The risks of acquiring hospital-acquired infections from wet breathing systems

Keywords::

• bacterial filter
• HME filter
• ventilator circuit
• ventilator colonization

Hospital-acquired infections are a significant cause of mortality [1,2] and morbidity [3], often resistant to standard antibiotic therapy [4], and are a topic of increasing public concern. A general anesthetic usually, and mechanical ventilation of the lungs in intensive care always, involves being connected to a breathing system, an intervention that is associated with the iatrogenic transmission of infection. This risk is increased in susceptible patient groups, including the elderly [5] and those with renal failure [6] or diabetes [7]. In response to this, the Association of Anesthetists of Great Britain and Ireland (AAGBI), the American Society of Anesthesiologists and others have produced guidelines designed to limit the transmission of hospital-acquired infection during anesthetic practice [8,9].

Anesthetic and other breathing equipment has been recognized as a potential vector for the transmission of infection for several decades. Earlier ventilators, with internal circuits that could not be sterilized, were implicated in the patient-to-patient transmission of TB [10]. In the early 1990s, four patients were infected with the hepatitis C virus after being connected to an anesthetic circuit that had been used earlier in the theater list on a patient with pre-existing hepatitis C [11]. In 1996, this case prompted the AAGBI to issue guidance that a filter should be placed between the patient and the breathing system, and that the filter should be changed between each patient [101]. This guidance remains current as of today [8].

Bacterial filters

The use of filters in anesthetic breathing circuits is intended to allow the same breathing system to be re-used for multiple patients, thereby saving cost and reducing waste. There are two main classes of filters, both of which can incorporate heat and moisture exchange elements. Pleated hydrophobic filters act by mechanically preventing particles and microorganisms from passing through due to their small pore size. Electrostatic filters are electromagnetically polarized, thereby using electrostatic forces to repel organisms [12].

The use of filters is not without controversy; filters increase respiratory dead-space and provide resistance to airflow [13]. Current guidelines from the American Society of Anesthesiologists state that there is “insufficient clinical outcome data to support the routine use of bacterial filters for breathing circuits” [9]. A randomized controlled trial in 230 mechanically-ventilated intensive care patients found no difference in the incidence of respiratory infection between patients ventilated using a bacterial filter and those with no filter in place [14].

Colonization of breathing circuits

Numerous studies have demonstrated contamination of breathing circuits on the machine side of the bacterial filter. A Canadian study, in which swabs were taken for culture from the machine and the patient sides of a bacterial filter after an operation, demonstrated that pathogenic bacteria were able to pass through the filter in up to one in 250 cases [15]. Two smaller studies did not, however, replicate the finding of contamination on the circuit side of a filter [16,17]. A more recent study involved ATP bioluminescence to assess contamination of 235 anesthetic filters, a technique borrowed from the food industry, demonstrating bacterial contamination of 9% of filters [18].

Standards for breathing system filters are set by the International Organization for Standardization. This involves a test in which the ability of the dry filter to prevent transmission of aerosolized sodium chloride particles is assessed [19]. However, this test does not replicate common circumstances encountered in clinical practice, during which the filter is moistened with airway secretions and condensation. Recent work in our department suggests that commercially available dry gas filters, whilst complying with requisite international standards [19], do not prevent the transmission of microorganisms in conditions commonly encountered in routine anesthetic practice [20]. In this study, a suspension of Candida albicans and coagulase-negative staphylococci was passed through six filters from three different manufacturers using minimal pressure. All filters allowed free passage of these microorganisms across them.

Hospital-acquired infection & breathing circuits

It is important to recognize that evidence of colonization of anesthetic equipment is not synonymous with, nor does it necessarily lead to, clinical infection presenting in recently anesthetized patients. This would require a pathogenic organism to pass from the respiratory tract of an initial patient (acting as an index case), through the filter to colonize the breathing system, before passing back through a second filter in the opposite direction in order to gain access to the respiratory tract of a second patient.

Demonstrating that colonization of modern breathing systems leads to clinically significant cross-infection is difficult. It has been shown that the level of equipment contamination with Acinetobacter baumannii in intensive care units correlates with infection in patients [21], and so the link is a plausible one. An Acinetobacter outbreak closed down David Scott’s cardiac surgery unit in the early 1990s; the result of water being used as a coupling medium between reusable transducers and their disposable sterile domes. Nevertheless, it remains the case that no case reports in the anesthetic literature exist of convincing cross-infection when new breathing filters are used for every patient, disposable breathing circuits are changed after use on a limited number of patients (in line with the manufacturers’ guidelines), and internal circuits can be sterilized.

Strategies to prevent the transmission of infection

An important question to ask is whether both the circuit and filter should be changed with every patient. In our opinion, this is likely to be unnecessary. In the absence of problems with current AAGBI guidelines (which advise changing the filter, but not the breathing circuit, after every patient), it is probably a waste of resources when anesthetizing patients with normal immune systems or who do not have potentially transmissible respiratory diseases such as TB. Furthermore, the debate about filters may be a distraction from other more pressing issues. The induction of anesthesia involves the gloved hands of the anesthetist holding the patient’s mouth whilst intubating, a procedure which exposes them to respiratory tract secretions and saliva from the patient. At the same time as this, it is necessary to adjust controls of the ventilator and anesthetic machine, and it is often not feasible to change gloves whilst doing this. Such practices contrast unfavorably with the aseptic surgical technique, in which the distinction between sterile and nonsterile equipment is absolute. Current infection control guidelines from the AAGBI suggest that anesthetic machines are cleaned only on a daily basis, but not necessarily between cases [8]. This allows for the possibility that anesthetic machine controls act as a fomite, transferring microorganisms from one patient to another. Recent work in our department has shown that even simple interventions, such as the cleaning of anesthetic equipment between cases, can reduce the incidence of positive cultures of pathogenic bacteria on anesthetic machines [22]. It may be that future strategies for preventing the transmission of pathogenic microorganisms in anesthetic practice should focus on this aspect. As for filters, their use has scientific, ethical and financial implications. Our department and others have demonstrated that bacterial filters may not work under standard anesthetic conditions. Despite this, current practice seems adequate and perhaps we need make no changes.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.