BACKGROUND: The monitoring of end-tidal CO2 tension (PETCO2) during high frequency jet ventilation (HFJV) has been unsatisfactory because of a small tidal volume and slow response time of CO2 analyser, although several authors have reported strategies of successful PETCO2 measurement during HFJV. The aim of this study was to assess the validity of tracheal CO2 tension (PtCO2) as a PaCO2 during HFJV. METHODS: We studied 24 patients undergoing laryngomicrosurgery during HFJV (rates: 100/min; I:E= 0.2; driving pressure: 0.25-0.35 MPa) through a 12 Fr. polyethylene injector placed 6-7 cm below the vocal cord. A gas sampling line was placed longitudinally against the injector and they were wrapped with aluminum foil. Continuous capnography was recorded during 20 minutes of HFJV. Every 5 minutes of HFJV, PtCO2 was obtained from the plateau value of CO2 wave after the stopping of JV and arterial blood gas analysis was done at 20 minutes of HFJV comparing PaCO2 to PtCO2. A Pearson's product moment correlation and regression analysis between PtCO2 and PaCO2 and the agreement between the two methods using Bland-Altman method were assessed. RESULTS: A regression analysis (R2=0.928) and a Pearson's product moment correlation (r=0.965, P<0.001) indicated a strong correlation of PtCO2 and PaCO2 during HFJV. The difference against a mean scatter diagram showed a relative good agreement between the two method (mean difference: 1.58 (SD 2.22) mmHg; limit of agreement: 2.86 and -6.02). CONCLUSIONS: PtCO2 obtained from a plateau of CO2 wave on capnography after interruption of HFJV can accurately reflect PaCO2 during HFJV in relative.