PO-06 - Cancer and the risk of venous thromboembolism in stroke patients

P Corraini; A G Ording; V W Henderson; S Szépligeti; E Hováth-Puhó; H T Sørensen

doi:10.1016/S0049-3848(16)30139-6

PO-06 - Cancer and the risk of venous thromboembolism in stroke patients

Thromb Res. 2016 Apr:140 Suppl 1:S178. doi: 10.1016/S0049-3848(16)30139-6. Epub 2016 Apr 8.

Authors

P Corraini¹, A G Ording¹, V W Henderson², S Szépligeti¹, E Hováth-Puhó¹, H T Sørensen¹

Affiliations

¹ Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark.
² Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark; Departments of Health Research & Policy (Epidemiology) and of Neurology & Neurological Sciences, Stanford University, Stanford, California, USA.

PMID: 27161694
DOI: 10.1016/S0049-3848(16)30139-6

Abstract

Introduction: The impact of comorbidity and in particular cancer on the risk of venous thromboembolism (VTE) after stroke is poorly understood.

Aim: We aimed to determine the impact of comorbidity, in particular cancer, on the risk of venous thromboembolism in stroke patients as the excess VTE rates not explained by stroke and comorbidity alone.

Materials and methods: We used Danish national databases to conduct a cohort study including 110,833 patients diagnosed with an incident stroke (72% ischemic) between 1995 and 2012. A comparison cohort of 545,960 members of the general population was matched to the stroke patients by date of diagnosis, year of birth, sex, and specific comorbidities using the Charlson Comorbidity Index and other VTE risk factors. We computed VTE cumulative risks, rates and rate ratios, as well as the interaction with comorbidity (as the excess VTE rates not explained by stroke and comorbidity alone) during five years of follow-up.

Results: Five-year VTE risks were 2.16% and 1.85% in the stroke and general population comparison cohorts, respectively. Three-month VTE rate ratios peaked at a 6-fold increase (95% confidence interval: 4.9;6.2) in stroke patients and remained increased by 52%-20% relative to the general population at subsequent follow-up. 20% to 38% of the three-month VTE rates in the stroke cohort were attributable to the interaction between stroke and comorbidity in patients with moderate (2-3) to high (≥4) Charlson Comorbidity Index scores. Non-metastatic and metastatic solid tumors accounted for most of the observed interaction with stroke, representing 44% and 60% of their attributable three-month VTE rates. No interaction between comorbidity and stroke was observed during subsequent follow-up to 5 years.

Conclusions: Cancer increased the risk of VTE within three months after the stroke.