Background/Purpose: There is increasing support in the literature for a neuropathic component to OA pain. Our primary objective was to determine whether pain at rest and pain on activity are differentially associated with neuropathic pain scores in individuals with end-stage hip and knee OA. In addition, we examined the associations of neuropathic pain scores with psychological factors, health status and sociodemographic characteristics. As there is evidence of sex specific effects on pain, we analysed men and women separately.

Methods: Study participants were 843 patients with hip or knee OA scheduled for total joint arthroplasty (TJA) in Toronto, Canada. In pre-surgery questionnaires, data were collected using the painDETECT scale for neuropathic pain, the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain subscale (items split into pain at rest (sitting, night pain) and pain on activity (standing, up/down stairs, walking on flat ground) and summed), the Hospital Anxiety and Depression Scale (HADS; depression only) and the Pain Catastrophizing Scale (PCS). Data on socio-demographics, health status (body mass index (BMI), comorbidity count, possible neuropathic comorbidities, symptomatic joint count) and medication use were also collected. Multivariable linear regression models were estimated for men and women to examine associations with neuropathic pain scores (higher indicates greater likelihood of neuropathic pain) by entering variables in predetermined blocks.

Results: Participants were of mean age 65.1 years; 57.1% were women. 55.3% of men and 57.6% of women were scheduled for knee TJA. Mean painDETECT scores were significantly higher (p=0.0001) for women (11.2±6.6 out of 38) than men (9.3±7.0), with 35.6% of women and 27.7% of men meeting established cut-offs for possible (painDETECT scores: 13-18) or likely (painDETECT scores: 19+) neuropathic pain. In regression modeling, pain at rest explained the most variance in neuropathic pain scores (men: 23.4%; women: 13.9%). In the final regression model for women, surgical joint (knee>hip; p=0.04), joint count, pain at rest, pain on activity, depression and pain catastrophizing were significantly associated with neuropathic pain score. In the final model for men, pain at rest, depression, pain catastrophizing and use of narcotics were significantly associated with neuropathic pain score. For women, the coefficient for pain at rest was 1.6 times greater than that for pain on activity. For men, pain on activity was not significantly associated (p=0.47) with neuropathic pain score.

Conclusion: Findings support that likely neuropathic pain is experienced by a notable proportion of patients with end-stage hip and knee OA. A greater neuropathic pain score was more strongly associated with pain at rest than pain on activity, particularly in men. Clinically, the presentation of pain at rest may suggest that a more thorough evaluation for potential neuropathic pain is warranted and may have implications for appropriate pain management. Findings also suggest that use of the WOMAC pain subscale as a single measure of pain burden (i.e. combined pain at rest and pain on activity) may need to be revisited.

« Back to 2017 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/neuropathic-pain-in-end-stage-hip-and-knee-osteoarthritis-differences-between-men-and-women-and-differential-associations-with-pain-at-rest-and-pain-on-activity/