The effect of high‐flow nasal oxygen flow rate on gas exchange in apnoeic patients: a randomised controlled trial

Abstract

SummaryHigh‐flow nasal oxygen can be administered at induction of anaesthesia for the purposes of pre‐oxygenation and apnoeic oxygenation. This intervention is claimed to enhance carbon dioxide elimination during apnoea, but the extent to which this occurs remains poorly quantified. The optimal nasal oxygen flow rate for gas exchange is also unknown. In this study, 114 patients received pre‐oxygenation with high‐flow nasal oxygen at 50 l.min‐1. At the onset of apnoea, patients were allocated randomly to receive one of three nasal oxygen flow rates: 0 l.min‐1; 70 l.min‐1; or 120 l.min‐1. After 4 minutes of apnoea, all oxygen delivery was ceased, tracheal intubation was performed, and oxygen delivery was recommenced when SpO2 was 92%. Mean (SD) PaCO2 rise during the first minute of apnoea was 1.39 (0.39) kPa, 1.41 (0.29) kPa, and 1.26 (0.38) kPa in the 0 l.min‐1, 70 l.min‐1 and 120 l.min‐1 groups, respectively; p = 0.16. During the second, third and fourth minutes of apnoea, mean (SD) rates of rise in PaCO2 were 0.34 (0.08) kPa.min‐1, 0.36 (0.06) kPa.min‐1 and 0.37 (0.07) kPa.min‐1 in the 0 l.min‐1, 70 l.min‐1 and 120 l.min‐1 groups, respectively; p = 0.17. After 4 minutes of apnoea, median (IQR [range]) arterial oxygen partial pressures in the 0 l.min‐1, 70 l.min‐1 and 120 l.min‐1 groups were 24.5 (18.6–31.4 [12.3–48.3]) kPa; 36.6 (28.1–43.8 [9.8–56.9]) kPa; and 37.6 (26.5–45.4 [11.0–56.6]) kPa, respectively; p < 0.001. Median (IQR [range]) times to desaturate to 92% after the onset of apnoea in the 0 l.min‐1, 70 l.min‐1 and 120 l.min‐1 groups, were 412 (347–509 [190–796]) s; 533 (467–641 [192–958]) s; and 531 (462–681 [326–1007]) s, respectively; p < 0.001. In conclusion, the rate of carbon dioxide accumulation in arterial blood did not differ significantly between apnoeic patients who received high‐flow nasal oxygen and those who did not.