This study was carried out to estimate the relationship between arterial Pco₂ (Paco₂) and end-tidal carbon dioxide (Petco₂) during prehospital controlled ventilation and also to evaluate variation of the gradient between Pco₂ and Petco₂ during prehospital transport.

Measurements of Petco₂ from capnography values and Paco₂ from arterial blood gases were registered at the beginning (T₀) and at the end (T_end) of out-of-hospital management. For all patients requiring invasive ventilation, the gradient between Pco₂ and Petco₂ was calculated for T₀ and T_end, the Paco₂ − Petco₂ variation between T_end and T₀ was also calculated.

One hundred patients were included in this study (mean age, 58.4 ± 16.4 years; 57 were male). There was no variation of the mean gradient (ΔPaco₂ − Petco₂) during transport (8.64 ± 13.5 mm Hg at T₀ and 7.26 ± 12.94 mm Hg at T_end). Thirty-six percent of patients (n = 36) had a gradient above +10 mm Hg, and for 6% of patients (n = 4) the gradient was lower than −10 mm Hg. The Paco₂ − Petco₂ gradient was not significantly different according to the pathology, but was significantly higher in hypercapnic patients compared with hypocapnic or normocapnic patients. In patients with severe head injury, the capnia was normalized in 80% of patients at the end of the transport according to the last blood gas result. In this subgroup the ΔPaco₂ − Petco₂ (T_end − T₀) gradient was stable between T₀ and T_end except in 20% of the patients for whom the ΔPaco₂ − Petco₂ was lower than −10 mm Hg. Fifty-four percent of critical care physicians had modified the respiratory setting after the first arterial blood gas results.

The Paco₂ cannot be estimated by the Petco₂ in the prehospital setting. There is wide variation in the gradient between Pco₂ and Petco₂ depending on patient condition, and over time, the relationship does not remain constant and thus cannot be useful in prehospital ventilation management.