Abstract
Objective:
One of the less understood adverse effects while taking opioids is the paradoxical increase in pain, known as opioid-induced hyperalgesia (OIH). We sought to determine whether pain sensitization mediates the relation of opioid use to pain severity in people with knee OA.
Methods:
We included participants in a NIH-funded cohort study of people with or at risk of knee OA. Participants were categorized into opioid and non-opioid analgesic groups at baseline. WOMAC pain 2 years later was assessed as the outcome. We used causal mediation analysis to assess the mediating role of pain sensitization, quantified by changes in pressure pain threshold (PPT) at the wrist and patella over 2 years, on the effect of opioid use on WOMAC pain 2 years later.
Results:
We included 296 opioid users and 1070 non-opioid analgesic users. Compared with non-opioid analgesic use, opioid use was associated with greater pain 2 years later. This relation was mediated by 0.05 and 0.08 unit changes in wrist PPT (95%CI: 0.01–0.10) and patellar PPT (95%CI 0.02–0.14), respectively. When we assessed any worsening in WOMAC pain score over 2 years, opioid use, compared with non-opioid analgesic use, had a 2% and 5% higher odds of experiencing any worsening pain mediated by changes in wrist PPT (95%CI, 0.99–1.04) and patellar PPT (95%CI, 1.01–1.09), respectively.
Conclusion:
Pain sensitization had small mediating effects on the paradoxical phenomenon of OIH, suggesting that pain sensitization may not play a major role and/or that PPT is an inadequate tool to assess OIH.
Introduction:
Knee osteoarthritis (OA), with its primary symptom of pain, causes significant disability and reduced quality of life(1). Although opioids are not recommended in treatment guidelines for the management of OA due to its limited long-term efficacy and substantial adverse effects(2–5), opioids are used in 40% of patients with knee OA(6, 7). Furthermore, some patients require opioid analgesics due to lack of other effective options. When opioids are prescribed, clinicians need to monitor for numerous potential adverse effects. One of the lesser understood adverse effects of opioids in knee OA is opioid-induced hyperalgesia (OIH)(8, 9), which is a seemingly paradoxical phenomenon of an increase in pain despite using opioids for their analgesic properties.
The presence of OIH can lead to clinicians increasing the dose of the opioid on the assumption that the dose was insufficient to manage the degree of pain. OIH was originally described in rats which were noted to have increased pain sensitivity after administration of opioids(10, 11). OIH is different from opioid tolerance and is characterized by a set of clinical symptoms: 1) an increase in pain severity, 2) an increase in pain sensitivity to external stimuli, and 3) spreading the increased pain sensitivity to other body locations(12, 13). In contrast, opioid tolerance is defined as an increased dose needed to achieve analgesia(12).
While what contributes to OIH is not entirely clear, a potential underlying mechanism of OIH is alterations in nociceptive signaling, such as pain sensitization. Pain sensitization is induced by neuroplastic changes with altered nociceptive signaling in the ascending pathways in the periphery (i.e., peripheral sensitization) as well as in the central nervous system (i.e., central sensitization), and causes an increase in pain sensitivity, widespread hyperalgesia, and allodynia(14). Because clinical symptoms of pain sensitization and OIH are similar, pain sensitization may contribute to OIH and account for this paradoxical experience. Quantitative sensory testing (QST), a clinical measurement of pain sensitization, has been examined in a few studies of OIH(13). Two studies demonstrated an increase in heat or cold pain sensitivity after one month of opioid use in individuals with chronic low back pain(15) and neuropathic radicular pain(16). A few cross-sectional studies demonstrated that opioid use was associated with pain sensitivity measured by QST and pain severity in individuals with chronic spinal pain(17) and chronic noncancer pain(18, 19).
Whether pain sensitization mediates the development of OIH (i.e., increased pain due to the use of opioids) in knee OA is not known. Such insights would be helpful to guide whether measures of pain sensitization may be useful as means to gauge development of OIH. Hence, we sought to longitudinally examine the potential mediating effects of pain sensitization on the relation of opioid use to pain severity in people with knee OA.
Materials and Methods
Study Sample
The Multicenter Osteoarthritis (MOST) Study is a National Institutes of Health-funded longitudinal cohort study of 3026 older adults aged between 50–79 years with or at risk of knee OA. Details of the study protocol have been published elsewhere(20, 21). The study protocol was approved by the institutional review boards at the University of Iowa, University of Alabama, Birmingham, University of California, San Francisco and Boston University Medical Campus.
We included data for participants from the 60- and 84-month study visits in MOST; the 60-month study visit was the first visit at which QST measures (described below) were obtained and is considered baseline for these analyses. To minimize confounding by indication, we included only participants who reported taking a form of pain medication (described below) for inclusion in our study.
Exposures
Pain medication use
Participants were categorized according to their medication use at baseline. Medication data were collected based on medication prescriptions that participants brought to the study visit that they had taken in the prior 30 days, and were coded according to the Iowa Drug Information Service (IDIS) ingredient code and name. An additional medication questionnaire queried participants about opioid and non-opioid medications they had taken within the 30 days prior to the study visit.
We identified participants as having taken opioids or non-opioid analgesics (e.g., nonsteroidal anti-inflammatory drug, Cox-2 inhibitors, salicylates, acetaminophen) based on these data sources (see Supplementary Table 1 for the list of the medications). Participants were considered to be opioid users if they had opioid prescriptions (based upon pill bottles brought into the study visit) or self-reported opioid use in the 30 days prior to baseline, regardless of dosage or the use of any non-opioid medications. The comparator group included participants who reported using only non-opioid analgesic medications.
Outcomes
Pain
The Western Ontario McMaster Universities Osteoarthritis Index (WOMAC) pain subscale score (with a range between 0 to 20) was used to measure knee-specific pain severity over the prior 30 days at baseline and follow-up 2 years later. The WOMAC pain subscale consists of 5 pain-related questions with a 0–4 Likert scale (none (0), mild-moderate (1–2) and severe-extreme (3–4)(22). Higher WOMAC scores reflect greater knee pain severity.
Mediator
Pressure Pain Threshold (PPT):
PPT is a reliable and common QST measurement to assess hyperalgesia using mechanical stimuli in knee OA(23, 24). We measured PPT at the right distal radioulnar joint (the wrist) and patella(23, 24). PPT at a peripheral site (e.g., knee) is thought to reflect peripheral with or without central sensitization, while PPT at a distant non-diseased body site (e.g., wrist) is thought to reflect central sensitization(23–26). PPT was assessed using a handheld pressure algometer (1 cm2 rubber tip; Wagner FDIX25) applied at a constant rate of 0.5 kg/second to the anatomic site being tested. PPT was defined as the point at which the participant verbally indicated that the pressure first changed to slight pain. Three trials were collected at each anatomic site and averaged. Lower PPT indicates greater pain sensitivity. PPT has excellent reliability with intraclass correlation coefficients of 0.85–0.90(24).
Statistical Analysis
We limited our analyses to knees that had pain at baseline (WOMAC pain ≥ 1/20) and used causal mediation analysis to quantify the extent to which the effect of opioid use on knee pain severity 2 years later may be mediated by changes in pain sensitization measured by a two-year change in PPT. We fit natural-effects models which allow for decomposition of the opioid effect into an indirect (mediated) effect (due to pain sensitization) and a remaining direct effect, not mediated by pain sensitization, on knee pain severity (Figure 1).
Figure 1:
Directed Acyclic Graph Representing Indirect and Direct Effects
QST: Quantitative sensory testing, changes in or baseline PPT at the wrist and patellae
WOMAC: Western Ontario McMaster Universities Osteoarthritis Index
To estimate the parameters of the fitted natural-effects models, we used an imputation-based approach developed by Vansteelandt et al.(27, 28). We examined the degree to which potential mediation by change in pain sensitization affected WOMAC knee pain 2 years later (indirect effect) as well as the direct effect of opioid use on the outcome using knee-based linear regression. We repeated analyses to assess any pain worsening over 2 years, defined as WOMAC pain 2 years later – WOMAC pain at baseline ≥ 1, with logistic regression.
In a sensitivity analysis, we included a no-pain medication group as the referent group to explore the extent to which the effects of pain medication use (opioid use and non-opioid pain medication use) on pain severity is mediated by pain sensitization as compared with no pain medication use. We also considered other medications related to pain (e.g., antidepressants, anxiolytics, muscle relaxants, gabapentin) as a non-opioid analgesic use group to more inclusively examine the relationships. Thus, in this sensitivity analysis, we had three exposure levels in the causal mediation analyses: opioid use, non-opioid pain medication use (including additional medications that could potentially be used for pain), and no-pain medication group.
All analyses were adjusted for the following potential confounders: age (continuous variable), sex (female or male), BMI (continuous variable), race (nominal variable as a proxy for numerous differential structural systemic societal impacts), clinic site (data collection sites of MOST study in either Iowa or Alabama), depressive symptoms (defined as ≥16 on Center for Epidemiologic Studies Depression Scale (29)), pain catastrophizing (defined as a score ≥ 1 on a 0–6 Likert scale from the Coping Strategies Questionnaire (30)), poor sleep quality (defined as very bad or fairly bad sleep in the last 7 days from a Likert scale sleep question), widespread pain (defined as pain present above and below the waist, on the right and left sides of the body, and axial pain using a body homunculus (32)), Kellgren-Lawrence grade (categorical variable), history of knee injury, and WOMAC knee pain severity prior to the baseline clinic visit (continuous variable). Statistical analysis was conducted using R version 3.6.2..
Results
There were 1366 participants (2386 knees) eligible for this study. Of those, 78.3% (1070) participants reported using a pain medication other than opioids in the prior 30 days (mean age 62.1, 69% female, mean BMI 31.2), while 21.7% (296) participants reported using opioids (mean age 61.5, 72% female, mean BMI 32.7). Baseline WOMAC knee pain in the non-opioid analgesic group was 4.10, compared to 5.76 in the opioid group (Table 1). Wrist PPT and patellar PPT at baseline were 3.23 and 4.57 kgf in the non-opioid analgesic group and 3.12 and 4.25 kgf in the opioid group, respectively.
Table 1:
Baseline Participant Characteristics
| Non-opioid users | Opioid users | |
|---|---|---|
| (n=1070) | (n=296) | |
| Mean (SD) | Mean (SD) | |
|
| ||
| Age | 62.1 (7.84) | 61.5 (7.84) |
|
| ||
| Female, n (%) | 715 (69%) | 213 (72%) |
|
| ||
| Body mass index, kg/m2 | 31.2 (6.17) | 32.7 (6.99) |
|
| ||
| WOMAC knee pain (0–20) | 4.10 (3.61) | 5.76 (4.41) |
|
| ||
| PPT wrist (Kgf) | 3.23 (1.39) | 3.12 (1.42) |
|
| ||
| PPT patella (Kgf) | 4.57 (2.02) | 4.25 (2.13) |
|
| ||
| Caucasian, n (%) | 717 (67) | 177 (60) |
|
| ||
| Clinic site: | ||
| Iowa | 539 (50) | 95 (32) |
| Alabama | 531 (50) | 201 (68) |
|
| ||
| Depressive symptoms, n (%) | 130 (12) | 89 (30) |
|
| ||
| Pain catastrophizing, n (%) | 578 (54) | 225 (76) |
|
| ||
| Poor sleep, n (%) | 854 (80) | 200 (68) |
|
| ||
| Widespread pain, n (%) | 575 (54) | 218 (74) |
|
| ||
| Kellgren-Lawrence Grades: | ||
| Grade 0, n (%) | 281 (26) | 84 (28) |
| Grade I, n (%) | 140 (13) | 39 (13) |
| Grade II, n (%) | 237 (22) | 67 (23) |
| Grade III, n (%) | 272 (25) | 63 (22) |
| Grade IV, n (%) | 140 (13) | 43 (14) |
|
| ||
| History of knee injury, n (%) | 321 (30) | 92 (31) |
WOMAC: Western Ontario McMaster Universities Osteoarthritis Index, PPT: Pressure Pain Threshold, Depressive symptoms: defined as ≥16 on Center for Epidemiologic Studies Depression Scale; Pain catastrophizing: defined as a score ≥ 1 on a 0–6 Likert scale from the Coping Strategies Questionnaire; Poor sleep: defined as very bad or fairly bad sleep in the last 7 days from a Likert scale sleep question; Widespread pain: defined as pain being present above and below the waist, on the right and left sides of the body, and axial pain using a body homunculus
Direct Effect of Opioid Use on Pain 2 years later
The estimate for the natural direct effect of opioid use on knee pain not mediated through pain sensitization suggested that baseline opioid use was associated with greater pain severity 2 years later, compared with non-opioid users (β=0.51, 95%: CI 0.07, 0.96, and β=0.49, 95%: CI 0.03, 0.95, depending on the mediator included in the model) (Table 2). For any worsening in WOMAC pain score over 2 years, estimates of natural direct effects suggested no association (OR=1.00, 95% CI: 0.75, 1.33 and OR=0.98, 95% CI: 0.74, 1.31).
Table 2:
Indirect (Mediating) and direct effects of opioid use on WOMAC knee pain severity compared to those with non-opioid use
| Exposure | Mediator | Outcome Definition (2 years later) | Estimate or OR (95%CI) | |
|---|---|---|---|---|
| Natural Mediating Effect | Natural Direct Effect | |||
| Medication Group at Baseline: Opioid Use (n=296) vs. Non-Opioid analgesic Use (n=1070) (REF) |
Change in wrist PPT | Pain severity, Beta Coef. | 0.05 (0.01, 0.10)* | 0.51 (0.07, 0.96)* |
| Change in patellar PPT | Pain severity, Beta Coef. | 0.08 (0.02, 0.14)* | 0.49 (0.03, 0.95)* | |
| Change in wrist PPT | Any worsening in pain, OR | 1.02 (0.99, 1.04) | 1.00 (0.75, 1.33) | |
| Change in patellar PPT | Any worsening in pain, OR | 1.05 (1.01, 1.09)* | 0.98 (0.74, 1.31) | |
Non-opioids analgesic use was served as reference group, Change in PPT: PPT 2 years later subtracted from PPT at baseline, OR: Odds Ratio, PPT: Pressure Pain Threshold
significant result
Coef.: Coefficient
All analyses were adjusted for the following baseline characteristics: age, sex, BMI, race, clinic site, depressive symptoms, pain catastrophizing, poor sleep, widespread pain, Kellgren-Lawrence grade, knee injury, and WOMAC knee pain severity.
Effects Mediated by Pain Sensitization
The estimates of the natural indirect (mediating) effects suggested that the use of opioids increased WOMAC knee pain 2 years later by 0.05 and 0.08 units which were mediated by changes in PPT over 2 years at the wrist (β=0.05, 95% CI: 0.01, 0.10) and patella (β=0.08, 95%: CI 0.02, 0.14), respectively (Table 2). When we assessed any worsening in WOMAC pain score over 2 years, the natural indirect effect estimate suggested an increased odds of knee pain by 2% and 5% in opioid users that was mediated by changes in PPT over 2 years at the wrist (OR=1.02, 95% CI: 0.99, 1.04) and patella (OR =1.05, 95% CI: 1.01, 1.09), respectively.
Sensitivity Analyses
For our sensitivity analyses, we identified 313 who reported no pain medication use, 1141 non-opioid pain medication users (including those with medications related indirectly to pain such as antidepressants, anxiolytics, muscle relaxants and gabapentin), and 296 opioid users. We found that, when compared with no pain medication use, both opioid use and non-opioid use at baseline were associated with greater WOMAC pain 2 years later. The magnitude of the direct effect of opioid use was larger than non-opioid use across mediators; however, the effects of opioid use on knee pain severity 2 years later were not mediated by pain sensitization (i.e., no mediating effects) in the sensitivity analysis (Supplementary Table 2)
Discussion
We used data from a large cohort of people with knee OA to determine whether pain sensitization mediated the association between opioid use and knee pain severity over 2 years. We found that opioid use was associated with greater pain severity 2 years later as compared to non-opioid users, and changes in PPT at the wrist and patella had mediating effects, though the mediating effect sizes were small and of unclear clinical significance. These findings suggest that pain sensitization, at least as measured by PPT, might not be a major mechanism of OIH nor a suitable means by which to screen for the development of OIH.
Pain sensitization has been considered to be a relevant mechanism for OIH development because of common neurobiological alterations. For example, opioids increase N-methyl-D-aspartate (NMDA) activity(12, 31), and elevated NMDA receptor activity is a major neurobiological alteration in pain sensitization(14, 32). Thus, OIH has been thought to be related to an increase in pain sensation(13). Nonetheless, reports of QST measures of pain sensitization in OIH have been inconclusive. In one study, those with chronic pain and opioid use had low heat pain thresholds (i.e., increased pain response to heat stimuli), but not increased pain response to cold stimuli nor increased cold and warm pain sensation(33). In contrast, other studies with opioid users or opioid-dependent participants demonstrated increased cold pain sensitivity, but not increased mechanical pain sensitivity measured by PPT(17, 34). A systematic review and meta-analysis concluded that differences in study findings might be related to the types of QST procedures and stimulus modalities(35). We focused on PPT as a mechanical stimulus since OA is largely a biomechanically-mediated disorder(1). Our results suggest that pain sensitization as measured by PPT does not play a major role in OIH and is not a suitable means for assessing those at potential risk for OIH.
Some evidence has suggested that biopsychosocial symptoms may contribute to developing OIH. For example, negative affect (e.g. depression, anxiety, pain catastrophizing) has been associated with an increase in the risk of developing OIH(17, 36, 37). Biologically, negative affect can impair opioid-mediated pain inhibition in the central nervous system(38) because of functional connectivities between brain regions responsible for those biopsychosocial symptoms and the midbrain that controls opioid-mediated pain inhibition(39). Therefore, evaluating biopsychosocial symptoms may provide a better understanding of OIH.
A strength of our study was the use of data from a large longitudinal cohort. To our knowledge, this is the first examination of whether pain sensitization explains the association between opioid use and pain severity longitudinally in knee OA. Another strength of our study was to implement of a rigorous causal inference-based approach to mediation analysis. We also have several limitations to acknowledge. We did not have data on duration and dose of opioid use, which might have provided better insights, although prior studies have reported the presence of OIH regardless of duration and dose(37, 40). Similarly, we did not have data on the initiation of opioid use, which might have provided insights into the mediating effects of new opioid users vs. experienced opioid users. We cannot rule out confounding by indication as a potential contributor to worse pain over time with opioid use. The time course over which OIH may occur may not have been adequately captured by our assessments that occurred at a two-year interval. Further, we did not incorporate alterations in descending pain modulation. Although pain sensitization was a rational target to investigate mechanisms of OIH as explained above, several prior studies suggested opioid users have more inefficient descending pain modulation than non-opioid users(41, 42).
In summary, compared to non-opioid users, opioid users reported greater pain severity 2 years later. We found a mediating effect of pain sensitization on developing OIH, though with only small effects of unclear clinical significance. Thus, pain sensitization is unlikely a major underlying mechanism of the paradoxical development of pain with opioids. Other mechanisms such as psychological factors such as depression, anxiety and pain catastrophizing may be additional avenues to be explored to better understand OIH to aid identification and management of OIH in those patients with knee OA for whom opioids are necessary as a pain management option.
Supplementary Material
Supplementary Table 1: List of opioid and non-opioid pain medications in MOST study dataset
Supplementary Table 2: Indirect (Mediating) and direct effects of opioid use and non-opioid analgesic use on WOMAC knee pain severity compared with no pain medication use
Significance and Innovation.
While 40% of people with knee OA use opioids, some people paradoxically experience an increase in pain, known as opioid-induced hyperalgesia (OIH). The underlying mechanisms for OIH in knee OA is unclear.
Pain sensitization may be a potential mechanism of OIH that may be related to common biological alterations such as enhanced NMDA receptor activity.
We found small mediating effects of pain sensitization as assessed with pressure pain threshold on the relation of opioid use to greater pain 2 years later.
Our findings suggest that pain sensitization is not a major contributor to OIH.
Acknowledgments
The authors acknowledge the staff at the UCSF Coordinating Center and all MOST participants.
Role of the funding source
The Multicenter Osteoarthritis Study was funded by the NIH (U01-AG18820, U01-AG18832, U01-AG18947, U01-AG19069 and AR47785). TN was supported by NIH/NIAMS P30 AR072571, K24 AR070892, R01 AG066010. SRJ was supported by grants from NIH/NIA (award number: R03AG060272) and NIH/NIAMS (award numbers: R21AR074578 and P30AR072571). Funding sources had no role in study design, data collection and analysis, data interpretation, or the decision to submit the manuscript for publication.
Footnotes
Conflict of Interest: None
References
- 1.Neogi T The epidemiology and impact of pain in osteoarthritis. Osteoarthritis and cartilage. 2013;21(9):1145–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Kolasinski SL, Neogi T, Hochberg MC, Oatis C, Guyatt G, Block J, et al. 2019 American College of Rheumatology/Arthritis Foundation Guideline for the Management of Osteoarthritis of the Hand, Hip, and Knee. Arthritis & Rheumatology. 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bannuru RR, Osani MC, Vaysbrot EE, Arden NK, Bennell K, Bierma-Zeinstra SMA, et al. OARSI guidelines for the non-surgical management of knee, hip, and polyarticular osteoarthritis. Osteoarthritis and cartilage. 2019;27(11):1578–89. [DOI] [PubMed] [Google Scholar]
- 4.Dowell D, Haegerich TM, Chou R. CDC guideline for prescribing opioids for chronic pain—United States, 2016. Jama. 2016;315(15):1624–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Sites BD, Beach ML, Davis MA. Increases in the use of prescription opioid analgesics and the lack of improvement in disability metrics among users. Regional Anesthesia & Pain Medicine. 2014;39(1):6–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wright EA, Katz JN, Abrams S, Solomon DH, Losina E. Trends in prescription of opioids from 2003–2009 in persons with knee osteoarthritis. Arthritis care & research. 2014;66(10):1489–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Bedard NA, Pugely AJ, Westermann RW, Duchman KR, Glass NA, Callaghan JJ. Opioid use after total knee arthroplasty: trends and risk factors for prolonged use. The Journal of arthroplasty. 2017;32(8):2390–4. [DOI] [PubMed] [Google Scholar]
- 8.Roeckel L-A, Le Coz G-M, Gaveriaux-Ruff C, Simonin F. Opioid-induced hyperalgesia: cellular and molecular mechanisms. Neuroscience. 2016;338:160–82. [DOI] [PubMed] [Google Scholar]
- 9.Mitra S. Opioid-induced hyperalgesia: pathophysiology and clinical implications. J Opioid Manag. 2008;4(3):123–30. [DOI] [PubMed] [Google Scholar]
- 10.Van Elstraete AC, Sitbon P, Trabold F, Mazoit J-X, Benhamou D. A single dose of intrathecal morphine in rats induces long-lasting hyperalgesia: the protective effect of prior administration of ketamine. Anesthesia & Analgesia. 2005;101(6):1750–6. [DOI] [PubMed] [Google Scholar]
- 11.Célèrier E, Rivat C, Jun Y, Laulin J-P, Larcher A, Reynier P, et al. Long-lasting hyperalgesia induced by fentanyl in rats preventive effect of ketamine. Anesthesiology: The Journal of the American Society of Anesthesiologists. 2000;92(2):465-. [DOI] [PubMed] [Google Scholar]
- 12.Colvin LA, Bull F, Hales TG. Perioperative opioid analgesia—when is enough too much? A review of opioid-induced tolerance and hyperalgesia. The Lancet. 2019;393(10180):1558–68. [DOI] [PubMed] [Google Scholar]
- 13.Katz NP, Paillard FC, Edwards RR. Review of the performance of quantitative sensory testing methods to detect hyperalgesia in chronic pain patients on long-term opioids. Anesthesiology. 2015;122(3):677–85. [DOI] [PubMed] [Google Scholar]
- 14.Latremoliere A, Woolf CJ. Central sensitization: a generator of pain hypersensitivity by central neural plasticity. J Pain. 2009;10(9):895–926. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Chu LF, D’Arcy N, Brady C, Zamora AK, Young CA, Kim JE, et al. Analgesic tolerance without demonstrable opioid-induced hyperalgesia: a double-blinded, randomized, placebo-controlled trial of sustained-release morphine for treatment of chronic nonradicular low-back pain. Pain®. 2012;153(8):1583–92. [DOI] [PubMed] [Google Scholar]
- 16.Elon Eisenberg M. A negative correlation between hyperalgesia and analgesia in patients with chronic radicular pain: is hydromorphone therapy a double-edged sword? Pain physician. 2013;16:65–76. [PubMed] [Google Scholar]
- 17.Edwards RR, Wasan AD, Michna E, Greenbaum S, Ross E, Jamison RN. Elevated pain sensitivity in chronic pain patients at risk for opioid misuse. The journal of pain : official journal of the American Pain Society. 2011;12(9):953–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Hooten WM, Sandroni P, Mantilla CB, Townsend CO. Associations between heat pain perception and pain severity among patients with chronic pain. Pain medicine (Malden, Mass). 2010;11(10):1554–63. [DOI] [PubMed] [Google Scholar]
- 19.Wang H, Fischer C, Chen G, Weinsheimer N, Gantz S, Schiltenwolf M. Does long-term opioid therapy reduce pain sensitivity of patients with chronic low back pain? Evidence from quantitative sensory testing. Pain physician. 2012;15(3 Suppl):ES135–43. [PubMed] [Google Scholar]
- 20.Felson DT, Niu J, Guermazi A, Roemer F, Aliabadi P, Clancy M, et al. Correlation of the development of knee pain with enlarging bone marrow lesions on magnetic resonance imaging. Arthritis and rheumatism. 2007;56(9):2986–92. [DOI] [PubMed] [Google Scholar]
- 21.Englund M, Felson DT, Guermazi A, Roemer FW, Wang K, Crema MD, et al. Risk factors for medial meniscal pathology on knee MRI in older US adults: a multicentre prospective cohort study. Annals of the rheumatic diseases. 2011;70(10):1733–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. The Journal of rheumatology. 1988;15(12):1833–40. [PubMed] [Google Scholar]
- 23.Neogi T, Guermazi A, Roemer F, Nevitt MC, Scholz J, Arendt‐Nielsen L, et al. Association of joint inflammation with pain sensitization in knee osteoarthritis: the Multicenter Osteoarthritis Study. Arthritis & rheumatology. 2016;68(3):654–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Neogi T, Frey-Law L, Scholz J, Niu J, Arendt-Nielsen L, Woolf C, et al. Sensitivity and sensitisation in relation to pain severity in knee osteoarthritis: Trait or state? Ann Rheum Dis. 2015;74(4):682–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Graven‐Nielsen T, Wodehouse T, Langford R, Arendt‐Nielsen L, Kidd B. Normalization of widespread hyperesthesia and facilitated spatial summation of deep‐tissue pain in knee osteoarthritis patients after knee replacement. Arthritis & Rheumatism. 2012;64(9):2907–16. [DOI] [PubMed] [Google Scholar]
- 26.Arendt-Nielsen L, Egsgaard LL, Petersen KK, Eskehave TN, Graven-Nielsen T, Hoeck HC, et al. A mechanism-based pain sensitivity index to characterize knee osteoarthritis patients with different disease stages and pain levels. European journal of pain (London, England). 2015;19(10):1406–17. [DOI] [PubMed] [Google Scholar]
- 27.Vansteelandt S, Bekaert M, Lange T. Imputation strategies for the estimation of natural direct and indirect effects. Epidemiologic Methods. 2012;1(1):131–58. [Google Scholar]
- 28.Loeys T, Moerkerke B, De Smet O, Buysse A, Steen J, Vansteelandt S. Flexible Mediation Analysis in the Presence of Nonlinear Relations: Beyond the Mediation Formula. Multivariate Behav Res. 2013;48(6):871–94. [DOI] [PubMed] [Google Scholar]
- 29.Radloff LS. The CES-D scale: A self-report depression scale for research in the general population. Applied psychological measurement. 1977;1(3):385–401. [Google Scholar]
- 30.Jensen MP, Keefe FJ, Lefebvre JC, Romano JM, Turner JA. One-and two-item measures of pain beliefs and coping strategies. Pain. 2003;104(3):453–69. [DOI] [PubMed] [Google Scholar]
- 31.Raffa RB, Pergolizzi JV Jr. Multi-mechanistic analgesia for opioid-induced hyperalgesia. Journal of clinical pharmacy and therapeutics. 2012;37(2):125–7. [DOI] [PubMed] [Google Scholar]
- 32.Arendt-Nielsen L, Morlion B, Perrot S, Dahan A, Dickenson A, Kress HG, et al. Assessment and manifestation of central sensitisation across different chronic pain conditions. European journal of pain (London, England). 2018;22(2):216–41. [DOI] [PubMed] [Google Scholar]
- 33.Chen L, Malarick C, Seefeld L, Wang S, Houghton M, Mao J. Altered quantitative sensory testing outcome in subjects with opioid therapy. Pain. 2009;143(1–2):65–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Krishnan S, Salter A, Sullivan T, Gentgall M, White J, Rolan P. Comparison of pain models to detect opioid-induced hyperalgesia. Journal of pain research. 2012;5:99. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Higgins C, Smith B, Matthews K. Evidence of opioid-induced hyperalgesia in clinical populations after chronic opioid exposure: a systematic review and meta-analysis. British journal of anaesthesia. 2019;122(6):e114–e26. [DOI] [PubMed] [Google Scholar]
- 36.Jamison RN, Edwards RR, Liu X, Ross EL, Michna E, Warnick M, et al. Relationship of negative affect and outcome of an opioid therapy trial among low back pain patients. Pain practice : the official journal of World Institute of Pain. 2013;13(3):173–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Wasserman RA, Brummett CM, Goesling J, Tsodikov A, Hassett AL. Characteristics of chronic pain patients who take opioids and persistently report high pain intensity. Regional anesthesia and pain medicine. 2014;39(1):13–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Edwards RR, Cahalan C, Mensing G, Smith M, Haythornthwaite JA. Pain, catastrophizing, and depression in the rheumatic diseases. Nature reviews Rheumatology. 2011;7(4):216–24. [DOI] [PubMed] [Google Scholar]
- 39.Cagnie B, Coppieters I, Denecker S, Six J, Danneels L, Meeus M. Central sensitization in fibromyalgia? A systematic review on structural and functional brain MRI. Seminars in arthritis and rheumatism. 2014;44(1):68–75. [DOI] [PubMed] [Google Scholar]
- 40.de Hoogd S, Valkenburg AJ, van Dongen EPA, Daeter EJ, van Rosmalen J, Dahan A, et al. Short- and long-term impact of remifentanil on thermal detection and pain thresholds after cardiac surgery: A randomised controlled trial. Eur J Anaesthesiol. 2019;36(1):32–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Ram KC, Eisenberg E, Haddad M, Pud D. Oral opioid use alters DNIC but not cold pain perception in patients with chronic pain - new perspective of opioid-induced hyperalgesia. Pain. 2008;139(2):431–8. [DOI] [PubMed] [Google Scholar]
- 42.Martel MO, Petersen K, Cornelius M, Arendt-Nielsen L, Edwards R. Endogenous Pain Modulation Profiles Among Individuals With Chronic Pain: Relation to Opioid Use. The journal of pain : official journal of the American Pain Society. 2019;20(4):462–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Table 1: List of opioid and non-opioid pain medications in MOST study dataset
Supplementary Table 2: Indirect (Mediating) and direct effects of opioid use and non-opioid analgesic use on WOMAC knee pain severity compared with no pain medication use