By Jeff Williams, PhD and Nathan Kenney
Introduction
The recent report from the Centers for Disease Control and Prevention (CDC) on “Microbes in Pool Filter Backwash” (May 17, 2013, Morbidity and Mortality Reports [MMWR], 62 (#19): 385-388) certainly stirred up the mass media airwaves and put something of a cloud over public perceptions of swimming pool safety, just as this year’s season was entering full swing. Reminding bathers about personal hygiene as a part of public pool use, in particular, is always a good thing, but this report’s results readily became sensationalized by content-hungry news channels. A rather unsurprising set of findings—at least to microbiologists—on a limited range of pool filter samples got converted from what should have been a news story in the ‘dog bites man’ category, into lurid coverage in the ‘poop in pools everywhere’ style. It probably ended up doing little to focus public attention on the behavioral changes that CDC strives so hard to bring about.
That’s unfortunate, because making such a beneficial and wholesome activity as swimming even more attractive and encouraging participation is critical to national efforts to get Americans off the couch and doing something healthy yet delightful. Raising fears about risks of what have come to be called recreational water infections (RWI) can be a setback to those efforts. That is what makes it important to put the MMWR data in a broader perspective, taking into account some of the nuances of the technology deployed in the report, and that is what we offer here. Nothing in our comments, however, should be taken as diminishing the growing and proper concerns about the steady increase in swimming pool RWIs that CDC has highlighted in recent years.
Interpreting high-tech findings and assessing the risk
CDC researchers collected 161 backwashed pool filter samples from Atlanta-area pools and tested them for biological contaminants using DNA probes. In this way, filtered particulates ended up being highly concentrated in a small amount of backwash water. They then used DNA fingerprinting to detect whatever was trapped on the filter medium with an extraordinary sensitivity. Many of the filters, doing their job properly, were found to have trapped bacteria to the extent that more than half of the samples showed evidence of the presence of E. coli, commonly used as an indicator of fecal contamination. About the same proportion contained fingerprints of Pseudomonas aeruginosa, a microbe that can be human-associated, sometimes causing disease, but which can occupy and multiply in a wide range of environmental niches.
Now, what these findings signify has to be framed by the limitations of the technology; DNA probes provide for quick and sensitive ID analysis of samples for specific microbes, but the meaning of results is limited without using other kinds of microbiological tools to arrive at an assessment of RWI infection risks. For example, DNA technology will give a positive signal in backwash water samples, whether the microbe is alive and potentially infectious or dead and harmless—killed, in this instance, most likely by the pool’s chlorine sanitizer doing its regular job, taking care of the inevitable microbial contamination that comes from bather load in the pool. It would have taken many more microbiological procedures to get a handle on whether the E.coli or the Pseudomonas fingerprints truly represented a RWI risk and signified that the sanitation/filtration system was faulty or not properly maintained, the kinds of signals that properly should raise alarms and are often discovered in the aftermath of disease outbreaks. No such outbreaks were reported in the Atlanta area in the summer of 2012, although it remains the case that many RWI incidents still go unrecorded. Nevertheless, the observation does suggest, at the very least, that there were not wholesale failures of proper sanitation and hygiene in the sampled region over this period. The bacteria detected were not in the primary disease agent category (such as E.coli O157;H7). All the anxieties raised by the headlines were not necessarily in a good cause. Finding traces of DNA ‘fingerprints’ of microbes in filter retentates does not necessarily point to unsafe pool management practices.
Their presence may instead be an indicator of the presence of human bodies in the pool. These kinds of results do not even assure us that intact germs were present, since remnants from the beneficial effects of sanitizers on microbes could give rise to positive signals in DNA tests. On the other hand, intact or not, alive or not, the fingerprints are undeniable, and in the long run, accumulation of this kind of evidence may end up proving useful in risk assessment. It certainly provides a new approach to characterization of an important public health issue associated with recreational water use.
It’s a fair proposition to make that adopting good personal hygiene practices could beneficially affect the amount of contamination the sanitation/filtration system must treat. But it remains to be determined whether showering ahead of pool bathing would significantly reduce the rate at which ultra-sensitive DNA probing turns up positive signals of human-related microbial material in filter backwashes. It is certainly a reasonable notion, but it will be a challenge to test it experimentally.
Common pool filter types
Chlorine: upside/downside
Significant, but almost lost in the high visibility of the pool- poop story, is the detection in some samples of parasite cyst fingerprints—significant because, in this instance, even the best chlorine sanitizing system will not inactivate Cryptosporidium oocysts. They are most likely to have a human feces origin in pool water and almost certainly would still be alive and infectious in the filter. Filtering them out and regularly backwashing them into the septic outflow is another component of ongoing, effective pool management. But here we encounter a higher risk to bathers and much less certainty about the capacity of the filter medium to capture and eventually remove oocysts. Close enough to to take its presence in backwash samples, even at a low rate (two percent), as a serious indicator of potential problems.
This risk is not readily diminished by adopting sanitizing alternatives to chlorine. For all its downsides, including objectionable smell, taste, eye and skin irritation and DPB generation among others, chlorine serves up an unmatched array of advantages in maintaining pool safety. It deserves its centerpiece role in recreational water sanitation. That it does not cope with the toughness of Crypto oocysts, added to the other shortcomings, has led to clamoring for more attractive and powerful replacements. That hasn’t proven easy and while practical alternatives are avail- able, they bring their own weaknesses to the challenge. Chemical agents with oocyst-inactivating power, such as chlorine dioxide and ozone, pose more operational demands on users than chlorine and cannot compare in their inability to sustain biocidal activity in the main body of pool water that is so readily accomplished with chlorine residuals. UV irradiation can destroy the infectivity of oocysts, but besides furnishing no residual protection to the recirculating water, it must operate by line-of-sight exposure to the inactivating rays. Any obstruction in the form of turbidity in heavily bather-loaded pools can undermine the utility of UV. All the inactivating power of UV has to wait upon the arrival of germs over the course of the recycling turn of all the pool water; in the meantime, those germs are free to cause trouble.
Clearer is safer
Which brings us to the final piece of the framework for our perspective on the CDC report: water clarity as a contributor to pool water safety, regardless of the method of disinfection being deployed. Turbidity is the enemy of effective pool safety manage- ment and efficient filtration is the way to counter it. Turbidity interferes with UV, consumes disinfectants (it’s a ‘demand’ for halogen, meaning less is available for sanitation) and contributes to the risk of pool drowning accidents by obscuring the view of bathers who get into trouble and sink out of sight. Filter back- washes should contain evidence for what the medium is taking out (just as the CDC researchers found). Helping that process by flocculating suspended particles is a worthy goal and can be achieved as a part of regular pool management. There are plenty of products available to make that happen. An illustration of what can be accomplished with a natural product formulation specifi- cally aimed at improving filtration efficiency, is shown in Figures 1 and 2. Cryptosporidium oocysts, enmeshed in large flocs created by pool water treatment with this biopolymer formulation (Figure 1), are now much more readily trapped on the filter and less likely to remain in the recirculating body of water and cause RWI out- breaks. Particles as small as E.coli can be aggregated by this kind of treatment into clumps (Figure 2) that end up being retained by, in this case, a flat filter membrane with five-micron pores that would normally allow these bacteria to pass through.
Figure 1 is a microscopic image of a flocculum generated with treatment of Cryptosporidium suspended in pool water. The green fluorescent areas are oocysts, enmeshed in a floc made up of two interacting biopolymers. The flocs are large enough to be retained by a coarse sand filter. Figure 2 is a microscopic image of a filter membrane after filtering a suspension of E.coli in treated pool water . The purple-stained clumps of flocced bacteria are retained on the membrane, whereas untreated bacteria normally pass straight through the five-micron pores.
Figures 1. (left) and 2 (right).
Cryptosporidium oocysts and E. Coli
Conclusion
Encouraging a multi-pronged approach to pool sanitation, with proper attention to disinfection, clarification and filtration, as well as emphasizing the role of personal hygiene measures for all bathers, is likely to lead to more widely enjoyed and enjoyable recreational water experiences. Properly implemented, these kinds of pool management practices do work. That would be a nice message to put out during the summer swimming season!
About the authors
Dr. Jeff Williams is the Chief Technology Officer and Sr. Vice President of R&D, HaloSource Inc. He is Emeritus Professor of Microbiology and Molecular Genetics at Michigan State University, where he was a teacher and biomedical researcher for almost 30 years before founding HaloSource in 1998.
Nathan Kenney is a Specialist in Water Microbiol- ogy, Division of Product Development, HaloSource Inc. His gained his BS in microbiology from the University of Washington, Seattle, WA.
Bather load bacterial sloughing explanation
Whenever a person swims in a pool, they will drop bacteria and viruses from their skin and hair (not to mention any incontinent accidents) and the vast majority of these microbes will be killed by the chlorine in the pool. The bacteria that live on the skin of healthy people are largely not harmful unless the person already has lowered immune defenses. People who fall into the category of having lowered immune defenses are typically very young children, seniors, pregnant women, AIDS patients and those on immunosuppressant medication.
It is not at all unusual that DNA probes of pool filters found bacteria as common as P. aeruginosa or Escherichia coli (E. coli), given that P. aeruginosa lives on most surfaces in civilization and E. coli is present in microscopic amounts on everything someone touches after going to the bathroom. Also, if 75 percent of samples could contain either E. coli or P. aeruginosa and there wasn’t a huge outbreak in the Atlanta area as a result, it suggests that these bacteria are truly not the main sources of swimming-related infections.
It should be made clear that there is a small sub-group of E. coli responsible for severe illness with vomiting and diarrhea called E. coli 0157:H7. These bacteria make people sick, and if they are found in a pool filter, it would indicate that someone was sick with diarrhea and vomiting symptoms when they went swimming. Almost any other
E. coli contamination would simply indicate that swimmers were not thoroughly cleaning themselves before using the pool. While this is unsanitary, it is not a cause for alarm if the pool is properly chlorinated and maintained.
Knowing the difference between normal E. coli and E. coli 0157:H7, it should be considered that the CDC study did not find any E. coli 0157:H7 with the DNA probe, but did find other non-specific E. coli. This shows that contamination in most pool filters was the result of dirty swimmers and not the result of severely sick people contaminating the entire pool, making it dangerous for all other swimmers. Luckily, in the case of either type of E. coli, chlorine will kill whatever is floating in the pool as long as the chlorine levels are kept appropriately high. Proper chlorine use is vital to keeping pools safely clean of the infectious and dangerous E. coli 0157:H7, as well as many other potential microbial contaminants.