Abstract
BACKGROUND: There may be significant difference between measurement of end-tidal carbon dioxide partial pressure (PetCO2) and arterial carbon dioxide partial pressure (PaCO2) during one-lung ventilation with low tidal volume for thoracic surgeries. Transcutaneous carbon dioxide partial pressure (PtcCO2) monitoring can be used continuously to evaluate PaCO2 in a noninvasive fashion. In this study, we compared the accuracy between PetCO2 and PtcCO2 in predicting PaCO2 during prolonged one-lung ventilation with low tidal volume for thoracic surgeries. METHODS: Eighteen adult patients who underwent thoracic surgeries with one-lung ventilation longer than two hours were included in this study. Their PetCO2, PtcCO2, and PaCO2 values were collected at five time points before and during one-lung ventilation. Agreement among measures was evaluated by Bland-Altman analysis. RESULTS: Ninety sample sets were obtained. The bias and precision when PtcCO2 and PaCO2 were compared were 4.1 ± 6.5 mmHg during two-lung ventilation and 2.9 ± 6.1 mmHg during one-lung ventilation. Those when PetCO2 and PaCO2 were compared were -11.8 ± 6.4 mmHg during two-lung ventilation and -11.8 ± 4.9 mmHg during one-lung ventilation. The differences between PtcCO2 and PaCO2 were significantly lower than those between PetCO2 and PaCO2 at all five time-points (p < 0.05). CONCLUSIONS: PtcCO2 monitoring was more accurate for predicting PaCO2 levels during prolonged one-lung ventilation with low tidal volume for patients undergoing thoracic surgeries.