Antiphospholipid syndrome in the paediatric population: performance of ACR-EULAR classification criteria in a controlled cohort

Introduction

Antiphospholipid syndrome (APS) is a systemic autoimmune disorder marked by thrombotic and non-thrombotic manifestations. In children, diagnosis remains challenging due to lack of validated paediatric criteria. As the 2006 revised Sapporo has been used without validation, and the 2023 American College of Rheumatology/European Alliance of Associations for Rheumatology (ACR-EULAR) classification system based on a more modern disease improves specificity; however, their performance and applicability in the paediatric population have not yet been extensively vetted.

Objective

This study evaluates the performance of the ACR-EULAR APS criteria in a paediatric cohort of APS and antiphospholipid antibody (aPL) patients, focusing on key diagnostic parameters and comparison with traditional criteria.

Methods

A retrospective review of 69 patients’ ≤18 years followed at the Sheba Medical Center (2011–2024) was conducted. Clinical, laboratory and serological data including anti-cardiolipin, anti-β2GPI, lupus anticoagulant as well as APS score (adjusted Global Antiphospholipid Syndrome Score, aGAPSS) were analysed. Patients were divided into APS patients (n=29) or aPL carriers (controls; n=40). APS diagnosis was made based on expert opinion and validated by two independent experts.

Results

In our cohort (mean age 16 years; 62% female). APS diagnosis was more common among those with a family history of APS (37.5% vs 17.5%; p=0.056). Comparing APS patients and aPL carriers, unprovoked venous thrombosis (OR 1.55, 95% CI 1.02 to 2.36) and aGAPSS OR 1.053 (1.031 to 1.076) and platelet number (OR 0.999, 95% CI 0.998 to 1) were the strongest predictors of APS. Moreover, non-thrombotic manifestations such as heart valve disease, livedo reticularis and diffuse alveolar haemorrhage were only present in the APS group. Catastrophic APS rate was 10.3% among our APS group. Mortality rate was 6.7% among APS group and 0% among aPL carriers.

In the paediatric population, the 2023 ACR/EULAR criteria demonstrated superior diagnostic performance compared with the 2006 Sapporo criteria, characterised by a higher positive likelihood ratio (18.23 vs 16.43) and a lower negative likelihood ratio (0.036 vs 0.136), effectively improving both the rule-in and rule-out capacity for APS diagnosis.

Conclusions

The superior sensitivity and likelihood ratio of the 2023 ACR-EULAR criteria suggest they may be a valuable tool for the classification of paediatric APS; however, further prospective validation is essential to optimise their clinical application.