Is AVPU comparable to GCS in critical prehospital decisions? – A cross-sectional study

• Published in: The American journal of emergency medicine Vol. 59; pp. 106 - 110
• Main Authors: Janagama, Srinivasa R., Newberry, Jennifer A., Kohn, Michael A., Rao, G.V. Ramana, Strehlow, Matthew C., Mahadevan, Swaminatha V.
• Format: Journal Article
• Language: English
• Published: Philadelphia Elsevier Inc 01.09.2022; Elsevier Limited
• Subjects: Ambulance services; AVPU; Clinical decision making; Consciousness; Cross-sectional studies; Data analysis; Emergency medical care; Fatalities; Field triage; Glasgow coma scale; Intensive care; Mortality; Pain; Patients; Prehospital AVPU; Quality improvement; Statistical analysis; Trauma; Trauma care; Unconsciousness; India; Prehospital AVPU; Glasgow coma scale; AVPU; Field triage; Trauma
• DOI: 10.1016/j.ajem.2022.06.042
• ISSN: 0735-6757

Advanced Trauma Life Support field triage utilizes the Glasgow Coma Scale (GCS) to assess the level of consciousness. However, prehospital care providers in low- and middle-income countries (LMICs) often use the Alert, Verbal, Pain, and Unresponsive (AVPU) scale to assess the level of consciousness. This study aimed to determine whether prehospital AVPU categorization correlates with mortality rates in trauma victims, similarly to GCS. In this cross-sectional study conducted between November 2015 and January 2016, we enrolled a convenience sample of prehospital trauma-related field activations. The primary outcome measure was the probability of death within 48 h for each category of AVPU. In a convenience sample of 4514 activations, 1606 (35.6%) met exclusion criteria, four did not have AVPU, and four did not have GCS, leaving 2900 (64.2%) trauma activations with both AVPU and GCS available for analysis. Forty-eight-hour follow-up data were available for 2184 (75.3%) activations out of these 2900. The 48-h mortality rates for each category of AVPU were 1.1% (Alert), 4.3% (Verbal), 17.9% (Pain), 53.2% (Unresponsive); and, for each GCS-based injury severity category, they were 0.9% (Mild, GCS 13–15), 8.1% (Moderate, GCS 9–12), 43.5% (Severe, GCS ≤ 8). Overall, there was a statistically significant difference in GCS for each category of AVPU (p < 0.001) except between patients responding to verbal commands and those responding to pain (p = 0.18). The discriminative ability of AVPU (AUC 79.7% (95% CI 73.4–86.1)) and GCS (AUC 81.5% (95% CI 74.8–88.2)) for death within 48-h following hospital drop-off were comparable. EMT assessments of AVPU and GCS relate to each other, and AVPU predicts mortality at 48 h. Future studies using AVPU to assess the level of consciousness in prehospital trauma protocols may simplify their global application without impacting the overall quality of care.