Elsevier

Water Research

Volume 171, 15 March 2020, 115439
Water Research

Residential water heater cleaning and occurrence of Legionella in Flint, MI

https://doi.org/10.1016/j.watres.2019.115439Get rights and content

Highlights

  • L. pneumophila culturability rates were low in Flint, MI homes in 2016.

  • Removing sediment from water heaters decreased Legionella gene markers in some homes.

  • Influent chlorine <0.4 ppm was associated with Legionella gene markers in homes.

  • Water temperature was a key factor controlling levels of Legionella gene markers.

Abstract

After the Federal emergency in Flint, MI was declared in early 2016 in response to elevated lead-in-water and incidence of Legionnaires’ disease, concerns arose that contaminants in residential water heaters could continue to contribute to poor quality tap water. Here, a comprehensive field survey of residential water heaters (n = 30) and associated water quality was conducted and the subsequent effects of an aggressive manual water heater clean-out was determined, including draining the tank and removing sediments via brushing and flushing. Before cleaning, inorganics accumulated in the tank sediments did not serve as a source of metals measured at hot water outlets. After cleaning, hardness- (calcium, magnesium, silica) and corrosion-associated inorganics (lead, iron, copper, aluminum, zinc) decreased by 64% in samples from sediment cleanout drain valves. Culturable L. pneumophila was only detected in 1 home (3.3%) prior to cleaning and 2 homes (6.7%) after cleaning, thus quantitative polymerase chain reaction was used to quantify potential effects on unculturable strains despite the limitation of differentiating live and dead cells. After the cleaning protocol, Legionella spp. and L. pneumophila gene numbers decreased or remained non-detectable in 83% and 98% of samples, respectively. Homes with less than 0.4 mg/L influent free chlorine tended to have quantifiable Legionella spp. gene numbers in water entering the home and had elevated L. pneumophila and Legionella spp. gene numbers throughout the home plumbing. Also, Legionella spp. and L. pneumophila gene numbers were highest for water heaters set at or below ∼42 °C and significantly decreased >51 °C, consistent with Legionella’s preferred temperature range. Examination of the only home that had culturable L. pneumophila both before and after the cleaning protocol revealed that the organism was culturable from several sample locations throughout the home, including in water representative of the water main. Notably, the home was located in close proximity to McLaren Hospital, where an outbreak of Legionnaires disease was reported, and the water heater had a setpoint within the Legionella growth range of 44.2 °C. Considering that other factors were more strongly associated with Legionella occurrence and water heater sediment was not detectably mobilizing to tap water, it was concluded that water heater cleaning had some benefits, but was not an overarching factor contributing to possible human health risks.

Introduction

After switching from Lake Huron to the Flint River as the drinking water source in April 2014, widespread distribution system water quality problems occurred in Flint, MI, ultimately leading to a Federal Emergency declaration (Edwards et al., 2015; Rose, 2015; Masten et al., 2016; FEMA, 2016). Citywide lead-in-water contamination and elevated lead in children’s blood (Pieper et al., 2017, 2018; Hanna-Attisha et al., 2016) prompted a switch back to Lake Huron source water purchased from Detroit in October 2015. In January 2016, authorities announced that outbreaks of Legionnaires’ disease had occurred in summers of 2014 and 2015. Although the switch back to the Lake Huron water source in October 2015 reinstated corrosion control and improved disinfectant residuals, there was lingering concern about persistent public health threats.

All available data collected in early 2016 demonstrated relatively low levels of L. pneumophila in small single-story buildings (Schwake et al., 2016a, Schwake et al., 2016b; Rhoads et al., 2017; FACHEP, 2016), suggesting that L. pneumophila occurrence and Legionnaires’ disease incidence were primarily associated with large buildings. However, the State of Michigan and the Federal Emergency Management Agency were concerned about possible high levels of Legionella bacteria in home water systems, especially due to the high rates of water main breaks and red water from 2014 to 2016. Nationally, it is believed that 96% of reported Legionnaires’ disease is not associated with outbreaks, with the origin of 64% of reported cases never reliably being identified (Hicks et al., 2011; Shah et al., 2018). Residential plumbing systems are a potential source of unidentified Legionnaires’ disease (Stout et al., 1987, 1992; Pedro-Botet et al., 2002; Straus et al., 1996). Byrne et al. (2018) detected L. pneumophila serogroup 6 strains in 12% of Flint homes (n = 130) in late 2016, which differed from serogroup 1 clinical strains detected by the urine antigen test. All available clinical isolates were also found to belong to serogroup 1, while isolates recovered from tap water in early 2016 were found to belong to serogroup 1 or serogroup 6 (Garner et al., 2019), but of a different sequence type of serogroup 6 than that reported by Byrne et al. (2018).

Following the Federal Emergency declaration in Flint, there was concern that Legionella could have colonized water heater sediments during the water crisis, creating a long-term reservoir for Legionella. Sediments are known to react with disinfectants and reduce their efficacy, while also providing nutrients, shelter, and surface area for attached microbial growth, including for Legionella (Lu et al., 2015; Qin et al., 2017). Sediments also decrease heat transfer and energy efficiency (Weingarten, 1992). After the Flint Water Crisis, there were also general concerns that accumulation of iron and lead from water main breaks and corrosion, might increase disinfectant loss, microbial growth, and lead exposure, although hot water is not considered by EPA to be potable water used for consumption (drinking or cooking).

Current recommendations for water heater cleanout, including flushing, vacuuming, and/or dissolving sediments, are designed to remove hardness scale (Weingarten, 1992; Widder and Baechler, 2013). Water heater flushing protocols were developed following the Elk River 4-Methylcyclohexanemethanol (MCHM) spill. However, the protocols were not rigorously designed, validated, or verified in the field and there are concerns that flushing might have even increased consumer exposure to MCHM (Omur-Ozbek et al., 2016). Eventually, science-based recommendations were developed (Casteloes et al., 2015; Hawes et al., 2017; Ragain et al., 2019) for the water soluble chemicals of concern, but they were not designed for removal of contaminants that settled or deposited in plumbing.

In surveying the occurrence of Legionella, it is also important to evaluate all relevant factors that could contribute to its proliferation. In particular, water heater set point has been identified as a critical parameter for Legionella colonization of residential plumbing systems, with settings less than 49 °C (120 °F) facilitating Legionella growth in the storage tanks (Lee et al., 1988). While Michigan code allows residential water heater temperature settings up to 60 °C (140 °F), default or resident-preferred temperatures are typically lower due to a desire to decrease energy demand, reduce scaling, and prevent scalding. Gas heaters tend to have lower prevalence of Legionella than electric heaters, likely due to heating from the tank bottom as opposed to side mounted electrical elements (resistors) that create stratification (Alary and Joly, 1991; Dufresne et al., 2011; Brazeau and Edwards, 2013). In addition, while free chlorine levels in the distribution system had improved markedly by summer 2016 relative to during the period the city was using the Flint River water (Rhoads et al., 2017; Zahran et al., 2018), residents were using little water due to high water bills, readily available bottled water, and concerns about the safety of bath water (Roy and Edwards, 2019). There was also concern that the improved disinfectant residuals were not being consistently delivered to the residential plumbing due to low water use.

Here we evaluated the hypothesis that a thorough water heater sediment cleaning event would improve water quality at hot water outlets by reducing inorganic contamination, disinfectant demand, and Legionella proliferation and release. We conducted a comprehensive field sampling campaign during the Flint, MI Federal Emergency response, deploying a rigorous water heater sediment cleaning protocol at 30 single-family Flint residences located near the epicenter of the reported Legionnaires’ disease outbreaks at McLaren Hospital (MDHHS, 2018). The goal was to provide insight into the potential for water heater flushing to reduce risk of exposure to Legionella and other water contaminants following major corrosion or other massive contamination events.

Section snippets

Sample collection

Two sampling campaigns were conducted in July 2016, including 30 homes near the two hospitals in Flint, MI where the majority of the Legionnaires’ disease cases and highest water ages were reported. Six samples were collected from each home before and 5–6 days after performing the water heater cleaning protocol. Three of the samples were stagnant first draws from the (1) cold kitchen outlet, (2) hot kitchen outlet, and (3) shower head or bathtub spout sample of mixed hot and cold water.

Qualitative notes on cleaning protocol

For new heaters (<1 year old) or a few heaters that had been flushed regularly by the homeowner (n = 3), only 30–45 min was required to perform the entire cleaning protocol and achieve clear flushed water. However, in the typical case (n = 21), the procedure required between 90 and 120 min before the water from the tank ran clear of sediment. For worst-case heaters (n = 6), clear water was never achieved and flushing was discontinued at 2 h.

During the second week of sampling, after the heaters

Conclusions

The one-time cleanout of sediment accumulated in Flint, Michigan water heater tanks:

  • Reduced the amount of particulate inorganics released from the drain valve;

  • Had no major impact on the level of inorganics sampled at other outlets (kitchen, shower) within homes, indicating that water heater sediments do not serve as a regular source of inorganics at hot water point of use outlets; and

  • At least temporarily reduced the occurrence and levels of Legionella spp. and L. pneumophila gene copy numbers,

Funding

This work was funded by a grant from the Michigan Department of Environmental Quality (MDEQ). The author’s views expressed in this publication do not necessarily reflect the views of the MDEQ.

Declaration of competing interest

The authors declare that they have no known competing financial interest or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

We would like to thank the Flint residents that participated in this study. We would also like to thank members of the Flint Water Study team for dedicating their time to conduct the sampling, in particular: Pan Ji, Rebekah Martin, Laurel Strom,Owen Strom, Ni Zhu, Kimberley Hughes, Mariana Martinez, David Otto Schwake, Jeffrey Parks, Dongjuan Dai, Haniyyah Chapman, Philip Smith, Ethan Edwards, Ailene Edwards, Connor Brown, and Kandance Donalson.

References (55)

  • R.E. Brazeau et al.

    Role of hot water system design on factors influential to pathogen regrowth: temperature, chlorine residual, hydrogen evolution, and sediment

    Environ. Eng. Sci.

    (2013)
  • B.G. Byrne et al.

    Prevalence of infection-competent serogroup 6 Legionella pneumophila within premise plumbing in Southeast Michigan

    mBio

    (2018)
  • K.S. Casteloes et al.

    Decontaminating chemically contaminated residential premise plumbing systems by flushing

    Environ. Sci. Water Res. Technol.

    (2015)
  • S.F. Dufresne et al.

    Sporadic Legionnaires’ disease: the role of domestic electric hot-water tanks

    Epidemiol. Infect.

    (2012)
  • M.A. Edwards et al.

    Synergistic Impacts of Corrosive Water and Interrupted Corrosion Control on Chemical/Microbiological Water Quality: Flint, MI

    (2015)
  • Federal Emergency Management Agency (FEMA)

    President Obama Signs emergency declaration for Michigan

  • Flint Area Community (FACHEP)

    Enhanced Disease Surveillance and Environmental Monitoring in Flint, Michigan: Phase I Project Report as Cited in Edwards, M.A. (2018) FACHEP vs. The People of the State of Michigan: Part II the Problematic Birth of FACHEP. Located on Page 5 of PDF Download “NIH Pauli Book Chapter FOIA Responses Archival Samples.”

    (2016)
  • E.D. Garner et al.

    Whole genome sequence comparison of clinical and drinking water Legionella pneumophila isolates associated with the Flint water crisis

    Emerg. Infect. Dis.

    (2019)
  • M. Hanna-Attisha et al.

    Elevated blood lead levels in children associated with the Flint drinking water crisis: a spatial analysis of risk and public health response

    Am. J. Public Health

    (2016)
  • J.K.C. Hawes et al.

    Predicting contaminated water removal from residential water heaters under various flushing Scenarios

    J. Am. Water Work. Assoc.

    (2017)
  • D. Hayes-Phillips et al.

    Factors influencing Legionella contamination of domestic household showers

    Pathogens

    (2019)
  • L.A. Hicks et al.
    (2011)
  • N.A. Kozak et al.

    Distribution of lag-1 alleles and sequence-based types among Legionella pneumophila serogroup 1 clinical and environmental isolates in the United States

    J. Clin. Microbiol.

    (2009)
  • M. LeChevallier

    Monitoring Distribution Systems for Legionella pneumophila Using Legiolert. Poster Presented at the American Water Works Association Annual Conference and Exposition

    (2019)
  • T.C. Lee et al.

    Factors predisposing to Legionella pneumophila colonization in residential water systems

    Arch. Environ. Health

    (1988)
  • T.C. Lee et al.

    Growth of 28 Legionella species on selective culture media: a comparative study

    J. Clin. Microbiol.

    (1993)
  • J. Lu et al.

    Molecular survey of occurrence and quantity of Legionella spp., Mycobacterium spp., Pseudomonas aeruginosa and amoeba hosts in municipal drinking water storage tank sediments

    J. Appl. Microbiol.

    (2015)
  • Cited by (0)

    1

    Current affiliation: The American Cleaning Institute, Washington, DC.

    2

    Current affiliation: Virginia Tech, Department of Population Health Sciences, Blacksburg, VA.

    View full text