Automated systems: an overview Robin A Howe Cardiff of presentation on Master Slide Insert name
AST standards • MIC taken as standard measure of susceptibility – Agar dilution – Broth dilution – Microbroth dilution
• ISO standard – Broth microdiluion – Mueller-Hinton Broth – Inoculum 5 x 105 cfu/m. L
AST standards • MIC taken as standard measure of susceptibility – Agar dilution – Broth dilution – Microbroth dilution • S/I/R Breakpoints defined by BSAC/EUCAST/CLSI
AST standards • MIC taken as standard measure of susceptibility – Agar dilution – Broth dilution – Microbroth dilution • S/I/R Breakpoints defined by BSAC/EUCAST/CLSI • MIC/BP supplemented by “expert rules” for clinical use
The rationale for automated systems • More reliable AST results – ? closer to “standard” – Reduced scope for “user” error – Reproducible • Improved ID linked to AST • Embedded expert rules Automated systems: an overview
Identification required for interpretation of BP & expert rules
Welsh variability in ID of Enterobacter/Klebsiella/Serratia
The rationale for automated systems • • • More reliable AST results Improved ID linked to AST Embedded expert rules Improved speed of results Increased number of agents tested Automated systems: an overview
Welsh Gram positive panel Welsh Gram negative panel
The rationale for automated systems • • More reliable AST results Improved ID linked to AST Embedded expert rules Improved speed of results Increased number of agents tested Laboratory workflow Staff skill mix Epidemiological software Automated systems: an overview
Available systems • Micro. Scan Walk. Away • Biomerieux Vitek 2 • BD Phoenix
Micro. Scan • 96 well plates • AST based on conventional microbroth dilution MIC • ID currently biochemical
Microscan • Conventional panels • Enterobacteriaceae, gram-negative nonfermenters, staphylococci, enterococci • Combo, MIC-only, ID-only • Rapid panels – 2 - 2. 5 hour ID (pre-formed enzymes) – AST for Gram negative organisms (MIC > 4 hrs) • Synergies plus panels – 2 - 2. 5 hour ID (pre-formed enzymes) – Broth microdilution “Read-when-ready” MIC – Resistance detection flagged in as few as 4. 5 hours – All results finalized in 16/18 hours • Additional – Yeast ID, Anaerobe ID, HNID – MICro. STREP plus ® - dry format – ESβL plus ® -ESβL Confirmatory Panel • Panels read by – Walk. Away® systems – auto. SCAN-4 – Manual
Microscan • Expert rules within Lab. Pro Alert system – Pre-loaded – customisable
Microscan • Expert rules within Lab. Pro Alert system – Pre-loaded – customisable
Vitek 2 • AST based on broth susceptibility with specific cards • ID currently biochemical • Growth/reactions “continuously” monitored
Vitek 2 • ID Cards available for – – – Gram positives Enterobacteriaceae Non-Enterobacteriaceae Neisseria/Haemophilus etc Anaerobes Yeasts • AST Cards available for – Gram positive cocci – Gram negative bacilli – Yeasts • Up to 22 antibiotics tested on each card
VITEK 2 – MIC Interpretation of Raw Data Growth Parameters MIC Raw Data % Change Slope Area SLPAR 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 TIME (hours)
Vitek 2 • Expert rules within the Advanced Expert System (AES) – Extensive Knowledge Base • >20, 000 MIC distributions • >2, 500 Phenotypes • >200 Resistance Mechanisms
Vitek 2 CARD Type Time to Result GN ID GP ID 2 to 10 hours 2 to 8 hours NH ID 6 hours ANC ID YST ID GN AST GP AST-YS 01 6 hours 18 hours 2 to 18 Hours 6 hours
Phoenix • AST based on broth susceptibility with specific cards • ID currently biochemical • Growth/reactions “continuously” monitored
Phoenix • AST only or Combo panels available for – GN (Enterobacteriaceae/ Nonfermenters) – GP (Staphylococci/ Enterococci/ Streptococci) – Streptococci ID/AST
Phoenix • Readings for each well taken and calculated every 20 minutes – – – – ID uses red, green, blue and UV light sources AST measures turbidity and color change Each well divided into pixels Readings taken for each pixel Entire process takes 7 minutes Pixel data analyzed for each well Readings for each well compared to previous reading and baseline • Well is complete when specific amount of change is not seen • Substrate/antimicrobic algorithm dependent – Instrument will post results when enough data collected
Phoenix • Expert rules within BDXpert based on CLSI/SFM/DIN/EUCAST rules bases
Comparison of systems
Performance • ID – Correct to genus >95% – Correct to Species>90% – Difficulties with difficult organisms • Non-fermenters • Shigella etc • AST – Concordant with other methods >95% – Specific issues with specific drug-bug combinations & certain instruments • VISA/VRSA • Beta-lactams & Pseudomonas
Performance • ID/AST performance relies on correct ID, correct AST plus appropriate expert interpretation – Incorrect ID can lead to incorrect AST • Up to 5% errors mean that results should be examined critically
Conclusions • Automated systems can provide – Reliable AST integrated with ID & expert rules – Improved speed of results • Caution about staff skill mix and interpretation of results • Microscan/ Vitek 2 /Phoenix have similar performance overall – Results should be examined critically
Скачать презентацию Automated systems an overview Robin A Howe Cardiff
6f275877a2dddf705a0243ef14bc3d80.ppt