Enterococci the Official Universal Marine Fecal Indicator: The Data Does Not Agree
A review of the science and a recommendation for change
R.H. Bennett Ph.D.
Applied Life Sciences LLC
In previous H2O Blogs, we discuss the problems with the use of the Enterococci Test for recreation water safety evaluations. Despite the science to the contrary, regulatory agency use of the Enterococci Test is engraved in stone and until the EPA or the State of Hawai‘i take the initiative it will remain in use and unfortunately continue to mislead, and misinform. It will continue to confuse anyone that tries to make sense of a measurement that is scientifically nonsensical.
The Enterococci test is a scientifically invalid risk assessment microbe.
If the Enterococci test were a medical diagnostic test, the FDA would not approve it. The FDA requires, for very valid reasons, that all diagnostic tests have very low false positive and false negative rates. Imagine the chaos and wrath that would ensue from a blood test to detect Pancreatic Cancer that had a 63% false positive rate? In effect that is the error rate for our recreation water safety test (Wade). Some officials might argue we need to err on the cautious side. This is a dangerous argument, as any test that has such imprecision will also result in a high false negative rate at the same time. That is to say, that as much a 37% of the results are false negative and significant risk is present and unannounced.
In figure 1 below is the risk estimate for illness (GI) and increasing Enterococci count in the recreation water. This is a statistical product from many studies called a meta-analysis (Wade).
A reasonable fit estimate would have the dots line up closely along the regression line. However, this real world data does not. If it did, it would have a good correlation of .7 or higher. This would suggest the risk is closely matched to the bacteria count of Enterococci. Instead, the correlation is poor 0.37. Thus we must conclude whatever is making people sick is not closely related to the Enterococci count when all studies are grouped as one. The Enterococci disease risk relationship is valid only when a treated sewage wastewater outfall is located in the water body. Where there is no Point Source discharge (a conduit), there is no relationship. Given this association the false positives and false negatives are inevitable. We have to wonder why the agencies have not been forthcoming with this confusing conundrum.
In fairness, however, the precise accuracy and precision of the Enterococci test are not known simply because in risk studies when a person gets sick from the recreation water, the precise cause of the illness or etiology cannot be determined where the illness is patient reported and after the fact. In other words, studies based on self-reporting of gastrointestinal illness are at best inconclusive. Moreover, researchers have assumed that the only risk from contaminated water is GI illness. Those that have measured other disease risks report skin and upper respiratory infections in addition to GI illness (Table 3).
To determine risk more precisely the study would have to track a large, diverse population of people in swimming with full-face immersion in Hawaiian waters and determine the precise cause of the illness. That would require definitive medical diagnosis when it occurred. Such trials are extremely expensive, and research labs are needed to confirm the diagnosis. This information void perpetuates the uncertainty and the misinterpretation of the current water quality test data.
Risk depends on where it is measured
Enterococci based swimmer risk assessments vary tremendously depending on where the study was conducted. The risk estimated in waters where there a known treated sewage water discharges is reasonably clear and consistent. In contrast, on beaches where point source wastewater discharges are not present the risk to swimmers based on Enterococci measures is much lower, inconsistent and much more variable. In this situation infections other than GI are far more common.
In the study data below, researchers reported all illness in bathers and non-bathers at a subtropical beach with no Point Source wastewater outfall. There was no association with the Enterococci counts and GI illness. However, there was a statistically significant association with Enterococci counts with skin illness and respiratory infections (Fleischer). Once again the data suggests we cannot attribute disease risk to the numbers bacterial genus alone or in settings without a source sewage discharge.
When sewage is known to be present in marine waters the likelihood of Enterococci indicating the presence of an enteric virus is quite good. On the border of Mexico and California, raw sewage from Tijuana Mexico flows into the sea. Ocean currents carry the wastewater a short distance into California waters. Researchers found a strong correlation between Enterococci counts and the presence of the Hepatitis A virus and Enterovirus. The correlation coefficient for the bacterial indicators and Enterovirus was 0.7 (a perfect correlation is 1.0) and highly statistically significant (Gersberg). Unlike the EPA studies, the border setting involved raw, untreated sewage and as such the numbers of virus and indicator bacteria were quite high. This provides good evidence that site-specific conditions significantly influence the interpretation of the microbe data and the risk of illness.
Of concern is the observation that young children are more likely than any other demographic to incur illness from contaminated recreation water. They are uniquely susceptible; spend more time in the water and more likely to swallow water (Wade).
Fecal indicator bacteria must not grow or persist in the environment
Confounding the use of enterococci is the data that shows this microbe can grow in beach sands that are wetted by tidal events (Desmarais). Any indicator that can grow in the environment fails the criteria for valid indicator microbes (Boehm).
Enterococci grow and persist in beach sand. Data from Hawai‘i Island and beaches on the West Coast confirm its ubiquitous presence. When the sand is disturbed by the tide, wave and human activity the microbes get suspended for a short while. On a calm day, the water column may have no detectable Enterococci. With surf events and many people in the water, the water column may become bacteria laden.
Enterococci are just one genus in the fecal microbiome. The human fecal microbiome is made up of over a trillion of bacteria thousands of species. (See http://www.hatsprobioticss.com) The other species including one from the marine ecosystem arise from the environment as their natural habitats. Consequently, when the water is tested for Enterococci, it will not and cannot indicate if the bacteria are from humans or the environment.
In the chart and data to the left above, researchers measured the species of Enterococci and several Southern California beach locations. When sewage works were near the beach, E. faecium predominated. Where beaches received urban runoff, and no sewage works present E. casseliflavus predominated (Moore, Lipp).
This suggests perhaps we should measure E. faecium specifically. Since it likely originates in the sewage works, it is likely to move with the virus present, and in this situation, the risk attributions may be valid.
For this hardiness, this organism does not meet the criteria for a fecal indicator (Boehm, Yates), as it is far too persistent in the environment, especially in stream sediments. Rainfall events are common causes of sewage spills it is the same rainfall event that causes transport of stream sediments to the nearshore waters. For these facts, any data on Cp must be interpreted with suspicion.
Data Interpretation is not universal, it is site specific
Researchers determined that enterococci concentrations measured in recreational marine waters polluted by treated wastewater be strongly correlated to the number of swimmers becoming sick with gastrointestinal illness (Byappanahalli Cabelli,). This relationship may hold for places like Honolulu were sewage spills, and leaks are common. However, in beach sites across the state where there are no wastewater treatment plants, there is some published data to guide the interpretation of enterococci counts. A Florida study reports no point source discharges of treated wastes and no OSDS in the area. In this prospective study Enterococci counts where shown to have a dose-response risk relationship not to GI disease but rather skin infections or rashes. The cause of the skin conditions was not determined (Colford).
The densities of OSDS further confound the interpretation of marine enterococci counts. In Hawai‘i tens of thousands of cesspits and septic systems are in use near coastlines of all islands. Besides, there are wet and dry sides of all of the islands, and there is an excellent diversity of soil conditions is which cesspits and septic leach fields are buried. The hydrologic conductivity at these sites also varies greatly. On the dry sides of most islands, the volcanic rock provides almost infinite drainage capacity. This may be moot as even deep clay loam soils of great agricultural valleys in the USA are not effective filters for virus particles (Yates). Scientists can only model virus filtration when and if the complete soil profile can be characterized geochemically (McCray). This process is impractical and very expensive.
The Hawai‘i Department of Health advocates for the transition from cesspits to septic systems as the solution to pollution from cesspits. Law in this regard has become public policy. The state is advised to evaluate the performance of septic leach fields in communities that use this type of onsite waste treatment and the movement of pathogens to the nearshore waters. There is a paucity of data for Hawai‘i conditions. One study conducted by the University of Hawai‘i in the early 1980ʻs demonstrated that soil typical to N. Oahu did not effectively filter bacteria for simple septic system leach fields (Tasato). Viruses are substantially small than bacteria by orders of magnitude.
In Florida, virtually all dwellings in the rural coastal communities use septic systems. Research on the movement of virus seeded into the septic systems reveals the septic systems fail to filter or otherwise impair the movement of virus into the nearshore waters. In Boot Key Harbor, viral phages were detected in a canal adjacent to the seeded septic tank within three h 15 min of the end of the seed period. For wastewater injection wells the, the rate of tracer migration from the injection well to this channel ranged from 66.8 to 141 m h−1. Both tracer studies showed a rapid movement of wastewater from on-site sewage treatment and disposal systems in a toward the Atlantic Ocean. These studies indicate that wastewater disposal systems currently in widespread use in the Florida Keys can rapidly contaminate the marine environment (Paul).
In coastal regions of Florida where there is a high density of OSDS (on-site disposal systems, septic, and cesspools) Enterovirus is commonly found in the receiving waters. The concentration of the virus is proportional to the density of the OSDS. Moreover, groundwater flows carry virus some distance to the receiving waters as well (Lipp).
In Oahu, Hawai‘i researchers documented swimmers infection risk from Enterovirus and Adenovirus transported to the sea in dry weather stream flows. A strong association was made with the density of OSDS including septic systems (Viau).
This data alone should raise a huge red flag and pause the states rush to judgment that requires septic systems in Hawaii’s coastal zones as the solution to cesspool contamination of groundwater and the sea.
A Simple Microbiological Fix
As has been shown, Enterococci are a genus that contains a great many species. Water quality research suggests that only 4 or 5 species predominate in the Marine environment. For very little expense and effort, the predominant species could be characterized in urban sewered areas, suburban non-sewered area, rural cesspool areas and those where septic systems are conventional.
Today bacterial genomics using next-generation technology identifies bacterial taxa that are unique to humans. That is they are not shared with animals or free-living in the environment. The work by genomic researchers depicted in the figure above shows (green dots) clusters of bacterial taxa that are unique to sewage. The blue dots show taxa humans share with other animals. The pink dots show taxa shared by all (Fisher). One of more the sewage taxa may become the definitive indicator for human waste and allow for near real-time monitoring.
It is long past time to apply well-established microbiological science to water quality measurements to more precisely estimate the disease risk from water recreation. The criteria above must be formed into public policy.
In the very near future, genomic technology will enable us to measure the pathogens of concern down to unique genotypes. The identification of a few species of Enterococci may well suggest the genomic tests that should be conducted. The pathogens in water that result from human wastes entering recreation waters are predominantly the Enterovirus Adenovirus and pathogenic Staph. aureus. Testing the waters in urban and rural regions for these disease agents is a matter of political wills, not available technology.
Boehm, Alexandria B., et al. "A sea change ahead for recreational water quality criteria." Journal of Water and Health 7.1 (2009): 9-20.
Byappanahalli, Muruleedhara N., et al. "Enterococci in the environment." Microbiology and Molecular Biology Reviews76.4 (2012): 685-706.
Cabelli, Víctor J., et al. "A marine recreational water quality criterion consistent with indicator concepts and risk analysis." Journal (Water Pollution Control Federation) (1983): 1306-1314.
Colford Jr, John M., et al. "Water quality indicators and the risk of illness at beaches with nonpoint sources of fecal contamination." Epidemiology 18.1 (2007): 27-35.
Desmarais, Timothy R., Helena M. Solo-Gabriele, and Carol J. Palmer. "Influence of soil on fecal indicator organisms in a tidally influenced subtropical environment." Applied and environmental microbiology 68.3 (2002): 1165-1172.
Fisher, Jenny C., et al. "Comparison of sewage and animal fecal microbiomes by using oligotyping reveals potential human fecal indicators in multiple taxonomic groups." Applied and environmental microbiology 81.20 (2015): 7023-7033.
Fleisher, Jay M., et al. "The BEACHES Study: health effects and exposures from non-point source microbial contaminants in subtropical recreational marine waters." International journal of epidemiology 39.5 (2010): 1291-1298.
Gersberg, Richard M., et al. "Quantitative detection of hepatitis A virus and enteroviruses near the United States-Mexico border and correlation with levels of fecal indicator bacteria." Applied and Environmental Microbiology 72.12 (2006): 7438-7444.
Jackson, Charlene R., Paula J. Fedorka-Cray, and John B. Barrett. "Use of a genus-and species-specific multiplex PCR for identification of enterococci." Journal of clinical microbiology 42.8 (2004): 3558-3565.
Lipp, Erin K., Samuel A. Farrah, and Joan B. Rose. "Assessment and impact of microbial fecal pollution and human enteric pathogens in a coastal community." Marine pollution bulletin 42.4 (2001): 286-293.
Love, David C., et al. "Human viruses and viral indicators in marine water at two recreational beaches in Southern California, USA." Journal of water and health 12.1 (2014): 136-150.
McCray, John. State of the science: review of quantitative tools to determine wastewater soil treatment unit performance. IWA Publishing, 2009.
Moore, D. F., J. A. Guzman, and C. McGee. "Species distribution and antimicrobial resistance of enterococci isolated from surface and ocean water." Journal of applied microbiology 105.4 (2008): 1017-1025.