1

2
Improvement to Method 1623 to Better
Detect Cryptosporidium Oocysts

Carrie Miller
Cryptosporidium Lab Technical Liaison
Technical Support Center
Standards & Risk Management Division
Office of Ground Water and Drinking Water

1
Long Term 2 Enhanced Surface Water
Treatment Rule (LT2 Rule)
•Source water monitoring for Cryptosporidium
• EPA Method 1623 or 1623.1 with required quality control
•2006 to 2012 EPA Laboratory Approval/Oversight
•2015 to 2021 Integration of Approval/Oversight with
States
• Consistent with all other regulated analytes
• States are interested

2
Can We Enhance Program-wide
Data Quality and Consistency?
• Challenging matrices
• Improve accuracy and precision
– Method components
– Laboratory performance

3
 Method 1623
Sample

Filter

Centrifuge Tubes

IMS

10 L 100 l
Slide
4
Determinative Assay
• Fluorescence
• Size and shape
• Nuclei and Sporozoites
• DAPI
• DIC

5
47 Approved Cryptosporidium U.S. Laboratories
for LT2 Round 1, 2006 - 2012

Commercial
University
Public Water System 6
State/Regional Lab
Method 1623 Variable in Reagent Water

7
Blind Matrix Spike, Method 1623 ~50 Laboratories

Diatomaceous
Tennessee River Sediment
Earth

8
Observed and LT2-Projected Recovery Distributions
2.0

LT2 Assumed (curve)

Round 1 Observed (histogram)
1.5
Probabilitiy (Mass or Density)

1.0
0.5
0.0

0.0 0.2 0.4 0.6 0.8 1.0

Oocysts Counted / Oocysts Spiked

9
 Proficiency Test Results for Cryptosporidium Laboratories
100

90 Percent Recovery
RSD
80

70
Percent (%)

60

50

40

30

20

10

Round

10
13
11
 Sodium hexametaphosphate (NaHMP)
• Reduces filter fouling during ultrafiltration
• Improves detection in waste and finished waters
• Improves the dispersion efficacy of surfactants by
– sequestering ions associated with water hardness
– lowering the surface tension
– increasing the zeta potential of particles

12
13
Recovery using 5.0% NaHMP
n≥8

44%

14
Distribution of Observed Recovery

15
 Review of Laboratory Practice
• Laboratories with a low frequency of low recovery
• Same laboratories also had high accuracy and
precision for PTs in matrix and reagent water
• Five of eight did an IMS rinse step to remove debris

16
17
 Pellet Wash in
Microcentrifuge Tube
No Wash Wash

remove visible obstructions in samples with

extraneous debris and microbiota

(oo)cysts

2008 18
 Expense for Method Modification
• Cost increase 25¢ per sample
• Processing Time increases ~20 minutes per batch
• Microscopy may decrease ~10 minutes per slide
• Theoretically it’s possible to save time
– E.g. 8 samples in a batch for 20 minutes – 10 minutes/slide
(80m) would yield an hour saved at the microscope

19
 Method 1623.1 Available June 2012

Sample

Filter

Centrifuge Tubes

IMS

10 L 100 l
Slide
20
Improved Recovery in Challenging Matrices
100
Method 1623 Method 1623.1

80
Percent Recovery

60

40

20

0
1 2 3 4 5 6 7 8 9
NTU Reservoir Reservoir Reservoir River River Reservoir River GWUI Reservoir
0.6 6.8 21.0 251.0 10.6 0.8 7.4 0.1 1.9
21
Multi-Laboratory Improvement with
Challenging Matrix
100
Method 1623 Method 1623.1

80
Percent Recovery

60

40

20

0
1 2 3
GWUI Reservoir River
NTU 0.2 0.4 104.7

22
 Matrix Effects
• Degree of improvement achieved with 1623.1
appears to vary from matrix to matrix.
• Turbidity alone does not have clear
relationship to either method.
• Additional data on the water sources may
further inform an understanding of the impact.

23
Experimental Design for Multi-Laboratory
Evaluation of the Modifications
• Live Harley Moon isolate of C. parvum
• Same lot of sampling capsules and reagents
• 14 Sources used by public water systems
• 140 samples analyzed
– 70 source water
– 70 reagent water

24
 Distribution of Source Waters

River
Reservoir/Lake
Ground

25
 Variation in
Source Water Samples
• Turbidity: 1 to 53 NTU
• Conductivity: 33 to 885 µS
• pH: 6.5 to 8.5

26
Good Quality Control Practices

• Trip control
• Positive and negative controls
• Data verification and validation
– no outliers were detected

27
Results: Reagent Water (n=56)

• 60% Mean recovery

• Range of recovery was 34% to 73%
• 16.2% average within lab RSD
• No oocysts in negative control samples

28
 Results: Source Water (n=53)

• 61% Mean recovery

• Range of recovery was 26% to 80%
• 12.7% average within lab RSD
• No oocysts found in un-spiked samples.

29
 Accuracy Increased 27 Percentage Points in Source Water

Method 1623 Method 1623.1

1999 2012

Mean % Recovery 40 (n=29) 60 (n=56)

Reagent Water Mean RSD (%) 24 16
Standard Deviation 9 9
8 sources 14 sources

Mean % Recovery 34 (n=14) 61 (n=53)

Source Water
Mean RSD (%) 25 13
Standard Deviation 9 7

30
Method 1623.1 Variable in Reagent Water

33
31
 Improved Accuracy in River Matrix

Method 1623 Method 1623.1

9/14/2009 9/21/2009 4/25/2011 9/14/2009 9/21/2009 4/25/2011

Independent Samples

32
 WI State Laboratory of Hygiene
Cryptosporidium PT Program
ISO 17043 Accredited

Cryptosporidium Recoveries

No. Labs using

Mean (%) Median (%) Std Dev
Envirochek HV

1623 28 64 68 11.4

1623.1 12 65 64 11.4

33
 Method Flexibility

• Select from options for procedural components

– Multi-lab validated
– Historical standardized procedure
• Additional alternate test procedures
– Side-by-side method comparisons
– 2010 EPA Guidance for conducting method studies

34
 Examples of Procedural Options

• Spiking Suspensions • Stain

– WI State Lab of Hygiene – Aqua-Glo™
– EasySeed™ – Crypt-a-Glo™
– AccuSpike™ – EasyStain™
– MeriFluor®

35
 Summary
• Improved recovery with 1623.1 was more pronounced
for samples with low initial recovery with 1623.
• Recovery improvements with Method 1623.1 depend
upon the characteristics of each matrix.
• Public water systems with good matrix spike
recoveries using Method 1623 should not need to
change to 1623.1.

36
 More Information
• miller.carrie@epa.gov
Office of Ground Water and Drinking Water, Technical Support Center
1. http://water.epa.gov/lawsregs/rulesregs/sdwa/lt2/lab_home.cfm
2. http://www.youtube.com/watch?v=SEIH9wblGjo
3. http://www.youtube.com/watch?v=U-pBHvBeazs

37