Helicopters and Injured Kids: Improved Survival with Scene Air - PowerPoint PPT Presentation

Helicopters and Injured Kids: Improved Survival with Scene Air Medical Transport in the Pediatric Trauma Population JB Brown, CM Leeper, JL Sperry, AB Peitzman, TR Billiar, BA Gaines & ML Gestring

Disclosures No conflicts to disclose

Background • Helicopter emergency medical services (HEMS) often transport injured children • Need to justify increased risk and cost compared to ground emergency medical services (GEMS) • Unclear evidence of benefit to support HEMS use in pediatric trauma patients

Objective Evaluate association of HEMS compared to GEMS transport with outcomes in pediatric trauma patients

Methods: Data Source & Population • National Trauma Databank 2007-2012 • Age ≤ 15 years • Transported from scene of injury by HEMS or GEMS • Excluded: Burns & DOA • Multiple imputation for missing data

Methods: Propensity Score Matching • Selection bias  HEMS patients more severely injured • Propensity score to predict likelihood of undergoing HEMS • Propensity score variables:  Age, sex, MOI, age-adjusted prehospital vital signs, prehospital time, ISS, pediatric level I center available, geographic region

Methods: Propensity Score Matching • Nearest neighbor 1:1 matching • Stratified matching within age groups  Infant/toddler (≤2 years)  Child (3 – 12 years)  Adolescent (>12 years) • Standardized differences to assess balance  |Std diff| <0.1  balance between groups

Methods: Outcomes • Primary outcome  In-hospital survival • Secondary outcome  Discharge disposition in survivors  Discharge to home vs.  Discharge to rehab or need for additional services

Methods: Analysis • Conditional logistic regression  HEMS vs GEMS  Adjusted for demographics, age-adjusted ED vital signs, ICU resources, urgent OR, NAT, complications  Accounting for clustering within centers

Methods: Subgroup Analysis • Logistic considerations for HEMS  proximity to trauma center • Evaluated patients with transport time >15min  Capture patients with potential to undergo HEMS transport • Repeated models for survival & discharge disposition

Study Population NTDB 07- 12 Age≤15 N=541,875 Non-HEMS/GEMS transport Transfer N=206,808 N=162,470 Burn injury DOA N=5,726 N=277 Study Population N=166,594 HEMS GEMS N=25,837 N=140,757 Treated Matched Control Matched N=25,700 (>99%) N=25,700 (18%)

Results: HEMS v GEMS Pre-Match HEMS GEMS p value N=25,837 N=140,757 Age (years) 10 (5, 14) 10 (4, 13) <0.01 <0.01 Mechanism (% blunt) 96 94 <0.01 Prehospital hypotension (%) 37 35 Prehospital abnormal HR (%) 21 17 <0.01 Prehospital abnormal RR (%) 34 33 <0.01 <0.01 Prehospital GCS 15 (12, 15) 15 (15, 15) <0.01 Prehospital time (min) 60 (47, 79) 42 (31, 59) <0.01 ISS 9 (5, 17) 5 (4, 9) Level I pediatric center (%) <0.01 50 37 Geographic region (%) <0.01 Midwest 15 25 - Northeast 11 15 - South 45 33 - West 29 27 -

Results: HEMS v GEMS Pre-Match HEMS GEMS p value N=25,837 N=140,757 Age (years) 10 (5, 14) 10 (4, 13) <0.01 <0.01 Mechanism (% blunt) 96 94 <0.01 Prehospital hypotension (%) 37 35 Prehospital abnormal HR (%) 21 17 <0.01 Prehospital abnormal RR (%) 34 33 <0.01 <0.01 Prehospital GCS 15 (12, 15) 15 (15, 15) <0.01 Prehospital time (min) 60 (47, 79) 42 (31, 59) <0.01 ISS 9 (5, 17) 5 (4, 9) Level I pediatric center (%) <0.01 50 37 Geographic region (%) <0.01 Midwest 15 25 - Northeast 11 15 - South 45 33 - West 29 27 -

Results: HEMS v GEMS Post-Match ISS Prehospital GCS Level I pediatric center South region Midwest region Northeast region Anatomic triage criteria Blunt mechanism Prehospital tachycardia Age Prehospital hypotension West region Prehospital apnea Sex Prehospital Time Prehospital tachypnea Prehospital bradycardia 0.0 0.1 0.2 0.3 0.4 0.5 Absolute Standardized Difference Before Matching After Matching

Results: HEMS v GEMS Post-Match ISS Prehospital GCS Level I pediatric center South region Midwest region Northeast region Anatomic triage criteria Blunt mechanism Prehospital tachycardia Age Prehospital hypotension West region Prehospital apnea Sex Prehospital Time Prehospital tachypnea Prehospital bradycardia 0.0 0.1 0.2 0.3 0.4 0.5 Absolute Standardized Difference Before Matching After Matching

Results: HEMS v GEMS Post-Match HEMS GEMS Standardized N=25,700 N=25,700 Difference Age (years) 10 (5, 14) 10 (5, 14) -0.002 Mechanism (% blunt) 96 95 0.027 Prehospital hypotension (%) 49 49 -0.006 Prehospital abnormal HR (%) 28 28 -0.011 Prehospital abnormal RR (%) 49 49 -0.007 Prehospital GCS 15 (13, 15) 15 (13, 15) -0.006 Prehospital time (min) 57 (43, 75) 44 (31, 62) -0.023 ISS 9 (5, 17) 9 (4, 17) -0.002 ICU admission (%) 40 31 0.191* Urgent OR (%) 15 13 0.068 Mechanical ventilation (%) 18 14 0.122* Survival (%) 96 94 0.081 * Indicates residual imbalance after matching

Results: HEMS v GEMS Post-Match HEMS GEMS Standardized N=25,700 N=25,700 Difference Age (years) 10 (5, 14) 10 (5, 14) -0.002 Mechanism (% blunt) 96 95 0.027 Prehospital hypotension (%) 49 49 -0.006 Prehospital abnormal HR (%) 28 28 -0.011 Prehospital abnormal RR (%) 49 49 -0.007 Prehospital GCS 15 (13, 15) 15 (13, 15) -0.006 Prehospital time (min) 57 (43, 75) 44 (31, 62) -0.023 ISS 9 (5, 17) 9 (4, 17) -0.002 ICU admission (%) 40 31 0.191* Urgent OR (%) 15 13 0.068 Mechanical ventilation (%) 18 14 0.122* Survival (%) 96 94 0.081 * Indicates residual imbalance after matching

Results: HEMS v GEMS Post-Match HEMS GEMS Standardized N=25,700 N=25,700 Difference Age (years) 10 (5, 14) 10 (5, 14) -0.002 Mechanism (% blunt) 96 95 0.027 Prehospital hypotension (%) 49 49 -0.006 Prehospital abnormal HR (%) 28 28 -0.011 Prehospital abnormal RR (%) 49 49 -0.007 Prehospital GCS 15 (13, 15) 15 (13, 15) -0.006 Prehospital time (min) 57 (43, 75) 44 (31, 62) -0.023 ISS 9 (5, 17) 9 (4, 17) -0.002 ICU admission (%) 40 31 0.191* Urgent OR (%) 15 13 0.068 Mechanical ventilation (%) 18 14 0.122* Survival (%) 96 94 0.081 * Indicates residual imbalance after matching

Results: HEMS v GEMS 1.72 1.10

Results: HEMS v GEMS 1.72 1.10

Results: HEMS v GEMS 1.72 NNT = 41 (95%CI 29 – 84) 1.10

Results: HEMS v GEMS 1.72 1.10

Results: HEMS v GEMS >15min 1.81 1.07

Results: HEMS v GEMS >15min 1.81 1.07

Results: HEMS v GEMS >15min 1.81 NNT = 38 (95%CI 27 – 89) 1.07

Results: HEMS v GEMS >15min 1.81 1.07

Summary • HEMS transport independently associated with survival  Transport 41 additional patients by HEMS to save one life • HEMS not associated with discharge disposition • Findings persist in patients with >15min transport time

Limitations • Retrospective • NTDB limitations • Unknown prehospital interventions • Selection bias  Propensity score matching

Conclusions • Injured pediatric patients undergoing HEMS scene transport have improved survival • Further study warranted  Investigate underlying mechanisms  Develop specific triage criteria for HEMS transport in pediatric trauma patients