Introduction: Unstimulated cortisol is commonly used as a screening test for adrenal insufficiency. In the UK over the last decade there has been a large increase in the numbers of requests for cortisol being made in both primary and secondary care. To increase the specificity of an unstimulated cortisol, and thus reduce unnecessary referrals and Short Synacthen Tests, the recommendation is that an early morning cortisol (EMC) is performed between 08:00 and 09:00 h. There is evidence that an EMC below <160 nmol/l is highly predictive of failing the SST and the corollary is seen with an EMC above >340 nmol/l. We analysed our cortisol data over a six-month period to evaluate the proportion of samples taken outside the recommended time period and evaluate the effect of timing on the cortisol result.
Methods: A retrospective analysis was performed of all serum cortisol samples processed in our Trust between November 2017 and April 2018. Cortisol samples taken as part of a hypoglycaemia screen or SST were excluded. Cortisol quantification was performed on the Abbott Architect i1000 chemiluminescent immunoassay (CVs <5%). Based on published data the results were grouped into <160 nmol/l, 160339 nmol/l or >339 nmol/l and the time each sample was taken collected. Before 09:15 h (to allow some leeway) was considered early morning cortisol and after 09:15 h random cortisol (RC).
Results: Overall 226 serum cortisol samples were analysed, 50% (114) were EMC and 50% (112) RC. The EMC group resulted in 36% of samples <160 nmol/l compared to 64% of the RC group. The reverse was seen in results >340 nmol/l, with 67% from the EMC group and 33% from the RC group. Of samples collected before 0915 h 46% (52/114) were >340 nmol/l, thereby confidently excluding adrenal insufficiency, compared to 23% (26/112) of those samples taken after 09:15 h.
Conclusions: Taking cortisol samples before 09:00 h significantly increases the specificity of the screening for adrenal insufficiency and avoids unnecessary referrals to endocrinology and SSTs. As a result of our study we have introduced a new autocomment issued on all cortisol samples with interpretative comments ONLY provided for those timed before 09:00 h.
27 - 29 Nov 2019
British Society for Paediatric Endocrinology and Diabetes