Background/Purpose: SLE is clinically heterogeneous and associated with increased hospitalizations. However, it remains unclear which patient subgroups drive this excess in unplanned healthcare use. We aimed to determine 1) whether elevated hospitalization rates are concentrated among patients with LN or aPL, and 2) whether individuals with non‐renal, aPL‐negative SLE exhibit hospitalization risks comparable to the general population.

Methods: Incident SLE cases (1995–2018) from a 27‐county region in the US Midwest were matched 1:2 to non‐SLE comparators on age, sex, race/ethnicity, and county. SLE onset was defined by fulfillment of the 2019 EULAR/ACR criteria with follow‐up starting after discharge if diagnosed during hospitalization. Using a time-dependent variable (Figure), patients contributed person‐time to the SLE state (non‐renal, aPL‐negative) until developing LN or aPL positivity. The aPL state began at detection of lupus anticoagulant or IgG anticardiolipin/β2‐glycoprotein‐1 (≥40 GPL). The LN state began at biopsy confirmation or proteinuria ≥500 mg/day. Patients meeting both LN and aPL criteria were analyzed in the LN state.The primary outcome was inpatient stays ≥24 hours. Follow-up continued until December 2023, emigration, or death. Conditional frailty models—adjusted for sex, age, white race, and smoking—compared admission hazards across SLE states vs. comparators; secondary models examined time-varying hazards post-discharge (0–30 days, 31 days–1 year, >1 year).

Results: Among 326 SLE patients and 652 comparators (median age 46; 79% female; 81% non-Hispanic White), SLE patients had 829 admissions over 3,272 person-years, vs. 799 admissions over 6,654 person-years among comparators (Table 1). Of all the SLE patients, 86 transitioned to the LN state (25 were also aPL-positive) and 69 to the aPL state. In fully adjusted models, the SLE state was associated with a 51% higher hazard of hospitalization compared to controls (HR 1.51, 95% CI 1.24–1.82; p < .001). aPL state conferred an 85% increase (HR 1.85, 95% CI 1.33–2.58; p < .001), and LN state the greatest rise at 159% (HR 2.59, 95% CI 2.01–3.32; p < .001). Time-stratified analyses demonstrated peak risk in the first 30 days post-discharge—hazard ratios of 1.82 for SLE, 2.66 for aPL, and 3.78 for LN—with gradual attenuation thereafter (Table 2).

Conclusion: In this population-based cohort, patients with lupus nephritis demonstrated the greatest increase, those with antiphospholipid antibodies an intermediate rise, and non-renal, aPL-negative SLE the smallest—yet still substantial—elevation. These findings highlight the need for stratified risk assessment in SLE: while everyone with SLE is at increased risk of rehospitalization, the greatest attention—and potentially the most aggressive preventive strategies—should be directed toward those with LN and aPL.

*Unless otherwise indicated, values are the number (%) of patients.

† Four patients did not have an SLEDAI calculated within 60 days following the index date. As a result, SLEDAI median and interquartile range (IQR) were calculated based on a total of 322 patients.

Figure. Timeline Representing the Contribution to Unplanned Hospitalizations Across Three Different Systemic Lupus Erythematosus (SLE) States: 1) Non-renal, antiphospholipid antibody-negative SLE (SLE State); 2)Lupus Nephritis (LN State); and 3) Antiphospholipid antibody-positive SLE (aPL State).

*Results are adjusted for sex, age at baseline, white race, and ever smoking. 30d= 30 days, 1y= 1 year, HR= Hazard Ratio, 95% CI= 95% Confidence Intervals.

†Result did not reach statistical significance.

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/differential-risk-of-inpatient-admissions-among-sle-subgroups-a-population-based-analysis/