Thrombocytopenia in systemic lupus erythematosus real-world insights from the nationwide RISE database

Abstract

Objectives

Systemic lupus erythematosus (SLE) is an autoimmune rheumatic disease with a highly heterogeneous presentation. Thrombocytopenia (TP) is an uncommon manifestation in SLE patients and is often associated with major organ involvement and poor prognosis. In this study, we aimed to evaluate the prevalence of TP in SLE patients and analyze its association with demographic and clinical factors.

Methods

We utilized data from the ACR’s Rheumatology Informatics System for Effectiveness (RISE) registry, a large, electronic health record-enabled database that collects data as part of routine clinical care. The study included patients who were ≥18 years old, had ≥2 SLE diagnostic codes recorded ≥30 days apart between 2016 and 2022, and had a valid complete blood count (CBC) recorded within 1 year of the second SLE code. Thrombocytopenia (TP) was classified based on the lowest platelet count within 1 year of the second SLE code: mild: 100,000–150,000/µL, moderate: 50,000–99,000/µL, and severe: <50,000/µL. We reported the frequency of TP by demographic and clinical characteristics. To identify factors associated with moderate-to-severe TP, we employed multi-level logistic regression models, controlling for covariates and accounting for clustering by practices.

Results

The study included 30,062 patients (91.2% female; mean age: 56 years, SD: 16). The frequencies of TP and severe TP were 9.3% and 0.5%, respectively. Patients with moderate-to-severe TP had significantly higher frequencies of male sex, African-American and Hispanic ethnicity, lupus nephritis, leukopenia, anemia, hypocomplementemia, anti-dsDNA positivity, end-stage renal disease, thrombosis, hemolytic anemia, antiphospholipid syndrome, and multiple comorbidities compared to other SLE patients. Multivariable logistic regression analysis demonstrated persistent independent associations of sex, race/ethnicity, multiple comorbidities, positive dsDNA, and hypocomplementemia with an increased risk of moderate-to-severe TP in SLE.

Conclusions

The RISE registry revealed that severe TP was less common in SLE patients from community practice compared to previously reported data from selected tertiary centers. Serologically active SLE, multiple comorbidities, male sex, and non-white race were identified as independent factors associated with TP in this database. Further studies are needed to elucidate the exact mechanisms and clinical significance of TP in SLE.