BACKGROUND: High-flow nasal oxygen (HFNO) is an emerging technology that has generated interest in tubeless anesthesia for airway surgery. HFNO has been shown to maintain oxygenation and CO2 clearance in spontaneously breathing patients and is an effective approach to apneic oxygenation. Although it has been suggested that HFNO can enhance CO2 clearance during apnea, this has not been established. The true extent of CO2 accumulation and resulting acidosis using HFNO during prolonged tubeless anesthesia remains undefined. METHODS: In a single-center trial, we randomly assigned 20 adults undergoing microlaryngoscopy to apnea or spontaneous ventilation (SV) using HFNO during 30 minutes of tubeless anesthesia. Serial arterial blood gas analysis was performed during preoxygenation and general anesthesia. The primary outcome was the partial pressure of CO2 (Paco 2) after 30 minutes of general anesthesia, with each group compared using a Student t test. RESULTS: Nineteen patients completed the study protocol (9 in the SV group and 10 in the apnea group). The mean (standard deviation [SD]) Paco 2 was 89.0 mm Hg (16.5 mm Hg) in the apnea group and 55.2 mm Hg (7.2 mm Hg) in the SV group (difference in means, 33.8; 95% confidence interval [CI], 20.6–47.0) after 30 minutes of general anesthesia (P < .001). The average rate of Paco 2 rise during 30 minutes of general anesthesia was 1.8 mm Hg/min (SD = 0.5 mm Hg/min) in the apnea group and 0.8 mm Hg/min (SD = 0.3 mm Hg/min) in the SV group. The mean (SD) pH was 7.11 (0.04) in the apnea group and 7.29 (0.06) in the SV group (P < .001) at 30 minutes. Five (55%) of the apneic patients had a pH <7.10, of which the lowest measurement was 7.057. No significant difference in partial pressure of arterial O2 (Pao 2) was observed after 30 minutes of general anesthesia. CONCLUSIONS: CO2 accumulation during apnea was more than double that of SV after 30 minutes of tubeless anesthesia using HFNO. The use of robust measurement confirms that apnea with HFNO is limited by CO2 accumulation and the concomitant severe respiratory acidosis, in contrast to SV. This extends previous knowledge and has implications for the safe application of HFNO during prolonged procedures.