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


International Survey

Responses were received from 766 society members (11% overall response rate), working in 60 countries (single response received from 19 countries). Response rates varied between the societies: PES, 21% (n = 290), SfE, 19% (n = 220), ESE, 13% (n = 220), ESPE, 3% (n = 36), ESA, <1% (n = 7) and APEG, 4% (n = 13). Responses were received from clinicians working in the USA (36%), UK (29%), mainland Europe (25%), North America (excluding the USA; 4%), Asia (3%), Australasia (3%), Africa (<1%) and South America (<1%). Endocrinologists who worked mainly or entirely with adults made up 52% of respondents and 45% worked mainly or entirely with children and/or adolescents (97% of USA respondents). The remaining 3% of respondents either did not indicate their patient base or were not clinicians.

The SST was the most popular test for assessing adrenal insufficiency (Table 2). It was used by 98% overall with 92% using the HDT, 43% the LDT and 37% both. The LDT was considerably more popular amongst paediatric endocrinologists (72%) compared with adult endocrinologists (17%). There was variation in LDT utility amongst respondents from different geographical regions: 76% of all respondents working in the USA used the LDT, 50% from the Middle East, 34% from mainland European countries, 30% from Australasia and 6% from the UK (82% UK paediatric endocrinologists in 2012 survey, not resurveyed). The most commonly utilised LDT dose was 1 μg (86% of question respondents) and an intermediate dose (between 5 and 15 μg) was used by 8%. Body surface area–based doses (0.1 –1 μg/m2) were used by 5%, 2% used weight–based calculations.

Respondents stated their rationale for using the HDT or LDT: the most popular reasons for using the HDT were diagnosis of primary adrenal insufficiency and congenital adrenal hyperplasia, or because it was standard procedure. The LDT was preferred to investigate secondary adrenal insufficiency. The majority administer the HDT by the intravenous route (81%), with 37% and 5% using intramuscular and subcutaneous routes, respectively.

Thirty different combinations of cortisol sampling times were specified for the HDT and 37 for the LDT (Figure 1). The most common times to sample were at 0, 30 and 60 minutes (HDT 46%, LDT 51%), while 17% of LDT respondents utilised a 20–minute sample in their protocol. The most commonly used interpretive threshold for adequacy of adrenal function (a "pass") was >500 nmol/L, used in 48% of HDT and 61% of LDT. More HDT users (27%) than LDT users (11%) utilised the higher threshold of >550 nmol/L. Similar proportions used thresholds below 500 nmol/L: HDT, 21% (range 374–475 nmol/L), and LDT, 25% (range 380–495 nmol/L).

Figure 1.

Chosen cortisol sampling times for respondents using high–dose and low–dose synacthen tests. Each bar represents the percentage of respondents (HDT, n = 716, and LDT, n = 284) who measure cortisol levels at the times provided. For clarity, not all combinations of timings have been included in the graph (HDT, n = 30 different combinations and, LDT, n = 37). HDT, high–dose test; LDT, low–dose test

Serum cortisol levels without stimulation were used in the diagnosis of adrenal insufficiency by 76% (Table 2). When asked to specify further (n = 290), 92% used morning serum cortisol and 19% random cortisol sampling. Paired ACTH and serum cortisol sampling was used by 71% of all respondents. Less popular tests included the insulin tolerance test (used by 36% of respondents: adult, 54%; paediatric 15%), glucagon stimulation test (27%), metyrapone test (4%), clonidine stimulation test (3%), corticotrophin releasing hormone test (2%) and depot (prolonged) synacthen test (1%).

Low–dose Synacthen Dilution Study

For 8 of the ten different dilution strategies, a marked intramethod variability in the final synacthen concentration was observed, with CVs of over 10% (Table 1). The least variable was method 6, with a CV of 2.1%; the most variable was method 10, with a CV of 109%. Optimal dilution would have yielded synacthen concentrations able to deliver a dose close to 1 μg (acceptable range, 0.9–1.1 μg). However, the method means ranged from 0.16 μg (least accurate) to 0.81 μg (most accurate; Table 1). The methods bearing results closest to the range chosen as acceptable were 1, 4 and 6 (Figure 2). Three methods (7, 9 and 10) had a mean concentration of less than half the expected dose ranging from 0.16 to 0.36 μg (Figure 2), reflecting substantial losses of synacthen. To assess any variation or reduction in synacthen dose resulting from the laboratory dilutions necessary for the ELISA quantification, 2 vials of synacthen (250 μg/mL) were diluted and samples run over 23 assays. This yielded results of 247 ± 11 μg/mL and 223 ± 12 μg/mL, and indicated that the wide variation in deliverable dose detected in samples was not due to inaccuracies in the required laboratory dilutions.

Figure 2.

Accuracy and variability in 1 μg low–dose synacthen dilution methods. For each method tested, except method 7, each individual point indicates the mean deliverable amount of synacthen as calculated from 3 samples taken from the final bag/syringe dilution. For method 7, each individual point relates to a single sample measurement. Each method mean was calculated from 5 separate dilution experiments and is depicted by a short black line. The unbroken line at 1 μg represents the expected amount of synacthen administered if dilutions were optimal. The broken lines represent the upper (1.1 μg) and lower (0.9 μg) limits of the accepted range of dose variability of ± 10%

Intrabag/syringe variability was high but unpredictable, with no part of the bag/syringe tending towards higher concentrated samples than another. Overall, top samples (n = 45) had a mean ± SD deliverable dose of 0.593 ± 0.298 μg synacthen, CV of 50.2%, middle samples (n = 45) 0.545 ± 0.286 μg, 52.5% and bottom samples (n = 45) 0.573 ± 0.293 μg, 51.3%.

Method 6 was the only one to use 5% dextrose as a diluent and was the least variable method (CV of 2.1%) and most accurate, with means closest to the desired 1 μg (0.79–0.84 μg). Six methods (n = 90 samples) involved a single dilution step, and together had a mean synacthen deliverable dose of 0.547 ± 0.319 μg, whilst 4 methods (n = 50) used double dilutions with an overall mean of 0.583 ± 0.24 μg (P = .46; 95% confidence interval (CI): −0.058 to 0.131 μg). When comparing the different initial volumes of the 1 mL ampoule of 250 μg/mL synacthen used for dilution, 6 methods (n = 90) used all 1 mL and resulted in a mean synacthen deliverable dose of 0.668 ± 0.212 μg. The remaining 4 methods (n = 50) used 0.5 mL or less and had a mean synacthen deliverable dose of 0.365 ± 0.318 μg (P < .0001; 95% CI: −0.404 to −0.204 μg). A bag of diluent, rather than a syringe, was utilised in 8 of the methods (n = 120), 4 of which (n = 60) used a large volume of diluent, ≥250 mL, and had a mean synacthen deliverable dose of 0.572 ± 0.314 μg, and 4 methods (n = 60) used a small volume of diluent, 50 mL, yielding a mean synacthen deliverable dose of 0.584 ± 0.283 μg (P = .837; 95% CI: −0.097 to 0.119 μg).