Predictive Tools for Thrombosis Recurrence in Antiphospholipid Syndrome Demonstrate Suboptimal Performance

The predictive performance of all 3 models was substandard for thrombosis recurrence.

Prediction of thrombosis recurrence among patients with antiphospholipid syndrome (APS) was substandard with use of the adjusted global APS (aGAPSS), Padua, and Caprini scores, according to study findings published in Rheumatic & Musculoskeletal Diseases Open.

Investigators aimed to validate use of the aGAPSS, Padua score, and Caprini score for prediction of thrombosis recurrence among patients with APS.

A single-center prospective cohort study was conducted, including patients with APS who had a known history of thrombotic events at the time of diagnosis. The primary study endpoint was the first recurrence of venous or arterial thrombosis. Thromboses were confirmed by imaging and physician diagnosis.

Cox regression analysis and the Harrell c-index were used to validate the 3 prediction models.

A total of 362 patients were included in the study. Mean patient age was 36.3 years, 57.7% were women, and median follow-up was 2.32 years.

The construction of new prediction models respectively for venous and arterial thrombosis recurrence in APS patients is required to guide treatment.

Patients presented with a total of 53 (14.6%) recurrent thrombotic events during the follow-up period, of which 32 were venous (8.84%) and 21 were arterial (5.80%). None of these recurrences were fatal.

Overall thrombosis risk was 5.0%, 14.3%, and 17.9% at years 1, 3, and 5, respectively. Venous and arterial thrombosis risks at years 1, 3, and 5 were 3.8% and 1.6%, 9.9% and 5.6%, and 12.8% and 6.9%, respectively

At the time of venous thrombosis diagnosis, 37.5% of patients received inadequate anticoagulation, compared with 47.6% of patients diagnosed with arterial thrombosis.

The Harrell c-index for thrombosis prediction was 0.54 (95% CI, 0.44-0.64) for aGAPSS, 0.54 (95% CI, 0.46-0.62) for the Padua score, and 0.50 (95% CI, 0.42-0.58) for the Caprini score.

The best predictor of venous thrombosis was the Padua score (Harrell c-index, 0.61; 95% CI, 0.53-0.69). The best predictor of arterial thrombosis was aGAPSS (Harrell’s C index 0.61; 95% CI, 0.47-0.75).

Study limitations included differing anticoagulation practices at the study site vs other centers and missing data on patient drug adherence, which could potentially influence thrombosis recurrence. Additionally, data on antiphospholipid antibody positivity was not validated due to the small sample size for thrombosis recurrence.

The study authors concluded, “The construction of new prediction models respectively for venous and arterial thrombosis recurrence in APS patients is required to guide treatment.”


Zhao Y, Huang C, Qi W, et al. Validation of three prediction models for thrombosis recurrence in antiphospholipid syndrome patients based on a prospective cohort. RMD Open. Published online July 28,2023. doi:10.1136/rmdopen-2023-003084