Chronic severe fatigue is a disabling comorbidity in autoimmune diseases such as systemic lupus erythematosus (SLE), primary Sjögren's syndrome (pSS), and multiple sclerosis (MS), sharing highly homologous underlying mechanisms characterized by neuro-immune microenvironment abnormalities and proinflammatory cytokine network cascade activation. Using MS—a classic central autoimmune disease—as a translational research model, this study aimed to quantify and prioritize the main effects of diverse exercise modalities and workload intensities on immune-mediated fatigue via network meta-analysis (NMA), thereby providing high-level evidence for targeted non-pharmacological interventions and personalized rehabilitation prescriptions in rheumatology.

Adhering to the PRISMA-NMA statement, a systematic search of core databases was conducted for randomized controlled trials (RCTs) concerning exercise interventions for immune-related fatigue. Following rigorous quality control and data extraction, 37 high-quality RCTs comprising 43 independent intervention arms and a total of 2,112 patients were included. A frequentist NMA model was employed to deconstruct interventions into 11 topological nodes: multicomponent training (SCT, and its moderate/high-workload subgroups SCT_M, SCT_H), resistance training (RES, and its subgroups RES_M, RES_H), high-intensity interval training (HIT), aerobic training (CAT), general rehabilitation (REH), mind-body interventions (YOP), and passive control (PC). Effect sizes were evaluated using standardized mean differences (SMD) and 95% confidence intervals (CIs). The P-score was utilized to quantify the cumulative probability of each node being the optimal intervention, and publication bias was assessed using comparison-adjusted funnel plots and Egger's regression test.

Based on the baseline and endpoint data of 2,112 patients, an 11-node closed-loop network with PC as the reference baseline was successfully constructed. Pooled effect size analysis demonstrated that, compared to PC, moderate-intensity multicomponent training (SCT_M) exhibited the most significant and robust main effect in fatigue improvement (SMD = -0.50, 95% CI [-0.87, -0.14]); overall multicomponent training (SCT) also showed strong statistical significance (SMD = -0.56, 95% CI [-0.86, -0.27]), with the point estimates for CAT, REH, YOP, and baseline RES ranging from -0.30 to -0.35. Spatial ranking via the P-score cumulative probability model established SCT (0.878), SCT_M (0.786), and HIT (0.703) as the top recommended tier. Subgroup analysis further revealed a non-linear dose-response characteristic regarding exercise workload intensity, where moderate-workload SCT_M significantly outperformed high-workload SCT_H (P-score: 0.786 vs 0.509). Crucially, the effect size for high-intensity single-muscle-group resistance training (RES_H) failed to reach statistical significance (SMD = -0.11, 95% CI widely crossing the line of no effect), and its P-score (0.222) approached that of the control group (0.074), indicating that stress workloads exceeding the threshold may exacerbate autoimmune decompensation and central fatigue. Furthermore, the adjusted funnel plot displayed excellent symmetry, and Egger's test yielded p = 0.9381, indicating that the current network model is devoid of significant confounding by small-study effects or publication bias, ensuring robust global consistency.

For autoimmune-mediated chronic fatigue, different exercise modalities and intensities present significant clinical heterogeneity and non-linear dose-response relationships. Moderate-intensity multicomponent training (SCT_M) yields the optimal benefit-risk profile; conversely, excessively high-workload interventions (e.g., high-intensity resistance) may induce adverse immune stress, leading to diminished therapeutic efficacy. The concept of an "immune-exercise workload adaptive window" distilled from this MS model provides critical, translational evidence-based support for developing precision rehabilitation guidelines for chronic fatigue in rheumatologic conditions like SLE and pSS.