2.2 Troubleshooting Flashcards
Result does not match clinical findings- approach
Generic template of answers
Caveat
- depends on the assay in question
- critical vs routine result/clinical impact/urgency
- Investigate
- post analytical; transcription, data entry error , correct patient
- analytical; QC on the day, calibration, machine issues, other samples in the run
- pre-analytical; correct patient, primary tube - If a mistake
- short term harm vs long term
- short term; redact report, amend, phone requester to explain
- - open disclosure
- CAR/incident form
- root cause analysis
- report at lab/management meeting - Solution
- education
- training
- algorithms
- action limits - Monitor situation
External QAP result falls outside of the allowable limits- approach
- Investigate
- post analytical: transcription error/data entry
- analytical run: QC, calibration,
- previous cycles; ongoing bias or random error
- pre analytical; correct specimen, transport/handling, reconstitution - If random
- repeat - If ongoing/systemic, drift
- equipment
- lot numbers of reagents
- method group - Clinical implication
External QAP result falls outside of the allowable limits- approach
- Investigate
- post analytical: transcription error/data entry
- analytical run: QC, calibration,
- previous cycles; ongoing bias or random error
- pre analytical; correct specimen, transport/handling, reconstitution - If random
- repeat - If ongoing/systemic, drift
- equipment
- lot numbers of reagents
- method group - Clinical implication
- QAP corrective action form/monitor
What are some of the pre-analytical errors?
patient factors- preparation, medications
collection factors- poor technique, contamination, incorrect tube, hemolysis, insuff volume, inadequate labelling, incorrect test
transit factors- failure to store, separation of sample, aliquot sufficient volume, freezing/thawing, temp
General approach to troubleshoorting
- post ana, ana, pre ana
- sourcing errors general principles
- specific method/platform considerations
- random vs systematic error
- then incident form/CAR/ raise in meeting and monitor**
What are some of the pre-analytical errors?
patient factors- preparation, medications
collection factors- poor technique, contamination, incorrect tube, hemolysis, insuff volume, inadequate labelling, incorrect test
transit factors- failure to store, separation of sample, aliquot sufficient volume, freezing/thawing, temp
automated aliquots- alignment, slash, contam
** liaise with specimen collection/clinical team
What are analytical errors?
Internal QC non conformity- outside allowable limits
Calibration
Platform- automated vs manual, failure to add, bubbles/blocked tubing,maintenance, calibration, mechanical fault (systematic),
Reagent-storage, expiry, faulty batch, contam (syst(
Operator-non compliance to SOP, bias, pipetting tech if manual
What are post analytical errors?
Transcription of results- comments, data entry verified/ second reader
Release of results
Interpretation- incorrect reading (training)
General format of dealing with a non-conformity
- identify
- investigate (what, who, when, clinical significance)
- immediate corrective action
- root cause analysis
- preventative strategy
- document
- discussion/management meeting
- monitor
- NATA will review
QC random error- what are the potential causes and approach?
Random error- sudden shift in QC
- QC material itself; wrong one, degraded/expired, evaporated
- reagent lot new
- calibration issue
- instrument issue
- operator issue on the day; pipette, incubation, temperature, variation reader
First step- check QC material, can re-run
If still out, can examine reagent/instrument/operator/other QCs (first party)
If other QCs affected on the instrument likely instrument issue and may require servicing
Or may need recalibration if QC despite re-run still out
Internal QC has failed with R4s- what is your approach?
Explain R4s
Reject run, need to ix before pursuing further tests
- need info is it kit/3rd party
- what about other levels of QC?
- other QCs on the platform ok?
Random error (reagent/instrument/operator); imprecision
Re-run
New lot
Previous known patient samples could be re-run
Results from different labs do not agree- what are the potential sources of variation for this result once lab error has been excluded?
Pre-analytical;
- different labs environmental- water, temp, centrifuge
- Biological variation of the sample to the detection kit
Method
- reagent variation; calibration/ curve
- batch to batch variation
- ag source ag/substrate, capture antibody
- avidity of ab, isotype detected
- nature of method and sensitivity/specificity
post analytical
- reference range
- cut offs
- reporting units
mention gold standard methods ie CIEP for ENA
Reagent recall question, results have already gone out, what would you do? apparoch
- immediate corrective action
- ID results affected/retraction/amendment
- stop testing current samples
- freeze samples coming in
- contingency plan
- communication to users - Clarify with company/manufacturer
- issue
- other results affected
- when new lot can be sent - Contingency
- get new lot and re run all/amend results
- may have to consider sendaway to another unaffected lab
4 . CAR form
- discuss at meeting
-close when able
Kit recall, company no longer making it in 3 months- what do you do? Apprach
New method • Investigate i. New method 1. Literature 2. EQAP 3. Colleagues 4. What platform is already available ii. 3 month worth of supply- how many tests is that, plan timeframe • Deciding on a new method i. Pre analytical- sample handling/storage ii. Analytical- ease of sourcing kit/ controls, method platform , analytical performance characteristics, safety, staff expertise/experience iii. Post analytical- LIS, interpretation iv. Costs/financial implications • Validation i. Cross validation 1. Of known pos/neg samples, EQAP samples, and also of samples from patients without disease 2. Depending on platform etc a. Precision b. Correlation c. Agreement
• If new method
i. Notify NATA/QAP
ii. New SOP
iii. Scope of practice
iv. Training
v. Communication
vi. Parallel testing for some time
What are some interferences in immunoassays?
Antibody related or non antibody related
Positive/neg interference
Antibody
- anti analyte ab
- anti animal ab (esp HAMA)
- mAbs
- autoantibody
- paraprotein
- cryoglobulin
- heterophile antibody
- cross reactivity
- RF
- hook/prozone effect
Non antibody
- medications/drugs
- contrast agent
- fibrinogen/CRP in EPG
- lipemia
- hyper bilirubinemia
- hemolysis
- other sample matrix
What are some of the methods to reduce interference from heterophile and HAAA?
- Removal of interfering antibody
- prior extraction
- immunoextraction on sepharose/protein A beads/suspension
- PEG
- heating 70-90deg - Addition of blocking agent from same species as antibody reagents
(monoclonal ab usually mouse ig)
- non immune serum, species sp polyclonal IgG
- species specific fragments of IgG
- heterophile blocking reagents HBR like scantibodies block all heterophiles in all isotypes
What you see an out of QC what are your steps?
- Decide whether to reject or accept first
- if Rejecting then do not allow run
- Stop until investigation complete - Investigate
- QC material; lot/correct/level, prep, storage, expired
- Reagent material; correct lot, prep, stored, expired
- calibration curve, standards prep, stored, loaded in order, lot # correct, expiry
- instrument; maintenance UTD, recent service, solution store/stability, problems on visual inspection - leak, plungers, bent probe
- operator
- environment; water system, waste , temp, humidity
For discrepant results, sometimes limitations of methodology play a role - why is this in immunological antibody testing?
uncertainty occurs in many aspects of process
antibodies
- difficult due to mol structure, weight, biological variability
- ag source; whole, cell extract, purified, recombinant
- antibody measuredl affinity, avidity, epitope specificity, isotype
- antibody detection system; polyclonal, monoclonal, affinity, conjugation, methodo variation re incubation time, choice substrate
standardisation issue- even when there is traceability, standards are limited
Discrepant results in IFA, why? potential sources
Lab error IFA
IFA limitation of assay
Pre analytical-
analytical
- substrate
– species, transfection, cross section /orientation, composite block
- fixative
- reagents expired
- serum dilution
- sampling error bubbles
- conjugation (fluorescence vs enzyme)
- quality of secondary antibody;(affinity avidity isotype)
- set up of UV microscopes (mercury vs LED light source, mag, numerical aperture)
Post analytical
- interpretation
- transcription
Need to check QCs, other wells ensure other runs ok
Ask if pipetting automated/slide mounting or manual
Can re-run sample
Try another lot of slides
Try a different slide
What about semi quantitative assays what causes uncertainty in these methods?
e.g. immunoblot, ELISA methods generating a ratio result
Generally
- methods
- reagents
- instrument
- calibrator/ref material
- control
- reporting units
- random error
- nature of analyte/behaviour
- substrate (recombinant/and or purified ag on nitrocellulose membrane)
- single point calibration usually
What about quantitative assays what are their sources of uncertainty and limitations?
- source of ag; if ag variation is reduced (CCP under single source) then reduced
- solid vs liquid phase, conc of ag, method of coating
- siotype being measured
- ## for sIgE prep of allergen
What are some of the things the labs can implement to overcome limitations of the assay and to ensure assay is optimised?
- validation including clinical sens/spec
- clinical input for interpretation
- use of internal QC material
- EQAP enrollment
- periodic re evaluation of assay
- expertise IFA reading, audit of reading
- regular maintenance
- validate manufacturer reference ranges
Name one instance where checkerboard titration may be useful/used
IFA
if antiserum too concentrated- may risk prozone effect, waste money, inc BG stain
too dilute, loss of sensitivity
use a known positive serum of defined titre to create a checkerboard with conjg antiserum.
appropriate dilution is the last but one before the fall off in sensitviity to allow long term storage without risking a drop in sensitivity
What kind of machine/instrument issues can cause problems?
Analytical> calibration> technical failure - sampling - mechanics - water/waste - reader issue/light source/cuvette
