Redefinition of the Normal Range for Serum Sodium

Abstract

Abstract The serum sodium concentrations of three groups of patients selected from laboratory records and of one group of outpatients selected for a prospective study were examined. The mean serum sodium concentration of inpatients with normal serum urea concentrations was 8.5 mEq./ L. lower than that of a group of healthy normal subjects. Part of this difference (2.6 mEq./L.) could be attributed to a nonspecific effect of illness. It was not possible to demonstrate any effect due to the hospital environment, but the results do not exclude the possibility of such an effect. It was concluded that all or most of the remaining difference was due to weighting of the inpatient data by low values of pathologic significance; the proportion of these low values was too great to allow a normal range to be extracted from the data by statistical methods. The concept of the normal range is discussed. It is suggested that two ranges are required for serum sodium, a normal range (137-147 mEq./L. in this laboratory) to make assertions about alterations in specific diseases, and a range derived from patients likely to have no manifest disturbances of salt and water metabolism (135-144 mEq./ L.) to detect such disturbances.