Rapid Phenotypic Antimicrobial Susceptibility Testing

Gunnar Kahlmeter

Martin Sundquist, Emma Jonasson, EUROSTAR-gänget

Rapid phenotypic AST

• Direct susceptibility testing

– AST directly on sample (mostly Urine)

– AST directly from the blood culture bottle

Cuts 24 hours off the standard procedure

• Potential problems:

– Inoculum??

– Purity of culture??

Rapid phenotypic AST

• “Early Reading” susceptibility testing

– Reading disk diffusion, or gradient tests, or automated testing earlier than stipulated in the standard method.

Cuts 24 hours off the standard procedure

• Potential problems:

– Some resistance mechanisms may not have time to express themselves phenotypically in 6, 7 or 8 hours

– Not standardised, but could be standardised by EUCAST

Why rapid ID and rapid AST? To increase the chance of patient survival, save time and

limit costs.

• Species ID determines the course of treatment, not only choice of antibiotic, but of many non-antimicrobial decisions.

• Empiric antimicrobial therapy will fail more often as resistance increases.

• To improve antibiotic stewardship – fewer cocktails, fewer ”to be on the safe side”, fewer ”trials-and-errors”

• Saves time

• Saves cost (Garcia-Vazquez SJID 2013)

• Because now we can!

• Every hour is important in the treatment of sepsis (septic shock) (Kumar CCM 2006)

• Inappropriate AB therapy 2 days longer hospital stay (Shorr CCM 2011)

• More Rapid Species ID and AST improves AB-therapy (Kerremans JAC 2008)

The lives saving diagnostic chain in blood stream infections

• The 24/7 rapid transportation (2 h) of blood culture bottles from patient to cupboard (incubation).

• The 24/7 access to cupboards for incubation and growth detection.

• Laboratory staff action on positive signals.

• Rapid ID – direct MT on bottles within 60 minutes of positive signal.

• Direct inoculation of AST from blood culture bottles.

• Early reading of AST – 6 to 8 hours of positive signal.

• Immediate alert and transfer of result.

The lives saving diagnostic chain in blood stream infections

• The 24/7 rapid transportation (2 h) of blood culture bottles from patient to cupboard (incubation).

• The 24/7 access to cupboards for incubation and growth detection.

• Laboratory staff action on positive signals.

• Rapid ID – direct MT on bottles within 60 minutes of positive signal.

• Direct inoculation of AST from blood culture bottles.

• Early reading of AST – 6 to 8 hours of positive signal.

• Immediate alert and transfer of result.

No EUCAST recommendation on direct AST

• EUCAST does not recommend: – Direct inoculation of AST from blood culture bottles.

– Early reading of AST – 6 to 8 hours of positive signal.

Since the inoculum is not standardised and the standard method is validated for 16 – 20 h incubation.

Urine - cultures are mixed and the pH variability may affect antimicrobial activity.

EUCAST

…considers…

•validating ”direct inoculation from blood culture bottles” and ”early reading” of disk diffusion.

…will not consider…

•validating ”direct inoculation from urines”

Direct AST of urines Each laboratory needs to

• Validate - confirm with standard methods

• Restrict to certain agent/species combinations • E.coli, Klebsiella pneumoniae

• Primary UTI agents (Piv-MEC, NIT, CDR, TRI, CIP)

• Define when to repeat/confirm result.

• Only report if pure culture and acceptable inoculum.

Direct AST from blood culture bottles

• 22/26 Swedish labora-tories perform Direct AST from positive blood cultures

• Early reading (after 6-8h incubation) is perfomed by 10 of 22

– Disk diffusion

– Vitek2

Åkerlund A Läkartidningen 2015

E.coli ATCC 25922 early reading

6h 18h

“Early reading – strategies”

1. Get it right – complete correlation - repeat testing not necessary.

2. Err on the side of caution – overcall resistance; use ECOFFs; repeat testing necessary

3. Identify “safe sensitive” – give the clinician a few safe alternatives; repeat testing necessary

EUROSTAR RAPID DISK

Metod

• EUCAST method with short incubation time

• 6h – E.coli

– K.pneumoniae

– Enterococci

• 8h – P.aeruginosa

– S.aureus

• 93% could be read after 6 or 8h

Lab

• Växjö – Manual reading

• Hvidovre – Manual reading

• AMC – Camera (Kiestra) with manual

adjustment. Reading every hour.

0

5

10

15

20

25

30

35

40

45

6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40

6 h

0

5

10

15

20

25

30

35

40

45

6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40

18 h

0

10

20

30

40

6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40

8h

E. coli vs. cefotaxime 5 μg

at 6, 8 and 18 h

E. coli vs. ciprofloxacin 5 μg

at 6, 8 and 18 h

0

5

10

15

20

25

30

35

6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44

6 h

0

5

10

15

20

25

30

6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44

18 h

0

5

10

15

20

25

30

6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44

8h

0

10

20

30

40

50

60

70

80

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

8h

0

10

20

30

40

50

60

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

18h

S. aureus vs. cefoxitin 30 μg

Disk diffusion – early reading

• After 6 – 8 hours

– Zone diameters for sensitive organisms are smaller than after 16-20 hours and

– Zone diameters for resistant organisms are bigger than after 16-20 hours

– The margin for errors shrink with 6-8h readings.

DIRECT AST DIRECTLY FROM URINE

DAST Urine

• Every day routine in many laboratories

• Increased usefulness with MALDI-TOF (direct?)

• Increased performance if pre-screening is performed?

DAST Urine

• ≥105 cfu/mL

• Singel species only

• 177 samples, 123 E.coli

• EUCAST CBP

• 855 tests

– 3 VME

– 5 ME

– 7mE

Sundqvist APMIS 2014

45

44 R I S

43

42 1

41

40 1

39 1 1 1

38 1 1 1 1

37 1 1 1

36 1 1 1 1 1 2

35 1 1 2 2

34 1 1 1 2 4 1 2 1 2 1

33 2 1 1 1 3

32 3 3 2 1 6 3 1 4 1

31 2 2 3 4 3 5 1 2 1 1 1

30 2 2 3 1 3 4 1 4 1

29 1 4 3 2 1 1

28 2 4 5 3 1 1 2

27 1 2 2 2 2 1 1

26 2 1 1 1

25 1 3 1

24 1 1 1 1

23 1 S

22 1

21

20 1 I

19 1

18 1 1

17 1 R

16

15

14

13

12

11 1 1

10 1 1

9

8

7

6 11

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

Ciprofloxacin 5 µg

Inhibition zone diameter (mm). Standard AST

Inhib

itio

n z

one d

iam

ete

r (m

m).

DS

T-U

Ciprofloxacin

DAST urine, Routine Örebro University Hospital april 2013

• 470 prov

• 53% Neg

• Gram pos 9%

• 10% Mixed cultures

• DAST recorded on 25%

98 E.coli (87%)

• 2 ME (TRI och MEC)

• 2/2 ESBL were found

10 Proteus spp

• 1 mE

Other Enterobacteriaceae

• No errors

• Rapid Species ID important

Sundman 2013

Antimicrobial Susceptibility Testing Directly from Urine

Samples – a Comparison between Standardised and

Direct Disk Diffusion Testing together with Direct

Species Identification using Matrix assisted laser

desorption/ionisation time of flight.

Jonas Olafsson

Klinisk mikrobiologi, Växjö

Linné-universitetet

Poster ECCMID 2014

44

43

42

41

40 1

39 1 1

38 1 1 1

37 1 1 1

36 1 1 2 1 1 2

35 1 1 2 2

34 1 1 1 2 4 1 2 1 2 1 1

33 2 1 1 1 3

32 3 3 2 1 5 3 1 4 1

31 1 2 3 4 3 5 1 2 1 1 1

30 2 2 2 3 4 1 4 1

29 1 4 3 2 1 1

28 2 3 5 3 1 1 2

27 1 2 2 2 2 1 1

26 2 1 1 1

25 1 2

24 1 1 1 1

23 1

22 1

21

20 1

19 1

18 1 1

17 1

16

15

14

13

12

11 1 1

10 1 1

9

8

7

6 10

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Inhibition zone diameter (mm)

EUCAST methodology

Inh

ibit

ion

zo

ne d

iam

ete

r (m

m)

D-A

ST

Ciprofloxacin 5 µg (n=188).

40

39

38

37

36

35

34

33 1

32

31 1 1 1 1

30 1 2 1 1 1 1

29 2 3 1 1

28 1 1 1 3 3 4 2 2

27 1 2 3 4 5 2 1 1 1

26 1 1 1 1 3 5 5 1

25 1 1 1 6 2 3 1

24 1 1 4 6 6 2 4 2

23 1 3 3 2

22 2 1 1 2 1 1

21 1 2 2

20 1 1 2 2

19 1

18

17 1 1 1

16 1

15 2 1

14

13

12

11 1

10 1

9 1

8

7

6 33

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

Inhibition zone diameter (mm)

EUCAST methodology

Inh

ibit

ion

zo

ne

dia

met

er (

mm

)

D-A

ST

Trimethoprim 5 µg (n=188).

Errors for P. mirabilis are marked with black boxes.

40

39

38

37

36

35

34

33

32 1

31

30 1 1 2 1 1

29 4 2 2 4 1 2 1 1

28 1 1 1 3 3 1

27 1 1 5 6 3 2

26 1 1 4 7 6 4 1

25 2 2 2 3 7 2 1

24 2 2 1 3 3 2 2 1

23 1 2 1 1 2 2 1 3

22 1 2 3 2 1

21 2 1 2

20 2 1 2 1 1

19 2 2 1 2 1 1 1

18 1

17 1

16 2 1

15 1

14 1

13 1

12 1

11 1

10

9

8 1

7

6 5

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

Inhibition zone diameter (mm)

EUCAST methodology

Inh

ibit

ion

zo

ne d

iam

ete

r (m

m)

D-A

ST

Mecillinam 10 µg

(E. coli, Klebsiella spp. and P. mirabilis, n=172).

40

39

38

37

36

35

34

33

32

31

30

29

28

27

26

25

24

23

22 2 1

21 1 3 4 1 3

20 1 4 13 2 7

19 8 7 16 5 2

18 1 1 4 7 15 8 4 3

17 2 8 5 3

16 2 1 3 6 3 1

15 1 3 1

14 1

13 1 1

12 1 1

11 1

10 1

9

8 1

7 1

6 16

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

Inhibition zone diameter (mm)

EUCAST methodology

Inh

ibit

ion

zo

ne d

iam

ete

r (m

m)

D-A

ST

Cefadroxil 30 µg (n=188).

Direkt resistensbestämning på urinprover för andra gramnegativer än Escherichia coli och

Klebsiella pneumoniae

Nadia Vickius

Oskar Ekelund

Poster ECCMID 2015

Klinisk mikrobiologi för Kronoberg och Blekinge

Upplägg

Konsekutiva urinprover med signifikant bakteriuri med någon av följande arter: P. mirabilis (n=29)

Enterobacter spp. (n=17)

Citrobacter spp. (n=22)

K. oxytoca (n=29)

Serratia marcescens (n=3)

Totalt (n=100)

Studien utförs under perioden mars 2013 - augusti 2014 (18 månader). Provsättning och avläsning utförs av på urinavdelningen tjänstgörande BMA.

D-AST vs. EUCAST, endast P. mirabilis Mecillinam 10 µg

35 R S

34

33

32

31

30

29

28

27 1

26 1

25 1 1

24 1 1 1

23 1 1 1 3 2 1

22 1

DST-U 21 2 1

20 1 1

19 1

18

17 1 S

16 1

15

14

13

12 R

11

10

9

8

7

6 3 1

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

EUCAST

Felklassificeringar av P. mirabilis (3/29 = 10.3%)

D-AST vs. EUCAST, endast P. mirabilis

Trimetoprim 5 µg

35 R I S

34 1

33

32

31

30

29 1

28

27

26 1 1

25 1 1

24 1 1

23

22 1

DST-U 21 1

20 1 S

19 2

18 1 1

17 1

16 1 I

15 1

14

13

12 R

11

10

9

8

7

6 11

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

EUCAST

Felklassificering av P. mirabilis (5/29 = 17%)

D-AST vs. EUCAST, Gentamicin 10 µg

35 R I S

34

33

32

31

30

29

28

27 1

26 1 1

25 1 3 2 1

24 1 1 2 3 2 2 1

23 2 1 1 4 1 3 2

22 2 2 1 2 3 2

DST-U 21 1 3 2 4 4 2 1 1

20 1 1 1 4 4 3 1

19 2 2 2 1 1

18 1 2 1 2 1 S

17 1 1

16

15 1 I

14 1

13

12 R

11

10

9

8

7

6

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

EUCAST

Felklassificering av P. mirabilis

Felklassificering av K. oxytoca

Problems with DAST on Urine

• Errors probably not a major problem if restricted to E.coli and Klebsiella pneumoniae

– Other species need more validation

• High cost if performed on every sample

– Pre-screening ??

– On demand/request

IN PIPELINE

Resistance detection using MALDI-TOF

• Detection of resistant subtypes (i.e. subtyping)

– CfiA +/- B.fragilis (Wybo 2011, Johansson JMM 2014)

• Detection of specific peaks associated with resistance (i.e. subtypning)

– Porin loss in K.pneumonaie (Cai JCM 2012)

• Detection of degradation of antibitics (i.e. Beta-lactamase detection)

– TEM, ESBL, Carbapenemases (Sparbier JCM 2012)

– Carbapenemases (Hrabak 2011-2014)

Susceptibility testing using MALDI-TOF

• Detection of changes in mass-spektra in the presence of breakpoint concentrations of antibiotics (i.e. quantitative MALDI)

– Incubate in BHI-broth +/- antibiotics 1h

– Centrifugation steps and extraction with standard

– Load on MALDI-tray and cover with HCCA

– Analyse and compare AUC

– SIR can be predicted using Breakpoint concentrations

Lange JCM 2014

AST and MALDI (so far…)

Advantage

• Rapid

• Broth

• Not thorough validated (only one study)

• Might be useful for specific ”hard to test” bacteria

Drawback

• Laborious

• Not quantitative

• Not automated

• Breakpoint concentrations

Microcalorimetry

• Measurement of Isothermal Microcalorimetry

• Breakpoint concentrations, 6h

• Urine as medium

Braissant JCM 2014

E.coli resistant to ampicillin and

ciprofloxacin

Inhibition Index

Combined Broth Dilution and PCR (FAST)

• 5ml BC broth centrifuged

• Suspension in MH +/-ab

• Breakpoint concentrations

• Incubation 0-8h

• 16SrRNA qPCR

• Meltcurve species

• Good results

• 9h

• 50.7 vs 66.3h

Beuving PLoS One 2011.

EJCMID 2014

Combined Broth Dilution and using PadLock probes and RCA

• Short (0.5-2h) preincubation of urine in LB-broth +/- ab

• Breakpoint concentrations CIP and TRI

• Addition of species specific PadLock Probes

• RCA

• Differences in ”Growth curves” SIR

• Good correlation

Mezger JCM 2015

Schematic illustration of the method.

Anja Mezger et al. J. Clin. Microbiol. 2015;53:425-432

Future perspectives

• Molecular based detection from whole blood – Black box

– Random access

– Affordable

– If accomplished: preliminary report in 6-8h?

• Culture (usually positive within 20h)

– MALDI-TOF + rapid DD definitive report in <30h?

But, to be rapid we need ”Microbiologistics”:

• Blood culture cabinets (also decentralized) available 24/7/365

• Microbiology open 24/7/365

• Skilled staff 24/7/365

• Not to forget other samples from the critically ill patient

• Take control of the transport organization and the referrals!

Thanks to

Växjö

•Emma Jonasson

•Gunilla Cederbrandt

•Robert Smyth

•Stina Bengtsson

•Jonas Olafsson

•Liselott Byhlén

•Åsa Johansson

•Erika Matuscheck

•Gunnar Kahlmeter

Hvidovre

Pia Littauer

Jenny Dahl Knudsen

AMC

Caspar Hodiamont

Caroline Visser

Nijmegen

Johan Mouton

Örebro

Kristina Sundman

Theresa Ennefors

Kerstin Falkenström