Diagnostic performance of morning basal serum cortisol for assessing adrenal reserve: a single-center experience from the United Arab Emirates

Abstract

Introduction:

Adrenal insufficiency can be life-threatening due to the lack of cortisol elevation in times of stress. The short Synacthen test (SST) is the most common diagnostic test for adrenal insufficiency. This study aimed to assess the diagnostic performance of morning basal serum cortisol during an SST to assess normal adrenal reserve in a local Emirati population.

Methods:

We retrospectively analyzed the electronic medical records of adult patients who underwent morning SST to assess adrenal reserve between August 2012 and August 2022.

Results:

This study included 344 patients (201 women) with a mean age of 49.1 ± 22.6 years. Based on the previously published cutoff data for SST peak values at 30 min (408 nmol/L for Beckman-Access used from 2012 to 2017 and 402 nmol/L for Roche-Cobas Generation-II used since 2018) as a gold standard to identify patients with adequate adrenal reserve, 106 patients (30.8%) were diagnosed with adrenal insufficiency and 238 (69.2%) with adequate adrenal reserve. Using the receiver–operator characteristics curve for morning cortisol, we identified a cutoff of 332 nmol/L, which corresponded to adequate Synacthen stimulation with 100% specificity and 73% sensitivity. Morning cortisol levels of <96 nmol/L corresponded to suboptimal adrenal response to SST, with 95% specificity and 55% sensitivity.

Conclusion:

We propose that a morning cortisol cutoff of >332 nmol/L can help identify patients with adequate adrenal reserve, thereby avoiding unnecessary SSTs.

Plain language summary

Using morning cortisol test to identify insufficiency

Adrenal are small glands present above the kidneys and they produce stress hormone called cortisol. The lack of this stress hormone can be life-threatening so it is important to identify this. Currently there is a cumbersome, time-consuming and expensive test called short synacthen test which is done to identify stress hormone deficiency, however our study aims to use a simple and robust basal cortisol test that can be done in clinic setting to identify adequate stress hormone reserve.

Keywords

Addison’s disease adrenal insufficiency adrenal reserve diagnostic cutoff morning cortisol short Synacthen test

Introduction

Adrenal insufficiency is a potentially life-threatening condition due to the lack of inherent cortisol elevation in response to any illness or stress. Adrenal insufficiency may still present as a diagnostic dilemma for healthcare providers, even 165 years after its first description in 1856.¹ It can be directly related to reduced adrenal secretion of cortisol (primary), interference in the secretion of adrenocorticotropin hormone (ACTH) from the pituitary gland (secondary; e.g., after traumatic brain injury or pituitary surgery), or a disruption in the secretion of corticotropin-releasing hormone by the hypothalamus (tertiary, e.g., after exogenous steroid intake).² The cosyntropin/short Synacthen test (SST), which involves testing of serum cortisol before and after injecting with 250 mcg cosyntropin (Tetracosactide), is frequently used to diagnose adrenal insufficiency and has replaced the gold standard insulin tolerance test in clinical practice.³ Determination of a morning cortisol value can help identify patients with sufficient adrenal reserve without the need for SSTs, and it has been proposed as a useful initial screening test; however, a consensus has not yet been established regarding the cutoff values due to varied results caused by different patient profiles, cortisol assay kits, and test timings.⁴ Recent guidelines have recommended a morning cortisol cutoff of 140 nmol/L paired with ACTH as a screening test for adrenal insufficiency. Other observational studies have indicated that morning cortisol levels of ⩽100 and ⩾450 nmol/L are useful markers to rule in and rule out the possibility of adrenal insufficiency with reasonable diagnostic accuracy. These tests may reduce the need for SSTs; however, the cutoff values for local population have not yet been verified.^5,6

This study aimed to assess the diagnostic performance of morning basal to peak cortisol levels during the SST and identify a morning cortisol threshold that can identify patients with adequate adrenal reserve in our local population with >98% specificity and >75% sensitivity.

Patients and methods

This was a retrospective study of all adult patients (aged ⩾16 years) who underwent SSTs in a tertiary care hospital in the United Arab Emirates (UAE) from August 2012 to August 2022. All tests were performed between 8 am and 11 am by well-trained medical nurses in outpatient or inpatient (excluding intensive care unit) clinical settings. Serum cortisol levels were measured at 0 (basal), 30, and 60 min after intravenous administration of 250 μg ACTH (Synacthen). All patients were identified using the electronic coding for medical procedures for SST. As per the sample size and study power calculations with margin of error of 3% and 95% confidence interval, we aimed to analyze a minimum of 203 patients who underwent SST. We identified 435 patients using the inclusion criteria, who had SST performed during the study period, and their electronic medical records were reviewed and confirmed by two physicians to ensure data quality. In case of any conflicting information, a specialist endocrinologist verified and confirmed the information pertinent to SST results and clinical information. Ninety-one patient records were excluded due to missing data regarding morning basal cortisol measurements or other incomplete clinical or biochemical data. Of the remaining 344 selected patients, 197 had undergone cortisol measurement using the Beckman-Access Assays from 2012 to 2017 while 147 underwent cortisol measurement using Roche-Cobas Generation-II from 2018 to 2022. The Beckman-Access Cortisol assay utilized an immunoenzymatic method with cross-reactivities for other steroids of up to 23.9% and intra-assay and inter-assay variabilities of 4.7% and 5.7%, respectively.⁷ The Roche-Cobas Generation-II assay employed an electrochemiluminescence competitive immunoassay method with all-steroid cross-reactivities of <13% and low intra-assay and inter-assay variabilities of 1.7% and 2.8%, respectively.⁸ The kit-specified and recently published new cutoff data for SST peak values at 30 min (⩾408 nmol/L or ⩾14.8 μg/dL for Beckman-Access and ⩾402 nmol/L ⩾14.8 μg/dL for Roche-Cobas Generation-II) were used to identify patients with adequate adrenal reserve.⁹ As per our SST local protocol, all patients undergoing SST had no glucocorticoids given within 24 h of testing and were not on oral contraceptive pills or having an acute stress state (like sepsis) that could impact their SST results.

Additional data extracted from the electronic medical records included demographic characteristics, baseline clinical and chemical results within 3 months before or after SST, and the reason for the SST. Data on comorbidities were retrieved from the medical notes and confirmed using the list of persistent diseases in the electronic medical records.

No patient identifiable data other than medical record numbers were collected and patient confidentiality was maintained throughout the study. The regional Research Ethics Committee approved the study and waived the requirement for consent due to the retrospective nature of the study and the lack of any intervention.

The continuous variables were presented as means and standard deviations or medians, and categorical variables were presented as numbers and percentages. The diagnostic performance of baseline cortisol levels to identify preserved adrenal function was evaluated using the receiver–operator characteristics (ROC) curves and area under the ROC analyses. Data were analyzed using the IBM SPSS statistical package for Windows, version 29 IBM Corp, Armonk, NY, USA. p-Values of <0.05 were considered to indicate statistical significance.

Results

We identified 344 patients who underwent SSTs for suspected adrenal insufficiency. The baseline characteristics are depicted in Table 1. The mean age of the patients was 49.1 ± 22.6 years, 58.4% were women, and 63.5% were Emirati. Most SSTs were performed as inpatient measurements (58.2%). The major indication for performing an SST was the presence of symptoms suggestive of possible adrenal insufficiency (51.2%). Approximately one-third of the patients had diabetes or hypertension as comorbidities. Using the previously published cutoff for SST at 30 min with cutoffs ⩾408 nmol/L for Beckman-Access and ⩾402 nmol/L for Roche-Cobas Generation-II assays,^7
–9 to define preserved adrenal function. In total, 106 (30.8%) patients were diagnosed with adrenal insufficiency. No significant differences in baseline characteristics, comorbidities, vital signs, or biochemical characteristics were detected between patients with adrenal insufficiency and patients with adequate adrenal reserve.

Table 1.

Clinical and biochemical features of patients undergoing the short Synacthen test (n = 344).

Variables	Total	Adrenal insufficiency present, 106 (30.8%)	Adrenal insufficiency absent, 238 (69.2%)	p-Value
Age, mean ± SD	49.1 ± 22.6	49.8 ± 22.3	47.4 ± 22.6	0.53
Female, n (%)	201 (58.4%)	57 (53.8%)	144 (60.5%)	0.18
Nationality, n (%)
Emirati	218 (64.2%)	60 (56.6%)	158 (66.4%)	0.20
Non-Emirati	126 (35.8%)	36 (43.4%)	80 (33.6%)
Short Synacthen clinical setting: n (%)
Inpatient	221 (58.2%)	69 (65.1%)	152 (63.9%)	0.83
Outpatient	123 (32.2%)	37 (34.9%)	86 (36.1%)
Indications for short Synacthen: n = 337 (%)				0.17
Previous suppressive steroid treatment	50 (14.5%)	22 (20%)	28 (12.3%)
Low morning cortisol	47 (13.7%)	16 (15.4%)	31 (13.1%)
Symptoms suggestive of adrenal insufficiency	176 (51.2%)	37 (35.6)	141 (59.5%)
Pituitary tumors/hypophysitis	22 (6.4%)	12 (11.5%)	11 (4.6%)
Other central nervous system tumors	10 (2.9%)	5 (4.8%)	5 (2.1%)
Cranial radiotherapy	4 (1.2%)	3 (2.9%)	1 (0.4%)
Adrenal disease	21 (6.1%)	9 (8.7%)	13 (5.5%)
Postoperative following pituitary surgery	7 (2.1%)	3 (2.9%)	4 (1.7%)
Comorbidities
Diabetes mellitus (n = 338)	126 (37.3%)	40 (36.6%)	86 (38.8%)	0.70
Hypertension (n = 341)	117 (34.3%)	35 (33.7%)	82 (34.6%)	0.87
Coronary artery disease (n = 340)	49 (14.4%)	13 (12.5%)	36 (15.3%)	0.51
Malignancy (n = 336)	72 (21.4%)	24 (23.8%)	48 (20.4%)	0.49
Vital signs
Heart rate (beat/min)	80 ± 12.4	79.6 ± 13.1	80.3 ± 12.1	0.95
Systolic blood pressure (mm Hg)	118.4 ± 20.3	112.8 ± 19.1	116.7 ± 20.7	0.06
Diastolic blood pressure (mm Hg)	68.8 ± 13.8	71.0 ± 11.2	67.8 ± 14.9	0.22
BMI (kg/m²)	29.7 ± 26.7	27.5 ± 7.3	30.8 ± 32.3	0.45
Biochemical results
Serum sodium (mmol/L)	136.1 ± 6.1	135.9 ± 6.4	136.3 ± 6.0	0.81
Serum potassium (mmol/L)	4.1 ± 0.6	3.9 ± 0.5	4.1 ± 0.6	0.07
Serum urea (mmol/L)	6.1 ± 6.4	5.3 ± 5.7	6.5 ± 6.6	1.02
Serum creatinine (µmol/L)	95.3 ± 111.6	89.4 ± 98.8	98.1 ± 117.4	1.00
Serum glucose (mmol/L)	7.0 ± 3.3	6.8 ± 3.1	7.2 ± 3.4	1.02
Short Synacthen test results
ACTH (n = 246) (pmol/L)	8.7 ± 29.3	12.1 ± 45.1	7.2 ± 18.9	0.21
Baseline cortisol (nmol/L)	228.8 ± 158.7	127.0 ± 144.4	276.0 ± 158.1	<0.001
Cortisol 30 min (nmol/L)	484.0 ± 357.3	265.8 ± 144.9	596.1 ± 381.6	<0.001
Cortisol 60 min (nmol/L)	561.2 ± 335.2	295.5 ± 138.9	694.0 ± 325.5	<0.001

ACTH, adrenocorticotropin hormone.

Bold p-value of < 0.05 is significant.

According to the ROC curve analysis of the diagnostic performance of basal cortisol, a cutoff of ⩾332 nmol/L corresponded to adequate adrenal reserve and normal SST with a specificity of 100% (Figure 1(a)), sensitivity of 73%, and area under the ROC of 0.86 with 95% confidence interval of 0.818–0.902.

Figure 1.

(a) Area under the ROC curve using a basal cortisol cutoff of ⩾332 nmol/L, corresponding to adequate adrenal reserve with a normal SST exhibits a specificity of 100% and a sensitivity of 73%. (b) Area under the ROC curve using a basal cortisol cutoff of <96 nmol/L corresponding to adrenal insufficiency with an inadequate response on the SST exhibits a specificity of 95% specificity and a sensitivity of 55%.

Selecting a lower cutoff value for basal cortisol of ⩾280 nmol/L in this model yielded a reduced specificity of 95% and sensitivity of 60%, which can be an acceptable trade-off if clinical acumen is applied. A cutoff of <96 nmol/L corresponded to adrenal insufficiency, which was confirmed on the SST, with a specificity of 95% (Figure 1(b)) and sensitivity of 55%. Therefore, if the clinical picture was suggestive of adrenal insufficiency, patients with a baseline cortisol level of <96 nmol/L can suggest adrenal insufficiency. In a separate sub-analysis, the morning cortisol cutoff of >332 nmol/L corresponded to adequate adrenal reserve on the SST with 100% specificity for both inpatient and outpatient SSTs, however, the sensitivity was 61.8% for inpatient (n = 221) and 88.4% for outpatient (n = 123), SSTs.

Among the 344 patients included in the study, 197 (57.3%) underwent SSTs using the Beckman-Access assay, of which 43 (21.8%) were diagnosed with adrenal insufficiency. The other 147 patients (42.7%) underwent SSTs using the Roche-Cobas Generation-II assay, of which 63 (44.1%) were diagnosed with adrenal insufficiency. Of the 238 patients with available 60-min cortisol levels, only 6 (n = 238, 2.5%) were at borderline at 30 min and passed clearly at 60 min; and all these patients were tested using the Beckman-Access assay and were classed as adrenally sufficient. There were no patients using the Roche-Cobas Generation-II tests who could not achieve the cutoff at 30 min but achieved it later at 60 min during the SST, indicating that 30-min test was sufficient for all these patients.

Discussion

This is the first study to assess the diagnostic performance of morning cortisol in predicting adequate response to the SST in the UAE population. SST is a time-consuming test, requiring the patient to stay in the clinic for >60 min. Many efforts have been made to develop a more time-efficient and cost-effective screening test. The modified SST (which measures the 30-min cortisol level) exhibits a diagnostic accuracy of 85.4%.¹⁰ In a previous study, morning cortisol levels of ⩾354 nmol/L predicted normal adrenal reserve with 100% sensitivity; thus, SSTs could be avoided in 37% of patients.¹⁰ In a study involving 122 patients, 60-min cortisol levels were unnecessary and added no value compared with the 30-min cortisol levels determined using the SST. Avoiding the 60-min cortisol can be cost-efficient, potentially saving £11 per patient.¹¹ In addition, indirect costs related to spending time in the clinic can be avoided, and healthcare staff time can be used more effectively.

The newer generations of assays exhibit greater specificity for cortisol and less reactivity to other steroids compared with the old polyclonal antibody assays. Therefore, following ACTH stimulation, the new peak 30-min cortisol cutoff values of 402 and 408 nmol/L have been established for Roche II and Abbott assays, respectively.⁹ The new cutoff values were used in our study to define SST pass. The 60-min cortisol levels determined using both assays were higher than the 30-min results. Based on our results, we propose that the 60-min cortisol measurement on the SST is unnecessary for the Roche-Cobas Generation-II kit, as 100% (n = 147) of patients passed the 30-min cortisol test. The 30-min cortisol measurement on the SST exhibits similar accuracy and saves both time and kit resources associated with the additional 60-min cortisol test on the SST. Even for the older Beckman-Access kit, 97% (191 of 197) of patients did not require 60-min cortisol measurement.

Several studies have focused on the diagnostic performance of the baseline morning cortisol levels as a predictor of adrenal reserve. Variable cortisol cutoff values, based on the assay type, test setting (outpatient vs inpatient), timing, and underlying etiology, have been reported. Sbardella et al.¹² reported that baseline cortisol levels for predicting SST with 100% specificity were 358 nmol/L for Siemens, 336 nmol/L for Abbott, and 506 nmol/L for Roche I in the outpatient setting with 30-min cortisol peaks to pass the defined test cutoff values of >450, >430, and >550 nmol/L, respectively. Another study using the Abbott assay confirmed that pretest morning cortisol (⩾343 nmol/L) can be a reliable predictor of SST outcome.¹³ A recent study evaluating home waking salivary cortisone to screen for adrenal insufficiency demonstrated that the comparator baseline serum cortisol cutoff value (⩾308 nmol/L) using the Roche II assay predicted normal SST response (peak cortisol defined as 430 nmol/L) with 96% specificity.¹⁴ Our results, incorporating the Abbott and Roche II assays, are in line with the previous studies; a baseline morning cortisol level of ⩾332 nmol/L corresponds to adequate adrenal reserve and obviates the need for ACTH stimulation.

Regarding a lower cutoff value for morning serum cortisol levels to confirm adrenal insufficiency, no clear consensus has been reached, and several studies highlighted the limitations of using baseline cortisol levels with variable cutoff values to confirm adrenal insufficiency.^4,15 Some studies suggest that SST adds little value in patients having 9 am serum cortisol level of <100 nmol/L,¹⁶ and the Endocrine Society guidelines support the use of 9 am serum cortisol level of <140 nmol/L.¹⁷ In an observational study, Fragoso et al.¹⁸ reported that morning cortisol values of 42–100 nmol/L confirmed adrenal insufficiency with 98%–100% specificity. Our results demonstrated that a morning cortisol value of ⩽96 nmol/L can identify patients with adrenal insufficiency with low sensitivity albeit high specificity.

Our study incorporated both outpatient and inpatient SSTs. While the SST results may be quite different in these two settings, all inpatients included in the study were stable (not in intensive care unit settings or having any sepsis/acute severe infection), and all patients shared similar indications for performing SST. Therefore, combining these patient populations was practical, given our small sample size.

Strengths and limitations

This is the first study in UAE to assess the diagnostic performance of morning basal serum cortisol to identify adequate adrenal reserve. All consecutive SST tests were performed in adults at a single center. Nevertheless, this study had several limitations. First, this was a retrospective study, which has inherent limitations, including missing data. Second, this study included all SSTs in both inpatient and outpatient settings due to the small sample size. Third, specific etiologies of adrenal insufficiency (like previous steroid use) may have different results and cutoffs. Fourth, due to the smaller number of patients, our study proposed a combined basal cortisol cut-off for the two assays used, Beckman-Access and Roche-Cobas Generation-II assays, although they both have slight variation in their previously published cutoff for SST at 30 min with cutoffs ⩾408 and ⩾402 nmol/L for Beckman-Access and Roche-Cobas Generation-II assays, respectively.

In terms of the characteristics and performance of the two tests, despite that the cross-reactivity, intra-assay and inter-assay variabilities looked better for Roche-Cobas Generation-II assay, the difference is small compared to Beckman-Access assay and both are widely used assays. Therefore, we believe that combining the data from the two assays may not affect the reliability of the results.

Future studies across different centers in the UAE with larger sample sizes are suggested which may allow subgroup analysis based on etiology and assays.

Conclusion

Our study provides the morning serum cortisol cutoff values for adrenal insufficiency and for normal adrenal reserve in our local population. Utilizing morning serum cortisol levels at these cutoff values can predict the outcome of SST in most of our patients. However, clinical acumen is required especially in patients with recent pituitary surgery or suspicion of secondary/tertiary adrenal insufficiency, and morning basal cortisol should not automatically replace the need for SST. In addition, the 60-min cortisol measurement may be unwarranted with the current cortisol assays such as the Roche-Cobas Generation-II and Beckman-Access assays. Thus, the modified SST (basal and 30-min test only) can be a time-saving alternative. Further large-scale prospective studies from neighboring regions are warranted to validate our findings.

Supplemental Material

sj-docx-1-tae-10.1177_20420188251335180 – Supplemental material for Diagnostic performance of morning basal serum cortisol for assessing adrenal reserve: a single-center experience from the United Arab Emirates

Supplemental material, sj-docx-1-tae-10.1177_20420188251335180 for Diagnostic performance of morning basal serum cortisol for assessing adrenal reserve: a single-center experience from the United Arab Emirates by Raya Almazrouei, Jawaher Alshamsi, Mohamed Abdelsalhen, Fatima Alkaabi and Adnan Agha in Therapeutic Advances in Endocrinology and Metabolism

Footnotes

Acknowledgements

None.

Declarations

ORCID iDs

Raya Almazrouei

Adnan Agha

Supplemental material

Supplemental material for this article is available online.

References

On the constitutional and local effects of disease of the supra-renal capsules. Br Foreign Med Chir Rev 1856; 18(36): 404–413.

Martin-Grace

Dineen

Sherlock

, et al. Adrenal insufficiency: physiology, clinical presentation and diagnostic challenges. Clin Chim Acta 2020; 505: 78–91.

Gleeson

Walker

Seckl

, et al. Ten years on: safety of short synacthen tests in assessing adrenocorticotropin deficiency in clinical practice. J Clin Endocrinol Metab 2003; 88: 2106–2111.

Montes-Villarreal

Perez-Arredondo

Rodriguez-Gutierrez

, et al. Serum morning cortisol as a screening test for adrenal insufficiency. Endocr Pract 2020; 26(1): 30–35.

Struja

Briner

Meier

, et al. Diagnostic accuracy of basal cortisol level to predict adrenal insufficiency in cosyntropin testing: results from an observational cohort study with 804 patients. Endocr Pract 2017; 23(8): 949–961.

Bornstein

Allolio

Arlt

, et al. Diagnosis and treatment of primary adrenal insufficiency: an endocrine society clinical practice guideline. J Clin Endocrinol Metab 2016; 101: 364–389.

RRID:AB_2802133. Beckman Coulter Cat# 33600, 2024. https://scicrunch.org/resolver/AB_2802133 (accessed 13 April 2025).

RRID:AB_2811288. Roche Cat# 11875116, https://scicrunch.org/resolver/AB_2811288 (2024, accessed 13 April 2025).

Javorsky

Raff

Carroll

, et al. New cutoffs for the biochemical diagnosis of adrenal insufficiency after ACTH stimulation using specific cortisol assays. J Endocr Soc 2021; 5(4): bvab022.

10.

Kumar

Carr

Wassif

Diagnostic performance of morning serum cortisol as an alternative to short synacthen test for the assessment of adrenal reserve; a retrospective study. Postgrad Med J 2022; 98(1156): 113–118.

11.

Dulaimi

Myint

Short synacthen tests: short 0 and 30 minutes versus conventional 0, 30 and 60 minutes. Clin Med (Lond) 2019; 19(Suppl. 2): 95.

12.

Sbardella

Isidori

Woods

, et al. Baseline morning cortisol level as a predictor of pituitary-adrenal reserve: a comparison across three assays. Clin Endocrinol (Oxf) 2017; 86(2): 177–184.

13.

Mathara Diddhenipothage

SAD

Beck

Loo

, et al. A morning cortisol is the most effective clinical predictor of short synacthen test outcome: a tertiary care centre experience. Clin Endocrinol (Oxf) 2023; 99(2): 142–151.

14.

Debono

Elder

Lewis

, et al. Home waking salivary cortisone to screen for adrenal insufficiency. NEJM Evid 2023; 2(2): EVIDoa2200182.

15.

Le Roux

Meeran

Alaghband-Zadeh

Is a 0900-h serum cortisol useful prior to a short synacthen test in outpatient assessment?

Ann Clin Biochem 2002; 39(Pt 2): 148–150.

16.

Mackenzie

Giford

Boyle

, et al. Validated criteria for the interpretation of a single measurement of serum cortisol in the investigation of suspected adrenal insufficiency. Clin Endocrinol 2019; 91(5): 608–615.

17.

Bornstein

Allolio

Arlt

, et al. Diagnosis and treatment of primary adrenal insufficiency: an endocrine society clinical practice guideline. J Clin Endocrinol Metab 2016; 101(2): 364–389.

18.

Fragoso Perozo

AFD

Fontes

Lopes

, et al. Morning serum cortisol role in the adrenal insufficiency diagnosis with modern cortisol assays. J Endocrinol Invest 2023; 46(10): 2115–2124.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB