Influence of Unmeasured Anions Identified by Stewart Principle on the Length of Postoperative Hospital Stay

Lee, Kyoung Min; Seo, Sung Ho; Lee, Seung Yun; Lee, Jun Geol; Kim, Tae Yop; Rhee, Ka young

Abstract

BACKGROUND
Calculation of the base excess (BE) and the anion gap (AG) is commonly used to identify the presence and to analyze the cause of metabolic acidosis in critically ill patients. However, the calculation of BE assumes normal water content, electrolytes, and albumin, changes in these values will change the calculated BE. Calculation of the AG does not control for changes in albumin and cannot distinguish plasma concentration changes of negatively charged protein (albumin) from that of other anions. Based on Stewart's physicochemical principles, Gilfix et al developed equations to calculate the BE caused by unmeasured anions (BEua) taking into account changes in free water, chloride, albumin, and PCO2 that theoretically should reflect metabolic changes better than the less complete biochemical measurements. This study was designed to evaluate the influence of BEua and other variables on the length of postoperative hospital stay. METHODS: The data from 100 consecutive patients were collected prospectively in patients who underwent intra-abdominal operations under general anesthesia and admitted to the adult intensive care unit. All samples were routine samples taken from arterial lines postoperatively and analyzed for arterial blood gas, plasma electrolytes, inorganic phosphates and albumin concentrations. BEua was calculated from the equations developed by Gilfix et al. We also calculated AGNa, K (Na++K+-Cl--HCO3-) and AGNa (Na+-Cl--HCO3-). Correlations between the length of postoperative hospital stay and these variables were studied using linear regression analysis. RESULTS: BEua and BE were significantly correlated with the length of ICU stay (r=0.295, p<0.01 and r=0.249, p<0.05). Neither AGNa, K nor AGNa was correlated with the length of ICU stay. Significant correlation was observed between the length of postoperative hospital stay and BEua (r=0.316, p<0.01), BE (r=0.288, p<0.01), AGNa, K (r=0.284, p<0.01), and AGNa (r=0.263, p<0.05). CONCLUSIONS: In this study BEua was significantly correlated with the length of ICU stay and postoperative hospital stay compared with other variables. This finding suggests that BEua may be used as a more reliable predictor of outcome in ICU patients.