Quality control method for medical examination and analysis

Abstract

The present invention relates to the field of medical examination and analysis, and in particular relates to a quality control method for medical examination and analysis. A quality-control product is reacted once, and a standard curve is fitted; and by means of comparing the fitted standard curve of the quality-control product with an original standard curve in a testing system to determine the degree of coincidence, whether quality control is acceptable is determined and the real state of the quality-control product is measured. The method of fitting standard curves on the basis of calibration products is abolished, such that the matrix difference between quality-control products and the calibration products is eliminated, quality control steps are simplified, the amount of quality-control products is reduced, the accuracy and stability of testing are improved, and the verification process of quality control management in the prior art is improved.

Description

A quality control method for medical test analysis Technical Field

[0001] The present invention relates to the field of medical testing and analysis, and in particular to a quality control method for medical testing and analysis. Background Art

[0002] The purpose of quality control is to ensure that the laboratory is correctly testing clinical samples, thereby ensuring accurate and reliable test results. Current quality control procedures for immunoassays and biochemical analyses generally require the use of at least two to three quality control products of varying concentrations along with samples every 24 hours. The test system already has a standard curve fitted with six calibrators. The quality control products are tested using this existing standard curve and the test results are plotted on a quality control chart to determine if the quality control has passed.

[0003] If the measured value falls near the center and within the upper and lower warning lines, quality control passes and the result is correct. Continue using the original standard curve. If the measured value falls outside the upper and lower control lines, the analysis is out of control and the result is unreliable. The following steps should be performed to investigate the cause, correct it, and then retest. Step 1: Retest the same control sample. This is primarily to determine if there is human error or accidental error. If the retest result is within the acceptable range, it is an accidental error. If the retest result is still outside the acceptable range, proceed to the next step. Step 2: Open a new bottle of control sample and retest the out-of-control item. If the result is normal, the original control sample may have expired, deteriorated due to prolonged storage at room temperature, or may be contaminated. If the result is still outside the acceptable range, proceed to the next step. Step 3: Open another batch of control sample and retest the out-of-control item. If the result is in control, there may be problems with the previous batch of control samples. Check the expiration date and storage environment to determine the cause. If the result is still outside the acceptable range, proceed to the next step. The fourth step is to maintain the instrument and re-test out-of-control items: check the instrument status, find out whether the light source needs to be replaced, whether the colorimetric cup needs to be cleaned or replaced, and perform other maintenance on the instrument. In addition, the reagents need to be checked. At this time, the reagents can be replaced to find out the cause. If the result is still not within the allowable range, proceed to the next step. The fifth step is to recalibrate and re-test out-of-control items: calibrate the instrument with a new calibration solution to rule out problems with the calibration solution. The sixth step is to ask an expert for help: If the first five steps fail to produce an in-control result, there may be more complicated reasons for the instrument or reagents. Contact the instrument or reagent manufacturer and request technical support. If the measured value falls between the upper and lower warning lines and the upper and lower control lines, although the analysis result is acceptable, there is a tendency for it to be out of control and attention should be paid.

[0004] Simply put: the existing quality control management verification process requires testing two to three concentrations of quality control products using a standard curve fitted with six concentrations of calibrators. The test results determine whether they are in control. If they are qualified, the original standard curve is continued to be used. If they are unqualified, the same quality control product needs to be re-measured first, and then the same batch of quality control products needs to be measured to eliminate quality control product problems. The calibrator reaction is then repeated six times to fit a new standard curve. The standard curve is then re-calibrated using two to three concentrations of quality control products. This process is repeated until quality control is qualified.

[0005] Calibrators and quality controls differ in their matrix, significantly interfering with the analytical process and affecting the accuracy of analytical results. Existing techniques use calibrators to fit a standard curve and quality controls for verification. This comparison can introduce bias due to the matrix.

[0006] The existing quality control technology has complicated operation steps, large consumption of quality control products and high prices. Many clinical laboratories have the requirement to reduce costs. Therefore, quality control methods that reduce the number of quality control operation steps and the amount of quality control products have very positive significance. Summary of the Invention

[0007] The technical solution for realizing the claims of the present invention requires two necessary conditions: (1) fitting the standard curve once; and (2) proving that the quality control products of different concentrations are all data points on the same standard curve.

[0008] In PCT patent applications (International Application No. PCT / CN2023 / 106493, Written Opinion of the International Searching Authority: All claims meet the requirements of novelty, inventive step, and industrial applicability; and International Application No. PCT / CN2023 / 106463, Written Opinion of the International Searching Authority: All claims meet the requirements of novelty, inventive step, and industrial applicability), a method of using a reaction curve mathematical model to fit a standard curve once has been confirmed.

[0009] An experiment using chemiluminescent reagents to verify that the data points of the control sample at different concentrations are all on the standard curve: a pre-created reaction curve mathematical model (see Figure 1) is used. The control sample at low, medium, and high concentrations is reacted once with the same substrate. Three standard curves are generated using the reaction curve mathematical model, and the three standard curves tend to overlap (see Figure 2). Within the same reaction curve mathematical model, standard curves generated using multiple control sample concentrations and the same reaction conditions tend to overlap. Comparing the degree of overlap between the standard curves can be used to determine quality control compliance and assess the true state of the control sample.

[0010] Because the quality control samples represent different data points on the same standard curve, the curve is restored using the reaction curve mathematical model after each reaction. This is also proven by the existing technology of using quality control samples to verify the standard curve. This can be used to verify the normal status of the quality control samples by comparing the degree of overlap between the reaction of different quality control samples and the fitted standard curve.

[0011] The present invention combines the above two technical features and applies them in quality control management. The quality control product reacts to a fitted standard curve. By comparing the degree of overlap between the standard curve fitted by the quality control product and the original standard curve in the detection system, it is determined whether the quality control is qualified and the true state of the quality control product is measured at the same time. If it is qualified, the original standard curve is used. If it is unqualified, another concentration of quality control product is selected for measurement, the standard curve is fitted, and the degree of overlap between the three standard curves is compared. If the standard curves fitted by two quality control products overlap, it means that the original standard curve in the detection system is offset, and the curve fitted by the quality control product is used; if the original standard curve overlaps with the standard curve fitted by a certain quality control product, it means that the state of the other quality control product is abnormal, and the original standard curve continues to be used.

[0012] The technical solution of the claim is as follows: The present invention provides a quality control method for medical test analysis, which is characterized by comprising the following steps: step 1, reacting a quality control product of any concentration once, and fitting a standard curve; step 2, using the standard curve fitted in step 1 to compare with the standard curve currently being used in the detection system, if the degree of overlap is within the allowable range, it means that the quality control is qualified, and the standard curve currently being used by the detection system is retained and continued to be used; step 3, if the degree of overlap exceeds the allowable range, it means that the quality control is unqualified, and using the quality control products of different concentrations in step 1 to react once, and fitting a standard curve; step 4, comparing the standard curve fitted in step 1, the standard curve fitted in step 3 and the standard curve currently being used by the detection system for degree of overlap; step 5, if the degree of overlap between the standard curve fitted in step 1 and the standard curve fitted in step 3 is within the allowable range, it indicates that the quality control product is in normal condition, and both standard curves can replace the standard curve currently being used in the original detection system; step 6, if the degree of overlap between the standard curve fitted in step 3 and the standard curve currently being used by the detection system is within the allowable range, it indicates that the quality control product is in abnormal condition, and the standard curve currently being used by the detection system is retained and continued to be used.

[0013] The medical test analysis is biochemical analysis and immunoassay.

[0014] The fitting standard curve is carried out in the same reaction system.

[0015] The comparison of standard curve coincidence refers to comparing the difference in slope and intercept of different standard curves, or comparing the difference in concentration of the same quality control product in different standard curves.

[0016] The allowable range of coincidence is the maximum error allowed for the quality control product.

[0017] It can be seen that the quality control product is in a qualified state, and the reaction is fitted to the standard curve once. The maximum error of the standard curve is also the error allowed by the quality control product. This error is definitely smaller than the error of the original calibration product fitting the standard curve, because the matrix difference between the calibration product and the quality control product is eliminated, making the detection more stable and accurate.

[0018] The present invention abolishes the method of fitting the standard curve with calibrators, eliminates the matrix difference between quality control products and calibrators, simplifies the quality control steps, saves the amount of quality control products, improves the detection accuracy and stability, and subverts the verification process of quality control management in the existing technology.

[0019] The following table compares the differences and progress of the present invention with the prior art.

[0020] Table 1 The differences and progress of the present invention compared with the prior art

[0021] BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1: Mathematical model of chemiluminescence reaction curve.

[0023] Figure 2: A: Standard curve of the low-value quality control; B: Standard curve of the mid-value quality control; C: Standard curve of the high-value quality control; D: Three standard curves of the low, mid, and high-value quality controls.

[0024] Figure 3: Multiple reaction curves of 25-100 U / L GALT quality control and 20%-40% H2O2.

[0025] Figure 4: Fitting standard curve of 50U / L GALT quality control.

[0026] Figure 5: Original standard curve in the detection system of Example 1.

[0027] Figure 6: Fitting standard curve of 100U / L GALT quality control.

[0028] Figure 7: Multiple reaction curves of 50-500 ng / mL AFP quality control and 14.1-42.3 mmol / L H2O2.

[0029] Figure 8: Fitted standard curve of 100 ng / mL AFP quality control.

[0030] Figure 9: Original standard curve in the detection system of Example 2.

[0031] Figure 10: Multiple reaction curves of 0.08-0.15 mmol / L glucose control and 2-6 U / mL glucose oxidase.

[0032] Figure 11: Fitted standard curve of 0.1 mmol / L glucose quality control.

[0033] Figure 12: Original standard curve in the detection system of Example 3.

[0034] Figure 13: Fitted standard curve of 0.15mmol / L glucose quality control. Modes for Carrying Out the Invention

[0035] The following examples are intended to deepen understanding of the present invention and do not represent the entire content of the present invention. The present invention includes but is not limited to the following examples.

[0036] In the embodiment, the degree of coincidence of the standard curves is measured by comparing the difference values ​​of the same quality control product in different standard curves with the maximum allowable error of the quality control product.

[0037] Example 1

[0038] Chemiluminescent enzyme immunoassay technology was used to prepare multiple reaction curves of different concentrations of galactose uridyltransferase (GALT) quality control products (25U / L, 50U / L, 75U / L, 100U / L) and different concentrations of H2O2 (20%, 25%, 30%, 35%, 40%) (see Figure 3). The marker was horseradish peroxidase (HRP) and the luminescent substrates were luminol and H2O2.

[0039] 50U / L of GALT quality control product was reacted with 30% H202 once, and a standard curve for GALT analysis was fitted using the reaction curve mathematical model (see Figure 4).

[0040] Substituting the reaction result of the 50U / L GALT quality control product into the original standard curve of the detection system (see Figure 5), the calculated detection value was 57.1U / L, with a relative deviation of 14.21%, which exceeded the maximum allowable error of the quality control product and was considered to be unqualified.

[0041] Another concentration (100 U / L) of the quality control product was selected to react once with 30% H202, and a standard curve for GALT analysis was fitted using the reaction curve mathematical model (see Figure 6).

[0042] Compare the degree of coincidence between the standard curves fitted by the two concentration quality control products and the original standard curve in the detection system.

[0043] Substituting the reaction result of the 100 U / L GALT quality control product into the original standard curve of the detection system, the calculated detection value was 108.94 U / L, with a relative deviation of 8.94%, which exceeded the maximum allowable error of the quality control product;

[0044] Substituting the reaction results of the 50U / L GALT quality control product into the standard curve fitted by the 100U / L quality control product, the calculated detection value was 51.72 U / L, with a relative deviation of 3.44%, which is within the allowable error range of the quality control product;

[0045] The reaction results of the 100U / L GALT quality control product were substituted into the standard curve fitted by the 50U / L quality control product, and the detection value was calculated to be 99.4 U / L, with a relative deviation of 0.6%, which is within the allowable error range of the quality control product.

[0046] Based on the above results, it is believed that the overlap of the standard curves fitted by two quality control products among the three standard curves is within the allowable range. It is believed that the two quality control products are in normal condition and the original standard curve in the detection system is invalid. One of the standard curves fitted by the two quality control products is used to replace the original standard curve in the detection system.

[0047] Example 2

[0048] Alpha-fetoprotein (AFP) was detected using acridinium ester direct chemiluminescence. A mathematical model was created for the reaction of multiple AFP control samples (50-500 ng / mL) with 14.1-42.3 mmol / L H₂O₂ (see Figure 7). The label was an acridinium ester, and the luminescent substrates were NaOH and H₂O₂.

[0049] A 100 ng / mL AFP quality control product was reacted once with 42.3 mmol / L H2O2, and a standard curve was fitted using the pre-made reaction curve mathematical model (see Figure 8).

[0050] The result of the original standard curve in the detection system (see Figure 9) for detecting 100 ng / mL AFP quality control product was 96.26 ng / mL, with a relative deviation of 3.74%, which is within the maximum allowable error range of the quality control product. It is considered that the degree of overlap between the original standard curve in the detection system and the fitted standard curve of the quality control product is within the allowable range, that is, the quality control is qualified, and the original standard curve in the detection system continues to be used.

[0051] Example 3

[0052] Glucose (Glu) quality control samples were tested using the Trinder reaction. Pre-fabricated mathematical models were generated for multiple reaction curves of 0.08-0.15 mmol / L glucose quality control samples and 2-6 U / mL glucose oxidase (GOD) (see Figure 10).

[0053] A 0.1 mmol / L glucose quality control product was reacted with 4 U / mL glucose oxidase once, and a standard curve for glucose detection was fitted through mathematical model calculations of multiple reaction curves (see Figure 11).

[0054] Substituting the reaction result of the 0.1mmol / L glucose quality control product into the original standard curve of the detection system (see Figure 12), the calculated detection value was 0.095mmol / L, with a relative deviation of 5.46%, which exceeded the allowable range and was judged to be unqualified quality control.

[0055] Another concentration of glucose control product (0.15 mmol / L) was selected to react once with 4 U / mL glucose oxidase, and a standard curve for glucose detection was fitted through mathematical model calculation of multiple reaction curves (see Figure 13).

[0056] Compare the degree of coincidence between the standard curves fitted by the two concentrations of quality control products and the original standard curve in the detection system.

[0057] Substituting the reaction result of the 0.15mmol / L glucose quality control product into the original standard curve of the detection system, the calculated detection value was 0.149mmol / L, with a relative deviation of 0.48%, which is within the maximum allowable error range of the quality control product;

[0058] The reaction result of 0.1mmol / L glucose quality control was substituted into the standard curve fitted by 0.15mmol / L glucose quality control and the calculated detection value was 0.095mmol / L, with a relative deviation of 5.47%, which exceeded the allowable range.

[0059] The reaction result of 0.15mmol / L glucose quality control was substituted into the standard curve fitted by 0.1mmol / L glucose quality control. The calculated detection value was 0.159mmol / L, with a relative deviation of 5.7%, which exceeded the allowable range.

[0060] In summary, the degree of coincidence between the 0.15mmol / L quality control product fitting standard curve and the original standard curve in the detection system is within the allowable range among the three standard curves. It is considered that the 0.1mmol / L quality control product is in an abnormal state, and the original standard curve is retained for continued use.

Claims

1. A quality control method for medical testing and analysis, characterized by comprising the following steps: Step 1: react once with a quality control product of any concentration and fit a standard curve; Step 2: compare the standard curve fitted in Step 1 with the standard curve currently being used in the detection system. If the degree of overlap is within the allowable range, it means the quality control is qualified, and the standard curve currently being used by the detection system is retained and continued to be used; Step 3: if the degree of overlap exceeds the allowable range, it means the quality control is unqualified, and the quality control products of different concentrations in Step 1 are reacted once and the standard curve is fitted; Step 4: compare the degree of overlap between the standard curve fitted in Step 1 and the standard curve fitted in Step 3 and the standard curve currently being used by the detection system; Step 5: if the degree of overlap between the standard curve fitted in Step 1 and the standard curve fitted in Step 3 is within the allowable range, it means the quality control product is in normal condition, and both standard curves can replace the standard curve currently being used in the original detection system; Step 6: if the degree of overlap between the standard curve fitted in Step 3 and the standard curve currently being used by the detection system is within the allowable range, it means the quality control product is in abnormal condition, and the standard curve currently being used by the detection system is retained and continued to be used.

2. The quality control method for medical examination and analysis according to claim 1, characterized in that: The medical test analysis is biochemical analysis and immunoassay.

3. The quality control method for medical examination and analysis according to claim 1, characterized in that: The fitting standard curve is carried out in the same reaction system.

4. The quality control method for medical examination and analysis according to claim 1, characterized in that: The comparison of standard curve coincidence refers to comparing the difference in slope and intercept of different standard curves, or comparing the difference in concentration of the same quality control product in different standard curves.

5. The quality control method for medical examination and analysis according to claim 1, characterized in that: The allowable range of coincidence is the maximum error allowed for the quality control product.

Images