Background/Purpose: Pain is a common symptom in childhood-onset systemic lupus erythematosus (cSLE) which impacts health-related quality of life. Its relationship to disease measures over time is not fully understood. We aimed to i) examine the course of patient-reported pain behavior (outward expressions of pain, e.g., grimacing) and pain interference (extent to which pain disrupts daily life, e.g., sleep, activities, mood) in patients with cSLE versus controls, and ii) explore relationships between these pain measures and disease characteristics in cSLE.

Methods: We conducted a secondary analysis of data collected from a prospective cohort study of patients with cSLE (meeting SLICC or ACR/EULAR SLE classification criteria) and age/sex-matched healthy controls. Patient-Reported Outcomes Measurement Information System (PROMIS) pain behavior and pain interference T-scores, and disease characteristics (disease duration, glucocorticoid use (current prednisone-equivalent dose, cumulative dose from time of SLE diagnosis), and Physician Global Assessment (PGA)) were collected at two time points: baseline (time of study enrollment) and 1-year follow up. Descriptive statistics summarized group characteristics. Within- and between-group differences in PROMIS scores were analyzed using Mann-Whitney and Wilcoxon signed-rank tests. Associations between PROMIS measures and disease characteristics were examined using Spearman’s rank correlation test.

Results: We included 51 youth with cSLE (mean age=15y (SD 1.7), 88% female) and 38 controls (mean age =15y (SD 1.7), 84% female). cSLE disease duration at baseline was median 24 months (IQR 12-45) (Table 1). Compared to controls, cSLE participants pain behavior (baseline: 37 vs 25, p < .001; one-year: 37 vs 29: p = .008) and pain interference (baseline: 46 vs 39, p = .007; one-year: 46 vs 41, p = .027) were significantly higher (Table 2). Scores remained stable over time within both groups, with no significant mean change at follow-up. For the cSLE group, disease duration negatively correlated with both pain measures at baseline (pain behavior: rho = -.372, p = .007; pain interference: rho = -.382, p = .005), but not at one year follow-up (Table 3). PGA significantly associated with both pain measures for all time points (pain behavior: baseline rho = .394, one-year rho = .389; pain interference: baseline rho = .459, one-year rho = .415, p < .005 for all). Current glucocorticoid dose at visit and cumulative glucocorticoid use did not correlate with pain.

Conclusion: Youth with cSLE report persistently elevated pain interference and pain behavior, underscoring the lasting impact of pain. The association of disease duration with pain weakened over time, initially a negative correlation at baseline and no correlation at one year. This may suggest that early disease manifestations causing pain are controlled (e.g. inflammation), however long-term pain manifestations of SLE may not be. Interestingly, PGA consistently positively correlated with pain indicating that clinicians are sensing and reporting on patient pain in their global assessment, beyond the early manifestations. These findings highlight the importance of furthering our understanding of pain and its causes in youth with SLE

Table 1: Demographic and Disease Characteristics of cSLE and Controls at Baseline and 1-Year Follow-Up

Table 2: PROMIS T-Scores at Baseline and 1-Year Follow Up for cSLE and Control Participants, and Within- and Between-Group Comparisons

Table 3: Exploratory Spearman Correlation Table between PROMIS Pain measures and Disease Characteristics across Baseline and 1-Year Follow-Up for cSLE Participants

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/longitudinal-exploration-of-pain-and-disease-characteristics-in-youth-with-childhood-onset-lupus/