$$\rightleftharpoonup{xx}$$
$$\longleftharp{xx}$$,
$$\longrightharp{xx}$$,
Average preparation time
The platelet screening process using a DLS system is summarized in Figure 1. Platelets are tested with the DLS system at the time of receipt from the blood supplier. As detailed in Table 1 , the average preparation time for trained users is 2 min 23 s while the clean-up and post-test work times are 14 s and 46 s, respectively. In total, the average user takes 3 min and 23 s of hands-on time per test.
| Activity | Active time | Walk-away testing time | TOTAL |
| Prepare DLS System | 14 s | | |
| Assemble sampling tool | 28 s | | |
| Obtain segment | 52 s | | |
| Fill capillary and start test | 49 s | | |
| | 5 min | |
| Clean-up | 14 s | | |
| Tag and inventory platelet bag | 46 s | | |
| | | |
| Total time needed per sample | 3 min 23 s | 5 min | 8 min 23 s |
Table 1: Active testing time breakdown. The test itself is a walk away test with an average duration of 5 min. The average user takes 3 min 23 s to prepare the DLS system for a test, obtain and test a sample following this protocol, and tag the platelet bag.
Precision
The precision of the DLS system was assessed at three microparticle levels, 0-7%, 12-25% and 28-75%. The clinically relevant range of %MP is 3 - 75%. Two operators tested low, medium, and high control samples for 16 operating days on two DLS systems in parallel. Samples were tested in duplicate, but in random order on each testing day.
Table 2 summarizes the within-device precision of the DLS measurements for Percent Microparticles (%MP) relative to platelets.
| Microparticle Content |
| Low | Medium | High |
| Mean %MP (%) | 4.4 | 19.5 | 53.8 |
| Standard deviation (%) | 1.8 | 2.6 | 5.8 |
| CV (%) | 40.4 | 13.2 | 10.6 |
Table 2: Within-device precision of Percent Microparticles (%MP). At very low microparticle content small platelets may contribute to %MP resulting in increased variability for low microparticle content samples.
Table 3 shows the within-device precision of the DLS measurements for average microparticle radius between 50-550 nm.
| Microparticle Content |
| Low | Medium | High |
| Mean Radius (nm) | 331 | 161 | 188 |
| Standard deviation (mm) | 133 | 41 | 25 |
| CV (%) | 40.1 | 25.2 | 13.5 |
Table 3: Within-device precision of microparticle radius. At very low microparticle content small platelets may contribute to Percent Microparticles (%MP) resulting in increased variability for low microparticle content samples.
Table 4 shows the reproducibility of DLS measurements for Percent Microparticles (%MP).
| Microparticle Content |
| Low | Medium | High |
| Mean %MP (%) | 4.4 | 29.2 | 53.6 |
| Reproducibility (%) | 1.5 | 2.3 | 5 |
| CV (%) | 35 | 11.8 | 9.4 |
Table 4: Reproducibility of Dynamic Light Scattering (DLS) measurements for Percent Microparticles (%MP).
Linearity
Figure 2 shows that DLS results are linear, i.e., fit a straight line with respect to the assigned values of the samples. Seven samples were prepared with differing microparticle content. Samples with high (MP7) and low (MP1) microparticle content and matching platelet concentration were prepared and mixed at different ratios to create intermediate samples (MPx). Samples were tested with flow cytometry as described previously19 and DLS and %MP results at each concentration were plotted as input and output. The coefficient of determination was found to be 0.985. Examples of DLS histograms for low and high microparticle content are shown in Figure 3A. The DLS results were confirmed by flow cytometry (Figure 3B).

Figure 2: Comparison of flow cytometry and DLS results. Linear relationship between %MP determined by Flow Cytometry (input) and DLS (output), the original DLS data from the two samples marked by the open symbol ○ are shown in Figure 3. Please click here to view a larger version of this figure.

Figure 3: Microparticle content to differentiate between activated and non-activated platelets. (A) DLS results of MP content in activated platelets (dashed line) was 57% compared to 4% in non-activated platelets (solid line). Tests were performed at a measurement temperature of 37 °C, plasma viscosity setting of 1.06 x10-3 Pa·s, and total intensity settings between 200-600 kHz. (B) Flow cytometry results (obtained as described previously19) of the same samples as shown in (A); in the forward scatter histograms P1 and P2 represent the MP and platelet gates, respectively; for activated platelets (left) 76% of events fell into the MP gate compared to 6% for non-activated platelets (right). The linear regression line suggests a constant relative contribution of background noise to %MP by flow cytometry leading to the consistently higher results. Please click here to view a larger version of this figure.
Specificity/Interference
The International Organization for Standardization (ISO) defines analytical specificity as the ability of a measurement procedure to detect or measure only the measurand while there are other quantities present in the sample. The analytical specificity of microparticle screening might be affected by red blood cells (RBC). DLS measures MP content relative to platelet content. RBC might interfere because the scattering contribution of RBC is included in the scattering contribution of platelets, reducing the relative contribution of MP. Regulatory limits for the allowable RBC content in blood products exist; a conservative conversion of the threshold recommended by AABB for allowable RBC concentration in platelet concentrates-2 mL of packed RBC in one unit of platelets-resulted in 8.0 x1010 cells/L (assumptions: RBC volume is 8.5 x10-14L, hematocrit of packed RBC is 68%, volume of platelet unit is 200 mL). The reported residual RBC concentrations in different products are well below this threshold46.
Three different donors donated platelets and red blood cells (RBC) on two different days, as eligible, for three independent experiments. The initial RBC content in the platelet concentrates (reference sample) was 0.05 - 0.15 x109 cells/L as determined with a hemocytometer. Five additional samples were created by spiking-in known quantities of RBC into aliquots of the platelet concentrate; target RBC levels in these samples were 1.0, 5.0, 10, 40 and 80 x109 cells/L.
The interference threshold of red blood cells was approximately 1.0 x1010 cells/L (Figure 4), which also correlated with the level at which the presence of red blood cells was visually evident (Figure 5). Above this level, %MP was underestimated which means that in visually red samples-which contain RBC rather than hemoglobin-the reported microparticle content will be too low.

Figure 4: Linear relationship between %MP (DLS) and RBC Concentration (cells/L). Increasing RBC concentrations of 0.1 x109, 1.0 x109, 5.0 x109, 1.0 x1010, 4.0 x1010, and 8.0 x1010 cells/L (from left to right) from 3 independent experiments (○ experiment 1, ● experiment 2, □ experiment 3, linear regression lines are shown for each experiment). Above 1.0 x1010 RBC/L leads to underestimation of %MP. Visual appearance of RBC containing samples is shown in Figure 5. Please click here to view a larger version of this figure.

Figure 5: Visual appearance of RBC containing samples. Redness of platelet samples containing RBC concentrations of 0.1 x109, 1.0 x109, 5.0 x109, 1.0 x1010, 4.0 x1010, and 8.0 x1010 cells/L from left to right. Please click here to view a larger version of this figure.
Accuracy
Accuracy is defined as the difference between a single measurement result and a true quantity value assigned to the sample. This measurement error includes a systematic component estimated by measurement bias and a random component estimated by a standard deviation. Thus, the accuracy of a measurement result is a combination of trueness and precision.
A bead standard was used to determine the accuracy of the DLS test. Accuracy was evaluated at two concentrations within the clinically relevant range of the assay of 3 - 75% MP. Reference bead mixtures were used to prepare samples of known concentrations because reference beads have a known size and concentration. Consequently, reference beads can be mixed to obtain samples with desired particle sizes and concentrations.
Standard polystyrene beads with a 125 nm radius were used to represent microparticles and beads with 1.5 µm radius were used to represent platelets. The accuracy of the microparticle assay was assessed with bead mixtures of approximately 20% and 50% MP content. The accuracy of the measured particle radii was compared to the particle radii documented on the Certificates of Analysis for the reference beads as shown in Table 5.
| 125 nm beads | 1.5 µm beads |
| Certificate of Analysis | | Certificate of Analysis | |
| 20% MP | 50% MP | | 20% MP | 50% MP |
| Mean Radius | 122.0 | 113.8 | 124.4 | 1.5 | 1.6 | 1.60 |
| Std Dev | 4.5 | 2.83 | 3.6 | 0.035 | 0.06 | 0.07 |
| CV | 3.60% | 2.48% | 2.89% | 2.20% | 3.74% | 4.05% |
| Accuracy | | 6.70% | 2.00% | | 6.50% | 6.30% |
Table 5: Accuracy of Dynamic Light Scattering (DLS) measurements. Size comparison for DLS results against certificate of analysis for beads with 125 nm radius and 1.5 µm radius in mixtures.