Development and Validation of a Dynamic Prediction Model for Massive Hemorrhage in Trauma.
Ontology highlight
ABSTRACT: Objectives
Early warning prediction of massive hemorrhages can greatly reduce mortality in trauma patients. This study aimed to develop and validate dynamic prediction models for massive hemorrhage in trauma patients.Methods
Based on vital signs (e.g., heart rate, respiratory rate, pulse pressure, and peripheral oxygen saturation) time-series data and the gated recurrent unit algorithm, we characterized a group of models to flexibly and dynamically predict the occurrence of massive hemorrhages in the subsequent T hours (where T = 1, 2, and 3). Models were evaluated in terms of accuracy, sensitivity, specificity, positive predictive value, negative predictive value, F1 score, and the area under the curve (AUC).Results
Results show that of the 2205 trauma patients selected for model development, a total of 265 (12.02%) had a massive hemorrhage. The AUCs of the model in the 1-h-group, 2-h-group, and 3-h-group were 0.763 (95% CI: 0.708-0.820), 0.775 (95% CI: 0.728-0.823), and 0.756 (95% CI: 0.715-0.797), respectively. Finally, the models were used in a web calculator and information system for the hospital emergency department.Conclusions
This study developed and validated a group of dynamic prediction models based on vital sign time-series data and a deep-learning algorithm to assist medical staff in the early diagnosis and dynamic prediction of a future massive hemorrhage in trauma.
SUBMITTER: Guo C
PROVIDER: S-EPMC9729037 | biostudies-literature | 2022
REPOSITORIES: biostudies-literature
Publications
Development and Validation of a Dynamic Prediction Model for Massive Hemorrhage in Trauma.
Guo Chengyu C Tian Maolin M Gong Minghui M Pan Fei F Han Hui H Li Chunping C Li Tanshi T
Emergency medicine international 20221130
<h4>Objectives</h4>Early warning prediction of massive hemorrhages can greatly reduce mortality in trauma patients. This study aimed to develop and validate dynamic prediction models for massive hemorrhage in trauma patients.<h4>Methods</h4>Based on vital signs (e.g., heart rate, respiratory rate, pulse pressure, and peripheral oxygen saturation) time-series data and the gated recurrent unit algorithm, we characterized a group of models to flexibly and dynamically predict the occurrence of massi ...[more]
Similar Datasets
Project description:BackgroundMassive hemorrhage is the main cause of preventable death after trauma. This study aimed to establish prediction models for early diagnosis of massive hemorrhage in trauma.MethodsUsing the trauma database of Chinese PLA General Hospital, two logistic regression (LR) models were fit to predict the risk of massive hemorrhage in trauma. Sixty-two potential predictive variables, including clinical symptoms, vital signs, laboratory tests, and imaging results, were included in this study. Variable selection was done using the least absolute shrinkage and selection operator (LASSO) method. The first model was constructed based on LASSO feature selection results. The second model was constructed based on the first vital sign recordings of trauma patients after admission. Finally, a web calculator was developed for clinical use.ResultsA total of 2353 patients were included in this study. There were 377 (16.02%) patients with massive hemorrhage. The selected predictive variables were heart rate (OR: 1.01; 95% CI: 1.01-1.02; P<0.001), pulse pressure (OR: 0.99; 95% CI: 0.98-0.99; P = 0.004), base excess (OR: 0.90; 95% CI: 0.87-0.93; P<0.001), hemoglobin (OR: 0.95; 95% CI: 0.95-0.96; P<0.001), displaced pelvic fracture (OR: 2.13; 95% CI: 1.48-3.06; P<0.001), and a positive computed tomography scan or positive focused assessment with sonography for trauma (OR: 1.62; 95% CI: 1.21-2.18; P = 0.001). Model 1, which was developed based on LASSO feature selection results and LR, displayed excellent discrimination (AUC: 0.894; 95% CI: 0.875-0.912), good calibration (P = 0.405), and clinical utility. In addition, the predictive power of model 1 was better than that of model 2 (AUC: 0.718; 95% CI: 0.679-0.757). Model 1 was deployed as a public web tool ( http://82.156.217.249:8080/ ).ConclusionsOur study developed and validated prediction models to assist medical staff in the early diagnosis of massive hemorrhage in trauma. An open web calculator was developed to facilitate the practical application of the research results.
| S-EPMC9661746 | biostudies-literature
Project description:Hemorrhage remains a major cause of potentially preventable deaths. Trauma and massive transfusion are associated with coagulopathy secondary to tissue injury, hypoperfusion, dilution, and consumption of clotting factors and platelets. Concepts of damage control surgery have evolved prioritizing early control of the cause of bleeding by non-definitive means, while hemostatic control resuscitation seeks early control of coagulopathy.Hemostatic resuscitation provides transfusions with plasma and platelets in addition to red blood cells in an immediate and sustained manner as part of the transfusion protocol for massively bleeding patients. Although early and effective reversal of coagulopathy is documented, the most effective means of preventing coagulopathy of massive transfusion remains debated and randomized controlled studies are lacking. Viscoelastical whole blood assays, like TEG and ROTEM however appear advantageous for identifying coagulopathy in patients with severe hemorrhage as opposed the conventional coagulation assays.In our view, patients with uncontrolled bleeding, regardless of it's cause, should be treated with hemostatic control resuscitation involving early administration of plasma and platelets and earliest possible goal-directed, based on the results of TEG/ROTEM analysis. The aim of the goal-directed therapy should be to maintain a normal hemostatic competence until surgical hemostasis is achieved, as this appears to be associated with reduced mortality.
| S-EPMC3439269 | biostudies-literature
Project description:ObjectiveDevelop a novel machine learning (ML) model to rapidly identify trauma patients with severe hemorrhage at risk of early mortality.BackgroundThe critical administration threshold (CAT, 3 or more units of red blood cells in a 60-minute period) indicates severe hemorrhage and predicts mortality, whereas early identification of such patients improves survival.MethodsPatients from the PRospective, Observational, Multicenter, Major Trauma Transfusion and Pragmatic, Randomized Optimal Platelet, and Plasma Ratio studies were identified as either CAT+ or CAT-. Candidate variables were separated into 4 tiers based on the anticipated time of availability during the patient's assessment. ML models were created with the stepwise addition of variables and compared with the baseline performance of the assessment of blood consumption (ABC) score for CAT+ prediction using a cross-validated training set and a hold-out validation test set.ResultsOf 1245 PRospective, Observational, Multicenter, Major Trauma Transfusion and 680 Pragmatic, Randomized Optimal Platelet and Plasma Ratio study patients, 1312 were included in this analysis, including 862 CAT+ and 450 CAT-. A CatBoost gradient-boosted decision tree model performed best. Using only variables available prehospital or on initial assessment (Tier 1), the ML model performed superior to the ABC score in predicting CAT+ patients [area under the receiver-operator curve (AUC = 0.71 vs 0.62)]. Model discrimination increased with the addition of Tier 2 (AUC = 0.75), Tier 3 (AUC = 0.77), and Tier 4 (AUC = 0.81) variables.ConclusionsA dynamic ML model reliably identified CAT+ trauma patients with data available within minutes of trauma center arrival, and the quality of the prediction improved as more patient-level data became available. Such an approach can optimize the accuracy and timeliness of massive transfusion protocol activation.
| S-EPMC10513183 | biostudies-literature
Project description:BackgroundThe prognosis of trauma patients is highly dependent on early medical diagnosis. By constructing a nomogram model, the risk of adverse outcomes can be displayed intuitively and individually, which has important clinical implications for medical diagnosis.ObjectiveTo develop and evaluate models for predicting patients with adverse outcomes of trauma that can be used in different data availability settings in China.MethodsThis was a retrospective prognostic study using data from 8 public tertiary hospitals in China from 2018. The data were randomly divided into a development set and a validation set. Simple, improved and extended models predicting adverse outcomes were developed, with adverse outcomes defined as in-hospital death or ICU transfer, and patient clinical characteristics, vital signs, diagnoses, and laboratory test values as predictors. The results of the models were presented in the form of nomograms, and performance was evaluated using area under the receiver operating characteristic curve (ROC-AUC), precision-recall (PR) curves (PR-AUC), Hosmer-Lemeshow goodness-of-fit test, calibration curve, and decision curve analysis (DCA).ResultsOur final dataset consisted of 18,629 patients (40.2% female, mean age of 52.3), 1,089 (5.85%) of whom resulted in adverse outcomes. In the external validation set, three models achieved ROC-AUC of 0.872, 0.881, and 0.903, and a PR-AUC of 0.339, 0.337, and 0.403, respectively. In terms of the calibration curves and DCA, the models also performed well.ConclusionsThis prognostic study found that three prediction models and nomograms including the patient clinical characteristics, vital signs, diagnoses, and laboratory test values can support clinicians in more accurately identifying patients who are at risk of adverse outcomes in different settings based on data availability.
| S-EPMC11334749 | biostudies-literature
Project description:Importance:Prehospital trauma triage protocols are used worldwide to get the right patient to the right hospital and thereby improve the chance of survival and avert lifelong disabilities. The American College of Surgeons Committee on Trauma set target levels for undertriage rates of less than 5%. None of the existing triage protocols has been able to achieve this target in isolation. Objective:To develop and validate a new prehospital trauma triage protocol to improve current triage rates. Design, Setting, and Participants:In this multicenter cohort study, all patients with trauma who were 16 years and older and transported to a trauma center in 2 different regions of the Netherlands were included in the analysis. Data were collected from January 1, 2012, through June 30, 2014, in the Central Netherlands region for the design data cohort and from January 1 through December 31, 2015, in the Brabant region for the validation cohort. Data were analyzed from May 3, 2017, through July 19, 2018. Main Outcomes and Measures:A new prediction model was developed in the Central Netherlands region based on prehospital predictors associated with severe injury. Severe injury was defined as an Injury Severity Score greater than 15. A full-model strategy with penalized maximum likelihood estimation was used to construct a model with 8 predictors that were chosen based on clinical reasoning. Accuracy of the developed prediction model was assessed in terms of discrimination and calibration. The model was externally validated in the Brabant region. Results:Using data from 4950 patients with trauma from the Central Netherlands region for the design data set (58.3% male; mean [SD] age, 47 [21] years) and 6859 patients for the validation Brabant region (52.2% male; mean [SD] age, 51 [22] years), the following 8 significant predictors were selected for the prediction model: age; systolic blood pressure; Glasgow Coma Scale score; mechanism criteria; penetrating injury to the head, thorax, or abdomen; signs and/or symptoms of head or neck injury; expected injury in the Abbreviated Injury Scale thorax region; and expected injury in 2 or more Abbreviated Injury Scale regions. The prediction model showed a C statistic of 0.823 (95% CI, 0.813-0.832) and good calibration. The cutoff point with a minimum specificity of 50.0% (95% CI, 49.3%-50.7%) led to a sensitivity of 88.8% (95% CI, 87.5%-90.0%). External validation showed a C statistic of 0.831 (95% CI, 0.814-0.848) and adequate calibration. Conclusions and Relevance:The new prehospital trauma triage prediction model may lower undertriage rates to approximately 10% with an overtriage rate of 50%. The next step should be to implement this prediction model with the use of a mobile app for emergency medical services professionals.
| S-EPMC6537785 | biostudies-literature
Project description:BackgroundWe aimed to establish risk factors for stroke-associated pneumonia (SAP) following intracerebral hemorrhage (ICH) and develop an efficient and convenient model to predict SAP in patients with ICH.MethodsOur study involved 1333 patients consecutively diagnosed with ICH and admitted to the Neurology Department of the First Affiliated Hospital of Wenzhou Medical University. The 1333 patients were randomly divided (3:1) into the derivation cohort (n = 1000) and validation Cohort (n = 333). Variables were screened from demographics, lifestyle-related factors, comorbidities, clinical symptoms, neuroimaging features, and laboratory tests. In the derivation cohort, we developed a prediction model with multivariable logistic regression analysis. In the validation cohort, we assessed the model performance and compared it to previously reported models. The area under the receiver operating characteristic curve (AUROC), GiViTI calibration belt, net reclassification index (NRI), integrated discrimination index (IDI) and decision curve analysis (DCA) were used to assess the prediction ability and the clinical decision-making ability.ResultsThe incidence of SAP was 19.9% and 19.8% in the derivation (n = 1000) and validation (n = 333) cohorts, respectively. We developed a nomogram prediction model including age (Odds Ratio [OR] 1.037, 95% confidence interval [CI] 1.020-1.054), male sex (OR 1.824, 95% CI 1.206-2.757), multilobar involvement (OR 1.851, 95% CI 1.160-2.954), extension into ventricles (OR 2.164, 95% CI 1.456-3.215), dysphagia (OR 3.626, 95% CI 2.297-5.725), disturbance of consciousness (OR 2.113, 95% CI 1.327-3.362) and total muscle strength of the worse side (OR 0.93, 95% CI 0.876-0.987). Compared with previous models, our model was well calibrated and showed significantly higher AUROC, better reclassification ability (improved NRI and IDI) and a positive net benefit for predicted probability thresholds between 10% and 73% in DCA.ConclusionsWe developed a simple, valid, and clinically useful model to predict SAP following ICH, with better predictive performance than previous models. It might be a promising tool to assess the individual risk of developing SAP for patients with ICH and optimize decision-making.
| S-EPMC10559607 | biostudies-literature
Project description:ImportanceMassive transfusion is essential to prevent complications during uncontrolled intraoperative hemorrhage. As massive transfusion requires time for blood product preparation and additional medical personnel for a team-based approach, early prediction of massive transfusion is crucial for appropriate management.ObjectiveTo evaluate a real-time prediction model for massive transfusion during surgery based on the incorporation of preoperative data and intraoperative hemodynamic monitoring data.Design, setting, and participantsThis prognostic study used data sets from patients who underwent surgery with invasive blood pressure monitoring at Seoul National University Hospital (SNUH) from 2016 to 2019 and Boramae Medical Center (BMC) from 2020 to 2021. SNUH represented the development and internal validation data sets (n = 17 986 patients), and BMC represented the external validation data sets (n = 494 patients). Data were analyzed from November 2020 to December 2021.ExposuresA deep learning-based real-time prediction model for massive transfusion.Main outcomes and measuresMassive transfusion was defined as a transfusion of 3 or more units of red blood cells over an hour. A preoperative prediction model for massive transfusion was developed using preoperative variables. Subsequently, a real-time prediction model using preoperative and intraoperative parameters was constructed to predict massive transfusion 10 minutes in advance. A prediction model, the massive transfusion index, calculated the risk of massive transfusion in real time.ResultsAmong 17 986 patients at SNUH (mean [SD] age, 58.65 [14.81] years; 9036 [50.2%] female), 416 patients (2.3%) underwent massive transfusion during the operation (mean [SD] duration of operation, 170.99 [105.03] minutes). The real-time prediction model constructed with the use of preoperative and intraoperative parameters significantly outperformed the preoperative prediction model (area under the receiver characteristic curve [AUROC], 0.972; 95% CI, 0.968-0.976 vs AUROC, 0.824; 95% CI, 0.813-0.834 in the SNUH internal validation data set; P < .001). Patients with the highest massive transfusion index (ie, >90th percentile) had a 47.5-fold increased risk for a massive transfusion compared with those with a lower massive transfusion index (ie, <80th percentile). The real-time prediction model also showed excellent performance in the external validation data set (AUROC of 0.943 [95% CI, 0.919-0.961] in BMC).Conclusions and relevanceThe findings of this prognostic study suggest that the real-time prediction model for massive transfusion showed high accuracy of prediction performance, enabling early intervention for high-risk patients. It suggests strong confidence in artificial intelligence-assisted clinical decision support systems in the operating field.
| S-EPMC9856486 | biostudies-literature
Project description:Non-accidental trauma (NAT) is deadly and difficult to predict. Transformer models pretrained on large datasets have recently produced state of the art performance on diverse prediction tasks, but the optimal pretraining strategies for diagnostic predictions are not known. Here we report the development and external validation of Pretrained and Adapted BERT for Longitudinal Outcomes (PABLO), a transformer-based deep learning model with multitask clinical pretraining, to identify patients who will receive a diagnosis of NAT in the next year. We develop a clinical interface to visualize patient trajectories, model predictions, and individual risk factors. In two comprehensive statewide databases, approximately 1% of patients experience NAT within one year of prediction. PABLO predicts NAT events with area under the receiver operating characteristic curve (AUROC) of 0.844 (95% CI 0.838-0.851) in the California test set, and 0.849 (95% CI 0.846-0.851) on external validation in Florida, outperforming comparator models. Multitask pretraining significantly improves model performance. Attribution analysis shows substance use, psychiatric, and injury diagnoses, in the context of age and racial demographics, as influential predictors of NAT. As a clinical decision support system, PABLO can identify high-risk patients and patient-specific risk factors, which can be used to target secondary screening and preventive interventions at the point-of-care.
| S-EPMC10356774 | biostudies-literature
Project description:ImportancePostpancreatectomy hemorrhage (PPH) due to postoperative pancreatic fistula (POPF) is a life-threatening complication after pancreatoduodenectomy. However, there is no prediction tool for early identification of patients at high risk of late PPH.ObjectiveTo develop and validate a prediction model for PPH.Design, setting, and participantsThis retrospective prognostic study included consecutive patients with clinically relevant POPF who underwent pancreatoduodenectomy from January 1, 2009, to May 20, 2023, at the University Hospital Mannheim (derivation cohort), and from January 1, 2012, to May 31, 2022, at the University Hospital Dresden (validation cohort). Data analysis was performed from May 30 to July 29, 2023.ExposureClinical and radiologic features of PPH.Main outcomes and measuresAccuracy of a predictive risk score of PPH. A multivariate prediction model-the hemorrhage risk score (HRS)-was established in the derivation cohort (n = 139) and validated in the validation cohort (n = 154).ResultsA total of 293 patients (187 [64%] men; median age, 69 [IQR, 60-76] years) were included. The HRS comprised 4 variables with associations: sentinel bleeding (odds ratio [OR], 35.10; 95% CI, 5.58-221.00; P < .001), drain fluid culture positive for Candida species (OR, 14.40; 95% CI, 2.24-92.20; P < .001), and radiologic proof of rim enhancement of (OR, 12.00; 95% CI, 2.08-69.50; P = .006) or gas within (OR, 12.10; 95% CI, 2.22-65.50; P = .004) a peripancreatic fluid collection. Two risk categories were identified with patients at low risk (0-1 points) and high risk (≥2 points) to develop PPH. Patients with PPH were predicted accurately in the derivation cohort (C index, 0.97) and validation cohort (C index 0.83). The need for more invasive PPH management (74% vs 34%; P < .001) and severe complications (49% vs 23%; P < .001) were more frequent in high-risk patients compared with low-risk patients.Conclusions and relevanceIn this retrospective prognostic study, a robust prediction model for PPH was developed and validated. This tool may facilitate early identification of patients at high risk for PPH.
| S-EPMC10701614 | biostudies-literature
Project description:Acute Kidney Injury (AKI) complicating major trauma is associated with increased mortality and morbidity. Traumatic AKI has specific risk factors and predictable time-course facilitating diagnostic modelling. In a single centre, retrospective observational study we developed risk prediction models for AKI after trauma based on data around intensive care admission. Models predicting AKI were developed using data from 830 patients, using data reduction followed by logistic regression, and were independently validated in a further 564 patients. AKI occurred in 163/830 (19.6%) with 42 (5.1%) receiving renal replacement therapy (RRT). First serum creatinine and phosphate, units of blood transfused in first 24 h, age and Charlson score discriminated need for RRT and AKI early after trauma. For RRT c-statistics were good to excellent: development: 0.92 (0.88-0.96), validation: 0.91 (0.86-0.97). Modelling AKI stage 2-3, c-statistics were also good, development: 0.81 (0.75-0.88) and validation: 0.83 (0.74-0.92). The model predicting AKI stage 1-3 performed moderately, development: c-statistic 0.77 (0.72-0.81), validation: 0.70 (0.64-0.77). Despite good discrimination of need for RRT, positive predictive values (PPV) at the optimal cut-off were only 23.0% (13.7-42.7) in development. However, PPV for the alternative endpoint of RRT and/or death improved to 41.2% (34.8-48.1) highlighting death as a clinically relevant endpoint to RRT.
| S-EPMC5827665 | biostudies-literature