Objective: It is important for preventing hyperventilation so as to improve the outcomes of patients with severe traumatic brain injuries. End-tidal CO2 (Petco2) reflects arterial CO2 (Paco2), noninvasively and continuously. The aim of this study is to examine the concordance between Paco2 and Petco2 among adults presenting to the emergency department with severe traumatic brain injury and to elucidate the patient groups who had discordance.
Methods: From July 1, 2006 to December 31, 2007, 77 patients who presented in the emergency department of the hospital after severe brain injury were enrolled. Glasgow Coma Scale, blood pressure, heart rate, respiration rate, body temperature, arterial blood gas variables, and serum lactate levels were measured under the mechanical ventilation support. Finally, Injury Severity Score and Abbreviated Injury Scale of head and chest were recorded. The concordance between Petco2 and Paco2 was analyzed using Bland-Altman plot. We defined that normal gap of Paco2 and Petco2 [P(a-et)co2] is -5 mm Hg to 5 mm Hg, and compared the normal and high P(a-et)co2 groups to find the factors affecting the P(a-et)co2.
Results: Sixty-six patients after exception of 11 patients with nontraumatic brain injury were analyzed. Ten patients (15.2%) were below 30 mm Hg of Paco2. Pearson's correlation coefficient between Paco2 and Petco2 was 0.666 (p < 0.001) and the concordance between Paco2 and Petco2 was 77.3%. The patients with high P(a-et)co2 showed significantly higher Injury Severity Score, lower systolic blood pressure, lower arterial pH, lower base deficit, and higher serum lactate levels than patients with normal P(a-et)co2. Patients with severe chest trauma, hypotension, and metabolic acidosis increased the risk of the discordance between Paco2 and Petco2. Twenty-eight patients who had no hypotension, no metabolic acidosis, and no severe chest trauma showed concordance rate of 100.0% between Paco2 and Petco2.
Conclusions: Severe chest trauma and hemodynamic and tissue perfusion state should be considered when attempting to monitor the hyperventilation in severe brain injury patients using Petco2.