The Hematologist

November-December 2018, Volume 15, Issue 6

Prevention of Central Venous Catheter-Related Thrombosis: Reality or Still a Dream?

Lori-Ann Linkins, MD, MSc(Clin Epi), FRCPC Associate Professor
McMaster University, Hamilton, ON, Canada

Published on: October 10, 2018

Kahale LA, Tsolakian IG, Hakoum MB, et al. Anticoagulation for people with cancer and central venous catheters. Cochrane Database Syst Rev. 2018;6:CD006468.

Central venous catheters (CVCs) are frequently needed to deliver chemotherapy, blood products, and parenteral nutrition to patients with cancer. They are also essential for drawing blood samples for treatment monitoring in patients who have poor peripheral venous access.

Conversely, the presence of a CVC increases the risk of upper extremity deep vein thrombosis. Estimates of the incidence of symptomatic catheter-related thrombosis (CRT) vary from 1 to 5 percent.1 Due to the morbidity of CRT, including limb edema, pain, and potential loss of line function, the search for an anticoagulant regimen to reduce the risk of this complication has been ongoing.

Dr. Lara A. Kahale and colleagues recently published an update of a 2014 systematic review of thromboprophylaxis in patients with cancer who use CVCs. Various regimens of prophylactic-dose low–molecular weight heparin (LMWH) and low-dose vitamin K antagonists (VKA; 1 mg/d) were evaluated in 13 RCTS enrolling 3,420 participants. The studies varied with respect to primary outcome measure (symptomatic vs. asymptomatic), duration of thromboprophylaxis (8 days-6 months), and whether screening (venography) was performed. A summary of the results for risk of symptomatic CRT and major bleeding up to three months is provided in the Table.

Comparison of Risk for Symptomatic CRT and Major Bleeding According to Prophylactic Anticoagulant Regimen
 Symptomatic CRTMajor Bleeding
ComparatorsNo. of PatientsRelative RiskLevel of Evidence (Grade)No. of PatientsRelative RiskLevel of Evidence (Grade)
LMWH* vs. nothing1,0890.43 (95% CI, 0.22-0.81)Moderate1,0181.49 (95% CI, 0.06 to 36.28)Very low
VKA vs. nothing1,2710.61 (95% CI, 0.23 to 1.64)Low1,0267.14 (95% CI, 0.88 to 57.78)Low
LMWH vs. VKA3271.83 (95% CI, 0.44 to 7.61)Very low2893.11 (95% CI, 0.13 to 73.11)Very low

*One study used unfractionated heparin.

Abbreviations: CRT, catheter-related thrombosis; LMWH, low-molecular weight heparin; VKA, vitamin K antagonist.

These results show that the only regimen that reduces the risk of symptomatic CRT is prophylactic-dose LMWH (level of evidence downgraded due to lack of blinding, incomplete data, and unclear allocation). The risk of major bleeding was not significantly increased with LMWH; however, the total number of events was low as illustrated by the wide confidence interval.

Does this mean patients with CVCs should receive LMWH to prevent CRT? I would argue it does not. This meta-analysis did not consider the burden or cost of LMWH. Given that CVCs are often required for months or even years, daily injection is not a small ask for patients with cancer. Furthermore, there is still uncertainty about the magnitude of benefit, with the estimate ranging from a 19 percent reduction to as much as a 78 percent reduction in CRT.

In my practice, the risk-benefit ratio does not currently favor prophylaxis in the average-risk patient with cancer. However, these results may indirectly support continuing LMWH in those diagnosed with CRT who are beyond the initial three-month treatment period and still require a CVC.

Could direct oral anticoagulants be used as an alternative to prevent CRT? While oral agents are likely to be better tolerated by patients, they would still carry a risk of bleeding, and unlike LMWH, there would be potential for interaction with some antineoplastic agents. Only time and future research will tell if direct oral anticoagulants can make prevention of CRT a reality.

References

  1. Geerts W. Central venous catheter-related thrombosis. Hematology Am Soc Hematol Educ Program. 2014;2014:306-311.

Conflict of Interests

Dr. Linkins indicated no relevant conflicts of interest. back to top