RESUMEN
BACKGROUND: Increased physiological dead space ventilation (VD /VT ) at exercise reflects pulmonary gas exchange impairment and is a sensitive marker of cardio-respiratory disease. VD /VT is typically not measured during routine cardiopulmonary exercise testing (CPET) because its calculation requires arterial blood gas analysis for determination of PaCO2 . Instead, dead space ventilation is indirectly evaluated as a determinant of the ventilation (VE)/VCO2 relationship, which also depends on the PaCO2 set point. We hypothesized that VD /VT calculations based on non-invasive transcutaneous PCO2 (PtcCO2 ) measurement had better diagnostic characteristics than the VE/VCO2 slope for the discrimination of healthy subjects from patients with COPD, a common disease associated with impaired pulmonary gas exchange. METHODS: Retrospective study of 19 healthy controls and 24 COPD patients who underwent CPET with continuous PtcCO2 monitoring. Areas under receiver operating characteristics curves (AUC) were calculated to assess diagnostic accuracy of CPET measurement for the discrimination of COPD and Controls. RESULTS: The AUC for PtcCO2 -based VD /VT at VT1 (0.977) was significantly higher than for the VE/VCO2 slope (0.660), SpO2 at peak exercise (0.913), decrease in inspiratory capacity (0.719), and ventilatory reserve (0.708). At a threshold of 0.24, the sensitivity and specificity of PtcCO2 -based VD /VT for the discrimination of COPD patients and healthy Controls were 100% and 84%, respectively. All Control subjects had PtcCO2 -based VD /VT ≤ 0.25. CONCLUSIONS: PtcCO2 -based VD /VT was the most accurate measurement to discriminate healthy controls from subjects with COPD, a chronic lung disease associated with altered pulmonary gas exchange. Non-invasive monitoring of PtcCO2 may be useful for routine CPET.