Instrument Validation

1
Instrument Validation
2
Pre-Installation Requirements

• Proper electrical wiring to meet manufacturers
  recommendation
• Constant line voltage
• Voltage regulator/stabilizer, if required
• Temperature controlled area to meet
  manufacturers specifications. Large
  fluctuations in temperature may adversely affect
  results.
• Adequate de-ionized water supply
• Proper waste disposal capability
• Adequate air circulation/ flow around instrument
• Dust free environment
• Stable floor / bench top

3
Out of the BoxFor the service Engineer

• Check instrument for visual damage
• Check for any loose parts or connections
• Make sure all computer boards are properly seated
• Check the wall socket to verify proper voltage
  outlet
• Plug instrument power cord into (voltage
  stabilizer) electrical supply
• Confirm the correct voltage on instrument
• Main power supply
• Photometric voltage
• Any other voltage supply that is pertinent to
  instrument functions

4
Out of the BoxFor the service Engineer

• Permit instrument to stabilize/equilibrate
• Let all components reach proper temperature
• Set in any parameters that may be required
• Ranges
• Temps

5
Begin Test Verification

• Follow manufacturers recommendation for reagent
  and control preparation
• Make a habit of reading package inserts
• Place reagents on instrument it required
• Place calibrators on instrument in proper order
  sequence

6
Begin Test Verification

• Run calibration
• If calibration is unsuccessful, then troubleshoot
• 1st determine of instrument or reagent problem
• Verify proper prep of calibrator
• If calibration is successful, then run controls
• Be sure to print out the calibration results and
  keep the printout for future reference
• If controls within stated value for analyzer and
  methodology, then perform linearity checks
• Review linear regression curve and enzyme curve.

7
(No Transcript)
8
Linearity Materials and Methods

• Purpose Verify the analytical measurement range
  (the lowest and highest result that can be
  accurately reported from the instrument.)
• Commercial materials with high values for
  analytes being performed
• Must not contain high levels of interfering
  substances such as bilirubin, etc.
• Separate materials for hematology and chemistry

9
Linearity Materials Methods

• Chemistry- bilirubin may be a separate material
  from material for other analytes
• Materials may be one vial with high level for
  several analytes and requires dilution or may be
  a set of samples ranging form low to high
• Make sure proper base matrix is used
• After results are obtained for each linearity
  sample, plot them on linear regression chart
• Check package inserts for expected values
• Review linear regression curve
• Troubleshoot if not at or near the manufacturers
  stated linearity for methodology

10
Linearity Study
Non-enzymatic (non-rate) Curve
Enzymatic (rate) Curve
11
Precision check

• Purpose check reproducibility of instrument
• Precision quick method
• Run 5 replicates of one patients sample free of
  elevated bilirubin, lipemia, and hemoloysis
• Calculate CV or SD
• Refer to the methodology in procedure manual for
  expected CV or SD
• Outside of limits, then troubleshoot
• Sampling system
• Reagent delivery
• Photometer / lamp

12
Carryover Study

• Place a sample with known high analyte value on
  the instrument
• Follow high-value sample with known low-value (as
  low as possible) sample for same analyte
• Check results If low-value sample is higher than
  expected, there is a problem with carryover.

13
Accuracy / Recovery Studies

• Run Controls
• Care must be taken that the materials used are of
  the same base matrix as that of the method
  calibrator
• If expected values are not obtained for the
  controls, return to the calibration procedure.

14
Correlation Studies Comparison of old
methodologies vs. automated methods

• Samples to Use
• Commercially prepared
• Need spread of ranges, e.g., low, high, normal
• Patient samples
• Must be free of interfering substances
• Need spread of ranges, e.g., from low to high
• Until all patient samples are collected
• Freeze at -20 degrees
• Do not use home-type freezer because of freeze
  thaw cycles will compromise sample

15
Correlation Studies

• Run a minimum of 20 different patient samples
• Run same samples by each method consecutively
  (same day)
• Plot results by linear regression
• X axis automated results
• Y axis Old method results
• Review linear regression curve
• How do the two methods compare?
• Is the slope and intercept acceptable?
• If the old method will be used for back-up, it is
  important to know normal range for each and how
  well the two methods agree or disagree
• How do you handle the procedures that do not
  correlate?

16
Correlation Studies
17
Correlation Studies