Heparin or 0.9% sodium chloride to maintain central venous catheter patency: a randomized trial.

OBJECTIVE To compare heparin (3 mL, 10 units/mL) and 0.9% sodium chloride (NaCl, 10 mL) flush solutions with respect to central venous catheter lumen patency. DESIGN Single-center, randomized, open label trial. SETTING Medical intensive care unit and Surgical/Burn/Trauma intensive care unit at Barnes-Jewish Hospital, St. Louis, MO. PATIENTS Three hundred forty-one patients with multilumen central venous catheters. Patients with at least one lumen with a minimum of two flushes were included in the analysis. INTERVENTIONS Patients were randomly assigned within 12 hrs of central venous catheter insertion to receive either heparin or 0.9% sodium chloride flush. MEASUREMENTS AND MAIN RESULTS The primary outcome was lumen nonpatency. Secondary outcomes included the rates of loss of blood return, inability to infuse or flush through the lumen (flush failure), heparin-induced thrombocytopenia, and catheter-related blood stream infection. Assessment for patency was performed every 8 hrs in lumens without continuous infusions for the duration of catheter placement or discharge from intensive care unit. Three hundred twenty-six central venous catheters were studied yielding 709 lumens for analysis. The nonpatency rate was 3.8% in the heparin group (n = 314) and 6.3% in the 0.9% sodium chloride group (n = 395) (relative risk 1.66, 95% confidence interval 0.86-3.22, p = .136). The Kaplan-Meier analysis for time to first patency loss was not significantly different (log rank = 0.093) between groups. The rates of loss of blood return and flush failure were similar between the heparin and 0.9% sodium chloride groups. Pressure-injectable central venous catheters had significantly greater rates of nonpatency (10.6% vs. 4.3%, p = .001) and loss of blood return (37.0% vs. 18.8%, p <.001) compared to nonpressure-injectable catheters. The frequencies of heparin-induced thrombocytopenia and catheter-related blood stream infection were similar between groups. CONCLUSION 0.9% sodium chloride and heparin flushing solutions have similar rates of lumen nonpatency. Given potential safety concerns with the use of heparin, 0.9% sodium chloride may be the preferred flushing solution for short-term use central venous catheter maintenance.