The effect of bone marrow aspiration strategy on the yield and quality of human mesenchymal stem cells

Abstract

Introduction: Large inter-donor differences exist in human mesenchymal stem cell (hMSC) yield and the response of these cells to osteogenic stimuli. The source of these differences may be clinical differences in stem cell characteristics between individuals or the aspiration procedure itself.

Methods: From a total of 23 donors, we aimed to take 2 consecutive 10-mL aspirates from the same site in 17 donors and in 6 donors we aimed to take a 5-mL and a 20-mL aspirate. The aspiration was stopped either when the syringe was full or when no more bone marrow came through. Mononuclear cell yield (MNC), MSC yield, and differentiation capacity were analyzed for intra-donor and inter-donor variation. We analyzed the effect of the dilution with peripheral blood by drawing 20 mL at once.

Results: There was a high correlation between the first and second aspiration volumes, and aspirates with a volume of less than 8 mL showed a large variation in cellular yield. The second 10-mL aspirate, and also 20-mL aspirates, contained a lower concentration of nucleated cells and yielded lower numbers of mesenchymal stem cells. No effect of the aspiration procedure on the biological characteristics of the mesenchymal stem cells was seen.

Conclusion: We recommend collection volumes of bone marrow aspirates of at least 8 mL to reduce the risk of obtaining aspirates with low cell numbers. From the same site, a second aspiration or an aspirate of > 10 mL can be drawn without any loss of biological quality due to dilution with peripheral blood.

Figure 1.

Correlation between the first and second aspiration volumes. Scatter plot of the first and second aspiration volumes. The correlation coefficient (Spearman's rho) is extremely high at 78% (CI: 48–92) but it should be noted that the correlation around 10 mL is artificial because the volume of the syringe was 10 mL.

Figure 2.

The effect of aspiration volume on the concentration of nucleated cells. Scatter plot of the aspiration volume against the concentration of nucleated cells. With higher aspiration volumes (> 8 mL), the concentration does not appear to increase. The concentrations seem to level off around the mean in the higher aspiration volumes. The mean concentration is 2.3 × 10⁷ (geometric mean). The 4 highest values are not within the 95% confidence interval.

Figure 3.

Concentration of nucleated cells per aspiration. Box plots showing the concentration of nucleated cells per mL in the first and the second aspirate with their range, first and third quartile, and mean. The second 10-mL aspirate contained 26% fewer cells on average than the first aspirate. Outlier depicted separately (°).

Figure 4.

Larger aspiration volumes are associated with reduced concentrations of nucleated cells. Box plots showing the concentration of nucleated cells per mL for the 5-mL and 20-mL aspirates with their range, first and third quartile, and mean. There was a statistically significant decrease in concentration of MNCs in the 20-mL aspirate.

Figure 5.

Mesenchymal stem cells from the first and second aspirates. Box plots showing range, first and third quartile, and median for the number of mesenchymal cells derived from the first and second aspirates. Outliers are depicted separately (* and o). After natural logarithmic transformation due to positively skewed data, there was no significant difference between the numbers of mesenchymal cells in the first and second aspirates.

Figure 6.

Proliferation rates of the first and second aspiration. Mean proliferation rates for the first and second aspirates. Almost no error was observed and there were only small differences between aspirates. These differences were not significantly different from each other.

Figure 7.

ALP induction per donor. The y-axis shows the percentage of ALP-positive cells gated in flow cytometric analysis. Box plots show the mean ALP levels for triplicate measurements from all donors combined for each modality. The range, first and third quartile, and the median are shown. Outliers are depicted separately (o). One-way ANOVA followed by Tukey's post hoc test showed no significant differences between the ALP levels of the first and second aspirates.