Central venous access devices (CVADs) are commonly employed in the management of cancer patients, facilitating administration of chemotherapy, parenteral nutrition, blood products and antibiotics, and stem cell collection/reinfusion. Previous studies have demonstrated certain cancer subtypes to have a higher risk of thrombosis, when incorporated into a risk scoring system. We sought to review the incidence and risk factors for catheter related thrombosis (CRT) in a large cohort of hematology and oncology patients using a retrospective study in an Australian tertiary hospital, including the influence of cancer subtype.

We reviewed data from a prospectively gathered CVAD database. The presence of CRT was defined as a positive duplex ultrasound scan either while line was in situ or within one week of line removal. Only patients symptomatic for CRT underwent USS imaging. We performed univariate and multivariate cox regression, and competing risk analysis to assess which parameters were associated with an increased risk of CRT.

5043 central lines were inserted into 3218 patients between Jan 2003 and Dec 2016 with the majority (72%) of lines being peripherally inserted central catheters (PICCs). Overall incidence of CRT was 3.55%, at a rate of 0.45 per 1000 line days. CRT developed at a median of 12 days following line insertion. In univariate analysis, PICCs were associated with the highest rates of CRT (HR 7.48, 95%CI 1.03-54.1, p=0.046) when compared to non-tunnelled lines, while implantable ports (HR 0.47 95%CI 0.03-7.90 p=0.597) and tunnelled catheters (HR 0.00) were not associated with increased thrombosis risk. Patient age <50 (HR 1.46, 95% CI 1.06-2.00, p=0.019) and number of prior CRT (HR 1.95, 95% CI 1.30-2.92, p=0.01) were associated with higher thrombosis rates in univariate analysis. In multivariate analysis, PICC lines (HR 8.63, 95% CI: 1.17-63.5 p=0.034) and number of prior CRT (HR 1.84, 95% CI 1.18-2.86, P=0.007) increased CRT risk. Age groups 50-61 (HR 0.56 95% CI 0.37-0.83 p=0.004) and > 61 (HR 0.65 95% CI 0.46-0.91 p=0.012) were associated with reduced risk compared to age group <50. We used the tumor score from the previously validated Khorana score and found mixed results, with intermediate, but not high-risk tumours associated with higher thrombosis risk (HR1.42 95%CI 1.03-1.95 p=0.031 and HR 0.65 95% CI 0.26-1.60 p=0,346 respectively). When individual cancer subtypes were analysed in univariate analysis, Hodgkin lymphoma (HR 2.43 95% CI 1.01-5.82 p=0.046) and germ cell tumours (HR 2.63 95% CI 1.05-6.55 p=0.038) were found to be associated with increased CRT rates. However, once controlling for age and line type in multivariate analysis, no subgroup of cancer was associated with higher CRT rates. Side of insertion was not associated with higher thrombosis risk (left vs right HR 0.79 95% CI 0.58-1.07 P=0.13). When the analysis was restricted to PICC lines (n=3642), we found no difference in significant predictors of CRT.

In conclusion, this is the largest published series reviewing CRT from a single centre. In patients with cancer, age <50, presence of a PICC line and prior CRT were associated with higher risk of CRT. Cancer subtype and insertion side were not significant predictors of thrombosis in this series.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract