The effect of systemic and local risk factors on triggering peripherally inserted central catheter–related thrombosis in cancer patients: A prospective cohort study based on ultrasound examination and structural equation modeling

Abstract

BackgroundThe negative synergistic effect of cancer and a peripherally inserted central catheter could significantly increase the incidence of thrombosis. Rather than identifying risk factors for peripherally inserted central catheter–related thrombosis, exploring the effect of these risk factors might be a promising method to improve the outcomes of thrombosis. ObjectivesTo analyze the effect of systemic and local risk factors on triggering peripherally inserted central catheter–related thrombosis in the first two weeks post–insertion in cancer patients. DesignA prospective cohort study. SettingThe study was conducted at a 4500-bed university-affiliated medical center in China. ParticipantsOne hundred seventy-three cancer patients with peripherally inserted central catheters were included. MethodsPeripherally inserted central catheter–related thrombosis was assessed using ultrasound at a series of timepoints, once every two days post–insertion. Data on age, body mass index, blood hypercoagulation, insertion attempts, catheter-to-vein ratio, and blood flow velocity were collected as risk factors. Descriptive statistics and structural equation modeling were used to describe the study samples and analyze the effects of systemic and local risk factors. ResultsAmong the 173 cancer patients included, 126 (72.8%) patients were diagnosed with peripherally inserted central catheter–related thrombosis. Most thromboses (n = 118, 93.7%) were detected within five days, and 100% were detected within nine days post–insertion. Structural equation modeling analysis showed that local risk factors [catheter-to-vein ratio (standardized path coefficient = 0.32, p < 0.05) and blood flow velocity (standardized path coefficient = −0.35, p < 0.05)] had a greater effect than systemic factors [age (standardized path coefficient = 0.13, p < 0.05) and blood hypercoagulation (standardized path coefficient = 0.17, p < 0.05)] on triggering peripherally inserted central catheter–related thrombosis. ConclusionPeripherally inserted central catheter–related thrombosis is quite common and can occur very early post–insertion in cancer patients. Among the common risk factors, local risk factors reflecting peripherally inserted central catheter technology itself had a greater effect than systemic risk factors reflecting predisposition to thrombosis.Clinical Registration: Clinical Trials ChiCTR1900024890.