View Abstract

A5133 - Reliability of Peripheral Venous Blood Gas Values in the Diagnosis of Ventilatory Failure
Author Block: M. Shahzadi1, A. Mogla1, H. Afsana1, V. P. Vasudevan2, A. Z. Rasheed3, L. N. Gerolemou4, F. Arjomand5; 1Pulmonary Division, Internal Medicine, The Brooklyn Hospital Center, Brooklyn, NY, United States, 2Pulmonary Division, Internal Medicine, The Brooklyn Hospital, Brooklyn, NY, United States, 3Internal Medicine, The Brooklyn Hospital Center, Brooklyn, NY, United States, 4The Brooklyn Hospital Center, Brooklyn, NY, United States, 5Brooklyn Hosp Ctr, Brooklyn, NY, United States.
RATIONALE: Arterial Blood gas analysis is a laboratory test used to assess acid-base status along with adequacy of ventilation and oxygenation among critically ill patients. Studies have validated that there is a fixed and predictable difference in pH (minus 0.05-0.06) and partial pressure of carbon dioxide (+ 5-6 mm Hg) between mixed central venous blood and arterial blood. However, blood drawn from a peripheral vein has variable and unpredictable differences from arterial blood in terms of pH and pressure of carbon dioxide. In practice, measurement of pH and partial pressure of carbon dioxide of blood drawn from peripheral vein can result in errors in diagnosis and treatment. METHODS: The study was conducted by the Pulmonary Division of the Department of Internal Medicine at The Brooklyn Hospital Center. This was a retrospective observational non-interventional chart review of all adult patients who had peripheral venous blood gas, basic metabolic profile and arterial blood gas done concurrently at The Brooklyn Hospital Center from January 2015 to September 2017. We reviewed Patient demographics, chronic comorbidities, clinical presentation, results of venous, arterial blood gas and metabolic panel. PaCO2 and pH analyzed from arterial blood were used as the gold standard for comparison. RESULTS: 100 patients were included in the study. 58(58%) were male; 42(42%) were female. The mean age was 60-69 years. 37(37%) patients had ventilator failure; 23(23%) had CKD; and 22(22%) had CHF. The difference between peripheral venous blood and arterial blood in pH is on average 0.07+/-0.03 (0.04-0.10), in PCO2 is on average 10+/-3 cmH2O (7-13) and in HCO3 is on average 2+/-1 (1-3). The gap on one hand between peripheral venous blood and arterial blood was much higher than the gap between central/mixed venous and arterial blood. 63(63%) were intubated on the basis of these results. The difference in HCO3 between basic metabolic panel and arterial blood is on average 3+/-1 (2-4). The difference in HCO3 between basic metabolic panel and venous blood is on average 4+/-1. CONCLUSION: Our study demonstrates that the difference in pH and PCO2 when comparing peripheral venous blood to arterial blood is unpredictable and ranges widely, and thus can lead to erroneous diagnosis and unnecessary treatment such as non-invasive or invasive ventilator support. Our study refutes the assumptions that PaCO2 can be reliably extrapolated from peripheral venous CO2 in therapeutic management decision.