Relationship between systemic inflammation and recovery over 12 months after an acute episode of low back pain

Abstract BACKGROUND CONTEXT Individual characteristics can influence outcomes after injury. Our previous work in individuals with early-acute low back pain (LBP) identified subgroups (clusters) with specific biopsychosocial features that recovered poorly or well by 6 months. PURPOSE This study extends on that work by revealing the short- and long-term trajectories of recovery and systemic inflammation of these participant clusters: (1) “inflammatory & poor sleep” (Cluster 1), “high TNF & depression” (Cluster 2), “high pain & high pain-related fear” (Cluster 3), and “low pain & low pain-related fear” (Cluster 4). STUDY DESIGN/SETTING Longitudinal cohort study. PATIENT SAMPLE Eighty-three individuals within 2 weeks of an acute episode of LBP – grouped into their a priori-defined cluster. OUTCOME MEASURES General participant characteristics (sex, age, body mass index, smoking history, previous LBP history); self-reported LBP (0–10 numerical rating scale, LBP-related disability (Roland-Morris Disability Questionnaire), depression (Center for Epidemiological Studies Depression Scale, pain catastrophizing (Pain Catastrophizing Scale), fear avoidance (Fear Avoidance Beliefs Questionnaire), pain self-efficacy (Pain Self-Efficacy Questionnaire), and sleep (Pittsburgh Sleep Quality Index); systemic inflammatory biomarkers (C-reactive protein [CRP], interleukin-6 [IL-6], interleukin-1β, tumor necrosis factor [TNF]). METHODS Participants provided blood for the measurement of CRP/cytokines, and completed questionnaires related to their pain/disability, psychological and sleep status. Blood measures were repeated 3-monthly for 9 months, and pain/disability were self-reported fortnightly for 12 months. Recovery (change in pain) and CRP/cytokines were longitudinally compared between clusters using mixed-models. Associations between baseline factors and follow-up CRP/cytokines levels were assessed with multiple regression. RESULTS Clusters 1 and 2 were associated, but oppositely, with recovery over the 12-months. Cluster 1 reported most recovery at every 3-monthly interval, whereas Cluster 2 reported least recovery. Cluster 1 had elevated CRP (and IL-6) at baseline that continued to decrease from 3 to 9 months. TNF was elevated early and persistently in Cluster 2. Baseline factors other than inflammation generally failed to predict follow-up inflammation. CONCLUSIONS Findings support the early role of CRP (and perhaps IL-6) in control of inflammation and recovery, and a pathological role of persistent TNF overexpression, which may be perpetuated by depressive-like behaviors.