Secondary Logo

Journal Logo


Reuse of a set for total intravenous anaesthesia: safe against bacterial contamination?

Eichler, W.; Schumacher, J.; Ohgke, H.; Klotz, K.-F.

Author Information
European Journal of Anaesthesiology: June 2004 - Volume 21 - Issue 6 - p 501-503


As modern intravenous (i.v.) anaesthetics, such as propofol, are expensive, application of undelivered drug to a second patient could greatly reduce costs. However, propofol is also an excellent medium for the growth of micro-organisms and requires a meticulous sterile-handling technique [1-3]. The vacuum produced during disconnection of syringes and cannulas or tube connections, acting in combination with venous pressure, can cause a reflux of the patient's blood into the syringe or connection line. Total intravenous anaesthesia (TIVA) sets (Fig. 1) were originally developed for two different drugs in two 50 mL syringes to be administered to consecutive patients. Reflux contamination during multiple application is supposed to be prevented by five non-reflux valves and between-use exchange of the line connecting the infusion system with the patient's cannula. In the present study, we investigated whether TIVA sets do indeed prevent reflux contamination in near clinical experimental conditions.

Figure 1
Figure 1:
The TIVA set. These sets allow administration of two i.v. anaesthetics in up to three consecutive patients. The five non-reflux valves (➞) combined with exchange of connection lines between applications are thought to prevent reflux contamination.

To simulate a highly infectious patient, we filled an infusion bag with 400 mL of a liquid culture containing 2.9 × 109 mL−1 of Streptococcus faecium. A culture medium of sterile casein-peptone-soya (CPS) broth was drawn into the connection lines (three per TIVA set) of each TIVA set (n = 15) (Codan, Lensahn, Germany) and the 50 mL syringes (n = 2 per connection line). An injection cannula fitted to the connection line of the TIVA set was inserted into the infusion bag containing the test organisms. The fluid surface level in the infusion bag was maintained at 10 cm above the infusion set to imitate an average patient's venous pressure. A syringe was placed in each syringe pump (Perfusor secura FT®; B. Braun Melsungen, Melsungen, Germany) that was adjusted to a flow rate of 10 mL h−1. The third line was connected to an infusion bag filled with Ringer's solution 500 mL that was infused at a rate of 70 mL h−1.

Procedure 1: After an infusion time of 1 min, the syringes in the pumps were replaced and the TIVA set was separated from the connection line which was then incubated under sterile conditions for 7 days at 37°C. This procedure was repeated twice, after which the TIVA set was also sealed under sterile conditions and incubated together with the syringes. This procedure was performed with 10 TIVA sets.

Procedure 2: To simulate long-term infusion conditions, the total application time was extended to 6 h (120 min for each syringe). The pressure in the infusion bag was increased using a compression cuff - twice to 50 mmHg and once to 150 mmHg for 120 s during each infusion period. This was done to simulate pressure changes that might occur during non-invasive blood pressure (BP) measurements or during venous occlusion for the insertion of further intravenous cannulae. Procedure 2 was repeated five times.

After incubation, all parts filled with CPS broth were examined visually for turbidity that would indicate bacterial growth. To exclude contamination by the investigator, the parts with bacterial growth were plated out to assure that only the test organism had grown. Fluid was collected for plating by puncturing the connection line in the middle and aspirating it into a syringe. Growth of the test organisms occurred in 26 of 30 (86.7%) connection lines in procedure 1, and in 15 of 15 connection lines in procedure 2 (100%). There was no growth of the test organisms in any of the TIVA set components including the syringes in either procedure.

Avoiding reuse use of infusion devices is considered the best measure to prevent transmission of infections in consecutive patients. However, it is still not unusual for the same infusion syringe and infusion system to be used for more than one patient in order to save costs of drugs and consumables [4]. The most common precaution against contamination is to exchange the distal tubing, with or without the use of non-reflux valves. From an economics viewpoint, a hygienically safe infusion device that allows serial infusions in consecutive patients would be highly desirable. We, therefore, investigated a TIVA set that might allow one to break with the practice of single use.

The test method used in the present study is highly sensitive for reflux contamination with the test bacteria. If reflux into a TIVA set had occurred, the high concentration of the test organism (2.9 × 109 mL−1) would enable detection of even a single test organism using the selected enrichment method. The maximum concentration of virus (HCB, HCV) in serum is up to 108 units mL−1. The minimum number for transmitting an infection with the most virulent virus, HCV, is thought to be at least 102 units. But even at the more than 10 times higher concentration of the test organism, not one of the test organisms was found in the TIVA set. The main drawback is that infections by viruses, being much smaller than microbes, may behave differently in the infusion device. On the other hand, contamination in this model most likely occurs due to a retrograde flow. The opening through which this happens compared to the size of either organism would be so large as not to be an impediment. Furthermore bacteria are more likely to show active locomotion and may be able to move upstream, something that viruses cannot do.

Measuring BP increases venous pressure in the affected limb to far above central venous pressure (CVP). Prolonged compression of the arm, for example for insertion of further intravenous cannulae, may cause the venous pressure to approach arterial pressure [5]. Despite this challenge, all TIVA sets were found to be sterile after procedure 2. The connection lines, in contrast, were all contaminated following the same procedure. Even the first short procedure led to contamination of 87% of the connection lines. The results of procedure 2 show that a total infusion time of 6 h combined with the simulation of non-invasive BP measurement invariably caused contamination of the connection line. If a previously used TIVA set is to be used for another patient, it is essential that the connection line be replaced. The actual TIVA sets showed no evidence of microbial contamination during the infusion procedures in this model. However, to ensure this result in practice, the TIVA set must be operated under the same standards of hygiene that applies for any infusion set. Improper infection control precautions among anaesthesiologists and other healthcare workers have been reported [3,6,7]. Whereas patient-to-patient transmission of bacteria is unlikely, improper aseptic handling of syringes, for example, can increase the risk of bacterial contamination. The enrichment method is highly sensitive but does not answer the question of how far the test organisms had spread in the connection line. One cannot exclude the possibility that they could have reached beyond the second non-reflux valve. The high risk of contamination of connection lines makes it dangerous to self assemble infusion systems and to use them in consecutive patients. Although changing the connection lines and the use of non-reflux valves might lower the risks, serial use remains problematical until the effectiveness of these measures has been better clarified. The manufacturer of the TIVA set now, therefore, recommends only single use. From this point of view, we consider a clinical study in infectious patients ethically unacceptable. However, the main risks of contamination remain lapses on the part of healthcare professionals, which can render even the best safety measures ineffective.

In summary, reflux contamination of the studied TIVA set by bacteria did not occur in this model setup, in which proper precautionary measures (sterile-handling technique, exchange of distal tubing) were observed. However, because the first non-reflux valve did not prevent contamination of the connection line and the same event at the next non-reflux valve cannot be excluded, the use of TIVA sets in more than one patient cannot be recommended at present.

W. Eichler

J. Schumacher

H. Ohgke

K.-F. Klotz

Departments of Anaesthesiology, Microbiology and Hygiene; University of Lübeck; Germany


1. Berry CB, Gillespie T, Hood J, Scott NB. Growth of micro-organisms in solutions of intravenous anaesthetic agents. Anaesthesia 1993; 48: 30-32.
2. Webb SA, Roberts B, Breheny FX, Golledge CL, Cameron PD, van Heerden PV. Contamination of propofol infusions in the intensive care unit: incidence and clinical significance. Anaesth Intens Care 1998; 26: 162-164.
3. Bennett SN, McNeil MM, Bland LA, et al. Postoperative infections traced to contamination of an intravenous anesthetic, propofol. N Engl J Med 1995; 333: 147-154.
4. Bach A. Syringe or lead change for TCI? Anaesthesist 1998; 47: 434-436.
5. Halliwill JR, Minson CT, Joyner MJ. Measurement of limb venous compliance in humans: technical considerations and physiological findings. J Appl Physiol 1999; 87: 1555-1563.
6. Berry AJ, Greene E. Universal precautions are not universally practiced by anesthesiologists. Anesth Analg 1995; 81: 205.
7. Henry B, Plante-Jenkins C, Ostrowska K. An outbreak of Serratia marcescens associated with the anesthetic agent propofol. Am J Infect Control 2001; 29: 312-315.
© 2004 European Academy of Anaesthesiology