Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE
Ecotoxicology and Human Environmental Health
Risk-based critical levels of Legionella pneumophila for indoor water uses Kerry A. Hamilton, Mark T. Hamilton, William Johnson, Patrick Jjemba, Zia Bukhari, Mark LeChevallier, Charles N. Haas, and Patrick L. Gurian Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.8b03000 • Publication Date (Web): 10 Jan 2019 Downloaded from http://pubs.acs.org on January 11, 2019
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 40
1 2 3
Environmental Science & Technology
Risk-based critical concentrations of Legionella pneumophila for indoor residential water uses
4 5
Kerry A. Hamiltona,b*, Mark T. Hamiltonc, William Johnsond, Patrick Jjembad,
6
Zia Bukharid, Mark LeChevallierd, Charles N. Haase, P.L. Guriane
7 8
a
9
b The
School for Sustainable Engineering and the Built Environment, Arizona State University, Biodesign Institute Center for Environmental Health Engineering, Arizona State
10
University
11
c Microsoft
Machine Learning Research Group, 1 Memorial Drive, Cambridge, MA 02142
12
d American
Water Research Laboratory, 213 Carriage Lane, Delran, New Jersey 08075
13
e
Drexel University, 3141 Chestnut Street, Philadelphia, PA 19104
14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
* Corresponding author. Kerry A. Hamilton. Mailing address: School for Sustainable Engineering ad the Built Environment, Arizona State University, 781 S Terrace Road, Tempe, Arizona 85281. E-mail address:
[email protected] 30 1 ACS Paragon Plus Environment
Environmental Science & Technology
Page 2 of 40
31
ABSTRACT
32
Legionella spp. is a key contributor to the United States waterborne disease burden. Despite
33
potentially widespread exposure, human disease is relatively uncommon, except under
34
circumstances where pathogen concentrations are high, host immunity is low, or exposures
35
to small-diameter aerosols occurs. Water quality guidance values for Legionella are
36
available for building managers but are generally not based on technical criteria. To address
37
this gap, a quantitative microbial risk assessment (QMRA) was conducted using target risk
38
values in order to calculate corresponding critical concentrations on a per-fixture and
39
aggregate (multiple fixture exposure) basis. Showers were the driving indoor exposure risk
40
compared to sinks and toilets. Based on aggregate fixture exposures, critical concentrations
41
depended on the dose response model (infection vs. clinical severity infection, CSI), risk
42
target used (infection risk vs. disability adjusted life years on a per-exposure or annual
43
basis), and fixture type (conventional vs. water efficient or “green”). Median critical
44
concentrations based on exposure to a combination of toilet, faucet, and shower aerosols
45
ranged from ~10-2 to ~100 CFU per L and ~101 to ~103 CFU per L for infection and CSI dose
46
response models, respectively. As infection model results for critical L. pneumophila
47
concentrations were often below a feasible detection limit for culture-based assays, the use
48
of CSI model results for non-healthcare water systems with a 10-6 DALY pppy target (the
49
more conservative target) would result in an estimate of ~20 CFU per L (arithmetic mean of
50
samples across multiple fixtures and/or over time). Single sample critical concentrations with
51
a per-exposure-corrected DALY target at each fixture would be 1050 CFU per L (faucets),
52
4.3 × 105 CFU per L (toilets), and 25.2 CFU per L (showers). The absence of detectable L.
53
pneumophila may be appropriate for healthcare or susceptible population settings.
54 55 56
Keywords: Opportunistic pathogens; reverse quantitative microbial risk assessment
57
(QMRA); Legionella pneumophila; monitoring; building water quality; green building; efficient
58
water fixtures 2 ACS Paragon Plus Environment
Page 3 of 40
59
Environmental Science & Technology
1. Introduction
60
The importance of opportunistic pathogens such as Legionella has been increasing in
61
recent years, with Legionella spp. identified in recent US Centers for Disease Control and
62
Prevention (CDC) reports as the most common cause of waterborne disease outbreaks in
63
the US 1, 2. Legionella causes illness primarily in individuals with underlying health
64
conditions, and/or the elderly. Infection occurs when aerosols containing the bacteria are
65
inhaled or aspirated by a susceptible host. Recent reviews of environmental sources of
66
Legionella infections for sporadic and outbreak-associated cases 3-5 have emphasized the
67
importance of building water and cooling tower design and maintenance.
68
Monitoring routinely for Legionella is not typically practiced in premise plumbing systems
69
except when legionellosis cases are associated with a facility. While Legionella occurrence
70
in premise plumbing systems is not uncommon 6-13, monitoring for Legionella spp. on a
71
routine basis may not be a cost-effective measure 14. However, in order to validate aspects
72
of a water safety plan or management strategy, knowledge regarding interpretation of
73
Legionella spp. sampling results can provide information regarding potential risks.
74
Concentrations of L. pneumophila in cold tap water have been reported up to ~105 gene
75
copies per L 12, 15, 16 and ~104 colony forming units (CFU) per L 8, 10, 11, 17, and up to ~107 CFU
76
per L in hot water 8, 18 .
77
Existing water quality guidance values for Legionella spp. are available for building water
78
quality managers to inform the interpretation of measurements made in their water systems,
79
with potable water values ranging from 102 to 105 colony forming units (CFU) per L
80
associated with various desired water management actions (Table 1). In one case, a French
81
guideline specifies faucets > toilets.
391
It is noted here that previous assessments have demonstrated that the appropriate
392
measure of risk is the average of multiple exposures 56. Therefore when few data are
393
available in the literature to assess the time variability of L. pneumophila concentrations and
394
doses as in the current case, the annual exposure dose in this study can be regarded as a
395
time-averaged arithmetic dose even if time variability is significant 23, 45. Consequently, while
396
critical concentration values calculated in association with per-exposure scenarios might be
397
interpreted as a single sample concentration, critical concentration values associated with
398
annual risk scenarios might be interpreted as average concentrations over multiple sampling
399
events. A major limitation of the guidance documents summarized in Table 1 is that specific
400
sampling locations, frequency and timing of samples taken at a given location, and
401
statistically rigorous interpretation of sampling results is not specified. Simulating a full three-
402
dimensional space for concentration, exposure frequency, and risk is recommended as a
403
follow-on to this analysis and could help to customize risk findings to sampling results
404
observed at a particular building or other setting.
405
As demonstrated here (Figure 4), the ultimate decision regarding concentration limit
406
values is a function of which risk target is used (e.g., 10-4 annual probability of infection or
407
10-6 DALY pppy). Therefore, an analysis of how timing and extent of sampling might affect
408
critical concentration conclusions is beyond the scope of the current set of models but is
409
recommended for further analysis. As actions within the context of a water management plan
410
are typically developed on a case-by-case basis, the current modeling approach can allow
411
for lower-risk facilities to adopt different cut-offs for action.
412
For the combined exposures to multiple fixtures, the concentration values (median
413
10-2 to 103 L. pneumophila per L) overlap with some of the current guidance values (102 to
414
105 per L, with species not specified), although the guidance values would be on the higher
415
end of the simulated ranges. Specifically, median critical concentrations calculated in the
416
current models ranged from ~10-2 to ~100 CFU per L and ~101 to ~103 CFU per L for 15 ACS Paragon Plus Environment
Environmental Science & Technology
Page 16 of 40
417
infection and CSI dose response models, respectively. All guidance values reviewed were
418
above infection dose response model estimates, and the corresponding risk values for
419
various concentrations specified in the guidance can be interpreted directly from Figure 3.
420
Most guidance values were consistent with, or had some management actions associated
421
with, a CSI aggregate exposure model critical median concentration of
