ENDOGENOUS OPIOID TONE IN PATIENTS WITH RHEUMATOID ARTHRITIS

Transcription

1 British Journal of Rheumatology 1996;35:436-^40 ENDOGENOUS OPIOID TONE IN PATIENTS WITH RHEUMATOID ARTHRITIS D. KASSEMOS, E. H. S. CHOY, A. B. GROSSMAN,* I. C. CHIKANZAt and G. S. PANAYI Rheumatology Unit, Guy's Hospital, United Medical and Dental Schools of Guy's and St Thomas Hospitals, *Department of Endocrinology, St Bartholomew's Hospital and ^Rheumatology Unit, Royal London Hospital, London SUMMARY We have previously shown that there is deficient hypothalamic-pituitary-adrenal (HPA) responsiveness in rheumatoid arthritis (RA) patients. The basis for this deficient response is not known. The purpose of the project was to investigate whether the defective HPA response in RA patients is the result of increased endogenous opioid tone secondary to chronic pain which can suppress corticotrophin-releasing hormone (CRH) production. We conducted a double-blind placebo-controlled cross-over trial to study the effect of the opiate antagonist, naloxone, on psychometric function together with plasma adrenocorticotrophic hormone (ACTH), cortisol and prolactin. Seven RA patients with active and established disease and eight healthy controls were studied. Each received either a bolus i.v. infusion of 20 mg naloxone or normal saline. After at least 72 h, they received naloxone if they had previously received normal saline or vice versa. The pain score was statistically significantly higher at baseline in the RA group compared with controls (5.7 ± 3.25 vs 0.35 ± 0.21, P < 0.001). No difference was found in the other psychometric assessments throughout the study. Patients receiving normal saline did not show any significant change in cortisol or ACTH. Cortisol and ACTH showed a sharp and significant rise after naloxone treatment in both RA and normal subjects (P < and P < 0.01), but no difference was observed between the two groups. The mean prolactin level showed no significant change in both groups after any treatment. We conclude that endogenous opioid tone does not appear to be a major contributor to the HPA defect in RA. However, the number of patients studied was small and this result will require confirmation from larger trials. KEY WORDS: Rheumatoid arthritis, Endogenous opioid, Opiate, Hypothalamus-pituitary-adrenal axis, Cortisol, Naloxone, ACTH, Prolactin. WE have previously shown that there is a deficient hypothalamic-pituitary adrenal (HPA) response in patients with rheumatoid arthritis (RA) [1]. The basis for this deficient response is not known, but appears not to be the consequence of chronic inflammation as it is not seen in patients with osteomyelitis or osteoarthritis (OA). Chronic painful conditions, such as RA, may result in increased endogenous opioid release. Endogenous opioids exert a negative tonic inhibition on corticotrophin-releasing hormone (CRH) secretion from the hypothalamus and cortisol secretion will thereby be inhibited [2]. If this is the case, infusion of the opiate antagonist naloxone should lead to increased cortisol secretion from the adrenal via increased CRH and adrenocorticotrophic hormone (ACTH) release from the hypothalamus and the pituitary, respectively [3]. The purpose of this project was to investigate the outcome of ACTH and cortisol secretion in RA and controls after injection of naloxone. In addition, since volunteers have experienced cognitive impairment and dysphoria during naloxone infusion [4, 5], psychometric testing, using the Profile of Mood Scale was administered before and after the naloxone infusion in all subjects. METHODS Study subjects We conducted a double-blind placebo-controlled cross-over trial to study the effect of naloxone on ACTH, cortisol, prolactin and psychometric function. Patients with definite RA, as defined by the 1987 ACR criteria [6], were recruited from the rheumatology out-patients clinic. They had active disease, as defined by the presence of at least three out of the four following criteria: (1) > 6 joints painful to touch or motion; (2) ^3 swollen joints; (3) >45min early morning stiffness; (4) > 28 mm erythrocyte sedimentation rate. Control subjects consisted of patients with soft tissue rheumatism or healthy volunteers. We excluded any subject who had thyroid or ovarian disorders, a history of psychiatric illness, oral corticosteroid therapy, cytotoxic therapy (except methotrexate or cyclophosphamide), androgen or oestrogen therapy, major tranquillizers, dopaminergic drugs, disorders of the HPA axis or pregnancy. The study protocol was approved by the Guy's and Lewisham Hospitals Research Ethics Committee. Signed informed consent was obtained from all subjects prior to participation. Submitted 5 September 1995; revised version accepted 7 December Correspondence to: G. S. Panayi, Division of Medicine, Rheumatology Unit, 4th Floor, Hunt's House, Guy's Hospital, St Thomas Street, London SE1 9RT. Study protocol Subjects fasted overnight and stopped all long-acting non-steroidal anti-inflammatory drugs (NSAIDs) and painkillers which contain opiates, such as Co-proxamol or Co-dydramol, 24 h before the test was performed British Society for Rheumatology

2 KASSIMOS ET AL.: ENDOGENOUS OPIOID TONE IN RA 437 TABLE I Demographic details of rheumatoid arthritis patients and normal controls No. of patients Male: female ratio Age, yr (mean) Disease duration, yr (mean) ESR (mm/h, mean) D-Penicillamine (no.) Sulphasalazine (no.) NSAIDs (no.) RA patients 7 4:3 55 ± Controls 8 4:4 52 ± 14-8 _ They attended the hospital at h and each received either a bolus i.v. injection of 20 mg naloxone or normal saline. After a minimum duration of 72 h, they received naloxone if they had previously received normal saline or vice versa. Pain and psychometric variables, which include relaxation, depression, tension, anxiety, anger, vigour, confusion, concentration, fatigue, happiness and sadness, were measured using standard visual analogue scales [4,5]. These were administered at times 15, +30 and 60 min postinfusion to assess pain and mood. Laboratory assessments Blood samples were taken 15 min prior to naloxone injection and afterwards at 15 min intervals for a total 800 -i of 120 min. Samples were collected into lithiumheparin tubes, immediately chilled to 4 C and centrifuged at 1500 r.p.m. at 4 C for 15 min. Plasma was aliquoted, quickly frozen in dry ice, and stored at 20 C until assayed for plasma ACTH, cortisol and prolactin. Serum cortisol was measured according to the method of Cunnah et al. [7] and expressed in nmol/1. The interassay coefficient of variation (CV) across the working range of the assay was 8% at 143 nmol/1 and 5.5% at 1042 nmol/1 with an interassay CV <5% at all values. Plasma ACTH was measured by immunoradiometric assay (Allegro HS-ACTH, Nichols Institute, San Joan Capistrano, CA, USA) and expressed in ng/1. The intra-assay and interassay CV values across the working range of the assay were <5%. Serum prolactin was measured by an immunoradiometric assay and expressed in mu/1. The intra-assay CV across the working range of the assay was <5%; the interassay CV was 6% at 1156mU/l, rising to 10.6% at 216 mu/1. Statistical analysis Results were expressed as the mean ± s.d. The data were statistically analysed in several ways: analysis of variance, paired Student's r-test for paired parametric variables and the Mann Whitney C/-test for non-paired non-parametric variables. P values of <0.05 were considered significant. "o B o O to JO Q_ Time (minutes) FKJ. 1. Plasma cortisol level (nmol/1) at different time points (the arrow indicates the time of naloxone infusion) in the four following groups:, RA patients given naloxone infusion; A, normal controls given naloxone infusion; T, RA patients given placebo infusion;, normal controls given placebo infusion.

3 438 BRITISH JOURNAL OF RHEUMATOLOGY VOL. 35 NO I 200 to E jo Q_ Time (minutes) FIG. 2. Plasma prolactin level (nu/1) at different time points (the arrow indicates the time of naloxone infusion) in the four following groups:, RA patients given naloxone infusion; A, normal controls given naloxone infusion; T, RA patients given placebo infusion;, normal controls given placebo infusion. RESULTS Seven patients with active established RA and eight controls were recruited (Table I). There was no change in any of the cardiovascular parameters (blood pressure, pulse rate) after the naloxone infusion, this is in agreement with previous data [2, 8-12]. Pain measurements The pain score was statistically significantly higher at baseline in the RA group (5.7 ± 3.25 cm) compared with controls (0.35 ± 0.21 cm, P < 0.001). No difference was found in other mood assessments between the RA patients and controls both before and after naloxone (data not shown). Hormonal measurements Basal. Serum prolactin was higher in RA patients (261 ± 125 mu/1) compared with controls (175 ± 36 mu/1), although this did not reach statistical significance. Neither cortisol (RA 502 ± 181 mu/1 vs control 487±125mU/l) nor ACTH (medians: RA 5.5 ng/1 and controls 8.0 ng/1) showed any significant difference before naloxone injection. After saline infusion. Patients receiving normal saline did not show any significant change in circulating cortisol (Fig. 1), prolactin (Fig. 2) or ACTH (data not shown). After naloxone infusion. Plasma cortisol showed a sharp and significant rise after naloxone treatment in both RA and normal subjects. In RA patients, cortisol increased statistically significantly from 462 ± 180 to 577 ± 172 nmol/1 (i> < 0.01) 30 min after naloxone infusion. The peak cortisol level in RA patients was 630 ± 152 nmol/1. In normal controls, cortisol increased from 439 ± 129 to 608 ± 172 nmol/1 (P < 0.01) 30 min after naloxone infusion. The peak cortisol level in normal controls was 620 ± 134 nmol/1. No statistically significant difference was observed between the two groups (Fig. 1). The mean prolactin level showed no significant change in either group after treatment (Fig. 2). Interestingly, RA patients showed higher prolactin levels than normal controls when receiving either naloxone or saline infusions. ACTH levels in patients and normal controls were in general very low. In both groups of patients, naloxone induced a marked rise in ACTH (> 20 ng/1) in two individuals in each group. However, in most individuals there were small and transient increases in ACTH (data not shown). There were no statistically significant differences between the two groups.

4 KASSIMOS ET AL:. ENDOGENOUS OPIOID TONE IN RA 439 DISCUSSION The most important finding of this study is that 20 mg of naloxone produced a significant increase in cortisol and ACTH in patients with RA and healthy controls, but no difference was observed between the two groups. We also showed that naloxone had no effect on the prolactin level in both groups. There was no effect on prolactin levels in either group, supporting previous data [13, 14]. There is little evidence that endogenous opioids are important regulators of prolactin release in man [13-15]. Interestingly, we found that the circulating prolactin level was higher in RA patients than controls, although this did not reach statistical significance. Our results confirmed previous data of Grossman et al. [2] showing that high-dose naloxone (16 mg) in normal subjects produced a peak level in plasma cortisol and ACTH [3, 16]. However, even at high doses, we cannot be absolutely sure that antagonism to opiate receptor occupancy is the only mode of action of naloxone [15]. A previous study showing that there was an additive effect of naloxone with CRH suggested that naloxone works through an ACTH-releasing factor independent of CRH, most probably the other recognized ACTH-releasing factor, vasopressin [17], or possibly both CRH and vasopressin [18]. The above data are also compatible with one previous report on the regulation of prolactin [19] and suggest that naloxone cannot be used as an endogenous surrogate for CRH as a test of the pituitary-adrenal axis [19]. We have shown that despite a significantly higher pain score at the baseline and throughout this study in the RA group compared with controls, we were unable to show any statistically significant difference in plasma cortisol and ACTH levels between RA patients and normal controls. Since RA patients had a higher pain score, they would be expected to have a higher endogenous opioid tone. This suggests that a high endogenous opioid tone does not account for the HPA axis defects in RA patients. However, it can be argued that RA patients with active disease should have a higher level of cortisol, their failure to do so indicating a blunted HPA response. Indeed, this could be explained by our previous finding that RA patients undergoing surgical stress showed a suboptimal cortisol response when compared to patients with osteoarthritis and osteomyelitis [18]. Nonetheless, the results of this study suggest that endogenous opioid tone is unlikely to be a major contributor to the HPA defect in RA, although this may be due to naloxone acting, at least in part, in releasing factors other than CRH. Hence, further studies are needed which investigate the vasopressin pathway in patients with RA. ACKNOWLEDGEMENTS The project was funded by the Arthritis and Rheumatism Council of Great Britain (project grant no. G0081). Dr D. Kassimos was supported by the Ministry of Defence of Greece. We would like to thank Drs T. Gibson, G. H. Kingsley and Professor R. Grahame for allowing us to study their patients. REFERENCES 1. Chikanza IC, Petrou P, Kingsley GH, Chrousos G, Panayi G. Defective hypothalamic response to immune and inflammatory stimuli in patients with rheumatoid arthritis. Arthritis Rheum 1992;35: Grossman A, Gaillard RC, McCarthney P, Rees LH, Besser GM. Opiate modulation of the pituitary-adrenal axis: effect of stress and arcadian rhythm. Clin Endocrinol 1982;17: Grossman A, Moult PJA, Cunnah D, Besser M. Different opioid mechanisms are involved in the modulation of ACTH and gonadotrophin release in man. Neuroendocrinology 1986;42: Martin del Campo AF, McMurray RG, Besser GM, Grossman A. Effect of the 12 hour infusion of Naloxone on mood and cognition in normal male volunteers. Biol Psychiatry 1992;32: McMurray RG, Newbould E, Bouloux P, Besser GM, Grossman A. High dose naloxone modifies cardiovascular and neuroendocrine function in ambulant subjects. Psychoneuroendocrinology 1991;16: Arnett FC, Edworthy SM, Bloch DA et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988^1: Cunnah D, Jessop DS, Besser GM, Rees LH. Measurement of circulating corticotrophin-releasing factor in man. J Endocrinol 1987;113: Ngai SH, Berkowitz B, Yang JC, Hempstread J, Spector S. Phannacokinetics of naloxone in rats and in man. Anesthesiology 1976;44: Bouloux PMG, Newbould E, Causon R et al. Different effect of high dose naloxone on the plasma adrenaline response to the coldpressor test. Clin Sd 1989;76: Delitala G, Trainer PJ, Oliva O, Fanciulli G, Grossman AB. Opioid peptide and adrenoceptor pathways in the regulation of the pituitary-adrenal axis in man.-a Endocrinol 1994;141: Cohen MR, Cohen RM, Pickar D, Weingart H, Murphy DL. High dose naloxone infusion in normals. Dose dependent behaviour, hormonal, and physiological responses. Arch Gen Psychiatry 1983;4(h Cohen NM, Cohen RM, Pickar D, Weingartner H, Murphy DL, Bunney VYT. Behavioural effect after high dose naloxone administration to normal volunteers. Lancet 1981;2: Blankstein J, Reyes F, Winter J, Faiman C. Effects of naloxone on prolactin and cortisol in normal women. Proc Soc Exp Biol Med 1980;164: Grossman A, Stubbs WA, Gaillard RC, Delitalia G, Rees LH, Bresser GM. Studies of the opiate control of prolactin, GH and TSH. Clin Endocrinol 1981;14: Sawynock J, Pinsky C, Labella FS. New review on the specificity of naloxone as an opiate antagonist. Life Sci 1979;25: Volanka V, Cho D, Mallya A, Bauman J. Naloxone increases ACTH and cortisol levels in man. N EnglJ Med 1979;3O0:

5 440 BRITISH JOURNAL OF RHEUMATOLOGY VOL. 35 NO Patcher VK, Mastorakos G, Brady LS, Rcdwine A, humans: evidence that naloxone stimulates endogenous Wilder RL, Chrousos G. Increased arginine vasopressin corticotrophin-releasing hormone. Clin Endocrinol secretion may participate in the enhanced susceptibility of 1995;61: Lewis rats to inflammatory disease. Neuroendocrinology 19. Chikanza IC, Petrou P, Chrousos G, Kingsley G, 1993;58: Panayi GS. Excessive and dysregulated secretion of 18. Hockings GI, Grice JE, Walters MM, Crosbie GV, prolactin in rheumatoid arthritis: Immunopathogenetic Torpy DJ, Jackson RV. A synergistic adrenocorti- and therapeutic implications. Br J Rheumatol cotrophin release to naloxone and vasopressin in normal 1993^2:445-8.