ABSTRACT
PURPOSE: To determine the correlation and agreement between perpheral venous and arterial values in emergency department patients. METHODS: The study was conducted at an urban teaching hospital emergency department with an annual volume of 80,000 patient visits. Prospective comparison of paired peripheral venous and arterial values were evaluated from October, 2009 to February, 2010. Peripheral venous and arterial samples were taken as simultaneously as possible when patients who were deemed by the attending doctor to require an peripheral arterial sample presented in the emergency department. Collected information included age, sex, vital signs, emergency department initial diagnosis, lactate, pH, base excess, bicarbonate, pO2, and pCO2. Statistical methods were Shapiro-Wilk test, Spearman rho test, linear regression analysis, and Bland-Altman plots. RESULTS: Among the 546 patients, 50 were excluded and 496 were included. The majority (59.5%) of patients were male. Correlation constant (rho) of pH and lactate were 0.907 and 0.901, respectively. Mean difference and CI (confidence interval) of pH were -0.04 and -1.34~1.56, respectively. Mean difference and CI of lactate were 0.11 mmol/L and -0.15~0.07 mmol/L, respectively. CONCLUSION: Moderate correlation and agreement between perpheral venous and arterial pH, lactate, base excess, and CO2 was evident. Especially, peripheral venous pH, lactate correlated very well and had reasonable agreement with peripheral arterial values to serve as substitutes.
Subject(s)
Humans , Male , Blood Gas Analysis , Emergencies , Hospitals, Teaching , Hydrogen-Ion Concentration , Lactic Acid , Linear Models , Prospective Studies , Statistics as Topic , Vital SignsABSTRACT
PURPOSE: To determine the correlation and agreement between perpheral venous and arterial values in emergency department patients. METHODS: The study was conducted at an urban teaching hospital emergency department with an annual volume of 80,000 patient visits. Prospective comparison of paired peripheral venous and arterial values were evaluated from October, 2009 to February, 2010. Peripheral venous and arterial samples were taken as simultaneously as possible when patients who were deemed by the attending doctor to require an peripheral arterial sample presented in the emergency department. Collected information included age, sex, vital signs, emergency department initial diagnosis, lactate, pH, base excess, bicarbonate, pO2, and pCO2. Statistical methods were Shapiro-Wilk test, Spearman rho test, linear regression analysis, and Bland-Altman plots. RESULTS: Among the 546 patients, 50 were excluded and 496 were included. The majority (59.5%) of patients were male. Correlation constant (rho) of pH and lactate were 0.907 and 0.901, respectively. Mean difference and CI (confidence interval) of pH were -0.04 and -1.34~1.56, respectively. Mean difference and CI of lactate were 0.11 mmol/L and -0.15~0.07 mmol/L, respectively. CONCLUSION: Moderate correlation and agreement between perpheral venous and arterial pH, lactate, base excess, and CO2 was evident. Especially, peripheral venous pH, lactate correlated very well and had reasonable agreement with peripheral arterial values to serve as substitutes.