• It remains unknown what impact saline flushing might have on venous physiology and injury.

  • Repeated vigorous flushing could cause endothelial injury, clot formation and catheter failure.

  • Flushing frequency (hourly vs. 5 hourly) did not impact catheter failure over 5 hours.

  • Frequent flushing increased prothrombin time, suggesting extrinsic coagulation factor consumption.

  • Sub-clinical thromboses formed adjacent to ~80% of PIVCs, irrespective of flushing frequency.


Peripheral intravenous venous catheters (PIVCs) are associated with a postinsertion failure incidence of 40%, yet the common maintenance and preventive strategy of saline flushing is poorly understood at a physiological level.


We developed a human model of bilateral cephalic vein cannulation to study the impact of varied PIVC flushing frequency (high frequency, HF; low frequency, LF) over 5 hours on catheter failure (primary outcome), coagulation, platelet aggregation, and local tissue injury. Ultrasound was used in a subset to assess vascular diameter/catheter to vein, blood flow velocity, and thrombus formation.


Out of 34 catheters in 17 adult participants, 1/17 (6%) LF catheters failed, which was not significantly different from HF catheters (0/17). Platelet function, activated partial thromboplastin time, and tissue factor were also not different (P > 0.05). However, prothrombin time (PT) increased with HF versus LF after 5 hours (P < 0.05). Ultrasound demonstrated luminal thromboses in veins experiencing both HF (6/7) versus LF (5/7) conditions after 5 hours with nonsignificant changes in vascular diameter and blood flow velocity over time. Conclusions: Although no difference in PIVC failure was observed between HF and LF flushing conditions over 5 hours, greater flushing frequency increased PT time, suggesting delayed activation or consumption of extrinsic coagulation factors. This study also demonstrated feasibility in assessment of luminal thromboses, which were remarkably prevalent after PIVC placement, and changes in vascular diameter and blood flow. This manuscript illustrates that the development of a sensitive human model will be of great use for exploring the impact of interventions on reducing PIVC failure in the future.

This content is only available as a PDF.
You do not currently have access to this content.