Background/Purpose: Rheumatology consultation to rule out SLE is common. However, in a substantial proportion of patients, SLE can be neither confirmed nor ruled out at initial visit. We followed a cohort of patients seen in consultation and thought to have possible SLE, through their most recent visits, to examine factors associated with developing SLE.

Methods: Our Lupus Center Registry contains data on 5,032 patients who have received billing codes for SLE (ICD-9 710.0) and rheumatologist reviews for ACR Criteria for Classification. Within the registry, we identified patients: ages 18-60, seen first from January 1, 1998 to December 31, 2008; thought to have possible SLE by a board certified rheumatologist at initial visit; < 4 SLE ACR criteria at that visit; and > 2 visits > 3 months apart. Extensive review of medical records through May 15, 2013 collected: visit dates, SLE manifestations, ACR criteria, autoimmune disease serologies, prescriptions, and final diagnoses per rheumatologists. Fisher’s exact and t-tests were employed to assess differences between patients with and without definitive SLE diagnoses. Multivariable logistic regression models were used to identify independent predictors of definite SLE.

Results: We identified 181 patients meeting inclusion criteria. At initial visit, mean age was 38.4 (SD 12.1) years, mean number of SLE ACR criteria was 2.6 (SD 0.9); 173 (95%) were female and 124 (68%) were white. Mean follow-up was 6.2 (SD 4.3) years. 136 (75%) of patients were prescribed hydroxychloroquine at some point. At last visit, 42 (23%) were diagnosed with SLE; 34 (19%) were thought not to have SLE, and the remaining 105 (58%) were thought still to have possible SLE. In univariable models, significantly higher proportions of patients with final definite SLE than those with possible SLE had positive ANAs, anti-dsDNA, arthritis and renal involvement, but fewer had neurologic involvement (Table). Definite SLE patients were also slightly younger (37 vs. 39 years) and were more likely to be non-white. A higher proportion of those with definite than with possible SLE had arthritis, anti-dsDNA and nephritis, and received hydroxychloroquine. In multivariable models, arthritis (OR 2.9, 95% CI 1.3-6.9), anti-dsDNA (OR 2.9, 95% CI 1.3-6.5) and renal involvement (OR 12.4 95% CI 1.1-135.8) were independent predictors of definite SLE. The most common diagnoses among those without ultimate SLE were fibromyalgia, Sjogren’s syndrome, MCTD and cutaneous lupus. Overall, 34 (18%) were also diagnosed with autoimmune thyroid disease.

Conclusion: Among patients with possible SLE at initial consultation, almost 25% were diagnosed with definite SLE within 6 years and 75% received hydroxychloroquine regardless of ultimate diagnosis. Arthritis, anti-dsDNA and renal disease were independent predictors of SLE. A better means of accurately identifying those who will progress to definitive SLE among those presenting with possible SLE is necessary.

 

Table. Characteristics at Last Follow-up of Patients seen for Possible SLE between 1998-2008 Clinical Definite SLE     N= 42 Possible SLE             N= 105 p value * Not SLE                    N= 34 p value** Mean Follow up years (±SD) 5.8  ± 3.4 5.6 ± 4.1 0.75 8.3 ± 5.0 0.02 Mean Age at first visit (±SD) 36.7 ± 10.5 38.9 ± 12.7 0.31 39.1 ± 11.9 0.36 Mean ACR criteria at follow up (+ SD)γ 4.1 ± 1.6 3.1 ± 1.2 0.0002 3.0 ± 1.4 0.006 Deaths in follow-up, n (%) 2(4.8%) 3 (2.86%) 0.67 0 (0%) 0.33 Female 41 (97.6%) 100 (95.2%) 0.62 32 (94.1%) 0.58 Race/Ethnicity      White 27 (69.23%) 72 (77.4%) 0.69 25 (78.1%) 0.46    African American 6 (15.4%) 10 (10.7%) 0.78 2 (6.2%) 0.28    Asian 3 (7.7%) 6 (5.4%) 0.71 3 (9.4%) 1.00    Hispanic 3 (7.7%) 5 (5.4%) 0.68 2 (5.4%) 1.00    Other/Multiple 3 (7.1%) 12 (11.4%) 0.55 2 (5.9 %) 1.00 Prescribed Hydroxychloroquine 37 (88.1%) 73 (69.5%) 0.02 26 (76.5%) 0.23 SLE Manifestations              Malar Rash 8 (19.1%) 13 (12.4%) 0.0.31 6 (17.6%) 1.00    Discoid rash 0 (0%) 2 (1.9%) 1.00 0 (0%) 0.58    Photosensitivity 10 (23.8%) 24 (22.9%) 1.00 10 (29.4%) 0.61    Oral ulcers 7 (16.7%) 9 (8.6%) 0.24 3 (8.8%) 0.49    Arthritis 32 (76.2%) 61 (58.1%) 0.05 14 (41.2%) 0.002    Serositis 9 (21.4%) 13 (12.4%) 0.20 5 (14.7%) 0.56    Renal disease 3 (7.1%) 0 (0%) 0.02 0 (0%) 0.25    Hematologic 19 (45.2%) 36 (34.3%) 0.25 13 (38.2%) 0.64    Neurological disease 0 (0%) 4 (3.8%) 0.47 5 (14.7%) 0.01    Positive ANA 42 (100%) 103 (98.1%) 1.00 26 (76.5%) 0.0009    Positive Anti-DNA 16 (38.1%) 20 (19.1%) 0.01 4 (11.8%) 0.017    Positive Anti-Sm 4 (9.5%) 2 (1.9%) 0.05 2 (5.9%) 0.68    Positive Anti-Ro/ Anti-La 11 (26.2%) 18 (17.1%) 0.25 8 (23.5%) 1.00    Raynaud’s 18 (42.9%) 46 (43.8%) 1.00 15 (44.1%) 1.00    Other Diagnoses    Autoimmune Thyroid Disease 6 (14.3%) 20 (19.1%) 0.63 6 (17.6%) 0.76    Cutaneous Lupus 0 (0%) 1 (0.9%) 1.00 3 (8.8%) 0.08    Fibromyalgia 4 (9.5%) 6 (5.7%) 0.47 7 (20.6%) 0.20    MCTD 0 (0%) 0(0%) 0.55 3 (8.8%) 0.09    RA 0 (0%) 0 (0%) 0.55 1 (2.9%) 0.45    Scleroderma/ Myositis 0 (0%) 3 (2.7%) 0.55 2 (5.4%) 0.19    Sjogren’s Syndrome 3 (6.7%) 3 (2.9%) 0.35 6 (17.6%) 0.28 γTwo patients diagnosed with SLE fulfilled < 4 ACR criteria: one developed WHO class IV nephritis and the other developed transverse myelitis. *p- value for Definite SLE vs. Possible SLE. ** p- Value for Definite SLE vs. Not SLE.  t -tests for continuous and Fisher’s exact for categorical variables.

---

« Back to 2013 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/development-of-systemic-lupus-erythamatosus-among-possible-systemic-lupus-erythamatosus-patients-seen-in-consultation-long-term-follow-up/