Background: The contribution of ethanol ([EtOH]) to the osmol gap (OG) is commonly described by the formula [EtOH (mg/dL)]/k, where k is assumed to be 4.6 (one-tenth of its molecular weight) if ethanol behaves ideally in solution. However, several studies on convenience samples of patients suggest that ethanol does not behave ideally and that k may be significantly different from this ideal constant.
Objectives: To determine prospectively the relationship between serum ethanol concentration and total serum osmolality in a group of healthy volunteers.
Methods: Experimental subjects ingested 20 mL of 100% ethanol diluted in sugar-free soda at a rate of one drink every 10 min, up to a maximum of seven drinks. Control subjects ingested 20 mL of water diluted in sugar-free soda at the same rate. Blood samples were obtained at baseline and then at every 20 min for 180 min to measure serum [EtOH] concentration, electrolytes, glucose, and osmolality (via freezing-point depression). The OG was calculated by subtracting predicted osmolality from measured osmolality. The OG was then divided by [EtOH] to determine the coefficient of ethanol's contribution to total serum osmolality.
Results: A total of 10 volunteers (five men and five women; mean age, 38.8 years, and range, 28-49 years) participated in and completed the study. Eight (four male and four female) were in the experimental group, and two (one male and one female) were in the control group. Mean peak [EtOH] was 229 mg/dL (median, 223.5 mg/dL; IQR, 171-273 mg/dL) and a linear relationship between [EtOH] and OG (Pearson coefficient of 0.98) was found. Using covariate correction for each subject's baseline OG, k was calculated to be 4.25 (95% CI, 4.13-4.38) averaged over all participants.
Conclusions: In this volunteer study, the coefficient describing the contribution of ethanol to serum osmolality (k) was found to be 4.25. This indicates that ethanol contributes more to total serum osmolality than would be predicted for an ideal solute.