RESUMO
PURPOSE: Identifying risk factors for premature totally implantable venous access device (TIVAD) catheter removal is crucial; however, because of the diversity of study methodologies, there is no consensus on such factors. The objective of the present study was to identify such risk factors by applying a cohort design study with a long-term follow-up period. METHODS: For this cohort study, we selected cancer patients who had newly implanted TIVADs between July 2008 and December 2008. The follow-up period lasted until September 2012. Univariate analysis was performed for age, gender, cancer type, TIVAD brand, puncture site, sidedness of puncture, and catheter tip position. The hazard ratio (HR) of potential risk factors was calculated using the Cox proportional hazards regression model, and Kaplan-Meier curves were applied for catheter survival analysis. RESULTS: Our study consisted of 240 people, with 5 people lost to follow-up. The cumulative premature catheter removal rate of all TIVADs was 9.8%, with the most common reason for premature removal being port-associated blood stream infection (PABSI), which proved to be highest in patients with hematology cancer (27.8%) and upper gastrointestinal cancer (19.4%). Suboptimal tip position (HR 5.13, 95% confidence interval 1.73-15.21) was also a risk factor for premature removal, and it was correlated with symptomatic TIVAD occlusion (p = 0.0004). CONCLUSIONS: PABSI was the most common reason for premature catheter removal, with a varied incidence rate between different cancer types. Suboptimal tip position was also a risk factor. Confirming the final tip position after implantation is crucial. Infection control is important for TIVAD care, especially in high-risk cancer patients.