By John Marx, MPH
Exposure to inhaled bacteria of the Legionella genus can result in two disease entities: Pontiac Fever, a self-limiting illness resulting from presumed immune reaction to inhaled bacterial antigens, and Legionella pneumonia (Legionnaires’ disease). Unlike Pontiac Fever, Legionella pneumonia is an active infection and has a case fatality rate of about 15 percent.
Legionella bacteria are naturally found in water and cases of Legionella pneumonia have been associated with aerosolization from cooling towers, evaporative condensers, misters, whirlpools, potable water, water fountains, humidifiers and respiratory therapy equipment. Rare cases have been attributed to aerosolized potting soil.
Since person-to-person spread does not occur with Legionella pneumonia, it remains one of the very few healthcare-associated infections (HAIs) linked exclusively to an environmental source – in this case, water. Unfortunately, the advantage of focusing prevention efforts on this defined reservoir is offset by the ubiquity of water in healthcare settings and the varied ways in which water systems may interact with the provision of patient care.
Experience has been an invaluable teacher as we have addressed the presence of Legionella in our hot potable water and a decorative water feature, two reservoirs which have challenged us in protecting patients from legionellosis. Following are a few points which experience has taught us over the past two decades.
Potable water is the most common source of healthcare-associated legionellosis. While direct inoculation of Legionella contaminated tap water into wounds, and ingestion of Legionella contaminated ice with subsequent aspiration have been reported as routes of infection, inhalation of Legionella species aerosolized from sink or shower taps is the presumed route of infection for most hospital-acquired cases.
Legionella are tolerant of chlorine levels added to drinking water by many municipalities and small numbers may be present even when municipal water treatment is functioning as designed. The extensive plumbing required to deliver water throughout a hospital building and its many rooms provides a staggering internal surface area with potential for biofilm development and accumulation of scale and it is within these areas that Legionella densities are highest. Areas of stagnation and sediment accumulation can enhance the environment in which Legionella can proliferate, particularly in areas where water temperature is between 25 degrees C and 42 degrees C.
In the mid-1990s we experienced a cluster of Legionella pneumonia cases among patients which molecular subtyping showed was linked to our hot potable water system. At that time, there was no recommendation to culture for Legionella in potable water since contamination levels do not correlate directly with incidence of disease. In other words, there was no defined threshold number of organisms in water above which intervention was required. Also at that time, there was no recommendation for use of copper/silver ionization as a control measure in hospitals due to a lack of long-term experience with metal ion generating technology in this setting.
In response to the cluster, we immediately performed two cycles of chlorine disinfection of the hot potable water system, each of which suppressed Legionella numbers for about a month but which did not reduce the proportion of sampling sites from which these organisms could be recovered. Hyperchlorination had knocked the total bacterial numbers down and immediately reduced risk, but did nothing to eradicate Legionella from any portion of the plumbing system.
It was not until a copper and silver ionization system was installed the following month that we saw a precipitous decline in Legionella numbers. This system releases low levels of metal ions, which are within drinking water limits, into the hot water system. We were able to recover very low levels of Legionella over the course of the next few years as we optimized operation of the system. While we have reduced the frequency of testing, Legionella has not been recovered from our hot potable water system in over a decade. Most importantly, we have not had a single case of hospital-acquired legionellosis since the system was installed.
Lessons about potable water treatment that we’ve learned the hard way:
- When official guidance on assessing or remediating a particular risk lags, or available guidance sources are conflictive, you may have no choice but to seek expert opinion based on the best evidence currently available to guide your decisions.
- Should a cluster of legionellosis occur in your facility, having a contingency plan in place ahead of time will expedite response.
- Temporary hyperchlorination of potable water can immediately reduce legionellosis risk, but is a short term solution.
- New technologies may be surprisingly effective, but still require monitoring to ensure that they are functioning properly.
Decorative Water Features
Unlike the provision of potable water, a decorative water feature is a completely non-essential addition to a healthcare environment and should always be viewed as such when weighing the risks and benefits of installing one. Like the potable water system mentioned above, water features can provide an environment suitable for Legionella to proliferate to high levels.
Decorative water features, notably water fountains and water walls, have been linked to legionellosis in both public settings and also within healthcare facilities. What we once envisioned as risk associated with relatively large fountains with obvious spray and aerosolization has now given way to the disconcerting knowledge that even small water features and those with minimal obvious aerosolization have been linked to legionellosis.
Also concerning is the observation that Legionella infections associated with decorative water features have occurred among members of the comparatively healthy general public. A water feature in a healthcare facility brings this potential source of infection into proximity with patient populations with underlying conditions and risk factors which increase their risk of acquiring Legionella pneumonia.
Taking current evidence into account, it becomes clear that if a healthcare facility chooses to install a decorative water feature, a number of obligations are incumbent upon the facility to manage this risk. Among those I would cite are the following:
- Maintain an automated disinfectant injection system.
- Select a disinfectant with a residual effect (e.g., bromine) to ensure that effective disinfectant levels are maintained in all areas where the water is circulated or collects. A single pass disinfecting system such as ultraviolet light can be used as an adjunct, but under no circumstances can it replace a disinfectant with residual effect.
- Select a disinfectant for which testing can be performed to readily monitor actual water levels.
- Monitor levels of disinfectant in the water at least weekly.
- Culture water monthly to verify disinfection efficacy.
As the headline of this article suggests, experience is the best teacher and the recommendations above are what we’ve gleaned after a number of years and with the advantage of 20/20 hindsight. A decorative water feature can be safely maintained, but not without vigilance, and the costs incurred to ensure safety over the life of the water feature should not be overlooked.
Our institution currently has one large fountain in place with an automated bromine injection system. Three smaller water features are indefinitely off-line due to our inability to ensure safe operation in their current configuration. We have not permitted installation of any new water features in our institution for over a decade.
Lessons about water features that we’ve learned the hard way:
- If disinfectant is manually added to a water feature at intervals, it will be forgotten at some point.
- If disinfectant addition is automated, the automated system will fail at some point.
- Disinfection failure will occur and if this failure is not recognized quickly by documenting a drop in disinfectant levels in the water, the final safety net to identify and address the increased risk is likely to be identifying high levels of bacteria in water samples obtained for culture.
- After the initial cheers engendered by installation of a water feature have quieted after a few months, the obligation for continuous automated disinfection, monitoring of disinfectant levels and microbiologic sampling will continue for the entire lifetime of the water feature.
- Give credit where it is due: It is crucial to maintain adequate support for, and communication with, facility staff who maintain building water systems. Ultimately, it is the man or woman with the wrench and the practical understanding of these systems who keeps them operating safely, and are often the first to recognize when something is amiss. While often overlooked for kudos, these are the folks who keep us all out of hot water.
John Marx, MPH, is the senior infection control practitioner and microbiologist at University of Wisconsin Hospital and Clinics. He has more than 20 years of experience in environmental microbiology and infection prevention.