JPH0146815B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to a method and apparatus for counting and measuring platelets in blood. Conventionally, when measuring platelets, methods such as the platelet-rich plasma method (PRP method), which requires sample pretreatment, and the visual method, which requires specimen preparation, have been used. In the PRP method, blood is collected in a capillary tube, red blood cells and white blood cells are sedimented by a static method or a centrifugal sedimentation method, and only the platelet-rich plasma layer is extracted.The number of platelets in this layer is counted using a particle counter. It counts and measures each piece. In this method, the suspended portion of platelets is concentrated in order to separate red blood cells and white blood cells, so the number of platelets counted and measured above is a corrected value related to the separately determined blood hematocrit value. After correction, it is converted into the number of platelets in blood per unit volume. Although this PRP method has relatively high accuracy and does not require skill in sample preparation, it takes time and effort to prepare platelet-rich plasma, and even a small amount of red blood cells can cause a large error, and conversion through correction is difficult. There are drawbacks such as being cumbersome, which makes it difficult to automate the PRP method. In the conventional visual method, a blood smear is created and counted directly under a microscope, but the distribution of platelets on the smear surface is uneven, and smearing requires skill. In addition, platelets tend to adhere to the glass surface of the preparation and are easily destroyed. The present invention is completely different from the conventional methods as described above, and aims to provide a measuring method and apparatus that can automatically and highly accurately perform platelet counting simply by diluting a blood sample similar to red blood cell counting. purpose. The present invention also realizes an automatic counting method in which the correction value used for counting and measuring the number of platelets is only the number of red blood cells, and the calculation of the final result is relatively simple, and an automatic counting device that is relatively inexpensive to manufacture. The purpose is to Furthermore, it is an object of the present invention to provide a counting method and device that is simpler and more accurate for screening measurements. The automatic measurement technology according to the present invention can be incorporated into an automatic measurement device for blood-related parameters that can simultaneously measure other blood-related parameters such as red blood cell count, white blood cell count, hemoglobin value, and hematric value. Sometimes it is also possible to measure the platelet count in a new manner. By the way, when counting and measuring platelets immediately by simply diluting whole blood, the diluted suspension sample is sucked into the detector and drifts into the micropore area where changes in electrical impedance are detected when particles pass through. Particles other than the attached platelets enter this region by a so-called entrainment phenomenon and generate a pulse signal, causing an error. For this reason, in order to prevent errors caused by the intrusion of this type of particles, conventional methods used a double structure inside the detector to suck all particles, such as red blood cells and platelets, through the micropores and from the periphery of the micropores. Efforts have been focused on structural improvements to the detection area, such as supplying a clean diluent. According to this, platelet counting can be performed accurately enough by simply discriminating the wave height of the electrical pulses generated by particles, but it is necessary to supply a clean diluent that is free of any foreign matter or particles, and to use a double-walled minute detection area. It has not been used in reality due to structural and manufacturing cost problems. The present invention statistically calculates in advance the increase in the number of output pulses due to the involvement of so-called red blood cells (and white blood cells) in the detection area where particle passage is detected from changes in electrical impedance in the same way as described above, and then increases the output by platelets. The principle is to correct the number of pulses generated by red blood cells (and white blood cells). Numerous experiments and calculations were iterated to arrive at such statistical techniques. When implementing the above detection method, the difference between the actual value of platelets measured by pulse height discrimination and the value measured by the conventional method is proportional to the number of red blood cells in the sample (the number of white blood cells is much smaller than this and can be ignored). Assuming that, the number of red blood cells
RBC/100 [×10 ⁴ /mm ³ ] was used to calculate the actual measured value of this method - the actual measured value of conventional method = K·RBC/100, and the average value = 4.66 was obtained, and the corrected platelet count was calculated again. As a result, the correlation coefficient before correction is r=
0.9072, and the correlation coefficient after correction by the average value was reduced to r=0.8474. This means that when the platelet count is measured by the pulse height discrimination described above, where the corrected count value of this method = the actual measured value of this method - a ₀ (a ₀ is a constant), the correlation coefficient with the conventional method is r = 0.9072. Not only did it show that the effect of erythrocytes was high, but it also appeared to be independent of the number of red blood cells. As mentioned above, since the error in the platelet count value is mainly due to the inclusion of the red blood cell count due to the so-called entrainment phenomenon, it was necessary to introduce this RBC count into a meaningful correction formula. Therefore, we investigated the recount reproducibility of the actual measured values using this method.
This was extremely bad, and it turned out that the cause was red blood cells that had gotten inside the detector. Therefore, we next attempted correction using least squares polynomial approximation, which includes the above-mentioned correction method. Actual measurement value of platelets using this method − Actual measurement value using conventional method = a ₀ + a ₁・RBC/100+a ₂・ (RBC/100) ² …+a _n (RBC/100) ^mAs described above, for the 48 samples, the first time of each sample, No correction for 2nd and 3rd counting, m=
0, m=1, m=2, and m=3, but in the first counting, r=0.8 in all cases, which was not very good. Furthermore, in the first counting, when no correction was made, the correlation coefficient r=0.8657, and r=0.8698 up to the correction term of m=1, which was almost unchanged.
On the other hand, in the third counting, when there is no correction, r
= 0.9250, up to the correction term of m = 1, r = 0.9772
I got better. Therefore, it has been found that it is sufficient to take the actually measured value of platelets corresponding to the third counting of the same sample and correct it up to m=1 using least squares polynomial approximation. These results are shown in the attached figures Figures 1 and 2.
Figure 3 and Table 1 below.

[Table] Next, the linearity of the same sample was investigated using this method.
In other words, when RBC (red blood cell count) and the actual value measured by this method are graphed as shown in Figure 4, the actual value measured by this method = b ₀ + b ₁ · RBC / 100, which is calculated by first calculating the actual value of this method minus the conventional value. Actual measured value =
Completely corresponds to a ₀ + a ₁ RBC/100. As is clear from Table 1 and Figure 3, the recount reproducibility of this method improves from the third measurement onward, so we tested the reproducibility after filling the detector with the sample to be measured. The stability of the measured values was superior to that of the conventional method. From the above points, diluting a blood sample (whole blood is acceptable) to prepare a suspension of blood cell particles, individually detecting the electrical pulse signals of the particles contained in this suspension, From the detected pulse signal, distinguish and count the first pulse output that does not reach a predetermined pulse height and the second pulse output that exceeds this predetermined pulse height, and calculate the actual measured value P of the first pulse output. From the formula PL = _{P n −} {a ₀ + a ₁・RBC/100+a ₂ (RBC/100) ² +… +a _n (RBC/100) ^m } from n (unit: 10 ⁴ /mm ³ ), platelet count PL is calculated. (Unit: 10 ⁴ /mm ³ ) was clearly required. Next, when measuring separate samples one after another,
Again, using the number of red blood cells as a correction parameter, the formula P _n - PL = a ₀ + a ₁・RBC(i)/100+a ₂・RBC(i-1)/10
0 +...+a _n・RBC(i-(m-1))/100 holds true. Here, RBC(i) is the number of red blood cells in the current sample, and RBC(i-1) is the number of red blood cells in the previous sample.
Further, RBC (i-(m-1)) is the number of red blood cells in each specimen from (m-1) times ago. As mentioned earlier in relation to the experimental results, considering the influence of the samples from the previous two times in terms of reproducibility, we applied the samples from this time, last time, the time before last, and the time before last to the above formula, RBC(i), RBC(i-1), RBC
(i-2), each coefficient a ₁ of RBC (i-3),
A ₂ , a ₃ , a ₄ and the constant a ₀ were statistically determined from numerous experiments. The results are as follows. a ₀ = 0.9, a ₁ = 0.632, a ₂ = 2.56, a ₃ = 0.579, a ₄ = 0.683, therefore, the formula PL = P _n − {0.9 + 0.632・RBC(i)/100 + 2.56・R
BC(i-1)/100 +0.579・RBC(i-2)/100+0.683・RBC
(i-3)/100}, and the same results as in FIG. 3 were obtained. Once a ₀ , a ₁ , a ₂ , a ₃ , and a ₄ are determined in this formula,
Even when measured using a different device, there was almost no deviation in the measurement results, and almost the same accuracy of platelet count measurement was obtained. The measurement conditions in this case are: the diameter of the fine hole in the detector is 80 μ;
The inner diameter of the detector is 6 mm, the amount of sample drawn into the detector at one time
The measurement time for each sample was approximately 11 seconds. Therefore, unless these conditions are changed, the constants and coefficients described above are universal, as is clear from FIG. On the other hand, if some of these conditions are changed, calculate P _n −PL for several dozen or more samples in advance, and then calculate a ₀ ,
We need to find ~a _n . The correlation coefficient obtained from the above equation was r=0.9445. Furthermore, from the above experimental results, it was found that by using the actual measurement examples up to the third time, a sufficiently accurate result could be obtained using the formula PL=P _n −0.0331×RBC−4.8. Here, the normal value for RBC (red blood cell count) is approximately 450 for men.
~510 [×10 ⁴ /mm ³ ], women about 395 to 465 [×10 ⁴ /mm 3 ]
mm ³ ], the correction value of the above formula is 19.7~ for men.
21.68, 17.87 to 20.19 for women, the respective average values are 20.69 and 19.03, and the corrected value when removing the distinction between men and women and taking the overall average is RBC = 395 ~
510 [×10 ⁴ /mm ³ ], so 19.86 is obtained. Therefore, if it is possible to differentiate between men and women, confirm that the red blood cell count is within the normal range (if abnormal, just display the abnormality), and then use the actual platelet count to confirm that it is within the normal range. By subtracting the corresponding correction values of 20.69 and 19.03, a sufficiently accurate platelet count can be obtained, and application of this simplified formula is effective in screening measurements. When measuring regardless of gender, simply
Just by subtracting 19.86, you can find the almost correct platelet count (unit: x10 ⁴ /mm ³ ). When it is determined that the platelet count shows an abnormal value, it is appropriate to remeasure or repeat the measurement using a conventional method or a more precise measurement method. As described above, it can be seen that screening-like platelet count measurement can be performed by simply calculating a ₀ and a ₁ for several dozen samples in advance using the formula PL=P _n -C. An embodiment of an apparatus for carrying out the method of the invention will be described below. FIG. 5 shows a simple device that counts and measures the number of platelets by simply subtracting a predetermined correction value when the number of red blood cells is within a normal range.
The output of the detection device 8 is to separately output a pulse output signal that has reached a predetermined discrimination level in the discrimination circuit 9 and a signal that has not reached the above-mentioned predetermined discrimination level, and calculate the former as the number of red blood cells (which may include white blood cells). A count determination circuit 15 determines whether the value is normal or not, while pulses in the platelet region are input to a count calculation circuit 16, where the above-mentioned correction value (C) is simply subtracted from the count value obtained here. Both the determination output from the count determination circuit 15 and the output from the count calculation circuit 16 are displayed or printed on the display circuit 17. In other words, the display circuit 17 displays the determination result of whether or not the number of red blood cells (which may include the number of white blood cells) is within the normal range (direct counts may be combined) and the corrected platelet count. be done. 6 and 7 show improvements in the vicinity of the detector 4. FIG. In FIG. 6, the inside and outside of the detector 18 are filled with the same suspension sample, and the number of red blood cells for correction calculation is always limited to that of the current sample. As a result, correction calculations can be easily performed because the red blood cell count of the previous sample is not affected at all. A non-liquid pipe 19 is provided at the lower end of the detector 18, and the sample inside the detector is discharged via a pot 20. The sample suspension liquid 2 is simultaneously introduced into the detector 18 and the sample chamber 23 through the introduction pipe 24 when the ports 21 and 22 are opened. The electrodes 26 and 27 can be electrically connected through a fine hole 28 formed in the wall of the detector 18. That is, when a predetermined amount of suspended liquid sample is sucked into the detector 18 from the sample chamber 23 by the quantitative device 29, when the blood cell particles pass through the detection area of the micropore 28, the electrical impedance between the electrodes 26 and 27 increases. The impedance at this time is taken out by the detection circuit 7 as a pulse output. While using the above devices, quantitative device 29, Kotoku 20, 2
1 and 22 may be electrically connected to the inside of the detector 18 by a suspension liquid 2 made of physiological saline or the like, but may be electrically connected to the outside of the detector 18 via a pot 25 or the like. There shall be no electrical leakage. In order to prevent this electrical leakage, it is preferable that, for example, the suspended liquid continues in the flow channel system as droplets as shown in FIG.

[Brief explanation of drawings]

Figures 1, 2, and 3 show a comparison between the platelet counts obtained by the method of the present invention and the conventional method.
A graph showing how the measurement accuracy improves when the influence of the red blood cell count is taken into consideration sequentially from the current time, the previous time, and the time before the previous time. Figure 4 shows how the platelet counts obtained by the method of the present invention and the conventional method differ depending on the number of red blood cells. Graph showing a comparison of whether or not the change has been made, including the presence or absence of correction, FIG. 5 is an embodiment of the device of the present invention, and FIGS. 6 and 7 are schematic longitudinal cross-sections showing improvements in the vicinity of the detector of the device of the present invention. It is a diagram. 1, 23...sample chamber or container, 2...floating suspension,
3... Micropore, 4... Detector, 6, 6', 26,
27... Electrode, 7... Detection circuit, 5, 29... Quantitative device, 8... Detection device, 9... Discrimination circuit, 17
... Display circuit, 15 ... Counting judgment circuit, 16 ...
Count calculation circuit.

Claims (1)

[Claims] 1. Diluting a blood sample to prepare a suspension of blood cell particles, detecting individual pulse signals of particles contained in the suspension, and calculating a predetermined pulse signal from the detected pulse signal. Discriminating and counting a first pulse output that does not reach the discrimination level and a second pulse output that is equal to or higher than the discrimination level, and determining that the number of red blood cells is a normal value from the actual measured value of the second pulse output. Confirm that it is within the range, and from the actual measured value of the first pulse output, use the formula PL [×10 ⁴ /mm ³ ]=Pm [×10 ⁴ /mm ³ ]−C (where PL is the platelet count ; Pm is an actual measurement value; C is a correction constant based on the red blood cell concentration in the suspension, which is 20.69 for men and 19.03 for women, and the overall average is
19.86) in the suspension liquid. 2. A detection device for detecting blood cell particles contained in a suspension obtained by diluting a blood sample as individual pulse signals based on changes in electrical impedance when passing through micropores of a detector, and detecting the individual pulse signals as individual pulse signals. A discrimination circuit for discriminating pulse output signals corresponding to areas of platelets that do not reach the discrimination level and pulse output signals corresponding to areas of red blood cells and white blood cells that are above the discrimination level; a counting determination circuit that determines whether the count value of the corresponding pulse output signal is within a normal range; a counting calculation circuit that counts the pulse output signal corresponding to the platelet region and simply subtracts a predetermined correction value; and a display circuit for displaying or printing whether the count values of red blood cells and white blood cells are within the normal range and corrected platelets based on the respective outputs of the counting calculation circuits. Features: Automatic platelet counting device in blood. 3. The automatic platelet counting device according to claim 2, wherein substantially the same suspension liquid is simultaneously supplied to the inside and outside of the detector, which are communicated through the micropores. Counting device.