Background: The qSOFA score (quickSOFA) was derived as a bedside prompt to help identify patients with suspected infection at risk for poor outcome and a qSOFA ≥2 points suggests a greater risk of death or prolonged ICU stay. This study assessed the association between end tidal carbon dioxide (ETCO2) measured at ED triage and qSOFA scores and outcome.
Methods: This prospective observational study enrolled a convenience sample of adult patients presenting to the ED of a tertiary care center over 30 months. Patients had elements of the qSOFA score measured along with initial vital signs including exhaled ETCO2 via capnography at triage. Outcome measures included in-hospital mortality and ICU admission. Area under the ROC curve (AUC) with 95%CI’s were calculated along with Spearman’s rho.
Results: There were 1136 patients enrolled prospectively and 1124 patients with complete data. Patient mean age was 56 (SD19), 53% were male, 64 (6%) patients were admitted to ICU and 26 (2.4%) experienced in-hospital mortality. There were 769 (68%) with qSOFA of zero; 273 (24%) with qSOFA of one; 76 (7%) with qSOFA of two and 6 (1%) with qSOFA of three. Mean levels of ETCO2 at triage in all patients was 34 (95%CI 33-34) mmHg. ETCO2 levels for patients with qSOFA scores of 0, 1, 2, and 3 respectively were 35 (34-35); 32 (30-33); 30 (27-32); and 21 (18-25) (p<0.001). The correlation between ETCO2 and qSOFA was -0.23 (p<0.001). The ETCO2 AUC for predicting mortality was 0.82 (0.72-0.91) and the qSOFA AUC was 0.83 (0.75- 0.91). The ETCO2 AUC for predicting ICU admission was 0.73 (0.67-0.80) and the qSOFA AUC was 0.76 (0.69-0.83).
Conclusion: Decreasing ETCO2 levels were significantly associated with increasing qSOFA scores. Moreover, ETCO2 predicted in-hospital mortality and ICU admission at par with aSOFA. Given that ETCO2 is quick and non-invasive, it has the potential to be used as a screening tool at triage for predicting mortality and ICU admission prior to calculating a qSOFA score.