METHOD AND DEVICE FOR DETERMINING RED BLOOD CELLS DEFORMABILITY

Abstract

The invention is related to a method for measuring the variability of the red blood cells deformability of an individual by determining the amount of red blood cells having a tank-treading motion in a population of red blood cells from a tested blood sample of said individual, and comparing the amount to a reference amount. The determination of the amount of red blood cells having a tank-treading motion is carried out using a visualisation means such as a brightfield microscope.

Claims

1. A method for the in vitro determination of the variation of the deformability of red blood cells of a tested individual, the method comprising the steps of: a) determining a first amount of red blood cells having a tank-treading motion in a population of red blood cells, said population of red blood cells being obtained from a sample of said tested individual, by performing the following steps: i. subjecting the population of red blood cells to a flow withing a circulation member having a boundary wall, the flow having a flow gradient and the boundary wall being perpendicular to the direction of the flow gradient, the flow having a direction and further having a wall shear rate from 1 s.sup.−1 to 50 s.sup.−1 and a viscosity from 10.sup.−3 Pa.Math.s to 10.sup.−1 Pa.Math.s, ii. recording in greyscale images, using a visualisation member connected to a recording member, the motions and trajectories of at least a part of the red blood cell population in an observation field of the visualisation member and obtaining at least one video being a sequence of successive images of the movement of the recorded red blood cells across the observation field, wherein the observation field is arranged perpendicularly to the direction of the flow gradient with a focus at a distance h where the flow has constant shear rate, wherein the images of the sequence comprise projected shapes of the red blood cells in a plane perpendicular to the direction of the flow gradient and parallel to the boundary wall of the circulation member, wherein the projected shapes of each red blood cell present a minor and a major axis, iii. individually tracking in all the images the projected shapes relating to at least a part of the recorded red blood cells, obtaining a motion sequence of individual images for each red blood cell and determining for each projected shape of a particular red blood cell a grey level intensity profile from the greyscale images, iv. analysing the evolution of the projected shapes of each tracked red blood cell in its motion sequence and classifying the motion type of the said red blood cell as follows: a red blood cell is classified as having a tank-treading motion on conditions that the ratio of the minor to major axis of its projected shape is greater than 0.45 in all images of its individual sequence and that the grey level intensity profile in at least 90% of its projected shapes has two extrema separated by at least 1 μm, a red blood cell is classified as having another motion on condition that the ratio of the minor to major axis of its projected shape is less than 0.45 in at least two images in the sequence, v. determining the first amount as the ratio of the number of red blood cells classified as having a tank-treading motion to the sum of the number of red blood cells classified as having a tank-treading motion and the number of red blood cells classified as having another motion, wherein steps ii. to v. are implemented by a control module and performed automatically, b) calculating the absolute value of the difference between said first amount of red blood cells having a tank-treading motion in a population of red blood cells, and a second amount being the calculated mean of the determined amount of red blood cells having a tank-treading motion obtained in reference populations of red blood cells, said calculated mean of the determined amount being obtained along with the value of the standard deviation of the red blood cells having a tank-treading motion obtained in the reference populations of red blood cells, wherein said reference populations are obtained either from at least two independent samples of said tested individual, or from at least one sample of at least two individuals, said at least two individuals being different from said tested individual, c) determining said absolute value is greater than twice the value of said standard deviation d) concluding the red blood cells of said population present a significant variation of their deformability compared to the red blood cells of said reference populations.

2. Method according to claim 1, wherein the circulation member is a parallepipedic flow chamber or a microchannel.

3. Method according to claim 2, wherein the observation field has a focus at most at 40% the height of the circulation member from the boundary wall.

4. Method according to claim 2, wherein the observation field has a length of at least the half period length L for enabling a red blood cell to show half a period of tumbling or a flip-flopping motion in the flow, L being determined according to the following formula:
L=3.Math.π.Math.h wherein h is the distance of the focus of the observation field from the boundary wall of the circulation member.

5. Method according to claim 2, wherein the at least one video recorded at step ii. lasts at least a time T required for a red blood cell to perform a half-period of tumbling or flip-flopping motion in the flow, T being determined according to the following formula: $T=3.Math.\frac{\pi }{\stackrel{.}{\gamma }}$ wherein {dot over (γ)} is the value of the wall shear rate of the flow.

6. Method according to claim 2, wherein the visualisation member comprises an objective with a depth of field, and wherein in step i. the red blood cells circulate in the circulation member in an extended layer centred at the distance h and of thickness corresponding to at most 5 times the depth of field of the objective.

7. Method according to claim 6, wherein in step i. the red blood cells are injected in the circulation member by a co-flow of two fluids configured to focus the red blood cells in the extended layer.

8. Method according to claim 2, wherein the visualisation member comprises an objective with a depth of field, wherein the recording member has a frame rate of recording, wherein in step i. part of the red blood cells circulate in the circulation member in an extended layer centred at the distance h and of thickness corresponding to at most 5 times the depth of field of the objective, wherein in step i. the rest of the red blood cells circulates outside the extended layer, and wherein the method further comprises the following sub-steps for excluding from the analysis step iv. the red blood cells circulating outside the extended layer: counting the number of individual images constituting the motion sequence of each red blood cell, excluding the red blood cells whose motion sequence has a number of individual images larger or smaller than N±1, wherein $N=\frac{n\times L}{\stackrel{.}{\gamma }\times h}$ wherein n is the image frame rate of the recording member, L is the length along the flow direction of the observation, {dot over (γ)} is the shear rate of the flow, and h is the distance of the focus of the observation field from the boundary wall of the circulation member.

9. Method according to claim 1, wherein step iii. comprises a step of excluding from analysing step iv. any projected shape whose largest dimension is less than 6 μm.

10. Method for the in vitro prediction of a vaso-occlusive crisis of a tested individual afflicted with a sickle-cell disease, the method comprising the steps of: a) determining a first amount of red blood cells having a tank-treading motion in a population of red blood cells, said population of red blood cells being obtained from a sample of said tested individual, by performing the following steps: i. subjecting the population of red blood cells to a flow within a circulation member having a bottom part, the flow having a flow gradient and the boundary wall being perpendicular to the direction of the flow gradient, the flow having a direction and further having a wall shear rate from 1 s.sup.−1 to 50 s.sup.−1 and a viscosity from 10.sup.−3 Pa.Math.s to 10.sup.−1 Pa.Math.s, wherein the circulation member is a parallepipedic flow chamber or a microchannel, ii. recording in greyscale images, using a visualisation member connected to a recording member, the motions and trajectories of at least a part of the red blood cell population in an observation field of the visualisation member and obtaining at least one video being a sequence of successive images of the movement of the recorded red blood cells across the observation field, wherein the observation field is arranged perpendicularly to the direction of the flow gradient with a focus at a distance h where the flow has constant shear rate, wherein the images of the sequence comprise projected shapes of the red blood cells in a plane perpendicular to the direction of the flow gradient and parallel to the boundary wall of the circulation member, wherein the projected shapes of each red blood cell present a minor and a major axis, iii. individually tracking in all the images the projected shapes relating to at least a part of the recorded red blood cells, obtaining a motion sequence of individual images for each red blood cell and determining for each projected shape of a particular red blood cell a grey level intensity profile from the greyscale images, iv. analysing the evolution of the projected shapes of each tracked red blood cell in its motion sequence and classifying the motion type of the said red blood cell as follows: a red blood cell is classified as having a tank-treading motion on conditions that the ratio of the minor to major axis of its projected shape is greater than 0.45 in all images of its individual sequence and that the grey level intensity profile in at least 90% of its projected shapes has two extrema separated by at least 1 μm, a red blood cell is classified as having another motion on condition that the ratio of the minor to major axis of its projected shape is less than 0.45 in at least two images in the sequence, v. determining the first amount as the ratio of the number of red blood cells classified as having a tank-treading motion to the sum of the number of red blood cells classified as having a tank-treading motion and the number of red blood cells classified as having another motion, wherein steps ii. to v. are implemented by a control module and performed automatically, b) calculating the value of the difference between a second amount being the calculated mean of the determined amount of red blood cells having a tank-treading motion obtained in reference populations of red blood cells, said calculated mean of the determined amount being obtained along with the value of the standard deviation of the red blood cells having a tank-treading motion obtained in the reference populations of red blood cells, and the said first amount of red blood cells having a tank-treading motion in a population of red blood cells, wherein said reference populations are obtained from at least two independent samples of said tested individual, in a period outside an episode of a vaso-occlusive crisis, c) determining said value of the difference is a positive value greater than twice the value of said standard deviation, d) concluding said tested individual is predicted is susceptible to experience, within 72 hours, a vaso-occlusive crisis.

11. Method according to claim 10, wherein the onset of a vaso-occlusive crisis is defined by the appearance of at least one first symptom selected from the group consisting of the appearance of a new bone pain affecting at least two territories, dyspnea or shortness of breath, and sputum.

12. Method according to claim 10, wherein said defined flow has a wall shear rate from 10 s.sup.−1 to 20 s.sup.−1.

13. Method according to claim 10, wherein said defined flow has a viscosity from 3*10.sup.−2 Pa.Math.s to 4.5*10.sup.−2 Pa.Math.s.

14. A method for the in vitro diagnosis of the response to a therapeutic treatment of a tested individual afflicted with a blood disorder affecting the deformability of the red blood cells and treated with said therapeutic treatment, said therapeutic treatment being directed to said blood disorder, the method comprising the steps of: a) determining a first amount of red blood cells having a tank-treading motion in a population of red blood cells, said population of red blood cells being obtained from a sample of said tested individual, by performing the following steps: i. subjecting the population of red blood cells to a flow within a circulation member having a bottom part, the flow having a flow gradient and the boundary wall being perpendicular to the direction of the flow gradient, the flow having a direction and further having a wall shear rate from 1 s.sup.−1 to 50 s.sup.−1 and a viscosity from 10.sup.−3 Pa.Math.s to 10.sup.−1 Pa.Math.s, wherein the circulation member is a parallepipedic flow chamber or a microchannel, ii. recording in greyscale images, using a visualisation member connected to a recording member, the motions and trajectories of at least a part of the red blood cell population in an observation field of the visualisation member and obtaining at least one video being a sequence of successive images of the movement of the recorded red blood cells across the observation field, wherein the observation field is arranged perpendicularly to the direction of the flow gradient with a focus at a distance h where the flow has constant shear rate, wherein the images of the sequence comprise projected shapes of the red blood cells in a plane perpendicular to the direction of the flow gradient and parallel to the boundary wall of the circulation member, wherein the projected shapes of each red blood cell present a minor and a major axis, iii. individually tracking in all the images the projected shapes relating to at least a part of the recorded red blood cells, obtaining a motion sequence of individual images for each red blood cell and determining for each projected shape of a particular red blood cell a grey level intensity profile from the greyscale images, iv. analysing the evolution of the projected shapes of each tracked red blood cell in its motion sequence and classifying the motion type of the said red blood cell as follows: a red blood cell is classified as having a tank-treading motion on conditions that the ratio of the minor to major axis of its projected shape is greater than 0.45 in all images of its individual sequence and that the grey level intensity profile in at least 90% of its projected shapes has two extrema separated by at least 1 μm, a red blood cell is classified as having another motion on condition that the ratio of the minor to major axis of its projected shape is less than 0.45 in at least two images in the sequence, v. determining the first amount as the ratio of the number of red blood cells classified as having a tank-treading motion to the sum of the number of red blood cells classified as having a tank-treading motion and the number of red blood cells classified as having another motion, wherein steps ii. to v. are implemented by a control module and performed automatically, b) calculating the value of the difference between said first amount of red blood cells having a tank-treading motion in a population of red blood cells, and a second amount being the calculated mean of the determined amount of red blood cells having a tank-treading motion obtained in reference populations of red blood cells, said calculated mean of the determined amount being obtained along with the value of the standard deviation of the red blood cells having a tank-treading motion obtained in the reference populations of red blood cells, wherein said reference populations are obtained from at least two independent samples of said tested individual, before the beginning of said therapeutic treatment, c) determining said value of the difference is greater than twice the value of said standard deviation, d) concluding said tested individual is responsive to the said therapeutic treatment.

15. Method according to claim 14, wherein the defined flow has a wall shear rate from 5 s.sup.−1 to 19 s.sup.−1.

16. Method according to claim 14, wherein the defined flow has a viscosity from 0.15*10.sup.−1 to 0.5*10.sup.−1 Pa.Math.s.

17. Method according to claim 14, wherein the blood disorder affecting the deformability of the red blood cells is selected from the group consisting of cell-sickle disease, thalassemia, acute coronary syndrome, bacterial sepsis, diabetes, malaria, stroke, paroxysmal nocturnal hemoglobinuria, haemolytic microangiopathy, red blood cell membrane disorders, enzymopathies, and hemoglobinopathies.

Description

LEGEND TO THE FIGURES

[0291] FIG. 1 is a schematic view of the three regimes of motion of RBC. Time sequences schematically representing the projected shapes in the shear plane of RBCs moving from left to right in a shear flow: tumbling (a), rolling (b) and tank-treading (c). The black dots on the RBCs together with arrow display the motion of a membrane element on the RBCs surface.

[0292] FIG. 2 is a schematic of the tank-treading experiment comprising a flow chamber where RBCs move and a microscope associated with a camera for recording images of said motion. The RBC stabilized motion is acquired at the end of the flow chamber (frame: 660×571 μm.sup.2) after they have moved for about 5 mm at a constant shear rate {dot over (γ)}. Left scale bar: 10 μm; right scale bar: 50 μm. 21-fps movies are analysed to classify RBC motion into three categories: Tank-Treading (top), Rolling (middle) and Flip-Flopping (bottom).

[0293] FIGS. 3A-3D represent the sensitivity of the proportion of RBCs in the tank-treading regime in the RBCs population from patients afflicted with sickle cell disease and healthy individuals. FIG. 3A is a graph representing the evolution of the ratio of RBCs found to be tank-treading f.sub.TT (Y-axis from 0.2 to 1) in the whole RBCs population as a function of the shear rate {dot over (γ)} (X-axis from 0 to 16) for three healthy samples (open circle symbols): from top to bottom HbAA11, HbAA3, HbAA4 from Table 1) and three SCD samples (solid circle symbols: from top to bottom HbSS31, HbSS9, HbSS8 from Table 1. FIG. 3B is a graph representing box plots of f.sub.TT (Y-axis) at a constant shear rate of 10 s.sup.−1 for nine healthy samples (individual HbAA 3-12 from Table 1) on the left box plot, and on the middle box plot fourteen SCD (patients HbSS 4-17 from Table 1), and on the right box plot 15 heterozygote thalassemic patients (HbpA3-17 from the Table 1). FIG. 3C is a graph representing the Evolution of f.sub.TT (Y-axis) at a constant shear rate of 15 s.sup.−1 as function of RBC mean density (X-axis) for 2 healthy (hollow circle symbols: individuals HbAA 1 and 5 from Table 1) and 2 SCD (solid circle symbols: patients HbSS 3 and 16 from Table 1) samples. FIG. 3D is a graph representing the evolution of f.sub.TT (Y-axis) at a constant shear rate of 15 s.sup.−1 as function of RBC state of dehydration (X-axis) for one healthy (hollow circle symbols: individual HbAA 6 from Table 1, h) and one SCD ((solid circle symbols: patient HbSS 17 from Table 1) samples. Dehydration is achieved by using a hyper-osmotic external buffer (>300 mOsm).

[0294] FIGS. 4A-4C represent the evolution of the proportion of tank-treading RBCs in the RBCs population in 7 SCD patients during vaso-occlusive events. Blood samples were harvested from patients during their stay at the hospital due to the occurrence of a crisis. For each graph is considered the following representation: solid circle symbols correspond to patient HbSS 22 from Table 1; solid square symbols correspond to patient HbSS 24 from Table 1; up triangle symbols represent patient HbSS 25 from Table 1; diamond symbols represent patient HbSS 19 from Table 1; criss-cross square symbols represent patient HbSS 20 from Table 1; down triangle symbols represent patient HbSS 23 from Table 1; hollow circle symbols represent patient HbSS 21 from Table 1. FIG. 4A is a graph representing the temporal evolution of f.sub.TT (Y-axis from 0.5 to 1) during and after crisis for five patients who had a single crisis. The X-axis (from −8 to 10) represents the days of hospitalisation for the vaso-occlusive crisis with 0 corresponding to the moment of hospitalisation. For patient HbSS 24, a blood sample has been obtained 6 days before being hospitalized for the vaso-occlusive crisis. FIG. 4B is a graph representing the temporal evolution of f.sub.TT (Y-axis from 0.5 to 1) during and after crisis for two patients who underwent a second crisis during their stay. The X-axis (from 0 to 18) represents the days of hospitalisation for a vaso-occlusive crisis with 0 corresponding to the moment of hospitalisation. FIG. 4C is a graph representing the f.sub.TT temporal evolution compiling data from (A) and (B) with abscissa and ordinates shifted for each curve: 0-abscissa is the estimated time of maximum f.sub.TT and 0-ordinate is the estimated base level.

[0295] FIGS. 5A-5B represent a device according to the invention. FIG. 5A is a top view and FIG. 5B is a front view. In FIGS. 5A and 5B, the different elements are represented transparently.

[0296] FIG. 6 is a histogram which represents the baseline for the ratio of RBCs that were found to tank-tread f.sub.TT (Y-axis from 0 to 1) for 9 homozygous SCD patients (X-axis from 1 to 9) over 6 months. For patient 1, 20 samples were recovered. For patient 2, 17 samples were recovered. For patient 3, 19 samples were recovered. For patient 4, 21 samples were recovered. For patient 5, 17 samples were recovered. For patient 6, 20 samples were recovered. For patient 7, 22 samples were recovered. For patients 8 and 9, 11 samples were recovered respectively. For each histogram is also represented the standard deviation. Each measurement was performed at a shear rate {dot over (γ)} of 10 s.sup.−1.

[0297] FIG. 7 is a graphical representation of the ratio of RBCs found to tank-tread f.sub.TT (Y-axis from 0 to 1) as a function of time (X-axis in days) for patient 1 represented in FIG. 6. Each dot represents a measurement on a sample of the patient.

[0298] FIG. 8 is a graphical representation of the ratio of RBCs found to tank-tread f.sub.TT (Y-axis from 0 to 1) as a function of time (X-axis in days) for patient 2 represented in FIG. 6. Each dot represents a measurement on a sample of the patient.

[0299] FIG. 9 is a graphical representation of the ratio of RBCs found to tank-tread f.sub.TT (Y-axis from 0 to 1) as a function of time (X-axis in days) for patient 3 represented in FIG. 6. Each dot represents a measurement on a sample of the patient.

[0300] FIG. 10 is a graphical representation of the ratio of RBCs found to tank-tread f.sub.TT (Y-axis from 0 to 1) as a function of time (X-axis in days) for patient 4 represented in FIG. 6. Each dot represents a measurement on a sample of the patient.

[0301] FIG. 11 is a graphical representation of the ratio of RBCs found to tank-tread f.sub.TT (Y-axis from 0 to 1) as a function of time (X-axis in days) for patient 5 represented in FIG. 6. Each dot represents a measurement on a sample of the patient.

[0302] FIG. 12 is a graphical representation of the ratio of RBCs found to tank-tread f.sub.TT (Y-axis from 0 to 1) as a function of time (X-axis in days) for patient 6 represented in FIG. 6. Each dot represents a measurement on a sample of the patient.

[0303] FIG. 13 is a graphical representation of the ratio of RBCs found to tank-tread f.sub.TT (Y-axis from 0 to 1) as a function of time (X-axis in days) for patient 7 represented in FIG. 6. Each dot represents a measurement on a sample of the patient.

[0304] FIG. 14 is a graphical representation of the ratio of RBCs found to tank-tread f.sub.TT (Y-axis from 0 to 1) as a function of time (X-axis in days) for patient 8 represented in FIG. 6. Each dot represents a measurement on a sample of the patient.

[0305] FIG. 15 is a graphical representation of the ratio of RBCs found to tank-tread f.sub.TT (Y-axis from 0 to 1) as a function of time (X-axis in days) for patient 9 represented in FIG. 6. Each dot represents a measurement on a sample of the patient.

[0306] FIG. 16 represents the ratio of RBCs found to tank-tread f.sub.TT (Y-axis from 0 to 1) as a function of time (X-axis in days) for a homozygous SCD patient (patient 10). Each dot represents a measurement on a sample of the patient. Each measurement performed at a shear rate {dot over (γ)} of 10 s.sup.−1. The horizontal solid lines represent the mean f.sub.TT value for an “out of crisis” period. The vertical line A represents the day when the patient's pain is a sign of crisis; the vertical line B represents the day of hospitalization; and the vertical line C represents the day of discharge from the hospital.

[0307] FIG. 17 represents the ratio of RBCs found to tank-tread f.sub.TT (Y-axis from 0 to 1) as a function of time (X-axis in days) for a homozygous SCD patient (patient 11). Each dot represents a measurement on a sample of the patient. Each measurement was performed at a shear rate {dot over (γ)} of 10 s.sup.−1. The horizontal solid lines represent the mean f.sub.TT value for an “out of crisis” period. The vertical line A represents the day when the patient's pain is a sign of crisis; the vertical line B represents the day of hospitalization; and the vertical line C represents the day of discharge from the hospital.

[0308] FIG. 18 represents the ratio of RBCs found to tank-tread f.sub.TT (Y-axis from 0 to 1) as a function of time (X-axis in days) for a homozygous SCD patient (patient 12). Each dot represents a measurement on a sample of the patient. Each measurement was performed at a shear rate {dot over (γ)} of 10 s.sup.−1. The horizontal solid lines represent the mean f.sub.TT value for an “out of crisis” period. The vertical line A represents the day when the patient's pain is a sign of crisis; the vertical line B represents the day of hospitalization; and the vertical line C represents the day of discharge from the hospital.

[0309] Hereafter it is described one example of such a device according to FIGS. 5A and 5B, and its method of operating.

[0310] A fingertip prick blood drop is first recovered on the finger of a tested individual and diluted in 1.5 ml of Dextran solution. The recovery can be done by a capillary tube (not shown).

[0311] Then, the blood is mechanically mixed with the Dextran solution by using a syringe 3 connected to the said capillary tube. The solution of dextran can be conditioned in unidose. The syringe is then filled in with the mixed RBC suspension.

[0312] Subsequently, the syringe 3 is set on a syringe pump 1 in a device 8 and connected to a flow chamber 5. The syringe pump 1 comprises a motor (not shown) for pushing the RBC suspension into the flow chamber 5 at a given viscosity and shear rate.

[0313] Afterwards, the device 8 is switched on which causes the activation of the motor of the syringe pump 1. Accordingly, said syringe pump 1 pushes the blood suspension contained within the syringe 3 into the flow chamber 5. Accordingly, the blood suspension will flow through the flow chamber 5 at the given shear rate and viscosity. The syringe pump 1 is configured to supply within the flow chamber 5 a flow having a wall shear rate from 1 s.sup.−1 to 50 s.sup.−1, and a viscosity from 10.sup.−3 Pa.Math.s to 10.sup.−1 Pa.Math.s.

[0314] The device 8 comprises a light 4, optical lenses 10, and a camera 11 which are switched on with the pump's motor to acquire a video of flowing red blood cells. The recorded video is stored in a computer 2. This computer 2 can be placed within the device 8 as shown in the figure or can be placed away. For example, the computer can be a smartphone.

[0315] The computer comprises an adequate software to analyze these videos. Different analyses can be done and include, but are not limited to, the percentage of the red blood cell with a tank-treading motion, pictograms, and recommendations.

[0316] The pictograms can represent a colour code, which is a function of the result of the analysis, such as red, yellow, green, with red meaning having rest and contact a doctor, yellow meaning to repeat the analysis within 24 hours and green meaning that the result of the analysis is good.

[0317] The recommendations can be to have rest, to repeat the analysis within 24 hours or to contact a doctor, in function of the result of the analysis.

[0318] The results of the analysis (% TT, pictogram, recommendation) are displayed on a screen. This screen can be part of the device 8, as shown in the FIGS. 5A and 5B, or be placed away. In case the computer is a smartphone, the said screen would be the screen of the smartphone.

[0319] The results can be stored in the computer and potentially, sent to a medical centre or a doctor.

[0320] The device 8 can further comprise an outlet 7 connected at the end of the flow chamber 5, in order to recover the RBC solution that has flowed through the chamber 5.

[0321] An inlet 6 can advantageously be connected to the tip of the syringe 3. In such a case, the syringe 3 is directly filled in within the device, so that there is no need to put the syringe 3 out of the device.

Example

[0322] Methods

[0323] Buffers

[0324] dextran (from Leuconostoc menseteroides, 2000 kDa, Sigma-Aldrich) was solubilized at 9% (wt/wt) in homemade PBS (osmolarity: 295±5 mOsm, pH: 7.4) by stirring at 50° C. for at least 2 hours. The dextran solution had a viscosity η.sub.o=39.2±0.7 10-3 Pa-s at a temperature of 20° C. and its density approximately matched RBC density, thus preventing cell sedimentation. Dehydration experiments were conducted using hyper-osmotic PBS buffers (up to 600 mOsm) which were prepared by concentrating all reagents in the PBS proportionally.

[0325] Blood Samples

[0326] Most tested blood specimens are a subset of residual samples referred to the Department of Genetics (Hôpital de La Timone, Marseille, France) for routine tests (haematological parameters for each individual/patient are given in Table 1 below). Blood was obtained from thirty homozygous subjects with SCD (HbSS), eight SCD patients during a vaso-occlusive event requiring hospitalization (Crisis), HbSS patients were selected preferentially with high % HbS (non-transfused) and low % foetal Hb (HbF), and fifteen heterozygous patients with thalassemia. Blood samples were mixed with EDTA when harvested and RBCs were then isolated within 24 hours by three consecutive washing steps in SAG-mannitol (SAGM, EFS, France) at a centrifugation rate of 500 g (10 min at 4° C.), then suspended in SAGM (haematocrit, Hct: 50%) for storage at 4° C., and used within 7 days. The blood samples were used as follows: individual HbAA 3 to 12, patients HbSS 4 to 17 and thalassemic patients were used in tank-treading experiments; Patients HbSS 8 and 19 to 25 were followed during vaso-occlusive events requiring hospitalisation.

TABLE-US-00001 TABLE 1 haematological parameters of tested subjects Patient/ Age Hct MCV MCH MCHC Individual (years) % HbS % HbF (L/L) (μm.sup.3) (pg/cell) (g/dL) Range N/A N/A N/A 0.37-0.47 80-98 27-32 30-36.5 HbSS 1 22 84 7.1 0.21 81.6 29.8 36.5 HbSS 2 33 72.4 18.1 0.19 101 38 37.6 HbSS 3 36 84.6 7.8 0.21 89.2 32.3 36.2 HbSS 4 34 82 9.2 0.2  85.9 30.3 35.3 HbSS 5 16 78.9 11.1 0.24 83.8 29.9 35.7 HbSS 6 14 78.9 12.1 0.21 79.3 26.7 33.6 HbSS 7 49 48.3 33 0.14 112 41.6 37.1 HbSS 8 31 71.3 11.9 0.21 80.9 24.6 30.4 HbSS 9 32 71.1 17.3 0.21 75.1 26.3 35 HbSS 10 Missing 63.8 24.9 0.22 89.7 32.2 35.9 HbSS 11 Missing 77.2 13.6 0.26 100.8 35.3 35 HbSS 12 22 84.5 4.6 0.23 67.9 23.3 34.3 HbSS 13 19 69.7 18.2 0.29 83.9 28.1 33.4 HbSS 14 41 73 10.8 0.28 113.4 38.9 34.3 HbSS 15  6 85.4 6.1 0.19 92.3 33.7 36.5 HbSS 16 Missing 83.9 3.4 0.19 90 32 35.6 HbSS 17 15 88.3 3.4 0.21 67.1 23.2 34.6 HbSS 18 32 71 11.3 0.18 91.5 32.3 35.3 HbSS 19 35 78.2 13.2 0.26 94.6 33.7 35.6 HbSS 20 23 74.1 7.4 0.25 86 32.1 37.3 HbSS 21 20 73.4 13.7 0.18 71.5 23.6 33 HbSS 22 25 55.2 1.1 0.23 70.9 23.9 33.6 HbSS 23 26 75.3 4.3 0.27 88.7 30.7 34.6 HbSS 24 33 24.1 3.1 0.24 80.5 25.9 32.2 HbSS 25 18 76 13.9 0.23 86.8 33.2 38.3 HbSS 26 Missing Missing Missing Missing Missing Missing Missing HbSS 27 51 66.3 20.2 0.23 107.1 38.2 35.7 HbSS 28 51 68.1 19.5 0.23 107.1 38.2 35.7 HbSS 29 51 74.8 13 0.22 96.1 35.8 37.3 HbSS 30 21 67.5 18.7 0.18 80.3 24.3 30.3 HbSS 31 missing 71.8 15.9 missing missing missing missing HbβS 1 10 72 9.9 0.24 69.9 23.2 33.2 HbβS 2 14 72.4 14.7 0.22 65.2 21.2 32.6 HbAA 1 39 0 0.8 0.39 88 30 34 HbAA 2 47 0 0.5 0.42 82.5 28.6 34.6 HbAA 3 35 0 0.5 0.42 86 30 34.4 HbAA 4 43 0 1.1 0.35 78 27 34.4 HbAA 5 13 0 0.4 0.35 81.6 27.9 34.2 HbAA 6 22 0 0.6 0.3  92.7 33.1 35.7 HbAA 7 34 0 0.5 0.42 86 30 34.4 HbAA 8-11 Volunteers, no information HbβA 1 Volunteer, no information HbβA 2 59 0 0.5 0.42 63 20.1 31.9 HbβA 3 Missing 0 1.1 Missing Missing Missing Missing HbβA 4 Missing 0 3.3 Missing Missing Missing Missing HbβA 5 Missing 0 1.2 Missing Missing Missing Missing HbβA 6 Missing 0 10.1 Missing Missing Missing Missing HbβA 7 Missing 0 1.2 Missing Missing Missing Missing HbβA 8 Missing 0 1.7 Missing Missing Missing Missing HbβA 9 Missing 0 1.4 Missing Missing Missing Missing HbβA 10 Missing 0 0.8 Missing Missing Missing Missing HbβA 11 Missing 0 0.7 Missing Missing Missing Missing HbβA 12 Missing 0 0.7 Missing Missing Missing Missing HbβA 13 Missing 0 1 Missing Missing Missing Missing HbβA 14 Missing 0 1.1 Missing Missing Missing Missing HbβA 15 Missing 0 unknown Missing Missing Missing Missing HbβA 16 Missing 0 1.3 Missing Missing Missing Missing HbβA 17 Missing 0 80.7 Missing Missing Missing Missing Hbββ 1 Missing 0 75.7 Missing Missing Missing Missing
packed red blood cells.

[0327] Flow experiments and microscopy

[0328] Flow experiments were performed as described previously (FIG. 2). Briefly, RBCs stored at 50% Hct were diluted (≈500×) in a solution of Dextran and injected in a parallelepiped quartz flow chamber (50×10×0.9 mm.sup.3) mounted on an inverted microscope (DMIRB, Leica). The fluid was driven by a syringe pump (11 Plus, Harvard Apparatus) at increasing wall shear rates ? starting from 1 s.sup.−1 up to 20 s.sup.−1.

[0329] First Experiment

[0330] RBCs were observed within 65 μm from the bottom wall (zone of constant shear rate) in brightfield microscopy (20× objective) along the direction of the flow gradient. Images were recorded at 21 fps with a camera and processed semi-automatically using Matlab home-made routines. Each RBC passing in the observation field of the microscope was individually tracked by a custom-made Matlab program and the projection of its shape in the observation plane was detected and fitted by an ellipse to identify its regime of motion.

[0331] The tank-treading motion of the red blood cells using custom-made Matlab tracking programs is determined as it follows:

[0332] The determination is carried out in two parts.

[0333] Part 1

[0334] This part corresponds to the pre-categorization of the MATLAB routine that sorts all cells into different categories: speed, remove, multiple, others, axis, size, suspicious, track correlation, area and tank treading.

[0335] The criteria for each category are:

[0336] Speed: the cell falls within this category if the cell moves more than twice as fast than the average between two frames. The speed of one cell is calculated by the program by measuring the displacement of the cell between two frames. This is done for each cell and for its entire trajectory;

[0337] Remove: the cell falls within this category if the cell trajectory across the observation field of the microscope is less than seven frames long;

[0338] Multiple: the cell falls within this category if the trajectory of two cells overlap;

[0339] Others: the cell falls within this category if the length ratio of the minor axis divided by the major axis of the cell is less than 0.5. In such a case, the motion of the cell corresponds either to rolling or to tumbling;

[0340] Axis: the cell falls within this category if the ratio between the main geometrical axes of the cell lies between 0.5 and 0.7. This implies in such a case, that the cell maintains a shape that is not characteristic for the typical motion types);

[0341] Size: the cell falls within this category if the major axis of the cell is smaller than a defined threshold (10 for instance). This threshold is given in pixel and can be adjusted to object size and used equipment;

[0342] Suspicious: the cell falls within this category if the cell presents an oddly shaped. This is carried out by comparing the original black and white image of the cell with a modified image of the same cell (the interior of the cell gets filled white, if the borders of the cell are well defined, the exterior region remains unchanged, if not, the whole picture turns white) to ensure integrity of cell boundaries;

[0343] Track correlation: the cell falls within this category if its motion is not consistent. This is carried out by calculating a mean and standard deviation for each cross correlation and normalizing them to 1. Since the cells move across the observation field of the microscope according to a line direction with nearly constant speed, the trajectory of the cell is very predictable and can be easily correlated by localizing the cell on frame n and looking for the cell on frame n+1 at the distance travelled between two frames. The intensity (arbitrary unit within the program) of every pixel of the cell is compared with the corresponding pixel of the cell in the next frame using an internal MATLAB algorithm giving an output between 0 (no correlation) to 1 (100% correlation) representing the comparability of the two images without unit. This is done for all images of the trajectory of each cell. If the mean of the comparability for all images of one trajectory of one cell is below 0.5 or the standard deviation below 0.1, the cell trajectory is moved in this category as it is unsure if the trajectory is consistent.

[0344] Area: the cell falls within this category if the cell diameter is smaller than a certain threshold (10 for instance). This threshold is given in pixel and can be adjusted to object size and used equipment;

[0345] Tank treading: by exclusion, the cells passing through all the above criteria fall within the tank-treading category.

[0346] Part 2

[0347] After the automated step of Part 1, one can check and delete cells or confirm or move their categories.

[0348] For example, one can check that each cell is followed during a minimum amount of frames (set in the program). One can check as well that the cell displays no length change on the entire trajectory across the observation field of the microscope.

[0349] Second Experiment

[0350] In a second experiment, red blood cells were observed in brightfield microscopy (20× objective) along the direction of the flow gradient, at a distance h from the bottom wall of the chamber of less than 100 μm, i. e. in an area of constant shear rate. Videos of moving red blood cells passing through the camera field of view were taken with a camera that records the projected shapes of the cells in the plane perpendicular to the direction of the flow gradient.

[0351] Experiments conducted by observing RBCs located at a distance h=65 μm from the bottom wall of the chamber, under a shear rate of ?=10 s.sup.−1, were performed with a square field of view of length equal to 660 μm and the recording lasted at least one minute. The length of the field of view and the recording time enabled, for each experiment, to observe the motion of a number of RBCs ranging from 250 to 1200 and to detect the motion during at least half a period of tumbling/flip-flopping RBCs (with {dot over (γ)}=10 s.sup.−1 and h=65 μm, the time T for a half rotation of a cell is 0.94 s and the distance L travelled by the cell is 613 μm). Movies were recorded at 21 fps with a camera and processed semi-automatically using Matlab home-made routines. The crossing time of the field of view of each RBC was 1.02 s, corresponding to 20 to 23 images recorded for each RBC.

Using a custom-made «tracking and segmenting» Matlab program, each object present in a movie was individually tracked and a bounding box was defined around it to generate a sequence of 20-23 images saved in a separate file. The shape of the object detected in each image of a given sequence from the difference between the intensity of the grey levels of each pixel of the image was fitted by an ellipse.

[0352] Each sequence was analyzed using custom-made Matlab automatic programs and the RBCs animated either by a tanktreading or by a tumbling/flip-flopping/rolling motion (called other) were determined as follows:

[0353] Sequences where the major axis of the ellipse observed on each image constituting the sequence is smaller than 6 m are eliminated from the analysis. Indeed, it is then considered that the object detected is not an RBC but an artifact (debris, dust)

The remaining sequences are considered to be sequences showing RBCs. [0354] If there are at least two images in the sequence for which the ratio between the minor axis and the major axis of the ellipse is less than 0.45, the RBC in the sequence is classified as ‘other’ (RBC viewed on the edge) [0355] If all images in the sequence have ellipses with a ratio of minor to major axis greater than 0.45, the grey level intensity profile of each ellipse along the segment aligned with the flow direction passing through the centre of the ellipse and bounded by the contour of the ellipse is analysed. The RBC biconcavity (non-constant cell thickness) results in a grey-level intensity difference between the periphery and the centre of the cell (dimple) (FIG. 2 top, left images). Therefore, if the grey level intensity profile of 90% of the ellipses in the sequence has two extrema separated by at least 1 μm, the RBC is considered to maintain a stable orientation and therefore tank-treads. It is classified in the ‘tank-treading’ category. Otherwise the sequence is eliminated and the RBC is not counted.

[0356] For each shear rate, 250 to 1200 RBCs were observed and classified as tank-treading or other, and the fraction of tank-treading RBCs (f.sub.TT) was calculated as the ratio of the number of cells classified as tank-treading to the sum of the number of cells classified as tank-treading and the number of cells classified as other.

[0357] Third Experiment

[0358] In another experiment, the RBCs were observed at a distance h=65 μm from the bottom wall of the chamber but were circulating at different heights in the flow chamber. Prior to the analysis of the sequences, it was necessary to eliminate the RBCs that were not in a layer located between 60 μm and 70 μm from the bottom of the chamber because the images of these cells were blurred and did not allow to determine the nature of the cell movements. Sequences with less than 20 images and more than 23 images were eliminated because they corresponded to RBCs circulating at a distance greater than 70 μm and less than 60 μm, respectively, and therefore blurred. The rest of the analysis was identical to the second experiment.

[0359] Statistics

[0360] For each shear rate, 250 to 1200 RBCs were observed and classified as tank-treading or not, and the fraction of tank-treading RBCs (f.sub.TT) was calculated.

[0361] Results

[0362] The inventors focus on two diseases, mainly SCD, and to a lesser extent, thalassemia. The inventors also focus on two physical physiologically relevant quantities, often associated with painful SCD crises, namely RBC density and internal water content.

[0363] Transition to Tank-Treading Regime: Effect of RBC Density, Water Content and SCD

[0364] The inventors observed the passage of collections of RBCs, including ISCs, (FIG. 2) and extracted the fraction of RBCs having reached the TT regime, which the inventors referred to as f.sub.TT, as a function of the shear stress. As shown in FIG. 3A, the f.sub.TT-increase with η.sub.o{dot over (γ)} exhibited two clearly distinct curves for healthy and SCD RBCs, thus suggesting that f.sub.TT could be a good candidate as a mechanical marker of SCD. The value of f.sub.TT measured at η.sub.o{dot over (γ)}=0.6 Pa (i.e. {dot over (γ)}=15 s.sup.−1) yields the strongest difference between the behaviour of SCD and healthy RBCs. The inventors chose this shear stress to characterize healthy and SCD RBC behaviours across many samples. At this shear rate, almost all HbAA RBCs (98.6±0.9%) are observed to be tank-treading versus only 70.4±12.9% for HbSS samples (FIG. 3B). The two distributions do not overlap, thus disclosing that f.sub.TT allows discriminating SCD from healthy RBCs. Moreover, the f.sub.TT distribution for HbSS RBCs is wide, revealing the heterogeneity of the mechanical properties displayed by SCD cells. Only 73.4±8.2% RBCs from heterozygotous thalassemic patients tanktread. This distribution does not overlap with that of the healthy RBCs, thus disclosing that f.sub.TT allows discriminating thalassemic from healthy RBCs.

[0365] Results from FIG. 3C show a marked decrease in f.sub.TT with cell density for both healthy and SCD samples. The f.sub.TT decreases with density is more drastic for SCD than for healthy RBCs.

[0366] Finally, f.sub.TT was found to be sensitive to RBC density and internal water content (figuring the state of dehydration of RBCs) as illustrated in FIGS. 3B and 3D. Here, RBC dehydration was achieved through hyper-osmolarity of the external medium, starting from the physiological 300 mOsm up to osmolarities of 600 mOsm. Results showed a marked decrease in f.sub.TT with cell density and dehydration for both healthy and SCD samples, the f.sub.TT decrease with density being more drastic for SCD than for healthy RBCs.

[0367] RBC dehydration is known to be a parameter relevant to the clinical condition of the sickle cell patient, favoring the occurrence of vaso-occlusive crises in SCD. Furthermore, the RBC density profile of SCD patients was shown to vary during vaso-occlusive crises. This suggested that f.sub.TT could be an easy-to-measure marker for questioning the clinical condition of a sickle cell patient during a vaso-occlusive crisis.

[0368] Fraction of TT RBCs as a Clinical Marker in SCD Crises

[0369] The inventors thus studied the temporal evolution of f.sub.TT on seven patients hospitalized for an acute SCD crisis as shown in FIGS. 4A and 4B. The first measurement, set at time=0, was done at the arrival of the patient at the hospital. In one case (patient HbSS 24), a measurement was done on the patient one week before he was admitted for an acute crisis. In another case (patient HbSS 20), the patient was admitted twice at the hospital for two crises that occurred two months apart. In two cases (Patient HbSS 23 and patient HbSS 21), the patients were hospitalized for a crisis and had another crisis during their stay at the hospital (FIG. 4B). Patient HbSS 23 was transfused at day 1, which, surprisingly, did not prevent the second crisis event. In all other cases of transfusion of patients, measurements were stopped after transfusion.

[0370] It clearly appears that f.sub.TT significantly varies during the course of the crisis. A maximum of f.sub.TT is typically observed. In general, this maximum is observed shortly after or shortly before hospitalization, as shown for patient HbSS 22, patient HbSS 19, patient HbSS 23, patient HbSS 21, and patient HbSS 24. The time at which this maximum is observed is taken in the following as the reference time at which the crisis occurs. In the case of patient HbSS 24, the maximum was probably reached before his/her late hospitalization but the value collected 7 days before the crisis allows highlighting the maximum during the crisis. Similarly, the maximum was probably also reached before hospitalization for the first crisis of patient HbSS 23 and patient HbSS 21. Patients HbSS 19 and HbSS 21 were monitored long enough after their crisis so that their f.sub.TT curves display a post-crisis decrease down to a constant level (base level). The two curves clearly show that this base level varies between patients, as already suggested by the strong dispersion of f.sub.TT measured on SCD patients (FIG. 3A). FIG. 4C pools all f.sub.TT curves shifted by subtraction of the base level to display the variation of f.sub.TT during a crisis, and shifted in time by defining the 0-time when f.sub.TT is at maximum. It shows that a variation between 6 and 10% is observed between the base level and the maximum of f.sub.TT reached during the crisis, and that f.sub.TT comes back to its base level within 2-3 days. Strikingly, a significant f.sub.TT decrease is observed several days before the crisis, that can reach 20% (patient HbSS 24) two days before the crisis. This shows that f.sub.TT is not only a relevant parameter of crisis but is also a crisis prediction marker.

[0371] Strikingly, f.sub.TT varies before and during vaso-occlusive events in SCD. The pre-crisis decrease is attributed to a larger number of RBCs with an enhance in membrane rigidity and/or in cytoplasmic viscosity. The strong increase in f.sub.TT during the crisis reveals that a large amount of these altered RBCs no longer circulate in the microvasculature, likely due to their lysis and/or their blockage in blood vessels. The blockage may be induced by an enhanced adhesiveness at the wall vessels and/or sequestration in clogged vessels. A strong asset of f.sub.TT is to be highly sensitive not only to the membrane rigidity, as it is the case for several methods, including ektacytometry, but also to cytoplasmic viscosity, a mechanical parameter that is strongly affected in SCD due to fibre formation of haemoglobin. Moreover, f.sub.TT is indirectly sensitive to RBC adhesiveness, a factor involved in vaso-occlusion, since HbSS non-deformable RBCs exhibit increased adhesion sites compared to HbSS deformable RBCs.

[0372] Evolution of the f.sub.TT Parameter on Patients During the “Out of Crisis” Period

[0373] The inventors studied the variability of the f.sub.TT parameter on 9 homozygous SCD patients over 6 months. Each patient has been weekly tested and at least 10 times over the 6 months.

[0374] Patient 1 is a woman of 33 years old with kidney disease, and who is treated with hydroxyurea, converting enzyme inhibitor and folic acid.

[0375] Patient 2 is a woman of 48 years old with kidney and liver disease, and who is treated with hydroxyurea and IEC.

[0376] Patient 3 is a woman of 48 years old with liver disease and retinopathy, and who is treated with hydroxyurea.

[0377] Patient 4 is a woman of 35 years old who is untreated.

[0378] Patient 5 is a man of 45 years old with liver and kidney disease and retinopathy, who has deep vein thrombosis, and is treated with hydroxyurea, angiotensin converting enzyme inhibitor and antiplatelet agent.

[0379] Patient 6 is a woman of 24 years old who is treated with hydroxyurea and folic acid.

[0380] Patient 7 is a woman of 25 years old who has deep vein thrombosis, and who is treated with hydroxyurea.

[0381] Patient 8 is a woman of 35 years old who is treated with hydroxyurea.

[0382] Patient 9 is a woman of 23 years old who has deep vein thrombosis, and who is treated with hydroxyurea and who was treated with darbepoetin alfa from day 1 to day 15 and then with apixaban.

[0383] These patients did not suffer a vaso-occlusive crisis during the tested period. The mean f.sub.TT values recorded for each patient and the associated standard deviations are represented in FIG. 6. Temporal evolution of f.sub.TT per individual are displayed in FIGS. 7 to 15.

[0384] From FIGS. 7 to 15, it can be deduced that the difference between the f.sub.TT mean and the f.sub.TT values measured for each patient is less than twice the value of the standard deviation when patients do not have a vaso-occlusive crisis: [0385] For patient 1, the mean f.sub.TT value is 0.805 and the standard deviation is 0.027; [0386] For patient 2, the mean f.sub.TT value is 0.786 and the standard deviation is 0.035; [0387] For patient 3, the mean f.sub.TT value is 0.840 and the standard deviation is 0.030; [0388] For patient 4, the mean f.sub.TT value is 0.711 and the standard deviation is 0.050; [0389] For patient 5, the mean f.sub.TT value is 0.853 and the standard deviation is 0.044; [0390] For patient 6, the mean f.sub.TT value is 0.776 and the standard deviation is 0.057; [0391] For patient 7, the mean f.sub.TT value is 0.724 and the standard deviation is 0.048; [0392] For patient 8, the mean f.sub.TT value is 0.820 and the standard deviation is 0.070; [0393] For patient 9, the mean f.sub.TT value is 0.658 and the standard deviation is 0.070.

[0394] Because in this example, the patients are heterogeneous in their diseases associated with SCD, in the treatments they receive, in their sex and age, and because for each of them the difference between the mean f.sub.TT value and each f.sub.TT value is less than twice the value of the standard deviation when they are not in crisis, it can be concluded that the f.sub.TT parameter is independent of the characteristics of these patients.

[0395] In addition, these results show that the f.sub.TT parameter is heterogenous and patient-dependent, but overall stable, with few fluctuations over time in the “out of crisis” period for each patient.

[0396] Evolution of the f.sub.TT Parameter in Patients in SCD Vaso-Occlusive Crisis

[0397] The inventors report in FIGS. 16, 17 and 18 the temporal evolution of f.sub.TT values in three patients who suffered vaso-occlusive crises during the period covered by the weekly follow-up study. Time sampling is complementary to that shown in FIG. 4. FIGS. 16 to 18 clearly show that each observed crisis is associated with a significant decrease in the f.sub.TT value during hospitalization.

[0398] For each patient, the mean f.sub.TT value and the associated standard deviation are calculated on the “out of crisis” period. For patient 10, the mean f.sub.TT value is 0,764 and the standard deviation is 0,035. For patient 11, the mean f.sub.TT value is 0,701 and the standard deviation is 0,049. For patient 12, the mean f.sub.TT value is 0.793 and the standard deviation is 0.038.

[0399] The results presented in these figures are in addition to those presented above in “Evolution of f.sub.TT parameter on patients in the “out of crisis” period”, in that when the patient is in an “out of crisis” period, the difference between the f.sub.TT mean and each measured value is less than twice the value of the standard deviation for each patient.

[0400] In FIG. 16, the patient suffered from a first crisis period from day 109 to day 111, followed by a second crisis period begins after day 132 and ends before day 139, during which no measurements were made. The difference between the mean f.sub.TT and the value measured at days 109 and 111 is more than twice the value of the standard deviation.

[0401] In FIG. 17, the patient suffered from a crisis period from day 1 to day 2. Here too, the difference between the f.sub.TT mean and the values measured at day 1 and 2 is more than twice the value of the standard deviation.

[0402] In FIG. 18, the patient suffered from a crisis period from day 28 to day 30. Here again, the difference between the f.sub.TT means and the values measured at day 28, 29 and 30 is more than twice the value of the standard deviation.

[0403] For each patient, the lower f.sub.TT values are consistent with the patient's pain, a sign of the crisis.

[0404] Overall, these results show that the f.sub.TT parameter is very sensitive to vaso-occlusive crises with a significant decrease in the f.sub.TT value during a crisis that resulted in the patient's hospitalization.