288 Phenotypic subgroups in systemic lupus erythematosus: an exploratory cluster analysis

2019 
Background Lupus is a complex, heterogeneous disease. We have developed a conceptual model to characterize SLE activity using two dimensions: Type 1 SLE includes active inflammatory manifestations of SLE, including arthritis, nephritis, and rashes; Type 2 SLE includes fatigue, myalgia, mood disturbance, and cognitive dysfunction, which can persist in the absence of inflammatory findings. We have grouped SLE patients into phenotypic clusters based on the extent of Type 1 and 2 SLE features. Methods Consecutive SLE patients meeting ACR or SLICC criteria in a university rheumatology clinic were included. Patients completed the Systemic Lupus Activity Questionnaire (SLAQ) and the ACR Fibromyalgia (FM) Diagnostic Criteria. The Fibromyalgia Severity Score (FSS) is the sum of the widespread pain (0–19) and symptom severity (0–9) scores. The physicians global assessment (PGA) was also recorded. Patients were grouped into clusters based on PGA (Type I SLE) and FSS (Type 2 SLE) using hierarchical clustering with Wards minimum variance method. Differences were estimated by ANOVA and Fishers exact test. Results The 419 SLE patients (92% female; mean age: 45 years) were classified into 7 clusters (figure 1). Minimal SLE: Clinical Remission (n=85): minimal Type 1 and 2 SLE. Clinical Remission with Fatigue (n=113): minimal Type 1 and mild Type 2 symptoms of fatigue (70%) and waking unrefreshed (58%). Predominantly Type 1: Moderate Type 1 SLE (n=56): active Type 1 SLE (43% arthritis, 55% anti-dsDNA+) and minimal Type 2 symptoms. Severe Type 1 SLE (n=31): severe Type 1 SLE, with active nephritis (39%), arthritis, new rashes, dsDNA+, and low C3/C4, with mild Type 2 symptoms, primarily fatigue (61%). Predominantly Type 2: Type 2 SLE (n=58): minimal Type 1 SLE and significant Type 2 symptoms including high widespread pain scores, fatigue (97%), waking unrefreshed (86%), depression (67%), cognitive dysfunction (59%). Mixed SLE: Moderate Mixed SLE (n=52): active Type 1 SLE (69% arthritis, 35% anti-dsDNA+) and active Type 2 SLE, with moderate widespread pain scores, fatigue (90%), waking unrefreshed (96%), forgetfulness (52%), and depression (49%). Severe Mixed SLE (n=24): active Type 1 SLE (54% arthritis, 25% proteinuria) combined with severe Type 2 SLE, with high widespread pain scores, depression (86%), fatigue (100%), and waking unrefreshed (100%). Conclusions Patient-reported measures can be identifying distinct clusters of patients with higher and lower levels of Type 1 and 2 SLE features. We have already found this approach useful in direct clinical care and are working to identify immunologic differences between clusters and optimal management protocols. Funding Source(s): None
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []
    Baidu
    map