Prognostic significance of central venous-to-arterial carbon dioxide difference during the first 24 hours of septic shock in patients with and without impaired cardiac function

Abstract

Objective: To investigate the prognostic significance of central venous-to-arterial carbon dioxide difference (cv-art CO 2 gap) during septic shock in patients with and without impaired cardiac function.

Methods: We performed a prospective cohort study in 10 French intensive care units. Patients suffering from septic shock were assigned to the impaired cardiac function group ('cardiac group', n =123) if they had atrial fibrillation (AF) and/or left ventricular ejection fraction (LVEF) <50% at study entry and to the non-cardiac group ( n =240) otherwise.

Results: Central venous and arterial blood gases were sampled every 6 h during the first 24 h to calculate cv-art CO 2 gap. Patients in the cardiac group had a higher cv-art CO 2 gap [at study entry and 6 and 12 h (all P <0.02)] than the non-cardiac group. Patients in the cardiac group with a cv-art CO 2 gap >0.9 kPa at 12 h had a higher risk of day 28 mortality (hazard ratio=3.18; P =0.0049). Among the 59 patients in the cardiac group with mean arterial pressure (MAP) ≥65 mm Hg, central venous pressure (CVP) ≥8 mm Hg and central venous oxygen saturation (ScvO 2 ) ≥70% at 12 h, those with a high cv-art CO 2 gap (>0.9 kPa; n =19) had a higher day 28 mortality (37% vs. 13%; P =0.042). In the non-cardiac group, a high cv-art CO 2 gap was not linked to a higher risk of day 28 death, whatever the threshold value of the cv-art CO 2 gap.

Conclusion: Patients with septic shock and with AF and/or low LVEF were more prone to a persistent high cv-art CO 2 gap, even when initial resuscitation succeeded in normalizing MAP, CVP, and ScvO 2 . In these patients, a persistent high cv-art CO 2 gap at 12 h was significantly associated with higher day 28 mortality.

Keywords: blood gas analysis; central venous-arterial CO2 difference; septic shock.