Measurement of urinary free cortisol using liquid chromatography-tandem mass spectrometry: comparison with the urine adapted ACS:180 serum cortisol chemiluminescent immunoassay and development of a new reference range.

Background: The measurement of urinary free cortisol (UFC) is commonly used in the investigation of possible Cushing's syndrome. With the recent availability of liquid chromatography-tandem mass spectrometry (LC-MS/MS) in hospital laboratories, we wanted to develop a specific UFC LC-MS/MS method and compare it with our current immunoassay method and develop a new LC-MS/MS reference range if required.

Methods: A UFC LC-MS/MS method using deuterated cortisol as an internal standard was optimized using solid-phase extraction as a clean-up procedure. The multiple reaction-monitoring transitions used for the detection of cortisol and deuterated cortisol were 363.1 > 121 and 365.1 > 121.8, respectively. The method was investigated regarding precision, linearity, sensitivity, recovery and interference. UFC was measured by the in-house urine adapted ACS:180 serum cortisol immunoassay and the developed LC-MS/MS method in 110 urine samples from patients being investigated for possible Cushing's syndrome.

Results: The within-batch precisions (n = 25) of the LC-MS/MS method were 7.6%, 4.5% and 3.3% at 25.0 nmol/L, 49.6 nmol/L and 344.6 nmol/L, respectively; the between-batch precisions (n = 10) were 9.4%, 9.4% and 8.4%, respectively, at these concentrations. The method is sensitive down to 5 nmol/L and linear up to at least 1000 nmol/L. The method showed adequate cortisol recovery and no interference from the numerous drugs and steroids tested. The total run time for 20 samples, including sample preparation, was 120 min. A scatter plot of paired UFC measurements on the LC-MS/MS and the ACS:180 gave the equation: LC-MS/MS = 0.408 (ACS:180) + 2.65, r2 = 0.6664. The 24-h measured UFC results on 110 samples (25 men and 85 women) were positively skewed. After log transformation the data were less skewed, and following back transformation of the lower 97.5th centile, the upper limit of normal was 165 nmol/24 h. The 95th centile of the untransformed data was 146 nmol/24 h (n = 110, 25 men and 85 women). Separated by sex, the 95th centile was 152 nmol/24 h for men (n = 25) and 141 nmol/24 h for women (n = 85).

Conclusions: We have developed a UFC LC-MS/MS method with a solid-phase extraction clean-up step. The method shows adequate performance and is suitable for routine laboratory use. The mixed sex (n = 110, men = 25, women = 85) reference range was up to 165 nmol/24 h or 146 nmol/24 h, depending on how the data are manipulated.