2014/12/19 10:09:30 發表

The Evaluation of The Application of AVE-763B

The Evaluation of The Application of AVE-763B

Hu Tieshi.

Clinical laboratory, the people’s hospital of Liaoning Province.

AVE-763B can enlarge urine formed elements in the flow counting cell through a microscope by using machine vision technology and input the images taken by CCD to the computer for automatic recognition, which is combined with manual judgment on suspicious cells. The various formed elements in urine can be identified and counted to obtain the quantity of each in unit volume. For quantitative analysis, there is no need to centrifuge the samples and the formed elements can be identified, classified and counted automatically. To test the credibility, we studied the precision, linearity, cross contamination, and accuracy of AVE-763B.

1. Material and Methods

1.1 Materials

1.1.1 Specimens

Fresh urine taken from outpatients of our hospital and healthy persons.

1.1.2 Instrument and reagents

AVE-763B urine formed element analyzer (produced by AVE Science ＆ Technology Co. Ltd.) and assorted reagents.

1.1.3 Test methods

Gather fresh urine samples and test in 2 hours.

1.2 Evaluation

1.2.1 Precision

Select 3 specimens, high (WBC2006/μl、RBC1206/μl、EP82/μl), middle (WBC364/μl、RBC210/μl、EP41/μl) and low value (WBC32/μl、RBC16/μl、EP8/μl) for each, mix them separately and test 20 times each. Calculate the CV.

1.2.2 Linearity

Choose 2 specimens, one containing WBC 1106/μl and another RBC 2168/μl, dilute them with the supernatant (obtained by centrifugation of normal urine for 10 minutes by 3000rpm) as 1:2, 1:4, 1:10, 1:20, 1:50 and 1:100. Test the diluted specimens and take down the values.

1.2.3 Cross contamination

Select specimens of high value and low value each for three and divide them into three groups randomly. Test the high value of each group for 3 times and record the value as H1, H2 and H3; then test the low value of each group for 3 times and record the value as L1, L2 and L3. Calculate the value of RBC, WBC, and EP cross contamination.

1.2.4 Accuracy and contrast tests

Select 30 specimens randomly and divide each sample into 2 after mixing perfectly, then test one with AVE-763B and another on the counting cell manually. Compare the results.

1.2.5 Negative coincidence

Centrifuge the specimens as recommend by NCCLS, then select 30 negative cases obtained from manual counting, calculate the negative coincidence.

1.2.6 Slight positive detection

Select a specimen containing RBC 500/μl, dilute it in 1:100 with the supernatant (obtained by centrifugation of normal urine for 10 minutes by 3000rpm). Test the diluted specimen 50 times and take down the values.

2. Results

2.1 Within batch precision of the high, middle and low values tested repeatedly for 20 times. Refer to Table 1.

                                                  Table 1 within batch precision evaluation

2.2 Linearity: make regression analysis between the testing value and theoretical value of the samples after dilution. Refer to Table 2.

                                                    Table 2 linearity evaluation 

2.3 Cross contamination：see Table 3 for cross contamination calculation。

                                              Table 3 cross contamination evaluation

2.4 see Table 4 for accuracy and contrast tests evaluation.

                                         Table 4 accuracy and contrast tests evaluation  

2.5 Negative coincidence

30 samples tested manually to be negative were all proved the same with AVE-763B. The negative coincidence was 100%.

2.6 Weak positive detection

The diluted specimen (concentration was about 5 /μl) was tested by AVE-763B for 50 times repeatedly and the results were all positive (See Table 5). The positive detection was 100%.

                                                      Table 5 slight positive detection evaluation

3. Discussions

The perfect reproducibility of AVE-763B through high, middle and low value reproducible tests showed that the test conditions of instrument were constant, the influencing factors were small, and the formed elements distributed in the counting cell evenly. The linear range of WBC and RBC were wide, fresh urine could be tested directly after mixing without dilution or condensation. A cross contamination less than 2% showed the high value sample had little effect on low value sample, and the automatic washing of the instrument was reliable and effective. The compare on 3 cases between instrumental and manual detections showed a good correlation and comparability with both negative coincidence and weak positive detection of 100%. No false positive results appeared.

The automatic recognition of AVE-763B can be affected a lot when the sample is of high concentration and causing cell adhesion. So when a report is verified and if there is some element which cannot be recognized by the instrument or be misjudged, manual recognition can be taken to solve the accuracy problem according to the instrument instruction. The parameter settings need to be improved gradually in practice to further improve the automatic recognition ability of AVE-763B. But as for the specimens with too many kinds of formed elements such as crystals, mucus, etc., the effectiveness has yet to be confirmed. The specimens had better to be mixed well manually before operated on the instrument and the test tube had better be U-shaped for mixing perfectly and for better reproducibility.

AVE-763B urine formed element analyzer is a reliable and highly automated instrument with a high testing speed. Since the testing sample needs no centrifugation and the report form is standard, it has shown a rather high clinical application value.

回復（）