New Laboratory Technologies · New Laboratory Technologies Karissa Culbreath, PhD D(ABMM) Scientific Director –Infectious Disease, TriCore Reference Laboratories Assistant Professor

New Laboratory Technologies

Karissa Culbreath, PhD D(ABMM)Scientific Director – Infectious Disease, TriCore

Reference LaboratoriesAssistant Professor – Department of Pathology,

University of New Mexico

Objectives

• Describe the current state of routine and molecular infectious disease testing

• Provide case examples of the benefits and challenges of molecular diagnostics

• Explore further advances in molecular and conventional testing to improve infection control

• Hospital Laboratory for 2 major health systems in New Mexico– Presbyterian Health System

• 8 Hospitals

– University of New Mexico Health System

• 2 Hospitals (Level 1 Trauma Center, Cancer Center)

• Commercial Physician and Hospital Clients throughout the state– Samples are delivered by courier

24/7

• Infectious Disease Laboratory Test Volume– >850,000 Infectious Disease tests

per year

TriCore Reference Laboratories

• Infectious Disease Laboratory Test Volume– >850,000 Infectious

Disease tests per year

• Esoteric and routine testing

• 18,000 molecular tests per month

• 37,000 culture-based tests per month

TriCore Clinical Microbiology Laboratory

Diagnostic Tools for Pathogens

Serology Antigen Detection

Culture Molecular Detection

How does it work? Detection of antibodiesDeveloped against pathogen

Detection of antigens produced by the pathogen

Isolation of pathogen through growth

Nucleic acidamplification detection of specific genetic target

Advantages Can detect acuteand past infection

Rapid, remains positive after treatment

Inexpensive High sensitivity,Usually rapid

Disadvantages Delay due to time to develop antibodies

Lower sensitivity, may be strain limitations

Time consuming,dependent on appropriate growth conditions

Expensive,Must know the target of interest

When to use With pathogens unable or difficult to culture

Specific diseasescenarios

Emergence of new pathogens, susceptibility testing, strain typing

When rapid/accurate diagnosis is necessary

You decide… PCR vs. Culture

• Patient with liver abscess secondary to cirrhosis

• Patient with 3-week history of cough consistent with pertussis

• Patient with diarrhea following chemotherapy

• Identification of patients colonized with MRSA for infection control

• Patient presents to the emergency room with symptoms suspicious for bacterial meningitis

From There to Here

“An individual who gets used to hard work can thereafter never live without it. Work is the foundation of everything in this world.” ~Louis Pasteur

From There to Here

“An individual who gets used to hard work can thereafter never live without it. Work is the foundation of everything in this world.” ~Louis Pasteur

PCR PCR

A Microbiologist or Physician

~ 2017 Microbiologist or Physician

10 Years Ago

Viral Loads HIV, HCV

Molecular Methods

Bacterial CultureC. difficile testingRespiratory VirusesHerpes VirusesHPVAFB and Fungal Cultures

Culture/Antigen/ Microscopy

MRSA Screening

Current State

Viral Loads HIV, HCV, CMV, etcGenotypingHPV DetectionHSV, VZVRespiratory VirusesStool Pathogens

Molecular Methods

Bacterial CultureAFB and Fungal Cultures

Culture/Antigen/ Microscopy

MRSA ScreeningHPVC. difficilePositive Blood Culture

The Next 3-5 Years

Viral Loads HIV, HCV, CMV, etcGenotypingHPV DetectionHSV, VZVRespiratory VirusesStool PathogensWhole Genome SequencingDirect Sequencing from Specimens

Molecular Methods

Bacterial Culture but fewer

Culture/Antigen/ Microscopy

MRSA ScreeningHPVC. difficilePositive Blood CulturePrimary Sterile Fluid SampleAFB and Fungal Cultures

Does MRSA Still Matter?

• 86% of invasive MRSA infections are hospital-acquired infections (HAI)

– 20% result in death

• Risk of surgical site infections is 30-fold higher in MRSA colonized patients

• Several states have legislative requirements for MRSA screening

• New impact from Affordable Care Act (ACA)

http://www.cdc.gov/hai/pdfs/toolkits/MRSA_toolkit_white_020910_v2.pdf

Natural History of MRSA Colonization

• Clearance of MRSA colonization: 12%-79%

• 50% of patients clear colonization at 88 weeks after initial documentation of colonization

• Risk factors for persistent colonization

– Comorbidities (decubitus ulcers)

– Residence at health care facility

Shenoy et al, BMC Infect Dis. 2014, 14:177

Community MRSA Rates

Community MRSA Rates 1.3 – 2.0%

MRSA Rate in ICU

ICU MRSA Rates – 4%-8%

MRSA Rate in ICU

ICU MRSA Rates – 4%-8%Nosocomial Acquisition – 2%

MRSA Rate in ICU

30% of MRSA positive will develop invasive disease

• Resistance to β-lactams in MRSA is conferred by the genetic element SCCmeccarrying the mecA gene

• mecA encodes an altered PBP2a which has a reduced affinity for β-lactam antibiotics resulting in resistance

• Gold standard for confirming MRSA is either detection of mecA or PBP2a

Genetics of MRSA

Performance and Utility of Culture

• Performance– Sensitivity 80%-89%

• Broth enrichment can increase sensitivity to near equivalence with NAAT, but increase TAT

– Specificity 92%-99%– Less expensive

• Benefits– Organism available for strain typing– Phenotypic test not limited by changes in molecular

characteristics

• Limitations– Longer TAT (24 hours)

Performance and Utility of Molecular Screening Method

• Performance– Sensitivity 82%-100%– Specificity 64%-100%

• Benefits– Rapid turn-around-time (1-3 hours)– Value is most evident when patients are not

preemptively isolated

• Limitations– May miss mecA drops outs– Emerging genetic resistant characteristics– Significantly more costly than culture

Emergence of mecC-MRSA

mecA PCR PBP2a Agglutination

Not detected Negative

FOX

OX

Emergence of mecC-MRSA

• mecC – emerged as a novel gene distinct from mecAthat confers resistance– Phenotypic analysis: Often cefoxitin resistant – oxacillin

susceptible– Molecular analysis: Distinct sequences won’t be detected

by molecular methods specifically targeted for mecA• Many assays have incorporated universal mec gene primers or add

mecC-specific primers

– Protein analysis: PBP2a encoded by mecC is distinct, resulting in false negative PBP2a results

• Further investigation for Cefoxitin resistant, PBP2a-negative S. aureus isolates

– Overall prevalence in humans is low but mecC must be recognized and tests must be able to detect

MRSA Screening Strategies

• Universal Screening– All patients upon admission, periodically thereafter

(i.e. weekly)

– More expensive

– Limited demonstration of impact

• Targeted Screening– Specific units or procedures (i.e. ICU)

– Less expensive

– Focuses on patients at highest risk for invasive infection

When and How to Screen?

Situation Need Possible Method

Limitations Benefits

Pre-SurgicalScreening

Decolonization and prophylaxis in elective procedures

Culture –MRSA/MSSA

May miss some low level colonized patients, not transferrable to emergency surgery

Low cost, faster TAT generally not needed

Universal Screening

Facilities without preemptive isolation or double beds

NAAT CostlyFast TAT necessary to cohort patients appropriately

Universal Screening (single bed facilities or with preemptive isolation)

Facilities with preemptive isolation, single beds

CultureMay miss some low level colonized patients

Less expensive, fast TAT not required for room assignment

Targeted Screening (i.e. ICU, ED)

Focus on high riskpopulations NAAT Cost

Focus cost on highest risk for greatest benefit

DecolonizationScreening

Identify patients to remove from isolation

CultureMay miss some low level colonized patients

Only detects viable organisms

• All NAAT screening methods are FDA-cleared for Nares/Nasal swab

• 30-40% of patients are exclusively colonized with MRSA at a different site

• Molecular testing is only as good as the specimen that is collected

All this screening, still MRSA infections, what are we missing?

Diagnostic Strategies for C. difficile Infections

GDH – Glutamate DehydrogenaseToxin A/B

DNA Targets

Current Testing Options

Two-Step Algorithm

• Screen for GDH and Toxin Production by EIA

• Reflex to molecular for discordant results

• Benefits– Detects toxin as a surrogate

of active disease only

• Limitation– Lower sensitivity could miss

C. difficile in low abundance

Molecular Testing Only

• Perform molecular testing on all sample submitted

• Benefits– Most sensitive method to

detect C. difficile

• Limitations– Could potentially pick up

colonized patients

Which method should be used?

2018 IDSA/SHEA Recommendation for C. difficile

• Pre and post analytical understanding of the patient, specimen and result are included in the guideline for test selection

• Well coordinated algorithms and response between laboratory and providers is required for optimal test interpretation

So… is it really CRE? Well, there’s

CRE and there’s

CRE…

Carbapenemase Detection

Understanding molecular terminology will save many headaches and heartaches

Carbapnemase Producing Organism• Carbapenemase Production

– Type and expression of carbapenemase enzyme (KPC, NDM, OXA)

– More likely to be plasmid mediated (transferable)

Carbapenem Resistant Organism• Carbapenem Resistance

– Bacterial species and presence of other resistance mechanisms (ESBL, AmpC)

– Reduced permeability/efflux pumps

– Presence of porin alterations

– Intrinsic resistance to carbapenems

• Proteus, Providencia, Morganella species have elevated imipenem MICs

Factors Influencing Carbapenem MIC

Ambler Class A Ambler Class B Ambler Class D

Serine Carbapenemase

Metallo-Carbapenemase

OXA Carbapenemase

KPCSMENMC-AIMIGESSFOSFCIBC

VIMGIMSIMNDMIMPSPM

OXAPSE

Detection of Carbapenemase

• Carbapenemase production usually initiates more aggressive infection control interventions

• CP-CRE possess a more stable and transferable for of resistance

• Understanding specific resistance mechanism is important for therapeutic decisions for novel antibiotics with activity against certain resistance genes

Carbapenemase Detection Methods

Modified HodgeTest (MHT)

CarbapenemaseInactivation Method (CIM)

Molecular (PCR) Whole Genome Sequencing

Accuracy Moderate High High High

Turn AroundTime

Next Day Next Day Same Day Several Days

Information Provided

Detection of carbapenemaseactivity

Detection of carbapenemaseactivity

Detection of specificcarbapenemasegenes

Detection of carbapenemresistance mechanisms

Limitations Poor sensitivity for NDM producers; poor specificity when AmpC present

None known Unable to detect novel carbapenemase

Quantity of bioinformatics data to analyze, especially in the setting of novel carbapenemase

Targeted Sequencing

• Targeted Amplicon Sequencing– 16s rRNA common bacterial target

– Broad coverage for most organisms• Loss of species resolution for some

genera

– Most common for clinical application from direct specimens

• Contamination from reagents and environment may be detected

– Application for microbiome analysis

Lefterova, MI et al. (2015) Next-Generation Sequencing for Infectious Disease Diagnosis and Management. J Mol Diagn 17:623-634

Broad Sequencing Sequencing

Lefterova, MI et al. (2015) Next-Generation Sequencing for Infectious Disease Diagnosis and Management. J Mol Diagn 17:623-634

• Whole Genome Sequencing– Unbiased sequencing of nucleic acid

(RNA and DNA)• Requires human sequences and

sequencing errors to the filtered

– Useful in cases when all other culture and molecular options have been exhausted

– Useful in characterization of organisms for strain typing in epidemiologic investigations

– May identify resistance markers that can help to predict susceptibility

Increased Detection of Pathogens: Impact of Direct Specimen Sequencing

Primary Pathogen Detected

Different Primary Pathogen Detected

Culture NegativeSequencing Positive

12 2 1

Source Culture Result Sequencing Result

Cerebellar Abscess S. intermediusA. aphrophilus

F. nucleatumCapnocytophaga spp.S. intermediusA. aphrophilus

Bile P. aeruginosa K. pneumoniaeC. braakiiP. aeruginosa

CSF Negative S. intermedius

Identification by sequencing can identify the cause of the infection, but there is no organism available for susceptibility testing

Challenges of Molecular Testing

• Increased detection of organisms, are they causing disease or innocent by-standers?

• Molecular testing will only answer the question you ask

• Will the test provide clinically actionable information?

• How do you treat a bug you’ve never heard of?

• Identification by sequencing but no organism available for susceptibility testing

• Routine Surveillance

– Active surveillance• Detection of Colonized

Patients

– Chart Review• Manual review of

previous culture results

• Informatics-Driven Surveillance

– Prior positive microbiology results with (EMR)

– System Alerts for risk factors

Informatics Driven Surveillance

Is the answer always more testing?

Informatics-Driven Surveillance

Maaike S. M. van Mourik et al. Clin Infect Dis. 2013;57:85-93

• Large sparsely populated geographical areas

• Remote locations

• Socioeconomic conditions

• Specimen collection and quality control

TriCore Reference Laboratory in New Mexico

Real-Time MDRO Database

Patient is admitted to healthcare facility

Database of MDRO Isolates

Au

tom

ated

IT Q

ue

ry History of MDRO

No History of MDRO

Preliminary Results

• In a 1 week query

– 2 facilities

– Approximately 8,897 patients screened

MRSA ESBL MDRO CP-CRE TOTAL

73 (0.82%) 31 (0.34%) 69 (0.77%) 1 (0.0001%) 174 (2.0%)

Screening Results

• 9 Patients routinely screened for MRSA (targeted screening practices)

• 4 patients potentially carriers of MRSA not on contact precautions

Facility A (culture) Facility B (NAAT)

MRSA Positive 2 3

MRSA Negative 2 2

Preliminary Results

• In a 1 week query

– Approximately 8,897 patients screened

MRSA ESBL MDRO CP-CRE TOTAL

73 (0.82%) 31 (0.34%) 69 (0.77%) 1 (0.0001%) 174 (2.0%)

• Isolated from urine• Isolated in outpatient setting• 5 months between CP-CRE isolate and

hospitalization• Admitted for suspected congestive heart failure• No cultures collected upon admission

Questions and Limitations

• Risk Stratification

– Increase Impact / Decrease Burden

– Refine risk

– Number of positive cultures in 365-day period

• Recent positive cultures (<30 days)

– Change notification/reporting timing for possible CRE

• Notification

– Notification to Infection Control?

– Notification to Bed Management System?

ARE ROUTINE CULTURES A THING OF THE PAST?

Automation in Microbiology

47

BEFORE AUTOMATION

A New Look for the Lab

AFTER AUTOMATION

Increased Pathogen Recovery with Automated Incubation

0

5

10

15

20

25

30

20

15/0

1

20

15/0

2

20

15/0

3

20

15/0

4

20

15/0

5

20

15/0

6

20

15/0

7

20

15/0

8

20

15/0

9

20

15/1

0

20

15/1

1

20

16/0

2

20

16/0

3

20

16/0

4

20

16/0

5

20

16/0

6

20

16/0

7

20

16/0

8

20

16/0

9

20

16/1

0

20

16/1

1

20

16/1

2

20

17/0

1

20

17/0

2

20

17/0

3

20

17/0

4

Iso

late

s

Aerococcus Isolates Per Month from Urine Cultures

Automation

Implementation

PreAutomaton PostAutomation

Average isolates per month

3.4 17.6

Isolates per 1,000 cultures

0.22 1.22

Age 77 76

Gender 25% M, 75% F 23.7% M, 77.1% F

WBC 69.0 60.4

Positive UA 48.1% 37.4%

Culbreath, K, unpublished data 2017

Increased Pathogen Recovery with Automated Incubation

January 2015 – March 2016

April 2016 –December 2016

N. gonorrhoeae

1 4

LeukocyteEsterase

Nitrate WBC Other GC Test Performed

Patient 1 Trace Negative 15-20 No

Patient 2 Large Negative 150 Yes

Patient 3 Large Negative >150 No

Patient 4 Not performed

Not performed

Not performed

No

Patient 5 Moderate Negative TNTC No

Culbreath, K, unpublished data 2017

Neisseria gonorrhoeae from WASPLab

Transformation of a Culture Plate to a Digital Image

1 mm

1 mm

1 mm

48

MegaPixe

l

12

MegaPix

el

3

MegaPix

el

High Resolution

Accurate coloration

Wide field depth

Quality Optics

What’s the value of a digital image?

• Competency assessment

• Training and review

• Clinical communication

• Colony recognition and identification

Algorithm-Based Colony Counting

• Separate Algorithm for Each Plate Side– Software identifies bi-plate and reads growth on each side using

different light/algorithm. Image is presented to users with the plastic separator oriented vertically.

Quantitation of Colonies Regardless of Morphology

y = 0.9539x + 2.1719R² = 0.9152

0

20

40

60

80

100

120

0 20 40 60 80 100 120

Man

ual C

ou

nt

Image Analysis Count

Colony Count

Colony Count

Linear (Colony Count)

Automated MRSA Detection

57,690 Sample Multi-Center Study• Automatic Detection and Segregation of

Positive from Negative MRSA Samples

Using Different Manufacturers’ Chromogenic agar

• Sensitivity of 100%• Specificity of 90-96% (varied by location)

• Detection of 153 Positive Plates that were

Missed Manually

Automated VRE Detection

104,730 Sample Multi-Center Study• Automatic Detection and Segregation of

Positive from Negative VRE Samples

Using Colorex VRE or Oxoid VRE

• Sensitivity of 100%• Specificity of 89.5% (varied by location)

• Detection of 498 Positive Plates that were

Missed Manually

Image Analysis on Chromogenic Media

Faron et al, J. Clin. Microbiol. March 2016 vol. 54 no. 3 620-624

Faron et al, J. Clin. Microbiol. October 2016 vol. 54 no. 10 2464-2469

Phenotypic Colony Recognition

Size

Morphology

Growth Characteristics

It’s E. coli!

Performance of Colony Recognition

Organism Classifications nCorrect

ClassificationPercent Unclassified Percent Misclassified Percent

Correct Gram Classification

Percent

Staphylococci 40 36 90% 4 10% 0 0% 40 100%

Candida species 38 37 97% 0 0% 1 3% 38 100%

Streptococci 40 27 68% 5 13% 8 20% 40 100%

Enterobacteriacae 72 64 89% 6 8% 1 1% 72 100%

Pseudomonas 37 31 84% 6 16% 0 0% 37 100%

Enterococci 40 40 100% 0 0% 0 0% 40 100%

• 267 Isolates were challenged against the Image Analysis Library

• Organisms were considered correct if within the appropriate classification

• Unclassified: No single classification provided

• Misclassified: Organism classified into the incorrect group

Image Analysis Concordance with Manual Interpretation

Value of Kappa

Level of Agreement

% of Data that are Reliable

0-0.20 None 0-4%

0.21-0.39 Minimal 4-15%

0.40-0.59 Weak 15-35%

0.60-0.79 Moderate 35-63%

0.80 -0.90 Strong 64-81%

Above 0.90 Almost Perfect

82-100%

Manual Interpretation Analysis

No GrowthNormal Flora/Contaminated

Culture Reviewfor ID/AST

% Agreement Kappa Value

Soft

war

e In

terp

reta

tio

n No Growth 2044 5 4 99.6% 0.98 (0.98-0.99)

NGUF/Contaminated (3 or

more organisms)7 360 8 96.0% 0.61 (0.59 -0.67)

Culture Review for ID/AST

15 194 2414 92.0% 0.92 (0.90- 0.93)

McHugh ML, Interrater reliability: the kappa statistic. Biochem Med. 2012 22(3):276-282

Summary

• Molecular testing is a valuable tool in the diagnostic arsenal

• Understanding the benefits and limitation of the testing is important for pre- and post- analytical decision making

• Laboratory informatics may be able to supplement unknown variables in infection control surveillance

• Routine culture and staining powered by artificial intelligence is being incorporated into routine laboratory workflow