Risk stratification and subgroup analysis of systemic lupus erythematosus patients with pregnancy: based on a 13-year multicenter cohort study.

OBJECTIVES

Systemic lupus erythematosus (SLE) presents with heterogeneous clinical manifestations and pregnancy outcomes, complicating its pregnancy management. This study aimed to identify distinct subgroups of pregnant SLE patients and explore their prognostic implications.

METHOD

In this multicenter cohort study across 45 centers in China, 25 baseline variables were incorporated into a cluster analysis (CA) considering collinearity and clinical relevance. Pregnancy outcomes were compared across clusters, and logistic regressions were conducted for total adverse pregnancy outcome (APO) in the overall cohort and each cluster.

RESULTS

Among 528 pregnancies in 499 patients, 72.7% were planned and 35.4% had positive antiphospholipid antibody (aPL). Living birth rate was 81.1%, and 41.5% experienced ≥ 1 APO. Four clusters were defined. Cluster 1 (n = 176), characterized by serological activity, had slightly higher fetal loss (20.5%). Cluster 2 (n = 117), with high aCL/aβ2GPI positivity but low LA, high planned pregnancy rate (91.5%), and more HCQ and aspirin use, had favorable outcomes. Cluster 3 (n = 194), with high planned pregnancy rate (85.1%) and lowest aPL positivity (28.4%), exhibited the best outcomes. Cluster 4 (n = 41), with lowest planned pregnancy (9.8%), most clinical activity, and highest aPL positivity (70.7%) including LA (53.7%), exhibited most APOs (75.6%), with hypocomplementemia further increasing risk (OR = 9.246, p = 0.022).

CONCLUSIONS

Four clinically meaningful SLE-pregnancy subgroups were identified, each with distinct APO risk. Our findings underscore the harm of unplanned pregnancy, especially clinical activity, and aPL positivity, particularly LA, thus promoted stratification and personalized management in SLE pregnancy. Key points • Four distinct subtypes of pregnant SLE patients were identified, each with different APO risks and predictors. • Major organ activity and LA positivity were strongly associated with higher APO risk, while isolated serological activity and non-LA aPL positivity conferred only modest risk. • HCQ appeared protective, especially in aPL-positive patients, whereas medium to high dose glucocorticoid may be harmful, particularly in those with stable disease. • These findings highlight the importance of early risk stratification and individualized management in SLE pregnancy.