To Örebro University

Background: One of the key therapeutic goals in systemic lupus erythematosus (SLE) is the prevention of organ damage. This core domain outcome is often evaluated with the Systemic Lupus International Collaborating Clinics/American College of Rheumatology (SLICC/ACR) Damage Index (SDI). Access to longitudinal organ damage data on large populations of unselected SLE patients, which currently remain unavailable, would enable real-world investigations of interventions on organ damage accrual.

Objectives: We aimed to develop a register-based organ damage index (RBODI) using International Classification of Diseases (ICD)-coded register data, and to evaluate its accuracy to estimate SDI scores from a well-characterised cohort. Furthermore, we used the RBODI to describe rates of organ damage accrual, as well as associations with mortality, in newly diagnosed patients with SLE in a population-based nationwide cohort in Sweden.

Methods: In collaboration with practitioners from six medical specialties, we translated original SDI items into ICD-10, Swedish Classification of Surgical and Medical Procedures (KVÅ), and Anatomical Therapeutic Chemical (ATC) codes to calculate a global organ damage score. These codes were retrieved from the National Patient Register (inpatient and outpatient visits) and Prescribed Drug Register, and the RBODI was calculated using similar rules as the SDI, i.e., using the same weighting system, and scoring damage occurring since SLE diagnosis only. Using SDI data from prevalent SLE cases from the Clinical Lupus Register in North-Eastern Gothia cohort (KLURING; 2021) as the gold standard for validation of the RBODI, we estimated the positive predictive value (PPV), sensitivity and specificity to detect the presence of organ damage (SDI=0 versus SDI >0). Among newly diagnosed patients with SLE from the National Patient Register (2005–2022; N=4421), we estimated 5-year cumulative incidences of organ damage overall, and by patient characteristics (age, sex, and year of diagnosis). Cox models were used to estimate the age- and sex-adjusted hazard ratio (HR) of first organ damage accrual (RBODI>0) five years after diagnosis associated with patient characteristics. Lastly, aiming to compare with previous reports from Sweden, among patients with SLE living in Sweden five years after diagnosis (N=3013) we estimated the association between presence of organ damage within the first five years of diagnosis and mortality.

Results: We identified 271 prevalent SLE cases with available SDI data in 2021 from KLURING (mean age at diagnosis 41.0±16.4 years, 86.7% female, mean time since diagnosis 16.7±9.1 years, and 62.7% had developed any organ damage). The RBODI displayed good accuracy to discriminate the presence of organ damage as scored with SDI, with a PPV of 85% (95%CI: 79–90%), sensitivity of 81% (95%CI: 74–86%), and specificity of 76% (95%CI: 67–84%; Figure 1). Among newly diagnosed patients with SLE in the nationwide cohort (mean age at diagnosis 47.5±19.8 years, 82.3% female), males (HR: 1.2; 95%CI: 1.1–1.4) and older individuals (>45 years versus ≤45 years; HR 3.4; 95%CI 3.1–3.8) had an increased risk of developing damage within 5 years of diagnosis, while there was no association with year of diagnosis (Figure 2). Patients with organ damage within the first five years of diagnosis had a 3-fold higher hazard of mortality compared with patients with a RBODI=0 during the same period (HR 2.7; 95%CI: 2.0–3.6).

Conclusion: Our novel RBODI accurately estimates SDI scores, and allows us to describe long-term trends in damage accrual in the largest cohort of incident SLE to date. The strong association between damage accrual early in the disease course and future mortality highlights the need for treat-to-target strategies that incorporate early and efficient interventions to prevent organ damage.