| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Haukeland University Hospital (K.L., O.L.M., E.S.H.), N-5021 Bergen, Norway; University Hospital Uppsala (G.G.-M., O.K.), SE-75185 Uppsala, Sweden; Rogaland Central Hospital (S.U.), N-4068 Stavanger, Norway; Haugesund Hospital (B.G.N.), N-5504 Haugesund, Norway; St. Olav’s Hospital, University Hospital of Trondheim (K.J.F.), N-7003 Trondheim, Norway; and University Hospital of Northern Norway (T.S.T.), N-9038 Tromsø, Norway
Address all correspondence and requests for reprints to: Dr. Kristian Løvås, Division of Endocrinology, Institute of Medicine, Haukeland University Hospital, N-5021 Bergen, Norway. E-mail: kristian.lovas{at}haukeland.no.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A lack of androgens may affect patients with Addison’s disease in a number of ways. We have recently published an observational study of subjective health status in 79 Norwegian patients with Addison’s disease (7), in which we found reduced general health status, reduced vitality perception, and increased fatigue compared with the general Norwegian population. Mental and social health was hardly affected. The women reported reduced physical functioning and more physical limitations than normal. No strong evidence exists that depressed mood is common in adrenal failure, although this has been associated with low DHEA levels in elderly women (8). Cognitive function in patients with adrenal failure was measured by Hunt et al. (9), who found good performance compared with normal subjects. The exact role of androgens in sexuality in women is still not established (10), and impaired sexuality has not been described as a feature of adrenal failure.
Both depletion of sex steroids and oversubstitution of glucocorticoids pose the risk of osteopenia. One study showed low bone mineral density in men, but not in women, with Addison’s disease (11), whereas two studies have shown decreased bone mineral density in postmenopausal women, but not in men (12, 13). Others have found normal bone mineral density in Addison’s disease, and variable correlation among disease duration, cumulative glucocorticoid treatment, and bone mineral density (14, 15).
Several clinical trials have been conducted to study DHEA replacement in age-advanced men and women and in diseases associated with low DHEA levels, but the clinical benefit is still controversial (2, 16). Two clinical trials of oral replacement therapy with 50 mg DHEA in adrenal failure have been published, and they indicate a favorable effect of DHEA on subjective health status (9, 17). Before these reports and based on the results of the dose-finding study by Gebre-Medhin et al. (18), we conducted a double-blind, parallel group clinical trial of treatment with 25 mg DHEA or placebo for 9 months. Treatment effects were assessed by measurement of subjective health status and sexuality, androgen levels, lipids, and markers of bone metabolism.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The study was designed as a multicenter, randomized parallel group clinical trial, as illustrated in Fig. 1
. Included were women, aged 18–70 yr, with confirmed primary or secondary adrenal failure and with subnormal DHEAS levels. Both hypogonadal and eugonadal women were included. Other replacement therapies should have remained unchanged for 6 wk before inclusion and were continued throughout the study period. Exclusion criteria were severe liver disease, malignant disease or other severe system disease, and actual or planned pregnancy.
|
The objective was to evaluate treatment effects of DHEA replacement, with subjective health and sexuality scales as primary outcome measures. Secondary outcome measures were androgen levels, markers of bone metabolism, and lipids. Clinical controls, registration of adverse events, and blood and urine sampling were performed at inclusion, during treatment (1, 4.5, and 9 months), and at a follow-up visit 3 months after discontinuation of study medications (12 months). Subjective health status [Short Form-36 (SF-36) and fatigue questionnaires] and sexuality (McCoy’s questionnaire) were recorded at inclusion, during treatment (4.5, and 9 months), and at follow-up 3 months after discontinuation of study medications (12 months). Side-effects were recorded at all controls except at inclusion. The patients returned surplus tablets for assessment of compliance. Blood samples were obtained in the morning, approximately 12 h after ingestion of DHEA.
Informed consent was obtained, and the protocol was approved by the regional committee of ethics in medical research and the Norwegian Medicines Agency.
Measurements
The SF-36 comprises 36 items formulated as questions or statements scored as numbers from 1–6 (19). Eight parameters with a range from 0–100 are calculated: physical functioning, role-physical or role limitations due to physical problems, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. The first 3 parameters primarily measure physical health, the last 3 parameters similarly measure mental health, whereas the general health and vitality scales are sensitive to both physical and mental health outcomes (19). Increasing scores represent increasing quality of life. Reference values for the Norwegian population were available (20).
The fatigue questionnaire records mental and physical fatigue as responses to 11 statements (21). Respondents are asked to express the extent of symptoms experienced the last month (less than usual, same as usual, more than usual, and much more than usual), with usual representing when they were last feeling well. To quantify the dimension of fatigue, each response is assigned a number from 0–3; the highest numbers represent the most severe fatigue. Total fatigue is the simple algebraic sum of all items, whereas physical fatigue and mental fatigue are the sums of the 7 and 4 items, respectively. Normative data for the Norwegian population were available (22).
A Norwegian version of the McCoy’s Sex Scale Questionnaire (23) was employed for assessment of sexuality. This questionnaire measures sexual experience during the previous 30 d by responses to seven items. Each item is assigned a score at a visual analog scale ranging from 1–7. Higher scores indicate higher degree of satisfaction. Two items were grouped into the variable desire, two items into the variable problems, and four items comprised the variable satisfaction. Normative data were not available.
Side-effects were recorded in a standardized form with four items (hirsutism, acne, sweat odor, and scalp itching), which were scored from none (0) to severe (4). For each patient the mean score across the four items were calculated (side-effect score).
At each visit safety blood parameters for liver and renal functions, glucose, lipids (total cholesterol, high density lipoprotein cholesterol, triglycerides), electrolytes, and blood counts were measured. Serum concentrations of hormones were measured by established immunoassay kits. In the case of androstenedione, ACTH, DHEAS, and aldosterone the kits were delivered from Diagnostic Products (Los Angeles, CA). In the case of cortisol, estradiol, testosterone, gonadotropic hormones (LH and FSH), TSH, free T4, and SHBG, the kits were delivered from AutoDelfia (Wallac, Inc., Turku, Finland). IGF-I was measured with a double antibody disequilibrium assay from DiaSorin, Inc. (Stillwater, MN). IGF-binding protein-3 was measured with a specific competitive protein binding RIA from Nichols Institute Diagnostics (San Juan Capistrano, CA). Osteocalcin was measured with competitive luminescence immunoassay from BRAHMS Diagnostica (Berlin, Germany), and deoxypyridinoline in urine was measured by solid phase chemiluminescent enzyme-labeled immunoassay from Diagnostic Products. The intra- and interassay coefficients of variation were less than 10% for the analyses of DHEAS, androstenedione, and testosterone within the female concentration range.
Randomization and blinding
Block randomization (n = 4) was performed, with stratification in two age groups (18–49 and 50–70 yr). Randomization and blinding were undertaken by Haukeland University Hospital pharmacy, and the randomization codes were kept in sealed envelopes until the end of the study. Laboratory data revealing the hormone levels were not available to the clinical investigators during the study.
Statistics
Time-dependent effects were tested by two-way ANOVA. When no significant differences during treatment (1, 4.5, and 9 months) were found, the individual average values during treatment were calculated. This was done to secure data from patients with early withdrawal, according to the principle of intention to treat. The treatment effects of the two groups were compared by t test. 
= 0,01 was selected as the level of significance, because multiple parameters were tested. No reports of treatment effects on subjective health status or sexuality were available for power calculations and estimation of sample size. Therefore, differences in treatment effects of 1 SD or more were assumed to be clinically relevant. The number of patients was limited due to the low prevalence of the diseases, but 40 patients would yield a statistical power of 0.67 to detect such effects at the 0.01 significance level. The data are presented as the mean ± SD throughout the text and tables.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The study design with participant flow is illustrated in Fig. 1
. The inclusion period was from October 1998 until February 2000, and the last patient visit was in March 2001. The randomization was successful with respect to age distribution, proportion of patients with gonadal failure, and disease etiology, as shown in Table 1. Compliance, as assessed by surplus tablet count, was above 85% for all patients, except one patient in the DHEA group who returned 30% of the capsules.
|
DHEA replacement yielded a wide variation in treatment effects in all SF-36 scores (Table 2), fatigue scores (Table 3), and sexuality scores (Table 4). All of the scales in the DHEA group improved nonsignificantly from baseline and also tended to be improved at 12 months (follow-up) compared with baseline. The subjective health and sexuality scores did not change significantly from 4.5 to 9 months in any of the groups. The effects of DHEA replacement were not different from the effects of placebo. Only the role-emotional scores of the SF-36 showed a trend toward benefit, but these scores did not fall to baseline at 12 months (follow-up).
|
|
|
Seventeen of 19 patients (89%) receiving DHEA experienced some kind of side-effects (side-effect score, 0.65 ± 0.84), in particular increased sweat odor (0.96 ± 0.73) and scalp itching (0.77 ± 0.84). However, 14 of 20 patients (70%) in the placebo group also experienced side-effects, but these were significantly milder compared with the DHEA group (side-effect score, 0.18 ± 0.23; P < 0.001). No serious adverse events occurred.
Laboratory findings
No significant change with time between 1, 4.5, and 9 months were found in any of the laboratory parameters tested. Therefore, the individual average levels during treatment (1, 4.5, and 9 months) were calculated for comparison of treatment effects, and these are the values that are presented here. In the DHEA group the androgen levels rose to the normal reference range for almost all patients during treatment (Fig. 2). The DHEA group had levels of DHEAS (6.1 ± 2.0 µmol/liter), androstenedione (5.7 ± 2.4 nmol/liter), and testosterone (0.93 ± 0.83 nmol/liter) that were significantly increased from baseline (P < 0.001 for all). This increase was also significantly different (P < 0.001) from the change from baseline in the placebo group (DHEAS, <0.8 µmol/liter at all visits; androstenedione, -0.2 ± 0.6 nmol/liter; testosterone, -0.02 ± 0.13 nmol/liter). In the DHEA group, SHBG levels were significantly lower than at baseline (76.5 ± 8.5 to 62.8 ± 1.8 nmol/liter; P = 0.0097), but this was not significantly different (P = 0.28) from the change in the placebo group (68.4 ± 31.9 to 62.5 ± 24.1 nmol/liter).
|
Total cholesterol (5.1 ± 1.0 to 5.2 ± 0.9 mmol/liter), high density lipoprotein cholesterol (1.6 ± 0.5 to 1.6 ± 0.5 mmol/liter), IGF-I (18.3 ± 6.6 to 18.9 ± 6.3 mmol/liter), and hematological, renal, and hepatic parameters did not significantly change during DHEA replacement compared with baseline values.
In both study groups, serum cortisol, ACTH, aldosterone, plasma renin activity, and electrolytes showed adequate replacement of glucocorticoids and mineralocorticoids. All patients were biochemically euthyroid. Estradiol and gonadotropic hormones were in the physiological range.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Arlt et al. (17) and Hunt et al. (9) both employed a two-period cross-over design to increase the statistical power of their trials. They were able to demonstrate statistically significant treatment effects, which, however, may be clinically irrelevant. From a clinical point of view, the cross-over design is not very robust in this type of study, with a high risk of violation of the blinding. If any carry-over effects from one trial phase to another can be anticipated, it is recommended to avoid the cross-over design (24).
Arlt et al. (17) highlighted improved overall well-being, depression, and anxiety in women with adrenal failure receiving 50 mg DHEA daily for 4 months. On these scales their patients scored worse than the normal population at baseline. Their data show that for these scales the scores were worse before the period of DHEA replacement than before the placebo period. If the differences between the scores after DHEA and placebo treatments are clinically relevant, then the differences in the baseline scores also require an explanation. For many of the scales the improvement during DHEA replacement was similar to the deterioration during placebo treatment. How should significant negative placebo effects be interpreted in this particular disease?
Hunt et al. (9) reported improved mood and fatigue in a trial with 50 mg DHEA daily for 3 months in women and men with adrenal failure, and this effect was most evident in the evening. However, according to their data, the improvement from baseline during DHEA replacement was similar in the morning and evening. Conversely, placebo was efficacious in the morning, but had a negative mean effect in the evening. Another serious methodological problem is that they did not provide baseline scores before the second period, which makes it is impossible to rule out carry-over effects.
Our patients received 25 mg DHEA in the evening and achieved morning androgen levels similar to those reported by the investigators who studied 50 mg DHEA given as a single morning dose (9, 17). We note that Arlt et al. (17) obtained higher levels of DHEAS (8.6 ± 1.1 µmol/liter, taken from figure) and testosterone (0.74 ± 0.09 nmol/liter) than Hunt et al. (9) (DHEAS, 4.62 ± 0.88 µmol/liter; testosterone, 0.46 ± 0.07 nmol/liter). These findings illustrate substantial differences in the preparations, diurnal variation in absorption, or measurement variations, which could be important for the evaluation of clinical effects. Obviously, dosage should be guided by the measurement of androgens, but the target levels of DHEA replacement in women with adrenal failure are not known, as the normal interindividual variations are large. Whether additional benefit can be gained by aiming for premenopausal androgen levels in postmenopausal women is another controversial and unsettled issue (2, 16). Moreover, the metabolism of DHEA may be reflected not by serum levels of active androgens, but, rather, by levels of their metabolites (25).
Whereas the patients in the study by Arlt et al. (26) had a slight, but significant, increase in osteocalcin levels, Hunt et al. (9) and we did not find such effects. We found minor reduction of deoxypyridinoline in the DHEA group, but this was not significantly different from the effect in the placebo group. Cross-sectional studies in Addison’s disease have not convincingly demonstrated decreased bone mineral density (11, 12, 13, 14, 15), and therefore, the clinical significance of DHEA depletion must be marginal. It is, however, conceivable that DHEA replacement could be beneficial for subgroups of patients with particular risk for osteoporosis.
All studies to date have been small and short-term, and randomized, parallel group studies with larger number of patients are needed to document beneficial effects of DHEA. Although we did not find significant effects of DHEA in our study, it is not possible to exclude a true benefit in the at large population of patients with adrenal failure. This is illustrated by the confidence intervals for the differences between DHEA and placebo treatments. It is also possible that effects could be demonstrated for subgroups of patients with severe androgen deficiency, such as patients with hypopituitarism or hypogonadism. If neuroprotection is an important biological role of DHEA(S) (3, 4), long-term studies will certainly be required to document clinically relevant effects. The imprecision of subjective health measurements is high, and small changes should be interpreted with caution. If a precise definition of the clinical syndrome of DHEA(S) deficiency were provided, more sensitive and disease-specific instruments to measure the effects of DHEA replacement could be constructed, which will probably be necessary for the demonstration of treatment effects.
The results of previous clinical trials discussed above have prompted the recommendations for DHEA replacement therapy in patients with adrenal failure (16, 27). Based on the results of our own study with 25 mg DHEA and a critical review of the previous trials with replacement of 50 mg DHEA in adrenal failure (9, 17), we conclude that the evidence of benefit for subjective health status is weak. The cross-over study design is highly questionable, as the high frequency of side-effects will unblind the randomization for both patient and doctor, and thus risk increasing the recorded differences.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: DHEA, Dehydroepiandrosterone; DHEAS, dehydroepiandrosterone sulfate; SF-36, Short Form-36.
Received May 17, 2002.
Accepted December 4, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  K.A. Walters, C.M. Allan, and D.J. Handelsman Androgen Actions and the Ovary Biol Reprod, March 1, 2008; 78(3): 380 - 389. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. M. Gurnell, P. J. Hunt, S. E. Curran, C. L. Conway, E. M. Pullenayegum, F. A. Huppert, J. E. Compston, J. Herbert, and V. K. K. Chatterjee Long-Term DHEA Replacement in Primary Adrenal Insufficiency: A Randomized, Controlled Trial J. Clin. Endocrinol. Metab., February 1, 2008; 93(2): 400 - 409. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Hahner, M. Loeffler, M. Fassnacht, D. Weismann, A.-C. Koschker, M. Quinkler, O. Decker, W. Arlt, and B. Allolio Impaired Subjective Health Status in 256 Patients with Adrenal Insufficiency on Standard Therapy Based on Cross-Sectional Analysis J. Clin. Endocrinol. Metab., October 1, 2007; 92(10): 3912 - 3922. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Lovas and E. S Husebye Continuous subcutaneous hydrocortisone infusion in Addison's disease Eur. J. Endocrinol., July 1, 2007; 157(1): 109 - 112. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Panjari and S. R. Davis DHEA therapy for women: effect on sexual function and wellbeing Hum. Reprod. Update, May 1, 2007; 13(3): 239 - 248. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Brooke, L. A. Kalingag, F. Miraki-Moud, C. Camacho-Hubner, K. T. Maher, D. M. Walker, J. P. Hinson, and J. P. Monson Dehydroepiandrosterone Improves Psychological Well-Being in Male and Female Hypopituitary Patients on Maintenance Growth Hormone Replacement J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 3773 - 3779. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E. Wierman, R. Basson, S. R. Davis, S. Khosla, K. K. Miller, W. Rosner, and N. Santoro Androgen Therapy in Women: An Endocrine Society Clinical Practice Guideline J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 3697 - 3710. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. K. Miller, B. M. K. Biller, C. Beauregard, J. G. Lipman, J. Jones, D. Schoenfeld, J. C. Sherman, B. Swearingen, J. Loeffler, and A. Klibanski Effects of Testosterone Replacement in Androgen-Deficient Women with Hypopituitarism: A Randomized, Double-Blind, Placebo-Controlled Study J. Clin. Endocrinol. Metab., May 1, 2006; 91(5): 1683 - 1690. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Basson Sexual Desire and Arousal Disorders in Women N. Engl. J. Med., April 6, 2006; 354(14): 1497 - 1506. [Full Text] [PDF]
|
|
|
|
|
|
W. Arlt Androgen therapy in women Eur. J. Endocrinol., January 1, 2006; 154(1): 1 - 11. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. W. van Thiel, J. A. Romijn, A. M. Pereira, N. R. Biermasz, F. Roelfsema, A. van Hemert, B. Ballieux, and J. W. A. Smit Effects of Dehydroepiandrostenedione, Superimposed on Growth Hormone Substitution, on Quality of Life and Insulin-Like Growth Factor I in Patients with Secondary Adrenal Insufficiency: A Randomized, Placebo-Controlled, Cross-Over Trial J. Clin. Endocrinol. Metab., June 1, 2005; 90(6): 3295 - 3303. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Handelsman Andro and the Prosteroids: Bolting the Stable Door J. Clin. Endocrinol. Metab., February 1, 2005; 90(2): 1249 - 1251. [Full Text] [PDF]
|
|
|
|
 |
|
 P. J. Schmidt, R. C. Daly, M. Bloch, M. J. Smith, M. A. Danaceau, L. Simpson St. Clair, J. H. Murphy, N. Haq, and D. R. Rubinow Dehydroepiandrosterone Monotherapy in Midlife-Onset Major and Minor Depression Arch Gen Psychiatry, February 1, 2005; 62(2): 154 - 162. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Christiansen, C. H. Gravholt, S. Fisker, N. Moller, M. Andersen, B. Svenstrup, P. Bennett, P. Ivarsen, J. S. Christiansen, and J. O. L. Jorgensen Very short term dehydroepiandrosterone treatment in female adrenal failure: impact on carbohydrate, lipid and protein metabolism Eur. J. Endocrinol., January 1, 2005; 152(1): 77 - 85. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Sexual Dysfunction Obstet. Gynecol., October 1, 2004; 104(4_suppl): 85S - 91S. [Full Text] [PDF]
|
|
|
|
|
|
W. Arlt and B. Allolio DHEA Replacement in Adrenal Insufficiency J. Clin. Endocrinol. Metab., August 1, 2003; 88(8): 4001 - 4001. [Full Text] [PDF]
|
|
|
|
|
|
K. Lovas, O. Kampe, and E. S. Husebye Authors' Response: DHEA Replacement in Adrenal Insufficiency J. Clin. Endocrinol. Metab., August 1, 2003; 88(8): 4001 - 4002. [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
