May 2003
Volume 45 Number 5

Microbiological Water Purifier Testing--The Quest for Surrogate Organisms
by Rick Andrew   Pages: 

Task groups of the NSF International Joint Committee on Drinking Water Treatment Units (DWTUs) have been in pursuit of standards for microbiological water purifiers for several years. The development of microbiological water purifier standards has proven to be a challenging project requiring a great deal of effort and expertise. Many wonder why the task of developing these standards is so complex. To many, it seems the task groups could simply build these standards on the foundation of already existing DWTU standards and complete them in relatively short order, but that isn’t the case.

The need for surrogates
Current DWTU standards deal primarily with performance claims for chemical reduction or mechanical filtration of particles. Testing involved to verify conformance to the standard for these performance claims is relatively straightforward in most cases—the product is challenged directly with the contaminant being evaluated. For example, a product being tested for benzene reduction under ANSI/NSF Standard 53 DWTUs - Health Effects is challenged directly with a solution of benzene in water, as would occur in the real world when such contamination occurs. Mechanical filtration claims are verified by challenging with particles of the appropriate size range, again as would occur in real world contamination situations.

The same isn’t necessarily true, however, of microbiological claims verification performance testing. The Standard 53 cyst reduction claim is a good example of this. Years ago, there were no good analytical methods for detecting and quantifying protozoan cysts. There were no commercial sources of cysts, nor were safe laboratory procedures available for handling cysts. Because of these difficulties, the joint committee decided to look for a surrogate for cysts. They examined Cryptosporidium, Giardia, toxoplasma and entamoeba, and determined that the smallest of these is Cryptosporidium in the range of 3-7 microns (mm). Because the technology covered by Standard 53 is that of mechanical reduction due to size exclusion, the joint committee determined particles in the size range of 3-4 mm would be an appropriate surrogate. If a filter can filter out particles in this size range, then it can filter out the four protozoan cysts that were examined, considering they’re all bigger in size than the surrogate particles. Under this philosophy came the adoption of Arizona road dust—a test media (see—as the surrogate for the Standard 53 cyst reduction test.

Testing for cysts
Since the “cyst test dust” test—as it has come to be known—was developed, another surrogate for cyst reduction under Standard 53 has been developed. Polystyrene microspheres (spheres) at 3 µm can now be effectively used as surrogates for cysts. They are the right size, and the spheres fluoresce under a microscope, making them easy to detect. Finally, there’s a third option that some laboratories have developed analytical and quantification methods for live Cryptosporidium. In addition to test methods, there are also commercial sources for the organisms and safe laboratory handling procedures. While this option is available, it’s still desirable to have a surrogate that’s less hazardous to handle such as spheres.

This example of cyst reduction testing shows us some of the reasons that surrogate organisms can be necessary when developing standards for microbiological reduction performance claims verification:
When many different, yet similar, organisms exist within a class: It may be possible to examine the class of organisms and select one of them to serve as a “worst case” surrogate for the others. In the class of protozoan cysts, Cryptosporidium was chosen as a worst case for mechanical filtration because it’s physically the smallest.
When there are not well developed, reliable analytical and quantification methods for certain organisms: It makes no sense to test with something that cannot be analyzed properly as performance cannot be validated.
When there’s no commercially available source of certain organisms: Standards must be developed such that any sufficiently equipped laboratory with the proper expertise can reasonably expect to be able to run any of the tests in the standard. If there’s no way to obtain the organisms necessary to run the test, then obviously no one can run the test and the standard isn’t serving a useful purpose.
When the organism cannot safely be handled by laboratory personnel: Many highly toxic chemicals and infectious organisms can be safely handled by well-equipped and properly trained laboratories. In these cases, research has been performed to develop these safe-handling procedures. There are also many infectious organisms that haven’t been studied enough to have safe handling procedures developed. For obvious reasons, these organisms cannot be used for testing to verify performance claims.

What makes a good surrogate?
Starting with the reasons for needing surrogate organisms as described above, it’s possible to develop a list of criteria for a good surrogate organism.
1. The surrogate organism must be a “worst case” organism for the class and the water treatment technology. Note that “worst case” organisms definitely do vary technology by technology. As discussed, for mechanical filtration, the smallest organisms are worst case organisms. For ultraviolet (UV) disinfection, those organisms that demonstrate the highest UV tolerance are worst case. For halogen-based technologies, resistance to halogen disinfection will determine the worst case and so on.
2. There must be established and reliable methods to detect and quantify the surrogate organism in water. 3. There must be a commercial source for the surrogate organism.
4. Laboratory staff must be able to safely handle the surrogate organism. The best surrogate organisms are non-pathogenic, and can be handled safely without any special precautions by laboratory staff.

Classes of organisms
In addition to protozoan cysts, there are two other classes of microorganisms for which surrogates are needed—virus and bacteria. The three classes of protozoan cysts, virus, and bacteria make up the spectrum of potential disease-causing microbes that can contaminate water. Because these classes of microorganisms have significantly different physical characteristics, they must be examined class-by-class when surrogate organisms are chosen.

Considering the three classes of organisms, different surrogates are required not only for each class of organism but for each treatment technology and all of the factors influencing choices of surrogate organisms, it isn’t surprising that the quest for surrogate organisms for microbiological water treatment product standards is a daunting one.

Gauging the progress
Extensive literature searches and consultation with experts in the field have resulted in a good deal of progress being made in the area of surrogate organism selection. Cryptosporidium has been selected as the surrogate for cysts for mechanical filtration performance testing. Brevundimonas diminuta and Klebsiella terrigena have been proposed as bacterial surrogates for mechanical filtration performance testing. In addition, PRD1 and MS2 coliphage have been proposed as viral surrogates for mechanical filtration performance testing. In all of these surrogate selections, the criteria as described above have been considered. The basis for the worst case determination has been small size.

As work on microbiological purification standards progresses, additional surrogates for other treatment technologies will be selected—again based on literature searches and consultation with experts in the field—considering all of the criteria discussed above. Given all of these criteria and the research and consideration given to selecting these surrogates, the resulting standards will be readily usable, conservative in testing so that real world performance will be as certain as possible, accurate in their implementation, and will be safe for laboratory workers involved in testing. Ultimately, this work will result in benefits to consumers, who will be able to buy certified devices that are capable of treating water from unknown sources and making it safe to drink. This is a daunting task, but one well worth pursuing and a goal well worth achieving.

About the author
Rick Andrew is the technical manager of the Drinking Water Treatment Units program at NSF International, of Ann Arbor, Mich. He has been with NSF for four years. His previous experience was in the area of analytical and environmental chemistry consulting. Andrew has a bachelor’s degree in chemistry and a master’s degree both from the University of Michigan. He can be reached at