Abstract
Knee osteoarthritis (KOA) is linked to an enhanced release of interleukin-6 (IL-6). Increased levels of IL-6 are associated with greater pain and insomnia. While total knee arthroplasty (TKA) typically results in the reduction of pain, for a subgroup of patients, pain does not improve. Understanding patients' propensity to upregulate IL-6 may provide insight into variation in the clinical success of TKA for improving pain, and insomnia may play an important modulatory role. We investigated the association between pre- and post-surgical changes in clinical pain and IL-6 reactivity, and whether change in insomnia moderated this association. Patients (n = 39) with KOA came in-person before and 3-months after TKA. At both visits, patients completed validated measures of clinical pain and insomnia, as well as underwent quantitative sensory testing (QST). Blood samples were collected to analyze IL-expression both before and after QST procedures to assess changes in IL-6 in response to QST (IL-6 reactivity). Patients were categorized into two groups based on change in clinical pain from pre- to post-surgery: 1) pain decreased > 2 points (pain improved) and 2) pain did not decrease > 2 points (pain did not improve). Based on this definition, 49 % of patients had improved pain at 3-months. Among patients with improved pain, IL-6 reactivity significantly decreased from pre- to post-surgery, whereas there was no significant change in IL-6 reactivity among those whose pain did not improve. There was also a significant interaction between pain status and change in insomnia, such that among patients whose insomnia decreased over time, improved pain was significantly associated with a reduction in IL-6 reactivity. However, among patients whose insomnia increased over time, pain status and change in IL-6 reactivity were not significantly associated. Our findings suggest that the resolution of clinical pain after TKA may be associated with discernible alterations in pro-inflammatory responses that can be measured under controlled laboratory conditions, and this association may be moderated by perioperative changes in insomnia. Randomized controlled trials which carefully characterize the phenotypic features of patients are needed to understand how and for whom behavioral interventions may be beneficial in modulating inflammation, pain, and insomnia.