Unconjugated and total oestriol (E3) have been determined in saliva (S) and plasma (P) by radioimmunoassay using a highly specific antiserum against E3. Total oestrogens (E = E1 + E2 + E3) in urine were analysed by the Kober-Ittrich method. No correlation (P> 0.05) was found between unconjugated E3 (S) and total E (U) as well as total E3 (S) and total E (U). Total E3 (S) and unconjugated E3 (P), total E3 (S) and total E3 (P) correlated weakly (P < 0.05 and > 0.01). The six remaining correlations were highly significant (P < 0.001 or < 0.0001). Oestradiol (Evans et al. 1980), progesterone (Walk¬ er et al. 1981; Luisi et al. 1981), testosterone (Landman et al. 1976), and cortisol (Walker et al. 1978; Katz & Shannon 1969) have been detected insaliva and found to reflect the biological activity of these hormones equally well or better than do the serum or urine levels. Steroids are supposed to reach the saliva by passive diffusion, thus salivary steroid levels should represent the active, non protein bound fraction of the total plasma steroid pool. Salivary E2 allowed the timing of ovulation (Evans et al. 1980), and progesterone levels in saliva were found to be superior in the evaluation of corpus luteum insufficiency (Luisi et al. 1981). In non-pregnant women, determination of salivary steroids is handicapped by the low concentrations (1/10 of plasma levels) in which they are present. Saliva was also found to be suitable material for therapeutic drug monitoring (Horning et al. 1977; Odlind & Johansson 1981). In pregnancy, E3 was found to be present in saliva mainly in the free, unconjugated form; salivary and plasma concen¬ trations of unconjugated E3 have correlated well in matched samples (Truran et al. 1982). The mean values of total E3 in saliva ranged between 2.77 and 7.16 nmol/1 from the 30th to the 40th week of gestation (Fischer-Rasmussen et al. 1981). Vining et al. (1983) found similar concentrations of unconjugated E3 in saliva and unbound unconjugated serum E3 levels. The aim of this study was to investigate correlations between salivary E3 (un¬ conjugated and total) concentrations and plasma (unconjugated and total) or urine E3 levels which are known to be suitable for monitoring risk preg¬ nancy. Saliva, because of its easy sampling, would represent a suitable biological material to deter¬ mine steroids, especially in pregnancy.
Material and MethodsSampling Saliva was obtained after the patients had chewed a sugar free chewing gum for 5 min. The samples were frozen and centrifuged before performing the assay to obtain a clear non-viscous supernatant. Blood samples were drawn at the same time as the saliva collection and ammonium heparinate was added as an anticoagulant. Twenty-four h urine samples were collected from 8 a. m. to 8 a. m. without adding any preservative.
Steroid assaysUrinary total oestrogens were determined by the KoberIttrich method using fluorescence detection (Ittrich 1960). E3 was extracted from saliva (0.5 ml) or plasma (0.2 ml) with 5 ml of ethylacetate. Aliquots of the extract (0...