QA of a Hematology Analyzer

May 30th 2014
M R Tiwari
B.Sc. (Microbiology), PGDMLT, M.Sc.(TQM)
Scientific Assistant ‘D’
Hematology, Composite Laboratory, ACTREC, TMC
Laboratory test results
(Good laboratory practice, GLP)
Clinical diagnosis
Patient management
Quality can be assured at
◦ Pre-analytical stage
◦ Analytical stage
◦ Post-analytical stage
Requisition form
Sample collection & transport
Sample receiving & rejection
Sample preparation
=


Definition
Reproducibility of a result
Achieved with
◦ Controls samples
 (Commercially procured or whole blood)

When required
◦ Calibration
◦ Daily quality control
◦ After troubleshooting
◦ After change of reagent



Definition
Closeness of a result to the true value
Achieved with
◦ Calibrator
When required
◦ Calibration
 Mean
– The target
 Standard
deviation – The acceptable limits (±2SD)
 Coefficient
of variation (CV %)
 variation at different levels can be compared.

X-axis - the days of the month (time interval)

Y-axis. - control observations
Control Chart's Inventor
In 1931,
Dr. Walter Shewhart, a
In 1950,
scientist at the Bell Telephone
S. Levey &
Laboratories, proposed
applying statistical based
E.R. Jennings
control charts to interpret
suggested the use in
industrial manufacturing
the clinical laboratory.
processes.
L-J chart interpretation
Westgard rules
Selection
Purchase
Installation
Use
Replace
Selection
Comparative studies
Cost per test
Through put
Reagent consumption
Sample volume
Good logistics
etc….
Purchase
Reagent rental / direct purchase
Loan availability
etc….
Installation
Installation Qualification (IQ)
Parts, accessories, manuals, computer, software, filter, labour, consumables shipped
& received, vendor’s verification against all materials received against the initial order,
hardware installation, environmental requirements, space, electricity, temperature.
Operational Qualification (OQ)
System power-up & shutdown, fluidic start-up & shutdown, no detectable leaks, absence of bubble
in the sheath fluid, sheath and waste level sensors are operational, pneumatic pressures are in range,
sample & sheath fluid rates meet their specifications, computer & its operating system starts up with no
errors, flowcytometer’s control software, photodiode & photo multiplier tubes (PMT) voltage controls &
detector responses should be tested, laser power of each laser line or wavelength to be tested, path
between the laser and flow cell is checked, manual and automated sample delivery, testing of barcode
reader, etc.
Performance Qualification (PQ)
Linearity, Carryover, Precision, Calibration, Method validation,.
◦
◦
◦
◦
◦
WBC (0.02 to 400)
RBC (0.0 to 7.0)
HGB (0.0 to 22.5)
PLT (0.2 to 3500)
RETIC% (0.2 to 24.5)
◦ %Carryover = [(L1-L3)/H]x100
◦ L1– the count of first aspiration of cell-free plasma
◦ L3- the count of third aspiration of cell-free
plasma
◦ H – the count of high pool

Carryover is defined
as a number of cells
remaining behind
following the cycling
of a blood sample.

This test is performed
to determine if one
sample interferes with
the accurate analysis
of the next sample.

Ideally, carryover shall
be very low.
Calibrator
Certified Reference Material (CRM) used to calibrate a
measurement on an analyzer.
Cal-Factors
If any deviation from calibration references is observed
necessary calibration correction factors are applied to
set the accuracy of the instrument.



Pre-calibration check
◦ Total maintenance of the instrument
◦ Reagents (replenish or replace)
◦ Calibrator (Check for expiry)
◦ Controls (Check for expiry)
Calibration procedure
◦ Carryover check
◦ Precision check
◦ Calibration in both
 (open and closed modes)
Post-calibration validation
◦ Run calibrator as samples
◦ Run 3 level controls
Use
Internal Quality control (IQC)
External Quality Assurance (EQA)
Sample analysis
Hematology analyzers
• Start up status and background count check
First step
(Monitor for acceptable background…..if required take corrective action)
• QC monitoring using multilevel controls
(Low level, normal level and high level)
Done once
(Monitor L-J chart, apply Westergard’s interpretation, monitor %CV…..if
required take corrective action)
• QC monitoring using retained sample
Periodic in b/w
check till shut
down
(Monitor %CV…..if required take corrective action)
• Start up status and background count check
First step
(Monitor for acceptable background…..if required take corrective action)
Done once
• QC monitoring using multilevel controls
(Low level, normal level and high level)
(Monitor L-J chart, apply Westergard’s interpretation, monitor %CV…..if
required take corrective action)
Warning rule = use other rules to inspect the control points
Rejection rule = “out of control”
◦ Do not run patient samples
◦ Identify and correct problem
◦ Do not report patient results until problem is solved and
controls indicate proper performance
• QC monitoring using retained sample
Periodic in b/w
check till shut
down
(Monitor %CV…..if required take corrective action)
Comparative study b/w the analyzers
(monitor % variation)
Rerun on same sample in different mode
Lot validation (after change of reagent)
with retained sample – precision check
Correlation with peripheral smears

All haematology analyzers
◦ Moving average analysis (Bull’s algorithm)

High end analyzers
◦
◦
◦
◦
◦
WBCP – WBCB
HGB (spectrophotometricaly)-HGB(cellular-laser)
MCHC – CHCM
PMN – (Neutrophils, Eosinophils)
MN – (Lymphocytes, Monocytes)


Follow the manufacturer’s manual
Perform maintenance regularly
◦
◦
◦
◦
◦
◦
Daily
Weekly
Biweekly
Monthly
Annually
As if required
Maintain record of maintenance

Try to correct at operator level if possible or
Call the service engineer for help

Maintain a record of any breakdown observed

◦
◦
◦
◦
◦
◦
Nature
Time
Error detected
Corrective action
QC checked
Service report
Normal Distribution Curve
or
Gaussian curve
Describes events or data that occur
symmetrically about the mean.
Out of 100 events
68.7 will fall within ±1 SD
95.4 will fall within ± 2 SD
99.7 will fall within ±3 SD
Z-Score
A-1
A-2
A-3
A-4
A-5
A-6
A-7
A-8
A-9
A-10
A-11
A-12
WBC
5.9
6.3
6.2
5.2
6.3
6.2
5.9
6.3
HGB
13.1
12.9
12.7
13.1
12.9
12.7
13.1
PLT
262
246
255
262
246
255
Z-SCORE
A-1
A-2
A-3
A-4
A-5
A-6
MEAN
SD
6.2
5.9
6.3
6.2
6.1
0.3
12.9
12.7
13.1
12.9
12.7
12.9
0.2
262
246
300
262
246
255
258.1
14.9
A-7
A-8
A-9
A-10
A-11
A-12
WBC
-0.55
0.70
0.39
-2.74
0.70
0.39
-0.55
0.70
0.39
-0.55
0.70
0.39
HGB
1.17
0.00
-1.17
1.17
0.00
-1.17
1.17
0.00
-1.17
1.17
0.00
-1.17
PLT
0.26
-0.81
-0.21
0.26
-0.81
-0.21
0.26
-0.81
2.82
0.26
-0.81
Z SCORE SCALING :< ±0.5 - Excellent performance
±0.5 to ±1.0 - Satisfactory
±1 to ±2 - Acceptable
> ±2 - Defect requiring attention
-0.21
Check for the IQC results during the period
when the EQAS sample was analyzed.
Follow the instructions from the nodal
organization.


A formal way of testing for aberrant results is known as `delta check`.
The blood count parameters should not differ from recent tests in the
previous 2-3 weeks by more than a certain amount.
For Hb and RBC
For WBC
For Platelet count
10 %
20-25 %
50 %
Assuming that the patient’s clinical condition has not altered significantly
The laboratory
The requester i.e.
takes
responsibility for
reporting the
results within the
specified turn
Coag4 hrs
CBC2hrs
the clinician is
notified in case of
delay in
BMA3wd
around time.
examination only
in such cases
where the delay
can compromise
TAT
patients care.
Quality Indicator – “TAT shall be monitored on periodic basis for continual improvement”
Laboratory test results
◦ Pre-analytical stage
Clinical diagnosis
◦ Analytical stage
◦ Post-analytical stage
Quality
Patient
management
Good Laboratory Practice