End-tidal carbon dioxide as a goal of early sepsis therapy

Objective To determine the use of end-tidal carbon dioxide (etco₂) as an end point of sepsis resuscitation.

Methods This was a prospective, observational, single-center cohort study of emergency department patients receiving treatment for severe sepsis with a quantitative resuscitation protocol. Three etco₂ readings were taken during a 1-minute time frame at 0, 3, and 6 hours of treatment. Linear regression was used to characterize the association between etco₂ and central venous oxygen saturation (SCVo₂) and lactate and also to determine the relationship between their change. Analysis of variance was used to determine the relationship between etco₂ and disposition.

Results Sixty-nine patients were included in our final analysis. For baseline values, linear regression failed to show a relationship between etco₂ and SCVo₂ (β = - 0.04, t(70) = - 0.53, P =.60) but showed a nearly significant relationship (β = - 0.51, t(70) = - 1.90, P =.06) with lactate. There was no significant relationship between etco₂ and SCVo₂ at 3 hours (β = 0.12, t(70) = 1.43, P =.16) or 6 hours (β = 0.05, t(64) = 0.82, P =.67). There was also no significant relationship between 6-hour change in etco₂ and change in SCVo₂ (β = 0.04, t(64) = 0.43, P =.67) or lactate (β = 0.04, t(59) = 0.52, P =.60) or disposition (F(4) = 0.78, P =.54).

Conclusion End-tidal carbon dioxide is unlikely to be a useful clinical end point for sepsis resuscitation, although it may be useful as a triage tool in suspected sepsis because baseline values may reflect initial lactate.