Predicción de hemorragia masiva. Índice de shock e índice de shock modificado

Could machine learning algorithms help us predict massive bleeding at prehospital level?

OBJECTIVE

Comparison of the predictive ability of various machine learning algorithms (MLA) versus traditional prediction scales (TPS) for massive hemorrhage (MH) in patients with severe traumatic injury (STI).

DESIGN

On a database of a retrospective cohort with prehospital clinical variables and MH outcome, a treatment of the database was performed to be able to apply the different AML, obtaining a total set of 473 patients (80% training, 20% validation). For modeling, proportional imputation and cross validation were performed. The predictive power was evaluated with the ROC metric and the importance of the variables using the Shapley values.

SETTING

Out-of-hospital care of patients with STI.

PARTICIPANTS

Patients with STI treated out-of-hospital by a out-of-hospital medical service from January 2010 to December 2015 and transferred to a trauma center in Madrid.

INTERVENTIONS

None.

MAIN VARIABLES OF INTEREST

Obtaining and comparing the "Receiver Operating Characteristic curve" (ROC curve) metric of four MLAs: "random forest" (RF), "vector support machine" (SVM), "gradient boosting machine" (GBM) and "neural network" (NN) with the results obtained with TPS.

RESULTS

The different AML reached ROC values higher than 0.85, having medians close to 0.98. We found no significant differences between AMLs. Each AML offers a different set of more important variables with a predominance of hemodynamic, resuscitation variables and neurological impairment.

CONCLUSIONS

MLA may be helpful in patients with HM by outperforming TPS.

Shock index, modified shock index, age shock index score, and reverse shock index multiplied by Glasgow Coma Scale predicting clinical outcomes in traumatic brain injury: Evidence from a 10-year analysis in a single center

OBJECTIVES

Early identification of traumatic brain injury (TBI) patients at a high risk of mortality is very important. This study aimed to compare the predictive accuracy of four scoring systems in TBI, including shock index (SI), modified shock index (MSI), age-adjusted shock index (ASI), and reverse shock index multiplied by the Glasgow Coma Scale (rSIG).

PATIENTS AND METHODS

This is a retrospective analysis of a registry from the Taipei Tzu Chi trauma database. Totally, 1,791 patients with TBI were included. We investigated the accuracy of four major shock indices for TBI mortality. In the subgroup analysis, we also analyzed the effects of age, injury mechanism, underlying diseases, TBI severity, and injury severity.

RESULTS

The predictive accuracy of rSIG was significantly higher than those of SI, MSI, and ASI in all the patients [area under the receiver operating characteristic curve (AUROC), 0.710 vs. 0.495 vs. 0.527 vs. 0.598], especially in the moderate/severe TBI (AUROC, 0.625 vs. 0.450 vs. 0.476 vs. 0.529) and isolated head injury populations (AUROC 0.689 vs. 0.472 vs. 0.504 vs. 0.587). In the subgroup analysis, the prediction accuracy of mortality of rSIG was better in TBI with major trauma [Injury Severity Score (ISS) ≥ 16], motor vehicle collisions, fall injury, and healthy and cardiovascular disease population. rSIG also had a better prediction effect, as compared to SI, MSI, and ASI, both in the non-geriatric (age < 65 years) and geriatric (age ≥ 65 years).

CONCLUSION

rSIG had a better prediction accuracy for mortality in the overall TBI population than SI, MSI, and ASI. Although rSIG have better accuracy than other indices (ROC values indicate poor to moderate accuracy), the further clinical studies are necessary to validate our results.

Predictors of transfusion in trauma and their utility in the prehospital environment: a scoping review

Background: Hemorrhage is a leading cause of preventable mortality from trauma, necessitating resuscitation through blood product transfusions. Early and accurate identification of patients requiring transfusions in the prehospital setting may reduce delays in time to transfusion upon arrival to hospital, reducing mortality. The purpose of this study is to characterize existing literature on predictors of transfusion and analyze their utility in the prehospital context.Objectives: The objectives of this study are to characterize the existing quantity and quality of literature regarding predictor scores for transfusion in injured patients, and to analyse the utility of predictor scores for massive transfusions in the prehospital setting and identify prehospital predictor scores for future research.Methods: A search strategy was developed in consultation with information specialists. A literature search of OVID MEDLINE from 1946 to present was conducted for primary studies evaluating the predictive ability of scoring systems or single variables in predicting transfusion in all trauma settings.Results: Of the 5824 studies were identified, 5784 studies underwent title and abstract screening, 94 studies underwent full text review, and 72 studies were included in the final review. We identified 16 single variables and 52 scoring systems for predicting transfusion. Amongst single predictor variables, fluids administered and systolic blood pressure had the highest reported sensitivity (100%) and specificity (89%) for massive transfusion protocol (MTP) activation respectively. Amongst scoring systems for transfusion, the Shock Index and Modified Shock Index had the highest reported sensitivity (96%), while the Pre-arrival Model had the highest reported specificity (95%) for MTP activation. Overall, 20 scores were identified as being applicable to the prehospital setting, 25 scores were identified as being potentially applicable, and seven scores were identified as being not applicable.Conclusions: We identified an extensive list of predictive single variables, validated scoring systems, and derived models for massive transfusion, presented their properties, and identified those with potential utility in the prehospital setting. By further validating applicable scoring tools in the prehospital setting, we may begin to administer more timely transfusions in the trauma population.

Step by step transfusion timeline and its challenges in trauma: A retrospective study in a level one trauma center

BACKGROUND

Hemorrhagic shock is the leading cause of preventable early death in trauma patients. Transfusion management is guided by international guidelines promoting early and aggressive transfusion strategies. This study aimed to describe transfusion timelines in a trauma center and to identify key points to performing early and efficient transfusions.

METHODS

This is a monocentric retrospective study of 108 severe trauma patients, transfused within the first 48 h and hospitalized in an intensive care unit between January 2017 and May 2019.

RESULTS

One hundred and eight patients were transfused with 1250 labile blood products. Half of these labile blood products were transfused within 3 h of admission and consumed by 26 patients requiring massive transfusion (≥4 red blood cells [RBC] within 1 h). Among these, the median delay from patient's admission to labile blood products prescription was -11 min (-34 to -1); from admission to delivery of labile blood products was 1 min (-20 to 16); and from admission to first transfusion was 20 min (7-37) for RBC, 26 min (13-38) for plasma, and 72 min (51-103) for platelet concentrates. The anticipated prescription of labile blood products and the use of massive transfusion packs and lyophilized plasma units were associated with earlier achievement of high transfusion ratios.

CONCLUSION

This study provides detailed data on the transfusion timelines and composition, from prescription to initial transfusion. Transfusion anticipation, use of preconditioned transfusion packs including platelets, and lyophilized plasma allow rapid and high-ratio transfusion practices in severe trauma patients.

Effect of Intra- and Post-Operative Fluid and Blood Volume on Postoperative Pulmonary Edema in Patients with Intraoperative Massive Bleeding

In patients with intraoperative massive bleeding, the effects of fluid and blood volume on postoperative pulmonary edema are uncertain. Patients with intraoperative massive bleeding who had undergone a non-cardiac surgery in five hospitals were enrolled in this study. We evaluated the association of postoperative pulmonary edema risk and intra- and post-operatively administered fluid and blood volumes in patients with intraoperative massive bleeding. In total, 2090 patients were included in the postoperative pulmonary edema analysis, and 300 patients developed pulmonary edema within 72 h of the surgery. The postoperative pulmonary edema with hypoxemia analysis included 1660 patients, and the condition occurred in 161 patients. An increase in the amount of red blood cells transfused per hour after surgery increased the risk of pulmonary edema (hazard ratio: 1.03; 95% confidence interval: 1.01–1.05; p = 0.013) and the risk of pulmonary edema with hypoxemia (hazard ratio: 1.04; 95% confidence interval: 1.01–1.07; p = 0.024). An increase in the red blood cells transfused per hour after surgery increased the risk of developing pulmonary edema. This increase can be considered as a risk factor for pulmonary edema.

The Initial Prehospital Management of Traumatic Brain Injuries in Kigali, Rwanda

INTRODUCTION

Traumatic brain injuries (TBIs) are an important cause of mortality and disability around the world. Early intervention and stabilization are necessary to obtain optimal outcomes, yet little is written on the topic in low- and middle-income countries (LMICs). The aim is to provide a descriptive analysis of patients with TBI treated by Service d'Aide Medicale Urgente (SAMU), the prehospital ambulance service in Kigali, Rwanda.

HYPOTHESIS/PROBLEM

What is the incidence and nature of TBI seen on the ambulance in Kigali, Rwanda?

METHODS

A retrospective descriptive analysis was performed using SAMU records captured on an electronic database from December 2012 through May 2016. Variables included demographic information, injury characteristics, and interventional data.

RESULTS

Patients with TBIs accounted for 18.0% (n = 2,012) of all SAMU cases. The incidence of TBIs in Kigali was 234 crashes per 100,000 people. The mean age was 30.5 (SD = 11.5) years and 81.5% (n = 1,615) were men. The most common mechanisms were road traffic incidents (RTIs; 78.5%, n = 1,535), assault (10.7%, n=216), and falls (7.8%, n=156). Most patients experienced mild TBI (Glasgow Coma Score [GCS] ≥ 13; 83.5%, n = 1,625). The most common interventions were provision of pain medications (71.0%, n = 1,429), placement of a cervical collar (53.6%, n = 1,079), and administration of intravenous fluids (48.7%, n = 979). In total, TBIs were involved in 67.0% of all mortalities seen by SAMU.

CONCLUSION

Currently, TBIs represent a large burden of disease managed in the prehospital setting of Kigali, Rwanda. These injuries are most often caused by RTIs and were observed in 67% of mortalities seen by SAMU. Rwanda has implemented several initiatives to reduce the incidence of TBIs with a specific emphasis on road safety. Further efforts are needed to better prevent these injuries. Countries seeking to develop prehospital care capacity should train providers to manage patients with TBIs.

Mortality in civilian trauma patients and massive blood transfusion treated with high vs low plasma: red blood cell ratio. Systematic review and meta-analysis

Introduction: Massive bleeding in civilian trauma patients leads to dilutional coagulopathy. Transfusion with high plasma: red blood cell (RBC) ratio has been effective in reducing mortality in war trauma patients. However, in civilian trauma the evidence is controversial. Objective: To assess the impact on mortality of high vs low plasma: RBC ratio transfusion, in civilian trauma patients with massive bleeding. Methods: A systematic review and meta-analysis, including observational studies and clinical trials, was conducted. Databases were systemically searched for relevant studies between January 2007 and June 2019. The main outcome was early (24-hours) and late (30-day) mortality. Fixed and random effects models were used. Results: Out of 1295 studies identified, 33 were selected: 2 clinical trials and 31 observational studies. The analysis of observational trials showed both decreased early mortality (odds ratio [OR] 0.67; 95% confidence interval [CI], 0.60–0.75) and late mortality (OR 0.79; 95% CI, 0.71–0.87) with the use of high plasma:RBC ratio transfusion, but there were no differences when clinical trials were evaluated (OR 0.89; 95% CI, 0.64–1.26). The exclusion of patients who died within the first 24hours was a source of heterogeneity. The Injury Severity Score (ISS) altered the association between high plasma: RBC ratio and mortality, with a reduced protective effect when the ISS was high. Conclusion: The use of high vs low plasma: RBC ratio transfusion, in patients with massive bleeding due to civil trauma, has a protective effect on early and late mortality in observational studies. The exclusion of patients who died within the first 24 hours was a source of heterogeneity.

Optimal Hemodynamic Parameter to Predict the Neurological Outcome in Out-of-Hospital Cardiac Arrest Survivors Treated with Target Temperature Management

Current guidelines suggest the maintenance of systolic blood pressure (SBP) at >90 mmHg and mean arterial pressure (MAP) at >65 mmHg in postcardiac arrest patients. There remains a lack of clarity regarding optimal values and timing of blood pressure parameters associated with the improvement of neurologic outcome. We investigated the association of time-weighted average (TWA) blood pressure parameters with favorable neurological outcome (FO) in postcardiac arrest patients. This was a registry-based observational study with consecutive adult out-of-hospital cardiac arrest (OHCA) survivors who were treated using targeted temperature management (TTM). During 72 hours of TTM period, we abstracted hemodynamic parameters such as SBP, diastolic blood pressure, pulse rate (PR), and MAP. Shock index (SI; PR/SBP) and modified shock index (MSI; PR/MAP) were calculated from each measured hemodynamics. Logistic regression was performed to assess the associations between TWA blood pressure parameters and FO, defined as cerebral performance category 1 or 2 at hospital discharge. Among the 173 patients (median age: 58 years; 64% male), 51 (29.3%) had FO in this study. MAP, SI, and MSI at 6 hours after return of spontaneous circulation (ROSC) showed considerable differences in patients with FO (MAP: 89.1 ± 14.7 vs. 83.6 ± 15.8 mmHg, p = 0.033, SI: 0.7 ± 0.2 vs. 0.9 ± 0.9, p = 0.002, MSI: 1.0 ± 0.3 vs. 1.2 ± 0.3, p ≤ 0.001). Among them, MSI, especially at 6 hours, had the highest area under the curve for prediction of FO (0.685; 95% confidence interval: 0.597-0.772, p < 0.001). Also, MSI <1.0 had a sensitivity of 64.7%, a specificity of 64.2% to predict FO. In comatose survivors of OHCA with TTM, MSI at 6 hours after ROSC had the highest prognostic value for neurologic outcome among blood pressure parameters.

Biomarkers in Shock Patients and Their Value as A Prognostic Tool; A Prospective Multi-Center Cohort Study

Objective

To investigate the prognostic value of clinical and laboratory tests in prediction of outcome in patients at day 30 post presentation to hospital with shock and to determine the prognostic value of mid regional pro-adrenomedullin (MR-proADM) on mortality prediction at 30 days in the same patient cohort.

Method

This prospective multicenter cohort study analyzed data from patients who had presenting with shock to the emergency departments of eleven urban, tertiary-care University hospitals in Spain between March, 2011 and May, 2011. Recruitment of patients was via convenience sampling. Inclusion criteria included age between 14 and 100 years with clinical diagnostic criteria of shock on admission. Various patient parameters were analysed, such as age, sex, past medical history. Other clinical variables were measured on arrival to hospital, including sequential organ failure assessment score (score SOFA), blood pressure, oxygen saturations, capillary refill time and shock index (SI). Laboratory variables investigated included base excess, MR-proADM, lactate, C-Reactive Protein (CRP) and procalcitonin (PCT).

Results

There were 212 patients included in the study from the eleven hospitals involved. The mean age was 72.2 years old and 60.4% of the patients were men. In the discriminant analysis only age, MR-proADM and PCT remained in the final discriminant equation. The separate analysis of MR-proADM showed that, in the non-survivors group, MR-proADM levels are significantly higher than those found in the group of survivors (p<0.001).

Conclusion

Age, PCT and MR-proADM were useful to predict short-term mortality in patients presenting to the emergency department shock. This suggests that PCT and MR-proADM in combination with the most common prediction models will improve prognostic value.

Shock index increase from the field to the emergency room is associated with higher odds of massive transfusion in trauma patients with stable blood pressure: A cross-sectional analysis

Background

The shock index (SI) is defined as the ratio of heart rate/systolic blood pressure. This study aimed to determine the performance of delta shock index (ΔSI), a difference between SI upon arrival at the emergency room (ER) and that in the field, in predicting the need for massive transfusion (MT) among adult trauma patients with stable blood pressure.

Methods

This study included registered data from all trauma patients aged 20 years and above who were hospitalized from January 1, 2009 to December 31, 2016. Only patients who were transferred by emergency medical service from the accident site with a systolic blood pressure ≥ 90 mm Hg at the ER were included. The 7,957 enrolled trauma patients were divided into 2 groups, those who had received blood transfusion ≥ 10 U (MT, n = 82) and those who had not (non-MT, n = 7,875). The odds ratios with 95% confidence intervals for the need for MT by a given ΔSI were measured. The plot of specific receiver operating characteristic (ROC) curves was used to evaluate the best cutoff point of ΔSI that could predict the patient’s probability of receiving MT.

Results

ROC curve analysis showed that a ΔSI of 0.06 as the cutoff point had the highest AUC of 0.61, with a sensitivity of 0.415 and specificity of 0.841. Patients with a ΔSI ≥ 0.00 had a significant 1.8-fold increase in need for MT than those patients with a ΔSI less than 0.00 (1.4% vs. 0.8%, p = 0.01). The larger the ΔSI, the higher the odds of need for an MT. Using the cutoff point of ΔSI of 0.06, patients with a ΔSI ≥ 0.06 had a significant 3.7-fold increase in need for MT than those patients with a ΔSI less than 0.06 (2.7% vs. 0.7%, p < 0.001).

Conclusions

This study indicated that, in trauma patients with stable blood pressure at the ER, the accuracy of prediction of the requirement for MT by ΔSI is low. However, the size of the delta is significantly associated with need for MT and a lack of improvement in the patient’s SI at the ER compared to that in the field significantly increases the odds of a need for MT.