The objective of this study was to examine the repeatability of blood gas (BG) parameters and their derived variables such as the central venous-to-arterial carbon dioxide tension difference (▵PCO2) and the ratio of ▵PCO2 over the central arteriovenous oxygen content difference (▵PCO2/C(a-cv)O2) and to determine the smallest detectable changes in individual patients.
A total of 192 patients with arterial and central venous catheters were included prospectively. Two subsequent arterial and central venous blood samples were collected immediately one after the other and analyzed using the same point-of-care BG analyzer. The samples were analyzed for arterial and venous BG parameters, ▵PCO2, and ▵PCO2/C(a-cv)O2 ratio. Repeatability was expressed as the smallest detectable difference (SDD) and the least significant change (LSC). A change in value of these parameters exceeding the SDD or the LSC should be regarded as real.
The SDDs for arterial carbon dioxide tension, arterial oxygen saturation, central venous oxygen saturation (ScvO2), and ▵PCO2 were small: ±2.06 mm Hg, ±1.23%, 2.92%, and ±1.98 mm Hg, respectively, whereas the SDDs for arterial oxygen tension (PaO2) and ▵PCO2/C(a-cv)O2 were high: ±9.09 mm Hg and ±0.57 mm Hg/mL, respectively. The LSCs (%) for these variables were 5.06, 1.27, 4.44, 32.4, 9.51, and 38.5, respectively.
The repeatability of all these variables was good except for PaO2 and ▵PCO2/C(a-cv)O2 ratio for which we observed an important inherent variability. Expressed as SDD, a ScvO2 change value of at least ±3% should be considered as true. The clinician must be aware that an apparent change in these variables in an individual patient might represent only an inherent variation.