Ashley Silva of Sanford, NC, a water treatment professional with 20 years experience, submitted the following:

1. No mention was made in the article at all about using chlorine dioxide as a combatant for Legionella.

2. Chlorine dioxide is the most effective biocide available for Legionella and biofilm, which needs to be removed for proper scale/corrosion/microbiological control in a cooling system.

3. High doses of chlorine are corrosive to most cooling systems' metallurgy and the corrosion products and Legionella have a negative synergism. Rule No. 1 for controlling Legionella is to have a clean, corrosion product (rust) free system. This should have been mentioned.

4. Chlorine dioxide has the following attributes:

· It is the most effective biocide against Legionella.

· It is the most effective biocide for the destruction of biofilm.

· It oxidizes and does not chlorinate.

· It is not affected by pH (high or low).

· It does not hydrolyze in water (it remains as a gas).

· It reacts within seconds with oxidizable materials.

· It does not form chloramines.

· It will destroy phenols without creating chlorinated phenols, does not form trihalomethanes (THMs), and it does not react with azoles.

· It does not react with ammonia or nitrogen.

· It does not react with organics (see above).

· It has a low corrosivity to metals (much lower than chlorine calcium hypo, sodium hypo or gas).

Frank Rosa of Aqua Technical Services Inc. in Liverpool, NY, offers the following:

· Infectious dose. The infectious dose for many bacterial-induced illnesses has never been determined - would you allow yourself to be injected with a solution containing one organism? If no, then why continue the argument against testing on that false premise?

· Not all Legionella species are associated with illness. There are reports that three additional species of Legionella have been found and identified, but since Legionella pneumophila sergroup 1 (LpSG1) is the most prevalent in outbreaks, it stands to reason that testing for this species is warranted. Medical workers have fixated on LpSG1, thus greater exposure has been given it. Many other species may be involved in illness, but go unreported since LpSG1 (plus the other SGs) has not been detected.

· Culture-based testing. While true, this does not negate the feasibility of using the Binax test for spot checks . . . this test takes one hour and provides a good indicator of LpSG1 count.

· Continuous chlorination. This is something health authorities keep revisiting and water treatment people keep supporting. However, Legionnaires' disease (LD) outbreaks due to contaminated potable water systems have already established that the chlorine used in potable water systems is not effective against LD. As it leaves the plant the chlorine is within 5-6 parts per million (ppm) to insure effective chlorine at distal ends of system.

The other issue is pH. If this is not addressed, so as to generate hypochlorous acid, then efficacy is lost.

Intermittent chlorination is OK, but hypochlorous acid must be generated or it is a lesson in futility. To do this will require acid application to the tower system - with no guarantee of efficacy.

· Bromine. This is a waste of time and money. If it takes 1 ppm of chlorine at pH 6.8 to address an issue, it will take 10 ppm bromine to do the same job based on reactivity of the bromine.

Nonoxidizing biocides are intended to keep a condenser water system within the fouling parameters of the condenser manufacturer. No commercially supplied biocide is offered for sale with claims of efficacy against LD by biocide manufacturers.

Author Charles Ascolese of BetzDearborn in Trevose, PA, offers this response:

Use of halogens figures prominently in the recommendations of various government agencies, professional organizations and numerous experts on Legionella risk reduction.

Halogens are key elements of preventative risk-reduction measures, as well as emergency disinfection procedures carried out in response to a disease outbreak.

A wide range of halogens is available for disinfection of industrial water systems, such as chlorine gas, liquid bleach, calcium hypochlorite, bromine and various solid halogen donor products. Chlorine dioxide is one form of halogen and is appropriate for certain applications, but it is generally much more expensive than gas or bleach, and does not appear to have the broad-based endorsement of authoritative bodies afforded chlorine from gas or bleach. No one agent can be considered a panacea for control of Legionella.

Our discussion centered on chlorine - from gas or liquid bleach - partly because chlorine is the most widely used disinfectant and because chlorine gas and liquid chlorine bleach are the most economical sources of halogen for water disinfection.

A common misconception is that chlorine from gas, bleach or solid donors becomes ineffective at alkaline pH. This is not true. When chlorine is applied to alkaline waters (pH >7.5), the hypochlorite ion (OCl-) predominates. This species is a strong oxidizer and germicide, albeit with slightly lower oxidizing and disinfecting power than hypochlorous acid (HOCl), which predominates at more neutral pHs.

The effect of pH on chlorine performance is minimal when it is applied continuously. When applied intermittently, a slightly higher chlorine residual will compensate for pH effects. Alternatively, bromine can be used since hypobromous acid predominates up to a pH of around 8.5. Bromine is very similar to chlorine in oxidizing power and germicidal performance. As is true for the hypochlorite ion, hypobromite, which predominates at pH above 8.5, also remains a strong oxidizer and germicide.

Some Legionella experts have called chlorine the "gold standard" for Legionella disinfection. Legionella bacteria are not resistant to chlorine; however, this organism may escape disinfection if it is sheltered by biofilm, deposits or corrosion products. To aid penetration of matrices that can protect Legionella bacteria, biodispersants and/or nonoxidizing biocides may be needed.

When applied in accordance with the US Environmental Protection Agency and state regulations, modern nonoxidizing biocides do not pose an environmental threat. These materials can greatly enhance biological control achieved using halogens alone. In fact, most published guidance recommends concurrent or alternating use of oxidizing and nonoxidizing biocides.

Organic biocide molecules, such as glutaraldehyde or isothiazolone, are not persistent, but are eliminated from the environment by hydrolysis, photodegradation, and biodegradation. A number of these molecules have demonstrated very good efficacy against Legionella under various test protocols at dosages well below the label maximum.

Nonetheless, responsible water treatment companies refrain from making claims about legionellosis prevention or eradication of Legionella bacteria because an EPA policy decision issued in 1980 prohibits such claims.

The value of routine Legionella monitoring for disease prevention in open recirculating cooling systems does not have a consensus of support among experts. In fact, government agencies and professional organizations do not currently recommend routine testing for Legionella in the absence of a known problem. The life cycle of this organism and the limits of test methods are such that the organism may not be detected or may be detected only sporadically, yet still represent a health risk.

Testing for Legionella is recommended to identify potential sources of disease, to evaluate the efficacy of disinfection procedures, and as a demonstration of responsible system care.

The Binax Equate field test has a high threshold of detection and is only designed to detect LpSG1. This means potentially significant populations of LpSG1 will go undetected. Further, the Binax system is unable to detect other pathogenic Legionella species. The Binax Equate test method is subject to false positives triggered by a common soil and water organism, Pseudomonas fluorescens.

When using the Binax Equate procedure, both negative results and positive results must be interpreted cautiously. This method may have some limited value in identifying potential "hot spots" for legionellosis risk, but should always be supported by confirmatory, culture-based testing performed by a competent Legionella testing laboratory. Experts in the Legionella field have stated that a useful real-time monitoring device that can be used in the field is likely two to three years from commercial availability.

When Legionella monitoring is performed, negative results or detection of only low levels of Legionella may be misinterpreted as indicating a system is "safe" and lead to a lapse in proper treatment. It is more prudent to assume open recirculating cooling systems are always capable of harboring Legionella bacteria and treat such systems accordingly, regardless of test result. It follows that a system operator should spend resources on cooling water treatments designed to prevent biofouling, corrosion and deposition. A properly treated, clean cooling system is the best insurance against health risks associated with growth of Legionella bacteria.