RESUMO
PURPOSE: We sought to identify clinical features associated with difficult subcutaneous port removals in children. METHODS: Ports placed between April 2014 and September 2017 at our institution were prospectively tracked for difficult removals. A case-control analysis was performed. Patients with ports that were difficult to remove (stuck; cases) were compared to biological sex and age-matched controls in a ratio of 1:3. Logistic regression determined the association between case/control status and clinical features adjusting for biological sex and age as covariates. A multivariable analysis was performed to identify independent associations. RESULTS: 57 stuck ports (28 extreme [10 endovascular intervention] and 29 moderate) and 171 controls were analyzed. Stuck ports were associated with a diagnosis of acute lymphoblastic leukemia (86% cases versus 22.2% controls; p < 0.001) and a longer placement duration (median 2.6 years [interquartile range (IQR) 2.5-2.6] versus 0.8 years [IQR 0.5-1.4]; p < 0.001). On univariate analysis, procedural and device features associated with stuck ports included subclavian access (71.9% cases versus 48.5% controls; p = 0.0126), a polyurethane versus silicone catheter (96.5% cases versus 79.9% controls; p = 0.001), and a rough catheter appearance at removal (92.6% cases versus 9.4% controls; p < 0.0001). A diagnosis of ALL and duration of line placement were associated with having a stuck port on multivariate analysis. CONCLUSION: Polyurethane central venous catheters placed for the two-year treatment of acute lymphoblastic leukemia may become difficult to remove. This constellation of factors warrants more extensive preoperative discussion of risk, endovascular backup availability, and scheduling for longer operating room time.