correlation analysis (SPSS/PC for Windows 95 version toms, jaw claudication, tongue claudication) were included. These patients were referred to the participat-.
Rheumatology 2000;39:624–631
The adrenal steroid status in relation to inflammatory cytokines (interleukin-6 and tumour necrosis factor) in polymyalgia rheumatica R. H. Straub, T. Glu¨ck, M. Cutolo1, J. Georgi2, K. Helmke3, J. Scho¨lmerich, P. Vaith4 and B. Lang Department of Internal Medicine, University Medical Center Regensburg, D-93042 Regensburg, Germany, 1Division of Rheumatology, Department of Internal Medicine, University of Genova, I-16136 Genova, Italy, 2Ostseeklinik, D-24349 Damp, 3Hospital Mu¨nchen-Bogenhausen, D-81925 Mu¨nchen and 4Department of Internal Medicine, University Medical Center Freiburg, D-79106 Freiburg, Germany Abstract Objectives. To determine the correlation between inflammatory cytokines and adrenal hormones in patients with polymyalgia rheumatica (PMR) and to compare the ratio of serum cortisol and androstenedione (ASD) or dehydroepiandrosterone sulphate (DHEAS ) in normal subjects with PMR patients. Methods. In 102 patients with PMR (32 beginning and 70 chronic disease) and 31 agematched and sex-matched healthy subjects, ASD, cortisol, DHEAS, interleukin-6 (IL-6), and tumour necrosis factor (TNF ) were measured by immunometric assays. Results. Serum levels of IL-6 were elevated in patients with PMR as compared with normal subjects (10.0 ± 1.6 vs 2.1 ± 0.1 pg/ml, P = 0.01), which was not found for TNF. In PMR patients, serum levels of IL-6 were positively correlated with serum levels of ASD (P < 0.001), cortisol (P < 0.001), and DHEAS (P = 0.038) irrespective of corticosteroid treatment. Serum levels of cortisol in relation to IL-6 were significantly lower in patients with chronic disease and long-standing corticosteroid administration as compared with patients with recent onset of the disease and without corticosteroid therapy (P < 0.01). Conclusions. In PMR, as expected, there was an increase in IL-6 serum levels that was associated with elevated serum levels of ASD, DHEAS, and cortisol which was more marked in patients with recent-onset disease and without corticosteroids. However, serum levels of cortisol in patients with and without corticosteroids were lower than expected by considering the inflammatory status (increased IL-6). This may indicate a change in the hypothalamic– pituitary–adrenal (HPA) axis responsiveness to inflammatory stimuli such as IL-6 during chronic disease. Furthermore, there seems to be a shift of biosynthesis to cortisol in relation to DHEAS or ASD in chronic disease. K : Polymyalgia rheumatica, Cortisol, Dehydroepiandrosterone sulphate, Androstenedione, Interleukin-6, Tumour necrosis factor.
Polymyalgia rheumatica (PMR) is a chronic inflammatory disease of unknown aetiology in elderly people. Although the striking feature of PMR is the development of the disease in patients older than 50 yr of age, ageassociated pathogenetic factors are not yet known. The age-specific incidence rate increases from 2.6 per 100 000 in the age group 50–59 yr to 44.7 per 100 000 in the age
group 80 yr and older [1], which suggests the possibility of age-associated factors in aetiology. The natural decline of several hormones such as dehydroepiandrosterone (DHEA) [2, 3] or androstenedione (ASD) [3] during ageing may represent one such factor. A pioneering study in this direction demonstrated that ageassociated decline of DHEA was linked to increased secretion of interleukin-6 (IL-6) in ageing mice [4] and we have confirmed this finding in humans [3]. Hence, endocrinosenescence may be an important pathogenetic element in patients with PMR. As in patients with
Submitted 10 June 1999; revised version accepted 20 December 1999. Correspondence to: R. H. Straub, Department of Internal Medicine I, University Medical Center, D-93042 Regensburg, Germany.
624
© 2000 British Society for Rheumatology
IL-6 and adrenal hormones in PMR
systemic lupus erythematosus [5, 6 ], rheumatoid arthritis [7], systemic sclerosis [8], inflammatory bowel disease [9], and Alzheimer’s disease or vascular dementia [10], low DHEA levels have been described in patients with PMR [11]. However, the interrelation between the natural decline of these adrenal hormones and the pathogenesis of chronic inflammatory diseases such as PMR, particularly in aged people, has yet to be examined. Another important characteristic of PMR is the rapid improvement after initial administration of corticosteroids in a dose range of about 20–30 mg of prednisolone (in giant cell arteritis ≥ 60 mg) [12] which is equivalent to about 80–120 mg of endogenous, adrenal gland-derived cortisol [13]. This amount of cortisol is three times the secretion rate of a healthy adrenal gland per day [13]. In cases of i.v. administration of cytokines such as tumour necrosis factor ( TNF ) [14] or IL-6 [15, 16 ] or during septic shock [17], the healthy human adrenal gland is capable of secreting up to 300 mg of endogenous cortisol daily. Hence, the question arises whether or not the adrenal glands of elderly patients with PMR are unable to produce the amount of cortisol necessary to control the inflammation. In view of these aspects, the aim of the present study was to determine the correlation between IL-6/TNF and immunosuppressive effector hormones of the hypothalamic–pituitary–adrenal (HPA) axis such as ASD, cortisol, and dehydroepiandrosterone sulphate (DHEAS ) in patients with PMR. Patients with recent onset of the disease prior to corticosteroid therapy were compared with patients with chronic disease who had undergone corticosteroid therapy. Furthermore, we compared serum levels of cortisol in relation to ASD or DHEAS between PMR patients with recent-onset and those with chronic disease.
Subjects and methods From a total number of 147 patients with PMR, 102 patients with complete data sets and without clinical signs of giant cell arteritis (artery tenderness, swollen scalp arteries, decreased artery pulses, fixed visual symptoms, jaw claudication, tongue claudication) were included. These patients were referred to the participating teaching hospitals specialized in rheumatology. All patients were clinically evaluated by a rheumatologist and PMR was diagnosed according to the criteria of Bird et al. [18]. Thirty-two patients had recent-onset PMR [disease duration, mean ± standard error of the mean (...), median: 0.64 ± 0.05 yr, 0.5 yr] and had never been treated with corticosteroids at the time blood was drawn as compared with 70 patients with chronic disease who had undergone corticosteroid therapy (disease duration, mean ± ..., median: 3.46 ± 0.40 yr, 2.0 yr, P < 0.001 as compared with patients without corticosteroids; prednisolone was administered before breakfast). Since patients under long-standing corticosteroid treatment had either low (50% of all patients) or moderate to high disease activity (20% of all patients
625
including patients with unstable disease or relapses), the daily prednisolone dose was variable according to clinical requirements. The two PMR patient groups did not differ in age or gender. Clinical and demographic data are shown in Table 1. For comparison, 31 healthy agematched and body mass index-matched control subjects were recruited and their state of health was validated by means of a 33-item questionnaire ( Table 1). The questionnaire addressed known diseases in the past and at present, current symptoms of diseases, current medication, alcohol intake, smoking habits, family history, and surgical history. Patients and healthy controls were not on hormone replacement therapy. Blood was drawn by venipuncture between 10.00 and 12.00 a.m. and serum was immediately stored at −80°C in adequate aliquots. Radioimmunometric assays for the quantitative determination of serum ASD (DPC Biermann, Bad Nauheim, Germany; detection limit: 0.14 nmol/l ) and serum cortisol (Coulter Immunotech, Marseille, France; detection limit: 10 nmol/l; cross-reactivity with prednisolone
