PAIN MEDICINE

Volume 3 • Number 2 • 2002

RESEARCH REPORT

Normalization of Serum Cortisol Concentration With Opioid

Treatment of Severe Chronic Pain

Downloaded from https://academic.oup.com/painmedicine/article-abstract/3/2/132/1940250 by guest on 22 December 2019

Forest Tennant, MD, DrPH, and Laura Hermann, BSN, RN, FNP
Veract Intractable Pain Centers, West Covina, California

ABSTRACT

Serum cortisol concentrations may be altered in severe, chronic pain due to excess stimulation of the
hypothalamic-pituitary-adrenal axis. Among 40 consecutive patients with severe, chronic pain 26
(65.0%) demonstrated abnormal serum cortisol concentration. After 90 days of treatment, only 7
(17.5%; p0.01) continued to show abnormal serum cortisol concentration indicating that serum
cortisol and other serologic abnormalities may serve as biologic markers of severe, chronic pain.

Introduction were referred while being treated with four to eight

One biologic marker, which may occur with severe,
chronic pain, is abnormal serum cortisol concentra-
tion [1-4]. Severe pain is a potent stressor, which
may induce excess stimulation of the hypothalamic-
pituitary-adrenal axis (HPA) causing elevated se-
rum cortisol concentration; and if adrenal reserve
or another component of the HPA diminishes over
time due to excess stimulation, serum cortisol con-
centration may be low [1,3,5]. This preliminary
study was done to determine if pain treatment with
opioid drugs may normalize high and low serum
cortisol concentrations found in uncontrolled se-
vere, chronic pain.
Methods
Forty consecutive, adult patients with a variety of
nonmalignant, painful conditions were referred for
intensive long-term medical management (Table
1). There were 21 males and 19 females whose ages
ranged from 25 to 60 years. Pain had been present
for a minimum of three years. Patients described
their pain as severe, uncontrolled, constant, and in-
curable, and that it interfered with sleep and some-
times forced them to be bed- or home-bound. All
had failed multiple, nonopioid treatments, and all
Reprint requests to: Forest Tennant, MD, Dr PH, Veract In-
tractable Pain Centers, 338 South Glendora Avenue, West
Covina, CA 91790. Tel: (626) 919-7476; Fax: (626) 919-7497.
daily dosages of a hydrocodone, codeine, or pro-
poxyphene preparation. The estimated morphine
equivalency was 50-100 mg/d. No patient was be-
ing treated with systemic corticosteroids.
In the first week of treatment, serum cortisol con-
centration was determined at 8:00 a.m. with the pa-
tient in a fasting state. After 90 days of treatment, an
8:00 a.m. fasting serum cortisol concentration was
again determined. Normal serum cortisol concentra-
tion was considered to range from 5.0-20.0 g/dL.
Treatment was initiated with one of the follow-
ing long-acting opioids: 1) Methadone; 2) Trans-
dermal fentanyl; or 3) Sustained-release oxycodone
or morphine. Breakthrough pain was treated with
one of the following short-acting opioids: 1) Oral
transmucosal fentanyl citrate; 2) Oxycodone; 3)
Hydromorphone; 4) Morphine; or 5) Hydroco-
done. Dosages of both long- and short-acting opio-
ids were titrated upward during a monthly clinic
visit to achieve improved pain control without pro-
ducing sedation or impairment. The only attempt
to evaluate pain control was the patient’s written
report that they had less pain, improved sleep, and
fewer bed- or house-bound days. Morphine equiva-
lency was calculated to range from 400-1000 mg/d.
Results
Initial serum cortisol concentrations ranged from
0.6–32.5 g/dL (mean: 13.9  10.3 SD). Fourteen

© American Academy of Pain Medicine 1526-2375/02/$15.00/132 132–134

Serum Cortisol Normalization With Opioid Treatment 133

Table 1 Causes of severe chronic pain in 40 patients (t (N  11)  2.62, P0.05), which was statistically
tested for cortisol abnormalities significant. The overall mean cortisol concentra-
Number (%) tion of the 40 patients did not change appreciably
of patients (13.9 10.3 SD g/dL versus 14.1  8.1 SD g/
Degenerative spinal disease postsurgery 18 (45.0%) dL), and no patient with normal cortisol concentra-
Degenerative spinal disease nonoperative 10 (25.0%) tion when entering treatment developed a low or
Osteoporosis 3 (7.5%)
Fibromyalgia 2 (5.0%)
high serum cortisol concentration during the 90-
Migraine-vascular headache 2 (5.0%) day study period.
Neuropathies 2 (5.0%)

Downloaded from https://academic.oup.com/painmedicine/article-abstract/3/2/132/1940250 by guest on 22 December 2019

Reflex sympathetic dystrophy 1 (2.5%)
Systemic lupus erythematosus 1 (2.5%)
Abdominal adhesions 1 (2.5%)
Total 40 (100%)
(35.0%) subjects demonstrated a low concentra-
tion—below 5 g/dL, and 12 (30.0%) subjects had
elevated serum cortisol concentrations—above 20.0
g/dL (Table 2). The mean serum cortisol concen-
tration of patients with low concentrations was 3.2 
1.41 SD g/dL, and the mean serum cortisol con-
centration of patients with high concentrations was
27.5  10.8 SD g/dL. At the follow-up testing, a
significantly greater number of patients had a nor-
mal early morning cortisol level—33 (82.5%) com-
pared with 14 (30.0%) prior to treatment (X2 (N 
1)  18.7, P0.01). The number of patients with a
low serum cortisol level was significantly lower—2
(5.0%) compared with 14 (35.0%) pretreatment (X2
(N  1)  11.33, P0.01), and the mean concen-
tration of cortisol in this group was higher—9.4 
3.7 compared with 3.2  1.41 (N  13)  5.90,
P0.01). The number of patients with high serum
cortisol after treatment was significantly lower—5
(12.5%) compared with 12 (30.0%) pretreatment
(X2 (N  1)  3.72, P0.10). Although this lower
proportion of patients with high serum cortisol
concentration was not statistically significant at the
P0.05 level, only 2 of 12 (16.7%) patients did not
have lower serum cortisol concentrations posttreat-
ment, and the mean serum concentration in this
group was lower—18.7  10.6 SD g/dL com-
pared with 27.5  10.8 SD g/dL pretreatment
Discussion
At the time of initial determination of early morning
serum cortisol concentration, patients gave a history
of poor pain control with low dosages of short-
acting opioids. Only 14 of 40 patients (35.0%) dem-
onstrated a normal early morning serum cortisol
concentration. Patients with high serum cortisol
concentrations confirm previous reports that se-
vere, constant pain is a potent stressor, which may
overstimulate the HPA [1-3]. Patients with low se-
rum cortisol concentrations may have depleted ad-
renal cortisol reserve due to constant, severe pain,
or some component of the HPA, other than the ad-
renal gland, may be responsible for this abnormal-
ity [3,6,8,9]. Regardless of the cause, deficient cor-
tisol serum concentration is a major concern, since
adrenal steroids may influence analgesic effective-
ness [10,11]. Fortunately, in our study, 19 of 26
(73.1%) patients with a low or high serum cortisol
concentration prior to treatment normalized fol-
lowing pain control with a combination of long-
and short-acting opioid therapy.
While this preliminary study suggests that ade-
quate opioid treatment can normalize serum cortisol
concentrations, it must be cautiously interpreted,
due to methodologic limitations. It is possible that
the treatment process of regular clinical attendance
and psychosocial support was instrumental in reliev-
ing stress, fear, and depression, which are known to
cause serum cortisol abnormalities [5,7,9,12]. Also,
no control group was utilized, and some patients
studied here had painful conditions, some of which
could cause cortisol abnormalities even when not ac-

Table 2 Normalization of serum cortisol concentrations before and after 90 days of pain treatment (N  40)
Before treatment After treatment Statistical significance
Number (%) with abnormal cortisol concentration 26 (65%) 7 (17.5%) P0.01
Number (%) with high cortisol concentration 12 (30.0%) 5 (12.5%) P0.10
Number (%) with low cortisol concentration 14 (35.0%) 2 (5.0%) P0.01
Mean of high cortisol cases (g/dL) 27.5  10.8 SD 18.7  10.6 SD P0.05
Mean of low cortisol cases (g/dL) 3.2  1.4 SD 9.4  3.65 SD P0.01
Overall mean cortisol cortisol cases (g/dL) 13.9  10.3 SD 14.1  8.1 SD not significant
134
companied by severe pain [9,12]. There was no at-
tempt to quantify the degree of pain relief other than
clinical impression of the authors based on patient
reports. Also, there was considerable heterogeneity
of medical diagnosis, and there was no evaluation
made of psychiatric or mood states. Another factor
that may have influenced the initial serum cortisol
concentrations was the use of low dosages of short-
acting opioids, since opioids are known to some-
times suppress pituitary-adrenal function [8,9,13,14].
Despite these limitations, this preliminary study sug-
gests that high or low early morning serum cortisol
concentration may be a useful, biologic marker for
severe, chronic pain, and pain treatment with opio-
ids is associated with normalization of abnormal se-
rum cortisol concentrations. Further studies should
be done to determine the role of cortisol and other
adrenal hormones and other objective serologic indi-
cators as biologic markers of severe, chronic pain [4].
References
1 Chapman RC, Gavrin J. Suffering: The contribu-
tions of persistent pain. Lancet 1999;353:2233–7.
2 Shenkin HA. Effect of pain on diurnal pattern of
plasma corticoid levels. Neurology 1964;14:1112–5.
3 Tennant F. Intractable pain is a severe stress state
associated with hypercortisolemia and reduced ad-
renal reserve. Drug Alcohol Depend 2000;60(suppl
1):220–1.
4 Glynn CJ, Lloyd JW. Biochemical changes associ-
ated with intractable pain. Br Med J 1978;1:280–1.
Tennant and Hermann
5 Murphy K. Chronic pain syndrome: What is it: And
how can we help? Am Pain Manage 1994;4:129–31.
6 Akil H, Shiomi H, Matthews J. Induction of the in-
termediate pituitary by stress: Synthesis and release
of a nonopioid form of beta-endorphin. Science
1985;227:424–8.
7 Liebeskind JC. Pain can kill. Pain 1991;44:3–4.
8 Buckingham JC. Secretion of corticotrophin and its
hypothalamic releasing factor in response to mor-
phine and opioid peptides. Neuroendocrinology 1982;
Downloaded from https://academic.oup.com/painmedicine/article-abstract/3/2/132/1940250 by guest on 22 December 2019
35:111–6.
9 Grossman A, Gaillard RC, McCartney P, Rees LH,
Besser GM. Opiate modulation of the pituitary-
adrenal axis: Effects of stress and circadian rhythm.
Clin Endocrinol (Oxf) 1982;17:279–86.
10 Holaday JW, Law PY, Loh HH, Li CH. Adrenal
steroids indirectly modulate morphine and beta-
endorphin effects. J Pharmacol Exp Ther 1979;208:
176–83.
11 Long JB, Holaday JW. Blood-brain barrier: Endog-
enous modulation by adrenal-cortical function. Sci-
ence 1985;227:1580–3.
12 Amsterdam JD, Winokur A, Abelman E, Lucki L,
Rickels K. Cosyntropin (ACTH alpha 1-24) stimu-
lation test in depressed patients and healthy sub-
jects. Am J Psych 1983;140:907–9.
13 Tennant F, Shannon JA, Nork JG, Sagherian A,
Berman M. Abnormal adrenal gland metabolism in
opioid addicts: Implications for clinical treatment. J
Psychoactive Drugs 1991;23:135–49.
14 McDonald RK, Evans ET, Weise VK, Patrick RW.
Effect of morphine and nalorphine on plasma hy-
drocortisone levels in man. J Pharmacol Exp Ther
1959;125:241–7.