Radioimmunoassay of steroid hormones is widely used but in application to clinical samples the technique can give erroneous results. In the diagnosis of congenital adrenal hyperplasia (CAH) due to 21-hydroxylase deficiency raised plasma levels of 17-hydroxyprogesterone (17-OHP) are supportive evidence for the disorder. In the immunoassay of 17-OHP there are particular difficulties with the assay of plasma from neonates due to interference from steroids such as 17-hydroxypregnenolone sulphate. Consequently most laboratories will not measure the steroid in the first 3 days of life and when a high result is recorded the sample should be processed again after solvent extraction that leaves conjugated steroids in the aqueous phase. We have developed a chemical method for 17-OHP based on isotope dilution (d2-17-OHP), solvent extraction and Sephadex LH-20 chromatography. Steroids are converted to heptafluorobuyrate derivatives before analysis with gas chromatography coupled to a mass spectrometer in selected ion monitoring mode. The recovery of steroid in spiked samples is 96-104%. The assay is based in general on 50 microlitres of sample. We endeavour in samples from newborn infants with ambiguous genitalia to get a comfirmed result within 24h of sample receipt. Blood samples may be taken from 12h after birth. Interassay variation is 1.6, 0.9, 1.4 and 1% at 4.8, 10.1, 31.2 and 85.9 nmol/L. Performances in 2 external quality schemes (monthly UKNEQAS - A score less than 50; fortnightly BioRad) are better in terms of accuracy and precision compared with radioimmunoassays. Twenty samples a day can be processed with the existing protocol. The technology can be refined for analysis of this and other intermediate steroids separately and together with further imrpovements predicted using LC and tandem mass spectrometry.
22 - 24 Mar 2004
British Endocrine Societies