Urinary free cortisol is unrelated to physiological changes in urine volume in healthy women.

To the Editor: Urinary free cortisol is largely accepted as an accurate means of assessing daily cortisol secretion. A more precise definition of this marker should be “urinary free corticoids” (UFCs), because all competitive binding assays overestimate free cortisol in urine (1 ). Recently, it has been argued that urine volume may be a confounding factor because high fluid intake has been reported to significantly increase UFCs in healthy subjects (2 ), and an earlier study also stated that UFC concentrations are closely related to changes in urine volume in children (3 ). Conversely, a recent study published in Clinical Chemistry reported that UFCs are not influenced by short-term water diuresis in male and female volunteers (4 ). We have therefore tried to establish whether spontaneous variations of urine volume, in absence of a possible stress caused by manipulations of water intake, may affect daily UFC concentrations. Three consecutive 24-h urine collections were obtained from 88 healthy women (age, 24.7 6 0.69 years, mean 6 SE; body mass index, 27.2 6 1.67 kg/m). All subjects had not taken drugs for the previous 3 months and drank freely during the study period. Only subjects with plasma cortisol within the reference values at 800, 1700, and 2400, cortisol circadian rhythm with suppressibility by 1 mg of dexamethasone overnight, and normal inflammatory indexes and blood chemistry were included in the study. Urinary creatinine and UFCs were measured by a colorimetric method (5 ) and by a dichloromethane extraction RIA (DPC) (6 ), respectively, and mean values were calculated from the 3 days. ANOVA followed by the Fisher post hoc test was used to establish intergroup differences. Correlations between the UFC concentration and independent variables were established by univariate and multivariate linear regression analysis. Subjects were divided by urine volume into three groups: 500–1000 mL (n 5 28), 1001–1500 mL (n 5 27), and 1501–2500 mL (n 5 33). Among groups, no differences were found in UFC values (39.2 6 3.07, 40.9 6 3.08, and 38.4 6 2.40 mg/day, respectively; P, not significant). In multiple linear regression, only mean plasma cortisol was positively associated with UFC concentration (r 5 0.27; P ,0.03). The UFC concentration was independent of age, body mass index, serum and urinary creatinine, and daily urine volume (Fig. 1). The timing of urine collection (24 h vs overnight), the different populations examined (adult women vs children), and differences in analytical method (use of a method that might be not suitable for UFCs) may partially account for the discrepancies with previous findings (3 ). Our results indicate that in healthy women, the UFC concentration is not related to daily urine output within the physiological range, i.e., 500–2500 mL. These variations in urine volumes, most frequently encountered in clinical practice, are therefore unlikely to exert any significant influence on UFC values. Further studies, hopefully based on chromatographic methods, are needed to define the variables that influence urinary free cortisol in oliguric and polyuric states.