Central venous catheters (CVCs) are temporary devices implanted into patients when easy or frequent intravenous access is needed. Doctors often use them. A Hickman line is an example of a CVC. A CVC is used, for instance, for monitoring patients in intensive care, or for giving chemotherapy or intravenous nutrition. However, such catheters can cause blood clots, which can block the line, increase the risk of infection and travel elsewhere in the body such as to the lung (this is called thromboembolism). Heparin is a drug that helps to prevent blood clots and may help prevent these unwanted consequences. But heparin can also cause serious adverse effects (bleeding, allergic reactions, fall in platelet count, etc.). Normal saline solution, a sterile solution of salt in water at a concentration suitable for the blood, is typically used for intravenous infusions. We wanted to know whether heparin helps prevent the unwanted effects of blood clots in CVCs, and if this benefit outweighs its risk of harms.
Six studies with a combined total of 1433 participants were included. The quality of the evidence ranged from very low to moderate for the main outcomes.
Our review found no compelling evidence of a decrease in the rate of blocking of CVCs flushed with heparin compared with CVCs flushed with sterile saline solution, nor of differences in the number of days the catheter lasted, the rate of thrombosis, rate of infection, mortality, bleeding rates or heparin-induced fall in platelet count.
We conclude there is no good evidence that heparin flushing of CVCs is better than flushing with sterile saline solution. As heparin is more expensive, the findings of this review do not support its use except in future clinical trials.
We found no conclusive evidence of important differences when heparin intermittent flushing was compared with 0.9% normal saline flushing for central venous catheter maintenance in terms of efficacy or safety. As heparin is more expensive than normal saline, our findings challenge its continued use in CVC flushing outside the context of clinical trials.
Heparin intermittent flushing is a standard practice in the maintenance of patency in central venous catheters. However, we could find no systematic review examining its effectiveness and safety.
To assess the effectiveness of intermittent flushing with heparin versus 0.9% sodium chloride (normal saline) solution in adults with central venous catheters in terms of prevention of occlusion and overall benefits versus harms.
The Cochrane Peripheral Vascular Diseases Group Trials Search Co-ordinator searched the Specialised Register (last searched December 2013) and the Cochrane Central Register of Controlled Trials (CENTRAL) (2013, Issue 11). Searches were also carried out in MEDLINE, EMBASE, CINAHL and clinical trials databases (December 2013).
Randomised controlled trials (RCTs) in adults 18 years of age and older with a central venous catheter (CVC) in which intermittent flushing with heparin (any dose with or without other drugs) was compared with 0.9% normal saline were included. No restriction on language was applied.
Two review authors independently selected trials, assessed trial quality and extracted data. Trial authors were contacted to retrieve additional information, when necessary.
Six eligible studies with a total of 1433 participants were included. The heparin concentrations used in these studies were very different (10-5000 IU/mL), and follow-up varied from 20 days to 180 days. The overall risk of bias in the studies was low. The quality of the evidence ranged from very low to moderate for the main outcomes (occlusion of CVC, duration of catheter patency, CVC-related sepsis, mortality and haemorrhage at any site).
Combined findings from three trials in which the unit of analysis was the catheter suggest that heparin was associated with reduced CVC occlusion rates (risk ratio (RR) 0.53, 95% confidence interval (CI) 0.29 to 0.94). However, no clear evidence of a similar effect was found when the results of two studies in which the unit of analysis was the participant were combined (RR 0.21, 95% CI 0.03 to 1.70), nor when findings were derived from one study, which considered total line accesses (RR 1.08, 95% CI 0.84 to 1.40). Furthermore, results for other estimated effects were found to be imprecise and compatible with benefit and harm: catheter duration in days (mean difference (MD) 0.41, 95% CI -1.29 to 2.12), CVC-related thrombosis (RR 1.22, 95% CI 0.74 to 1.99), CVC-related sepsis (RR 1.02, 95% CI 0.34 to 3.03), mortality (RR 0.77, 95% CI 0.45 to 1.32) and haemorrhage at any site (RR 1.37, 95% CI 0.49 to 3.85).