Outcome of Comparison Study Of Various Equations For Serum Osmolality

Abstract

Background: Serum  osmolality  measured by cryoscopic technique in laboratory is the reference method.In clinical settings,serum osmolality measurement is not feasible at bedside. In  normal subjects, sodium, potassium  glucose, and urea are the primary circulating solutes. These solute concentrations can be used to predict measured osmolality if no other solutes present at high millimolar concentrations. Many equations of serum osmolality have been proposed. The osmolal gap (OG) is the difference between measured osmolality and calculated osmolality. The major use of OG is to screen for  presence of exogenous toxic substances and to screen  alcohol intoxication cases. Aim/Objective: The purpose of this study was to compare the calculated osmolality using various formulae with the measured osmolality to determine which calculated formula fit best with measured osmolality. Materials and Methods: Serum osmolality results from January 2015 to December 2015 were extracted from the laboratory information system (LIS). Serum osmolality performed simultaneously with renal and liver function tests, serum electrolytes and plasma glucose were included. Serum osmolality measured for patients with the history of drug abuse and poisoning were excluded from the study. 405 serum osmolality results were chosen and  divided into two groups. Group 1 included 205 data with normal serum osmolality, renal, liver function tests and plasma glucose level less than 7.8 mmol/L. For the second group (n=200), data with low serum osmolality (n=90) and high serum osmolality (n=80) and normal serum osmolality (n=30) were included. Group 1 data was to identify which equation correlated with the measured osmolality and the Group 2 data to study the performance of equation that correlated with the measured osmolality.

Results: Only four out of 19 formulae were identified as optimal by having the mean OG ≤ 2 mOsm/kg. The Smithline-Gardner formula (2Na+ Glu + BUN) showed the smallest osmolal gap with mean bias 0.3 mOsm/kg. The Dorwart-Chalmers formula incorporated in most autoanalysers for calculation of osmolality underestimated  compared to measured osmolality. Conclusion: We recommend Smithline-Gardner formula for calculation of osmolal gap, as the OG gap is close to zero, simple, easy to calculate at bedside and easily incorporated in the Laboratory Information System.