Updated on 6/24/19 and 8/14/21
The World Health Organization (WHO) defines “heterotrophs” as micro-organisms that require organic carbon for growth and Heterotrophic Plate Count (HPC) as colony counts using a simple, culture test on an organic carbon medium (WHO 2017). The bacteria population recovered in HPC tests can vary widely with temperatures, incubation times, nutrients, and methods used by the laboratory.
People are exposed to HPC bacteria much more from food than from water (WHO 2003). Tens of thousands of colony-forming units (CFUs) per gram or milliliter have been found in dairy, meat, and vegetable products (Reynolds 2002).
Are HPC bacteria harmful?
After two days in Geneva in April 2002, a group of public health experts and microbiologists convened by WHO and NSF International to discuss the role of HPC in managing water systems reached a consensus that,
“There is no evidence, either from epidemiological studies or from correlation with the occurrence of waterborne pathogens, that HPC values alone directly relate to health risk” (WHO 2003).
The basis for the WHO group’s conclusion is simply that disease is caused only by pathogens, not by harmless bacteria:
“There is no evidence for a health effects link in the absence of pathogen contamination” (WHO 2003).
HPC versus pathogens
Although most disease-causing (pathogenic) bacteria are heterotrophic, typical HPC tests will not detect pathogens because special media and techniques are needed to grow those organisms:
“There are opportunistic pathogens that may regrow in water but that is not detected in HPC measurements, including strains of Legionella and nontuberculous mycobacteria. …There is no evidence that HPC levels per se, as measured by established procedures, have a direct relationship to… organisms such as legionellae, Pseudomonas aeruginosa, and non-tuberculous mycobacteria” (WHO 2003).
Donahue found no consistent relationship between HPC and Legionella pneumophila in domestic (potable) water (Donahue 2021).
Some studies have suggested high levels of harmless HPC bacteria can actually protect humans by out competing pathogens (Camper 1985). A University of Arizona study showed that HPC bacteria were antagonistic to pathogenic strains of Salmonella and E. Coli, reducing counts of pathogens added to water by more than 10-fold in one day and 10,000-fold in two days. Payment’s studies showed that faucet filters reduced gastrointestinal illness by up to 40% despite increasing HPC counts (Payment 1991; Payment 1997). Donahue found that Legionella pneumophila positivity was higher in water with HPC concentrations ranging from 101–1000 CFU/mL than in water with HPC exceeding 1001 CFU/mL ((Donahue 2021).
The role of HPC in water management programs (WMPs)
Although HPC has not been found to correlate directly with pathogens, it can be used to assess conditions (e.g., stagnation, disinfection) that affect microbial growth in a building water system (AWWA 2020, Donahue 2021).
The maximum level of HPC allowed in US public water systems without filtration is 500 CFU/mL (per the US EPA Safe Drinking Water Act), but a maximum HPC level in building plumbing (potable) systems has not been defined.
For the type of water management program (WMP) outlined in ANSI/ASHRAE Standard 188 and the Centers for Disease Control and Prevention (CDC) toolkit (CDC 2017)—and required for hospitals and nursing homes that receive reimbursements from the Centers for Medicare & Medicaid Services (CMS)—tests are generally performed for one of two purposes:
- To validate the overall effectiveness of the WMP in accomplishing its primary objective (e.g., to minimize Legionella), or…
- To monitor the performance of a control measure. For example, if a WMP control measure is to maintain hot water temperatures at all faucets within a certain range (limit), temperatures could be measured at faucets periodically to see if they are within that range.
The distinction between validation tests and control measure performance tests is important in determining the role of HPC in a WMP.
For a water management program (WMP) designed to minimize Legionella risk, consider each water system type separately–based on scientific evidence, guidelines, and best practices–in deciding the role of HPC tests for control measure monitoring or WMP validation.
Plumbing (domestic water) systems
Studies have shown no direct or consistent correlation between HPC results and Legionella in domestic hot or cold water systems. Within biofilms, however, van der Kooij found Legionella colony count correlated significantly with HPC as well as with the total cell count (TCC) and adenosine triphosphate (ATP) concentration (van der Kooij 2017). Donahue’s data showed that Legionella and Mycobacterium species were found more often in water samples with high HPC (Donahue 2019). Selvaratnam did not find a significant correlation between HPC and Legionella, mycobacteria, or Pseudomonas (Selvaratnam 2018).
Although there is not enough evidence to assume pathogen control based solely on negative or low HPC tests, comparing HPC and pathogen test results for a given facility’s domestic water system could provide useful information. For example, seeing a consistent correlation could allow a facility to make use of HPC test results between Legionella sampling rounds.
WHO points out that HPC tests can indicate stagnation, inadequate temperatures, or low disinfectant levels in domestic water systems but other tests (e.g., temperature; chlorine) will likely be more useful for control measure monitoring.
Whirlpool spas
Data studied by Dr. Richard Miller of the University of Louisville (Miller 2002) have shown HPC to be a good indirect indicator of Legionella in whirlpool spas. Although Legionella tests will provide more direct and reliable feedback on Legionella concentrations in whirlpool spas, HPC tests could be used as an additional (and perhaps more frequent) validation test. ASHRAE Guideline 12-2020 (section 7.4.1) and ASHRAE Standard 188 (section 7.3.5.1) recommend a maximum HPC level of 200 CFU/mL in spas.
Cooling towers
HPC tests should not be used to validate Legionella control in cooling tower systems because studies have shown no reliable correlation between HPC and Legionella counts. In Garnett’s study of 100 cooling towers, Legionella counts greater than 100 CFU/mL were found only in towers with an HPC exceeding 10,000 CFU/mL and Legionella counts higher than 1,000 CFU/mL were found only in towers with an HPC exceeding 100,000 CFU/mL (Garnett 1993). But of the 1,100-plus towers studied by Miller, towers that tested positive for Legionella generally had a lower HPC than towers that tested negative. What’s more, some of the towers with extremely high Legionella counts had an HPC near zero (Miller 1993). Witherell found the same results as Miller—high Legionella counts with low HPC—indicating certain water treatment programs may reduce bacteria competing with Legionella for food, but not reduce Legionella, allowing Legionella to flourish (Witherell 1986).
Although they are not an appropriate validation method for Legionella in cooling towers, HPC tests may be required by regulations (e.g., in New York) and have been recommended for monitoring cooling tower treatment (CTI 2008; WHO 2007).
How do you use HPC in your water management program? Please comment below.
For Additional Reading
ASHRAE. 2021. Legionellosis: Risk Management for Building Water Systems. Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers. Available for purchase at Legionellosis: Risk Management for Building Water Systems
ASHRAE. 2020. Guideline 12-2020. Managing the Risk of Legionellosis Associated with Building Water Systems. Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers.
AWWA. 2020. Responding to Water Stagnation in Buildings with Reduced or No Water Use. American Water Works Association.
Camper AK, et al. 1985. Growth and Persistence of Pathogens on Granular Activated Carbon Filters. Appl. Environ. Microbiol. 50:6, 1378-1382.
CDC. 2017. Developing a Water Management Program to Reduce Legionella Growth & Spread in Buildings: A Practical Guide to Implementing Industry Standards. Atlanta: Centers for Disease Control and Prevention. Available at Developing a Water Management Program to Reduce Legionella Growth & Spread in Buildings
CTI. 2020. Legionellosis Guideline: Practices to Reduce the Risk of Legionellosis from Evaporative Heat Rejection Equipment Systems. Houston, TX: Cooling Technology Institute.
Garnett HM, Gilmore K, Liu J. 1993. Legionella: An Unwelcome Pollutant. Environmental Technology 11; 393-400.
Reynolds KA. 2002. HPC Bacteria in Drinking Water: Public Health Implications? Water Conditioning & Purification, July.
Miller RD, Kenepp KA. 1993. Risk Assessments for Legionnaires Disease Based on Routine Surveillance of Cooling Towers for Legionellae. Presented at the 4th International Symposium on Legionella, 1992. In: Barbaree, J. M., R. F. Breiman, and A. P. DuFour, eds. Legionella: Current Status and Emerging Perspectives. Washington, D.C.: American Society for Microbiology; 40-43.
Miller RD, Koebel DA. 2002. Prevalence of Legionella in Whirlpool Spas: Correlation with total bacteria numbers. In: Marre R, et al, eds. Legionella. Washington, DC: ASM Press, 275-279.
Payment P, Richardson L, Siemiatycki J, et al. 1991. A Randomized Trial to Evaluate the Risk of Gastrointestinal Disease due to Consumption of Drinking Water meeting current Biological Standards. American Journal of Public Health vol. 81, no. 6.
Payment P, Siemiatycki J, Richardson L, et al. 1997. A Prospective Epidemiological Study of the Gastrointestinal Health Affects due to the Consumption of Drinking Water. International Journal of Environmental Health Research 7, 5-31.
Witherell LE, Novick LF, Stone KM, et al. 1986. Legionella in Cooling Towers. Journal of Environmental Health, November/December; 134-139.
Rusin P, Rose J, Haas C, Gerba C. 1997. Risk assessment of opportunistic bacterial pathogens in drinking water. Rev Environ Contam Toxicol. 152:57-83.
Selvaratnam S, Coughlin M, Sotkiewicz E. 2018. Practical Recommendations for Meeting the Objectives of the CMS Memorandum. Presented at the annual convention of the Association of Water Technologies.
US EPA Safe Drinking Water Act (SDWA).
WHO. 2003. Heterotrophic Plate Counts and Drinking-water Safety: The Significance of HPCs for Water Quality and Human Health. Published on behalf of the World Health Organization by IWA Publishing, London. Available at Heterotrophic Plate Counts and Drinking-water Safety.
WHO. 2007. Legionella and the prevention of legionellosis. Geneva: World Health Organization. Available at Legionella and the prevention of legionellosis.
WHO. 2017. Drinking Water Parameter Cooperation Project. Support to the revision of Annex I Council Directive 98/83/EC on the Quality of Water Intended for Human Consumption (Drinking Water Directive). Available at Drinking Water Parameter Cooperation Project.
Thanks for sharing, Matt. I agree that there is little if any correlation between HPC and specific opportunistic pathogens such as Legionella. Validation testing for HPC can be useful as a general indicator of the overall effectiveness of the water management program. Depending on the results of the analysis, this can confirm whether bacterial growth is likely to occur in critical areas in your water system.
Thanks very much for your input, Andrew. I assume you are referring to plumbing systems in particular. The Payment studies are really interesting in showing that HPC bacteria could actually be protective. I hope to see more research on the relationship between certain harmless bacteria and pathogens, and advanced testing technologies, which together should help facilities and consultants interpret and apply test results to reduce health risk. But you make a good point that HPC good possibly be used to monitor a control measure, particularly one related to stagnation or organic carbon.
Hi Matt,
thanks for a very balanced review. One point I would add is that water samples taken from an outlet and tested for Legionella will reflect what is in the water flowing through the tap. Contamination of the outlet from other sources is highly unlikely. However HPC results may reflect the condition of the hands that last touched the outlet – or splash back from the drain on to the outlet. This adds another level of uncertainty in understanding what a ‘High’ HPC result actually means. It also raise the question of what we are sampling from – the water or the outlet?
Do you have any ideas on representative sampling for HPC? Surface disinfection before sampling? remove aerators? sampling after a flushing period? I’m afraid I no answers spring to mind on this but I would value any thoughts you have.
cheers
Richard
Excellent points, Richard. Since I don’t sample for HPC in domestic water systems, I haven’t thought through methods. For Legionella sampling, I definitely would not want to disinfect the faucet tip or remove the aerator prior to sample collection, but I suppose it might make sense for HPC, to minimize the sources of background contamination you mentioned. Taking those measures would not tell you what a person is exposed to but would give a better indication of what’s in the system, which for HPC is the more appropriate question to answer. Thanks very much!
Hello Matt,
The relationship between HPC, P. aeruginosa, and Legionella is quite fascinating in my opinion. My novel understanding from some research is that high HPC levels are a good indicator that there is biofilm present somewhere in your water system, or at least the potential for formation. Since Legionella are nearly ubiquitous in all water systems, I believe we should be focusing on eliminating these biofilms over Legionella in general. When Legionella is allowed to multiply rapidly within a protozoa rich biofilm, disperse itself into a water system, and become aerosolized at outlets is when outbreaks occur. It is very likely that a WMP that focuses solely on maintaining low levels of Legionella may unknowingly create more resistant strains of Legionella over time.
I work for a environmental testing and am currently doing RnD work to create a system for sampling, testing, and helping our customers create viable WMPs. HC Info has been an invaluable resource for us in this endeavor, keep up the great work!
Regards,
Chase
Thank you for commenting, Chase. I agree that HPC levels relative to Legionella and Pseudomonas, as well as mycobacteria, is interesting and understudied. I know efforts are underway to gather and study data and look forward to the results. Several studies have shown you are correct that biofilm is important in managing Legionella risk.
Does it make a difference how the water samples are collected? Grab samples take a very small portion of water. I have heard of a product made of olefin that is said to capture bacteria over time while repelling water. So the sample taken is more representative of what is actually in the water. Whether it be a basin, like a cooling tower or a length of piping feeding a faucet.
Could this sampling procedure make a difference in testing?
Similar to swab samples of biofilm, that procedure might provide useful data for water system management but a water sample of what building occupants are actually exposed to is a better indicator of risk. To validate a water management plan for Legionella control, I suggest using a method has been proven reliable. Thank you for commenting