The Journal of ExtraCorporeal Technology Issue 55-4 | Page 21

T . Matsumoto et al .: J Extra Corpor Technol 2023 , 55 , 167 – 174 171

Table 2 . Simultaneous reproducibility with test liquids 1 – 7 in Experiment 1

Measurements were performed 10 times for each test liquid and each measurement method , and CVs (%) were calculated . For the COP measurements using the Test Method , all CVs were in the range of 1.44 ± 0.66 %. For measurements obtained using the colloid osmometer , all CVs were in the range of 0.99 % ± 0.34 %. Abbreviations : COP , colloid osmometer ; SD , standard deviation ; CV , coefficient of variation ; Hct , hematocrit value .

Correlation

COP as measured by the test method and the colloid osmometer demonstrated a strong correlation ( Pearson product-moment correlation coefficient R 2 = 0.997 ; p < 0.01 ). The formula for the regression line is as follows ( Figure 5 ):

y ¼ 0:5351251 þ 0:9813299 x : ð2Þ

Measurement of time

Although the standby time was 3 min , solvent transfer was completed in under 1 min for test liquid 8 . Solvent transfer time increased as the concentration of the test liquid increased . With test liquid 10 , moderate solvent migration appeared to occur after 1 min , but no solvent migration could be observed after 2 min had passed .

Discussion

Evaluation of measurement accuracy ( simultaneous reproducibility )

Simultaneous reproducibility was used to evaluate the accuracy of the measured values . Each measurement was within 3 %, which indicates high reproducibility ( Tables 2 and 3 ). The presence of little variability indicates that there may not be many factors contributing to variations in the measurements , even when the person obtaining the measurement changes . This suggests that this measurement method is highly reliable . The absolute value of the colloid osmometer measurement error is ± 5 mmHg . It has an error rate of 10 % in calculating albumin concentrations [ 19 ]. Because the systematic error rate of the test method was unknown , we could not accurately capture all the errors . However , the measurement result was similar to that of the colloid osmometer , which has a measurement error rate of

Figure 4 . Correlation between COP measurements from the colloid osmometer and by the test method used in Experiment 1 . n = 70 ; Pearson ’ s correlation : p < 0.01 ; R 2 = 0.998 . COP , colloid osmotic pressure . y = �0.815074 + 1.1085785 x .

less than 3 % CV . Thus , the systemic error rate of the test method is expected to be comparable to that of the colloid osmometer .

Over extended periods of use , the measurement accuracy may be impacted by the performance of the ultrafiltration membranes , which are not constant . The concentration polarization model can explain the filtration performance of semipermeable membranes ; substances that cannot pass through the membrane concentrate on the membrane surface increasing transmembrane resistance and decreasing filtration performance . Additionally , fouling is caused by the adhesion of proteins to the surface and interior of the membrane , which similarly increases the membrane passage resistance and reduces the permeability and solute permeability [ 20 – 22 ]. These factors will be affected by the filtered flow rate , blood flow rate , blood concentration , and elapsed time . During our experiment , the ultrafiltration membrane was loaded using variable rates of blood flow . In addition , the blood concentration and elapsed time were varied .