Point-of-care bedside gas analyzer: limited use of venous pCO2 in emergency patients

Background: Because arterial punctures are more painful, venous blood gas analysis has been proposed as an alternative.

Objectives: To determine if venous pCO(2) can replace arterial pCO(2) in emergency patients using a bedside blood gas analyzer.

Methods: This is a cross-sectional study. We recruited patients who were deemed by the attending Emergency Physicians to require arterial blood gas analysis to determine their ventilation or acid-base status. A venous and an arterial blood gas sample were drawn from the patient, temporally as close to each other as possible. Both samples were then analyzed using the same bedside blood gas analyzer immediately after collection.

Results: There were 122 paired samples obtained. The strength of the association between arterial and venous pCO(2) is r = 0.838 (p = 0.001). The Bland-Altman bias plot methods for agreement show a mean difference of 3.3 mm Hg with two standard deviation limits of agreement being -17.4 to 23.9; 93.4% of the pCO(2) values fell within two standard deviation limits. Venous pCO(2) below 30 mm Hg had a 100% (95% confidence interval [CI] 90.5-100) sensitivity and 100% (95% CI 80.7-100) Negative predictive value to rule out hypercarbia, defined as arterial pCO(2) > 45 mm Hg.

Conclusions: There is significant variability in the arteriovenous difference of pCO(2) readings, hence, arterial punctures are still needed to specifically determine the arterial pCO(2).