Thrombosis in adult primary immune thrombocytopenia in Korea: incidence and risk factors in a 20-year retrospective cohort.
Source: PubMed, NCBI / U.S. National Library of Medicine
Immune thrombocytopenia (ITP) is primarily a bleeding disorder, yet thrombotic events occur and remain challenging to predict, particularly in Asian populations and across real-world treatment patterns. We evaluated thrombosis incidence and associated clinical and therapy-related factors in Korean adults with primary ITP. We conducted a retrospective cohort study of adults (≥ 19 years) with hematologist-confirmed primary ITP at Seoul National University Bundang Hospital from April 1, 2003 to December 31, 2023. Thrombotic events were identified by diagnostic codes and confirmed by review of imaging and clinical documentation. Patients were classified into five ever-treated groups: never treated (Group 1), first-line only (Group 2), immunosuppressive agents without TPO-RA (Group 3), TPO-RA monotherapy (Group 4), and TPO-RA combination therapy (Group 5). Incidence rates were calculated per 100 person-years (PY). Predictors were assessed using Firth's penalized logistic regression. Among 1,038 patients (median age 57 years; 42.6% male), median follow-up was 1.4 years (IQR, 0.3-4.5) with 3,185 PY. Fourteen patients experienced thrombotic events (19 total; 4 arterial, 15 venous), yielding an incidence of 0.44 per 100 PY (95% CI, 0.24-0.74). No events occurred in never-treated patients; events occurred only in Groups 2-5 (p < 0.001). In multivariable analysis, atrial fibrillation (OR 6.01), splenectomy (OR 6.97), and higher ordinal treatment-intensity gro
Abstract
Immune thrombocytopenia (ITP) is primarily a bleeding disorder, yet thrombotic events occur and remain challenging to predict, particularly in Asian populations and across real-world treatment patterns. We evaluated thrombosis incidence and associated clinical and therapy-related factors in Korean adults with primary ITP. We conducted a retrospective cohort study of adults (≥ 19 years) with hematologist-confirmed primary ITP at Seoul National University Bundang Hospital from April 1, 2003 to December 31, 2023. Thrombotic events were identified by diagnostic codes and confirmed by review of imaging and clinical documentation. Patients were classified into five ever-treated groups: never treated (Group 1), first-line only (Group 2), immunosuppressive agents without TPO-RA (Group 3), TPO-RA monotherapy (Group 4), and TPO-RA combination therapy (Group 5). Incidence rates were calculated per 100 person-years (PY). Predictors were assessed using Firth's penalized logistic regression. Among 1,038 patients (median age 57 years; 42.6% male), median follow-up was 1.4 years (IQR, 0.3-4.5) with 3,185 PY. Fourteen patients experienced thrombotic events (19 total; 4 arterial, 15 venous), yielding an incidence of 0.44 per 100 PY (95% CI, 0.24-0.74). No events occurred in never-treated patients; events occurred only in Groups 2-5 (p < 0.001). In multivariable analysis, atrial fibrillation (OR 6.01), splenectomy (OR 6.97), and higher ordinal treatment-intensity group (OR 2.16 per step) were independently associated with thrombosis (all p ≤ 0.027). Thrombosis was uncommon overall but clustered in high-risk contexts-atrial fibrillation, post-splenectomy status, and higher treatment-intensity patterns, particularly TPO-RA-based regimens. The association between treatment intensity and thrombosis most likely reflects a higher-risk clinical context-including refractory disease, greater inflammatory burden, and more intensive clinical surveillance-rather than a direct causal effect of therapy. These findings support individualized risk assessment, integrating comorbidity profiles and treatment trajectories during periods of therapy escalation and platelet instability.