Skip navigation

Focus on Drips

Many patients who receive invasive medical treatment or diagnostic investigations undergo subcutaneous venous cannulation. In some situations a venous cannula may be in place for a few minutes or hours, as for patients undergoing general anaesthesia, sedation before surgery, noxious clinical procedures, or diagnostic radiological investigations. However, in some clinical situations long term cannulation for weeks or months may be required - in the critically ill or cancer patients receiving long term opiate infusions for pain relief or chemotherapy.

Patients and nursing staff are only too well aware of the negative aspects associated with venous cannulation, including pain and discomfort and infection. These may be underestimated and under-reported. Patients are known to complain of pain and discomfort not only during the actual procedure itself, but whilst the cannula is in place and symptoms may continue weeks and months after it has been removed. Venous cannulation is a common procedure which appears to have been the subject of few RCTs. It is an ideal subject for nursing research, as despite traditionally being a medical responsibility it is now increasingly done by nurses in their extended role.

RCT search

Bandolier , in its search for evidence on things that are common, has this month conducted searches for randomised controlled trials (RCTs) for venous cannulation. MEDLINE was searched from 1991, using the terms CATHETERISATION PERIPHERAL, RANDOM* and VENOUS and VENOUS CANNULATION and RANDOM*. This brought to light some interesting studies on this subject, and the most interesting are reviewed.

Patient position and faint

Some patients undergoing venous cannulation faint. Can the proportion who faint be altered by simple things like sitting or lying? One study randomised 300 pre-surgical patients aged 18 to 40 years to a lying or sitting position during insertion of a venous cannula [1]. Blood pressure and heart rate were monitored and recorded before, during and for 6 minutes after the procedure.

The incidence of vasovagal symptoms (nausea, dizziness, sweaty, pallor, hot, cold, syncope) was 12.6% (95%CI 7.4 - 17.7%; 20/159) in the sitting patients and 2.1% (95%CI 0.0 - 4.5%; 3/141) in the recumbent patients. Two sitting patients (1.3%) experienced frank syncope. Symptomatic patients were more likely (39.1%) to have had a previous history of fainting than asymptomatic patients (8.3%).

If all patients underwent cannulation in the recumbent position, 87% of all vasovagal reactions could be prevented.

Local anaesthetic for cannulation?

Pain on cannula insertion depends to some extent on needle size - but subcutaneous injection of local anaesthetic itself causes pain. Is there a point at which the pain from injection of anaesthetic is worse than the pain from venous cannulation?

Sixty patients about to undergo general anaesthesia were randomised to receive intravenous cannulation with Venflon 18, 20 and 22 gauge cannulae [2]. One hand was cannulated without anaesthetic and the other hand received a subcutaneous injection of 1% lignocaine, and this also was randomised. Patients had their eyes closed, and were asked to grade which was the more painful.

This elegant experiment clearly demonstrated that irrespective of needle size, most patients would benefit from local anaesthetic before venous cannulation.

In another study topical amethocaine cream was compared to EMLA cream for pain relief, when given before insertion of a venous cannula [3]. The results showed that both treatments gave good analgesia with median visual analogue pain intensity scores of about 10 mm (of 100 maximum possible).

Helpful heparin?

Does heparin, when added to the fluids infused through a peripheral catheter, reduce local catheter-related problems and extend catheter life?

A positive result for both questions came from a well conducted study of a trial which examined the effects of heparin (final concentration 1 unit/mL) or saline added to infusion fluids just before they were given to patients. This study [4] fails to say that it was randomised, though a detailed study of the methods indicates that it almost certainly was. Indistinguishable flasks of heparin and saline were given to nurses to add to intravenous fluids except phenytoin, amiodarone or aminoglycoside antibiotics where there was incompatibility with heparin.

The results showed that catheters had a longer life with heparin (by an average of 33 hours {95%CI 9 - 56 hours}), that there were fewer complications, and that when complications occurred with heparin that happened an average of 43 hours later than with saline (95% CI 6 - 80 hours).

Butterfly needle or Teflon cannula?

A 25 gauge butterfly needle has been tested against a Teflon cannula with a 26 gauge introducer needle and 24 gauge Teflon cannula in a crossover RCT [5] in palliative care for subcutaneous opiate infusions. Patients were randomised to receive either the butterfly at 45 degrees or the Teflon cannula at 90 degrees in the subclavicular or abdominal regions. When there were signs of toxicity - redness, swelling, tenderness, bruising or leaking - the alternate needle type was used at a different site.

Twenty patients completed both parts of the study. The mean duration of the butterfly was 5.3 days and for the Teflon cannula was 11.9 days. Patient and nursing preferences were both heavily in favour of the Teflon cannula. The Teflon cannula, though, could not cope with large volumes of infusate (>30-50 mL/hr), and cost more.

Antibiotic bonding reduces intravascular catheter infection

Perhaps the most interesting RCT came from a JAMA article in 1991 [6]. Catheters were pre-treated with cationic surfactant and then an anionic antibiotic cefazolin was bonded onto the surface before insertion. The treated catheters were compared with untreated catheters in central venous sites in 93 patients and arterial catheters in 85 patients in an ITU.

Catheters were removed on the seventh day or earlier if not needed, after decontamination of the site of insertion with 70% alcohol, and then sent to the microbiology laboratory in sterile tubes. Microbial growths were sought on 2 cm sections by staff blind to the nature of the catheter.

There were 81 control catheters, 14% of which were infected. There were 97 antibiotic catheters, of which 2% were infected. Antibiotic bonding significantly reduced the cumulative risk of infection of the catheter.

The message seems to be lie down, use local anaesthetic, use heparin and Teflon cannulae for long-term use.

Dawn Carroll
Research Sister, Pain Relief Unit, Oxford


  1. SE Rapp, J Pavlin, ML Nessly, H Keyes. Effect of patient position on the incidence of vasovagal response to venous cannulation. Archives of Internal Medicine 1993 153: 1698-1704.
  2. N Harrison, BT Langham, DG Bogod. Appropriate use of local anaesthetic for venous cannulation. Anaesthesia 1992 47: 210-2.
  3. J Molodecka, C Stenhouse, JM Jones, A Tomlinson. Comparison of percutaneous anaesthesia for venous cannulation after topical application of either amethocaine or EMLA cream. British Journal of Anaesthesia 1994 72: 174-6.
  4. JA Nieto-Rodriguez, MA Garcia-Martin, MD Barreda-Hernandez, MJ Hervàs, O Cano-Real. Heparin and infusion phlebitis: a prospective study. Annals of Pharmacotherapy 1992 26: 1211-4.
  5. K Macmillan, E Bruera, N Kuehn, P Selmser, A Macmillan. A prospective comparative study between a butterfly needle and a Teflon cannula for subcutaneous narcotic administration. Journal of Pain and Symptom Management 1994 9: 82-4.
  6. GD Kamal, MA Pfaller, LE Rempe, PJR Jebson. Reduced intravascular catheter infection by antibiotic bonding. A prospective randomised controlled trial. Journal of the American Medical Association 1991 265: 2364-8.

next story in this issue