Kiinische Wochenschrift

Klin. Wschr. 54, 149-157 (1976)

© by Springer-Verlag 1976

OBERSICHTEN Red Cell Aggregation in Blood Flow * I. New Methods of Quantification H. Schmid-Sch6nbein**, G. Gallasch, J.v. Gosen, E. Volger, and H.J. Klose Physiologisches Institut der Universit~it Miinchen, 8000 M/inchen 2

Erythrocytenaggregation und Blutstrdmung. L Neue Quantifizierw~gsmethoden Zusammenfassung. Das rheologische Verhalten von normalen und pathologischen Erythrocytenaggregaten in viskometrischer Str6mung (kiinstliche Str6mung in Platte-Kegel-Kammer) wird mittels direkter mikroskopischer Beobachtung (ira ,,Rheoskop"), Viskometrie nnd Photometrie untersucht. Erhebliche Unterschiede finden sich bei dem mikrorheologischen Verhalten, wfihrend im makrorheologischen Verhalten (Blut-Viskometrie) nur diskrete und oft komptexe Differenzen zwischen normalen und pathologischem Blut gemessen werden. Sowohl im normalen wie im pathologischen Blut ist die Erythrocytenaggregation ein reversibler Prozet3, vorausgesetzt, dab ausreichende Scherkr~ifte vorhanden sind. Der Einflug der normalen und pathologischen Aggregate ist eine komplexe Funktion des Schergrades, der Scherzeit, des Hfimatokritwertes und der Plasmaviskositfit. Pathologische Erythrocytenaggregate bilden sich schneller und extensiver aus als normale Aggregate. Sie haben oft eine Tendenz, bei langsamer Str6mung zu wachsen und sind hochgradig scherungsresistent: Schliisselw6rter." Blutrheologie, arterielle VerschluBkrankheiten, Diabetes mellitus, Myelom, photometrische Aggregometrie, Rouleaux-Bildung, Rouleanx-Klumpen, Rouleaux-Netzwerke, scheinbare Blutviskositfit, Schwangerschaft.

Summary. The rheological behavior of normal and pathological red cell aggregates in viscometric flow (artificial flow in cone plate chamber) is studied by direct microscopy, (rheoscopy) viscometry and photometry. Marked differences between normal and pathological blood are measured in the microrheological properties of red cell aggregates; only discreet differences are measured by blood viscometry (macrorheology). Both in normal and abnormal blood, red cell aggregation is a reversible process in the presence of adequate shear fmces; their respective influences on apparent blood viscosity at low rates of shear are complex functions of shear rate, shear time, hematocrit and plasma viscosities. Pathological red cell aggregation (RCA) forms more rapidly and extensivelythan normal RCA. The pathological aggregates frequently have a tendency to grow at low rates of shear and they are highly shear resistant. Key words: Apparent blood viscosity, arterial vascular disease. blood rheology, diabetes mellitus, myeloma, photometric aggregometry, pregnancy~ rouleaux clumps, rouleaux formation, rouleaux networks. * Presented in parts as an invited lecture during the 1st International Congress of Biorheology, Lyon (France) September I972. The lecture was part of a session dedicated to the memory of the late Professor Aharon Katschalsky. ** Supported by Deutsche Forschungsgemeinschaft: Present address: Abtlg. Physiologie der RWTH Aachen, 51 Aachen, Melatenerstr. 211 --213.

Introduction The p h e n o m e n o n o f red b l o o d cell aggregation is the basis o f the oldest medical test o f blood in vitro. The medieval practice o f " v e n a e sectio" was based not only on the theoretical assumption o f a " d y s c r a s i a o f the b l o o d " but rather on the detailed observation o f withdrawn blood. W h e n e v e r sedimentation occurred before clotting, the " c r u s t a i n f l a m m a t o r i a " f o r m e d above the red cells and was p r o o f to the physician that there was a disturbance in the distribution o f the four " h u m o r s " . R o b i n F a h r a e u s [7, 8], w h o rediscovered this test and its rationale half a century ago, introduced the m e a s u r e m e n t o f the sedimentation rate as a measure o f the suspension o f anticoagulated blood, This test is since then uncontested as the m o s t p o p u l a r diagnostic screening test in clinical medicine. P r o b a b l y observed already by the founders o f intravital m i c r o s c o p y (c.f. 5), the in vivo clumping o f cells [or blood " s l u d g i n g " [14] has been seen regularly in all cases o f pathologically disturbed microcirculation, e.g. in all inflammatory, toxic, anaphylactic and h e m o r h a g i c reactions o f small blood vessels in experimental animals and in h u m a n patients.

Basic Blood Rheology The understanding o f the complex flow behavior o f blood, a concentrated suspension o f red cells in plasma, has m a d e remarkable process during the last decade. Extending concepts initiated by Hess [12] and by B i n g h a m [2], it is n o w accepted that the " a n o m a l o u s " viscosity o f h u m a n blood is based on the bipotential flow properties o f the erythrocyte. This cell m u s t be regarded [10] as a m e m b r a n e shell, filled incompletely with a fluid (primarily hemoglobin solution) o f low viscosity. In rapid blood flow, the red celt behaves m u c h like a fluid d r o p : it is oriented, participates in flow and thereby minimizes the viscous energy dissipation of b l o o d (for a review see Dintenfass [5], Chien [4] and Schmid-Sch6nbein [18]). In the pres-

150

H. Schmid-Sch6nbein et al. : Red Cell Aggregation in Blood Flow. I

ence of high molecular weight plasma proteins, any flow retardation leads to a reversible red cell aggregation. The red cells adhere and form primary rouleaux; these in turn form secondary structures. Such threedimensional red cell-rouleaux-networks under conditions of prestatie flow greatly increase the apparent viscosity of bulk blood (v.i.). This rise in viscosity is as reversible as the cell adhesion; nevertheless, aggregation phenomena, when occurring in the retarded microcirculation, may have very significant hemodynamic consequences. If, for example, such aggregates straddle on a bifurcation of a microscopic blood vessel, they can make blood flow impossible. Consequently, uncoagulated blood assumes the properties of a solid, i.e. it can withstand finite forces without flowing or yielding. The described phenomena apply for normal human blood taken from healthy subjects; and, in principle, for blood from patients with intensified aggregation as well. However, this intensified aggregation cannot be established unequivocally by even the most sophisticated rotational viscometers available to date. This has many reasons that have been discussed in detail elsewhere [20]. The most important are: 1. in most cases, intensified aggregation is accompanied by simultaneous increases in plasma viscosity 2. the influence of even slight differences in hematocrit level excels all other effects and 3. phase separation between plasma and cell aggregates occur rapidly and quite extensively, so that meaningful measurements of blood viscosity are often impossible.

Quantitative Aggregometry The relative ineffectiveness of viscometric methods has prompted us to develop new methods. Firstly, a counterrotating "rheoscope" chamber [21] allows microscopic observation of the flowing blood at high magnification. Secondly, photometric methods were employed [13, 19, 23]. They are based on the well known phenomenon that the optical density of a particulate suspension can be decreased either by orientation of particles (e.g. by oriented red cells in high flow) or by aggregation or "flocculation" of particles (rouleaux and aggregates of rouleaux) at slow flow. Similar photometric techniques have also been employed by Brinckman et aI. [3], Dognon [6], Ofstadt [16], Usami [24], and Berman [1], however, the most versatile instrument was introduced by Klose [13]. The effects of aggregation, desaggregation and cell alignment on the optical properties of blood are recorded. It became possible to measure both rate and extent of aggregate formation

as well as aggregate dispersion from the photometric records and correlate these to the incident rate of shear (zero to about 500 sec- 1). With the help of these techniques in concert, the kinetics of artificial red cell aggregation (induced by dextrans of various molecular weights [26], polyalbumin [25], and of fibrinogen, ~2-globulin and IgM) have already been explored [20]. The present report deals with the in vitro flow behavior of red cell aggregates in blood from three patient groups (myocardial infarction, diabetes, myeloma) in which strong intravascutar red cell aggregation ("blood sludging") has notoriously been observed [11, 14] and which frequently present a reduced suspension stability of their blood in vitro. This pathologically intensified aggregation was compared to the "physiological" aggregation. The complex flow behavior even under simplified flow conditions is not easily dealt with in numerical terms and can best be seized by microcinematography. 1 A rather rigid experimental protocol was therefore chosen which centers on four rheological situations: 1. Aggregate formation in stasis, 2. Aggregate theology at slow flow (7 sec- ~), 3. Aggregate dispersion, 4. Flow behavior following dispersion. Methods and Materials

The flow behavior of red cell aggregates was studied in human, venous blood, anticoagulated with EDTA (1.5 rag/l) and obtained from healthy volunteer subjects (students and members of the staff aged 17 to 58 years) as well as from patients. Blood from myeloma patients was withdrawn in the 1. Dept. of Medicine, University of Munich, Germany (courtesy of Dr. Fateh Mogadahm), blood from diabetics was collected at the Krankenhaus Munich-Schwabing (courtesy of Prof. Dr. H. Mehnert). All blood samples were examined within 4 hours after withdrawal ; viscosity was measured both as taken from the patient or after careful adjustment of the hematoerit level to 45% (unless stated otherwise). This was done after centrifugation and removal of the buffy coat by appropriate removal or addition of plasma. Without hematocrit adjustment, the results of normal and abnormal blood cannot be compared and a detailed microrheological analysis is meaningless. The microhematocrit method was used, no corrections for trapped plasma were made. In addition, the same experiments were carried out on defibrinated blood: red cells were washed once 1 A16 mmcotourcinefilmshowingthemostpertinent featuresofred cell aggregation was shown during the conference and can be borrowed upon request.

H. Schmid-Schrnbein etat. : Red Cell Aggregation in Blood Flow. I

in isotonic, buffered saline and were resuspended in the serum of the same patient or subject. The apparent viscosity of the blood samples and plasma was measured in the conventional Wells-Brookfield cone-plate viscometer (23-230 sec- ~) as well as in modified instrument with Couette-attachment (8 - 160 sec- ~) and guard-ring as described elsewhere [22]. The rheology of red cell aggregation in defined viscometric flow at shear rates between 7 and 460 sec- ~ was examined photometrically in a transparent coneplate chamber described elsewhere [19, 21], in which the light transmission is measured as a function of shear rate. For details of the rather complex theoretical basis of this method the reader is referred to the original papers [13, 19, 23]. In a similar, counterrotating transparent chamber (Schmid-Sch6nbein [21] the flow behavior of red blood cells was examined microscopically at magnifications between 4 0 - 6 0 0 x, at various shear rates ( 7 - 5 0 0 x); these observations were also correlated to measurements of the light transmission [21].

Data Processing The transparency of red cell suspensions was shown [13, 19] to have a minimum whenever the flowing cells

151

are dispersed and aligned in random fashion. Transparency increases when the cells are either oriented (in high shear flow) or aggregated (in low shear flow or in stasis). As human red blood cells are physiologically aggregated at low rates of shear, shear forces must be applied to disperse hydrodynamically the red blood cells. The forces necessary to produce the dispersed state of minimum transparency (rr~i,) is therefore a direct measurement of the adhesive forces acting between red cell aggregates: these increase substantically in many diseased states. Besides the determination of the shear rate (Trmi,) and shear stress (~Tm~.) of hydrodynamic desaggregation, the following values were taken. The light transmission (measured as photovoltage) after 1 rain flow at 7 sec- t and in stasis (normalized by the minimum light transmission (T~,) at the shear rate of hydrodynamic dispersion), was taken as a measure of the extent of aggregation at slow flow (TTr.~) and in stasis (To~e0, respectively. Similarly, the light transmission at 460 sec- 1 T46or~0 was taken as a measure of cell orientation and alignment in flow [13, 19] (Table 1). The velocity of aggregate formation is measured by recording the increase in photovoltage (dV) as a

Table 1. Quantification of the behavior of red cell aggregation (RCA) in vicsometlSc flow by transmission photometry

1. "orientation transparency"

A. Static Parameters a photovoltage corresponding to viscometric flow at 460 sec- 1

T46o~o~-l[mV]

2. transparency at '° random cell flow"

Minimum photovoltage measured in any blood sample in viscometric flow

Tm~[mV]

3. "aggregation transparency" in slow flow

photovoltage corresponding to viscometric flow at 7 sec- 1, measured 60 sec following switch from rapid flow (460 sec- i)

rT~oc-l[mV]

4, "aggregation transparency" in stasis

photovoltage in the static sample measured 60 sec following switch from rapid flow

To~oo-4mV]

1. shear resistance of RCA

B. Kinetic Parameter a) shear rate of hydrodynamic desaggregation = shear rate corresponding to random cell flow

,yTmin[sec 1]

b) shear stress of hydrodynamic desaggregation. Viscometrically determined shear stress corresponding to ~Tmin

ZTrainldyn" cm- 2] ti/2[sec]

aggregation constant

a) half time of the decay of the first derivative of dV/dt = i [36] where 1= ]0" e- ~ at 0 sec-1 In2 b) 2 = - - -

corrected aggregation constant = potency to aggregation in stasis

c) 2 multiplied by the viscosity of the continuous phase = 2t/0

2- tto [dyn. c m 2]

3. velocity of shear induced aggregate with growth at 7 sec-1

increase in photovoltage in unit time during flow at 7 sec - 1 normalized by T,,i~ (AT7% Tmi,/min)

% Train/rain

4. ratio indicating shear induction (T7 > To) or shear dispersion (7"7 < To) of R C a

transparency after 1 min of flow (7 sec- 1 TT) divided by transparency after 1 rain of stasis (0 sec- i, To) FSAR, (Flow to Stasis Aggregation Ratio)

(TT/To)

2. rate of spontaneous aggregate formation in stasis

r1/2

a For comparative purposes, all static parameters are normalized by division with the T~i, value (T~l at 460, 7, 0 sec- 1)

[sec- 1]

H. Schmid-Sch6nbeinetal. : Red Cell Aggregation in Blood Flow. I

152 function of time (t). dV/dt = i shows a maximum value immediately after flow stop and then decays exponentially [](t)= ]o.e -~1 with time [231. The half time of this decay (tl/2) is taken as the half time of aggregate reformation. An aggregation constant (2= In 2/q/2) is computed, and corrected for the viscosity of the respective plasma (2"qo)- For a detailed elaboration of this concept, see [23]. In many pathological samples showing continuous aggregate growth in slow flow (7 sec-1), the light transmission does not show a steady state value but rather increases with time. The slope of this increase (% Tmln/min) is taken as a measure of aggregate growth rate. Accordingly, the light transmission in such samples after 1 min of flow (TT), is higher than that in stasis (To). The ratio (Tv/To=FSAR, s. p. 153) is therefore > 1.0, while it is always < 1.0 in normal blood. Table 1 summarizes all parameters listed above. Results and Discussion

1. Aggregate bbrmation in Stasis In normal human blood the red cell aggregates (RCA) are made up of roulzaux and networks of rouleaux formed by end to side attachment (Fig. 1 A).

These aggregates are separated by gaps of free plasma, and are the cause of the greater transparency of aggregated blood. If normal human blood is observed in stasis following rapid shearing (460 sec- 1, sufficing to disperse all RCA), the aggregation begins within less than one second (as determined by sequential photographs) and after 10 sec practically all cells are integrated. To form aggregates, the cells must be passively brought into physical contact by sedimentation, Brownian motion or residual motion of the fluid. In no case, any sort of active or magnet-like attraction was observed. On the other hand, once part of a static rouleaux, the cells were not seen to be spontaneously released. Platelets, platelet aggregates and leukocytes were not seen to participate in RCA but they are frequently immobilized in the plasma gaps between the rouleaux structures. In pathological blood, the velocity and intensity of RCA formation is altogether increased; the primary formation of rouleaux, however, is very similar to that found in normal blood. Qualitative differences are related to the mode of secondary network formation. In many cases, only the extent of ramification and cross-linking is intensified, still based o n end-to-side attachment. In other samples, however, the rouleaux are aligned in a side-to-side rather than end-to-side

A

D C Fig. 1. (A) (Interference contrast optics, 100 x objective)Network of normal human red cell aggregates (RCA): primary aggregation into typical rouleaux, secondary aggregation due to end-to-side attachement. (B). Pathological RCA (myeloma) short rouleaux and at the free ends of rouleaux, erythrocytes are drawn into hemispherical caps irregular clumping of red cells. (C) Meshes of pathological red cell aggregates. Extendedred celt chains bridgingplasma gaps (cholangitis with disseminated intravascular coagulation).(D) Patholgical red cell aggregates at high shear flow (230 sec-1) (myocardialinfarction). Incomplete hydrodynamic dispersion: flocs and short rouleaux persist, individualcells irregularlydeformed

H. Schmid-Sch6nbein etal. : Red Cell Aggregation in Blood Flow. I

fashion, which results in coarse, rather discontinuous clumps separated by extensive, cell free plasma gaps. The term "agglomeration" might aptly be used for this type of aggregation. In such cases, the terminal cells of rouleaux are often deformed and resemble a hemisperical cap (Fig. 1 B). A third type of secondary aggregation results in extensive, continuous meshes of red blood cells. Here, both end-to-side and side-toside attachment is seen, there are many gaps within the aggregate, encircled by rouleaux which are sometimes passively extended (Fig. 1 C). Observation of the aggregation process at high magnification reveals that the pathologically intensified adhesiveness leads to a rapid condensation of the RCA. Following hydrodynamic dispersion and stop, the cells first form smM1 aggregates and flocs, which then collide, attach each other by a tilting and tractile motion which phenomenologically appears like an elastic recoil. As plasma must be removed from the intrato the inter-aggregate space by this condensation, active work seems to be involved in the adhesive process. In all cases of pathological RCA, the velocity of RCA is much higher than in the normal blood [23], within 1 - 2 sec after flow stop, all cells have become part of an aggregate. Consequently, the half time of aggregate retbrmation (as defined on p. 6 and

153

in [23]), is markedly reduced (Table 2). Occasionally in the myeloma patients, there were quite normal values, presumably due to the fact that the excessive plasma viscosity in those samples hindered Brownian motion and/or sedimentation of individual cells. The extent of RCA (To~t), as measured exactly one minute after flow stop, was in general not higher in the patient than in the normal group. This is explained by the fact that even in normal human blood practically all red blood cells are integrated into aggregates. This fact will have to be taken into consideration when evaluating viscometric results on blood in prestatic flow.

2. Aggregate Rheotogy at Slow Flow (7 sec- 1) In normal human blood, the maximum size of aggregates, plasma gaps and consequently the maximum transparency of the blood is seen in stasis [ 1 9 21]. Upon induction of flow, the size of RCA decreases steadily until at a certain critical shear rate (v.i.) all aggregates are dispersed. The size of the individual RCA for each given constant shear rate remains constant through a continuous uptake and dismissal of individual cells or small rouleaux. Accordingly, the

Table 2. Flow behavior of red cell aggregates in blood from normal subjects, diabetics and myeloma patients Normal controls n=26

Diabetics n = 12

Myeloma patients n = 19

A. Aggregate formation in stasis T0re~

1.32,+_0.06

1.29 +0,04 n.s.

1.26+0,12 p < 0.025

tl/2 (sec) )~"~o (dyn. cm ~ 2 . 1 0 - 2

3.49 ± 1.03 0.287 +- 0.135

1.69_+ 0.79 0.78 _+0.59 p > 0.005

0.94 + 0.39 1.25 _+0.56 a p > 0.001

7'7r~i

1.I5 +_0.05

B. Aggregate theology at 7 s e c ~ 1.18 +0.05 p < 0.05

1.21_+0.15 p < 0.0025

A Tv(% T~,i~/min)

0.065 _+0.067

-

1.54 +- 2.63 p: u

FSAR

0.87l +-0.031

0.918 + 0.043 p < 0.001

0.956_+..0.119 b p < 0.00t

C. Aggregate dispersion and cell alignment v7~,~°(dyn/cm2)

2.56 ± 0.06

2.85 _+0.70

6.51 +- 4.88

7460 ret

1.233 + 0.089

1,2 t 1 ± 0.061 n.s,

1.084 2:0.053 p 1,0)

H. Schmid-Sch6nbein

et a L :

Red Cell Aggregation in Blood Flow. I

loids. In experiments on artificial RCA, e2 macroglobulin consistenly caused an RCA in the form of discontinuous clumps [20], whereas IgM [20] and fibrinogen (v. Hugo and Schmid-Sch6nbein, unpublished) in even the highest concentrations never cause clumps but rather continuous strands. Dextran with a molecular weight of 250000 Dalton causes clump aggregation, below 110000 strand aggregation. Reduction of the zeta-potential (e.g. by neuraminidase treatment) of RCA also causes the formation of clump aggregation (unpublished observation). As the hemodynamic behavior of RCA is most likely affected by the two types of intensified RCA, these deserve further studies [26]. In the blood samples tested in the present study, both types of RCA and transients were observed. Statistical evaluation of these observations is seem in Table IIB. The extent of aggregation (TTrel) is slightly but significantly higher in the two patients groups. In spite of a considerably range of the results (reflecting the two patterns discussed above), both the aggregate growth (A TT) and the FSAR are considerably higher in the diabetics and the myeloma blood, with the largest differences seen in the rate of aggregate growth (A TT).

155

3. Aggregate Dispersion In normal blood, the rouleaux are broken down into short rouleaux (doublets, triplets and quatruplets) at a shear rate of about 50 sec- a. At this shear rate, a mixture of such red cell flocs with individual cells that either rotate randomly or are aligned with flow is observed in the "rheoscope". After only a slight additional increase in shear rate ( 7 0 - 9 0 sec-1), all cells are seen aligned and eventually deformed in flow. Since this alignment gives rise to a secondary increase in light transmission [13, 19] and Fig. 3, shaded area) the shear rate necessary to produce the minimum light transmission can be monitored objectively. The shear stress corresponding to this shear rate thus equals the forces necessary to keep red cell aggregates dispersed : a force proportional to the maximum adhesive forces acting between any pair of red blood cells [13, 19]. As can be seen from Fig. 4, higher forces are necessary to produce the dispersed state in blood from a myeloma patient. In a number of very shear resistant blood samples, it was seen that the small, almost spherical "flocs" (doublets, triplets etc.) were constantly rotating in flow and were not dispersed even at very

Fig. 3. Kinetics of pathological RCA : Photomontage as explained in the legend of Fig. 2 : The aggregation is flow than in stasis (FSAR > 1.0) (myeloma blood)

more

pronounced in slow

H. Schmid-Sch6nbein etal. : Red Cell Aggregation in Blood Flow. I

156

• 1.6-

?

Humcn whole blood Hct45% (n:24}

O 3gG- 3,t - Porcproteinemia

37%

I

-

O3

12£ FUJ > ~t.0 LLI cr

0{

Hydrodynamic Des(xjgregation 1 3,0

/fil

10,9 [dyn/cm 2]

I

1

10 100 SHEAR RATE [sec-~]

Fig. 4. Schematicrepresentationof the light transmissionof human blood as a functionof shear rate. Relativetransmission.Photovoltage at any given shear rate dividedby minimumlight transmission (correspondingto random cell orientationand aggregatedispersal. Shaded area=normal controls: A shear rate of 55+18 sec-~ is required for hydrodynamicdesaggregation,which is equivalent to a shear stress of 3.0 dyn/cm2. Myeloma blood: 10.9dyn/cm2 are required for hydrodynamicdispersion (Trine.). Note strong time effectscompare to Fig. 3)

high shear rates (Fig. 1 D). Consequently, the extent of the secondary increase in transparency (T46orel) is less pronounced. Even though again considerable scatter was seen in the normal and the two patient groups, the shear stress necessary to keep red cell aggregates dispersed is higher in diabetes, highest in the myeloma patients (Table 2 C). T460re1 is significantly lower in the myeloma group: a finding correlated to an increase in relative apparent viscosity at 160 sec- t (v.i.).

4. Effect of Defibrination." Red Cells Suspended in Serum As established by Wells et al. [27] and Merrill [15] there is little viscometric evidence of red cell aggregation in normal human blood after removal of the

plasma fibrin0gen (red cells suspended in defibrinated serum). In t h e " rheoscope" disseminated rouleaux are observed among many individual cells in stasis and there are no continuous networks. A shear rate of 18 _+ 10 sec- 1 is required to disperse the rouleaux. In strict contrast to this, in blood from patients, red cell aggregation with all its rheological consequences occurs when carefully washed red cells are resuspended in serum. Confirming earlier viscometric results by Wells and Schmid-Sch6nbein [27], we again have found photometric evidence of RCA below a shear rate of 30 sec- 1 in defibrinated blood from diabetic patients. In the "rheoscope", participation of all erythrocytes in rouleaux are seen, both in diabetic and myeloma serum. Table 3 shows the statistical evaluation of the diabetic samples compared to the "norm a l " group: with strong differences in all parameters. In many myeloma patients, the RCA in serum is stronger than that of whole normal blood. This finding corroborates earlier ones by many authors [7, 8, 9, 17] and underlines the significance of non-clottable serum proteins in the pathological RCA (a subject beyond the scope of the present communication).

Conclusions

New microrheological methods have revealed profound functional differences between the red cell aggregation occurring as a physiological phenomenon in normal human blood and that occuring under various disease states. These were observed microscopically in blood subjected to viscometric flow in a transparent, counter-rotating "rheoscope" chamber. Due to the effect of aggregation phenomena on the optical behavior of the suspension, the rate and extent of red cell aggregation can be monitored objectively by photometry. An added dividend of the photometric recording is the possibility to follow rapidly occurring changes in the flow behavior of individual cells and cell aggregates. Such rapid changes occur after sudden changes in the rate of shear; the inertia of the parts in the traditional rotational viscometers is far too great

Table 3. Influence of serum proteins on RCA (as determined by measuring the aggregation kinetics of RBC suspended in serum). Hct 45%, 37° C Normal controls RBC in plasma n=25 zrmi,(dyn/cm 2) tl/2(sec) FSAR

2.06 _ + 0 . 4 9 3.49+-1.04 0.871___0.8

Diabetics RBC in serum n=16 0.826+_0.18 10.I +7.46 0.87 +-0.05

RBC in plasma n=32

RBC in serum n=10

2.89_+0.63 1.57+_0.56 0.918±0.04

1.73_+0.81 5.68 +-2.t3 0.901±0.03

H. Schmid-SchSnbein etal. : Red Cell Aggregation in Blood Flow. I

to follow such rapid transients, they were therefore missed by conventional rheological methodology. So far, the present review has dealt with blood samples before and after defibrination in which the hematocrit value was adjusted to one standard value (45%). The rheological behavior in stasis, under conditions of low shear, at the point of hydro-dynamic disaggregation, and in high shear was described. The behavior of the cell aggregates was complicated enough under these highly simplified flow conditions to justity this procedure for the sake of clarification. On the basis of these observations, the much more difficult analysis of the in vivo behavior of cell aggregates shall be attempted. Part II of this review will deal with the hydro-dynamic effects of red cell aggregates on blood flow in vitro and some aspects of their hemo-dynamic relevance in vivo.

References 1. Berman, H.J., Fuhro, R.I. : Quantitative red blood cell aggregomerry of human and hamster bloods. Proc. VIIth Conf. on Microcirculation (Aberdeen t972) Karger, Basel, p. 1t7-125 (1973) 2. Bingham, E.C., Roepke, R.R. : The rheology of blood. III. J. gen. Physiol. 28, 79-93 (1944) 3. Brinkman, R., Zijtstra, W.G. and Jansonius, N.J. : Quantitative evaluation of the rate of rouleaux formation of erythrocytes by measuring light reflection" Syllectometry"). Proc. Kon. Ned. Akad. Wet., Set. C, 66, 236-248 (1963) 4. Chien, S. : The present state of blood rheology. In: ttemodilution, theoretical basis and clinical application. K, Messmer and H. Schmid-Schanbein, eds., Karger, Basel, p . 1 - 4 0 (1972) 5. Dintenfass, L. : Blood microrheology -viscosity factors in blood flow, ischemia and thrombosis. London, (Butterworths) (197I) 6. Dognon, A. : Granulom~trie optique de la suspension sanguine. Nouvelle technique d'6tude. C.R. Ac. Sc. (Paris) 268, 974-975 (1969) 7. Fahraeus, R.: Suspension stability of the blood. Acta med. scand. 55, 1-228 (t921) 8. Fahraeus, R. : The suspension stability of blood. Physiol. Rev. 9, 241-274 (1929) 9. Frimberger, F.: Unspezifit/it der Blutsenkungsreaktion und deren Abhfingigkeit yon LinearkolIoiden. )~rztt. Forschg. 15, 296-304 (1961) 10. Fung, Y.C, : Theoretical considerations of the elasticity of red cells and small blood vessels. Fed. Proc. 25, 1761-1772 (1966) 11. Gelin, L.E. : Intravascular aggregation and capillary flow. Acta chir. scan& 113, 463-465 (1957) 12. Hess, R.W. : Gehorcht das Blut dem allgemeinen StrSmungsgesetz der Fliissigkeiten? Pfltiger's Arch. ges. Physiol. 162, 187244 (1915) 13. Klose, H.J., Volger, E., Brechtelsbauer, H., Heinich, L., SchmidSch6nbein, H. : Microrheology and light transmission of bIood.

I57 I. The photometric effects of red cell aggregation and red cell orientation. Pfli.iger's Arch. 333, 126-139 (1972) 14. Kinsely, M.H. : Intravascular erythrocyte aggregation (blood sludge). In: Handbook of physiology, Sect. 2, Vol. III, W.F. Hamilton and P. Dow, eds., American. Physiological Society, Washington D.C., p. 2249-2292 (1965) 15. Merrill, E.W. : Rheology of blood. Physiol. Rev. 49, 863-888 (1969) 16. Ofstad, J. : The measurement of oxygen saturation and hemoglobin concentration by photometry of whole blood (Chapt. IV), A.S. John Griegs Buktrykker, Bergen (1965) 17. Ruhenstroht-Bauer, G.: Mechanismus und Bedeutung der beschleunigten Erythrocytensenkung. Kli. Wschr. 44, 531-539 (1966) t8. Schmid-SchSnbein, H., Wells, R.E. : Rheological properties of human erythrocytes and their influence upon the "anomalous" viscosity of blood. Ergebn. Physiol. 63, 146-219 (197I) 19. Schmid-Sch6nbein, H., Volger, E., Klose, H.J. : Microrheology and light transmission of blood. II. The photometric quantification of red cell aggregate formation and dispersion on flow. Pfliiger's Arch, 333, 140 - 155 (1972) 20. Schmid-SchSnbein, H., Gallasch, G., Volger, E., Klose, H.J. : Microrheology and protein chemistry of pathological red cell aggregation ("blood sludge") studied in vitro. Biorheology 10, 213-227 (1973) 21. Schmid-SchSnbein, H., Gosen, J.v., Heinich, L., Klose, H.J., Volger, E. : A counter-rotating "rheoscope" chamber for the study of the microrheology of blood cell aggregation by microscopic observation and microphotometry. Microvasc. Res. 6, 366-376 (1973) 22. Schmid-SchSnbein, H. : A simple device allowing blood viscometry at low rates of shear with the Wells-Brookfield viscometer. Res. exp. Med. 161, 49---57 (1973) 23. Schmid-SchSnbein, H., Kline, K.A., Heinich, L., Volger, E., and Fischer, T. : Microrheology and light transmission of blood. Iti. The velocity of red ceil aggregate formation. Pfliiger's Arch. 354, 299-317 (1975) 24. Usami, S., Chien, S.: Optical reflectometry of red blood cell aggregation under shear flow. Proc. VIIth Conf. on Microcirculation (Aberdeen 1972), Karger, Basel, 91-98 (1973) 25. Volger, E., Schmid-SchSnbein, H., Merishi, J.N. : Artificial red cell aggregation caused by reduced salinity: production of a Polyalbumin. Bibl. anat. 11, 296-302, Karger, Basel (1973) 26. Volger, E., Schmid~SchSnbein, H., Gosen, J.v, KIose, H.J., and Kline, K.A.: Microrheology and Light transmission of blood. IV. The kinetics of artificial red celt aggregation induced by Dextran. Pfl/iger's Arch. 354, 319-337 (1975) 27. Wells, R.E., Schmid-SchSnbein, H., Goldstone, J. : Flow behavior of red cells in pathologic sera: existence of a yield shear stress in the absence of Fibrinogen. In : Theoretical and clinical hemorheology. Proc. 2nd Int. Conf. Hemorheology, Heidelberg 1969, Springer, Berlin-Heidelberg-New York, 1971 Professor Dr. H. Schmid-Sch6nbein Abteilung Physiologie der Medizinischen Fakultfit der RWTH D-5100 Aachen Melatener Stral3e 211 Federal Republic of Germany