Identifying trigger cues for hospital blood transfusions based on ensemble of machine learning methods.

BACKGROUND: Traumatic shock is the leading cause of preventable death with most patients dying within the first six hours from arriving to the hospital. This underscores the importance of prehospital interventions, and growing evidence suggests prehospital transfusion improves survival. Optimizing transfusion triggers in the prehospital setting is key to improving outcomes for patients in hemorrhagic shock. Our objective was to identify factors associated with early in-hospital transfusion requirements available to prehospital clinicians in the field to develop a simple algorithm for prehospital transfusion, particularly for patients with occult shock. METHODS: We included trauma patients transported by a single critical care transport service to a level I trauma center between 2012 and 2019. We used logistic regression, Fast and Frugal Trees (FFTs), and Bayesian analysis to identify factors associated with early in-hospital blood transfusion as a potential trigger for prehospital transfusion. RESULTS: We included 2,157 patients transported from the scene or emergency department (ED) of whom 207 (9.60%) required blood transfusion within four hours of admission. The mean age was 47 (IQR = 28 - 62) and 1,480 (68.6%) patients were male. From 13 clinically relevant factors for early hospital transfusions, four were incorporated into the FFT in following order: 1) SBP, 2) prehospital lactate concentration, 3) Shock Index, 4) AIS of chest (sensitivity = 0.81, specificity = 0.71). The chosen thresholds were similar to conventional ones. Using conventional thresholds resulted in lower model sensitivity. Consistently, prehospital lactate was among most decisive factors of hospital transfusions identified by Bayesian analysis (OR = 2.31; 95% CI 1.55 - 3.37). CONCLUSIONS: Using an ensemble of frequentist statistics, Bayesian analysis and machine learning, we developed a simple, clinically relevant prehospital algorithm to help identify patients requiring transfusion within 4 h of hospital arrival.

Identifying Trigger Cues for Hospital Blood Transfusions Based on Ensemble Learning Methods.

Background: Traumatic shock is the leading cause of preventable death with most patients dying within the first 6 hours. This underscores the importance of prehospital interventions, and growing evidence suggests prehospital transfusion improves survival. Optimizing transfusion triggers in the prehospital setting is key to improving outcomes for patients in hemorrhagic shock. Our objective was to identify factors associated with early in-hospital transfusion requirements available to prehospital clinicians in the field to develop a simple algorithm for prehospital transfusion, particularly for patients with occult shock. Methods: We included trauma patients transported by a single critical care transport service to a level I trauma center between 2012 and 2019. We used logistic regression, Fast and Frugal Trees (FFTs), and Bayesian analysis to identify factors associated with early in-hospital blood transfusion as a potential trigger for prehospital transfusion. Results: We included 2,157 patients transported from the scene or emergency department (ED) of whom 207 (9.60%) required blood transfusion within 4 hours of admission. The mean age was 47 (IQR = 28-62) and 1,480 (68.6%) patients were male. From 13 clinically relevant factors for early hospital transfusions, four were incorporated into the FFT in following order: 1) SBP, 2) prehospital lactate concentration, 3) Shock Index, 4) AIS of chest (sensitivity = 0.81, specificity = 0.71). The chosen thresholds were similar to conventional ones. Using conventional thresholds resulted in lower model sensitivity. Consistently, prehospital lactate was among most decisive factors of hospital transfusions identified by Bayesian analysis (OR = 2.31; 95% CI 1.55-3.37). Conclusions: Using an ensemble of frequentist statistics, Bayesian analysis and machine learning, we developed a simple, clinically relevant, prehospital algorithm to help identify patients requiring transfusion within 4 hours of hospital arrival.

Prehospital Lactate is Associated with the Need for Blood in Trauma.

Background: Traumatic hemorrhage is the leading cause of preventable death, and its effects are often evident within the first 24 hours of hospital admission. We investigated the relationship between prehospital lactate measurement and administration of hospital blood products and life-saving interventions (LSIs) within 24 hours of hospital admission.Methods: We included trauma patients with recorded prehospital venous lactate transported by a single critical care transport service to a Level I trauma center between 2012 and 2019. We abstracted vital signs, mission type, anatomic location of injury, prehospital administration of crystalloid and blood products, and hospital LSIs started within 24 hours of admission. We used logistic regression to determine the association of prehospital lactate and in-hospital administration of blood products in groups with or without hypotensive patients. We investigated the effect of prehospital lactate concentration on secondary outcomes such as LSIs and mortality.Results: We included 2,170 patients transported from the scene or emergency department (ED), of whom 1,821 (84%) were normotensive. The median concentration of prehospital lactate was 2.10 mmol/L for the main population (IQR = 1.40-3.30) and 2.00 mmol/L for the normotensive subgroup (IQR = 1.30-2.90). A higher prehospital lactate concentration was associated with higher odds of needing early hospital blood products in the whole study population (OR = 1.12, (95% CI 1.06-1.20), p < 0.01) and in the normotensive subgroup (OR = 1.13, (95% CI 1.03-1.22), p = 0.01). These positive associations were also observed with the secondary outcome of hospital LSIs, and higher prehospital lactate was also associated with higher odds of mortality (OR = 1.32, (95% CI 1.20-1.45), p < 0.01).Conclusions: Higher concentrations of prehospital lactate were associated with the need for in-hospital blood transfusion within 24 hours of admission. The relationship between lactate and blood transfusion persisted among normotensive patients. Further work is needed to incorporate prehospital lactate into decision support tools for prehospital blood administration.