Water sampling for Legionella can be useful in helping assess risks and in determining whether or not preventive and corrective measures are working. Having an action plan based on results of Legionella sampling can alert you to increased risks and whether or not disinfection procedures should be implemented. Not sampling obviously tells you nothing about your programs -- until a case of LD occurs. The aforementioned sounds quite logical and simple, and is an assumption that Legionella sampling should be routine for any monitored system. Such is not the case.

Legionella experts have debated the issue of sampling and routine testing for years. CDC advocates sampling after LD has been found (confirmed) so as to locate the source of legionellae that caused it and take remedial action. They do not encourage sampling in the absence of confirmed LD cases. Other experts disagree with this and advocate a more proactive approach of conducting periodic sampling (so-called "routine sampling"), even if no cases of LD have been detected.

What has stood in the path of any real consensus being achieved amongst the experts are the following current facts and understanding of Legionella and LD:

There is no specific infectious density known for Legionella or clearly established correlation between test culture or direct fluorescent antibody (DFA) test results of Legionella and risk of contamination.

Legionella is frequently present in water supplies without causing disease, so routine testing and obtaining a positive count does not mean LD will occur and may even produce a false sense of alarm and lead to costly corrective actions being undertaken. On the other hand, obtaining negative results does not assure that LD cannot occur and may even provide a false sense of security and lead to relaxation of preventive maintenance.

Interpretation of results in routine Legionella sampling is still questionable due to:

(a) use of different bacteriologic methods amongst laboratories;

(b) variable results between culture and direct fluorescent antibody methods;

(c) variable culture results from differing sites within the same system; and

(d) variations in the counts of legionellae isolated from a single site. In addition, potentially infectious Legionella in some water samples may not grow on the microbiological medium specifically formulated to grow Legionella.

The risk of illness following exposure to a given Legionella source is influenced by a number of variables and factors other than just the concentration of organisms in a sample. Host susceptibility, Legionella strain virulence, and efficiency of legionellae transmission (to host) are integral to LD risks and disease progress.

Routine testing can be a double-edged "liability sword:" testing and getting positive results may establish a legal liability if a disease case occurs; yet the testing may prevent negligence charges from applying. However, not testing in the presence of other factors may leave you guilty in the face of defending a responsible LD case.

Those in favor of routine sampling believe that since the risk of Legionnaires’ Disease is greater when there are high levels of legionellae in water, it makes sense to take measures that will minimize legionellae in water and to check legionellae levels periodically to make sure the preventive measures are working. They contend that sampling results, although sometimes inconclusive, may at other times provide life-saving information.
Most experts, however, including those in the "anti-routine" sampling group, would agree that there should be a consideration to routinely test any system that presented enough LD risk factors or other pertinent information. This would be based on a thorough review and assessment of the system, its operations and surroundings. Relevant factors would include: raw water quality, system design and operations, fouling history and potential, host population and susceptibility, LD case history, etc.

James R. Watson, Ph.D., provides the following Legionella sampling and testing information in his LEGIONELLA UPDATE – 2000 as reported by Microbiological Consultation Services, Inc. (MCS):

"Although small numbers of legionella bacteria probably pose a very low risk to healthy individuals, corrective action should be kept in mind whenever legionellae are isolated from a water sample. Although the scientific community cannot agree on what number of legionella bacteria is acceptable, we believe colony counts, as expressed in colony forming units (cfus) per ml of water, can be used as a loose guide for deciding when to implement corrective action. Legionella pneumophila colony counts for cooling tower specimens may be interpreted as follows:

>100 cfus/ml = large number of legionella bacteria.
10-100 cfus/ ml = moderate number of legionella bacteria.
<10 cfus/ml = small number of legionella bacteria."