7 Levey–Jennings Chart and Westgard Rules
Tải bản đầy đủ - 0trang
4.7 LeveyÀJennings Chart and Westgard Rules
■
are commonly used for calibration. One calibrator must be a zero
calibrator and the highest calibrator must contain a concentration of the
analyte at the upper end of the analytical measurement range. Other
calibrators usually have concentrations in between the zero calibrator and
the highest calibrator, and represent normal values of the analyte as well
as values expected in a disease state (for drugs, values below therapeutic
range, between therapeutic ranges, and then toxic range).
Controls are materials that contain a known amount of the analyte. The
matrix of the control must be similar to the matrix of the patient’s
sample; for example, matrix of the control must resemble serum for
assays conducted in serum or plasma.
A LeveyÀJennings chart is commonly used for recording observed values of
controls during daily operation of a clinical laboratory. A LeveyÀJennings
chart is a graphical representation of all control values for an assay during an
extended period of laboratory operation. In this graphical representation,
values are plotted with respect to the calculated mean and standard deviation, and if all controls are within the mean and 6 2 SD, then all control
values are within acceptable limits and all runs during that period will have
acceptable performance (Figure 4.3). In this figure, all glucose low controls
were within acceptable limits for the entire month. The LeveyÀJennings chart
must be constructed for each control (low and high control or low, medium,
and high control) for each assay the laboratory offers. For example, if the laboratory runs two controls (low and high) for each test and offers 100 tests,
then there will be 100 3 2, or 200 LeveyÀJennings charts each month.
Usually a LeveyÀJennings chart is constructed for one control for one month.
The laboratory director or designee must review all LeveyÀJennings charts
each month and sign them for compliance with an accrediting agency.
93.6
+ 3 SD
90.4
+ 2 SD
87.2
+ 1 SD
84
Mean
80.8
– 1 SD
77.6
– 2 SD
– 3 SD
74.4
0
5
FIGURE 4.3
LeveyÀJennings chart with no violation.
10
15
Days
20
25
30
55
56
CHAPTER 4:
Laboratory Statistics and Quality Control
Table 4.1 Westgard Rules
Violation
Comments
Accept/Reject
Run
Error
Type
12s
One control value is outside 6 2 SD limit, but other control within
6 2 SD limit
One control exceeds 6 3 SD
Both controls outside 6 2 SD limit, or two consecutive controls
outside limit
One control 12 SD and another 22 SD
Four consecutive control exceeding 11 SD or 21 SD
Ten consecutive control values falling on one side of the mean
Accept run
Random
Reject run
Reject run
Random
Systematic
Reject run
Reject run*
Reject run*
Random
Systematic
Systematic
13s
22s
R4s
41S
103
*Although these are rejection rules, a laboratory may consider these violations as warnings and may accept the runs and take steps
to correct such systematic errors.
However, if technologists review results of the control during a run and
accept the run if the value of the control is within an acceptable range established by the laboratory (usually a mean of 6 2 SD), then the laboratory
supervisor can review all control data on a daily basis; usually the supervisor
reviews all control data weekly.
Usually Westgard rules are used for interpreting a LeveyÀJennings chart, and
for certain violations a run must be rejected and the problem resolved prior
to resuming testing of a patient’s samples. Various errors can occur in
LeveyÀJennings charts, including shift, trend, and other violations
(Table 4.1). The basic principle is that control values must fall within
6 2 SD of the mean, but there are some situations when violation of
Westgard rules occurs despite control values that are within the 6 2 SD limits of the mean. Usually 12s is a warning rule and occurs due to random error
(Figure 4.4), and other rules are rejection rules. In addition, shift (Figure 4.5)
and trend (Figure 4.6) may be observed in LeveyÀJennings charts, indicating
systematic errors where corrective actions must be taken. When 10 or more
consecutive control values are falling on one side of the mean, a shift is
observed (103 rule). In addition, when a 10 3 violation is observed, it may
also indicate a trend when control values indicate an upward trend.
4.8 DELTA CHECKS
Delta checks are an additional quality control measure adopted by the computer of an automated analyzer or the laboratory information system (LIS)
where a value is flagged if the value deviates more than a predetermined
limit from the previous value in the same patient. The limit of deviation for
4.8 Delta Checks
96.8
12S violation
Glucose control,mg/dL
93.6
+ 3 SD
22S violation
90.4
+ 2 SD
87.2
84
+ 1 SD
Mean
80.8
– 1 SD
77.6
– 2 SD
41S violation
74.4
– 3 SD
13S violation
71.2
0
5
10
20
15
Days
25
30
FIGURE 4.4
LeveyÀJennings chart showing certain violations.
93.6
+ 3 SD
90.4
+ 2 SD
+ 1 SD
87.2
Shift
84
Mean
80.8
– 1 SD
77.6
– 2 SD
– 3 SD
74.4
0
5
10
15
20
25
30
Days
FIGURE 4.5
LeveyÀJennings chart showing shift of control values.
+3 SD
Glucose control, mg/dL
93.6
+2 SD
90.4
Trend
87.2
+1 SD
84
Mean
80.8
–1 SD
77.6
–2 SD
–3 SD
74.4
0
5
FIGURE 4.6
LeveyÀJennings chart showing trend.
10
15
Days
20
25
30
57
58
CHAPTER 4:
Laboratory Statistics and Quality Control
each analyte is set by laboratory professionals. The basis of the delta check is
that the value of an analyte in a patient should not deviate significantly from
the previous value unless certain intervention is done; for example, a high
glucose value may decrease significantly following administration of insulin.
If a value is flagged as a failed delta check, then a further investigation
should be made. A phone call to the nurse may address issues such as erroneous results due to collection of a specimen from an IV line or collection of
the wrong specimen. Quality control of the assay must also be addressed to
ensure that the erroneous result is not due to instrument malfunction.
The value of a delta check is usually based on one of the following criteria:
■
■
■
■
Delta difference: current valueÀprevious value should be within a
predetermined limit.
Delta percent change: delta difference/current value.
Rate difference: delta difference/delta interval 3 100.
Rate percent change: delta percentage change/delta interval.
4.9 METHOD VALIDATION/EVALUATION
OF A NEW METHOD
Since April 2003, clinical laboratories must perform method validation for
each new test implemented in the laboratory even though such tests have
FDA approval. The following are steps for method validation as well as
implementation of a new method in the clinical laboratory:
■
■
■
■
Within-run assay precision must be validated by running low, medium,
and high controls, or low and high controls 20 times each in a single
run. Then mean, standard deviation, and CV must be calculated
individually for low, medium, and high control.
Between-run assay precision must be established by running low,
medium, and high control, or low and high control once daily for
20 consecutive days. Then mean, standard deviation, and CV must be
calculated.
Although assay linearity is provided by the manufacturer, it must be
validated in the clinical laboratory prior to running patient specimens.
Linearity is essentially the calibration range of the assay (also called
“analytical measurement range”). In order to validate the linearity, a
high-end calibrator or standard can be selected and then diluted to
produce at least four to five dilutions that cover the entire analytical
measurement range. Then, if the observed value matches the expected
value, the assay can be considered linear over the stated range.
The detection limit should be traditionally determined by running a zero
calibrator or blank specimen 20 times and then determining the mean
4.10 How to Interpret the Regression Equation?
■
and standard deviation. The detection limit (also called the lower limit of
detection) is the mean 12 SD value. However, the guidelines of the
Clinical Laboratory Standard Institute (CLSI, E17 protocol) advise that a
specimen with no analyte (blank specimen) should be run; then the
Limit of Blank (LoB) 5 Mean 1 1.654 SD. This should be established by
running blank specimens 60 times, but if a company already established
a guideline, then 20 runs are enough. Limit of Quantification is usually
defined as a concentration where CV is 20% or less [4].
Comparison of a new method with an existing method is a very
important step in method validation. For this purpose, at least 100
patient specimens must be run in the laboratory at the same time with
both the existing method and the new method. It is advisable to batch
patient samples and then run these specimens by both methods on the
same day, and, if possible, at the same time (by splitting specimens).
Then results obtained by the existing method should be plotted in the
x-axis (reference method) and corresponding values obtained by the new
method should be plotted in the y-axis. Linear regression is the simplest
way of comparing results obtained by the existing method in the
laboratory and the new method. The linear regression equation is the line
of best fit with all data points. A computer can produce the linear
regression line as well as an equation called a linear regression equation,
which is the equation representing a straight line (regression line),
Equation 4.9:
y 5 mx 1 b
■
ð4:9Þ
Here, “m” is called the slope of the line and “b” is the intercept. The
computer calculates the equation of the regression line using a least
squares approach. The software also calculates “r,” the correlation
coefficient, using a complicated formula.
4.10 HOW TO INTERPRET THE REGRESSION
EQUATION?
The regression equation (y 5 mx 1 b) provides a lot of important information regarding how the new method (y) compares with the reference method
(x). Interpretations of a linear regression equation include:
■
■
Ideal value: m 5 1, b 5 0, and y 5 x. In reality this never happens.
If the value of m is less than 1.0, then the method shows negative bias
compared to the reference method. Bias can be calculated as 1 2 m; for
example, if the value of “m” is 0.95, then the negative bias is
1 2 0.95 5 0.05, or 0.05 3 100 5 5%.
59