We investigated saliva as a possible contaminant causing unexplained increases in serum potassium (K) concentration in our laboratory. Random saliva samples were collected from healthy adult volunteers and assayed for K and sodium (Na). Most samples were also analyzed for 25 general chemistry tests. Concentrations of some biochemical analytes were much lower in saliva than in serum [alkaline phosphatase, alanine aminotransferase, Na, chloride (Cl), bicarbonate (HCO3), pH (based on concentration)] and some were comparable with serum [aspartate aminotransferase, total magnesium (Mg), total and ionized calcium (Ca), uric acid, urea nitrogen]. Glucose, ionized Mg, albumin, and total protein were unmeasurable with the routine laboratory methods. Amylase, K, and inorganic phosphorus (Pi) were much higher in saliva than in serum. The salivary pH (mean 7.01) was close to the acid dissociation constant of H2PO4- is less than or greater than HPO4(2-) + H+ (pKa is approximately 7.2) indicating the concentrations of these two phosphate anions are approximately equimolar, that is 1 mmol of Pi carries 1.5 negative charge (equal to 1.5 meq). The calculated mean difference between the measured equivalents of cations and anions (anion gap, AG) was similar to serum (4.2 vs. 6.6 meq/L) only when phosphate anion was included in the calculation.The positive AG values indicate that unmeasured anions are commonly present in the saliva (e.g., thiocynate in smokers and vegetarians, fluoride from drinking water and toothpaste). The value of the salivary AG was not dependent on the pH (p equals 0.067). Significant relationship was found between phosphate anion and several cations: K (p is less than 0.0001), total Ca (p=0.0033), bound Ca (p is less than 0.0001), total Mg (p = 0.020) and between Cl anion and K cation (p = 0.034). The concentrations of phosphate anion, K, total Mg, total and bound Ca decreased with increasing pH (p is less than 0.05), but Na and Cl were not affected by pH changes (p equals 0.45). The relation between Pi and ionized Ca was not significant (p=0.853). The decrease of ionized Ca with increasing pH (p=0.047) was most likely due to increased protein binding of Ca. Thus, phosphate is the main complexing anion in saliva and Ca-K-Mg-P complexes are deposited during the remineralization process that occurs at higher salivary pH. Further, contamination of blood or any other body fluid (such as urine) with saliva may cause not only erroneously high amylase results but also falsely high K and Pi results.

Agency
National Institute of Health (NIH)
Institute
Clinical Center (CLC)
Type
Intramural Research (Z01)
Project #
1Z01CL010001-24
Application #
6103667
Study Section
Cognition and Perception Study Section (CP)
Project Start
Project End
Budget Start
Budget End
Support Year
24
Fiscal Year
1998
Total Cost
Indirect Cost
Name
Clinical Center
Department
Type
DUNS #
City
State
Country
United States
Zip Code
Lane, Jason W; Rehak, Nadja N; Hortin, Glen L et al. (2008) Pseudohyperphosphatemia associated with high-dose liposomal amphotericin B therapy. Clin Chim Acta 387:145-9
Rehak, Nadja N; Cecco, Stacey A; Hortin, Glen L (2008) Photolysis of bilirubin in serum specimens exposed to room lighting. Clin Chim Acta 387:181-3