Am J Physiol Heart Circ Physiol 288: H584 –H590, 2005.
First published September 30, 2004; doi:10.1152/ajpheart.00690.2004.
Aggregate formation of erythrocytes in postcapillary venules
Sangho Kim,1 Aleksander S. Popel,2 Marcos Intaglietta,1 and Paul C. Johnson1
1
Department of Bioengineering, University of California, San Diego, La Jolla, California; and
Department of Biomedical Engineering, Johns Hopkins University, Baltimore, Maryland
2
Submitted 13 July 2004; accepted in final form 17 September 2004
hemodynamics; in vivo blood rheology; in vivo microscopy
Current knowledge of the process of aggregate formation is
based on in vitro studies using various systems including
transparent cone-and-plate viscometers, glass capillary tubes,
or narrow-gap flow chambers. These instruments make it
possible to visualize the time course and degree of aggregation
formation (1– 4, 12, 14, 31, 33, 34). These investigations report
a wide range of time scales (5–200 s) for aggregate formation,
indicating that the rate of red blood cell aggregation might be
associated with the dimensions and geometry of the experimental system. As a consequence, direct application of these
findings to aggregate formation in venular networks is not
feasible.
To our knowledge, there have been no studies on time
course and process of aggregate formation in vivo. The purpose of the present study, therefore, was to provide such
information. Postcapillary venules were selected for the study
based on the assumption that aggregates would not be present
in capillaries and this would be the first site in the venous
network where aggregation could develop. The study was done
using the spinotrapezius muscle preparation of the rat, an
animal whose blood normally shows negligible red blood cell
aggregation tendency (6). Intravenous infusion of dextran 500
was used to induce aggregation at levels seen in the human.
MATERIALS AND METHODS
RED BLOOD CELL AGGREGATION
is a prominent feature in humans
and other athletic species but not sedentary species (30).
Numerous in vitro studies have shown that aggregation of
blood increases as shear rate decreases. Aggregation also
depends on hematocrit and the concentration of macromolecules in the plasma or suspending medium (7, 10, 15, 16, 19,
25, 32). However, the circumstances in which aggregation
occurs in vivo and its possible importance in circulatory
function are not well understood.
Previous whole organ studies in the dog and cat, in which
red blood cell aggregation normally occurs, have shown that
venous vascular resistance in skeletal muscle increases as
arterial pressure and blood flow rate decrease and provided
evidence that this effect is dependent on red blood cell aggregation (11, 23, 24, 35). From these studies, it was postulated
that aggregate formation increased resistance to flow in the
venular network. Recently, Bishop and co-workers (9) reported
that, in rodents treated with high molecular mass dextran to
induce aggregation, velocity profiles of blood flow in venules
became blunted at low flow rates and red blood cells formed
aggregates, which would explain, in part, the dependence of
venous resistance on flow rate.
Experimental setup. An intravital microscope (Ortholux II, Leitz)
transilluminated with a 100-W mercury lamp (model 1149, Walker
Instruments, Scottsdale, AZ) was used with an Olympus ⫻40 waterimmersion objective and a long working distance condenser (Instec,
Boulder, CO), which have numerical apertures of 0.7 and 0.35,
respectively. Figure 1 shows the schematic diagram of the microscope
system. A blue filter (transmittance peak 400 nm, model 59820,
Spectra-Physics, Stratford, CT) was used to enhance the contrast
between the red blood cells and the background field. Focused images
viewed on a computer monitor were projected onto a high-speed video
camera (model FASTCAM ultima SE, Photron USA) capable of
recording up to 4,450 frames/s and storing up to 24,576 frames in the
internal memory and was viewed on a computer monitor. This
arrangement provided a full-frame field of 200 ⫻ 200-␮m2 view. The
high-speed video camera was connected to and controlled by a
microcomputer (Intel Pentium IV). All the image files stored in the
computer were transferred to a compact disk (CD-Writer Plus 7200e;
Hewlett-Packard).
Animal preparation. Eight male Wistar-Furth rats weighing 162 ⫾
25 g were used for the present study. Animal handling and care were
provided in accordance with the procedures outlined in the Guide for
the Care and Use of Laboratory Animals (National Institutes of
Health, 1996). The study was approved by the local Animal Subjects
Committee. The rats were anesthetized with an intraperitoneal injection of 50 mg/kg pentobarbital sodium (Abbott). Additional anesthetic
Address for reprint requests and other correspondence: P. C. Johnson, Dept.
of Bioengineering, Univ. of California, San Diego, La Jolla, CA 92093-0412
(E-mail: [email protected]).
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
H584
0363-6135/05 $8.00 Copyright © 2005 the American Physiological Society
http://www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.32.247 on June 17, 2017
Kim, Sangho, Aleksander S. Popel, Marcos Intaglietta, and
Paul C. Johnson. Aggregate formation of erythrocytes in postcapillary venules. Am J Physiol Heart Circ Physiol 288: H584 –H590,
2005. First published September 30, 2004; doi:10.1152/ajpheart.
00690.2004.—The purpose of the present study was to obtain information on erythrocyte aggregate formation in vivo. The movements of
erythrocytes in postcapillary venules of the rat spinotrapezius muscle
at various flow rates were recorded with a high-speed video camera
before and after infusion of dextran 500. To distinguish aggregates,
the following criteria were used: 1) a fixed distance (4 ␮m) between
the center points of two adjacent cells, 2) lack of visible separation
between the adjacent cells, and 3) movement of the adjacent cells in
the same direction. Without dextran 500 infusion, 11 and 5% of
erythrocytes formed aggregates in low (33.2 ⫾ 28.3 s) and high
pseudoshear (144.2 ⫾ 58.3 s) conditions, respectively, based on the
above criteria. After dextran 500 infusion, 53% of erythrocytes satisfied the criteria in the low pseudoshear condition (26.5 ⫾ 17.0 s) and
13% of erythrocytes met the criteria in the high pseudoshear condition
(240.0 ⫾ 85.9 s), indicating erythrocyte aggregation is strongly
associated with shear rate. Approximately 90% of aggregate formation occurred in a short time period (0.15– 0.30 s after entering the
venule) in a region 15 to 30 ␮m from the entrance. The time delay
may reflect rheological entrance conditions in the venule.
AGGREGATE FORMATION OF RED BLOOD CELLS IN VIVO
H585
was administered intravenously throughout the experiment as needed.
The animal was placed on a heating pad to maintain body temperature
of 37°C during surgery. A tracheal tube (ID ⫽ 1.57 mm, OD ⫽ 2.08
mm) was inserted to assist breathing, the jugular vein was catheterized
for administration of anesthetic or dextran 500 during the course of
the experiment, and the carotid artery was catheterized to measure
arterial pressure and withdraw blood as needed for pressure reductions. All catheters were filled with a solution of heparinized saline
(30 IU/ml) to prevent clotting. The rat spinotrapezius muscle was
exteriorized and prepared for intravital microscopy as described in
previous studies (8, 9). A temperature probe was placed beside the
muscle so that temperature could be monitored and maintained at
37°C during the experiment by a heating element attached to the
animal platform.
Pressure, hematocrit, and aggregation measurements. Arterial
pressure was continuously recorded using a physiological data-acquisition system (MP 100 System, BIOPAC Systems, Goleta, CA). The
pressure data were transferred and stored in a microcomputer during
the experiment. Blood samples of ⬃0.1 ml were withdrawn from the
carotid artery catheter during an experiment for hematocrit and degree
of red blood cell aggregation measurements, which were taken during
control as well as after infusion of Dextran 500. Hematocrit was
determined after centrifugation with a microhematocrit centrifuge
(Readacrit, Clay Adams). The degree of red blood cell aggregation
was assessed from duplicate measurements on a 0.035-ml blood
sample with a photometric rheoscope (Myrenne Aggregometer,
Myrenne, Roentgen, Germany). The aggregation index (M) obtained
in the present study was based on the 10-s setting.
Adjustment of red blood cell aggregability and hematocrit. Experiments were performed on animals both under normal (nonaggregating) conditions and after induction of red blood cell aggregation with
AJP-Heart Circ Physiol • VOL
dextran 500 (average molecular mass 460 kDa; Sigma) dissolved in
saline (6%) and infused in 50-mg/kg increments over the course of
2–3 min (total 200 mg/kg body wt). This represents a plasma dextran
concentration of ⬃0.6%. Hematocrit and aggregation index values
were determined 15 min after dextran 500 infusion. There was no
discernable adverse reaction (e.g., visible swelling of the limbs) to the
dextran 500 infusion in any of the rats used in these investigations.
To obtain clear and distinct images of both individual red blood
cells and aggregates in the venule, it was necessary to reduce local
hematocrit. This was achieved by reduction of systemic hematocrit
with infusion of autologous plasma obtained from a donor animal. The
blood volume was maintained by exchanging the same amount of
blood with plasma (1–2 ml). Mean arterial pressure was monitored
during and after blood withdrawal and plasma infusion until a steadystate value, which was not significantly different from the control
arterial pressure, was reestablished.
To estimate the tube hematocrit in the postcapillary venules,
symmetric cylindrical vessel geometry was assumed. The dimensions
of the postcapillary venule, the mean red blood cell volume of the rat
(⬃52 ␮m3) (5), and the number of red blood cells present in the
venule at a specific moment were used for the estimation.
Experimental protocol. Before the experiment began, a blood
sample was taken from the carotid artery to obtain control values of
hematocrit and aggregation index. Skeletal muscle postcapillary
venules with two to three supply capillaries in the field of view were
selected for the study on the basis of the criteria of stable flow as well
as clear focus and contrast of the image. The microscope was focused
on the equatorial plane of the venule, and the high-speed video camera
was used to record the movements of red blood cells in both the
capillaries and venule for ⬃30 s. The frame rate was adjusted to
maximize image clarity for the specific flow and background condi-
288 • FEBRUARY 2005 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.32.247 on June 17, 2017
Fig. 1. Schematic diagram of the experimental setup. Details
are described in Experimental setup. The high-speed video is
capable of recording red blood cell movements in both capillaries and postcapillary venules at a wide range of flow rates.
H586
AGGREGATE FORMATION OF RED BLOOD CELLS IN VIVO
Fig. 2. Videomicrograph and schematic diagram of a representative postcapillary venule with 2 supply capillaries. A blue filter was used to enhance the
contrast between the red blood cells and the background field. The microscopic
system provided a full-frame field of 200 ⫻ 200 ␮m2. The postcapillary venule
image shown here was a portion of the field. The contrast of image was
increased for publication.
AJP-Heart Circ Physiol • VOL
Similarly, venular wall position was determined from the transillumination image. However, in the latter case, an averaged value from five
measurements was used to minimize human measurement error. Error
associated with marking the center of red blood cell images on the
image analysis software was estimated to be ⫾0.5 ␮m based on 10
repeated measures on a group of 5 cells.
The mean velocity of each red blood cell was obtained from the
position data of the cells and time interval between two position
measurements. The velocities were averaged and used for the mean
velocity (V) of blood flow in the venule. Pseudoshear rate in the
venule was calculated from the following equation
␥⫽
V
D
(1)
where ␥ is pseudoshear rate and D is average diameter. The average
diameter was determined from five measurements at different locations along the venule using the image analysis software.
Criteria for red blood cell aggregate formation. To determine
whether adjacent red blood cells form an aggregate, the following
criteria were used: 1) a fixed distance (4 ␮m) between the center
points of two adjacent cells based on measurements in stationary
aggregates in vitro, 2) lack of visible separation between the adjacent
cells, and 3) movement of the two adjacent cells in the same direction.
To establish the first criterion, a blood sample was removed from
a dextran-treated animal via the carotid artery into a heparinized
syringe and a drop was placed on a glass slide and covered with a
glass coverslip. The field containing the in vitro red blood cell
aggregates was recorded for ⬃10 s and converted into image files. The
distance between centers of two adjacent red blood cells in selected
aggregates was measured using image analysis software. The mean
distance between centers of two unstressed red blood cells obtained
from the in vitro experiment was 3.4 ⫾ 0.1 ␮m. Bishop and coworkers (8) reported that the minimum width of red blood cells in
50-␮m skeletal muscle venules, representing cell thickness, was ⬃3.7
␮m. Considering the error of human measurement for the cell location
in our study, which was ⫾0.5 ␮m, a value of 4 ␮m or less between
centers of two cells was used as the criterion for aggregate formation.
Shown in Fig. 3 is a schematic diagram of the different cell
configurations for each case (nonaggregating cells, aggregate, or
possible aggregate) we observed in vivo. The distance between the
centers of adjacent cells, as shown in Fig. 3, was directly calculated
from the position data of the cells passing through the postcapillary
venule in the field of view. The aggregate case (see Fig. 3C) consisted
of either two or more cells in parallel with the flow direction or
crescent-shaped cells in tandem. In the case of possible aggregate, a
few unusual cell configurations with two crescent-like or/and ovallike cells (see Fig. 3B) were observed, whereas the configurations
designated as possible aggregates showed similar shapes but larger
center-to-center distances (⬎4 ␮m) compared with apparent aggregates. If there was distinct separation between two adjacent red blood
cells as shown in Fig. 3A, the cells were classified as nonaggregating
cells.
Statistical analysis. A paired t-test was used to determine differences in experimental and physiological parameters between normal
and dextran-treated animals. The degree of aggregation was compared
in four groups as follows: normal arterial pressure, normal arterial
pressure with dextran 500 infusion, reduced flow condition, and
reduced flow condition with dextran 500 infusion. At least two
animals were used for each condition that provided the experimental
data from three different venules. All of the data are reported as
means ⫾ SD. The statistical significance of differences in the aggregate formation of red blood cells was determined by comparing the
percentage of aggregates in each case, using a one-way ANOVA test
with subsequent Scheffé’s, Fisher’s least significant difference, and
Bonferroni tests. All statistical tests were done using a commercially
288 • FEBRUARY 2005 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.32.247 on June 17, 2017
tions. The frame rates of 500 –2,250 frames/s were used in this study.
In general, higher frame rates gave us more distinct edge definition of
red cells but lower contrast with background. Thus the frame rates of
500 –750 frames/s were used for reduced arterial pressures and higher
frame rates were used for normal arterial pressures.
The microcirculatory observations were repeated at reduced flow
rates. For this purpose, blood was removed via the carotid artery into
a heparinized syringe. An average of 1.0 ⫾ 0.5 ml of blood was
withdrawn at a rate of ⬃2 ml/min. After the reduced pressure
stabilized and steady blood flow was reestablished, the video image of
the vessels was recorded for ⬃1 min. The blood withdrawn was
reinfused after the recording, and in all cases, the arterial pressure was
monitored until it returned to a stable value. After completion of the
above procedures, red blood cell aggregation was induced with
infusion of dextran 500 and the experimental protocol described
above was repeated. In some instances, it was necessary to find
another suitable venule due to either unsteady flow or cessation of
flow in one or more of the supply capillaries of the venule used in the
control study.
Determination of red blood cell and wall positions. Digital video
clips stored in the control unit of the high-speed camera were transferred to a hard drive in the microcomputer for analysis and storage.
Using Adobe Premier 5.1, the video files were then converted to
image files (either TIFF or BMP format). Image magnification was
determined from the recorded image of a stage micrometer under
transillumination. Figure 2 shows a videomicrograph of a typical
vessel studied under transillumination (top) and a schematic diagram
of the vessel (bottom), respectively.
From an image such as shown in Fig. 2, top, x-y coordinate data for
each red blood cell image were obtained using an image analysis
software package (SigmaScan Pro 4.0, SPSS). Depending on the flow
velocity and framing rate, every fifth to fiftieth frame was chosen for
the determination of instantaneous positions of each red blood cell
coming from capillaries and passing through the postcapillary venule.
A total of 12 postcapillary venules were used in the present study and
for each venule, 24 –30 red blood cells were selected and analyzed.
AGGREGATE FORMATION OF RED BLOOD CELLS IN VIVO
Fig. 3. Typical shapes of red blood cells observed in the study for 3 cases. A:
nonaggregating cells. B: possible aggregates. C: aggregates. The arrows shown
between 2 red cells indicate the center-center distance between the cells.
RESULTS
Hematocrit, degree of aggregation, and arterial pressure.
As noted in MATERIALS AND METHODS, for better observation of
aggregation formation, normal systemic hematocrit (42.4 ⫾
2.1%) was reduced by hemodilution with plasma. The reduced
hematocrit in the control period (27.1 ⫾ 2.4%) was not
significantly different from that after dextran treatment (27.7 ⫾
3.2%, P ⬎ 0.05). The mean diameter and length of postcapillary venules selected in the present study were 12.0 ⫾ 1.4 and
83.1 ⫾ 35.7 ␮m, respectively. Based on these values, the tube
hematocrit was 9.4 ⫾ 2.1% for the 12 postcapillary venules
used in the study. The thickness of the endothelial glycocalyx
was not considered in this calculation but its possible effect is
considered in DISCUSSION.
Under control conditions for normal rats, the index of
aggregation, M, was 0.0 in all cases, and the mean arterial
pressure was 113.3 ⫾ 5.8 mmHg. The reduced mean arterial
pressure for normal rats was 56.3 ⫾ 17.4 mmHg. There were
no significant differences (P ⬎ 0.05) between the mean arterial
pressures of normal and dextran-treated animals during the
control and reduced flow situations. In dextran-treated rats, the
index of aggregation was 16.9 ⫾ 3.9 and the arterial pressures
were 110.0 ⫾ 8.7 and 55.7 ⫾ 9.0 mmHg during control and
reduced flow situations, respectively. No significant difference
(P ⬎ 0.05) in the degree of aggregation was found between
normal and reduced flow conditions in dextran-treated rats.
Time course of red blood cell aggregate formation. The time
course of red blood cell aggregate formation in a postcapillary
venule is shown in Fig. 4. A sample group of eight red blood
cells in two capillaries at the entrance region of the venule was
identified (see Fig. 4A) and monitored over time. Once the
individual cells entered the venule, as shown in Fig. 4, B–D,
the separation between red blood cells tended to decrease and
some cells formed aggregates. After red blood cells formed
aggregates, the adherent cells moved together in the same
direction. The distance between the red blood cells that formed
an aggregate was monitored until the aggregate moved out of
focus, which was usually ⬃80 ␮m from the entrance.
As shown in Fig. 4, the boundaries of individual red blood
cells were distinct before they formed aggregates but became
indistinct in the region between adjoining cells once aggregates
formed. If the adjacent cells maintained a distance ⬍4 ␮m and
moved in the same direction with no distinct separation in
between, the two (or more) cells were defined as an aggregate.
AJP-Heart Circ Physiol • VOL
Distance between the centers of two adjacent red blood cells
in an aggregate. A total of 306 red blood cells were analyzed.
Between 72 and 78 red blood cells were investigated for each
of four conditions: nondextran condition at control and reduced
arterial pressure and dextran treatment at control and reduced
arterial pressure. Figure 5 shows the distance changes between
the centers of two adjacent red blood cells over time for three
cases: aggregate, possible aggregate, and nonaggregating cells.
Note that the distance measurement covers the period in which
the red blood cells were flowing in capillaries as well as in a
postcapillary venule as shown in Fig. 5. The distance data
indicate that in two cases the red blood cells came from two
different capillaries and in one case from the same capillary.
For the case in which an aggregate formed, the distance
between centers of the two cells entering from two supply
capillaries decreased gradually and eventually stabilized below
4 ␮m in the postcapillary venule. The two cells tumbled after
aggregate formation and the center-to-center distance varied
between ⬃2 and 4 ␮m. In the possible aggregate case, the
distance fluctuated around 4 ␮m but did not stay below this
value while the other criteria were met. For the case of
nonaggregating cells, the distance between the cell centers was
always larger than 4 ␮m.
Red blood cell aggregation in postcapillary venules. As
shown in Fig. 6, without dextran 500 infusion, 11 and 5% of
red blood cells formed aggregates in low (␥ ⫽ 33.2 ⫾ 28.3 s)
and high pseudoshear (␥ ⫽ 144.2 ⫾ 58.3 s) conditions,
respectively, based on the criteria. With dextran 500 infusion,
53% of red cells satisfied the criteria in the low pseudoshear
condition (␥ ⫽ 26.5 ⫾ 17.0 s), whereas 13% of red cells met
the criteria in the high psuedoshear condition (␥ ⫽ 240.0 ⫾
85.9 s), indicating red cell aggregation is highly associated
with shear rate. The percentages of aggregating and nonaggregating cells with dextran treatment at reduced arterial pressure
were significantly different (P ⬍ 0.01) from those for any of
the other three conditions. The latter were not significantly
different (P ⬎ 0.05) from each other.
Fig. 4. Time history of red blood cell aggregate formation in a postcapillary
venule. Representative videomicrographs of 13.7-␮m postcapillary venule at
t ⫽ 0.00 s (A), t ⫽ 0.10 s (B), t ⫽ 0.20 s (C), and t ⫽ 0.32 s (D) after dextran
treatment at a pseudoshear rate of ⬃10 s. The dotted lines indicate the walls of
blood vessels. The red cell images shown here were also enhanced.
288 • FEBRUARY 2005 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.32.247 on June 17, 2017
available software Package (SPSS 11.0 for Windows). For all tests,
P ⬍ 0.05 was considered statistically significant.
H587
H588
AGGREGATE FORMATION OF RED BLOOD CELLS IN VIVO
rotation in 12% of aggregates and 14% of individual cells in
the ⬃70-␮m length of the venule we observed.
DISCUSSION
Rate of red blood cell aggregate formation. Figure 7 shows
red blood cell aggregate formation as a function of time and
position in postcapillary venules. In this analysis, the aggregation data used were only for dextran treatment at reduced
arterial pressure. There was no red blood cell aggregate formation at the immediate entrance region of postcapillary
venules. Aggregate formation began rather suddenly at ⬃30
ms elapsed time when the distance of the red blood cell from
the entrance was 15 ␮m and then increased rapidly as the
elapsed time increased to ⬃100 ms and the distance traveled
increased to 22 ␮m. In fact, ⬃90% of the 17 aggregates we
observed were formed in the region between 15 and 30 ␮m of
the entrance to the postcapillary venule and it took less than
0.3 s to form all aggregates. In the remaining 30-␮m segment
we were able to visualize, the rate of aggregate formation was
very low despite the fact that ⬃50% of the red blood cells were
still in the nonaggregated state.
The open triangle symbols in Fig. 7 represent the larger
aggregates, showing that ⬃50% of total aggregates were
formed by three or more red blood cells. In general, aggregates
were first formed by two cells joining. As those binary aggregates moved downstream, six became larger by individual red
blood cells or aggregates joining the binary aggregate. In two
instances, two binary aggregates coupled to each other, and in
one instance, three cells combined simultaneously to form one
large aggregate. The largest aggregate observed in this study
consisted of five red blood cells. Rotary motion (tumbling) was
observed in both aggregates and individual red cells. Although
the speed of such motion could not be quantified, we observed
AJP-Heart Circ Physiol • VOL
Fig. 6. Experimental results of red blood cell aggregate formation in postcapillary venules based on the 3 criteria. Values are means ⫾ SD; n ⫽ 3 in each
group. *P ⬍ 0.01.
288 • FEBRUARY 2005 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.32.247 on June 17, 2017
Fig. 5. Distance changes between the centers of 2 adjacent red blood cells over
time for 3 cases: aggregate, possible aggregate, and nonaggregating cells. Data
points in the gray area represent red blood cells in capillaries, whereas other
data points are from red blood cells in a postcapillary venule. Dotted line
indicates 4 ␮m, one of criteria used in the present study.
The purpose of the present study was to provide information
on the time course and process of red blood cell aggregate
formation in vivo. To our knowledge, this is the first such
study. Postcapillary venules were chosen for the study because
this is the first site in the venous network where aggregation
could develop. No red blood cell aggregates were observed in
supply capillaries. We show that most of the aggregate formation occurs in a region 15–30 ␮m from the entrance to
postcapillary venules. It appears that a much shorter time
period is required for red blood cells to form aggregates in vivo
compared with those commonly reported for in vitro studies
(5–200 s) (1– 4, 12, 14, 31, 33, 34).
The major source of uncertainty in distinguishing aggregates
based on the criteria presented in MATERIALS AND METHODS was
in determining the center-to-center distance of 4 ␮m. The
distance measurement was based on the horizontal plane coordinate, although actual distance should include the z coordinate as well. Therefore, this measurement might underestimate
the actual separation between two red blood cells at different
levels moving with the same velocity in a postcapillary venule.
To minimize error due to overlap of the erythrocyte images,
hematocrit was reduced to a level at which the images of red
blood cells were distinct.
The degree of red cell aggregability induced by high molecular mass polymers is dependent on hematocrit. It has been
AGGREGATE FORMATION OF RED BLOOD CELLS IN VIVO
pointed out that the endothelial glycocalyx, which is not
observable with light microscopy, could alter local hematocit
and change flow patterns in small vessels such as postcapillary
venules and capillaries (17, 28). In the present study, assuming
a 0.5-␮m-thick glycocalyx in the venule as suggested by
Damiano and co-workers (17, 26), the tube hematocrit would
be 11.3 ⫾ 2.6%, a 20% increase compared with the estimated
value of 9.4 ⫾ 2.1% without a glycocalyx.
It is well known that red blood cell aggregates are formed in the
presence of long-chain macromolecules such as dextran 500. In
vitro studies on the formation of aggregates show a strong relationship between aggregate size and shear rate (4, 8, 22). Goldsmith and Marlow (22) reported that, in suspensions of ⬍2%
hematocrit, only single cells were observed at shear rates above
⬃50 s, whereas large aggregates with 10 –20 red blood cells were
seen at shear rates below ⬃4 s. More recently, Yedgar and
co-workers (12) found that red blood cells (hematocrit ⫽ 10%)
formed small aggregates with three to four cells at a shear rate of
⬃70 s. Bishop et al. (8) found that aggregates were two to three
times wider than the diameter of a single red blood cell at the low
pseudoshear rates tested, less than ⬃10 s, whereas at pseudoshear
rates above ⬃50 s the average width of individual particles was
not significantly different from the red blood cell diameter. These
previous results agree with the results shown in Fig. 5, supporting
the criteria for aggregate formation.
AJP-Heart Circ Physiol • VOL
The time required to begin aggregate formation reported
from in vitro studies is longer than we report and varies with
the system used. In previous studies, red blood cell aggregation
was observed following a sudden change in shear rate, which
might require a longer time period for flow stabilization (21,
36), compared with steady flows at low shears in the present
study. Alonso and co-workers (1–3) found that cell-free plasma
spaces resulting from red cell aggregation started to develop 5
to 10 s after sudden reduction of flow rate in straight capillary
tubes (25- to 100-␮m ID). Chen and co-workers (12, 14)
observed no significant change of particle sizes in human blood
in the first 10 –20 s after stopping flow in a 40-␮m gap flow
chamber (hematocrit ⫽ ⬃10%) and a sudden development of
aggregation after that.
In a study of whole blood viscosity with a Contraves
viscometer, red cell aggregates in human blood (hematocrit ⫽
47%) appeared to form in ⬃2 s with shear rate reduction to
0.7 s (13). However, it may take less time to form binary
aggregates like those observed in our study because the shear
rate of 0.7 s used in that study was low enough to form large
three-dimensional aggregates. Hematocrit is also important.
Shiga and co-workers (33, 34) found that the development of
one-dimensional short rouleaux (composed of 2 or 3 red blood
cells) required longer than 150 s at a steady shear rate of 7.5 s
in the transparent cone-and-plate viscometer with a hematocrit
of 0.25%. The much shorter time required for red blood cell
aggregation to begin in the present study (30 ms) and reach
completion (0.3 s) likely reflects the higher hematocrit and/or
smaller scale of the system than in previous in vitro studies.
An important finding of our study, as shown in Fig. 7, is the
lack of red blood cell aggregate formation at the entry region
15 ␮m from the branching point of postcapillary venules. Das
and co-workers (18) performed a numerical study of blood
flows in small venular bifurcations and showed that the velocity profiles became fully developed ⬃1.5 diameters downstream from the bifurcation. Therefore, it appears that the
absence of aggregation in the entrance region reflects the
distance required for convergence of the streamlines that brings
the red blood cells close together.
Another significant finding is that aggregation was essentially complete 30 ␮m from the branch point, although 47% of
the cells had not formed aggregates. Because the tube hematocrit for the venules used in this study was relatively low
(9 –11% depending on assumed glycocalyx width), not all cells
would come in contact with another cell. The age of the red
blood cells could be involved. Meiselman (27) demonstrated
that relatively young red blood cells had approximately three
times lower aggregability compared with older cells.
A few aggregates (5 and 11%, respectively, of total cells)
according to our criteria were formed in postcapillary venules
of normal rats at control and reduced shear rates, although the
aggregability (M) of the blood measured with the Myrenne
Aggregometer was zero. It is possible that these aggregates
represent short-lived red cell clusters brought in close contact
solely by hemodynamic forces as suggested by Osterloh et al.
(29). They reported a mean particle size considerably greater
than that of red cells at high shear rates in pre- and postcapillary vessels (⬃25 ␮m) using a spatial Fourier analysis of light
intensity patterns. These clusters are not linked together by
mechanisms engendered by macromolecules.
288 • FEBRUARY 2005 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.32.247 on June 17, 2017
Fig. 7. Aggregate formation of red blood cells as a function of time and
position in postcapillary venules (11.2 ⫾ 2.3 ␮m). F Indicate the percentage of
red cells in all aggregates, whereas ‚ represent the percentage of larger
aggregates (3–5 cells).
H589
H590
AGGREGATE FORMATION OF RED BLOOD CELLS IN VIVO
ACKNOWLEDGMENTS
The authors thank Dr. P. Cabrales, K. Flores, S. Dunning, J. H. Park, and
R. Pang for technical assistance.
GRANTS
This work was supported by National Heart, Lung, and Blood Institute
Grants HL-52684, HL-62354, and HL-64395.
REFERENCES
1. Alonso C, Pries AR, and Gaehtgens P. Time-dependent rheological
behaviour of blood flow at low shear in narrow horizontal tubes. Biorheology 26: 229 –245, 1989.
2. Alonso C, Prices AR, and Gaehtgens P. Time-dependent rheological
behavior of blood at low shear in narrow vertical tubes. Am J Physiol
Heart Circ Physiol 265: H553–H561, 1993.
3. Alonso C, Prices AR, Kiesslich O, Lerche D, and Gaehtgens P.
Transient rheological behavior of blood in low-shear tube flow: velocity
profiles and effective viscosity. Am J Physiol Heart Circ Physiol 268:
H25–H32, 1995.
4. Barshtein G, Wajnblum D, and Yedgar S. Kinetics of linear rouleaux
formation studied by visual monitoring of red cell dynamic organization.
Biophys J 78: 2470 –2474, 2000.
5. Baskurt OK. Deformability of red blood cells from different species
studied by resistive pulse shape analysis technique. Biorheology 33:
169 –179, 1996.
6. Baskurt OK, Farley RA, and Meiselman HJ. Erythrocyte aggregation
tendency and cellular properties in horse, human, and rat: a comparative
study. Am J Physiol Heart Circ Physiol 273: H2604 –H2612, 1997.
7. Baskurt OK, Meilselman HJ, and Kayar E. Measurement of red blood
cell aggregation in a “plate-plate” shearing system by analysis of light
transmission. Clin Hemorheol Microcirc 19: 307–314, 1998.
8. Bishop JJ, Nance PR, Popel AS, Intaglietta M, and Johnson PC.
Relationship between erythrocyte aggregate size and flow rate in skeletal
muscle venule. Am J Physiol Heart Circ Physiol 286: H113–H120, 2004.
9. Bishop JJ, Nance PR, Popel AS, Intaglietta M, and Johnson PC. Effect
of erythrocyte aggregation on velocity profiles in venules. Am J Physiol
Heart Circ Physiol 280: H222–H236, 2001.
AJP-Heart Circ Physiol • VOL
10. Boynard M and Lelievre JC. Size determination of red blood cell
aggregates induced by dextran using ultrasound backscattering phenomenon. Biorheology 27: 39 – 46, 1990.
11. Cabel M, Meiselman HJ, Popel AS, and Johnson PC. Contribution of
red blood cell aggregation to venous vascular resistance in skeletal muscle.
Am J Physiol Heart Circ Physiol 272: H1020 –H1032, 1997.
12. Chen S, Barshtein G, Gavish B, Mahler Y, and Yedgar S. Monitoring
of red blood cell aggregability in a flow-chamber by computerized image
analysis. Clin Hemorheol 14: 497–508, 1994.
13. Chen S, Dormandy J, Ernst E, and Matrai A. Clinical Hemorheology:
Biophysics. Hingham, MA: Kluwer Academic, 1987, chapt. 2.
14. Chen S, Gavish B, Zhang S, Mahler Y, and Yedgar S. Monitoring of
erythrocyte aggregate morphology under flow by computerized image
analysis. Biorheology 32: 487– 496, 1995.
15. Cloutier G, Qin Z, Durand L, and The BG. Power Doppler ultrasound
evaluation of the shear rate and shear stress dependences of red blood cell
aggregation. IEEE Trans Biomed Eng 43: 441– 450, 1996.
16. Cloutier G and Shung KK. Study of red cell aggregation in pulsatile flow
from ultrasonic Doppler power measurements. Biorheology 30: 443– 461,
1993.
17. Damiano ER. The effect of the endothelial-cell glycocalyx on the motion
of red blood cells through capillaries. Microvasc Res 55: 77–91, 1998.
18. Das B, Enden G, and Popel AS. Stratified multiphase model for blood
flow in a venular bifurcation. Ann Biomed Eng 25: 135–153, 1997.
19. Fontaine I, Savery D, and Cloutier G. Simulation of ultrasound backscattering by red cell aggregates: effect of shear rate and anisotropy.
Biophys J 82: 1696 –1710, 2002.
20. Fung YC. Biomechanics: Mechanical Properties of Living Tissues, 2nd
Edition. New York: Springer, 1993.
21. Gaspar-Rosas A and Thurston GB. Erythrocyte aggregate rheology by
transmitted and reflected light. Biorheology 25: 471– 487, 1988.
22. Goldsmith HL and Marlow J. Flow behavior of erythrocytes. I. Rotation
and deformation in dilute suspensions. Proc R Soc Lond B Biol Sci 182:
351–384, 1972.
23. House SD and Johnson PC. Microvascular pressure in venules of skeletal
muscle during arterial pressure reduction. Am J Physiol Heart Circ Physiol
250: H838 –H845, 1986.
24. Johnson PC and Hanson KM. Effect of arterial pressure on arterial and
venous resistance of intestine. J Appl Physiol 17: 503–508, 1962.
25. Kim SY, Miller IF, Sigel B, Consigny PM, and Justin J. Ultrasonic
evaluation of erythrocyte aggregation dynamics. Biorheology 26: 723–
736, 1989.
26. Long DS, Smith ML, Pries AR, Ley K, and Damiano ER. Microviscometry reveals reduced blood viscosity and altered shear rate and shear
stress profiles in microvessels after hemodilution. Proc Natl Acad Sci USA
101: 10060 –10065, 2004.
27. Meiselman HJ. Red blood cell role in RBC aggregation: 1963–1993 and
beyond. Clin Hemorheol 13: 575–592, 1993.
28. Mulivor AW and Lipowsky HH. Role of glycocalyx in leukocyteendothelial cell adhesion. Am J Physiol Heart Circ Physiol 283: H1282–
H1291, 2002.
29. Osterloh K, Gaehtgens P, and Pries AR. Determination of microvascular flow pattern formation in vivo. Am J Physiol Heart Circ Physiol 278:
H1142–H1152, 2000.
30. Popel AS, Johnson PC, Kameneva MV, and Wild MA. Capacity for red
blood cell aggregation is higher in athletic mammalian species than
sedentary species. J Appl Physiol 77: 1790 –1794, 1994.
31. Schmid-Schönbein H, Gaehtgens P, and Hirsch H. On the shear rate
dependence of red cell aggregation in vitro. J Clin Invest 47: 1447–1454,
1968.
32. Sennaoui A, Boynard M, and Pautou C. Characterization of red blood
cell aggregate formation using an analytical model of the ultrasound
backscattering coefficient. IEEE Trans Biomed Eng 44: 585–591, 1997.
33. Shiga T, Imaizumi K, Harada N, and Sekiya M. Kinetics of rouleaux
formation using TV image analyzer. I. Human erythrocytes. Am J Physiol
Heart Circ Physiol 245: H252–H258, 1983.
34. Shiga T, Imaizumi K, Maeda N, and Kon K. Kinetics of rouleaux
formation using TV image analyzer. II. Rat erythrocytes. Am J Physiol
Heart Circ Physiol 245: H259 –H264, 1983.
35. Thulesius O and Johnson PC. Pre- and postcapillary resistance in
skeletal muscle. Am J Physiol 210: 869 – 872, 1966.
36. Thurston GB. Advances in Hemodynamics and Hemorheology: Viscoelastic Properties of Blood and Blood Analogs. London: Jai, 1996,
chapt. 1.
288 • FEBRUARY 2005 •
www.ajpheart.org
Downloaded from http://ajpheart.physiology.org/ by 10.220.32.247 on June 17, 2017
Approximately 13% of cells met the criteria for aggregation at
the high pseudoshear rate after dextran treatment. Not surprisingly, these aggregates formed further downstream, between 40
and 55 ␮m from the entrance to the postcapillary venule. Osterloh
and coworkers (29) found that the pattern size of particles in preand postcapillary vessels increased at pseudoshear rates both
higher and lower than ⬃150 s after infusion of dextran 250 (800
mg/kg body wt). The presence of aggregation at high pseudoshear
rates in their study and ours may be explained not only by the
hemodynamic forces mentioned above but also by the macromolecules of dextran once the hemodynamic forces bring red blood
cells into close contact.
Previous studies have shown that the formation of red cell
aggregates in the venules contributes significantly to the pressure drop in that bed, thus influencing the overall energy
dissipation in the venous circulation (9, 11). In addition to the
effect of increased particle size on velocity profiles, the tumbling of aggregates disturbs the flow and leads to the dissipation of energy, resulting in increased blood viscosity at low
shear rates (20). We observed in this study that a few (12%) of
the aggregates tumbled. However, the distance that the aggregates could be observed in the postcapillary venules after
aggregate formation was short, and this might have contributed
to the low percentage of tumbling aggregates observed. It
appeared that the tumbling motion of aggregates or individual
red cells was very slow compared with their movements in a
flow direction.