PREVALENCE OF POSTOPERATIVE CHRONIC PAIN AFTER TOTAL HIP OR KNEE ARTHROPLASTY

Abstract

<p><strong>Background: </strong>Total joint arthroplasty (TJA) is considered the treatment of choice to alleviate pain and improve function of patients with osteoarthritis. However,recent evidence suggests that a significant proportion of patients continue to report pain, or worsening of their symptoms well after their joint replacement. We call this chronic pain “<em>phantom joint pain</em>” as it persists despite the fact hat the affected joint has been replaced.</p> <p>Chronic pain of neuropathic origin may be a consequence of surgery or in patients with osteoarthritis (OA); there may be a combination of nociceptive and neuropathic pain (NP) mechanisms. As there are no definitive physiological indicators for NP or gold standards for diagnosis, Guidelines on Neuropathic Pain Assessment advocate the use of screening tools to evaluate the patient’s pain experiences and potentially characterize various pain features.</p> <p>Despite suggestions that phantom joint pain post TJA is a common problem there is limited information about its prevalence among Canadians. To date there are no studies that have characterized neuropathic vs. non- neuropathic chronic pain features in a TJA population.</p> <p><strong>Purpose: </strong>The purpose of this work was to determine the<em> </em>prevalence of chronic pain following total hip (THA) or knee (TKA) arthroplasty, and to identify the proportion of the cohort with chronic pain whose symptoms suggested the pain was of neuropathic origin. In addition we evaluated the ability of the NP Subscale of the McGill pain questionnaire [NP-MPQ (SF-2)] to identify individuals with NP vs. Non NP in the TJA population.</p> <p><strong>Methods:</strong> A retrospective cohort study (2-4 years post joint replacement) of 148 participants with primary unilateral TJAidentified from a large joint arthroplasty database (n=1143). Chronic pain was defined as post surgical pain reported 6-12 months following surgery to be 3 or higher (out of 5) on the Oxford Hip/Knee Scores, and that pain was the same or worse than reported preoperatively. A postal survey was used to administer the NP-MPQ (SF-2)and the Self-Administered Leeds Assessment of Neuropathic Signs and Symptoms (S–LANSS) (1.5-3.5 years post TJA). S-LANSS was the “non reference standard” for classification of neuropathic pain. Human research ethics approvals from Sunnybrook Health Sciences Centre and McMaster University/Hamilton Health Sciences were obtained prior to the study.</p> <p><strong>Results</strong>: The response rate to the postal survey to identify those with chronic pain of neuropathic origin was 53%. Thirteen percent of individuals experienced chronic pain; among individuals with chronic pain, neuropathic subtype was found in 28% (S-LANSS ≥ 12) - 43% [NP-MPQ (SF-2) ≥ 0.91]. Receiver Operating Characteristic (ROC) analysis for NP-MPQ (SF-2) yielded an area under the curve of 0.89 (95% CI: 0.82, 0.97). A cut off score of 0.91 NP-MPQ (SF-2),<strong> </strong>maximized sensitivity (89.5%) and specificity (75.0%). Our results revealed moderate correlation (r=0.56; 95% CI: 0.40, 0.68) between the S-LANSS and NP-MPQ (SF-2)scores in patients with NP post TJA.</p> <p><strong>Conclusion: </strong>Based on our results, a considerable percentage of individuals (13%) experience chronic pain following TKA and THA. Moreover, among individuals with chronic pain symptoms, a significant proportion (28-43%) of those experience pain that appears to have a neuropathic component, even 1.5 to 3.5 years following surgery.Overall prevalence of NP in TJA was 3.3 to 4.5%. The NP-MPQ (SF-2)subscale demonstrated “good” discriminatory ability, thus it might be useful in identifying patients with NP following TJA. Moderate association exists between the scales and this could affect prevalence rates in studies; or diagnosis of NP of individual patients based on the criterion used.</p>