Peripherally Inserted Central Catheters Sharply Increase Risk of Childhood VTE Events

In children who received central venous catheters, peripherally inserted central catheters (PICCs) appear to be associated with a higher risk of venous thromboembolism (VTE), compared with tunneled lines (TLs), according to findings from the multicenter CIRCLE study. These results, which were published in Blood by Julie Jaffray, MD, from the Children’s Hospital Los Angeles, and her colleagues, also suggest that having either a prior history of VTE, multi-lumen central venous catheter (CVC) compared with single-lumen, or leukemia compared with other diagnoses increases the risk of VTE in children who receive CVCs.

“[CVCs] are necessary and important devices often required in the care of medically complex and acutely ill children. … Unfortunately, these devices can lead to serious complications,” Dr. Jaffray and co-authors explained. While VTE, including of deep venous thrombosis and pulmonary embolism, has been rare in children, “recent data demonstrate that the incidence of VTE has increased by 70% to 200% in the past 2 decades with ~80% of cases in children caused by CVCs.”

To evaluate the incidence rate of symptomatic, catheter-related VTE in pediatric patients, the investigators conducted the multicenter, observational CIRCLE cohort study in 1,742 children older than 6 months who had a newly placed PICC or TL between October 2013 and June 2018 at one of four tertiary care centers in the U.S.

The researchers assessed incidence of CVC-related VTE in PICCs versus TLs (primary endpoint), as well as incidence of central line-associated bloodstream infections (CLABSIs), catheter malfunctions in PICCs and TLs, and risk factors for CVC-related VTE.

During the study period, a total of 1,967 CVCs were placed; 183 children had two or more CVCs placed. At time of CVC placement, the median patient age was 6.4 years (range = 0.6-17.9). The most common diagnosis leading to the need for a CVC was cancer (41%).
As the researchers expected, most of the CVCs (64%; n=1,257) were PICCs, and the remaining 36% (n=710) were TLs. The authors noted that PICCs likely were chosen more often due to the ability to place and remove PICCs at bedside with little or no sedation, compared with TLs.

The incidence rate of catheter-related VTE within 6 months of placement was 5.9%. Among all CVC types, the median time from insertion to VTE diagnosis was 15.5 days (range = 1-162). Most events (n=75; 80%) occurred with PICCs, for an incidence rate of 9.0%, compared with an incidence rate of 2.9% with TLs. Overall, the risk of VTE was significantly higher with PICCs compared with TLs (hazard ratio [HR] = 8.5; 95% CI 3.1-23; p<0.001).

For CLABSI, the incidence rate in the entire cohort was 17%. Again, patients with PICCs had a significantly greater likelihood of having this adverse event, compared with patients with TLs (22% vs. 15%; HR=1.6; 95% CI 1.2-2.2; p=0.002). CVC malfunctions also were higher with PICCs (HR=2.0; 95% CI 1.6-2.4; p<0.001).

In multivariable analyses, the following variables were associated with a higher risk of VTE:

  • having a PICC inserted (HR=8.5; 95% CI 3.1-23; p<0.001)
  • prior history of VTE (HR=23; 95% CI 4-127; p<0.001)
  • multi-lumen CVC (HR=3.9; 95% CI 1.8-8.9; p=0.003)
  • leukemia (HR=3.5; 95% CI 1.3-9.0; p=0.031)

Other CVC characteristics, such as CVC tip location, access location, and catheter brand, were not associated with an increased risk of VTE.

“We don’t want to tell others to never use a PICC, because there are definitely many patients where it’s safer and better to use this option,” said Dr. Jaffray. “PICCs can be placed at bedside and can be done with nursing, so there’s definitely still a role for them, but we want to encourage [clinicians] to think about what lines are being placed [and whether central venous access is necessary].” In many instances, she added, patients will tolerate a regular peripheral IV instead of needing a more invasive line like a PICC. “In those cases, I would strongly encourage clinicians to avoid putting in a central access.”

“We want to encourage [clinicians] to think about what lines are being placed [and whether central venous access is necessary].”

—Julie Jaffray, MD

Based on these findings, she also suggested addressing other modifiable risk factors for CVC-related VTE, including limiting multi-lumen CVCs, using strategies for preventing CLABSIs, or placing a TL over a PICC when possible.

A limitation of the study included the lack of surveillance imaging for the assessment of thrombosis in asymptomatic patients who received TLs. In addition, the investigators suggested that the exclusion of temporary, non-tunneled CVCs (which are frequently used in critically ill children) potentially limited the generalizability of findings across a broader patient population. Dr. Jaffray added that the CIRCLE study lacked information about PICC and TL use in neonates, the pediatric patient population at the highest risk for VTE.

Despite the study’s limitations, Dr. Jaffray noted that the research was unique in its investigation of PICCs versus TLs because it looked at placement of these lines in different types of patients with varying medical histories, compared with previous studies that looked only at patients with specific characteristics. “Our study captured a broader patient population,” she concluded.

The authors report no relevant conflicts of interest.


Jaffray J, Witmer C, O’Brien SH, et al. Peripherally inserted central catheters lead to a high risk of venous thromboembolism in children. Blood. 2020;135:220-226.