JPH02304354A - Analysis of blood data - Google Patents

Abstract

PURPOSE:To exactly and efficiently inspect blood by inputting the data from a blood cell counter to an analyzing instrument and making decision as to whether the count value of white cells is affected by the hemolysis defect of red cells or not by specific decision conditions. CONSTITUTION:The data from the blood cell counter 1 is inputted to a data analyzing instrument 2 and whether the count value of the white blood cells is affected by the hemolysis defect of the red blood cells or not is decided. The result thereof is displayed. The instrument 2 sets the boundaries between the ghost components and small-size white blood cells in the white blood cell grain size distribution and designates the number of the particles in the boundaries as YWL, a mean red blood cell hemoglobin concn. as MCHC, and the total number of platelets as PLT; further, the instrument sets the boundaries between the red blood cells and the platelets in the platelet grain size distribution and designates the number of the particles in the boundaries as YPU. The instrument decides that the red blood cells affect the count value of the white blood cells if equation I or equation II is held when (a) to (f) are designated as constants. The blood inspection is exactly and efficiently executed in this way.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention relates to a blood data analysis method for obtaining specimen information useful in clinical tests from data obtained by counting blood cells or the like.

[Conventional technology]

2. Description of the Related Art Blood cell counters that count the number of white blood cells, red blood cells, etc. in blood are widely used in the field of clinical testing, and the obtained data is used as diagnostic data. The basic function of a blood cell counter is to obtain numerical data such as white blood cell count, red blood cell count, hemoglobin It, and platelet count, but recently it has become possible to also obtain particle size distribution charts for white blood cells, red blood cells, and platelets. It is becoming. A particle size distribution diagram, for example, shows the size of particles on the horizontal axis (X-axis).
This is a graph showing the relationship between particle size and frequency, with frequency plotted on the vertical axis (Y-axis).

For example, the standard particle size distribution of white blood cells is represented by a particle size distribution diagram as shown in FIG. 5, and this particle size distribution curve shows various changes when there is an abnormality in the sample. Therefore, based on this particle size distribution map, for example, neutropenia,
Various abnormalities can be determined, such as neutrophilia, lymphopenia, and lymphocytosis.

[Problem to be solved by the invention]

However, conventionally, such abnormalities have been determined by visually observing the particle size distribution map, and with this method, one has to rely on experience, so the determination of abnormalities may be uncertain. This results in individual differences and is inefficient. Furthermore, the representation format of particle size distribution is not necessarily uniform among multiple types of blood cell counters, and therefore particle size distribution charts are generally difficult to use.

By the way, when counting white blood cells, a sample for white blood cells that is prepared by subjecting blood to pretreatment such as dilution and hemolysis to destroy red blood cells is used. In this case, the red blood cell contents are eluted and the red blood cell membrane is reduced in size, and the platelets are also reduced in size.

These reduced red blood cells and platelets become blood shadow components (ghost components). When blood cells are counted using such a sample for white blood cells, the distribution of the ghost component appears in the part where the particle size is small in the particle size distribution diagram. However, since the size of white blood cells is larger than the ghost component,
White blood cells can be counted without the influence of red blood cells and platelets.

However, if poor hemolysis occurs, for example, when the blood contains yielded red blood cells, ghost components with large particles appear, which interferes with white blood cell counting. In such cases, it is not possible to accurately diagnose using the particle size distribution of white blood cells.
The only way to judge whether or not hemolysis has occurred is through the experience of the person looking at the particle size distribution map.

An object of the present invention is to provide a blood data analysis method that allows it to be easily determined that red blood cells are affecting the white blood cell count due to poor hemolysis as described above.

[Means to solve the problem]

The blood data analysis method of the present invention includes a sample for red blood cells and platelets obtained by diluting blood, a sample for white blood cells obtained by diluting and hemolyzing blood to reduce the size of red blood cells and platelets, and a sample for hemoglobin obtained by diluting and hemolyzing blood. From these, the particle size distribution of red blood cells, platelets, and white blood cells, as well as the average red blood cell type pigment concentration' are determined, and in the white blood cell particle size distribution, the ghost component consisting of the reduced red blood cells and platelets and the small white blood cells are determined. Set a boundary and let the number of particles at this boundary be YWL, let the mean red blood cell pigment concentration be MCHC, let the total number of platelets be PLT, and in the platelet particle size distribution, set the boundary between red blood cells and platelets and calculate the number of particles at this boundary. is YPU, and a, b, c, d, and e are constants, and if YWL>a and b>MCHC>c or YWL>d and PLT<e and YPU<f, red blood cells are counted as white blood cells. A characteristic [effect] characterized in that it is determined that the ghost component has an influence. When a relatively large ghost component remains in a white blood cell sample due to poor hemolysis of red blood cells, the difference between the ghost component and small white blood cells in the white blood cell particle size distribution The boundary becomes unclear, and as a result, the number of particles increases in the so-called trough of the distribution where the number of particles is small between the two distributions, that is, the part set as the boundary. Therefore, first, as a condition for determining poor hemolysis, YWL>a can be listed, where the number of particles at the boundary is YWL and a is a constant.

Cases in which poor hemolysis occurs are classified into cases of bell-shaped red blood cells and other cases. First, in the case of bell-shaped red blood cells, poor phlebotomy occurs in a hemoglobin sample that undergoes hemolysis treatment in the same way as a white blood cell sample, and this becomes a turbid component that causes scattering, resulting in an apparently high hemoglobin amount. This increases red blood cell hemoglobin concentration.

Therefore, when poor hemolysis occurs due to the influence of bell-shaped red blood cells, MCHC>c where the mean corpuscular hemoglobin concentration is MCHC and C is a constant. Furthermore, if the mean corpuscular hemoglobin concentration M CHC is higher than a certain value, there is a possibility of red blood cell agglutination, so it is necessary to exclude such cases. In this invention, in consideration of the above, in determining whether or not red blood cells affect the count of white blood cells, first, in the first case, Y W L > a and b > MCHC > c. If so, it is determined that it is having an impact.

The second case is when large platelets are present, although red blood cells may affect the white blood cell count due to poor hemolysis even in cases other than linear red blood cells.

Platelets affect the white blood cell count when platelet aggregation occurs and when the platelet count is abnormally high. If the constant that also holds in this case is d, then YWL>d holds true. Therefore, with this judgment condition alone, it is not possible to judge whether or not red blood cells have an influence on the white blood cell count. Particle size distribution is referred to. That is, when large platelets are present or platelet aggregation occurs, the distribution of platelets and the distribution of red blood cells overlap, and as a result, the number of particles Y P tJ at the boundary between the two increases. Therefore, if the total number of platelets is PLT and the constants e and f are such that Y W L > d, P L T < e, and YPU < f, then red blood cells, not platelets, affect the white blood cell count. It can be determined that the

In this invention, when at least one of the determination conditions in each of the first or second cases is satisfied,
It is determined that red blood cells are influencing the white blood cell count.

〔Example〕

FIG. 1 is a block diagram showing the overall configuration for implementing an embodiment of the present invention. Data from the blood cell count@box 1 is input to the data analysis device 2 as a serial signal.

In the data analysis device 2, the serial signal is input to a data communication circuit 4 via a line driver 3 and converted into a parallel signal. This parallel signal is sent to the CPU
(Central processing unit) 5 and arithmetic means 7 comprising memory 6
is input. In this calculation means 7, the data from the blood cell counter 1 is analyzed, and the results are printed out to a printing device 9 via an input/output interface 8, and are also output to a CRT (cathode ray tube) or liquid crystal display device. The image is displayed on a display device 10 such as the following.

The contents output to the printing device 9 or the display device IO include detection data in the blood cell counting bag [1 and various messages based on the analysis results thereof. This message includes: ■Abnormal message indicating that an abnormality clearly exists in the sample.

■ Suspect message indicating that there is a high possibility that an abnormality exists in the sample, ■ Normal message indicating that the sample is normal with high accuracy, and ■ Above ■ ~
There is an action message that instructs the next action to be taken in response to the message (①).

The calculation means 7 is prepared with the above-mentioned messages (1) to (2) corresponding to each of white blood cells, red blood cells, and platelets, and the suspect messages regarding white blood cells will be explained in more detail below.

Regarding white blood cells, the following suspect messages are available.

(a) RBCIntf: Red blood cells may affect the white blood cell count.

(b) EO/BASO/B P, a s t s
:Eosinophil.

There is a suspicion that either basophils or blast cells are increasing.

(c) MONO/EO/Beasts: monocytes,
There is a suspicion that either eosinophils or blast cells are increasing.

(d) TncroMonocytes: An increase in monocytes is suspected.

(e) BNasts/1mm, Ce1ls: The presence of blasts or immature leukocytes is suspected.

(f) ReacL, Lymph: Presence of atypical lymphocytes is suspected.

(g) 1mm, Neut, /Le f tSh
ift: Immature neutrophils are present, and the nucleus is shifted to the left (L6ft
5hift).

(h) NRBC/PLT C4! umps/Fi
b.

Intr: There is nucleated red blood cells, platelet aggregation, or fibrin precipitation, which is suspected to be affecting the white blood cell count.

(i) NRBC: The presence of nucleated red blood cells is suspected.

(j) Incr, Eo: An increase in eosinophils is suspected.

Qc) Large Cells: The presence of large white blood cells is suspected.

6'n) TtMiss-3et: A setting error in the boundary T2 is suspected (the boundary Tt will be described later).

In this embodiment, (a) "RBCIntf"
This is related to the determination when outputting the suspect message ".". The counting of white blood cells, red blood cells, and platelets in the blood cell counter 1 and the data processing thereof are performed according to the following procedure.

First, blood collected from a specimen is subjected to anticoagulation treatment, and then subjected to the blood cell counter 1. Blood cell counter! 1, pretreatment such as blood dilution and hemolysis is performed, and the sample subjected to this pretreatment is introduced into a detection section having micropores, where blood cells are counted.

For white blood cell samples, dilution and hemolytic treatment destroy red blood cells that can get in the way when counting white blood cells.

That is, the contents of the red blood cells are eluted, the red blood cell membrane is reduced in size, and the platelets are further reduced in size. In the hemolytic treatment, white blood cells also shrink, but since they have a nucleus, they remain as particles. By performing this hemolysis treatment under certain conditions, the lymphocytes in the white blood cell sample are made into relatively small particles, the neutrophils are made into relatively large particles, and other monocytes and eosinophils are made into particles in between. It can be made into particles of a certain size.

Furthermore, between monocytes and eosinophils, eosinophils can be made into slightly larger particles.

The white blood cell sample subjected to such hemolytic treatment is introduced into the leukocyte system detection section in the hemocytometer 1, and the red blood cell and platelet samples that have not been subjected to hemolytic treatment are introduced into the red blood cell system detection section. Each detection unit has, for example, a first chamber and a second chamber separated by a partition wall having micropores, and allows the sample to pass through the partition wall, and the first chamber and the second chamber are separated when particles pass through the partition wall. It is designed to detect changes in impedance between. This change in impedance corresponds to the size of particles passing through the micropores, and therefore the size of particles passing through the micropores can also be detected. The signal from such a detection unit is converted from analog to digital, etc.
By aggregating each digital value, white blood cells,
Particle size distributions of red blood cells and platelets can be obtained.

The standard distribution of particles in a leukocyte sample after hemolysis is shown in FIG. The white blood cell sample contains ghost component G (shrinked red blood cell membrane and small blood #Ii).
, small particles SC (such as lymphocytes).

Medium-sized particles MC (such as monocytes and eosinophils) and large particles LC (such as neutrophils) are mixed together, each with variations in size. The particle size distribution obtained from a leukocyte sample having such a distribution is a composite of the above-mentioned components, and the particle size distribution diagram is shown in FIG. 3.

That is, reference numerals 11 and 12.1 correspond to small particles SC1, medium particles MC, and large particles LC, respectively.
The mountains shown in 3 will be created.

There is a boundary WL at the valley part of such a particle size distribution curve.

By setting T,,T,, it is possible to count the number of small, medium, and large particles in white blood cells.0If there is no valley in the particle size distribution curve, set the boundary at the inflection point. However, in medium-sized particles MC, monocytes are distributed toward relatively small size, and eosinophils are distributed toward relatively large size.

Generally, the valley of the particle size distribution moves under the influence of the frequency of adjacent particle groups. That is, for example, particle A
Assume that the particle size distribution of a particle group including particles B and particles B, which are larger than particles A, is as shown in FIG.

However, the mountain indicated by reference numeral 2A is caused by particle A, and the peak indicated by reference numeral 1B is caused by particle B.

In addition, the frequency of particle B is different in FIG. 4 (+1. (2)), and the frequency of particle B is greater in the case shown in FIG. 4 (2).

As is clear from the comparison of Figure 4 (1) and (2), the valley ■ of the particle size distribution increases as the frequency of particles B increases.
shift towards the distribution of That is, when the frequency of the particle is constant, it is possible to determine whether the frequency of the particle B increases or decreases by the shift of the valley.

Furthermore, due to the increase in the frequency of the particles B, the frequency at the valley ■ becomes a high value. This can also be used as a criterion for determining an increase or decrease in the number of particles B.

Figure 5 shows the white blood cell sample! ! An example of a typical particle size distribution is shown. The horizontal axis is the particle size, and the vertical axis is the relative frequency (%) with the peak of the distribution as 100, which is the boundary W.
L, T+, and Tz are set in the valley between the distributions of ghost component G, small white blood cells SC1, medium white blood cells MC, and large white blood cells LC, respectively.

Numerical data regarding this leukocyte sample are as follows.

White blood cell count 63 XIO"/μ2 Small white blood cell ratio 36.8% Medium white blood cell ratio 10.0% Large white blood cell ratio 53.2% Small white blood cell count 23 XIO"/μ2 Medium white blood cell count 6 XIO"/μl Large white blood cell count 34 XiO" /μl Figure 6 shows the standard particle size distribution for red blood cell and platelet samples.
The illustration is similar to that of FIG. 5. In samples for red blood cells and platelets, platelets P appear as relatively small particles, and red blood cells R appear as relatively large particles. In this case, a boundary RL is set in the valley between the two, platelets and red blood cells are classified, and red blood cells and larger particles are separated by the boundary R.
Classified with U. The particle size distribution regarding platelets is shown in FIG. 7, where boundaries PL and PU are set, and particles in the area above the boundary PU are classified as red blood cells R. Boundary PL
Noise components are distributed in the following areas. An example of numerical data obtained from red blood cell and platelet samples is shown below.

Red blood cell count 417 XIO'/μl hemoglobin! 13.1 g/a Hematolift value 39.4 % Mean corpuscular volume 94.5 fρ Average directional pigment amount 31.4 pg Mean corpuscular hemoglobin concentration 33.2 g/a Red blood cell distribution width 44.1 fj! Platelet count 23
．． 3 XIO'/μl platelet distribution width 11.5
fffi average platelet volume 10.5fj! Large platelet ratio: 27.4% The red blood cell distribution width is the distribution width at a frequency of 20%. Further, the large platelet ratio is the ratio of the number of platelets of a predetermined size, for example, 12fj2) or more, to the total number of platelets. The amount of hemoglobin is obtained from a hemoglobin sample subjected to hemolytic treatment, and in the hemocytometer 1, the hemoglobin sample is introduced into its colorimetric measurement section and the amount of hemoglobin is measured. As the hemoglobin sample, a part of the white blood cell sample may be used, or it may be prepared separately using a dedicated hemolytic agent. In the latter case, a sample is prepared that is more suitable for measuring the amount of hemoglobin by suppressing the generation of ghost components and stabilizing the light absorption characteristics.

Each data regarding the sample for white blood cells and the sample for red blood cells and platelets is data given to the data analysis device 2 from the blood cell counting bag Wl.

White blood cells are composed of multi-Wi particles, and their content ratio varies greatly. Furthermore, depending on the disease, cells that do not normally appear may appear. Therefore, the particle size distribution of the leukocyte sample is highly variable, and abnormalities such as various diseases can be determined from the particle size distribution of the leukocyte sample.

In the following, the procedure up to the output of RBCIntf, shown in (a) in the above-mentioned suspect message will be explained.

This suspect message indicates that red blood cells may be influencing the white blood cell count. As mentioned above, white blood cell samples are subjected to hemolytic treatment, but if red blood cell hemolysis occurs during this hemolytic treatment, a relatively large ghost component remains in the white blood cell sample, which causes white blood cells and ghosts to be separated. This makes it difficult to distinguish between the components and adversely affects the accuracy of white blood cell counting.

Therefore, when determining "RBCIntf," it is sufficient to detect the above-mentioned defective hemolysis.

The particle size distribution of the leukocyte sample in the case of poor hemolysis is shown in FIG. The first sign of poor hemolysis appears in the particle size distribution of the leukocyte sample as an increase in the relative frequency YWL of the valley where the boundary WL is set.

That is, the relative power YWL in FIG. 8 is greater than the relative power YWL in the case of FIG.

If defective hemolysis is further classified into two types, the first is a case of bell-shaped red blood cells, and the second is a case where defective hemolysis occurs regardless of whether the cell is a bell-shaped red blood cell.

(i) Case of bell-shaped red blood cells In this case, the determination is made with reference to the mean red blood cell hemoglobin concentration as well as the relative frequency at the boundary WL in the white blood cell particle size distribution. This mean corpuscular hemoglobin concentration M CHC
(g# is calculated as ct from the hemoglobin amount HGB (g/a) and hematocrit (11Hc, t (%)).

In the case of bell-shaped red blood cells, poor phlebotomy may also occur in the sample for hemoglobin, and therefore a larger ghost component than in the normal case may be generated in the sample. This becomes a turbid component and causes scattering during colorimetric measurements, resulting in a phenomenon in which hemoglobin IHGB becomes apparently high. Therefore, the value of mean corpuscular hemoglobin concentration MCHC becomes higher than normal.

Therefore, by evaluating this mean corpuscular hemoglobin concentration MCHC together with the relative frequency YWL at the boundary WL in the white blood cell particle size distribution, it is possible to determine whether hemolysis is caused by bell-shaped red blood cells.

However, when the mean corpuscular hemoglobin concentration MCHC is excessively high, there is a possibility of red blood cell agglutination, so such cases must be excluded. Therefore, with a, b, and c as constants, YWL>a(%)... (2)
and b (g/a)>MCHC>c (g/a)
When .=(3), it can be determined that poor hemolysis due to bell-shaped red blood cells has occurred. constants a, b
.

It is desirable that C be set as follows.

a... 15-25 b... 40.1 C... 37.0 It has been confirmed that extremely accurate judgments can be made by setting in this way. .

Note that the smaller the constants a and c tend to be, the more over-detection will occur, and the larger the constants a and c, the more the detection will be missed. Furthermore, when the mean corpuscular hemoglobin concentration MCHC exceeds 40, it is considered that red blood cell agglutination has occurred.

(11) When red blood cells affect white blood cell counts due to factors other than yield red blood cells, insufficient hemolysis of red blood cells leaves many larger ghost components than in the case of sickle cells, and these ghost components are distributed in small white blood cells. This is the case when a large portion of the two overlaps, making it impossible to distinguish between the two. However, since platelets are not destroyed by hemolysis like red blood cells, a similar phenomenon occurs when the platelet count is abnormally high, when large platelets are present, or when platelet aggregation occurs. Therefore, in order to more reliably detect defective hemolysis of red blood cells, it is necessary to exclude cases caused by the influence of platelets.

FIG. 9 shows the particle size distribution of platelets when large platelets are present or when platelet aggregation occurs. When large platelets or platelet aggregation are present, in the platelet particle size distribution, the relative frequency YPU at the upper boundary PtJ for classifying platelets, which are small particles, and red blood cells, which are large particles, becomes higher than normal. That is, in this case, the relative power YPU at the boundary PU is greater than the relative power YPU in the normal case shown in FIG.

Therefore, defective hemolysis of red blood cells can be determined as follows. In other words, taking into consideration the case where the platelet count PLT is abnormally high, YWL>d (%) ... (4
) and PLT<e (XIO'/μjri -(5) and YPU<f(%)... (6)
If this is the case, it is determined that the phlebotomy is defective. Constant d is 75-95
It is preferable that the constant e be selected to be a value of about 60 and the constant r be a value of about 15 to 25.

Each judgment regarding the hemolysis of red blood cells according to (i) and (i) above is performed in the data analysis bag W2, and if poor lysis of red blood cells is recognized by any of the judgments, it is determined that the red blood cells affect the count of white blood cells. RBCInt is given to the printing device 9 or the display device 10.
f," suspect message is output. At this time, at the same time, "Verify
Res, - is output and the tester is informed of the measurement result (Res
prompt for confirmation of the unit).

As described above, according to this embodiment, it is determined based on certain determination conditions whether or not the red blood cells are affecting the white blood cell count due to insufficient hemolysis of the red blood cells, and this determination is made by the data analysis device 2. The result is displayed on the display device 1.
It is displayed and output as 0, etc. Therefore, such a determination can be made accurately without depending on the experience of the examiner, and there are no individual differences among the people viewing the particle size distribution.

Furthermore, since the above-mentioned determination is performed automatically, so to speak, the work efficiency of the inspection can be significantly improved.

〔Effect of the invention〕

As described above, according to the blood data analysis method of the present invention, it is determined whether or not red blood cells have an influence on the white blood cell count based on certain determination conditions. It becomes possible to perform the process accurately and efficiently without relying on other factors. This makes it possible to perform blood tests with good workability and accuracy.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the overall configuration for carrying out an embodiment of the present invention, FIG. 2 is a diagram showing the standard distribution of particles in a leukocyte sample after hemolysis treatment, and FIG. 3 is a particle size distribution diagram corresponding to the particle distribution state shown in Figure 2, Figure 4 is a particle size distribution diagram to explain changes in the particle size distribution curve due to changes in particle size distribution, and Figure 5 is a standard for white blood cell samples. Figure 6 is a particle size distribution diagram showing the standard particle size distribution of samples for red blood cells and platelets, Figure 7 is a standard particle size distribution diagram of platelets, and Figure 8 is the ghost component. A particle size distribution diagram showing the particle size distribution of a leukocyte sample when
FIG. 9 is a particle size distribution diagram showing the particle size distribution of platelets when large platelets are present or platelet aggregation occurs. l... Blood cell counter, 2... Data analysis device No. 1
Entry point of zubroom 2 zu grains) - 3rd figure) human crowd □ 誼+の八さigo ichigo 5 zuzoshishi jingucho (f12) - Goro zu a'+ human pupil ( fλ) - Figure 7 Shin)'s human life (fR) - Figure 8 s9 @shi's 8''ff (fR) =

Claims (1)

[Claims] A sample for red blood cells and platelets obtained by diluting blood, a sample for white blood cells obtained by diluting and hemolyzing blood to reduce the size of red blood cells and platelets, and a sample for hemoglobin obtained by diluting and hemolyzing blood. From these, the particle size distribution of red blood cells, platelets, and white blood cells as well as the average red blood cell hemoglobin concentration are determined, and in the white blood cell particle size distribution, boundaries are set between the ghost component consisting of the reduced red blood cells and platelets and small white blood cells, and this is determined. The number of particles at the boundary is YWL, the mean corpuscular hemoglobin concentration is MCHC, the total number of platelets is PLT, the boundary between red blood cells and platelets is set in the platelet particle size distribution, and the number of particles at this boundary is YPU, a, b , c, d, and e are constants, and if YWL>a and b>MCHC>c or YWL>d and PLT<e and YPU<f, it is determined that red blood cells are affecting the white blood cell count. A blood data analysis method characterized by: