Monte Minute‎ > ‎Monte Minute‎ > ‎

08/30/2011 - VBG and lactate

posted Sep 28, 2011, 5:31 PM by Chief Resident   [ updated Sep 28, 2011, 5:33 PM by Purnema Madahar ]
In intake, we wondered what the utility of a venous lactate level is in the prediction of hyperlactatemia measured on an arterial blood gas, in this case as an indicator for bowel ischemia.

In the attached, excellent paper from Montefiore's ER, the authors compared peripheral venous and arterial lactate levels in 74 ED patients. They found good correlation (r2=0.89) but not great agreement: the mean difference between venous and arterial lactate was only 0.22 mmol/L, the differences in individual patients were between -1.3 and 1.7 mmol/L. As the authors note: 'it must be remembered that correlation is a measure of association rather than a true measure of agreement'.

When lactate levels were divided into normal and abnormal, venous lactate had a sensitivity of 94%, specificity of 57%, positive LR 2.2 and negative LR of 0.1. In other words: venous lactate might be useful to make hyperlactatemia much less likely but not to make it more likely.

I also attached a paper by Younger that is critiqued in the Montefiore paper and forms a good contrast in methodology - it also draws a much more optimistic conclusion.

Comments