The clinical relevance of pulmonary infection caused by microbial contamination of anaesthesia tubing systems is still controversial [1–5]. After surgery only in a few cases can pneumonia be traced back to a contaminated breathing system when no protective devices were used [6–8]. Nevertheless, the necessity for minimizing the potential risk of cross-infection is generally accepted. Hygienic recommendations extend from using sterile disposable tubes, desinfection/sterilization of the whole circle system or at least some of its parts after each use to single use of breathing system filters (BSF) between the endotracheal tube and the Y-piece, and cleaning the systems at the end of the day [9–13]. To maintain the physiological clearance function of mucosa and upper airways, BSF can be combined with heat and moisture exchangers (HMEF). Some models contain electret material enhancing the filtration efficiency by incorporating fixed electrical charges. Compared with common depth filters, electrets are less dense at equivalent filtration capability for airborne particles resulting in lower flow resistance . However, it is still uncertain whether the performance of electrets is reduced by the presence of condensed water, which is abundant in rebreathing circle systems. In a randomized controlled clinical study we investigated the contamination of anaesthesia breathing systems, comparing different types of HMEF with electret filter (DAR Hygrobac, Hygrobac S and Medisize Hygrovent S), and the contamination when no filtering device was used.
Patients and material
This prospective controlled study was approved by the local ethics committee. In a first step 150 consenting males and females undergoing elective surgery were assigned at random to one of 3 groups with differently configured anaesthesia breathing systems. In group 1 (control group) no HMEF was used. In group 2 and 3 different HMEF with the same electret filter material (Filtrete 200G®, 3 m, Breada, Netherlands) were placed between the Y-piece and the flexible endotracheal tube connection (10 cm length). The HMEFs were previously selected for their excellent heat and moisture exchanging properties . They only differed in the heat and moisture exchanging material, consisting either of a calcium chloride (CaCl2) coated paper roll (DAR Hygrobac S®, DAR, Mirandola, Italy, group 2) or a CaCl2 coated natural sponge (DAR Hygrobac®, DAR, Mirandola, Italy, group 3). In a second step another 100 HMEF (Hygrovent S®, Medisize, Hillegom, The Netherlands) with a calcium chloride coated paper roll and electret filter (Filtrete 200G®, 3 m, Breda, Netherlands) were investigated under routine clinical conditions as a confirmation group (group 4).
The clear PVC tubes (165 cm length, 22 mm internal diameter, smooth inner surface) were reinforced with an external spiral and glued to the 22 mm ports of the Y-piece. Side-stream CO2 monitoring was conducted with a PVC tube (2.5 m length, 1 mm internal diameter) connected either with a T-adapter between the Y-piece and the flexible connection (group 1) or with a luerlock connection on the distal side of the HMEF (group 2, 3 and 4). Inspiratory and expiratory circle system connecting ports of the breathing tubes were hygienically separated with common depth filters (DAR Sterivent®, DAR, Mirandola, Italy). All breathing systems (with or without HMEF) were assembled by the supplier and delivered in sterile conditions. The packaging was opened shortly before use.
After tracheal intubation, the patient's lungs were ventilated using one of the tubing systems previously described. The anaesthesia ventilator, manual ventilation bag and all the gas-carrying tubes of the circle system (Dräger 8 ISO, Dräger, Lübeck, Germany) were decontaminated according to the manufacturers specifications before use. All patients received balanced anaesthesia with fentanyl and additional volatile anaesthetics (halothane or isoflurane). After denitrogenization, ventilation (FiO2 0.3) was performed with reduced fresh gas flow. CO2 was absorbed with soda lime which was recharged shortly before anaesthesia. The membranes of the HMEFs were orientated in a vertical position. The worst case situation was considered to be the accumulation of condensation on both sides of the HMEFs. At the end of anaesthesia the tubing system was sterile plugged and transferred to the microbiological laboratory for further investigations.
Withdrawal and incubation of samples (see Fig. 1)
Tracheal aspirate (TA) (all groups)
After anaesthesia, tracheal secretion was obtained by sterile endotracheal aspiration, the catheter tip was cut off and incubated for 24 h at 37°C in a test tube with 10 mL of sterile nutrition broth (bouillon, pepton, dextrose, NaCl 0.9%, pH7).
Swabs from breathing system tubes (P1, P2, P3) (not performed in group 4)
Sterile swabs with 0.9% NaCl solution were used to collect samples from the inside the flexible connector (P1), distal (machine) side of the HMEF (P2, if applicable), and inner walls of the Y-piece (P3). All samples were incubated for 24 h at 37°C in test tubes with 10 mL of nutrition broth (see above).
Rinsing of breathing system tubes (P4) (all groups)
The inspiratory and expiratory tubes and the Y-piece were filled with 250 mL of sterile 0.9% NaCl-solution and sterile-plugged at the Y-piece. After carefully rinsing the whole tubing system the solution was filtered through a 0.45 μm cellulose-nitroacetate filter membrane (Sartorius, Göttingen, Germany). The filter was incubated for 24h at 37°C in 100 mL nutrition broth (see above).
Detection of microorganisms
Cultures of all samples were grown on Columbia agar with 5% sheep blood, Endo agar and Sabouraud agar with gentamicin and chloramphenicol (bio-Mérieux, Marcy l'Étoile, France) and incubated for at least 24 h at 37°C. Positive results were qualitatively tested with standard morphological, biochemical and serological methods:
- macroscopic evaluation of growth at 4°C and 42°C;
- microscopic evaluation after Gram staining (magnification 800 ×);
- oxidase evaluation with Patho Tec-C (Organon Technika, Eppelheim, Germany);
- detection of ferments with api 20 STREP, ID32 STAPH (bio-Mérieux, Marcy l'Étoile, France), enterotube 2 Roche, mycotube Roche, oxi/Ferm tube 2 Roche (Hoffmann-La Roche AG, Grenzach-Wyhlen, Switzerland);
- Mannit-Kochsalz-Phenolrot-Agar (Merck, Darmstadt, Germany);
- plasmacoagulase with staphslide-test (bio-Mérieux, Marcy l'Étoile, France);
- Lancefield classification with slidex strepto-Kit A.B.C.D.F.G (bio-Mérieux, Marcy l'Étoile, France).
The discovery rate of tracheal aspirate microorganisms in the tubes was compared between group 1 (control group) and groups 2 and 3 (with HMEF). Differences were statistically evaluated with the Mann–Whitney U-test as a nonparametric test method (P<0.05). The data of group 4 were used for confirmation.
Age, gender, duration of anaesthesia, fresh gas flow and fraction of rebreathing were distributed equally among the randomized groups (group 1, 2 and 3, see Table 1). No patient had fever at the start of surgery. The number of patients who had received antibiotic treatment for pneumonia within 3 months before or continuously until the day of operation is shown in Table 1. There was a significantly higher incidence of patients with a history of pneumonia (positive X-ray, leucocytosis >10000 μL−1, fever >38°C) before surgery in group 1 than in group 2 and 3 (P<0.005). However, none of these patients showed persistent X-ray signs or leucocytosis. On the other hand, significantly more patients with leucocytosis but no history of pneumonia were seen in group 2 and 3 (P<0.005). Other known pulmonary diseases such as Chronic Obstructive Pulmonary Disease (COPD), bronchial carcinoma, bronchopulmonary fistula or atelectasis were equally distributed among the randomized groups. Regardless of the patients' history, most received antibiotic prophylaxis with cephalosporine (cephazoline) intra-operatively.
The data in the confirmation group (group 4) did not appear to vary; however, they were not compared statistically with the randomized groups.
Migration of microorganisms
At the end of anaesthesia visible condensate was seen in the inspiratory and expiratory tubes of all patients and on both sides of the HMEFs, if installed.
In 112 patients (75%) 26 different bacterial species were found in the tracheal aspirate (see Table 2); there were no significant differences among the three randomized groups (150 patients). In group 1, which had no filter, the tracheal aspirates of 38 patients (76%) showed positive results. In 13% of these patients (five patients) identical species were not only detected in the flexible connector but also at different locations in the anaesthesia tubing system. The distribution of the species was neither homogeneous nor did it follow any systematic correlation with the duration of anaesthesia or pre-operative pulmonary disease. In group 2 and 3 the incidence of microorganisms in the tracheal aspirate was similar to group 1: 40 (82%) and 34 (67%) positive findings, respectively. In contrast with the results in group 1, no migration of tracheal microorganisms across the HMEF into the tubes was detected in either group with HMEFs between flexible connector and Y-piece. The differences between the three randomized groups were statistically significant (P<0.005).
The results in group 4 confirmed these findings: indeed, 72% of the 100 patients showed positive results in the tracheal aspirate, but no migration into the tubes via the HMEF could be proved with one exception: in this case the CO2 adapter had been moved from the patient side to the machine side of the HMEF during anaesthesia.
In all groups bacterial species which were not related to the tracheal findings were sporadically detected at different locations: aerobic spore-forming species, S. epidermidis and aerococci in group 1, S. epidermidis and aerococci in group 2, lactococci cremoris, aerococci and not serogrouped streptococci in group 3, aerobic spore-forming and aerococci species in group 4.
The results demonstrate the high incidence of bacterial contamination of breathing systems if no filter devices are used: in the control group microorganisms originating from the patients' airways were able to migrate into the tubing system in 13% of the cases. The existing risk was additionally confirmed by one case in group 4, where the CO2 connector was moved mistakenly from one side of the HMEF to the other. There was neither a tendency for a particular species to be detected nor was detection dependent on the duration of anaesthesia, or the preexistence of pulmonary disease. Additionally it must be assumed that the warm and moist conditions in anaesthesia circle systems provide excellent conditions for the persistence and reproduction of microorganisms. Therefore the necessity for effective contamination prevention is obvious for the avoidance of further cross-contamination.
There are different approaches to the prevention of cross infection caused by contaminated breathing systems. The most common is the decontamination or sterilization of reusable breathing systems or at least parts of them. The use of clean disposable tubes is equivalent and reduces the costs of decontamination, but it must be regarded as an environmental problem. A less costly alternative may be the use of breathing system filters between the endotracheal tube and the breathing system: BSFs not only prevent initial contamination of the tubing system but also prevent colonization of the patient's airways if a contaminated breathing system happens to be used. If the filtration efficiency is high, the breathing system need not be decontaminated or changed after each use [13,16]. BSF with additional heat and moisture exchanger properties (HMEF) have a supplementary benefit for the patient: by almost instantly warming and humidifying the inspired air, they maintain the physiological mucociliar clearance and integrity of the mucus which is fundamental for the intact airway's infection defence mechanisms [15,17–19]. Two types of filter materials are commonly used: coated glass fibres ('hydrophobic' filters) have proved to be highly effective—even preventing the passage of aggregated and contaminated water. Split polypropylene fibres with fixed electrical charges ('electret' filters) show lower flow resistance , however, the filtration efficiency in clinical use has not been established. Furthermore, the less dense web structure does not prevent condensed water from being forced through the filter. Thus, although there is no risk of blockage, it might be possible for water to penetrate the filter carrying microorganisms. Such an adverse situation exists, when HMEFs are used in anaesthesia circle systems with considerable amounts of condensate on both sides of the filter.
This study shows that the investigated HMEFs effectively prevent migration of the patient's tracheal microorganisms into the breathing system. Nevertheless, it should be remembered that absolute protection cannot be achieved by any one hygienic measure, since there are many other possible sources of contamination. This is indicated by the occurence of sporadically detected microorganisms which have no correlation with the tracheal secretion species, i.e. airborne or cutaneous. They presumably originate from secondary contamination from external sources.
Although the tested HMEFs showed a similar filtration efficiency under clinical conditions, the results cannot generally be applied to all types of HMEFs. Because the results of other investigations conflict [20–23], it is essential to establish laboratory tests for the evaluation of filtration properties of BSF/HMEF under comparable conditions.
1 Bengtson JP, Brandberg A, Brinkhoff B, Sonander H, Stenquist O. Low-flow anaesthesia does not increase the risk of microbial contamination through the circle absorber system. Acta Anaesthesiol Scand
2 Dryden GE. Risk of contamination from anesthesia circle absorber: an evaluation. Anesth Analg
3 Du Moulin GC, Saubermann AJ. The anesthesia machine and circle system are not likely to be sources of bacterial contamination. Anesthesiology
4 Feeley TW, Hamilton WK, Xavier B, Moyers J, Eger EI. Sterile anesthesia breathing circuits do not prevent post-operative pulmonary infection. Anesthesiology
5 Pandit SK, Metha S, Agarwal SC. Risk of cross-infection from inhalation anaesthetic equipment. Br J Anaesth
6 Beck A, Zadeh AJ. Infection by anaesthetic apparatus. Lancet
7 Nielsen H, Jacobsen JB, Stokke DB, Brinklov MM, Christensen KN. Cross-infection from contaminated anaesthetic equipment. Anaesthesia
8 Olds JW, Kisch AL, Eberle BJ, Wilson JN. Pseudomonas aeruginosa
respiratory tract infection acquired from contaminated anesthesia machine. Am Rev Respir Dis
9 Deverill CEA, Dutt KK. Methods of decontamination of anaesthetic equipment: daily sessional exchange of circuits. J Hosp Infect
10 Lumley J. Decontamination of anaesthetic equipment and ventilators. Br J Anaesth
11 Parmley JB, Tahir AH, Dascomb HE, Adriani J. Disposable vs. reusable rebreathing circuits: advantages, disadvantages, hazards and bacterial studies. Anesth Analg
12 Roberts RB. The anaesthetist, cross infection and sterilization techniques—a review. Anaesth Intens Care
13 Shiotani GM, Nicholes P, Ballinger CM, Shaw L. Prevention of contamination of the circle system and ventilators with a new disposable filter. Anesth Analg
14 Baumgartner H, Löffler F, Umhauer H. Deep-bed electret filters—the determination of single fibre charge and collection efficiency. IEEE Trans Elec Ins
15 Rathgeber J, Züchner K, Kietzmann D, Weyland W. Heat and moisture exchangers for humidifying the inspired air of intubated patients in intensive care medicine. An investigation of humidification properties of HME under clinical conditions. Anaesthesist
16 Borghi V, Ferrari E, Vacondio P, Meacci M, Squadrini F. Bacterial removal efficiency of filter-HME for anaesthesia and intensive care in 24-hour tests. Acta Anaesthesiol Ital
(Suppl. I): 93–97.
17 Chalon J, Loew DAY, Malebranch J. Effects of dry anaesthetic gases on tracheobronchial ciliated epithelium. Anaesthesiology
18 Forbes AR. Temperature, humidity and mucus flow in the intubated trachea. Br J Anaesth
19 Kleemann PP. The climatisation of anesthetic gases under conditions of high flow to low flow. Acta Anaesthesiol Belg
20 Bygdeman S, Euler C, Nyström B. Moisture exchangers do not prevent patient contamination of ventilators. A microbiological study. Acta Anaesthesiol Scand
21 Hedley RM, Allt-Graham J. A comparison of the filtration properties of heat and moisture exchangers. Anaesthesia
22 Luttropp HH, Berntman L. Bacterial filters protect anaesthetic equipment in a low-flow system. Anaesthesia
23 Saravolatz LH, Pohlod DJ, Conway W, Haberaecker W, Markowitz NP, Popovich Jr J. Lack of bacterial aerosols associated with heat and moisture exchangers. Am Rev Respir Dis