The role of transcutaneous carbon dioxide measurement during exercise.

Abstract

Rationale: Arterial blood gas (ABG) sampling is the gold standard of measuring arterial carbon dioxide pressure (PaCO₂). However, it is invasive and has potential for complications. Since transcutaneous monitoring of carbon dioxide (TcCO₂) is easy to perform and non-invasive, it could be an alternative to PaCO₂.

Objective: We hypothesized that TcCO₂ could be used to discriminated between hypocapnia and hypercapnia during exercise.

Methods: This cross-sectional study included 124 outpatient referrals for cardiopulmonary exercise testing with ABG at rest and at peak exercise. The TcCO₂ was measured continuously using SENTEC V-STATS^TM. ABG was marked at the moment of extraction. The tests were grouped into rest (n= 65), peak exercise (n= 112) and both (n= 53) based on ABG collected. We used PaCO₂ value of 1.3 kPa (4.0 - 5.3 kPa) to discriminate between hypocapnia and hypercapnia at peak exercise.

Results: The Bland-Altman analysis of PaCO₂ and TcCO₂ at rest showed a mean difference of -0.01 kPa with limits of agreement (LOA) -0.73 – 0.72. At peak exercise, mean difference is -0.07 kPa with LOA -0.82 – 0.67. The TcCO₂ showed a strong and significant correlation with PaCO₂ at rest and peak exercise (r= 0.768, p=<0.0001 and r=0.807, p=<0.0001, respectively). The change in TcCO₂ from rest to peak exercise showed a moderate and significant correlation with the change in PaCO₂ (r= 0.61, p=<0.0001). In 70% of the cases, the direction of change in TcCO₂ is the same as in PaCO₂.

Conclusion: The mean difference between PaCO₂ and TcCO₂ is small. But the range of 1.49 kPa at peak exercise is higher than the predetermined acceptable range. Using TcCO₂ alone does not discriminate between hypocapnia or hypercapnia during exercise.