Transcription

2013 Laboratory Accreditation Program Audioconferences and WebinarsAnalytical Measurement Range: Examples and ApproachesWilliam Castellani, MD, FCAPMarch 20, 2013www.cap.org

Objectives Define and describe the analyticalymeasurement rangeg Compare this range to CLIA’88 requirements forcalibration and calibration verification Discuss strategies for satisfying accreditationrequirementsq 2013 College of American Pathologists. All rights reserved.2

Conflicts Nothingg to disclose 2013 College of American Pathologists. All rights reserved.3

It All Begins Here “Calibration” means a pprocess of testingg and adjustingjgan instrument or test system to establish a correlationbetween the measurement response and thegconcentration or amount of the substance that is beingmeasured by the test procedure. “Calibration verification” means the assaying of materialsoff knownkconcentrationt ti ini theth same manner as patientti tspecimens to substantiate the instrument or testsystem’s calibration throughout the reportable range forpatientti t testt t results.lto Centers for Medicare and Medicaid Services, State Operations Manual,Appendix C 2013 College of American Pathologists. All rights reserved.4

ReportablepRangeg Reportableprangeg means the spanp of test result valuesover which the laboratory can establish or verify theaccuracy of the instrument or test system measurementresponse.responseo CLIA ’88, Sec. 493.2, Definitions 2013 College of American Pathologists. All rights reserved.5

Reportable Range Two components:po The primary range of measurement Analytical measurement range “Linear” rangeo Anything done to the system to expand this range DilutionDil ti Concentration– The span of values that can be reliablymeasured using these modifications is thereportable range in CLIA’88 2013 College of American Pathologists. All rights reserved.6

What Must We Do? Validate or verifyyo Reportable range: as part of methodvalidation/verification Initial documentation of the analyticalmeasurement range Documentation of any dilution or concentrationprotocol that can be done to expand the range ofvalues that can be reported by the methodo Analytical measurement range [verification]: every sixmonths thereafter (when necessary) 2013 College of American Pathologists. All rights reserved.7

How Do We Do It? Set criteria of acceptancep Established protocol MMedicaldi l relevancelo All this should be established by the laboratorydirectoro All this should be documented formallyo The actual review mayy be delegated,g, thoughg finalauthorization may be reserved for the director 2013 College of American Pathologists. All rights reserved.8

Reportable Range Mayy decrease the lower limit of the analyticalymeasurement range by:o Concentrating the sample Amicon concentrator Extraction and resuspensiono Increasing the ratio of sample to reagent Altering the programming of the instrument 2013 College of American Pathologists. All rights reserved.9

Reportable Range More commonly,y, mayy increase the upperpp limit of theanalytical measurement range by:o Decreasing the ratio of sample to reagento Diluting the sample before analysis 2013 College of American Pathologists. All rights reserved.10

Reportable Range Most often, the manufacturer pprovides the information ormechanism for this modificationo Autodilution/autoconcentrationo Dilution protocolo Concentration protocol GGoodd laboratoryl b tpracticeti wouldld iincludel d verifyingif i ththattthese modifications work Going beyond the modifications stated in themanufacturer’s instructions for use requires laboratoryvalidation 2013 College of American Pathologists. All rights reserved.11

AMR & RR ExampleshCGASTAMR: 3-1000 mIU/mlAMR: 4-900 U/LRR lower limit: 5RR lower: 4RR upperpp limit:not definedDilute and reassay untilvaluel iis obtainedbt i dRR upperpp limit:1,000,000Values above upper RRreportedt d as greatert thanth1,000,000 2013 College of American Pathologists. All rights reserved.12

General Principles Establish a target valueo May use a patient sample’ssample s result as the “target”targeto May use peer group mean of PT materialo Mayy be established byy the pprovider of the material Establish an acceptable range around the targeto Mayy be an absolute ((arbitrary)y) rangeg [[10%]]o May use precision data for control material near thetargeto May be provided by the manufacturer Document your protocol (approved by director) 2013 College of American Pathologists. All rights reserved.13

Verifying the Reportable Range Documentingg that the modifications to the samplep ormethod produce reliable results Begin with verifying the dilution protocolo Use the method manufacturer’s recommended diluentand, if given, dilution factoro Determine an acceptance range for the dilution result If the manufacturer does not allow for dilution, then theprotocollaboratoryy must validate its own p 2013 College of American Pathologists. All rights reserved.14

Dilution Protocol Take samples near theupper limit of theanalytical range Manually dilute andassay Compare against thetarget using theacceptanceprangeg 2013 College of American Pathologists. All rights reserved.15

Dilution Protocol Dilution Limito If the manufacturer gives either a maximum dilution ora specific dilution, verify to the limit Report all results exceeding this diluted range as upper limito If the manufacturer gives no guidelines: Laboratory director may specify a maximum dilutedvalue or dilution protocol May dilute until you get a value (no upper limit) 2013 College of American Pathologists. All rights reserved.16

No Upper Limit Take samples that exceed theupper limit of the analytical rangerange.Manually dilute and assay 10(should get a valid answer at thehigh end).Manually re-dilute 10.Compare against the target usingthe acceptance range.You have verified a 100 dilution;at this point, the sample matrix isdiluent, not serum/plasma. Anyfurther dilution would notsignificantly change the matrix.Document this claim. 2013 College of American Pathologists. All rights reserved.17

Verify Autodilution Method Take ppatient samplespthat pprovide results on autodilution Manually dilute the sampleo Do not need to use same dilution factor as theautodiluter Compare the autodilution and manual dilution resultsusing the manual dilution result as target value 2013 College of American Pathologists. All rights reserved.18

Verify Autoconcentration MethodMethod, Part 1 For those methods where the lower limit is extended byyincreasing the ratio of sample to reagent Take low samplespand seriallyy diluteo Suggest x 2 dilution Using the neat result of the sample as the targettarget, look fordeviations from linearity 2013 College of American Pathologists. All rights reserved.19

Example: Serum Immunoglobulin A Manufacturer’s range: 40 – 800 mg/dLDilutionValue 1Value 2MeanTheoreticalNeat171171171171 287888886 443444443 820202021 1688811 323335 2013 College of American Pathologists. All rights reserved.20

Verify Autoconcentration MethodMethod, Part 2 Once yyou’ve established the maximum concentrationthat will give reliable results, test samples using themodified concentration protocol programmed into theanalyzer May need to first dilute sample into lower part of theanalytical measurement range manually before placingon analyzer 2013 College of American Pathologists. All rights reserved.21

Example: Serum Immunoglobulin ANeat resultManualdilutionResult (standardprotocol)Result(autoconcentration)171 8202351 22624110None11011288None8889 2013 College of American Pathologists. All rights reserved.22

CLIA Requirement for Calibration/Calibration Verification 493.1255: Calibration and calibration verificationproceduresdare requiredi d tto substantiateb t ti t ththecontinued accuracy of the test system throughoutthe laboratory’sy reportableprangeg 2013 College of American Pathologists. All rights reserved.23

Calibration/Calibration Verification Calibration: Establishes the relationship betweenanalytel t contentt t andd instrumenti tt measurementt signalil Calibration verification: Confirms that the currentcalibrationlib ti settingsttiremaini validlid 2013 College of American Pathologists. All rights reserved.24

CAP Interpretation of CLIA Calibration Verification Validate the set ppoint ((CAP Calibration Verification)) Prove response over the entire analytical measurementrangeg The Laboratory Accreditation Program considerscalibration verification to be secondaryy to calibrationo If calibration satisfies the CLIA requirements forcalibration verification, no further action is necessary 2013 College of American Pathologists. All rights reserved.25

Calibration Verification Requirements Sec. 493.1255(b)(2)( )( ) [Perform[and document calibrationverification procedures] Using the criteria verified orestablished by the laboratory o (i) IIncludingl di ththe number,b ttype, andd concentrationt ti offthe materials, as well as acceptable limits forcalibration verification; ando (ii) Including at least a minimal (or zero) value, a midpoint value, and a maximum value near the upperlimit of the range to verify the laboratory's reportablerange of test results for the test system 2013 College of American Pathologists. All rights reserved.26

Frequency of Calibration OR Calibration Verification At least every 6 months When recommended by manufacturer After major maintenance or service When QC indicates At complete reagent changechange, unless laboratoryhas data showing lot uniformity At change of critical reagents (laboratorydefines) 2013 College of American Pathologists. All rights reserved.27

When is Calibration Sufficient?QuantitativeAssay?NoYesCalibrateat leastevery 6months?NoSpecial case: cutoffPerform calibrationverification at leastevery 6 monthsYes 2013 College of American Pathologists. All rights reserved.28

Calibrate at Least Every 6 MonthsMonths YesUse atleast 3llevels?l ?NoYesDoneMaterialsspan AMR(low-midhigh)?NoVerify AMRusingmaterialsnear low,mid-pointid i tand highvalues atleast every6 months 2013 College of American Pathologists. All rights reserved.29

ExamplesLaboratory calibrates:With 1 point dailyVerifycalibration?NoVerify AMR?YesWith each new lot change (8-10 months)using 2 calibrators that span the AMRYesYesEvery 3-4 months with 5 calibrators thatspan the AMRNoNoMonthly with 2 calibrators that span theAMRNoYesEvery 3 months with 3 points (50, 150,300) but reports 25-60025 600NoYesNoTwo concentrationsaround the cut-offEvery run using kit material to determinea positive/negative cut-off 2013 College of American Pathologists. All rights reserved.30

Requirements for [CAP] Calibration Verification Target values Appropriate matrixo Calibratorso Materials provided by the vendor for the purpose of calibrationverificationo Previously tested unaltered patient/client specimenso Primary or secondary standards or reference materials with matrixcharacteristics and target values appropriate for the methodo Third party reference material if commutableo PProficiencyfi ittestingti materialt i l withith matrixt i characteristicsht i ti andd ttargettvalues appropriate for the method Controls usually do not satisfy this requirement unlessspecifically designed by the manufacturer for this purpose 2013 College of American Pathologists. All rights reserved.31

Requirements for Verification of the AnalyticalMeasurement Range Target values Appropriate matrixo Linearity material of appropriate matrixo Proficiency testing survey materialo Previously tested patient specimens, unalteredo Previously tested patient specimens, altered by admixture with otherspecimens,idildilution,tispiking,ikior otherth ttechniqueh io Primary or secondary standards or reference materials with matrixcharacteristics and target values appropriate for the methodo Calibrators used to calibrate the analytic measurement systemo Control materials, if they adequately span the AMR 2013 College of American Pathologists. All rights reserved.32

Verification of the Analytical Measurement Range “Guidelines for analytey levels near the low and highgrange of the AMR should be determined by thelaboratory director. Factors to consider are the expectedanalytic imprecision near the limitslimits, the clinical impact oferrors near the limits, and the availability of testspecimens near the limits. It may be difficult to obtainspecimens with values near the limits forf some analytes(eg, T-uptake, free thyroxine, free phenytoin, prolactin,FSH,, troponin,p, pOp 2)). In such cases,, reasonableprocedures should be adopted based on availablespecimen materials.”o Chemistry and Toxicology checklist,checklist 9/25/12 2013 College of American Pathologists. All rights reserved.33

How Do You Determine the Upperand Lower Limit Targets? Determined by available material:o Define the linear range as going from the low to thehigh target sample Fixed range:o Within 10% of the top end and 1% of the bottom end Clinical use and decision points 2013 College of American Pathologists. All rights reserved.34

Upper Limit Target Based on Clinical Decision Point Decision point inthe lower quarterof the linearrange:o Adult totalbilirubinUse materialwithin 80% of theupper limit of thelinear range 2013 College of American Pathologists. All rights reserved.35

Upper Limit Target Based on Clinical Decision Point Decision point inthe mid-half of thelinear range:o Most drugsUse materialwithin 90% of theupper limit of thelinear range 2013 College of American Pathologists. All rights reserved.36

Upper Limit Target Based on Clinical Decision Point Decision point inthe upper quarterof the linearrange:o Neonatalbilirubino Special case:β hCGβ-hCGUse materialwithin 95% of theupper limit of thelinear range 2013 College of American Pathologists. All rights reserved.37

Ongoing Verification of the Upper Limit of theAnalytical Range When a samplep exceedingg the current verified upperpp limit(but within the manufacturer’s range) is assayed,perform two dilutions into the currently verified range(eg 2 and 3)(eg, If the two dilutions come within defined targets, the upperlimit has been extended to the value of this new sample 2013 College of American Pathologists. All rights reserved.38

Ongoing Verification of the Upper Limit of theAnalytical Range Manufacturer’srange: Currently verifiedrange:ModifiedffromfMartin Kroll, Boston Medical CenterC 2013 College of American Pathologists. All rights reserved.39

Ongoing Verification of the Upper Limit of theAnalytical Range Manufacturer’srange: Newly verified range:ModifiedffromfMartin Kroll, Boston Medical CenterC 2013 College of American Pathologists. All rights reserved.40

Extending the Verification Range Available material withtarget values may notreach the upper limit ofthe analyticalmeasurement range Example: urine sodiumo manufacturer’s range: 0 – 200 mmol/L 2013 College of American Pathologists. All rights reserved.41

Extending the Verification Range Prepare a stock sodiumsolution of 200 mmol/L Do two serial x 2 dilutions(100 and 50 mmol/Ltarget) Assay each level and plot 2013 College of American Pathologists. All rights reserved.42

Verifying the Lower Limit of the Analytical Range Negativegmaterialo Patient sample without the analyte in questiono Surrogate material Diluent Water or aqueous IV fluid– Matrix of questionable value for most methods Diluted low patient samples 2013 College of American Pathologists. All rights reserved.43

Lower Limit Target Based on Clinical Decision Point Relativelystraightforwardif there is alower clinicaldecision pointUse material 50% of the lowerclinical decisionpoint 2013 College of American Pathologists. All rights reserved.44

What about the MidMid-Point?Point? The regulationgonlyy states that one mid-pointpvalue isrequired The checklist does not further define how this valueshould be obtained 2013 College of American Pathologists. All rights reserved.45

Mid-PointMidPoint Material The most common approachppis 1:1 mixtures of the lowand high material This ratio of 1:1 is strictlyy “convenient;”; for those testswhere a critical decision point exists, using a ratio thatbrings the target of the mid-point material close to thisdecision point should be considered The difficulty is in creating an exception to standardlaboratory protocol 2013 College of American Pathologists. All rights reserved.46

Mid Point MaterialMid-Point Most commercial products provide four or more targetsamples, though this is not required by regulation A similar distribution of target values can be obtained bysequential mixing2.5Low101.0High1.0202.02.5Mid1.01.00.5LowMid LoMid-Lo 2013 College of American Pathologists. All rights reserved.0.51.00.5Mid0.5Mid HiMid-HiHigh47

Summary Establish protocols addressing the requirementso material that will be usedo how the targetg values will be obtainedo how the acceptance ranges will be determined Document the protocolso Reportable range: Method validation procedureo Calibration verification (CLIA): Qualitycontrol/assurance procedure Approved by director 2013 College of American Pathologists. All rights reserved.48

Thank you!Questions? 2013 College of American Pathologists. All rights reserved.49