard ratio, 1.63; 95% confidence interval, 1.32-2.01) compared with the general population. Patients with PD with TBI incidence had a higher Charlson Comorbidity Index than did the general population with TBI (P < 0.0001). Fall is the major cause of TBI in patients with PD. CONCLUSIONS: This study shows that patients with PD have a high risk of TBI, and the major cause of TBI in patients with PD is fall.

Increased Risk of Anxiety or Depression After Traumatic Spinal Cord Injury in Patients with Preexisting Hyperlipidemia: A Population-Based Study.

OBJECTIVE: Anxiety or depression (AD) is a common complication after traumatic spinal cord injury (tSCI). This study sought to investigate the role of preexisting hyperlipidemia in new-onset AD after tSCI using a longitudinal population database. METHODS: This retrospective cohort study used Longitudinal Health Insurance Database data from January 1997 to December 2011. The case and comparison groups were individuals who experienced tSCI and who did and did not have preexisting hyperlipidemia, respectively. Kaplan-Meier curves were plotted, and log-rank test was used to compare the differences between these 2 groups. A Cox regression model was used to estimate the relative risk of AD. RESULTS: A total of 26,892 adult patients were enrolled in this study. After 1:3 matching with age and gender, it showed 1) tSCI patients with preexisting hyperlipidemia have a 1.32-fold adjusted hazard ratio (HR) compared with those without hyperlipidemia (P < 0.05); 2) The Kaplan-Meier plot in tSCI patients with hyperlipidemia were more likely to develop the new-onset AD than those without hyperlipidemia during the follow-up period (P = 0.0003); and 3) the stratified analysis showed the risk of AD among patients with tSCI aged 18-34 years (HR, 3.2; 95% confidence interval (CI) 1.2-8.9), male patients (HR, 1.3; 95% CI 1.1-1.6), and higher Charlson's comorbidity index (CCI) score (CCI > 2; HR, 1.9; 95% CI 1.2-2.9), and those with a history of stroke (HR, 1.7; 95% CI 1.0-2.7). CONCLUSIONS: Preexisting hyperlipidemia is an independent predictor of new-onset AD in patients with tSCI, especially in those who are younger, male, have a higher CCI score, and have stroke.