Background/Purpose: Sleep and pain have been shown to be reciprocally related. Studies have suggested a stronger effect of sleep impairment on chronic pain such as fibromyalgia and widespread pain than pain effects on sleep quality. In this study, we attempted to disentangle the longitudinal association of sleep quality with knee pain development and worsening related to knee OA from chronic widespread pain.

Methods: In the Multicenter Osteoarthritis Study of participants with or at risk of knee OA, sleep quality was assessed at the 60-month study visit (baseline) using a single questionnaire item on a four-point Likert scale from the Pittsburgh Sleep Quality Index, i.e. the overall quality of sleep in the past 7 days and 2) in the Center for Epidemiologic Studies Depression Scale Revised (CESD) for frequency of restless sleep. We categorized sleep quality as ‘fairly good’ and ‘very good’ for the upper two category responses, and ‘bad’ (referent group) for the lower two category responses combined together. Chronic widespread pain (WSP) was defined as having pain above and below the waist, on both sides of the body, and in the axial region. We defined our outcomes as follows: 1) knee pain worsening as relative change ≥ 14% or absolute change ≥ 2 in WOMAC pain score; and 2) incident consistent frequent knee pain (CFKP, knee pain on most days in past month at both the clinic visit and telephone screen ~30 days prior to the clinic visit and not present at baseline). We examined the relation of sleep quality to each knee pain outcome over a 2-year period stratified by baseline WSP status using logistic regression with Generalized Estimating Equations to account for correlations between two knees within an individual, and adjusted for potential confounders (see Table).  

Results: We included 2329 participants (4658 knees) with valid values for sleep quality [mean (SD) age: 62.1 (7.9), BMI: 30.9 (6.1), 60.5% female, 84.3% White, 41% with WSP]. The spearman correlation for sleep quality and restless sleep was 0.69 (p<0.001). There was a significant interaction between sleep quality and baseline WSP for both outcome measures (p<0.01). In those who had knee pain at baseline, better sleep quality was associated with a lower risk of knee pain worsening regardless of baseline WSP (p for trend <0.04), with an effect that appeared to be stronger among those with WSP. No significant association was found for risk of incident CFKP in either stratum of WSP (see Table). Similar results were observed using restless sleep in CESD.

Conclusion: Although better sleep quality was associated with a lower risk of worsening knee pain regardless of chronic widespread pain, sleep quality did not affect the risk of developing new knee pain. Objective measures of sleep quality and knee pain are warranted to study the role of sleep in knee pain in OA. Nonetheless, these data suggest that for those with knee pain, sleep quality improvement should be considered for knee pain management.