ABSTRACT
BACKGROUND: Venous blood gases (VBG) are commonly utilised, particularly in the emergency setting, to assess and monitor patients at risk of ventilatory failure with limited evidence regarding their clinical utility in the assessment of ventilatory status over time. AIMS: This study aims to assess agreement between arterial and venous pH and partial pressure of carbon dioxide (pCO2) both before and after physiological stress, at each time point, and within the same subject between paired samples before and after bronchoscopy. METHODS: Prospective study of 30 patients undergoing flexible bronchoscopy under conscious sedation. Paired arterial and venous samples taken before and after bronchoscopy were analysed utilising descriptive statistics and bias plot (Bland-Altman) analysis to assess limits of agreement. RESULTS: Compared with baseline, post-bronchoscopy arterial blood gas and VBG showed reduced pH (-0.05 ± 0.05 and -0.04 ± 0.04 respectively) and increased arterial and venous pCO2 (5.9 ± 6.7 and 3.5 ± 5.5 mmHg respectively), the differences being statistically significant (P = 0.035). There was statistical agreement between arterial blood gas and VBG parameters; however, the limits of agreement were wide at rest and, for pCO2, widened further post-bronchoscopy. CONCLUSION: Sequential VBG provide an unpredictable means for assessing pCO2 in patients undergoing flexible bronchoscopy. Previously noted poor agreement between arterial and venous pCO2 worsens following physiological stress, with sequential VBG likely to underestimate changes in ventilatory status in patients with acute respiratory compromise, suggesting limited utility as a means for monitoring changes in ventilation.
