Singapore Med J 2009; 50(10) : 962
O r i g i n a l A r t i c l e
Jalan Yaacob Latif,
Kuala Lumpur 56000,
Norasyikin AW, MD,
Norlela S, MD, MMed
Rozita M, MD, MMed
Masliza M, MBChB,
Shamsul AS, BScMed,
Nor Azmi K, MBBS,
Consultant and Head
Dr Norasyikin A Wahab
Tel: (60) 3 9170 2304
Fax: (60) 3 9173 7829
Adrenal insufficiency in acute coronary
Norasyikin A W, Norlela S, Rozita M, Masliza M, Shamsul A S, Nor Azmi K
Introduction: Acute coronary syndrome (ACS)
is an acute stressful condition which stimulates
the hypothalamus-pituitary-adrenal axis that
regulates neurovascular and hormonal responses.
Functional hypoadrenalism has been shown to
be associated with significant morbidity and
mortality in the critically-ill patient, but there is
to date, no known study done to determine its
prevalence in patients with ACS.
Methods: 37 patients who fulfilled the diagnostic
criteria of ACS were subjected to the low-dose
(1 μg) ACTH stimulation test (LDT), followed by
a standard-dose (250 μg) ACTH stimulation test
(SDT) two hours later.
Results: 14 (37.8 percent) patients had ST acute
myocardial infarction, eight (21.6 percent)
patients had non-ST elevation myocardial
infarction, and 15 (40.5 percent) patients had
unstable angina. Based on an increment of less
than 250 nmol/L post-SDT, no patient had adrenal
insufficiency. However, using a similar criteria
with the LDT, eight (21.6 percent) patients had
adrenal insufficiency. Four patients died during
the study and they had very high cortisol levels.
The diagnosis of adrenal insufficiency is not
associated with any significant morbidity and
mortality in our group of patients.
Conclusion: Utilising the LDT, adrenal insufficiency
is present in 21.6 percent of patients admitted
with ACS. However, this is not associated with
any significant morbidity and mortality.
Keywords: ACTH stimulation test, acute coronary
syndrome, adrenal insufficiency, cortisol
Singapore Med J 2009; 50(10): 962-966
The hypothalamus-pituitary-adrenal (HPA) axis response
is a self-defensive mechanism to maintain cardiovascular
and cellular homoeostasis or adapt to illness and stress.
Even a minor degree of adrenal insufficiency has been
shown to be associated with mortality in critically ill
patients. Circulating levels of glucocorticoids in patients
who are critically-ill are often more than three times
higher than those of healthy individuals.(1) This is not only
due to an increased cortisol production by the activation
of the HPA axis, but also due to a reduced hepatic
degradation and feedback mechanism, corticosteroid-
binding globulin activity and a loss of the normal
diurnal pattern of secretion.(2-5) Another cause of cortisol
elevation may be a shift in adrenal steroid synthesis, from
androgens and mineralocorticoids to cortisol.(6,7) High
levels of glucocorticoids during stress have negative
effects on many organ systems and may impair the
immune system, and affect protein, carbohydrate and fat
metabolism, as well as myocardial functions.(8) Patients
with critical illnesses would have an elevated serum
cortisol concentration, which is positively correlated with
the severity of the illness and negatively correlated with
survival. The incidence of reported adrenal insufficiency
in critical illness varies from 1% to 9%.(9)
Functional adrenal insufficiency or relative adrenal
insufficiency is used to describe the subnormal production
of corticosteroid during critical illness, in the absence of
structural defects in the HPA axis. This is due to high
levels of inflammatory cytokines, resulting in the direct
inhibition of adrenal cortisol synthesis and mediating
tissue-specific corticosteroid resistance.(10) Previous
studies have shown that either a basal cortisol level of
< 550, 690, 825 nmol/L or an incremental response of
< 250 nmol/L after cosyntropin stimulation predicted
a poor outcome, and identified patients who responded
favourably to glucocorticoid administration.(11-13) Only one
previous study reported on the prevalence of functional
hypoadrenalism in acute myocardial infarction (AMI),(14)
but none had occurred in the spectrum of acute coronary
syndrome (ACS). Hence, this study was done to evaluate
the prevalence of adrenal insufficiency in ACS patients
and to correlate its results with the inhospital mortality.
This study was performed in the coronary care unit of our
institution which is a tertiary centre, and it was approved
by the institutional Ethics Committee and Research. All
Singapore Med J 2009; 50(10) : 963
Table I. Baseline demographical data.
Table II. The prevalence of adrenal insufficiency by using
the patients gave written informed consent prior to the
onset of the study. 37 patients (25 male [67.6%] and 12
female [32.4%]) who fulfilled the criteria of ACS were
recruited. Patients were excluded if they were on herbal
medication or steroid treatment up to six months prior to
the admission; had a history of hypothalamus-pituitary,
adrenal or liver diseases (ALT ≥ twice normal upper
limit), systemic inflammatory response syndrome (SIRS;
temperature ≥ 38°C or ≤ 36°C, heart rate > 90 beats/min,
respiratory rate ≥ 20 breaths/min, PaCO2 ≤ 32 mmHg,
white blood cell count ≥ 12,000/μL or ≤ 4,000/μL or >
10% immature forms) and cardiogenic shock (Killip IV);
and were in an unconscious state or required inotropic
support. AMI (ST elevation MI [STEMI]) was diagnosed
according to the World Health Organization criteria.(15)
Unstable angina (UA) was diagnosed based on the
Braunwald classification and non-STEMI (NSTEMI) if
they did not meet the AMI and UA criteria.(16)
obtained. All the subjects had a low-dose (1 μg) ACTH
On admission, a random serum cortisol level was
stimulation test (LDT) done, followed by a one-hour
interval before having a standard dose (250 μg) ACTH
stimulation test (SDT) done. A 23G branula was inserted
and a baseline serum cortisol was obtained within 48 hours
of admission. The baseline serum cortisol was drawn
at zero min (0LD), 1 μg ACTH was given intravenously
and further serum cortisol levels were drawn at 30 min
(30LD) and 60 min (60LD). Two hours after 0LD, the SDT
was started with the second baseline serum cortisol level
drawn at 120 min (0SD), followed by the administration of
250 μg ACTH intravenously. Three further serum cortisol
levels were drawn at 150 min (30SD), 180 min (60SD) and
210 min (90SD).
A 120-minute interval between the two doses of
ACTH allowed cortisol levels to return to baseline levels
before the second dose and permitted the two tests to
be performed under as similar clinical circumstances as
possible. The order of the tests was maintained throughout
the study as the SDTs caused a more sustained rise in
cortisol, which would have necessitated increasing
the time interval between the two tests. For the LDT, a
vial of 250 μg ACTH (Alliance Pharmaceuticals Ltd,
Chippenham, UK) was diluted in normal saline solution to
a concentration of 0.5 μg/mL and was used immediately.
The serum cortisol was immediately separated and
stored at −20°C until it was assayed. The cortisol was
measured using commercially-available chemiluminescent
enzyme immunoassays (Immulite, Diagnostic Product
Corp, Los Angeles, CA, USA). The quality control
samples were 91–141 nmol/L for low level, 254–386
nmol/L for moderate level and 750–1,126 nmol/L for
high level. All of the eight cortisol samples from a single
patient were analysed together in the same analysis. The
cortisol samples from all 37 patients were assayed during
a single-batch analysis to avoid inter-assay variation.
The diagnosis of adrenal insufficiency was made using
the following criteria: baseline cortisol levels of < 550
nmol/L; cortisol response following LDT, increment of
cortisol < 250 nmol/L and peak cortisol < 700 nmol/L;(17)
and following SDT, increment of cortisol < 250 nmol/L(18)
and peak cortisol < 938 nmol/L.(12) Statistical analysis
was performed using the Wilcoxon rank test for repeated
measurements. The Mann-Whitney test was used to
determine the significance between two groups (survival
and non-survival) and for numerical variables. A p-value
of < 0.05 was deemed to be of statistical significance.
The baseline demographical data of all the subjects are
presented in Table I. 25 (67.6%) patients were male and
12 (32.4%) were female. The age range was 34–70 (mean
and standard deviation 53.32 ± 10.90) years. 17 (45.9%)
patients were Malay, 13 (35.1%) were Chinese, and the
remaining seven patients (18.9%) were Indian. 14 (37.8%)
had AMI, eight (21.6%) had NSTEMI, and 15 (40.5%)
had UA. In this study, all of our AMI patients received
thrombolysis as a mode of emergent reperfusion strategy.
The prevalence of adrenal insufficiency was 51.4% (19
patients), if the baseline cortisol of < 550 nmol/L was
taken. When the peak cortisol response to LD and SD
Singapore Med J 2009; 50(10) : 964
were taken, 18.9% (seven patients) and 10.8% (four
patients) prevalence were respectively noted. Upon
taking an increment of cortisol of < 250 nmol/L, 21.6%
(eight patients) and no patients were found to have adrenal
insufficiency with LD and SD, respectively (Table II).
There was no significant difference between the median
serum cortisol levels of the 37 patients at 0LD and 0SD (p =
0.065), indicating that the serum cortisol had returned to
the baseline levels before the SDT started (Fig. 1). There
was significant correlation between serum cortisol at
baseline and brain natriuretic peptide (BNP) (p = 0.046).
There were four mortalities within seven days of their
coronary events. Three of them had NSTEMI and one
had extensive MI. Two of these patients had underlying
diabetes mellitus. There was no significant increment
of serum cortisol at 30 min (p = 0.068) and 60 min (p
= 0.068) from the baseline in non-survivors using LDT
(Fig. 2). Non-survivors were found to have significantly
higher baseline cortisol (median cortisol 1,217.60 nmol/
L, interquartile range [IQR] 853.0–1,631.73), compared
to the survivor group (median cortisol 474.00 nmol/L,
IQR 361.50–592.50) (p = 0.003) (Figs. 2 & 3). The BNP
level was significantly higher in the non-survivor group
(median BNP 4.12 vs. 0.24 pg/ml; p = 0.045). There were
no significant correlations between mortality and the left
ventricular fraction, creatine kinase (CK), CK-myocardial
band (MB) fraction, troponin T or T peak levels in the MI
group. There was also no significant difference between
the Killip classification and mortality (p = 0.63).
The objective of this study was to determine the prevalence
of adrenal insufficiency by using LDT and SDT in patients
with ACS. The higher prevalence of males with ACS has
been reported in numerous other studies. There were
various criteria used to diagnose adrenal insufficiency in
previous studies. However, those studies were conducted
in severely-ill patients, and none has been done on ACS
patients so far. In this respect, the cut-off cortisol level
which indicates adrenal insufficiency in severely-ill
patients is still debatable. The proposed random levels
ranged from 276 nmol/L to 938 nmol/L, but several
studies have suggested that a threshold of 414 nmol/L best
identifies persons with clinical features of corticosteroid
insufficiency or who would benefit from corticosteroid
Barquist and Kirton considered a baseline cortisol
level of < 414 nmol/L in critically-ill patients to indicate
adrenal insufficiency, regardless of the stimulated cortisol
the? standard-dose?ACTH? stimulation? test? in? survivors? and?
Fig. 1 Graph?shows?the?comparison?of?low-dose?and?standard-?
Singapore Med J 2009; 50(10) : 965
levels. In this study, patients with a baseline cortisol level
of between 414 nmol/L and 550 nmol/L and a response
of < 690 nmol/L after 30 min of ACTH stimulation were
considered as having adrenal insufficiency.(19) However,
Oelkers(20) and Grinspoon and Biller(21) recommended
that a basal level of 500 nmol/L or a post-corticotropin
stimulation plasma cortisol level of ≥ 500 nmol/L was
adequate to rule out adrenal insufficiency. Bourne et al
used a peak cortisol of < 700 nmol/L and an increment of <
250 nmol/L, using a LDT as the cut-off points, to diagnose
adrenal insufficiency,(17) while Annane et al recommended
that a lack of increment of 250 nmol/L suggested an
inadequate response for the SDT.(18) To date, no similar
study had been done to determine the cortisol level in ACS
patients. As a result, we used the cortisol level in critically-
ill patients to make a diagnosis of adrenal insufficiency. In
comparing the various criteria used in our study, 51.4%
of our patients had adrenal insufficiency (baseline cortisol
level < 550 nmol/L). This figure was higher compared to
Rivers et al’s study, which found that 14% of critically-
ill patients had functional adrenal insufficiency, using the
However, using the peak cortisol response (< 700
nmol/L with a low-dose and < 938 nmol/L with a standard-
dose), we found that seven (18.9%) and four (10.8%)
patients had adrenal insufficiency, respectively. Based on
this result, we would have missed adrenal insufficiency
in four patients with the SDT as compared to the LDT.
These results were lower compared to a previous study
by Bourne et al which reported a 39.4% prevalence of
functional adrenal insufficiency.(17) Using the criteria
of cortisol increment < 250 nmol/L post-corticotropin
stimulation, eight (21.6%) of our patients had adrenal
insufficiency with the LDT, but none with the SDT. All of
these patients had a baseline cortisol level < 680 nmol/L.
These results were much lower compared to that reported
by Bourne et al, where the prevalence was 70.4%.(17) Their
study involved 71 septic shock patients, and the LDT was
used. In contrast, Rothwell et al found that 19% of their
septic shock patients had functional adrenal insufficiency,
using the SDT.(23) The reason for this discrepancy may
be related to the differences in the study population, the
background of medical problems and the small sample
sizes. In this study, low-dose ACTH was used to diagnose
adrenal insufficiency instead of the standard-dose ACTH,
because the latter would have resulted in a false-negative
All the patients had the LDT, followed by the SDT
with a two-hour interval between the two tests. The time
interval between the tests was to allow the cortisol levels
to return to the baseline prior to starting the second test
using a standard dose. There was no significant difference
between the serum cortisol levels at 0 min with a low-
dose and at 0 min of the standard dose (p = 0.10), which
suggests that the two tests were performed independent
of each other. Crowley et al observed that the maximum
cortisol level response in a LDT was at 15 minutes in the
majority of their normal subjects.(24) For the LDT in this
study, blood levels were taken at 0, 30 and 60 min. All
our patients attained the maximum response at 30 min.
The study by Beale et al showed there was no significant
difference between the serum cortisol at 0 and 60 min of a
LDT.(25) However, in our patients, the serum cortisol were
significantly higher at 30 and 60 min with the LDT (p =
0.005), indicating that the patients had responded to the
ACTH given and that their values did not return to baseline
even at 60 min. With the SDT, the increment of cortisol
persisted even after 90 min, indicating a more prolonged
response. This validated our approach of not beginning the
test with the standard dose, as the prolonged response will
definitely interfere with the results of the second test.
In this study, four (10.8%) patients with a high
baseline cortisol level (> 1,217.6 nmol/L) and a small
increment post-corticotropin stimulation, died within
five days of their acute coronary event. These deaths were
unexpected in three of them at the time of the cortisol
sampling and ACTH stimulation tests. Two of them had
recurrent NSTEMI, while the third succumbed to initial
NSTEMI and extensive MI. Whether these findings can
be used to further stratify ACS or AMI is still premature,
as more numbers are needed to arrive at a firm conclusion.
Currently, patients who had thrombolysis treatment are
stratified according to the Thrombolysis In Myocardial
Infarction trial (TIMI) guidelines. This will influence
the need for a more aggressive intervention in the form
of percutaneous coronary intervention or coronary artery
bypass grafting. It is interesting to speculate whether the
addition of random cortisol or cortisol response to the LDT
could complement the treatment guidelines of TIMI. The
patients who demonstrated adrenal insufficiency based on
the LDT and SDT did not suffer any excess in morbidity
and mortality despite the well-known associations in other
critical illness studies. In conclusion, a high cortisol level
above 1,217 nmol/L in patients with ACS on admission
tends to be associated with an inhospital mortality. The
prevalence of adrenal insufficiency in our cohort of
ACS varied, depending on the criteria used – using a
baseline cortisol level below 550 nmol/L gave the highest
prevalence of 51.4%, while an increment of 250 nmol/L
on a LDT gave a value of 21.6%. Peak cortisol during the
LDT and SDT gave a prevalence of 18.9% and 10.8%,
Singapore Med J 2009; 50(10) : 966
We would like to thank the Dean of the Medical Faculty,
National University of Malaysia, for allowing us to
publish this paper. This project was funded by the Faculty
of Medicine, National University of Malaysia (project
code no: FF 011/ 2005).
1. Vermes I, Beishuizen A, Hampsink RM, Haanen C. Dissociation
of plasma adrenocorticotropin and cortisol levels in critically ill
patients: possible role of endothelin and atrial natriuretic hormone.
J Clin Endocrinol Metab 1995; 80:1238-42.
2. Wang P, Ba ZF, Jarrar D, et al. Mechanism of adrenal insufficiency
following trauma and severe hemorrhage: role of hepatic 11beta-
hydroxysteroid dehydrogenase. Arch Surg 1999; 134: 394-401.
3. Perrot D, Bonneton A, Dechaud H, Motin J, Pugeat M.
Hyepercortisolism in septic shock is not suppressible by
dexamethasone infusion. Crit Care Med 1993; 21:396-401.
4. Schein RM, Sprung CL, Marcial E, Napolitano L, Chernow B.
Plasma cortisol levels in patients with septic shock. Crit Care Med
5. Lamberts SWJ, Bruining HA, de Jong FH. Corticosteroid therapy
in severe illness. N Engl J Med 1997; 337:1285-92.
6. Drucker D, Shandling M. Variable adrenocortical function in acute
medical illness. Crit Care Med 1985; 13:477-9.
7. Parker LN, Levin ER, Lifrak ET. Evidence for adrenocortical
adaptation to severe illness. J Clin Endocrinol Metab 1985;
8. Sam S, Corbridge TC, Mokhlesi B, Comellas AP, Molitch ME.
Cortisol levels and mortality in severe sepsis. Clin Endocrinol
9. Knapp PE, Arum SM, Melby JC. Relative adrenal insufficiency
in critical illness: a review of the evidence. Curr Opin Endocrinol
Diabetes 2004; 11:147-52.
10. Cooper MS, Stewart PM. Corticosteroid insufficiency in acutely
ill patients. N Engl J Med 2003; 348:727-34.
11. Rivers EP, Gaspari M, Saad GA, et al. Adrenal insufficiency in
high-risk surgical ICU patients. Chest 2001; 119:889-96.
12. Marik PE, Zaloga GP. Adrenal insufficiency during septic shock.
Crit Care Med 2003; 31:141-5.
13. Annane D, Sébille V, Charpentier C, et al. Effect of treatment with
low doses of hydrocortisone and fludrocortisone on mortality in
patients with septic shock. JAMA 2002; 288: 862-71.
14. Chang SS, Liaw SJ, Bullard MJ, et al. Adrenal insufficiency in
critically ill emergency department patients: A Taiwan preliminary
study. Acad Emergency Med 2001; 8:761-4.
15. Clinical practice guideline on acute myocardial infarction 2001.
Putrajaya: Ministry of Health Malaysia, 2001.
16. Clinical practice guideline on UA/NSTEMI 2002. Putrajaya:
Ministry of Health Malaysia, 2002.
17. Bourne RS, Webber SJ, Hutchinson SP. Adrenal axis testing and
corticosteroid replacement therapy in septic shock patients--local
and national perspectives. Anaesthesia 2003; 58:591-6.
18. Annane D, Sébille V, Troché G, et al. A 3-level prognostic
classification in septic shock based on cortisol levels and cortisol
response to corticotropin. JAMA 2000; 283:1038-45.
19. Barquist E, Kirton O. Adrenal insufficiency in the surgical
intensive care unit patient. J Trauma 1997; 42:27-31.
20. Oelkers W. Adrenal insufficiency. N Engl J Med 1996;335:1206-12.
21. Grinspoon SK, Biller BM. Clinical review 62: Laboratory
assessment of adrenal insufficiency. J Clin Endocrinol Metab
22. Rivers EP, Blake HC, Dereczyk B, et al. Adrenal dysfunction in
hemodynamically unstable patients in the emergency department.
Acad Emerg Med 1999; 6:626-30.
23. Rothwell PM, Udwadia ZF, Lawler PG. Cortisol response
to corticotropin and survival in septic shock. Lancet 1991;
24. Crowley S, Hindmarsh PC, Holownia P, Honour JW, Brook CG.
The use of low doses of ACTH in the investigation of adrenal
function in man. J Endocrinol 1991; 130:475-9.
25. Beale E, Zhu J, Belzberg H. Changes in serum cortisol with age in
critically ill patients. Gerontology 2002; 48:84-92.