PURPOSE: The difference between central venous and arterial partial pressure of carbon dioxide [P(cv-a)CO2] has been demonstrated to correlate with the adequacy of fluid resuscitation in septic shock. We hypothesized that P(cv-a)CO2 independently predicts central venous oxygen saturation (ScvO2) in patients with severe sepsis.
METHODS: A prospective observational study was performed on 221 consecutive patients admitted to the intensive care unit with septic shock between September 2007 and August 2008. Three hundred simultaneous measurements of arterial and central venous blood gases, serum lactate, and hemoglobin were performed on admission. Univariable and multivariable stepwise linear regression analyses were performed with ScvO2 being the dependent variable, and P(cv-a)CO2, serum lactate, age, and hemoglobin being independent variables. Regression diagnostics were performed and there were no violations of regression assumptions. Data presented as mean±SD, unless otherwise indicated, and significance was defined as p < 0.05.
RESULTS: Patients age and APACHE II scores were 56±19 yrs and 16±9, respectively. ScvO2 was 67±11% and serum lactate was 2.7±3.4 mmol/l. P(cv-a)CO2, age, and serum lactate were all independent predictors of ScvO2.
CONCLUSION: Increased P(cv-a)CO2 on ICU admission is an independent predictor of decreased ScvO2 in patients with severe sepsis.
CLINICAL IMPLICATIONS: P(cv-a)CO2 could be used as a surrogate marker for adequate resuscitation in patients with severe sepsis.
DISCLOSURE: Jamal Alhashemi, No Financial Disclosure Information; No Product/Research Disclosure Information