[Prediction of PaO₂/FiO₂ ratio from SpO₂/FiO₂ ratio adjusted by transcutaneous CO₂ measurement in critically ill children].

OBJECTIVE: To evaluate if the inclusion of the transcutaneous CO(2) tension measurement (PtcCO(2)) can improve partial pressure of oxygen/ fraction of inspired oxygen ratio [PaO(2)/FiO(2) (P/F)] prediction from pulse oximetry saturation/FiO(2) ratio [SpO(2)/FiO(2) (S/F)].
METHODS: Retrospective analysis of blood gas data from critically ill children. PaO(2), SpO(2), FiO(2) and PtcCO(2) from 40 samples in 8 patients were analysed. A multiple linear regression model was performed to predict P/F ratio from S/F ratio and PtcCO(2). Using the equation obtained, S/F ratio values were calculated for P/F ratios of 200 and 300 and different levels of PtcCO2. Receiver Operator Characteristic (ROC) curves were made to analyse the diagnostic values of P/F ratio (200 and 300).
RESULTS: The linear regression model was: P/F=37.277+(1.072×S/F) - (1.567×PtcCO2); P<.0001; R(2)=0.469. Using the equation, for a PtcCO(2) value of 40 mmHg, P/F ratios of 200 and 300 corresponded to S/F ratios of 295.1 and 426.5, respectively. Computed P/F ratio less than 256.7 had 84.6% sensitivity and 85.2% specificity for the diagnosis of P/F ratio less than 200. Computed P/F ratio less than 297.6 had 89.7% sensitivity and 82% specificity for the diagnosis of P/F ratio less than 300.
CONCLUSION: PtcCO(2) has a significant influence on the prediction of P/F ratio from S/F ratio. Prospective studies with more patients are needed to validate these results.