International Survey on High– and Low–Dose Synacthen Test and Assessment of Accuracy in Preparing Low–Dose Synacthen

Alexandra S. Cross; E. Helen Kemp; Anne White; Leanne Walker; Suzanne Meredith; Pooja Sachdev; Nils P. Krone; Richard J. Ross; Neil P. Wright; Charlotte J. Elder


Clin Endocrinol. 2018;88(5):744-751. 

In This Article


This is the largest international survey of diagnostic tests for adrenal insufficiency to date. Although the response rate of 11% was low, this was a survey of society members some of whom are not in clinical practice and the response rate is in keeping with similar internet surveys.[15,16] There were geographical variations in responses. Not all endocrine societies approached distributed the survey and this has contributed to the imbalance in paediatric and adult endocrinologist responses from certain regions.

The SST was the most popular test for assessing HPA axis function and has been growing in popularity amongst endocrinologists, increasing from 24% in 1988,[1] 69% in 1993,[2] 59% in 2005,[17] to 98% in this survey and 100% of paediatric endocrinology centres in the UK in 2012.[3] It is regarded now as the "standard" test for adrenal insufficiency.[4] This is the first international survey to distinguish proponents of the HDT from the LDT. Whilst the HDT is used by 92% of respondents, and is the test of choice for diagnosing primary adrenal insufficiency, the LDT is used by 43%. Similar proportions of survey respondents practised as adult and paediatric endocrinologists. The LDT is popular amongst paediatric endocrinologists, 72% compared with 17% of adult endocrinologists, resonating the results of the British Society for Paediatric Endocrinology and Diabetes (BSPED) survey, where 82% used the LDT.[3] This may reflect respiratory guidelines, which recommend the LDT for assessment of adrenal function in children on inhaled corticosteroids.[18,19]

The sampling times and diagnostic cut–offs practised by the majority of respondents were in keeping with Endocrine Society guidelines,[4] which state a peak cortisol less than 500 nmol/L at 30 or 60 minutes indicates adrenal insufficiency. Deviations from these guidelines were seen in 52% of HDT and 39% LDT users for cut–off and <1% HDT and 5% LDT users for timing. The tendency to employ lower diagnostic thresholds for serum cortisol is likely to reflect a change in practice to locally derived cut–offs, dependent on the assay platform used. Additionally, clinicians review the SST results in the context of the clinical suspicion of adrenal insufficiency.[20,21]

Responses were received from people working in 60 countries and 6 continents, demonstrating a range of practises, resource settings and patient populations. There was a preponderance of responses from endocrinologists working in Europe and the USA; therefore, the survey may not be truly representative of worldwide practice. Additionally, national practice cannot be assumed in the 136 countries with no respondent and 19 countries with a single respondent.

This study has shown a high intermethod variability between different commonly employed dilution strategies for the low–dose SST. The variation in dose was from 0.16 to 0.81 μg when the dose should be 1 μg; thereby in all cases, the dilution methods used provide inadequate dosing, with doses up to 7–fold less than required. There was variation when the same method was used to make up the 1 μg dose 5 times (intramethod variability) and variation when individual samples from the same final solution were compared (intrabag/syringe variability), inferring inadequate mixing. This inaccuracy in dosing and variability between and within dilution methods may result in false–positive synacthen tests with potentially important clinical sequelae.

When similar methods (eg volume of diluent, proportion of synacthen ampoule used, number of dilution steps) were grouped and compared, only the initial volume of synacthen was shown to significantly affect the final concentration: dilution methods using the full ampoule gave significantly higher concentrations and closer to the desired concentration. The most accurate and least variable method was the only one to use 5% dextrose, suggesting that dextrose may be the most suitable diluent for making up low–dose synacthen. However, this would require further investigation along with other possible diluents for synacthen.

The plateau of the synacthen/cortisol dose–response curve is thought to begin at approximately 5 μg of synacthen.[12] The lowest dose of synacthen to maximally stimulate the adrenal gland has been found to be between 0.5 and 1 μg.[12,22–25] The supraphysiological dose of 250 μg of synacthen employed by the HDT means that even marked variation in the actual dose delivered to the patient is unlikely to manifest clinically. However, the doses employed in the LDT are much closer to the amounts needed to produce a maximal adrenal response and thus, small variations in the administered dose may have clinical ramifications, with the potential of false–positive diagnoses of adrenal insufficiency. Using the results of this study, a patient undergoing a 1 μg LDT, using dilution methods 7, 9 or 10, may receive between 0.16 and 0.36 μg of synacthen. These 3 methods used half or less of the synacthen ampoule, with methods 7 and 9 using 0.2 mL or less, a volume too small to draw up accurately using 1 mL ward syringes.

Intrabag/syringe variability was high but similar between different parts (top, middle, bottom), suggesting mixing inadequacy but no specific area the synacthen settled in. In laboratory practice, mixing of constituents similar to those used in this study may take place over many hours with the use of specialised equipment, to be assured of uniform distribution throughout the diluent.

There is no "standard" way to make up the 1 μg synacthen dose. The method of adding 250 μg/mL to 250 mL of 0.9% saline (method 3), described by Dickstein et al[12] on introducing the 1 μg test in 1991, was later recommended by the meta–analysis of Kazlauskaite et al,[8] but was neither the most popular method in the 2012 British survey[3] nor the most accurate method in the current study.

Other sources of variation have been considered. These include potential losses caused by the adherence of synacthen to plastic, reported to be between 21.6% and 58.6% and proportional to the length of the device.[13,14] This study made up low–dose synacthen under replicated ward conditions, using plastic syringes. Additional plastic laboratory equipment was used in the dilutions prior to ELISA analysis, potentially adding to the losses. However, the "control" samples diluted from a vial of synacthen with laboratory equipment showed very little variation and only minimal losses. Pharmaceutical industry standards require that an ampoule of 250 μg/mL synacthen contains between 95% and 105% of the declared content, 237.5 and 262.5 μg, respectively (Mallinckrodt Pharmaceuticals, Dublin, Ireland), and this variation may be amplified when diluting the synacthen to physiological doses.

Ward, rather than specialised, calibrated laboratory equipment was used for simulation purposes, reflecting current clinical practice, but other variables were controlled as far as possible. The synacthen was kept refrigerated until the point of use and a single investigator performed all dilutions for each individual method. The additional dilutions required to run the samples on the ELISA were performed under strict laboratory conditions and by a single investigator. In the reality of a less controlled, busy clinical environment ambient temperatures may vary, synacthen may degrade in sunlight or if left out of the refrigerator and many different personnel may perform the dilutions, all potentially increasing the inaccuracy of dilution and variability further. A systematic review has shown preprepared syringes for intravenous medication can reduce errors in the preparation and administration by 21%.[26]

Our international survey showed the synacthen test is employed by 98% of endocrinologists, with 43% using the LDT. Our dilution study demonstrated considerable variation and inaccuracy when preparing the low dose of synacthen. The least variable methods were 1, 4 and 6 (Table 1). Although method 6 used 5% dextrose, the effect of diluent needs to be investigated further before any recommendations can be made. In addition, it would be expected that controlled laboratory/pharmacy conditions would impact positively on the accuracy of the delivered dose.