The anion gap is not a reliable screening test for lactic acidosis in emergency department patients

Abstract

Study objectives: Lactic acidosis portends a poor prognosis in trauma, sepsis, and other shock states and is useful for triaging and resuscitating emergency department (ED) patients. The anion gap is commonly taught to be a good predictor of lactic acidosis. However, some ICU studies show that the anion gap may be insensitive to lactic acidosis. We seek to determine whether the anion gap is a sensitive predictor of lactic acidosis in the ED setting. Methods: This was a retrospective cohort design using a laboratory database and explicit medical record review in a military academic ED with a 50,000 annual census. Participants were all ED patients over a 7-month period who had lactic acidosis level drawn for clinical indications. Patients were excluded if the anion gap and lactic acidosis values were drawn more than 60 minutes apart or for presence of ketoacidosis or an anion gap–inducing toxic ingestion. Baseline demographic data, hospital mortalitye the outcome measures. Venous or arterial lactic acidosis specimens were analyzed on a Vitros 950 using colorimetric methodology (normal up to 2.5 mmol/L). The anion gap was calculated by [Na–Cl–CO 2 ] on the Vitros 950 using potentiometric methodology. Sensitivity analyses with 95% confidence intervals (CIs) of the anion gap for detecting presence of lactic acidosis were calculated for the traditional anion gap normal value (anion gap present if >12.0) and for the lower anion gap normal value when using newer ion-selective electrode assay (anion gap present if >6.0). Results: We performed an interim data analysis of 207 patients, of whom 61 were excluded by protocol. Of the 146 patients remaining, the average age was 66 years (range 16 to 94 years), and 48.6% of patients were women. There were 33 cases of lactic acidosis (22.6%) in this highly selected ED patient population. The most common diagnosis in the lactic acidosis cohort was severe sepsis. Hospital mortality was significantly higher in patients with lactic acidosis (24.2%) than in patients without lactic acidosis (2.6%; P Conclusion: Using the traditional definition of anion gap greater than 12, an electrolyte panel with calculated anion gap is poorly sensitive for presence of lactic acidosis. When using the newly revised anion gap of greater than 6 on ion-selective assays, the anion gap is more sensitive but very nonspecific for detection of lactic acidosis (indeed, 84% of our highly selected ED patients had an elevated anion gap >6). Therefore, with either defined cutoff value, following an anion gap screen stepwise approach is inappropriate. Therefore, we recommend ordering a specific lactic acidosis assay immediately on suspicion of a shock state, instead of awaiting the anion gap results. Furthermore, emergency physicians should be advised of the new normal value of anion gap is [Na–Cl–CO 2 ] greater than 6 when using ion-specific electrolyte assays.