Abstract
Objectives
Pain sensitization contributes to the complex osteoarthritis (OA) pain experience. The relation between hand OA imaging features and clinically assessed pain sensitization is largely unexplored. We aimed to examine the association of structural and inflammatory features of hand OA with local pressure pain thresholds in the Nor-Hand study.
Methods
The cross-sectional relation of structural radiographic severity (Kellgren-Lawrence grade 0–4 and absence/presence of erosive joint disease) and ultrasound-detected inflammatory severity (grey scale synovitis grade 0–3 and absence/presence of power Doppler activity) to pressure pain threshold of two finger joints was examined in multilevel regression analyses adjusted for age, sex and BMI.
Results
In total, 570 joints from 285 persons were included. Greater structural and inflammatory severity was associated with lower pressure pain thresholds (adjusted beta (95% CI) −0.5 (−0.6, −0.4) per Kellgren-Lawrence grade increase, −1.4 (−1.8, −0.9) for erosive vs non-erosive joints, −0.7 (−0.9, −0.6) per grey-scale synovitis grade increase and −1.5 (−1.8, −1.1) for joints with vs without power Doppler activity).
Conclusions
Greater severity of structural and inflammatory hand OA pathologies was associated with lower pressure pain thresholds at finger joints, indicating pain sensitization. Our results indicate that pain sensitization might be driven by structural and inflammatory pathology in hand OA.
INTRODUCTION
Pain is the main symptom for patients with hand osteoarthritis (OA) and represents a major health care challenge[1]. About 14% of women and 7% of men between 40 and 84 years of age are estimated to have symptomatic hand OA[2]. Although OA is one of the most prevalent chronic pain conditions worldwide, treatment options remain symptomatic and both traditional analgesics and non-pharmacological strategies have limited effect on pain or problematic side effects. The lack of effective analgesics may be due to our poor understanding of the determinants of OA related pain. Increased knowledge of the mechanisms causing OA pain is therefore much needed to develop new and better strategies to manage and prevent pain.
A peripheral nociceptive input is traditionally believed to cause OA pain and both structural and inflammatory changes in finger joints are associated with pain[3–5]. However, by which mechanisms joint pathologies mediate pain is unclear. Alterations in the peripheral and central sensory nervous system, called peripheral and central sensitization, allows pain signaling to be facilitated and cause increased pain experience. These mechanisms may be induced by injury, from for example mechanic pressure or inflammatory cytokines, and are proposed to explain why OA pain become chronic and persistent for a subgroup of patients. Clinically assessed signs of pain sensitization, using quantitative sensory testing methods, is related to presence and severity of OA pain in knee and hand OA[6, 7]. Pain sensitization is acknowledged as a clinically important treatment target. Yet, whether there are certain pathological features that cause sensitization and whether these are potential targets for prevention or treatment is largely unknown.
In animal studies, OA-related tissue damage and inflammation have been associated with peripheral sensitization to mechanical stimuli[8]. The excitation threshold for local nociceptors and the transmission of pain signals is lowered and cause increased sensitivity to painful stimuli (hyperalgesia) and painful sensation from normally non-painful stimuli (allodynia), consequently aggravating OA pain. Pain sensitization is difficult to investigate in humans because of the complexity of the many factors that influence pain perception. Sensory testing of mechanical pressure pain threshold at the skin in close proximity to an affected joint is considered to reflect mechanisms of peripheral and/or central sensitization[9]. MRI-detected inflammation in knee OA, but not radiographic severity, was associated with development and worsening of local pressure pain sensitivity at the knee [10]. In contrast, a small hand OA study found that greater structural damage was associated with greater local sensitivity to mechanical pressure pain stimuli[11]. However, the study sample was small (n=13) and they did not report any data of inflammation. Inflammation in hand OA is important and might precede damage of cartilage and bone as an inducer of sensitization.
More knowledge about the mechanisms by which pain sensitization occurs in OA is needed, especially for hand OA where the role of inflammation in the pathogenesis of pain sensitization is unknown. Hence, our objective was to explore the cross-sectional association of structural radiographic features and ultrasound-detected inflammatory features with local pressure pain threshold in a large hand OA study. Second, we aimed to examine whether the observed associations were different between joints with and without pain.
METHODS
Study design and population
We used baseline data from the Nor-Hand study, including 300 persons with hand OA. Detailed inclusion and exclusion criteria have been published[12]. Participants received oral and written information and provided their written informed consent to participate. The Norwegian Regional Committee for Medical and Health Research Ethics (Ref. no: 2014/2057) approved the study.
Pressure pain threshold (PPT) of painful and non-painful finger joints
We tested PPTs at two joints of each participant among the 2nd-5th distal and 1st-5th proximal interphalangeal (DIP/PIP) joints; the joint they reported to be “the most painful one on a daily basis”, and further a non-painful joint. If none of the joints were reported to be painful the joint with the most severe clinical OA (swelling and/or bony enlargements) was chosen. If none of the joints were pain-free, the joint with the least pain and no or least clinical OA was chosen. A handheld algometer (FPIX25, 1cm2 flat rubber probe) was applied in a perpendicular direction on the dorsal aspect of the joint with increasing pressure (0.5kilogram/second). The participants were instructed to say “stop” when the pressure first changed to slight pain. The average value (kg/cm2) of three tests from each joint was recorded[12]. In a subset of 9 participants, test-retest reliability of PPT was found to be moderate to good (intraclass correlation coefficient 0.52–0.61).
Hand radiographs
All participants underwent bilateral posteroanterior hand radiographs. One experienced reader (IKH) scored all hand joints for OA severity on a 0–4 scale using a modified Kellgren-Lawrence (KL) scale[2] and the DIP/PIP joints using the Verbuggen-Veys (VV) anatomical phase score. Joints in the erosive or remodelling phases were defined as erosive[13]. The DIP/PIP joints on 20 radiographs were re-assessed for intra-reader reliability, which was excellent (KL grade 0–3: kappa with linear weighting = 0.92, absence/presence of erosions yes/no: kappa = 0.98).
Ultrasound
On the same day as PPT testing, a trained medical student (NRA) performed the ultrasound examinations using a General Electric (GE) Logic S8 ultrasound machine with a linear 6–15MHz probe and a preset for optimal grey-scale (GS) synovitis and power Doppler (PD) (pulse repetition frequency 0.6kHz, frequency 7.7MHz). Initial scorings were done in consensus with an experienced ultrasonographer (AM).
The hand examination was performed with the patient’s hands resting flat. All hand joints were scanned dorsally with longitudinal projection from the radial to the ulnar side of each joint. An additional transverse scanning was performed when presence of pathology was uncertain. GS synovitis and PD signals were scored on 0–3 scales[14]. Due to low frequency of PD activity grade 2–3, we dichotomized this variable (grade 0 vs 1–3). Inter-reader reliability of DIP/PIP joints in 10 participants between the medical student (NRA) and the expert (AM) was good (prevalence and bias adjusted kappa values for categorical variables with linear weighting; GS synovitis grade 0–3 = 0.80, absence/presence of PD activity = 0.79).
Statistical analysis
Our study sample includes two joints per person. PPTs of two joints within one person are likely to correlate. To account for this within-person effect, we conducted mixed model regression analyses. The association between each structural and inflammatory imaging feature (independent variables) and PPT (dependent variable) was examined with adjustment for age, sex and body mass index (BMI). To explore the potential confounding effect of inflammation in analyses of radiographic OA and vice versa we repeated the analyses including KL grade and GS synovitis in the analyses on inflammation and structural features, respectively. We also explored whether additional adjustment for non-steroidal anti-inflammatory drugs (NSAIDs) altered the associations between inflammation and PPT. Finally, to explore how pain influence the associations, we performed separate analyses for the painful and the non-painful joints. Analyses were done using STATA/SE 15.
RESULTS
Fifteen of 300 persons in the cohort had missing quantitative sensory testing data due to equipment error (n=9), incomplete examination (n=1) and incomplete information on assessed joints (n=5). Hence, 570 joints from 285 participants were included in analyses (Table 1).
Table 1.
Demographic and clinical characteristics of the study population at person level and joint level.
| Study participants included, N=285 | |
|---|---|
| Age, median (IQR) years | 61 (57, 66) |
| Sex, n (%) women | 251 (88) |
| Fulfilment of ACR hand OA criteria, n (%) | 268 (94) |
| Body mass index, mean (SD) | 26 (5) |
| Symptom duration, median (IQR) years | 6 (3, 13) |
| Numeric Rating Scale of hand pain last 24 hours, mean (SD) [0–10] | 3.8 (2.3) |
| Regular use of analgesics, n (%) | |
| Acetaminophen | 11 (4) |
| Oral or topical nonsteroidal anti-inflammatory drugs | 35 (12) |
| Opioids or opioid-like drugs | 5 (2) |
| Antiepileptics, TCA and/or SNRI | 15 (5) |
| KL sum score, median (IQR) [0–128] 1 | 28 (16, 43) |
| Erosive OA, n (%)2 | 101 (35) |
| Grey Scale synovitis sum score, median (IQR) [0–90] 1 | 3 (1, 7) |
| Number of joints with Power Doppler activity grade 1–3, median (IQR) [0–30]1 | 1 (0, 3) |
| Finger joints included3, N=570 | |
| Joints with Kellgren-Lawrence grade n (%) | 290 (51) |
| Erosive joint disease, n (%) | 63 (11) |
| Grey Scale synovitis grade 1–3, n (%) | 147 (26) |
| Power Doppler activity grade 1–3, n (%) | 98 (17) |
IQR; interquartile range, ACR; American College of Rheumatology, OA; osteoarthritis, SD; standard deviation, TCA; tricyclic antidepressants, SNRI; serotonin-norepinephrine reuptake inhibitor.
Including the bilateral distal and proximal interphalangeal, metacarpophalangeal, first carpometacarpal and scaphotrapeziotrapezoidal joints.
Erosive hand OA = at least one interphalangeal joint in the Verbuggen-Veys erosive or remodelling phases.
Two joints per participant; The most painful joint and a random non-painful joint among the 2nd-5th distal and 1st-5th proximal interphalangeal joints.
Radiographic pathology and PPT
Higher KL grade as a continuous variable was statistically significantly associated with lower PPT values (beta = −0.5, 95% CI −0.6, −0.4). Joints with radiographic OA (KL=2, 3 and 4), but not doubtful OA (KL=1), had statistically significantly lower PPTs than joints with no radiographic OA (KL=0) (Table 2). Similarly, the PPT values were statistically significantly lower in erosive vs. non-erosive joints (Table 2). Additional adjustment for GS synovitis led to small reductions in the strength of the estimates, but the associations remained statistically significant (Table 2).
Table 2.
Associations between OA features (radiographic severity and ultrasound-detected inflammation) and PPT in the same finger joints, N=570.
| PPT mean kg/cm2 (SD) | Adjusted beta (95% CI) | Additional adjustment of KL grade/GS synovitis1, beta (95% CI) | |
|---|---|---|---|
| KL | |||
| Grade 0 (n=187, 33%) | 4.9 (2.1) | Ref. | Ref. |
| Grade 1 (n=93, 16%) | 4.7 (2.0) | −0.3 (−0.6, 0.1) | −0.1 (−0.5, 0.2) |
| Grade 2 (n=137, 24%) | 4.7 (2.1) | −0.5 (−0.9, −0.2) | −0.4 (−0.7, 0.0) |
| Grade 3 (n=79, 14%) | 3.5 (1.6) | −1.6 (−2.0, −1.1) | −1.2 (−1.7, −0.7) |
| Grade 4 (n=74, 13%) | 2.9 (1.3) | −2.0 (−2.4, −1.6) | −1.4 (−1.9, −0.9) |
| Erosive disease | |||
| No (n=507, 89%) | 4.6 (2.1) | Ref. | Ref. |
| Yes (n=63, 11%) | 2.9 (1.2) | −1.4 (−1.8, −0.9) | −0.7 (−1.1, −0.2) |
| GS synovitis | |||
| Grade 0 (n=423, 74%) | 4.7 (2.1) | Ref. | Ref. |
| Grade 1 (n=72, 13%) | 3.9 (1.9) | −0.9 (−1.3, −0.5) | −0.3 (−0.7, 0.1) |
| Grade 2 (n=48, 8%) | 3.3 (1.2) | −1.4 (−1.9, −1.0) | −0.9 (−1.4, −0.4) |
| Grade 3 (n=27, 5%) | 2.5 (1.5) | −2.0 (−2.6, −1.4) | −1.2 (−1.8, −0.6) |
| PD activity grade 1–3 | |||
| No (n=472, 82%) | 4.7 (2.1) | Ref. | Ref. |
| Yes (n=98, 17%) | 3.1 (1.4) | −1.5 (−1.8, −1.1) | −0.9 (−1.2, −0.5) |
Mixed-effects multilevel regression of two joints (units) per person (cluster). All analyses are adjusted for age, sex and BMI.
Analyses of KL and erosive OA are adjusted for GS synovitis grade while analyses of GS synovitis and PD activity are adjusted for KL grade. PPT; pressure pain threshold. SD; standard deviation. CI; confidence interval. GS; grey scale. KL; Kellgren-Lawrence grade. PD; power Doppler.
Ultrasound detected inflammation and PPT
Greater severity of GS synovitis grade was associated with lower PPT values (beta = −0.7, 95% CI −0.9, −0.6). Joints with GS synovitis grade 1, 2 and 3 had statistically significantly lower PPT than joints without GS synovitis (GS synovitis grade 0), also after additional adjustment for KL grade (Table 2). Similar associations were found for PD activity (Table 2). Additional adjustment for regular use of NSAIDs did not alter the results (data not shown).
Sensitivity analyses
In separate analyses of the painful finger joints (n=285), the strength of the associations remained similar as in the main analyses (Table 3).
Table 3.
PPT values across structural and inflammatory severity for painful (n=285) and non-painful (n=285) finger joints separately.
| PPT mean kg/cm2 (SD) | Adjusted beta (95% CI) | Additional adjustment of KL grade/GS synovitis1, beta (95% CI) | |
|---|---|---|---|
| Painful finger joints | |||
| KL | |||
| Grade 0 (n=63) | 4.5 (1.9) | Ref. | Ref. |
| Grade 1 (n=41) | 4.3 (1.8) | −0.3 (−1.1, 0.4) | −0.3 (−1.0, 0.4) |
| Grade 2 (n=62) | 4.6 (2.2) | 0.0 (−0.7, 0.7) | 0.0 (−0.7, 0.7) |
| Grade 3 (n=55) | 3.2 (1.6) | −1.4 (−2.1, −0.7) | −1.2 (−1.9, −0.5) |
| Grade 4 (n=64) | 2.8 (1.2) | −1.9 (−2.5, −1.2) | −1.6 (−0.4, −0.9) |
| Continuous scales (grade 0–4) | −0.5 (−0.6, −0.3) | −0.4 (−0.6, −0.2) | |
| Erosive disease | |||
| No (n=231) | 4.1 (2.0) | Ref. | Ref. |
| Yes (n=54) | 2.8 (1.2) | −1.3 (−1.9, −0.7) | −1.0 (−0.6, −0.2) |
| GS synovitis | |||
| Grade 0 (n=159) | 4.3 (2.0) | Ref. | Ref. |
| Grade 1 (n=57) | 3.8 (2.0) | −0.5 (−1.1, 0.1) | 0.0 (−0.5, 0.6) |
| Grade 2 (n=43) | 3.3 (1.2) | −1.0 (−1.6, −0.3) | −0.3 (−0.9, 0.4) |
| Grade 3 (n=26) | 2.5 (1.5) | −1.8 (−2.6, −1.0) | −1.1 (−1.9, −0.3) |
| Continuous scales (grade 0–3) | −0.6 (−0.8, −0.3) | −0.3 (−0.5, −0.0) | |
| PD activity grade 1–3 | |||
| No (n=201) | 4.2 (2.0) | Ref. | Ref. |
| Yes (n=84) | 3.0 (1.3) | −1.3 (−1.8, −0.8) | −0.8 (−1.3, −0.3) |
| Non-painful finger joints | |||
| KL | |||
| Grade 0 (n=124) | 5.1 (2.3) | Ref. | Ref. |
| Grade 1 (n=52) | 5.0 (2.1) | −0.1 (−0.7, 0.6) | −0.0 (−0.7, 0.6) |
| Grade 2 (n=75) | 4.8 (2.0) | −0.3 (−0.9, 0.4) | −1.2 (−0.8, 0.4) |
| Grade 3 (n=24) | 4.2 (1.6) | −1.0 (−1.9, −0.1) | −0.9 (−1.9, 0.0) |
| Grade 4 (n=10) | 3.6 (1.3) | −1.4 (−2.7, −0.0) | −1.3 (−2.8, 0.1) |
| Continuous scales (grade 0–4) | −0.3 (−0.5, −0.1) | −0.2 (−0.5, −0.0) | |
| Erosive OA | |||
| No (n=271) | 5.0 (2.1) | Ref. | Ref. |
| Yes (n=9) | 3.6 (1.4) | −1.2 (−2.6, 0.2) | −1.1 (−2.5, 0.3) |
| GS synovitis | |||
| Grade 0 (n=264) | 5.0 (2.1) | Ref. | Ref. |
| Grade 1 (n=15) | 4.3 (1.4) | −0.7 (−1.8, 0.4) | −0.1 (−1.3, 1.1) |
| Grade 2 (n=5) | 3.4 (1.2) | −1.4 (−3.3, 0.4) | −1.3 (−3.2, 0.6) |
| Grade 3 (n=1) | 3.7 (NA) | −1.1 (−5.2, 2.9) | −1.1 (−5.2, 3.0) |
| Continuous scales (grade 0–3) | −0.6 (−1.3, −0.0) | −0.4 (−1.1, 0.2) | |
| PD activity grade 1–3 | |||
| No (n=271) | 5.0 (2.1) | Ref. | Ref. |
| Yes (n=14) | 3.9 (1.4) | −1.0 (−2.1, 0.2) | −0.5 (−1.7, 0.7) |
Linear regression. All analyses are adjusted for age, sex and BMI.
Analyses of KL and erosive OA are adjusted for GS synovitis grade while analyses of GS synovitis and PD activity are adjusted for KL grade. SD; standard deviation. CI; confidence interval. KL; Kellgren-Lawrence grade. GS; grey scale. NA; not applicable. PD; power Doppler.
In the non-painful joints (n=285), we found a similar trend of lower PPT with increasing KL grade as a continuous variable (adjusted beta = −0.3, 95% CI −0.5, −0.1), presence of erosions (adjusted beta = −1.2, 95% CI −2.6, 0.2), increasing GS synovitis grade as a continuous variable (adjusted beta = −0.6, 95% CI −1.3, −0.0) and presence of PD grade 1–3 (adjusted beta = −1.0, 95% CI −2.1, 0.2). Pathologies were less frequently present in these non-painful joints (Table 3), and fewer of the associations reached statistical significance.
DISCUSSION
Both structural and inflammatory hand OA features, independent of each other, were associated with lower PPT at finger joints, and may represent possible drivers of pain sensitization. We also demonstrated that the relation of greater joint pathology to greater local pain sensitivity was similar in joints with and without pain.
Previous hand OA studies have shown that structural features and inflammatory severity on radiographs, magnetic resonance imaging (MRI) and ultrasound are strongly associated with joint tenderness on palpation[3–5]. Our results are the first to support these findings with a semi-objective quantitative measure of pain sensitization. While the Doyle Index evaluates presence of pain or pain on a 0–3 scale on pressure or passive joint movement[15], PPT will determine the exact threshold at which increasing pressure first feels slightly painful. PPT testing, a recognized measure of pain sensitivity in pain research, is more standardized and nuanced with a scale value and could be more sensitive to change than joint tenderness, though we acknowledge that the potential added value of PPT needs further exploration.
Our results are in line with a small study of 13 hand OA patients, where significant correlations between KL grade and PPT at the same interphalangeal joint were found[11]. Other studies have explored the associations between knee OA pathology and local pain sensitivity. MRI-detected synovitis was associated with lower PPT at the patella and predicted a significant reduction in PPT after 2 years[10]. In contrast to the strong associations we observed between radiographic features and PPT values, several studies on knee OA have not been able to demonstrate such an association between radiographic knee OA and PPT after accounting for potential confounders and pain severity[10]. While the differences in our results between the painful and non-painful joints should be interpreted with caution due to potential issues of precision, the stronger associations with the painful joints may indicate an important role of pain symptoms itself beyond radiographic abnormalities, similar to prior findings at the knee.
By using the PPT testing method we could for the first time demonstrate that even in joints without self-reported pain, radiographic structural severity and ultrasound-detected inflammatory severity was associated with local pain sensitivity. These are new and important findings that may indicate that pain sensitization is an early feature in the pathogenesis of pain. Future longitudinal studies are needed to explore whether low PPT in pain-free joints predict the development of self-reported pain.
A limitation of the study is the cross-sectional design. However, the observed dose-dependent associations and the unlikeliness that pain sensitivity causes joint pathology supports a true relationship. Further, the study population has a wide range of disease severity making it possible to present dose-response data which otherwise could have been difficult to uncover. This study was confined to explore primarily peripheral sensitization through joint level associations. Local PPT was only tested of two finger joints per participant. This was a pragmatic choice. DIP/PIP joints are the joints with highest prevalence of OA, and we considered the selection of the most symptomatic and a non-symptomatic joint to be sufficient to represent the local mechanisms we were examining. Still, it is important to acknowledge that PPT assessed adjacent to a site of pathology also comprises of the individual’s central pain sensitization. Although our study implicates that preventing structural changes and treating inflammation might have clinical consequences, the relation to central sensitization is still unknown. A study examining the relation of hand OA joint pathologies and PPT at distant non-diseased sites, or other quantitative sensory testing modalities of central sensitization (e.g. temporal summation), would add a distinction between the relation to peripheral and central sensitization.
Our results have potential implications for future research and therapeutic approaches. Pain sensitization is a potential treatment target both indirectly and directly. Indirectly, disease modifying drugs targeting structural and inflammatory disease activity could alter pain sensitization and consequently pain. Directly, mechanisms by which pain sensitization occurs are potential treatment targets themselves. Recent years have revealed several promising targets that are mediators of pain sensitization, e.g. nerve growth factor, tropomyosin-related kinase A receptor and ion-channels[1]. So far, only one clinical trial of disease or symptom modifying drugs in hand OA have included characterization of pain sensitization[16]. Future clinical trials could benefit from including quantitative sensory testing of pain sensitization as a predictor of treatment efficacy, a stratification-tool to evaluate subgroup effects or as an inclusion criterion to select the right pain phenotype for the intervention in question.
In summary, this is the first study to demonstrate an independent association of structural and inflammatory hand OA features with lower local PPTs indicating pain sensitization. The associations were similar in joints with and without pain. These results complement preclinical evidence that pain sensitization, especially peripheral, might be driven by structural and inflammatory features. Future research should investigate the role of pain sensitization as a potential target for hand OA pain management or prevention.
Acknowledgements:
The authors would like to thank the study participants in the Nor-Hand study. A thank you is also extended to the project coordinators Elisabeth Mulrooney and Janicke Magnus, as well as Nicolai Ravn Aarskog (medical student) for performing ultrasound examination of the participants.
Funding:
This study was funded by funds from the South-East Norway Regional Health Authority, Pahles foundation, Simon Fougner Hartmanns Family foundation and Trygve Gythfeldt’s research foundation. Dr. Neogi was supported by NIH K24 AR070892, R01 AR062506, and P30 AR072571.
Footnotes
Conflicts of interest: None declared
References
- 1.Miller RE, Block JA, Malfait AM. What is new in pain modification in osteoarthritis? Rheumatology. 2018;57(suppl_4):iv99–iv107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Haugen IK, Englund M, Aliabadi P, et al. Prevalence, incidence and progression of hand osteoarthritis in the general population: the Framingham Osteoarthritis Study. Ann Rheum Dis. 2011;70(9):1581–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Haugen IK, Slatkowsky-Christensen B, Boyesen P, et al. Cross-sectional and longitudinal associations between radiographic features and measures of pain and physical function in hand osteoarthritis. Osteoarthritis Cartilage. 2013;21(9):1191–8. [DOI] [PubMed] [Google Scholar]
- 4.Haugen IK, Slatkowsky Christensen B, Boyesen P, et al. Increasing synovitis and bone marrow lesions are associated with incident joint tenderness in hand osteoarthritis. Ann Rheum Dis. 2016;75(4):702–8. [DOI] [PubMed] [Google Scholar]
- 5.Kortekaas MC, Kwok WY, Reijnierse M, et al. Pain in hand osteoarthritis is associated with inflammation: the value of ultrasound. Ann Rheum Dis. 2010;69(7):1367–9. [DOI] [PubMed] [Google Scholar]
- 6.Arendt-Nielsen L Pain sensitisation in osteoarthritis. Clin Exp Rheumatol. 2017;35(107):S68–S74. [PubMed] [Google Scholar]
- 7.Steen Pettersen P, Neogi T, Magnusson K, et al. Peripheral and Central Sensitization of Pain in Individuals With Hand Osteoarthritis and Associations With Self-Reported Pain Severity. Arthritis Rheumatol. 2019;71(7):1070–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Fu K, Robbins SR, McDougall JJ. Osteoarthritis: the genesis of pain. Rheumatology. 2018;57(suppl_4):iv43–iv50. [DOI] [PubMed] [Google Scholar]
- 9.Suokas A, Walsh D, McWilliams D, et al. Quantitative sensory testing in painful osteoarthritis: a systematic review and meta-analysis. Osteoarthritis Cartilage. 2012;20(10):1075–85. [DOI] [PubMed] [Google Scholar]
- 10.Neogi T Structural correlates of pain in osteoarthritis. Clin Exp Rheumatol. 2017;35 Suppl 107(5):75–8. [PubMed] [Google Scholar]
- 11.Wajed J, Ejindu V, Heron C, et al. Quantitative sensory testing in painful hand osteoarthritis demonstrates features of peripheral sensitisation. Int J Rheumatol. 2012;2012:703138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Gløersen M, Mulrooney E, Mathiessen A, et al. A hospital-based observational cohort study exploring pain and biomarkers in patients with hand osteoarthritis in Norway: The Nor-Hand protocol. BMJ open. 2017;7(9):e016938. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Verbruggen G, Veys EM. Erosive and non-erosive hand osteoarthritis. Use and limitations of two scoring systems. Osteoarthritis Cartilage. 2000;8 Suppl A:S45–54. [DOI] [PubMed] [Google Scholar]
- 14.Keen HI, Lavie F, Wakefield RJ, et al. The development of a preliminary ultrasonographic scoring system for features of hand osteoarthritis. Ann Rheum Dis. 2008;67(5):651–5. [DOI] [PubMed] [Google Scholar]
- 15.Bijsterbosch J, Wassenaar M, le Cessie S, et al. Doyle Index is a valuable additional pain measure in osteoarthritis. Osteoarthritis Cartilage. 2010;18(8):1046–50. [DOI] [PubMed] [Google Scholar]
- 16.Sofat N, Harrison A, Russell MD, et al. The effect of pregabalin or duloxetine on arthritis pain: a clinical and mechanistic study in people with hand osteoarthritis. J Pain Res. 2017;10:2437–49. [DOI] [PMC free article] [PubMed] [Google Scholar]