RHEUMATOID ARTHRITIS

Central Sensitization in Patients with Rheumatoid Arthritis: A Systematic Literature Review

Mira Meeus, PhD,*, Stijn Vervisch, PT,* Luc S. De Clerck, PhD, Greta Moorkens, PhD, Guy Hans, PhD, and Jo Nijs, PhD*,,

Objective: The goal of the present study is to systematically review the scientic literature addressing central sensitization and central nociceptive processing in patients with rheumatoid arthritis (RA). Methods: To identify relevant articles, we searched PubMed and Web of Science. The search strategy was a combination of terms of the following groups: Rheumatoid arthritis, inammatory joint pain, or arthritis; AND (central) sensitization, (central) hypersensitivity, central hyperexcitability, pain modulation, pain processing, neural inhibition, or pain physiopathology; AND pain, nociception, hyperalgesia, pain threshold, or algometry. Articles fullling the inclusion criteria were screened for methodologic quality with specic checklists to evaluate different study designs (2 independent raters). Results: Twenty-four full-text articles were included, of which the majority were case-control studies, followed by nonsystematic reviews, cross-sectional studies, and case reports. Methodologic quality was very heterogeneous. Preliminary evidence for generalized hyperalgesia in RA is available. In addition, the mechanism behind impaired central nociceptive processing remains rather obscure. The role of cytokines and neuropeptides especially remains to be elucidated. Windup appears to develop more easily in RA, but evidence in support of impaired nociceptive inhibition and cognitive emotional sensitization (sensitization due to cognitive bias) is scarce. Conclusions: The symmetrical manifestation of the disease, the poor relation between disease activity and symptoms, and the generalized hyperalgesia at both articular and nonarticular sites for different kinds of stimuli are indicative of the presence of central sensitization in RA patients. Further research is required to provide rm evidence in support of various aspects of central sensitization in humans with RA. 2012 Elsevier Inc. All rights reserved. Semin Arthritis Rheum 41:556 567 Keywords: pain inhibition, chronic pain, central sensitization, pain processing, rheumatoid arthritis, systematic review

*Department of Human Physiology, Faculty of Physical Education and Physiotherapy, Vrije Universiteit Brussel (VUB), Brussels, Belgium. Division of Musculoskeletal Physiotherapy, Higher Institute of Physiotherapy, Department of Health Care Sciences, Artesis University College Antwerp (AHA), Antwerp, Belgium. Department of Immunology, Allergy and Rheumatology, University of Antwerp (UA), Antwerp, Belgium. Department of Internal Medicine, University Hospital Antwerp (UZA), Antwerp, Belgium. Multidisciplinary Pain Center (PCT), University Hospital Antwerp (UZA), Antwerp, Belgium. Department of Rehabilitation and Physiotherapy, University Hospital Brussels, Brussels, Belgium. The authors have no conicts of interests to disclose. Address reprint requests to: Mira Meeus, PhD, Department of Human Physiology, Faculty of Physical Education and Physiotherapy, Vrije Universiteit Brussel, Building L-Mfys, Pleinlaan 2, 1050 Brussels, Belgium. E-mail: Mira.Meeus@vub.ac.be.

heumatoid arthritis (RA) is characterized by a symmetrical distribution of joint inammation in conjunction with ongoing pain, as well as increased pain during movement and light pressure to the articular margin of the joint (for reviews, see (1)). Pain associated with RA can occur spontaneously or can be evoked by gentle stimulation of the joint when it is moved within its normal working range (2). Furthermore, pain and tenderness are not only present in joints directly affected but also in surrounding, apparently normal, tissues. Referred pain syndromes may also occur. The magnitude of symptoms may not necessarily correlate with the severity of the underlying disease and symptoms may persist even when disease exacerbations have apparently settled (1). These ndings suggest that there could be a facilitation or amplication of the transmission of nociceptive infor-

556

0049-0172/12/$-see front matter 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.semarthrit.2011.08.001

M. Meeus et al.

557

mation, in that hyperalgesia, allodynia, and referred pain arise, all typical manifestation of sensitization. Under clinical conditions such as inammation, the nervous system does not simply react to noxious stimuli, it undergoes changes that modify the neuronal processing of nociceptive stimuli (functional neuroplasticity) (3). The heightened state of reactivity to innocuous/noxious stimuli at the site of inammation is referred to as peripheral sensitization (2). Peripheral Sensitization Any joint pathology can produce severe pain resulting from sensitization of primary afferent nociceptive neurons (4). Peripheral sensitization is mediated by local components of the local inammatory reaction released by the microenvironment of the inamed and injured tissue, such as bradykinin, prostaglandin, and probably neuropeptides and cytokines (2,5). The peripheral terminals of A and C bers, which are activated only by noxious stimuli under noninamed conditions, express many receptors and ion channels that recognize the various inammatory mediators in the vicinity. Thus, the local inammation results in the release of multiple factors that activate local nerve terminals involved in pain perception (2). In a state of inammation-evoked neuroplasticity, sensory stimuli such as light pressure or movements in the working range of the joint may evoke much larger responses in the nervous system than under normal conditions (3,6). Inactive, silent, or sleeping nociceptors alter their function during inammation. A bers, which normally respond only to tactile and not to noxious stimuli, can mediate a long-lasting tactile allodynic state during chronic inammation (2,7). Consequently, the message noxious stimulus is signaled to the spinal cord by innocuous stimuli, which are usually not painful (for details, see (3)). In summary, peripheral sensitization implies increased spontaneous activity, decreased activation threshold of the nociceptive bers, increased responsiveness to suprathreshold stimuli, and increased local release of neuropeptides on stimulation and recruitment of silent nociceptive bers (8). Hypothesis: Central Hyperexcitability in RA Peripheral sensitization is widely held to be responsible for pain and tenderness over inamed or damaged joints, but it does not account for other symptoms observed in RA (1) and for enhanced responses to afferent inputs from noninamed regions adjacent to or even remote from the inamed site (3). Furthermore, the autonomic dysfunction and bilateral involvement in RA are strongly suggestive for impaired pain processing in the central nervous system (CNS). Continuously increased nociceptive impulse activity, as in RA, could lead to peripheral and subsequently long-lasting central sensitization, as well as to an increased activity of the sympathetic nervous system

(3,8), giving rise to chronic, persistent pain, stress-related features, endocrine and behavioral symptoms, fatigue, and mood disorders (7). Chronic widespread pain in some other chronic idiopathic conditions, such as bromyalgia, appears to result from abnormal central pain processing, rather than from damage and injury to anatomical body structures (9). Hyperalgesia, allodynia, and referred pain are indeed symptoms frequently observed in idiopathic chronic pain syndromes like bromyalgia or chronic fatigue syndrome (10,11). The absence of a clear source of nociception and the fact that pain in these syndromes lacks a distinct spatial localization is indicative for central sensitization (12). Besides the presence of enhanced windup, widespread hyperalgesia, and allodynia, different studies also reported dysfunctional pain inhibition and enhanced pain facilitation in these patients (for reviews, see (12,13)). Objective In patients with RA, however, neurophysiologic mechanisms underlying pain remain unclear. Experimental animal models of inammatory arthritis suggest that changes of neuronal sensitivity at both peripheral and central levels may be important (3,14). Evidence indicating that central plasticity of the nociceptive system is contributing to pain in humans with RA seems, however, less obvious (1,15). The goal of the present study is to systematically review the scientic literature addressing central nociceptive processing in patients with RA. It is hypothesized that the pain and various other symptoms are due to sensitization of the CNS. Does the scientic literature provide evidence indicative of central sensitization in RA? We do not review possible therapeutic targets, but instead we focus on the clinical manifestations of those with RA. Clinical/ neurophysiologic manifestations and assessment methods, such as algometry, generalized hyperalgesia, windup, efcacy of pain inhibition, etc compared with normal controls or other chronic pain populations, can be informative regarding the involvement of central sensitization in patients with RA. Literature Search Methods Given the expectancy of chiey observational studies (given the research question), the review was conducted according to the recommendations of the Meta-Analysis of Observational Studies in Epidemiology group (16). Qualication of Searchers/Raters Literature was searched and screened by the rst author (MM), who had obtained a PhD with a dissertation regarding central sensitization (in chronic fatigue syndrome), including a systematic review (17). Methodologic quality was assessed by 2 researchers (MM and SV). The second rater achieved the degree of

558
Table 1 Entry Terms for Systematic Literature Search RA (MeSH) AND AND AND AND AND AND AND AND AND AND AND AND AND AND AND AND sensitization pain nociception hyperalgesia algometry central sensitization hypersensitivity central sensitivity central hyperexcitability pain processing pain modulation neural inhibition pain threshold pain physiopathology (MeSH/Majr) sensitization pain

Central sensitization in patients with RA

AND pain

AND central

Inammatory joint pain Arthritis NOT osteoarthritis

RA, rheumatoid arthritis.

AND central sensitization

Master of Science and was trained in assessing methodologic quality. In the case of disagreement, a third opinion was provided by the last author (JN), who obtained a degree of PhD and is experienced in chronic pain research. Search Strategy To identify relevant articles concerning central pain processing in patients with RA, PubMed (http://www. ncbi.nlm.nih.gov/entrez) and Web of Science (http:// isiwebofknowledge.com) were searched in October 2009. The search strategy was based on a combination of the search terms, derived from PICO. It consisted of the following keywords and MeSH terms: Rheumatoid arthritis, inammatory joint pain or arthritis AND (central) sensitization, (central) hypersensitivity, central hyperexcitability, pain modulation, pain processing, neural inhibition, pain physiopathology AND pain, nociception, hyperalgesia, pain threshold; algometry was used to identify articles concerning central pain processing in RA. The construct of the search strategy is presented in Table 1. In addition, reference lists of relevant published articles were searched to make the search as complete as possible. Study Selection To be included in the review, an article had to meet the following criteria: (1) subjects of the study had to be adults (humans) diagnosed with RA to obtain a population as heterogeneous as possible; (2) the author(s) studied the concept of central pain processing in these individuals; (3) the studies were presented in English, Dutch, French, or German; (4) only studies in the last 20 years were allowed; and (5) articles were full-text reports, and not abstracts, letters, or editorials. If any of the 5 inclusion criteria were not fullled, then the article was excluded from the literature review.

Study Process We screened the titles and abstracts of all citations identied in the search. The full-text article was retrieved if the citation was considered potentially relevant. Each full-text article was evaluated whether it fullled the inclusion criteria. Articles were categorized by the reviewer following study design (meta-analysis/review/(randomized) clinical trial/longitudinal study/case-control study/cross-sectional/case report). Assessment of Study Quality Different checklists were used to assess the methodologic quality of different study designs. Quality assessment of case-control studies or cohort studies was performed using the Newcastle-Ottawa Scale (NOS), a scale specic for the respective studies, which uses a star rating system to judge quality based on 3 aspects of the study: selection of groups, comparability, and ascertainment of the outcomes of interest. The validity of this tool has been previously established (18). The NOS is 1 of the more comprehensive instruments for assessing the quality of nonrandomized studies: the 8-item instrument consists of 3 subscales, namely, selection of subjects (4-item), comparability of subjects (1-item), and assessment of outcome/exposure (3-item). A maximum of 9 starts can be awarded. The NOS has been recommended by the Cochrane Non-Randomized Studies Methods Working Group and it is partly validated and primarily used to appraise cohort studies and case-control studies (18). Cross-sectional studies were evaluated with the relevant criteria of NOS checklist for cohort studies. This means criteria concerning comparability, control group, or follow-up were thrown out. Reviews were evaluated with the specic checklist provided by the Dutch Cochrane Centre (19). Randomized controlled trials are evaluated with the Jadad scale (20), as recommend by the former QUORUM

M. Meeus et al.
Table 2 Methodological Quality Criteria Case-Control Studies (18) S1: Adequate case denition* S2: Representativeness of cases* Cohort/Cross-Sectional Studies (18) S1: Representativeness exposed cohort* S2: Selection of nonexposed cohort* S3: Ascertainment of exposure* S4: Outcome of interest not present at start* Ca: Controlled for age* Cb: Controlled for . . .* O1: Assessment outcome* O2: Duration follow-up* O3: Adequate follow-up* RCT (20) Randomized? (1) Double blind? (1) Reviews (19)

559

S3: Selection of Controls* S4: Denition of Controls* Ca: Controlled for age* Cb: Controlled for . . .* E1: Ascertainment of exposure* E2: Method case control* E3: Nonresponse rate*

Description withdrawals/ dropouts (1) Randomization described and appropriate? (1) Blinding described and appropriate? (1) Randomization described and inappropriate? (1) Blinding described and inappropriate? (1)

Clearly-focused question? (1) Right type studies included? (1) Need to continue? Identied all relevant studies? (1) Quality assessment? (1) Results combined? (1) Presentation results? (1) Results precise? (1) Extrapolation possible? (1) Important outcomes considered? (1) Need for change policy? (1) Maximum: 10 points

Maximum: 9 stars

Maximum: 9 stars/3 stars

Maximum: 5 points

S1 to 4, criteria for selection; Ca, Cb, criteria for comparability; E1 to 3, criteria for exposure; O1 to 3, criteria for outcome. Italic criteria, criteria selected for cross-sectional studies.

guidelines (21). The criteria for the different studies are presented in Table 2. Methodologic quality was assessed by 2 independent, blinded researchers (MM and SV), ie, they were not acquainted with each others evaluation of the search results. After rating the selected articles, the results of both researchers were compared and differences were analyzed. In the case of disagreement, the reviewers screened the article a second time and the point of difference was discussed. Both reviewers could argue and convince the other to obtain a consensus. When consensus could not be reached, a third opinion was provided by the last author (JN). Data Abstraction Finally, the results were analyzed and the existing knowledge concerning central pain processing in RA was summarized. The methodologic quality of the different studies was taken into account when interpreting the ndings of the different studies. RESULTS Study Selection The study selection is presented in Figure 1. Twenty-four full-text articles were included in the qualitative synthesis of the review. Those 24 articles were screened and reviewed. Most studies were excluded based on the rst inclusion criterion: adult humans with RA.

Study Characteristics The properties of the 24 studies that fullled all inclusion criteria are presented in Table 3. Of the 24 studies, 12 were case control studies, 2 cross-sectional, 2 case reports, 6 reviews, 1 pilot trial (combined with an animal experiment that was not included), and 1 combined case control and randomized controlled trial. Methodologic Quality The methodologic quality was assessed based on the criteria presented in Table 2. In most cases (92.5% or 174 of the 188 items), the 2 researchers agreed. After a second review and a comparison of the 14 differences, the reviewers reached a consensus for 8 items. The remaining 6 points of discussion were solved after a third opinion. The nal score of each study is presented in Table 3, with the explanation for the loss of points. The pilot trial and the 2 case reports were not evaluated, given the lack of a checklist for this study design, probably due to the awareness of the low validity intrinsic to the study design. The reviews obtained overall low scores, because none were systematic reviews. Because of that, there was no need to continue after the rst 2 questions. Case-control studies often lost points as the result of an inappropriate selection of their control subjects, representativeness of the patient group, and comparability of the different groups. The cross-sectional studies were hard to evaluate with the checklist, but the major concern was that outcome measures were self-reported measures.

560
Records idened through database searching (n = 684 )

Central sensitization in patients with RA

Addional records idened through other sources (via expert) (n = 1 )

Idencaon

Records aer duplicates removed (n = 424 )

Screening

Records screened (n = 424 )

Records excluded (n = 364 )

Eligibility

Full-text arcles assessed for eligibility (n = 60 )

Full-text arcles excluded, with reasons (n = 36 )

Included

Studies included in qualitave synthesis (n = 24 )

Figure 1 Flowchart study selection (adapted from Moher et al., 2009). (Color version of gure is available online.)

Evidence Suggestive for CNS Involvement Besides listing the different search results and their characteristics, this review attempts to list the present knowledge on central pain processing in RA. Because of the small amount of available studies and the variety of design and quality, all 24 studies are included in the substantive overview, regardless of their study design or methodologic quality. However, methodologic quality or level of evidence is taken into account when interpreting the content of the different articles. To allow a deeper interpretation and translation of the study results, characteristics of the included RA patients are presented in Table 4. Clinical Manifestation of RA Although RA is often considered an example of nociceptive pain (22), the magnitude of symptoms is not always related to the severity of the underlying disease and symptoms may occur spontaneously or persist even when disease exacerbations are reduced (1). Furthermore, patients with RA exhibit symmetrical disease in both their clinical symptoms of pain and stiffness as well as radiographic damage. Bilateral neural loops releasing proinammatory cytokines such as substance P and calcitonin gene-related peptide could be responsible (23). Inversely, patients with paralyzing lesions of the central or peripheral nervous system who later develop RA are spared of joint involvement in the paretic limbs (7). Onehalf of the RA patients regularly visiting a sauna report an exacerbation of pain up to 12 hours after the heat exposure. This suggests a general response, such as general sympathetic activity, rather than the activation of peripheral afferents (24).

Hyperalgesia and Allodynia In patients with RA, decreased pain thresholds have been found in both structures overlying inamed joints and noninamed tissues. This nding is replicated in various studies using different assessment tools (ie, pressure algometers, von Frey bers, thermal stimuli). Four studies using pressure algometry at both nonarticular tissues and at articular/inamed sites (possibly peripheral sensitization) reported hyperalgesia to pressure (2529) Using von Frey bers, mechanical allodynia at inamed tissue was reported in the study of Hendiani and coworkers (30) and at nonarticular sites in response to capsaicin in 2 other studies (1,23). Thermal hyperalgesia, at both nonarticular and articular sites (26,28), was reported in 2 studies. All these studies reported hyerpalgesia or allodynia in RA patients compared with a healthy control group, besides the study of Alstergren and coworkers (29), which compared seropositive versus seronegative RA patients. This enhancement of pain sensitivity appears to be magnied in individuals with RA of longer duration (28). Temporomandibular (TMJ) joint pressure pain thresholds (PPTs) appeared to be systematically modulated, mediated by circulating serotonin (5-HT) and not by local inammatory activity. Furthermore, TMJ PPTs were related to the glabella PPT and were not or only weakly related to other TMJ pain parameters (resting pain and movement pain) (29). Pinprick hyperalgesia induced by capsaicin is substantially greater over the forearms of patients with RA compared with normal controls (1). The random order of placebo or capsaicin application is likely to avoid several expectations, beliefs, or cognitions regarding the outcome

M. Meeus et al.

Table 3 Characteristics of the Included Studies Author Dhondt et al (1999) Study Design Case control RCT Case control Case control Case control Case control Case control Case control Case control Case control Case control Case control Case control Case control Cross-sectional Cross-sectional Case report Case report Pilot trial Review Review Review Review Review Review, Congress report Quality 6/9 4/5 6/9 7/9 5/9 7/9 5/9 2/9 6/9 3/9 5/9 2/9 4/9 4/9 2/3 2/3 / / / 1/10 0/10 0/10 1/10 0/10 0/10 Not systematic Not systematic Not systematic Not systematic Not systematic Not systematic S 2, 3, 4 Randomization not described C a, b; E 3 C a, b S 1, 2, 4; C b S 2, 3 S 1, 2, 3, 4 S 1, 2, 3, 4; C a, b; E 3 S 2, 3, 4 S 2, 3, 4; C a, b; E 3 C a, b; E 1,3 S 1, 2, 3, 4; C a, b; E 3 S 2, 3, 4; C a, b S 1, 2, 3, 4; E 3 O1: self-report, not blinded O1: self-report, not blinded Lacuna Study Aim PPT: RA (30) CON (30) PPT after oscillation (15 RA)/rest (15 RA) PPT: RA sero (74) RA sero (60) Response to pain (cortisol, IL-6, TNF): RA (19) CON (21) PPT: RA (15) CON (18) SLE (16) FM (16) Mechanical sensation and PT: RA (27) CON (28) OA (28) Wind-up: RA (12) CON (12) Response to pain (rCBF): RA (6) CON (6) DNIC: RA 1 year (11) RA 5 year (10) CON (21) Capsaicin-induced hyperalgesia: RA (35) CON (57) Pain response to sauna heat: RA (59) PNP (61) CNP (13) Capsaicin-induced contralateral hyperalgesia: RA (14) CON (14) Endomorphine expression in knee joint tissues: RA (10) OA (10) Wind-up: RA (25) CON (25) PPT disease activity and personality: RA (44) Pain depression brain activation: RA (20) Stiffness in phantom limb (3) Response to pain (opiate) (3) Nociceptive CNS activity after TNF- neutralization in RA (5) Molecular pathways in inammatory and rheumatic pain Pathophysiology of (non-) nociceptive rheumatologic pain Catastrophizing and pain Phantom limb pain unexplained rheumatologic pain Neuroplasticity and neurogenic inammation Neuro-endocrine-immune mechanisms and neuroplasticity

Alstergren et al (2008) Edwards et al (2009) Garcia-Fernandez et al (2009) Hendiani et al (2003) Hummel et al (2000) Jones et al (1997) Lefer et al (2002) Morris et al (1997) Nurmikko and Hietaharju (1992) Shenker et al (2008) Straub et al (2008) Wendler et al (2001) Konttinen et al (1992) Schweinhardt et al (2008) Haigh et al (2003) Jones et al (1991) Hess et al (2011) Bingham et al (2009) Dessein et al (2000) Edwards et al (2006) McCabe et al (2004) Niissalo et al (2002) Straub et al (2002)

561

The lacuna columns explain the loss of points; for the case-control studies the reasons are expressed in codes according to the criteria presented in Table 1. (P)PT, (pressure) pain threshold; , change; RA, rheumatoid arthritis; CON, control subjects; SLE, systemic lupus erythematosus; FM, bromyalgia; OA, osteoarthritis; PNP, peripheral neuropathic pain; CNP, central neuropathic pain; rCBF, regional cerebral blood ow; DNIC, diffuse noxious inhibitory controls; CNS, central nervous system.

562

Table 4 Characteristics of the RA Patients in the Included Studies Author Dhondt et al (1999) Alstergren et al (2008) Age in years 55.5 10.9 (34 to 70) Sero: Md 55 yr Sero: Md 45 yr 51.7 12.2 57 9.06 46.85 2.30 (20 to 66 yr) 41 (27 to 59 yr) 62 12.2 RA1: 47 yr (20 to 68 yr) RA5: 53 yr (37 to 67 yr) 57 (33 to 78 yr) 59.3 12.1 Gender 24 &, 6( Sero: 48&, 12( Sero: 66&, 8( 11&, 80( 15& 24&, 3( 8&, 4( 6& RA1: 7&, 4( RA5: 9&, 1( n 35 (gender?) 38&, 21( 9&, 5( 7&, 3( 21&, 4( 43&, 10( 14&, 6( n3 (gender?) n3 (gender?) 5& 23, 24, 25 yr n 1: minimal disease activity n 2: tested during are-up and in pain-free state DAS28 6 (deduced from graphs) Tender joint count 13 and swollen joint count 9 10.5 year (3 to 23 yr) 10.6 yr (2 to 37 yr) 5.33 yr 1.53 (0.25 to 35 yr) 1 yr 4 to 30 yr RA1: 1 yr RA5: 5 to 21 yr 3 mo ACR (1990) ARA (1988) ARA (1988) Ritchie index 33 22 RA1: CRP 18 mg/L n 5: mild* n 4: moderate n 2: severe RA5: CRP 28 mg/L n 5: mild n 2: moderate n 3: severe Duration 5 mo to 30 yr Sero: Md 8 yr Sero: Md 10 yr 8.3 yr Criteria ACR (1990) ACR (1990) Disease Activity EULAR joint counts: Tender joint count 6.7 5.9 and Swollen joint count 7.3 5.8 Sero: Sero: Painful joint regions 4 Md 6 ESR 32 mm/h 20 mm/h CRP 14 mg/L 15 mg/L Low-moderate activity: DAS28 3.1 1.4 CRP 3.3 3.9 g/mL

Edwards et al (2009) Garcia-Fernandez et al (2009) Hendiani et al (2003) Hummel et al (2000) Jones et al (1997) Lefer et al (2002)

ACR (1990)

Morris et al (1997) Nurmikko and Hietaharju (1992) Shenker et al (2008) Straub et al (2008) Wendler et al (2001) Konttinen et al (1992) Schweinhardt et al (2008) Haigh et al (2003) Jones et al (1991)

ARA (1988) ARA (1988) DAS 5.12 ESR 20.2 mm/h 9.4

52 (28 to 66 yr) 67.0 3.1 45 yr (20 to 66 yr) 51 12 57 11.7 63, 77, 79

5.9 yr (2 to 29 yr)

ARA (1988) ACR (1990) ARA (1988) ARA (1988)

Central sensitization in patients with RA

Ritchie Index Md 13.0 ESR (after 1 and 2 hr) Md 13 and 45 mm/h ESR 35 mm/h 21 CRP 16 mg/L 12 Kaarelas joint score index 11 4 Low BDI: DAS 28 High BDI: DAS 28 6.1 0.5 5.4 1.1

Hess et al (2011)

56.3 8.2

8.5 yr 3.3

Md, median; ACR, American College for Rheumatology; ARA, American Rheumatism Association; OK, patients were allowed to continue medication; no, patients taking medication were excluded or had to refrain taking them; DAS 28, disease activity score using 28 joint counts; *, dened following assessors global assessment of disease activity; BDI, Beck Depression Inventory; ESR, erythrocyte sedimentation rate; CRP, C-reactive protein.

M. Meeus et al.

563

of a certain application and thus the hyperalgesic response seems to be the result of the substance injected, rather than being due to other factors. On the other hand, Shenker and coworkers (23) described areas of contralateral hyperalgesia and allodynia occurring in a signicant proportion of both healthy volunteers and patients with RA following an intradermal injection of capsaicin. However, these results are generated by a case-control study with very low methodologic quality. Hypoesthesia Paradoxically to the hyperalgesia, several studies report simultaneous cutaneous hypoesthesia in the involved joint in RA (28,30). Sensory aberrations are rather present in patients with longer disease duration. Patients with a short duration of RA only present increased sensitivity to pressure pain in structures overlying the inamed joint, whereas patients with a longer duration of RA also had increased sensitivity to innocuous cold and decreased sensitivity to light touch in the area overlying the inamed joint compared with the corresponding homologous area in controls, suggesting widespread altered central processing of somatosensory functions (28). Possible Mechanisms Causing Central Sensitization in RA The generalized nature of the enhanced sensitivity to pain observed in these patients suggests alterations in pain processing at the level of the CNS (26). These ndings may be due to a disorder of endogenous inhibitory systems, windup as the result of repeated noxious stimuli, and/or overactivity in descending pain facilitatory pathways, thus affecting pain processing in the entire individual (28). Effects of Repeated Noxious Stimuli In comparison with healthy controls, RA patients exhibit increased cortical response amplitudes when trains of painful intranasal stimuli with gaseous CO2 were presented with an interstimulus interval of 2 seconds. Hummel and coworkers (31) interpreted these results in terms of a decreased adaptation to painful stimuli, which in turn might contribute to chronic pain in RA. Habituation to painful stimuli appeared to be less pronounced in patients. These observations would rather reect adaptive brain processes than peripheral effects, because an increase of the patients responses was only found with trains of phasic painful stimuli, but not with single tonic stimuli, and non-RA-inamed tissue (nasal mucosa) was stimulated (15,31). Indirect evidence for enhanced temporal summation or windup is also available in studies where PPTs are measured repeatedly. During 2 consecutive trials or 2 trials spaced at 15 minutes, a clear tendency toward a reduction of the PPT was observed in RA patients and patients with

bromyalgia. In contrast, women with SLE and those without pain showed levels of PPT that were maintained stable (27). Also, in the study of Dhondt and coworkers (25), repeated measurement of the PPT in RA patients showed a decrease in all subjects, which was slightly less pronounced after a single manual oscillation session than after rest, possibly as the result of gate control or a more global inhibition of nociception. Given the evidence provided in 4 different case-control studies with moderate quality, enhanced windup seems plausible in patients with RA. Endogenous Nociceptive Inhibition High-intensity stimulation of C and A nociceptive bers through the administration of a heterotopic noxious stimulus results in an antinociceptive endogenous response. This causes a generalized inhibition of the wide dynamic range neurons, leading to pain relief at sites not even initially involved in nociception. This process is called diffuse noxious inhibitory control (DNIC) (22). A dysfunction of DNIC-like mechanisms in humans may be induced and maintained by nociceptive input from joints. Nevertheless, Lefer and coworkers (28) found no dysfunction of DNIC in RA patients. The response to a heterotopic noxious conditioning stimulus was similar (increased PPT) in both patients with short and long disease duration and their corresponding controls, indicating unaltered function of DNIC-related mechanisms (28). Dessein and coworkers (22) hypothesized that the observed hypoactivity of the stress system in RA, with inappropriately low cortisol production, would mediate the process of central sensitization. Decreased cortisol production would be expected to mediate the central sensitization process by increasing glutamate concentrations, nerve growth factor, and prostaglandins (22). This is, however, only hypothesized on the basis of a nonsystematic review (22) and partly countered by a methodologic good case-control study in which the cortisol response to several experimental pain procedures was not different from controls (26). Concerning 5-HT, the literature only reports their role in central nociceptive inhibition, but it is not specically studied in patients with RA. 5-HT is released in the spinal cord to inhibit the central sensitization process directly and through the use of opioidergic interneurons, by way of both pre- and postsynaptic inhibition (8,22). It appeared to modulate TMJ PPTs in RA patients, but in an opposite way in seronegative or seropositive patients (29). Cognitive Emotional Sensitization Enhanced pain facilitation may be caused by cognitive emotional sensitization. Activity in descending pathways can be modulated, for example, by forebrain products such as cognitions, emotions, attention, and motivation. Every strongly relevant individual concern, like fear, can

564

Central sensitization in patients with RA

yield cognitive bias or cognitive-emotional sensitization (32). Chronic pain, in itself, could cause distress and activation of the stress response, resulting in a lower pain threshold. Moreover, hypervigilance could increase pain sensitivity in chronic pain patients. However, its role appears to be small or inexistent in RA (1). In response to noxious heat pain, RA patients show remarkably damped cortical and subcortical responses compared with the control group. Signicant differences between the 2 groups were observed in the prefrontal and anterior cingulated and cingulofrontal transition cortical areas, which is distinct from patients with psychogenically maintained pain or atypical facial pain. This suggests that different physiologic and psychologic mechanisms are operating to modify responses to noxious stimulation in these 2 types of pain, probably because of differences in processing of affective and attentional components of nociceptive processing. On the 1 hand, certain personality traits may play a role. Atypical facial pain patients would be more depressed and respond to hyperattentional affective noxious and nonnoxious inputs; the converse is the case with the RA group (33). Nevertheless, depressive symptoms are related to the cerebral processing of provoked joint pain, also in patients with RA. Activation in limbic areas and brain regions concerned with diseaserelevant processing during provoked joint pain is related to depressive symptoms in RA. The medial prefrontal cortex mediates the association between depressive symptoms and 1 component of clinical pain. Based on these results and on previous evidence that some of these regions are important for the emotional augmentation of pain, it is suggested that these brain regions contribute to the maintenance and exacerbation of pain in RA (34). Substantial changes in opioid receptor binding are found in the cortical and subcortical structures, which are either part of or closely related to the limbic system or the medial pain system, or both, in patients with active RA (35). Furthermore, centers of the limbic system, such as the cingulate and insular cortex, appear to be modulated by tumor necrosis factor alpha (TNF-) activity, because TNF- blockade blocked nociceptive activation of the limbic system in a pilot study in 5 women with RA. The cingulate cortex is particularly involved in forming and processing of emotions and memory, whereas the insular region is important for the subjective sense of the inner body, including the experience for pain and emotions. These ndings suggest that not only pain responses are affected by TNF- blockade but also CNS structures relevant for creating pain perception, emotions, and body experience (36). Besides certain personality traits, perceptions, and emotions, coping strategies also may modulate central pain processing. Edwards and coworkers (37) reviewed the effects of catastrophizing on the experience of pain and pain-related sequelae in those with RA. Catastrophizing is positively related, in both cross-sectional and prospective studies, to the reported severity of pain,

affective distress, muscle and joint tenderness, painrelated disability, and, potentially, inammatory disease activity. Moreover, these associations generally persist after controlling for symptoms of depression. Some of the proposed mechanisms interfere with paincoping and benecial health behaviors, the increase of attention to pain, and amplied pain processing in the CNS. Catastrophizing could promote sensitization or interfere with nociceptive inhibition (37). In a crosssectional study, pain tenderness scores were, however, not affected by the subjects personality (locus of control, coping, etc) (38). Controversial results regarding the role of personality and coping strategies seem to exist and the level of evidence (eg, case report, nonsystematic review, crosssectional studies) and the methodologic quality of the different studies are not high, and therefore, further study into this topic is required. Cytokines and Neuropeptides Neuropeptides are synthesized in dorsal root and autonomic ganglion neurons and transported to peripheral nerve terminals. Sensitized peripheral nerve terminals are particularly prone to release these substances. In turn, other bioactive factors are released, like cytokines, prostaglandins, NO, and CO, which are capable of acting back on nerve terminals and bers. Chronic inammation alters the processing of neuropeptides, leading to long-term functional/structural changes of innervation, changes in neurovascular regulation, and modulation of immune function. Ultimately, this process can lead to central sensitization and amplication of inammatory responses and activation of the HPA (Hypothalamus Pituitary Adrenal) axis and neuroendocrine responses. Neuropeptides affect effector cells by regulating their cytokine release and, vice versa, peripherally released cytokines can induce elevated cytokine expression levels in the CNS (7). In RA patients compared with controls, acute pain induction is associated with elevations in serum TNF- levels that last for at least 1 hour. These results indicate that the normal enhanced release of pro-inammatory cytokines (at least for TNF-) may be magnied in the context of RA (26). In contrast, 24 hours after neutralization of TNF- (infusion of a monoclonal antibody to TNF-), nociceptive CNS activity in the thalamus and somatosensoric cortex, but also the activation of the limbic system, is blocked, while clinical and laboratory markers of inammation were not yet affected. It seems that (36) while these pro-inammatory cytokines may contribute to pain sensitization, expression of anti-inammatory endomorphins is also abundant in inamed tissue from patients with RA. These endomorphins inhibit synovial secretion of interleukin (IL)-6 and IL-8 in patients with RA and have subsequent anti-inammatory effects. Nevertheless, endomorphin-positive cells were more abundant in osteoarthritis than in RA and the inhibition of IL-8 was weaker in RA. This difference might contribute to increased pain

M. Meeus et al.

565

processing in RA as compared with osteoarthritis. The role of these endomorphins in pain processing remains, however, rather obscure, because there is no doseresponse relation and it seems that lower doses of the peptides exerted stronger anti-inammatory effects. Possibly, opioidergic pathways are down-regulated due to receptor desensitization (39). DISCUSSION The goal of the present systematic literature search was to review the scientic literature addressing central nociceptive processing in patients with RA. Most authors seem to agree with the primary and secondary hyperalgesia in RA, regardless of the method used (2530) Generalized hyperalgesia and allodynia may be the result of long-lasting nociceptive bombardment from inamed joints, giving rise to peripheral and central sensitization of nociceptive afferents, as well as to an increased activity of the sympathetic nervous system (25,28). All these results suggest a cardinal role for systemic modulation, rather than for local inammatory factors (29). Also, other physical factors, such as joint deformity, are unlikely to inuence the results, as the site of testing in the capsaicin study of Morris and coworkers (1) was well away from inamed or damaged tissues. In addition, the lack of placebo-induced hyperalgesia in the placebo-controlled double-blind study seems to exclude the possibility that hypervigilance or psychologic factors may have contributed (1). The observed contralateral responses in the study of Shenker and coworkers (23) are, however, unlikely to reect a global change in pain thresholds as only the topographically precise contralateral area was sensitized and there was no sensitization of anatomically mismatched areas. The responses are therefore a result of a specic contralateral effect rather than a generalized sensitization of the CNS. The period between the capsaicin injection and testing the contralateral side was several minutes, which is too long an interval to differentiate whether the contralateral response either occurs directly through the spinal cord or is mediated via long supraspinal pathways (23), because both pathways could have been recruited during this time period. A shorter time interval would possibly only allow a response directly mediated by the spinal cord, whereas supraspinal mechanisms would not have been recruited (23). The latter study was the sole study that did not nd generalized hyperalgesia, but hyperalgesia was restricted to the contralateral area of the affected/stimulated area. This restricted contralateral area deserves more attention in the future to conrm or reject the present ndings. Simultaneously to the allodynia and hyperalgesia, studies report cutaneous hypoesthesia (28,30). Descending inhibitory systems, which become activated after tissue insult to counteract the enhanced excitation of peripheral afferents, may alter somatosensory perception, to prevent the development of even lower pain thresholds (30). An inhibitory effect of nociceptive activity on mechanorecep-

tive input is supported by ndings of decreased sensitivity to light touch in a painful area following induction of pain in healthy subjects (28). Preliminary indirect evidence for the involvement of the CNS is provided by reporting the generalized hyperalgesia. The question remains as to which mechanisms could be responsible for the abnormal central pain processing. Based on the gathered literature, the main problem seems to be the inammation-evoked peripheral sensitization, leading to windup of the dorsal horn. Any joint pathology can produce severe pain resulting from sensitization of primary afferent nociceptive neurons (4). Peripheral sensitization is mediated by local components of the local inammatory reaction released by the microenvironment of the inamed and injured tissue, such as bradykinin, prostaglandin, and probably neuropeptides and cytokines (2,5). The peripheral terminals of A and C bers, which are activated only by noxious stimuli under noninamed conditions, express many receptors and ion channels that recognize the various inammatory mediators in the vicinity. Thus, the local inammation results in the release of multiple factors that activate local nerve terminals involved in pain perception (2). In a state of inammation-evoked neuroplasticity, sensory stimuli such as light pressure or movements in the working range of the joint may evoke much larger responses in the nervous system than under normal conditions (3,6). Inactive, silent, or sleeping nociceptors alter their function during inammation. A bers, which normally respond only to tactile and not to noxious stimuli, can mediate a longlasting tactile allodynic state during chronic inammation (2,7). The message noxious stimulus is signaled to the spinal cord by innocuous stimuli, which are usually not painful (for details, see (3)). Consequently, peripheral sensitization implies increased spontaneous activity, decreased activation threshold of the nociceptive bers, increased responsiveness to suprathreshold stimuli, and increased local release of neuropeptides on stimulation and recruitment of silent nociceptive bers (8). Due to this continuous bombardment of nociceptive information, spinal neurons become sensitized. RA patients indeed showed increased responses to repetitive stimuli, indicative for increased efcacy of windup (15,25,27,31). Endogenous nociceptive inhibition seems to function rather normally (26,28), although the functioning of cortisol, 5-HT, and opioids in nociceptive inhibition in RA needs further clarication. Stress responses in RA patients are complex and vary as a function of the stimulus. Thus, rather than a global generalized hyporesponsiveness of the HPA axis to stress in RA, there appears to be a signicant stimulus specicity to stress response proles (26). Further clarication is also needed for neuropeptides, the role of which is reviewed by Niissalo and coworkers (7), but their function in central pain processing remains rather speculative. The same goes for the endomorphins, detected in inamed tissues of RA patients (39). The relation with pro-inammatory cytokines, such as IL-6 or

566

Central sensitization in patients with RA

IL-8 and TNF-, of which release is magnied in RA in response to pain (26), may be important in central nociceptive processing (36). Neutralization of TNF- rapidly seems to downregulate nociceptive brain activity, long before it achieves anti-inammatory effects in the joints of RA patients (36). Finally, it is not clear to what extent pain cognitions, behaviors, or personality traits modulate nociception in RA. Cognitive emotional sensitization may be important in other chronic pain populations, especially in those without a clear source (12), but seems to be less obvious in RA. The inuence of attention, depression, locus of control, etc, on pain remains controversial in RA. TNF- may, however, play a role in nociceptive activation of the limbic system (36). Although we recognize the relevance of animal studies in understanding pathophysiologic mechanisms and in studying drug efcacy, we chose not to include animal studies in the present review. In chronic pain research, studies in laboratory animals with experimentally induced arthritis or pain are only partially predictive or valid for human outcomes (40). These experimentally induced conditions may not adequately model the natural disease process that leads to pain (41). The comprehensive assessment of pain in humans extends beyond pain severity to include how pain interferes with the patients functioning through daily living. This evaluation of spontaneous pain and its impact on daily living as the pain process evolves parallels the human condition in a way that the animal models do not. Mood disturbance might also be a core symptom of chronic pain in humans. It is important that chronic pain models approach as closely as possible the diverse symptoms of the associated chronic-pain condition (eg, neuroendocrine disturbance, sleep patterns, and mood imbalance) (40). The goal of the present study was to gain insight in the multifactorial nature of chronic pain and central pain processing in humans, including the subjective nature of chronic pain and psychosocial factors associated to chronic pain. Therefore, animal studies were not included. The multifactorial and subjective nature of chronic pain is also reected in the fact that in all studies disease activity in women is higher as expressed by the standard measures, which is primarily due to the psychometric properties of the instruments (tender joint count, etc) and not really due to a more severe disease course in women, as objective measures often seem to be the same in female and males. Furthermore, women are known to have higher pain levels and different coping mechanisms than men (42). Given the fact that predominantly women are affected by RA, further studies should account for gender differences while studying pain processing in RA. Besides gender differences, age, disease duration, etc, may play a role. Therefore, Table 4 might be interesting while interpreting the data of the different studies. Finally, we did decide to include all study designs and all quality levels in the present review, given the shortage of articles on this topic. Of course, the results should be

interpreted in the light of this limitation. This way, we wanted to underline the dearth of knowledge, even when all articles were considered. Only preliminary evidence is currently available and further research is warranted. To conclude, this review shows that progress has been made toward understanding the neurophysiologic mechanisms of chronic pain in RA, but far more research must be done. Many studies concentrated on peripheral mechanisms or were conducted on animals. Studies on central nociceptive processing in adults with RA are scarce, situated in the lowest levels of the hierarchy of evidence based medicine, or of poor methodologic quality. The 1 fact that seems rather clear is the existence of generalized hyperalgesia and allodynia, indicative of the involvement of the CNS. Several hypotheses have been proposed and tested, but results are not always consistent and the frontiers of knowledge became clear in the present systematic review. Future research should try to clarify which mechanisms could be responsible for the chronic generalized pain that is not always related to the severity of the underlying disease and that can occur spontaneously or in nonaffected body parts. Unlike other sensory systems, where neuronal plasticity usually improves performance, plasticity of the nociceptive system seems to produce the opposite, ie, in the condition of an underlying chronic arthritis, it appears as if the nociceptive system contributes to pathologic pain (15). Elucidation of such processes should lead to a better management of central pain processing in the future. ACKNOWLEDGMENT Mira Meeus has a postdoctoral research fellowship from the Research Foundation Flanders (FWO). REFERENCES
1. Morris VH, Cruwys SC, Kidd BL. Characterisation of capsaicininduced mechanical hyperalgesia as a marker for altered nociceptive processing in patients with rheumatoid arthritis. Pain 1997; 71(2):179-86. 2. Cheng HY, Penninger JM. DREAMing about arthritic pain. Ann Rheumat Dis 2004;63(Suppl 2):ii72-ii5. 3. Schaible HG, Grubb BD. Afferent and spinal mechanisms of joint pain. Pain 1993;55(1):5-54. 4. Bingham B, Ajit SK, Blake DR, Samad TA. The molecular basis of pain and its clinical implications in rheumatology. Nat Clin Pract 2009;5(1):28-37. 5. Straub RH, Schaible HG, Wahle M, Schedlowski M, Neeck G, Buttgereit F. [Neuroendocrine-immunologic mechanisms in rheumatic diseasesa congress report]. Z Rheumatol 2002; 61(2):195-200. 6. Schaible HG. Why does an inammation in the joint hurt? Br J Rheumatol 1996;35(5):405-6. 7. Niissalo S, Hukkanen M, Imai S, Tornwall J, Konttinen YT. Neuropeptides in experimental and degenerative arthritis. Ann NY Acad Sci 2002;966:384-99. 8. Konttinen YT, Kemppinen P, Segerberg M, Hukkanen M, Rees R, Santavirta S, et al. Peripheral and spinal neural mechanisms in arthritis, with particular reference to treatment of inammation and pain. Arthritis Rheum 1994;37(7):965-82. 9. Giesecke T, Gracely RH, Grant MA, Nachemson A, Petzke F, Williams DA, et al. Evidence of augmented central pain process-

M. Meeus et al.
ing in idiopathic chronic low back pain. Arthritis Rheum 2004;50(2):613-23. Banic B, Petersen-Felix S, Andersen OK, Radanov BP, Villiger PM, Arendt-Nielsen L, et al. Evidence for spinal cord hypersensitivity in chronic pain after whiplash injury and in bromyalgia. Pain 2004;107(12):7-15. Meeus M, Nijs J, Huybrechts S, Truijen S. Evidence for generalized hyperalgesia in chronic fatigue syndrome: a case control study. Clin Rheumatol 2010;29(4):393-8. Meeus M, Nijs J. Central sensitization: a biopsychosocial explanation for chronic widespread pain in patients with bromyalgia and chronic fatigue syndrome. Clin Rheumatol 2007;26(4):465-73. Nijs J, Meeus M, Van Oosterwijck J, Ickmans K, Moorkens G, Hans G, et al. In the mind or in the brain? Scientic evidence for central sensitization in chronic fatigue syndrome. Eur J Clin Investig 2012;42(2):203-12. Neugebauer V, Schaible HG. Evidence for a central component in the sensitization of spinal neurons with joint input during development of acute arthritis in cats knee. J Neurophysiol 1990; 64(1):299-311. Wendler J, Hummel T, Reissinger M, Manger B, Pauli E, Kalden JR, et al. Patients with rheumatoid arthritis adapt differently to repetitive painful stimuli compared to healthy controls. J Clin Neurosci 2001;8(3):272-7. Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000; 283(15):2008-12. Meeus M, Nijs J, Meirleir KD. Chronic musculoskeletal pain in patients with the chronic fatigue syndrome: a systematic review. Eur J Pain 2007;11(4):377-86. Wells GA, Shea B, OConnell D, Petersen J, Welch V, Losos M, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomized studies in meta-analyses. [cited; Available from: http://www.ohri.ca/programs/clinical_epidemiology/ oxford.htm Dutch Cochrane Centre. [cited; Available from: http://www. cochrane.nl/nl/newPage1.html] Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996; 17(1):1-12. Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF. Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement. Quality of Reporting of Meta-analyses. Lancet 1999;354(9193):1896-900. Dessein PH, Shipton EA, Budd K. Nociceptive and non-nociceptive rheumatological pain: recent developments in the understanding of pathophysiology and management in rheumatoid arthritis and bromyalgia. Pain Rev 2000;7:67-79. Shenker NG, Haigh RC, Mapp PI, Harris N, Blake DR. Contralateral hyperalgesia and allodynia following intradermal capsaicin injection in man. Rheumatology (Oxford) 2008;47(9):1417-21. Nurmikko T, Hietaharju A. Effect of exposure to sauna heat on neuropathic and rheumatoid pain. Pain 1992;49(1):43-51. Dhondt W, Willaeys T, Verbruggen LA, Oostendorp RA, Duquet W. Pain threshold in patients with rheumatoid arthritis and effect of manual oscillations. Scand J Rheumatol 1999;28(2):88-93.

567
26. Edwards RR, Wasan AD, Bingham CO 3rd, Bathon J, Haythornthwaite JA, Smith MT, et al. Enhanced reactivity to pain in patients with rheumatoid arthritis. Arthritis Res Therapy 2009; 11(3):R61. 27. Garcia-Fernandez E, Godoy-Izquierdo D, Prez-Garcia M, JimnezAlonso J, Lopez-Chicheri I, Godoy JF. Differences in pressure-pain threshold between healthy women and patients with bromyalgia syndrome, systemic lupus eryhtematosus, and rheumatoid arthritis. J Musculoskel Pain 2009;17(2):139-54. 28. Lefer AS, Kosek E, Lerndal T, Nordmark B, Hansson P. Somatosensory perception and function of diffuse noxious inhibitory controls (DNIC) in patients suffering from rheumatoid arthritis. Eur J Pain 2002;6(2):161-76. 29. Alstergren P, Fredriksson L, Kopp S. Temporomandibular joint pressure pain threshold is systemically modulated in rheumatoid arthritis. J Orofac Pain 2008;22(3):231-8. 30. Hendiani JA, Westlund KN, Lawand N, Goel N, Lisse J, McNearney T. Mechanical sensation and pain thresholds in patients with chronic arthropathies. J Pain 2003;4(4):203-11. 31. Hummel T, Schiessl C, Wendler J, Kobal G. Peripheral and central nervous changes in patients with rheumatoid arthritis in response to repetitive painful stimulation. Int J Psychophysiol 2000; 37(2):177-83. 32. Brosschot JF. Cognitive-emotional sensitization and somatic health complaints. Scand J Psychol 2002;43(2):113-21. 33. Jones AK, Derbyshire SW. Reduced cortical responses to noxious heat in patients with rheumatoid arthritis. Ann Rheumat Dis 1997;56(10):601-7. 34. Schweinhardt P, Kalk N, Wartolowska K, Chessell I, Wordsworth P, Tracey I. Investigation into the neural correlates of emotional augmentation of clinical pain. NeuroImage 2008;40(2):759-66. 35. Jones AK, Liyi Q, Cunningham VV, Brown DW, Ha-Kawa S, Fujiwara T, et al. Endogenous opiate response to pain in rheumatoid arthritis and cortical and subcortical response to pain in normal volunteers using positron emission tomography. Int J Clin Pharmacol Res 1991;11(6):261-6. 36. Hess A, Axmann R, Rech J, Finzel S, Heindl C, Kreitz S, et al. Blockade of TNF-alpha rapidly inhibits pain responses in the central nervous system. Proc Natl Acad Sci USA 2011;108(9): 3731-6. 37. Edwards RR, Bingham CO 3rd, Bathon J, Haythornthwaite JA. Catastrophizing and pain in arthritis, bromyalgia, and other rheumatic diseases. Arthritis Rheum 2006;55(2):325-32. 38. Konttinen YT, Honkanen VE, Gronblad M, Keinonen M, Santavirta N, Santavirta S. The relation of extraarticular tenderness to inammatory joint disease and personality in patients with rheumatoid arthritis. J Rheumatol 1992;19(6):851-5. 39. Straub RH, Wolff C, Fassold A, Hofbauer R, Chover-Gonzalez A, Richards LJ, et al. Antiinammatory role of endomorphins in osteoarthritis, rheumatoid arthritis, and adjuvant-induced polyarthritis. Arthritis Rheum 2008;58(2):456-66. 40. Blackburn-Munro G. Pain-like behaviours in animals how human are they? Trends Pharmacol Sci 2004;25(6):299-305. 41. Brown DC, Boston R, Coyne JC, Farrar JT. A novel approach to the use of animals in studies of pain: validation of the canine brief pain inventory in canine bone cancer. Pain Med 2009;10(1):133-42. 42. Berkley KJ. Sex differences in pain. Behav Brain Sci 1997;20(3): 371-80.

10.

11. 12. 13.

14.

15.

16.

17. 18.

19. 20.

21.

22.

23. 24. 25.