Persistently high venous-to-arterial carbon dioxide differences during early resuscitation are associated with poor outcomes in septic shock

• Published in: Critical care (London, England) Vol. 17; no. 6; p. R294
• Main Authors: Ospina-Tascón, Gustavo A, Bautista-Rincón, Diego F, Umaña, Mauricio, Tafur, José D, Gutiérrez, Alejandro, García, Alberto F, Bermúdez, William, Granados, Marcela, Arango-Dávila, César, Hernández, Glenn
• Format: Journal Article
• Language: English
• Published: England BioMed Central Ltd 13.12.2013; BioMed Central
• Subjects: Aged; Blood flow; Carbon dioxide; Carbon Dioxide - blood; Cardiac Output; Complications and side effects; Female; Hemodynamics; Humans; Lactic Acid - blood; Male; Middle Aged; Multiple Organ Failure - etiology; Oxygen - blood; Oxygen Consumption; Physiological aspects; Prognosis; Prospective Studies; Septic shock; Shock, Septic - blood; Shock, Septic - complications; Shock, Septic - mortality; Shock, Septic - therapy; Survival Analysis; United States
• ISSN: 1364-8535; 1466-609X; 1364-8535
• DOI: 10.1186/cc13160

Venous-to-arterial carbon dioxide difference (Pv-aCO2) may reflect the adequacy of blood flow during shock states. We sought to test whether the development of Pv-aCO2 during the very early phases of resuscitation is related to multi-organ dysfunction and outcomes in a population of septic shock patients resuscitated targeting the usual oxygen-derived and hemodynamic parameters. We conducted a prospective observational study in a 60-bed mixed ICU in a University affiliated Hospital. 85 patients with a new septic shock episode were included. A Pv-aCO2 value ≥ 6 mmHg was considered to be high. Patients were classified in four predefined groups according to the Pv-aCO2 evolution during the first 6 hours of resuscitation: (1) persistently high Pv-aCO2 (high at T0 and T6); (2) increasing Pv-aCO2 (normal at T0, high at T6); (3) decreasing Pv-aCO2 (high at T0, normal at T6); and (4) persistently normal Pv-aCO2 (normal at T0 and T6). Multiorgan dysfunction at day-3 was compared for predefined groups and a Kaplan Meier curve was constructed to show the survival probabilities at day-28 using a log-rank test to evaluate differences between groups. A Spearman-Rho was used to test the agreement between cardiac output and Pv-aCO2. Finally, we calculated the mortality risk ratios at day-28 among patients attaining normal oxygen parameters but with a concomitantly increased Pv-aCO2. Patients with persistently high and increasing Pv-aCO2 at T6 had significant higher SOFA scores at day-3 (p < 0.001) and higher mortality rates at day-28 (log rank test: 19.21, p < 0.001) compared with patients who evolved with normal Pv-aCO2 at T6. Interestingly, a poor agreement between cardiac output and Pv-aCO2 was observed (r2 = 0.025, p < 0.01) at different points of resuscitation. Patients who reached a central venous saturation (ScvO)2 ≥ 70% or mixed venous oxygen saturation (SvO2) ≥ 65% but with concomitantly high Pv-aCO2 at different developmental points (i.e., T0, T6 and T12) had a significant mortality risk ratio at day-28. The persistence of high Pv-aCO2 during the early resuscitation of septic shock was associated with more severe multi-organ dysfunction and worse outcomes at day-28. Although mechanisms conducting to increase Pv-aCO2 during septic shock are insufficiently understood, Pv-aCO2 could identify a high risk of death in apparently resuscitated patients.