Treatment of catheter related thrombosis: A systematic review, meta-analysis, and national survey

Catheter-related thrombosis (CRT) is a known complication of central venous catheters and peripherally inserted central catheters, yet optimal treatment remains uncertain. We conducted a systematic review and national survey to assess current CRT management strategies.

Following the PRISMA guidelines, we searched three databases through October 2024 for studies on CRT associated with central venous catheters or peripherally inserted central catheters. Meta-analyses and subgroup analyses were performed by anticoagulant type. A national survey among Dutch intensive care and hematology physicians explored current treatment practices.

Of 4123 records screened, 34 observational studies were included, mostly involving patients with cancer. The venous thromboembolism recurrence rate per 100 patient-years was higher in patients with cancer (14.1; 95% confidence interval, 11.4- 17.4; I² = 35.1) vs patients without cancer (2.0; 95% confidence interval, 0.6-6.0; I² = 10.3; P = .0002). Recurrence was comparable between direct oral anticoagulants (DOACs) and low-molecular-weight heparin/vitamin K antagonists (LMWH/VKAs), at 11.0 vs 7.6 (P = .14). Major bleeding occurred in 10.5 vs 13.1 (P = .45), and clinically relevant nonmajor bleeding in 26.2 vs. 22.4 (P = .70), for DOACs vs LMWH/VKAs, respectively. All studies were observational, most at high risk of bias. Survey data showed LMWH was preferred for symptomatic CRT (50%), with treatment lasting 8 days to 6 months. In asymptomatic CRT, anticoagulant type and duration were left to physician discretion in 64% of cases.

Treatment with LMWH/VKA or DOACs shows similarly low venous thromboembolism recurrence, although rates are higher in patients with cancer. Bleeding was substantial and comparable across therapies. Evidence is limited by observational bias. Survey data show that LMWH predominates for CRT, with variable duration. Well-designed randomized controlled trials are warranted.