Washing in hypotonic saline reduces the fraction of irreversibly-damaged cells in stored blood: a proof-of-concept study

Deep ensemble learning enables highly accurate classification of stored red blood cell morphology

Changes in red blood cell (RBC) morphology distribution have emerged as a quantitative biomarker for the degradation of RBC functional properties during hypothermic storage. Previously published automated methods for classifying the morphology of stored RBCs often had insufficient accuracy and relied on proprietary code and datasets, making them difficult to use in many research and clinical applications. Here we describe the development and validation of a highly accurate open-source RBC morphology classification pipeline based on ensemble deep learning (DL). The DL-enabled pipeline utilized adaptive thresholding or semantic segmentation for RBC identification, a deep ensemble of four convolutional neural networks (CNNs) to classify RBC morphology, and Kalman filtering with Hungarian assignment for tracking changes in the morphology of individual RBCs over time. The ensembled CNNs were trained and evaluated on thousands of individual RBCs from two open-access datasets previously collected to quantify the morphological heterogeneity and washing-induced shape recovery of stored RBCs. Confusion matrices and reliability diagrams demonstrated under-confidence of the constituent models and an accuracy of about 98% for the deep ensemble. Such a high accuracy allowed the CNN ensemble to uncover new insights over our previously published studies. Re-analysis of the datasets yielded much more accurate distributions of the effective diameters of stored RBCs at each stage of morphological degradation (discocyte: 7.821 ± 0.429 µm, echinocyte 1: 7.800 ± 0.581 µm, echinocyte 2: 7.304 ± 0.567 µm, echinocyte 3: 6.433 ± 0.490 µm, sphero-echinocyte: 5.963 ± 0.348 µm, spherocyte: 5.904 ± 0.292 µm, stomatocyte: 7.080 ± 0.522 µm). The effective diameter distributions were significantly different across all morphologies, with considerable effect sizes for non-neighboring classes. A combination of morphology classification with cell tracking enabled the discovery of a relatively rare and previously overlooked shape recovery of some sphero-echinocytes to early-stage echinocytes after washing with 1% human serum albumin solution. Finally, the datasets and code have been made freely available online to enable replication, further improvement, and adaptation of our work for other applications.

Erythrocyte interaction with titanium nanostructured surfaces

Titanium and its alloys are used to make different medical devices such as stents, artificial heart valves, and catheters for cardiovascular diseases due to their superior biocompatibility. Thrombus formation begins on the surface of these devices as soon as they encounter blood. This leads to the formation of blood clots, which obstructs the flow of blood that leads to severe complications. Recent advancements in nanoscale fabrication and superhydrophobic surface modification techniques have demonstrated that these surfaces have antiadhesive properties and the ability to reduce thrombosis. In this study, the interaction of erythrocytes and whole blood clotting kinetics on superhydrophobic titanium nanostructured surfaces was investigated. These surfaces were characterized for their wettability (contact angle), surface morphology and topography (scanning electron microscopy (SEM)), and crystallinity (glancing angled X-ray diffraction (GAXRD)). Erythrocyte morphology on different surfaces was characterized using SEM, and overall cell viability was demonstrated through fluorescence microscopy. The hemocompatibility of these surfaces was characterized using commercially available assays: thrombin generation assay thrombin generation, hemolytic assay hemolysis, and complement convertase assay complement activity. The results indicate that superhydrophobic titanium nanostructured surfaces had lower erythrocyte adhesion, less morphological changes in adhered cells, lower thrombin generation, lower complement activation, and were less cytotoxic compared to control surfaces. Thus, superhydrophobic titanium nanostructured surfaces may be a promising approach to prevent thrombosis for several medical devices.

Centrifugation-free washing: A novel approach for removing immunoglobulin A from stored red blood cells

Washed red blood cells (RBCs) are indicated for immunoglobulin A (IgA) deficient recipients. Centrifugation-based cell processors commonly used by hospital blood banks cannot consistently reduce IgA below the recommended levels, hence double washing is frequently required. Here, we describe a prototype of a simple, portable, disposable system capable of washing stored RBCs without centrifugation, while reducing IgA below 0.05 mg/dL in a single run. Samples from RBC units (n = 8, leukoreduced, 4-6 weeks storage duration) were diluted with normal saline to a hematocrit of 10%, and then washed using either the prototype washing system, or via conventional centrifugation. The efficiency of the two washing methods was quantified and compared by measuring several key in vitro quality metrics. The prototype of the washing system was able to process stored RBCs at a rate of 300 mL/hour, producing a suspension of washed RBCs with 43 ± 3% hematocrit and 86 ± 7% cell recovery. Overall, the two washing methods performed similarly for most measured parameters, lowering the concentration of free hemoglobin by >4-fold and total free protein by >10-fold. Importantly, the new washing system reduced the IgA level to 0.02 ± 0.01 mg/mL, a concentration 5-fold lower than that produced by conventional centrifugation. This proof-of-concept study showed that centrifugation may be unnecessary for washing stored RBCs. A simple, disposable, centrifugation-free washing system could be particularly useful in smaller medical facilities and resource limited settings that may lack access to centrifugation-based cell processors.

Fluorescence Exclusion: A Simple Method to Assess Projected Surface, Volume and Morphology of Red Blood Cells Stored in Blood Bank

Red blood cells (RBC) ability to circulate is closely related to their surface area-to-volume ratio. A decrease in this ratio induces a decrease in RBC deformability that can lead to their retention and elimination in the spleen. We recently showed that a subpopulation of “small RBC” with reduced projected surface area accumulated upon storage in blood bank concentrates, but data on the volume of these altered RBC are lacking. So far, single cell measurement of RBC volume has remained a challenging task achieved by a few sophisticated methods some being subject to potential artifacts. We aimed to develop a reproducible and ergonomic method to assess simultaneously RBC volume and morphology at the single cell level. We adapted the fluorescence exclusion measurement of volume in nucleated cells to the measurement of RBC volume. This method requires no pre-treatment of the cell and can be performed in physiological or experimental buffer. In addition to RBC volume assessment, brightfield images enabling a precise definition of the morphology and the measurement of projected surface area can be generated simultaneously. We first verified that fluorescence exclusion is precise, reproducible and can quantify volume modifications following morphological changes induced by heating or incubation in non-physiological medium. We then used the method to characterize RBC stored for 42 days in SAG-M in blood bank conditions. Simultaneous determination of the volume, projected surface area and morphology allowed to evaluate the surface area-to-volume ratio of individual RBC upon storage. We observed a similar surface area-to-volume ratio in discocytes (D) and echinocytes I (EI), which decreased in EII (7%) and EIII (24%), sphero-echinocytes (SE; 41%) and spherocytes (S; 47%). If RBC dimensions determine indeed the ability of RBC to cross the spleen, these modifications are expected to induce the rapid splenic entrapment of the most morphologically altered RBC (EIII, SE, and S) and further support the hypothesis of a rapid clearance of the “small RBC” subpopulation by the spleen following transfusion.