Hereditary spherocytosis (HS) is a heterogeneous group of inherited anemias that originates from defective anchoring of transmembrane proteins to the cytoskeletal network of the red blood cell (RBC). The defective anchoring is predominantly caused by a mutation in the genes coding for ankyrin (ANK1), α-spectrin (SPTA1), β-spectrin (SPTB), band-3 (SLC4A1), or protein 4.2 (EPB42) (Perrotta et al., 2008). These mutations compromise the vertical linkages between the lipid bilayer and the cytoskeletal network, leading to destabilization of the membrane, increased vesiculation and subsequent membrane loss. The progressive membrane loss leads to formation of dense spherical-shaped RBCs (spherocytes) with reduced deformability (Chasis et al., 1988; Eber and Lux, 2004; Huisjes et al., 2018).

The spleen plays an intricate role in the pathophysiology of HS. Normally, this organ functions as a quality control for RBCs. During the 120-day lifespan of healthy RBCs, membrane surface area, surface area-to-volume ratio, and deformability decrease because of release of essentially hemoglobin-free microvesicles. RBCs with increased density and reduced deformability are eventually trapped in the narrow endothelial slits of the spleen, leading to clearance of aged RBCs (Eber and Lux, 2004; Mebius and Kraal, 2005). The compromised vertical linkages in HS accelerate the loss of membrane and deformability, leading to premature destruction of RBCs in the spleen. Therefore, splenectomy is an effective treatment, and removal of the primary site of RBC destruction generally improves clinical symptoms (Musser et al., 1984; Eber and Lux, 2004; Perrotta et al., 2008). Nevertheless, the risks and benefits should be carefully assessed as splenectomy results in a permanently increased risk of infections caused by encapsulated bacteria and long term risk for cardiovascular events (Perrotta et al., 2008; Schilling et al., 2008).

The effects of splenectomy on RBC rheology and RBC related parameters in HS have been studied to a limited extent. It is known that splenectomy improves the RBC count, hemoglobin (Hb) levels, and hematocrit, and that it reduces mean corpuscular hemoglobin concentration (MCHC) and the percentage of reticulocytes (Reliene et al., 2002; Li et al., 2016; Zaninoni et al., 2018; Huisjes et al., 2020). On a cellular level it has been shown that the size of RBCs increases following splenectomy, and that microspherocytes can no longer be detected (Sugihara et al., 1984). However, splenectomy has little effect on correcting the cytoskeletal membrane defect (Reliene et al., 2002). More recent studies have shown that RBC deformability as measured by osmotic gradient ektacytometry was not improved after splenectomy (Zaninoni et al., 2018; Huisjes et al., 2020). An important limitation of these studies was the fact that they compared cohorts of splenectomized and non-splenectomized patients; longitudinal studies on the response to splenectomy of individual HS patients are scarce (Li et al., 2016).

In this study, we investigated individual responses to splenectomy in a group of five HS patients, with particular focus on RBC functional properties as determined by ektacytometry. Our results indicate that the Cell Membrane Stability Test (CMST), which measures the RBCs response to high shear stress, is able to detect substantial functional improvement of the RBC membrane after splenectomy, by showing a partly restored ability to shed membrane, a feature of healthy RBCs. We suggest that the CMST represents a novel biomarker of RBC membrane health in HS, and may be used to assess the efficacy of treatment.

In the present study, we report on the longitudinal effects of splenectomy in five HS patients. We specifically focused on cellular properties related to membrane health with the use of two different forms of ektacytometry: osmotic gradient ektacytometry and the CMST. In particular the CMST results revealed a novel feature of HS RBCs, i.e., the loss of the ability to shed membrane, and improvement of this in vitro cellular property following splenectomy. Membrane-shedding capacity in this test is assessed by the loss of deformability that occurs during prolonged exposure of RBCs to high shear stress. We suggest that improved membrane-shedding capacity after splenectomy reflects improved RBC membrane health, and ΔEI as measured by the CMST may thus serve as a novel clinically relevant biomarker.

The longitudinally observed increase in RBC count, Hb and reticulocyte count after splenectomy corresponds well with results from previous studies where splenectomized and non-splenectomized patient groups were compared (Zaninoni et al., 2018; Huisjes et al., 2020). In addition, our patients also showed a decrease in MCHC directly after splenectomy, which continued to decrease in the following months. This implicates that the internal viscosity or cellular density of HS RBCs is reduced after splenectomy, which is supported by the reduction in the percentage of hyperchromic cells. Little is known about the effect of splenectomy on in vivo RBC vesiculation in HS, an important pathophysiological feature, but previous studies demonstrated that RBC vesiculation caused an increase in internal viscosity (MCHC) through shedding of RBC-derived microvesicles (Bosch et al., 1994; Alaarg et al., 2013; Bosman, 2013). Hence, both the decrease in MCHC and hyperchromic cells could indicate that in vivo vesiculation of RBCs in HS is reduced after splenectomy. In turn this could explain the observed improvement in ΔEI in the CMST after splenectomy, which reflects improved ability to shed membrane in vitro.

More detailed analysis of the longitudinal effects of splenectomy was obtained by osmotic gradient ektacytometry. This technique is generally considered as the gold standard in the diagnosis of HS (DaCosta et al., 2016; Lazarova et al., 2017; Llaudet-Planas et al., 2018; Zaninoni et al., 2018), and its parameters EI_max, O_min, and O_hyper are considered biomarkers of, respectively, total membrane surface area, surface area to volume ratio, and RBC hydration status. Upon splenectomy most of these parameters were not affected, only O_hyper was significantly increased (Figures 2D, 4D). O_hyper and MCHC are known to have an inverse correlation with each other (Zaninoni et al., 2018). Both the increase in O_hyper and the decrease in MCHC indicate that splenectomy improves the hydration state/intracellular viscosity. In line with this, an increase in the AUC was observed in the longitudinal cohort, whereas AUC remained unchanged after splenectomy in the case-control cohort. Our findings partly contradict with previous studies were both O_hyper and AUC remained unaltered after splenectomy (Zaninoni et al., 2018; Huisjes et al., 2020). This could be explained by the non-longitudinal design of these latter studies in which individual differences in response to splenectomy become less apparent.

We next investigated the effect of splenectomy on the ability of HS RBCs to respond to mechanical stress. For this we used the CMST, an ektacytometry based test that was previously used to study membrane stability in HS by studying resealed RBC ghosts (Mohandas et al., 1982; Chasis and Mohandas, 1986). We demonstrate that RBCs from non-splenectomized HS patients show no or only a modest decrease in EI after prolonged exposure to shear stress, in contrast to HC RBCs which display a substantial decrease in EI under these conditions. We hypothesize that HS RBCs from non-splenectomized HS patients are more dense and rigid due to in vivo vesiculation that is accelerated by the spleen, and, therefore are less able to shed membrane in vitro (Waugh and La Celle, 1980). This is confirmed by the microscopic evaluation of HS RBCs during CMST measurement, which shows that, in contrast to healthy RBCs, HS RBCs do not change morphologically (Figure 3C). Further support for this hypothesis is obtained from the significant correlation of ΔEI and hyperchromic cells (Figure 3D).

Following splenectomy HS RBCs showed an increase in ΔEI (Figure 3B) and after more than 3 months ΔEI was not significantly different compared to HC. This suggests that after splenectomy HS RBCs have regained part of the ability to shed membrane in vitro, which may be related to a partly restored ability of HS RBCs for de novo synthesis of lipids (Cooper and Jandl, 1969; Sugihara et al., 1984; Takashi and Yoshihito, 1984). The increase in ΔEI could also indicate a change in RBC population due to the absence of the quality control function of the spleen; Instead of shedding micro vesicles in vivo in the spleen (i.e., splenic conditioning), this now occurs in vitro under the supraphysiological conditions in the CMST. In both cases cellular characteristics as obtained by the CMST measurements indicate an improvement after splenectomy, in a sense that they behave more like normal healthy RBCs. Furthermore, the degree of increase in ΔEI after splenectomy we observed could also be dependent on genetic defect (Ingrosso et al., 1996; Reliene et al., 2002). Our findings were strengthened by CMST results on a large cohort of HS patients that showed that this test was able to discriminate between splenectomized and non-splenectomized HS patients (Figure 4A). This implies that the CMST may represent a novel biomarker of HS, which was further supported by the correlation of ΔEI and % reticulocytes (r = 0.66, p = 0.004). The ability to distinguish splenectomized patients from non-splenectomized patients can be valuable in an era where different types of splenectomy are explored. Partial splenectomy, either through embolization of splenic arteries or through (laparoscopic) removal of a part of the spleen, might ameliorate symptoms and improve anemia while maintaining splenic phagocytic function (Tchernia et al., 1993; Pratl et al., 2008). However, data of several small studies are inconclusive regarding the remaining immunological capacity of the spleen after partial splenectomy even though hemolysis is decreased (Guizzetti, 2016). Larger studies that also include functional analysis of the spleen and functional analysis of RBCs, i.e., the CMST, are warranted to accurately assess the efficacy of (partial) splenectomy or embolization, and to investigate whether changes in RBC properties or the nature of the underlying molecular defect (Ingrosso et al., 1996; Reliene et al., 2002) contribute to the clinical benefit of splenectomy.

In conclusion, we report on the longitudinal effects of splenectomy on HS RBC characteristics and function as studied by ektacytometry. Our data shows that before splenectomy the HS RBC population is more heterogeneous, cells are more rigid, have increased intracellular viscosity and reduced deformability. Functional analysis of HS RBCs using osmotic gradient ektacytometry and CMST further shows that splenectomy improves the hydration state of HS RBCs and allows cells to regain the ability to shed membrane. In particular the CMST reflects an yet-undescribed distinct RBC characteristic and holds promise as a novel biomarker for membrane health in HS that could be helpful, together with a comprehensive clinical evaluation and appropriate follow-up, to assess the effect of different treatments such as embolization and (partial) splenectomy, and that may be related to clinical severity given the correlation of ΔEI and reticulocyte count. Larger studies are warranted to establish if the CMST can be used to improve the assessment of clinical severity and/or is able to contribute to a better understanding of phenotypic differences in HS.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by METC University Medical Center Utrecht. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

RH, MB, RW, and MR designed the study. RH, JV, ML, MB, EB, and MR collected clinical and laboratory data. JB, RH, TM, BO, and MR performed laboratory experiments. MCB, RW, and MR analyzed the data and wrote the manuscript. All authors edited the manuscript and approved the final version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.