Cancer-associated thrombosis in patients with implanted ports: a prospective multicenter French cohort study (ONCOCIP)

Abstract

The need to accurately identify cancer outpatients at high risk of thrombotic complications is still unmet. In a prospective, multicenter cohort study (ONCOlogie et Chambres ImPlantables [ONCOCIP]), consecutive adult patients with a solid tumor and implanted port underwent 12-month follow-up. Our primary objective was to identify risk factors for (1) catheter-related thrombosis, defined as ipsilateral symptomatic upper-limb deep-vein thrombosis with or without pulmonary embolism, and (2) venous thromboembolism other than catheter-related, defined as any symptomatic superficial- or deep-vein thrombosis (other than catheter-related) or pulmonary embolism, and incidental pulmonary embolism. All events were objectively confirmed and centrally adjudicated. Rate assessments integrated competing risk of death. Overall, 3032 patients were included (median age: 63 years; women: 58%). The most frequent cancer locations were breast (33.7%), lung (18.5%), and colorectal (15.6%), cancer being metastatic in 43.2% of patients. Most patients (97.1%) received chemotherapy. By 12 months, 48 (1.6%) patients had been lost to follow-up and 656 (24.6%) had died; 3.8% (n = 111) of patients had experienced catheter-related thrombosis, and 9.6% (n = 276) venous thromboembolism other than catheter-related. By multivariate analysis, use of cephalic vein for catheter insertion predicted catheter-related thrombosis, whereas ongoing antiplatelet therapy was protective; risk factors for venous thromboembolism other than catheter-related were advanced age, previous venous thromboembolism, cancer site, and low hemoglobin level or increased leukocyte count before chemotherapy. In conclusion, this large prospective cohort study showed a high rate of venous thromboembolism in patients with a solid tumor and implanted port. Risk factors for catheter-related thrombosis differed from those for venous thromboembolism not catheter-related. This trial was registered at www.clinicaltrials.gov as #NCT02025894.

Implanted vascular access device related deep vein thrombosis in oncology patients: A prospective cohort study

“…In both studies, the majority of patients were diagnosed with breast cancer or colorectal cancer [10,16]. A recently published prospective cohort study examining the rate of IVAD related VTE in 3032 patients across France found a rate of 3.8% [15]. This rate is higher than what was found in this study, and their population included patients who were on prophylactic and therapeutic anticoagulation, which may have decreased the rate of IVAD-related UEDVT.…”

“…The primary outcome of this study was symptomatic IVAD-related UEDVT defined as a symptomatic VTE in any deep vein throughout the course of the catheter, or symptomatic pulmonary embolism (PE) associated with the IVAD-related UEDVT [15]. These include the axillary, brachiocephalic, jugular and subclavian veins on the ipsilateral side of the IVAD insertion.…”

Direct oral anticoagulant for the prevention of thrombosis in ambulatory patients with cancer: A systematic review and meta‐analysis

“…Third, both trials included symptomatic upper extremity DVT as part of the composite VTE outcome. Because of its low incidence and frequency co‐occurrence with other outcome events, it is challenging to pool the results as a separate endpoint; however, the ONCOCIP prospective cohort study showed that upper extremity DVT likely had different incidence and risk factors than PE and lower extremity DVT . Fourth, the utility weights in the EUT model were derived in older studies with predominately non‐cancer patients receiving either warfarin or LMWH .…”

“…B, Risk ratio for clinically relevant non-major bleeding (during on-treatment period) it is challenging to pool the results as a separate endpoint; however, the ONCOCIP prospective cohort study showed that upper extremity DVT likely had different incidence and risk factors than PE and lower extremity DVT. 32 Fourth, the utility weights in the EUT model were derived in older studies with predominately non-cancer patients receiving either warfarin or LMWH. 17 Although these utility weights were found to be similar in a more recent study, very few patients in either study included cancer patients receiving DOAC; this remains an important area of future research.…”

Risikofaktoren für Chemotherapie-assoziierte venöse Thrombosen bei gynäkoonkologischen Patientinnen

“…In einer anderen multizentrischen prospektiven Studie an 3032 Patienten differieren die Risikofaktoren für das Auftreten einer Katheter-assoziierten (an der oberen Extremität in der Umgebung des Ports) Thrombose von denen für das Auftreten einer nicht Katheter-assoziierten Thrombose [18]: Ein unabhängiger Risikofaktor für die portassoziierten VTs war die Implantation des Ports in die V. cephalica. Bei den nicht portassoziierten VTs ergaben sich u. a. das Vorliegen einer Anämie (Hb-Wert < 10 g/dl) sowie einer Leukozytose (> 11 × 10 9 /l) als unabhängige Risikofaktoren [18]. In unserer Studie zeigte sich im χ 2 -Test ein auffällig starker Zusammenhang zwischen dem Vorliegen eines Ports und dem Auftreten eines Gefäßverschlusses, wobei die genaue Lokalisation der VT (z.…”