Predictors of cord blood unit cell content in a volume unrestricted large series collections: a chance for a fast and cheap multiparameter selection model

Abstract

Background: Cord blood plays a very important role in stem cell transplantation and therapy with an emerging implication also in regenerative medicine. The number of cells available in a single cord blood unit (CBU), in particular, the CD34+ and total nucleated cell (TNC) content influences the transplantation clinical outcome. We analysed a very large series of CBUs, collected for private banking without any specific volume restriction, to deeply investigate the best predictors of cord blood stem cells content.

Methods: Maternal and neonatal clinic laboratory data of a total 2583 UCBs were obtained from the InScientiaFides cord blood bank based in Republic of San Marino. Univariate and multivariate analysis were conducted to better interpret the data and to build a predictive model to select, the CBU with high CD34+ content.

Results: Our univariate analysis shows that seasonality and the geographical area affects the quality of umbilical cord blood. Gestational age, babie's gender and birth weight have a positive correlation with CB TNC content. The babie's birth weight affects positively also CD34+ content and CBU volume while the cesarean delivery affect the CB volume only. Our predictive model, based on multivariate analysis, shows that male babie's, gestational age lower to 39 weeks, cesarean delivery and CBUs with a content of TNC higher than 3.44 × 10⁸ (group A) have a significant higher CD34+ content than group B (female babie's, gestational age higher than 39 weeks and vaginal delivery). The group A have a 37.5% of CBUs with a concentration of CD34+ > 2 × 10⁶, while no CBUs with high concentration of CD34+ were detect in group B.

Conclusion: This study, conducted on a very large series of CBUs without any specific volume constraint, highlighted the prenatal and maternal factors that significantly influence the quality of the CBU collected. Specifically, it highlights that volume is not the best predictor of CD34+ CBU content; for this reason it cannot be taken into consideration alone for the analysis of the collected samples. Our final aim is to identify relevant factors, immediately available, that help to choice UCB with high CD34+ cell content, especially in simultaneous deliveries.

Keywords: CD34+; Newborn; Predictive model; Prenatal and maternal factors; TNC; Transfusion; Umbilical cord blood; Volume.

Fig. 1

Correlation among the UCB characteristics (N = 2583). Significant positive correlation were observed between (A) CD34+ and TNC (r = 0.5368; p < 0.001); (B) volume and TNC (r = 0.5149; p < 0.001); (C) volume and CD34+ (r = 0.3175; p < 0.001); (D) Positive correlation observed among the TNC content and CD45 + (r = 0.8022; p < 0.001)

Fig. 2

Correlation among the UCB characteristics (N = 2583) and obstetrics and neonatal factors. Significant positive correlation were observed between (A) TNC and gestational age (r = 0.0213; p < 0.0001); birth weight and volume (B) (r = 0.1024; p < 0.0001); TNC (C) (r = 0.1152; p < 0.0001) and CD34+ (D) (r = 0.0772; p < 0.0001)

Fig. 3

Relation between babies gender and type of delivery with UCB volume and TNC. One-way ANOVA analysis (using Tukey Method post hoc test) showed that child gender affects TNC content (A) while birth modality affects the volume (B). Particularly females showed a higher TNC content than males (A) and the cesarean delivery (B) showed more abundant volumes compared to vaginal. All reported values are expressed as mean ± standard deviation (**p < 0.005; **** p < 0.0001)

Fig. 4

Correlation among the UCB characteristics and seasonal and geographical data. One-way ANOVA analysis (using Tukey Method post hoc test) showed that CBU volume collected in winter and autumn are more abundant than those collected in spring. No significant differences were observed between UCB collected in summer and those collected in other months (A). CBU collected in central-southern Italy showed higher volumes (B) and higher TNC/CD34+ content (C, D) than those of northern Italy. For CBU collected in italian islands we did not observe any significant difference as compared to the rest of the country in terms of volume and TNC/CD34+. All reported values are expressed as mean ± standard deviation (**p < 0.005; ***p < 0.001)