1. The Legiolert Method: Description, Validations and Drinking Water Study ASDWA, March 29, 2018

2. What is Legionella?
Bellevue-Stratford Hotel, 2000 attendees, 221 contracted/34 dead

3. Where does Legionella live?
Potable Water
Nonpotable Water

4. Biofilm – stable structures and breeding grounds

5. Infectivity – Legionella are aerolized
Shower aerosol
Cooling tower aerosol

6. Why should you be concerned with Legionella?
The # of cases of legionellosis have increased 286% over the past 14 years
8,000 to 18,000 people contract legionellosis in the U.S. each year (estimated)
5-15% of the known cases of legionellosis are fatal; up to 25% if contracted in healthcare
Number of Legionnaires’ disease cases by month and 12-month moving average, EU/EEA,
2011−2015
Number of cases
12-month moving average
Reported cases of legionellosis per 100,000 population, by year — United States, 2000–2014
Vital Signs: Deficiencies in Environmental Control Identified in Outbreaks of Legionnaires’ Disease — North America, 2000–2014
European Center for Disease Protection and Control (ECDC). Annual Epidemiological Report for 2015 Legionnaires’ disease.

7. L. pneumophila is the # 1 waterborne pathogen
Waterborne outbreaks associated with drinking water by state/jurisdiction,
LP included since 2000, ALL fatalities have been caused by Lp.
Benedict KM, Reses H, Vigar M, et al. Surveillance for Waterborne Disease Outbreaks Associated with Drinking Water — United States, 2013–2014. MMWR Morb Mortal Wkly Rep 2017;66:1216–1221

8. ECDC Legionnaires' disease in Europe, Surveillance Report 2009-2015
Not all Legionella are equally pathogenic:
L. pneumophila vs. other species: culture-confirmed cases
ECDC Legionnaires' disease in Europe, Surveillance Report
Year n
% L. pneumophila
2009
448 99
2010
652 98
2011
600 96
2012
661
2013
691
2014
777 95
2015
890
Species
n (2015)
% of cases
L. pneumophila
855
96.1
L. anisa 1
0.1
L. bozemanii 2
0.2
L. longbeachae 7
0.8
L. other species 13
1.5
L. species unknown 12
1.3
European Center for Disease Protection and Control (ECDC)
Invoke WHO recommendations here as well (doc available upon request)

9. World Health Organization (WHO)
WHO Recommendation paper September 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)
“Focus verification monitoring on L. pneumophila instead of Legionella spp. This is the approach taken by France, for example (Government of France, 2017). L. pneumophila is the most significant causative agent of legionellosis in Europe.”
“When a microbiological parameter is defined taxonomically, the parameter is much less prone to ambiguous results. This also allows for the development of alternative methods, as the endpoint is defined, while in the case of Legionella spp. the endpoint is defined by the culture method. Note that this has also been the rationale for changing from faecal coliforms to E. coli in the previous revision of the Directive.”
Testing for Lp is most protective of human health and more cost effective for utilities, e.g. they are not shutting down or mitigating based on non-pathogenic bacteria.

10. Legionella is everywhere, but it can be managed
Treatment is done primarily to control, not to eradicate in a single event

11. Agencies and organizations know the importance of managing the risk of Legionnaires’ disease, but there are currently no rules for this in the US for public water systems, just reference in 40 CFR 141. But the new standards and public news stories are pointing the finger at PWS’, should PSW test to understand their risk and role in managing LD?

12. Testing for Legionella: What do most labs do
Testing for Legionella: What do most labs do? The Gold Standard = Culture
Most internationally regulated methods are strong supporters of a culture-based method

13. Different methods = different results
- Updated 2017
Examples of results on different media with contaminated samples

14. IDEXX Legionella test- Legiolert

15. Legiolert™, a next generation culture test
Ideal for routine monitoring and compliance
Highly specific, little background interference
Highly reproducible and repeatable
Simple to use, color reaction similar to Colilert®
7 days to yield a confirmed result
Can be used for potable and nonpotable matrices
Matrix-specific protocols
Quantification by most probable number (MPN), equivalent to CFU
Counts of up to 2,272 per test, much higher than petri plates
Accreditation available through programs, AIHA and TNI
Detects Legionella pneumophila: All serogroups
IDEXX Confidential

16. Reaction with L. pneumophila
Legiolert platform
Unique 100 mL “Quanti-Tray” device
6 large wells (overflow)
90 small wells (resolution)
Counts L. pneumophila; from MPN
Blister pack reagent with 12 month shelf life
Reaction with L. pneumophila
Negative Sample

17. Legiolert protocol: potable water
Hardness Supplement
100 mL water sample
100 mL water sample
100mL Legiolert +
sample
Determine water hardness
Add 0.33 or 1mL hardness supplement
Add Legiolert reagent to water sample
7 days =
Confirmed result
Pour into Quanti-Tray, seal, and incubate 39°C

18. Legiolert protocol: nonpotable water
Water sample
Legiolert pretreat-ment
100mL sterile diluent
100mL Legiolert +
sample
Add Legiolert reagent to
100 mL sterile diluent
Mix 2 mL sample with 2 mL pretreatment, wait 60 seconds
Add 2 mL treated water sample
7 days =
Confirmed result
Pour into Quanti-Tray, seal, and incubate 37°C

19. Validation of Legiolert
Legiolert performance ─ ISO Validation Report
Validation of Legiolert
ISO/TR 13843:2000(E) Water Quality – Guidance on validation of microbiological methods
Metric
Value
Sensitivity
98%
Specificity
> 99%
False positive rate
< 0.01%
False negative rate
4.20%
Efficiency
Repeatability
< 0.01
Reproducibility
ISO statistics interpretation
Sensitivity = Of all the wells where there were truly Legionella, how many did Legiolert find?
Specificity = Of all the wells that were truly negative, how accurate was Legiolert?
FP rate = Of all positives, how many were false?
FN rate = Of all negatives how many were false (failures to detect)?
Efficiency (E) = What fraction of all results were the true results?
Sensitivity = TP/TP+FN a / (a+b) = 639 / (639+14) = 0.98
Specificity = TN/FP+TN d / (c+d) = 317 / (0+317) = > 0.99
Selectivity = log[TP+FP/TP+FN+FP+TN] log10 [(a+c) / (a+b+c+d)] = log [(639+0) / ( )] =
FP rate = FP/TP+FP c / (a+c) = 0 / (639+0) = < 0.01
FN rate = FN/FN+TN b / (b+d) = 14 / (14+317) = 0.042
Efficiency (E) = TP+TN/TP+FN+FP+TN (a+d) / (a+b+c+d) = ( ) / ( ) = > 0.99
Repeatability: closeness of the agreement between the results of successive measurements of the same measure and carried out under the
same conditions of measurement (same user)
Reproducibility: closeness of the agreement between the results of measurements on the same measure and carried out under changed conditions of
measurement (different user)
a = number of positive wells found to contain L. pneumophila (true positives);
b = number of negative wells found to contain L. pneumophila (false negatives);
c = number of positive wells found not to contain L. pneumophila (false positives);
d = number of negative wells found not to contain L. pneumophila (true negatives).
Link to Report:
IDEXX Confidential

20. Legiolert Trials Potable and Nonpotable water Independent laboratories
All methods confirmed by secondary culture and serotyping when necessary
Split samples analyses
Samples were obtained from sites/customers during routine business/laboratory operations in order to mimic performance in a real world scenario
Data segregated for L. pneumophila by isolate speciation
IDEXX Confidential

21. ISO-11731-1/2 Trials – Potable Water
Trial 2 – Legiolert vs. ISO /2
6 labs
846 samples
100mL and 1mL analyzed for both methods
1991/2034 wells confirmed
97.9% true positive
Statistic
Relative difference
Lower 95% CI
Upper 95% N
Legiolert 100mL vs. ISO 100mL Lp
89.5%
73.0%
106.1%
444
Legiolert 1mL vs. ISO 1mL Lp
-2.0%
-14.4%
10.4%
277
Legiolert 100mL vs. Best Lp
6.6%
-8.0%
21.1%
527
Legiolert 100mL vs. Best all Lspp
-8.1%
-24.1%
7.9%
537
0/199 wells FP
Legiolert detection of L. pneumophila only is still not statistically different compared to detection of all L. spp by ISO /2 in this study

22. ISO-11731-1/2 Trials – Potable Water
Trial 2 – Legiolert vs. ISO /2
McNemar’s Analysis (presence/absence)
Higher incidence of false negatives w/ISO in 100mL and best condition
Legiolert 100mL vs. MF 100mL - L. pneumophila
Legiolert
chi2
p-value
Reject/Accept Null + -
27.38
0.0000
Reject
ISO
349 22 73
284
Legiolert 1mL vs. MF 1mL - L. pneumophila
1.222
0.2689
Accept
178 55 44
484
Legiolert 100mL vs. ISO best - L. pneumophila
10.98
0.0009
445 26 56
281
13/56 Lp+ ISO- were>1cfu/mL
3/26 LP- ISO+ were>1cfu/mL

23. What about the rest of the world?
Legiolert Experience Program (LEP) – small scale validations of Legiolert vs. global standard methods
Participant
Country
# of data points
Waters tested
Pathwest
Australia 23
P/NP
FASS
In Progress P
MUV
Austria 31
CNR
France 80
ISHA
Idrogeolab
Italy 41
ISS
124
LAIST
Portugal
129
PUB
Singapore 29
PHE UK 96
ALSglobal
130
IDEXX Confidential

24. Conclusions Legiolert – accurate and easy
Ideal for routine monitoring, very easy to use and just as accurate and sensitive as cumbersome plate culture methods
Detects L. pneumophila, the primary clinical and environmental disease-causing agent
Easy test means more laboratories will be able to offer this as a reliable culture option
IDEXX Confidential

25. Legiolert Drinking Water Study
IDEXX Confidential

26. Legiolert Drinking Water Study
Project Objectives
Allow utilities to gain experience with the Legiolert test.
Gain experience with training and QA/QC procedures
Obtain feedback from utilities on the ease of use for the Legiolert test
Development of a Legionella monitoring program
Obtain preliminary information on Legionella occurrence in distribution systems
Obtain information on Legionella serotypes in distribution systems
IDEXX Confidential

27. Study Locations

28. Study Design
Each location participated in training conducted by IDEXX personnel and conducted QA/QC tests with known positive and negative controls.
Incubator temperature and humidity
Development of a sampling plan to collect samples representative of the distribution system
Locations that were routinely monitored (e.g., TCR sites, storage tanks, etc.)
Raw Water
Treatment plant effluent

29. Sample Collection Samples Quantity Raw water 4 Plant effluent
Distribution system 48
Total tests 56
# of utilities 12
Study total # tests
672

30. Water Quality Parameters
Parameters to test
Test performed
L. pneumophila (MPN/mL)
Legiolert
TC/EC (MPN/mL)
Colilert
HPC (MPN/mL)
Simplate for HPC
Free/Total chlorine
DPD Method
Temperature
Thermometer pH
pH strip or electrode
Total organic carbon
SM5310B

31. Results to Date 7 of 12 utilities reporting
265 samples analyzed – no positive results
Raw water
Plant effluent
Distribution system samples
Chlorine residuals ranged from:
0.15 to 1.7 mg/L for 4 of the free chlorine systems
0.4 to 2.6 mg/L for 3 of the chloramine systems
Temperatures ranged from 2 to 20oC
pH levels ranged from 6.9 to 9.5
Speak to the fact that even zeros mean something; unsure of what results would be in warmer weather. AND we’re not looking in suspect areas or buildings

32. Responding to Positive Samples
Serotype positive samples by mailing positives to the American Water laboratory.
Compare water quality data (free/total chlorine, etc.) for the site compared to historical levels. Determine if any anomalies exist.
A repeat sample should be collected from the sample site within 24 hours of reporting the positive Legiolert sample.
If the repeat sample is negative, the site should be re-sampled at the regular interval per study protocols.
A second positive Legiolert sample should trigger a Level 1 TCR assessment. Check in with DOH/regulator.
This is very similar to a TCR Level I Assessment, this is very familiar to the ADWA group
IDEXX Confidential

33. Responding to Positive Samples
Flush the area the positive site. If the positive is from a reservoir, consider draining and cleaning the reservoir.
Initial response is based on presence / absence for L. pneumophila. Reassess if study results prompt a re- assessment of this approach, OSHA guideline concentrations and European Union guidance can be considered in establishing possible action triggers.
Multiple detections of L. pneumophila in the distribution system should trigger close consultation between the utility and regulator regarding follow up actions and communication to other parties.
Multiple detections should trigger action similar to a TCR Level II Assessment
IDEXX Confidential

34. OSHA Guidelines
Table III:7-1. COLONY FORMING UNITS (CFU) OF LEGIONELLA PER MILLILITER
Action
Cooling tower
Domestic water
Humidifier 1
100 10 2
1,000
Action 1:  Prompt cleaning and/or biocide treatment of the system.
Action 2:  Immediate cleaning and/or biocide treatment. Take prompt steps to prevent employee exposure

35. EU Guidelines
Point out these limits are per LITER, previous slide is per ML.
European Technical Guidelines for the Prevention, Control and Investigation of Infections Caused by Legionella species. June 2017.

36. Discussion Questions
Does the design of this study seem reasonable and what else could be done to get a more complete picture of distribution systems? Warmer weather testing? Additional systems?
Understanding that drinking water isn’t sterile, and L pneumophila exists at some low level, what actions seem reasonable, based on some pre-set level of occurrence?
There is a trade-off between monitoring all Legionella species using a very complicated methodology, and focusing on the principle pathogen, L. pneumophila, with an easy to use test.
Does it make sense to use an easier test that focuses on the pathogen?
Let’s not re-do crypto and look for everything, let’s look for the human pathogenic bacteria, L pneumophila
IDEXX Confidential

37. Mark W. LeChevallier, Ph.D.
Dan Broder, Ph.D.
IDEXX Water R&D
Phone:
Mark W. LeChevallier, Ph.D.
Dr. Water Consulting, LLC
Phone: