Air medical response to traumatic brain injury: a computer learning algorithm analysis.

BACKGROUND: The role of air medicine in traumatic brain injury (TBI) has been studied extensively using trauma registries but remains unclear. Learning algorithms, such as artificial neural networks (ANN), support vector machines (SVM), and decision trees, can identify relationships between data set variables but are not empirically useful for hypothesis testing. OBJECTIVE: To use ANN, SVM, and decision trees to explore the role of air medicine in TBI. METHODS: Patients with Head Abbreviated Injury Score 3+ were identified from our county trauma registry. Predictive models were generated using ANN, SVM, and decision trees. The three best-performing ANN models were used to calculate differential survival values (actual and predicted outcome) for each patient. In addition, predicted survival values with transport mode artificially input as "air" or "ground" were calculated for each patient to identify those who benefit from air transport. For SVM analysis, chi was used to compare the ratio of unexpected survivors to unexpected deaths for air- and ground-transported patients. Finally, decision tree analysis was used to explore the indications for various transport modes in optimized survival algorithms. RESULTS: A total of 11,961 patients were included. All three learning algorithms predicted a survival benefit with air transport across all patients, especially those with higher Head Abbreviated Injury Score or Injury Severity Score values, lower Glasgow Coma Scale scores, or hypotension. CONCLUSION: Air medical response in TBI seems to confer a survival advantage, especially in more critically injured patients.

Head-injured patients who "talk and die": the San Diego perspective.

BACKGROUND: Head-injured patients who "talk and die" are potentially salvageable, making their early identification important. This study uses a large, comprehensive database to explore risk factors for head-injured patients who deteriorate after their initial presentation. METHODS: Patients with a head Abbreviated Injury Score (AIS) score of 3+ and a preadmission verbal Glasgow Coma Scale (GCS) score of 3+ were identified from our county trauma registry during a 16-year period. Survivors and nonsurvivors were compared with regard to demographics, initial clinical presentation, and various risk factors. Logistic regression was used to explore the impact of multiple factors on outcome, including the significance of a change in GCS score from field to arrival. In addition, patients were stratified by injury severity and hospital day of death to further define the relationship between outcome and multiple clinical variables. RESULTS: A total of 7,443 patients were identified with head AIS 3+ and verbal GCS score 3+. Overall mortality was 6.1%. About one-third of deaths occurred on the first hospital day, with more than one-third occurring after hospital day 5. Logistic regression revealed an association between mortality and older age, more violent mechanisms of injury (fall, gunshot wound, pedestrian versus automobile), greater injury severity (higher head AIS and Injury Severity Score), lower GCS score, and hypotension. In addition, mortality was associated with a decrease in GCS score from field to arrival, the use of anticoagulants, and a diagnosis of pulmonary embolus. Two important groups of "talk-and-die" patients were identified. Early deaths occurred in younger patients with more critical extracranial injuries. Anticoagulant use was also an independent risk factor in these early deaths. Later deaths occurred in older patients with less significant extracranial injuries. Pulmonary embolus also appeared to be an important contributor to late mortality. CONCLUSIONS: More severe injuries and use of anticoagulants are independent risk factors for early death in potentially salvageable traumatic brain injury patients, whereas older age and pulmonary embolus are associated with later deaths.

The predictive value of field versus arrival Glasgow Coma Scale score and TRISS calculations in moderate-to-severe traumatic brain injury.

BACKGROUND: Glasgow Coma Scale (GCS) scores are widely used to quantify level of consciousness in the prehospital environment. The predictive value of field versus arrival GCS is not well defined but has tremendous implications with regard to triage and therapeutic decisions as well as the use of various predictive scoring systems, such as Trauma Score and Injury Severity Score (TRISS). This study explores the predictive value of field GCS (fGCS) and arrival GCS (aGCS) as well as TRISS calculations using field (fTRISS) and arrival (aTRISS) data in patients with moderate-to-severe traumatic brain injury (TBI). METHODS: Major trauma victims with head Abbreviated Injury Scores of 3 or greater were identified from our county trauma registry over a 16-year period. The predictive ability of fGCS with regard to aGCS was explored using univariate statistics and linear regression modeling. The difference between aGCS and fGCS was also modeled against mortality and the composite endpoint using logistic regression, adjusting for fGCS. The predictive value of preadmission GCS (pGCS), defined as either fGCS or aGCS in nonintubated patients without a documented fGCS, with regard to mortality and a composite endpoint representing the need for neurosurgical care (death, craniotomy, invasive intracranial pressure monitoring, or intensive care unit care >48 hours) was determined using receiver-operator curve (ROC) analysis. Finally, fTRISS and aTRISS predicted survival values were compared with each other and to observed survival. RESULTS: A total of 12,882 patients were included. Mean values for fGCS and aGCS were similar (11.4 and 11.5, respectively, p = 0.336), and a strong correlation (r = 0.67, 95% CI 0.66-0.69, p < 0.0001) was observed between them. The difference between fGCS and aGCS was also predictive of outcome after adjusting for fGCS. Good predictive ability was observed for pGCS with regard to both mortality and neurosurgical intervention. Both fTRISS and aTRISS predicted survival values were nearly identical to observed survival. Observed and fTRISS predicted survival were nearly identical in patients undergoing prehospital intubation CONCLUSIONS: Values for fGCS are highly predictive of aGCS, and both are associated with outcome from TBI. A change in GCS from the field to arrival is highly predictive of outcome. The use of field data for TRISS calculations appears to be a valid methodological approach, even in severely injured TBI patients undergoing prehospital intubation.

Traumatic brain injury outcomes in pre- and post- menopausal females versus age-matched males.

Gender differences in outcomes from major trauma have been described previously, and exogenous female hormone administration appears to be neuroprotective following traumatic brain injury (TBI). This analysis explored outcomes in pre- and post-menopausal females versus age-matched males. A total of 13,437 patients (n = 3,178 females, n = 10,259 males) with moderate-to-severe TBI (head AIS > or = 3) were identified from our county trauma registry. Overall mortality was similar between males and females (22% for both). Logistic regression was used to compare gender outcome differences, with a separate analysis performed for premenopausal (< 50 years) versus postmenopausal (> or = 50 years) patients, and after stratification by decade of life. No statistically significant difference in outcomes was observed for pre-menopausal females versus males (odds ratio [OR] 1.06; 95% confidence interval [CI] 0.83, 1.35; p = 0.633), but outcomes were significantly better in postmenopausal females versus males (OR 0.63, 95% CI 0.48-0.81, p < 0.001) after adjusting for age, mechanism of injury, Glasgow Coma Scale (GCS), hypotension (SBP < or = 90 mm Hg), head Abbreviated Injury Score (AIS), and Injury Severity Score (ISS). Stratification by decade of life revealed the gender survival differential inflection point to occur between ages 40-49 (OR 1.06, 95% CI 0.66-1.71, p = 0.798) and ages 50-59 (OR 0.38, 95% CI 0.20-0.74, p = 0.005). In addition, Revised Trauma Score and Injury Severity Score (TRISS) was used to calculate probability of survival (PS); all patients were then stratified by decade of life, and males and females were compared with regard to mean survival differential (outcome - PS). The identical pattern of improved outcomes in post-menopausal but not pre-menopausal females versus age-matched males was observed. These data suggest that endogenous female sex hormone production is not neuroprotective.

Early ventilation and outcome in patients with moderate to severe traumatic brain injury.

OBJECTIVES: An increase in mortality has been reported with early intubation in severe traumatic brain injury, possibly due to suboptimal ventilation. This analysis explores the impact of early ventilation on outcome in moderate to severe traumatic brain injury. DESIGN: Retrospective, registry-based analysis. SETTING: This study was conducted in a large county trauma system that includes urban, suburban, and rural jurisdictions. PATIENTS: Nonarrest trauma victims with a Head Abbreviated Injury Score of > or =3 were identified from our county trauma registry. INTERVENTIONS: Intubated patients were stratified into 5 mm Hg arrival PCO(2) increments. Logistic regression was used to calculate odds ratios for each increment, adjusting for age, gender, mechanism of injury, year of injury, preadmission Glasgow Coma Scale score, hypotension, Head Abbreviated Injury Score, Injury Severity Score, PO(2), and base deficit. Increments with the highest relative survival were used to define the optimal PCO(2) range. Outcomes for patients with arrival PCO(2) values inside and outside this optimal range were then explored for both intubated and nonintubated patients, adjusting for the same factors as defined previously. In addition, the independent outcome effect of hyperventilation and hypoventilation was assessed. MEASUREMENTS AND MAIN RESULTS: A total of 890 intubated and 2,914 nonintubated patients were included. Improved survival was observed for the arrival PCO(2) range 30-49 mm Hg. Patients with arrival PCO(2) values inside this optimal range had improved survival and a higher incidence of good outcomes. Conversely, there was no improvement in outcomes for patients within this optimal PCO(2) range for nonintubated patients after adjusting for all of the factors defined previously. Both hyperventilation and hypoventilation were associated with worse outcomes in intubated but not nonintubated patients. The proportion of arrival PCO(2) values within the optimal range was lower for intubated vs. nonintubated patients. CONCLUSIONS: Arrival hypercapnia and hypocapnia are common and associated with worse outcomes in intubated but not spontaneously breathing patients with traumatic brain injury.

The impact of aeromedical response to patients with moderate to severe traumatic brain injury.

STUDY OBJECTIVE: Aeromedical crews offer an advanced level of practice and rapid transport to definitive care; however, their efficacy remains unproven. Previous studies have used relatively small sample sizes or have been unable to adequately control for the effect of other potentially influential variables. Here we explore the impact of aeromedical response in patients with moderate to severe traumatic brain injury. METHODS: This was a retrospective analysis using our county trauma registry. All patients with head Abbreviated Injury Score of 3 or greater were included; interfacility transfers were excluded. The impact of aeromedical response was determined using logistic regression, adjusting for age, sex, mechanism, preadmission Glasgow Coma Scale score, head Abbreviated Injury Score, Injury Severity Score, and the presence of preadmission hypotension. Propensity scores were used to account for variability in selection of patients to undergo air versus ground transport. Patients with moderate and severe traumatic brain injury, as defined by head Abbreviated Injury Score and Glasgow Coma Scale score, were compared. Finally, aeromedical patients undergoing field intubation were compared with ground patients undergoing emergency department (ED) intubation. RESULTS: A total of 10,314 patients meeting all inclusion and exclusion criteria and with complete data sets were identified and included 3,017 transported by aeromedical crews. Overall mortality was 25% in the air- and ground-transported cohorts, but outcomes were significantly better for the aeromedical patients when adjusted for age, sex, mechanism of injury, hypotension, Glasgow Coma Scale score, head Abbreviated Injury Score, and Injury Severity Score (adjusted odds ratio [OR] 1.90; 95% confidence interval [CI] 1.60 to 2.25; P<.0001). Good outcomes (discharge to home, jail, psychiatric facility, rehabilitation, or leaving against medical advice) were also higher in aeromedical patients (adjusted OR 1.36; 95% CI 1.18 to 1.58; P<.0001). The primary benefit appeared to be in more severely injured patients, as reflected by head Abbreviated Injury Score and Glasgow Coma Scale score. Improved survival was also observed for air-transported patients intubated in the field versus ground-transported patients given emergency intubation in the ED (adjusted OR 1.42; 95% CI 1.13 to 1.78; P<.001). CONCLUSION: Here we analyze a large database of patients with moderate to severe traumatic brain injury. Aeromedical response appears to result in improved outcomes after adjustment for multiple influential factors in patients with moderate to severe traumatic brain injury. In addition, out-of-hospital intubation among air-transported patients resulted in better outcomes than ED intubation among ground-transported patients. Patients with more severe injuries appeared to derive the greatest benefit from aeromedical transport.

The impact of prehospital endotracheal intubation on outcome in moderate to severe traumatic brain injury.

BACKGROUND: Although early intubation to prevent the mortality that accompanies hypoxia is considered the standard of care for severe traumatic brain injury (TBI), the efficacy of this approach remains unproven. METHODS: Patients with moderate to severe TBI (Head/Neck Abbreviated Injury Scale [AIS] score 3+) were identified from our county trauma registry. Logistic regression was used to explore the impact of prehospital intubation on outcome, controlling for age, gender, mechanism, Glasgow Coma Scale score, Head/Neck AIS score, Injury Severity Score, and hypotension. Neural network analysis was performed to identify patients predicted to benefit from prehospital intubation. RESULTS: A total of 13,625 patients from five trauma centers were included; overall mortality was 22.9%, and 19.3% underwent prehospital intubation. Logistic regression revealed an increase in mortality with prehospital intubation (odds ratio, 0.36; 95% confidence interval, 0.32-0.42; p < 0.001). This was true for all patients, for those with severe TBI (Head/Neck AIS score 4+ and/or Glasgow Coma Scale score of 3-8), and with exclusion of patients transported by aeromedical crews. Patients intubated in the field versus the emergency department had worse outcomes. Neural network analysis identified a subgroup of patients with more significant injuries as potentially benefiting from prehospital intubation. CONCLUSION: Prehospital intubation is associated with a decrease in survival among patients with moderate-to-severe TBI. More critically injured patients may benefit from prehospital intubation but may be difficult to identify prospectively.