Methodological aspects of oxygen gradient ektacytometry in sickle cell disease: Effects of sample storage on outcome parameters in distinct patient subgroups

1 Introduction

Sickle cell disease (SCD) is caused by a point mutation in the β-globin gene. The two main forms are homozygous sickle cell anemia (HbSS) and compound heterozygous hemoglobin SC disease (HbSC). Red blood cells (RBCs) from patients with HbSS contain abnormal hemoglobin S (HbS), which polymerizes under deoxygenation. HbS polymerization causes a mechanical distortion of RBCs, which results in a crescent-like or sickle shape. Such RBCs are fragile and poorly deformable. RBCs from patients with HbSC contain equivalent amounts (40–45%) of both HbS and hemoglobin C. HbC has the particularity to form crystals under oxygenated, as well as deoxygenated, conditions [4, 6]. In addition, HbC promotes RBC dehydration through activation of the KCl cotransporter [5]. RBC rheological properties of HbSS and HbSC patients and associations with clinical severity have been studied. Both SCD genotypes have reduced RBC deformability at low and high shear stresses compared to healthy individuals [10]. RBC deformability at low shear stress has been found to be similar for the two different genotypes but HbSS patients have a lower RBC deformability at high shear stress than HbSC patients [1, 3]. However, all these RBC rheological studies have been performed in normoxic conditions, while RBC rheological properties would likely be sensitive to oxygen levels as HbS containing RBCs can sickle under deoxygenation thereby becoming more rigid and less deformable. Recently a method has been developed to assess RBC deformability as a function of oxygen tension (pO₂) [9]. This method, called oxygen gradient ektacytometry (oxygenscan), is currently under development for its use in the field of SCD to test the efficacy of (new) targeted and curative therapies as well as to aid in classifying clinical severity [8]. Given the potential use of oxygen gradient ektacytometry in patient care, methodological standardization is strongly needed. It is essential that we identify pre-analytical factors that could influence outcome parameters, and develop strategies and methodology to allow comparison of results between different laboratories and devices. We recently showed that several parameters of oxygen gradient ektacytometry are sensitive to the number of RBCs used for the analysis, as well as to device settings, such as the speed of deoxygenation applied to the RBC suspension [7]. In this previous study we also tested the effect of blood storage on oxygen gradient ektacytometry-derived parameters, showing minimal changes in key oxygenscan parameters. This study was limited by the heterogeneity and relatively small numbers of SCD patients included; 8 HbSS patients, 20 transfused HbSS patients and 2 HbSC patients [7]. RBCs from each of these samples may behave differently under deoxygenation and reoxygenation, and blood storage could have a different impact on each subgroup. Indeed, the study compared the oxygenscan responses during blood storage between distinct populations of SCD patients.

Blood from 40 non-transfused HbSS patients (HbS: 80.0±9.7%), 16 chronically transfused HbSS patients (HbS: 57.0±14.0%), and 14 HbSC patients (HbS: 49.2±4.2%, HbC: 42.3±3.0%) at steady-state was collected in EDTA tubes. The study was conducted in the Sickle Cell Centre in Lyon, in accordance with the guidelines set by the Declaration of Helsinki and approved by the Regional Ethics Committees (L14-127). Measurements were performed on fresh samples (i.e. within 4 hours after collection, samples maintained at room temperature) and after 24 hours of storage (4°C). Stored samples were adjusted to room temperature before analyses. The Laser Optical Rotational Red Cell Analyzer (Lorrca, RR Mechatronics, Zwaag, The Netherlands) with oxygenscan module was used. This ektacytometer determines RBC deformability (expressed as Elongation Index, EI) as a function of continuously changing oxygen tension. A volume of blood ranging between 20 and 30μL, standardized to a fixed RBC count of 200x10⁶, was mixed with 5 mL of high viscous (±30 cP, pH 7.37–7.45, osmolarity 282–286 mOsm/kg at 22°C) Oxy-Iso polyvinylpyrrolidone (PVP) suspension and subjected to a shear stress of 30 Pa at 37°C. The oxygen partial pressure (pO₂) was gradually decreased from 160 mmHg to 20 mmHg (deoxygenation), and then allowed to return to normoxic values as recommended [7]. The diffraction pattern was analyzed by the computer and several parameters were derived [9]: 1) EImax; the RBC deformability at normoxia; 2) EImin; the lowest RBC deformability reached upon deoxygenation; 3) Point of Sickling (PoS); the pO₂ at which RBC deformability decreases below 5%of EImax during deoxygenation; 4) DeltaEI; the difference between EImin and EImax; 5) Recovery; the percentage of EImax recovered during reoxygenation.

A Wilcoxon T test was performed to investigate the effect of storage. Correlations between the percentage of HbS or HbC and the magnitude of changes in oxygenscan parameters were performed using Spearman coefficient. A p-value <0.05 was considered significant.

Main results are presented in Fig. 1. Our results show that analysis after 24 hours of blood storage compared to measurements on fresh samples significantly increased EImin in non-transfused HbSS and HbSC patients, and EImax in all three groups. Recovery was also significantly affected by blood storage in HbSS patients, although the magnitude of change remained rather low (2±5.6%). PoS and DeltaEI were not significantly affected by blood storage. No correlation was observed between the magnitude of changes in the oxygenscan parameters affected by blood storage and the percentage of HbS or HbC, independently of the group of patients considered.

OPEN IN VIEWER

Fig. 1

Oxygenscan parameters measured on fresh blood (<4 hours after sampling) and after blood storage (24 hrs at 4°C) in HbSS, transfused HbSS and HbSC patients. Data are represented as medians, quartiles and range. Significant difference: ^*p < 0.05; ^**p <0.01; ^***p < 0.001.

The present study clearly shows that blood storage affects several oxygen gradient ektacytometry-derived parameters, in particular EImin and EImax. The largest effect was noted in untransfused HbSS and HbSC patients. Despite the fact that the magnitudes of changes were not related to the percentage of HbS or HbC, we observed that RBCs from transfused HbSS individuals were less affected by blood storage. This finding can be explained by the fact that blood samples from transfused HbSS patients contain a considerable number of normal RBCs, which are less sensitive to sample storage. Since HbS and HbC are highly sensitive to oxygen levels and metabolic changes, one would have expected to observe a decrease of EImin and EImax with blood storage. However, RBCs from SCD patients are more fragile than normal RBCs [2], and one could hypothesize that the slightly better mean RBC deformability observed 24 hrs after blood sampling would be the consequence of the lysis of the most fragile and rigid RBCs in the whole suspension. Parameters reflecting the dynamic behavior of sickle RBCs upon deoxygenation and reoxygenation, including the point of sickling were not affected by blood storage, suggesting that the extent of sickling/unsickling is maintained, even after 24 hours of storage. This is important because several laboratories are not able to perform measurements within 4 hours due to shipping time.

Collectively, since the time between blood sampling and analysis affects some key oxygen gradient ektacytometry-derived parameters, in particular in non-transfused SCD patients, we recommend that each individual laboratory performs oxygenscan measurements at a standardized time point.