Complication Rates Among Central Venous Access Devices Compared

Get Permission

In the CAVA trial reported in The Lancet, Moss et al found that among central venous access devices used for the delivery of systemic anticancer therapy, totally implanted ports (PORTs) were associated with significantly reduced rates of complications compared with Hickman-type tunneled catheters (Hickman) and peripherally inserted central catheters (PICCs).

Study Details

The open-label multicenter trial included 1,061 patients aged 18 years or older receiving systemic anticancer therapy for ≥ 12 weeks for solid or hematologic malignancies; they were enrolled between November 2013 and February 2018. Patients were randomly assigned to use of a central access device according to four options:

  • Hickman vs PICC vs PORT (2:2:1)
  • PICC vs Hickman (1:1)
  • PORT vs Hickman (1:1)
  • PORT vs PICC (1:1).

Random assignment was stratified for center, body mass index, cancer type, device use history, and treatment mode. The primary outcome was complication rate, a composite of infection, venous thrombosis, pulmonary embolus, inability to aspirate blood, mechanical failure, and other factors, assessed until device removal, withdrawal from the study, or 1-year follow-up. Comparisons of complication rates were analyzed for PICCs vs Hickman (noninferiority), PORTs vs Hickman (superiority), and PORTs vs PICCs (superiority).

Photo credit: Getty

Key Findings

Among 424 patients in the PICC vs Hickman comparison, PICC was used in 212 and Hickman in 212. Among 556 in the PORT vs Hickman comparison, PORT was used in 253 and Hickman in 303. Among 346 in the PORT vs PICC comparison, PORT was used in 147 and PICC in 199. Across comparisons, 87% to 97% of patients had solid tumors. 

In the PICC vs Hickman noninferiority analysis, complications were observed in 110 (52%) of 212 patients with PICCs vs 103 (49%) of 212 with Hickmans. Although the observed difference was less than the 10% noninferiority margin, noninferiority of PICC was not confirmed (odds ratio [OR] = 1.15, 95% confidence interval [CI] = 0.78–1.71), potentially representing inadequate statistical powering. PICCs were associated with higher rates of inability to aspirate blood (21% vs 16%) and mechanical failure (15% vs 3%). Hickman use was associated with a higher rate of all infections (30% vs 11%). Similar rates of venous thrombosis, pulmonary embolism, and other complications were observed in the two groups.  

In the PORT vs Hickman analysis, PORTs were superior, with complications observed in 73 (29%) of 253 patients with PORTs vs 131 (43%) of 303 with Hickmans (OR = 0.54, 95% CI = 0.37­–0.77). PORTs were associated with lower rates of laboratory-confirmed bloodstream infection (6% vs 16%) and exit site infection (4% vs 9%) and a higher rate of suspected catheter-related bloodstream infections (8% vs 5%). Venous thrombosis occurred in 1% vs 2% of patients (P = .56). Rates of other complications were similar in the two groups.


  • PORTs were associated with significantly lower complication rates compared with both Hickmans and PICCs.
  • Complication rates were similar with PICCs vs Hickmans, although noninferiority of PICCs was not established.

In the PORT vs PICC analysis, PORTs were superior, with complications observed in 47 (32%) of 147 patients with PORTs vs 93 (47%) of 199 patients with PICCs (OR = 0.52, 95% CI = 0.33–0.83). PORTs were associated with lower rates of venous thrombosis (2% vs 11%, P = .0024) and mechanical failure (3% vs 11%). PORTs were associated with a higher rate of infection of any type (12% vs 8%), although the mean number of infections per catheter week was 0.02 in both groups.

Overall, the median durations of devices remaining in place were more than 350 days for PORTs, approximately 160 days for Hickmans, and approximately 120 days for PICCs, with the difference partly reflecting a lower rate of removal due to complications.

The investigators concluded: “For most patients receiving systemic anticancer therapy, PORTs are more effective and safer than both Hickman and PICCs.”

Jonathan G. Moss, FRCR, of the Institute of Cardiovascular and Medical Sciences, University of Glasgow, is the corresponding author for The Lancet article.

Disclosure: The study was funded by the UK National Institute for Health Research Health Technology Assessment Programme. For full disclosures of the study authors, visit

The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.