Eye (1991) 5, 151-158

Blood Rheology and Ischaemia

G.B.NASH Birmingham

Summary Ischaemia is essentially the failure of tissue to obtain a sufficientoxygen supply for its function. In the context of rheology, which is the study of the deformation and flowof materials, this implies failure to deliver the blood, rather than failure to oxygenate the blood or extract oxygen from it. Resistance to the delivery of blood is generally considered to have vascular and rheological components. Vascular effects on resist­ ance may often be dominant, and there is wide appreciation of the ischaemic con­ sequences of vascular obstruction and narrowing, for example in atherosclerotic disease. However, rheological factors can vary widely between individuals and in I 2 disease, and such variations have the potential to influence oxygen supply. , Here, the rheological factors which affect blood flow are reviewed and their role in the development of ischaemia is discussed, with particular reference to the eye where possible.

Rheological Factors Affecting Blood Flow influence in microvessels with diameters of The rheological properties of the blood can the order of the cellular dimensions. At the conveniently be divided into bulk properties level of the microcirculation it may not be and cellular properties (Table I). Broadly appropriate to consider the viscosity of the speaking, the bulk properties (blood viscosity blood as a continuum, but may be better to and its determinants) dominate flow in large consider separately the flow properties of the vessels, whereas cellular factors have greater cellular constituents and the plasma.

Table I. Rheological properties of the blood and their major determining factors

Bulk Properties �Flow resistance in large vessels Factors: Haematocrit Plasma Viscosity Blood Viscosity Red Cell Aggregation I Red Cell Deformation

Cellular Properties �Resistance to capillary entry and obstruction of microvessels Factors: Red Cell Deformability Intrinsic Structure Red Cell Adhesiveness { Plasma Factors Intrinsic Structure White Cell Deformability Functional State (Activation) White Cell Adhesiveness { Plasma Factors

From: Department of Haematology, The Medical School, University of Birmingham, Birmingham B15 2IT.

Correspondence to: Dr. G. B. Nash, Department of Haematology, The Medical School, Vincent Drive, Birmingham B15 2IT. 152 O. B. NASH

Bulk Properties Plasma Viscosity Blood viscosity is determined by haematocrit Variations in the viscosity of plasma depend (and leukocrit in some hyper-leukocytic on variations in its protein constituents. In states), plasma viscosity and red cell aggrega­ normal plasma, the main contributions are tion (which occurs at low shear rates). The from globulins, albumin and fibrinogen, in ability of the red blood cells to deform and that order, and there exists a relatively narrow align in flow greatly reduces the viscosity of normal range of viscosities (Table 11).6 Plasma the blood from the level it would be if these viscosity is raised in a vast range of acu,te and cells were non-deformable, but there are few, chronic infections, and most degenerative and if any, clinical conditions where abnormal neoplastic diseases.6 The changes depend deform ability has a significant affect on bulk mainly on elevation of fibrinogen and have viscosity. The most marked abnormalities in quite small though easily detectable effects on red cell mechanics are generally linked with plasma viscosity. They may, however, have anaemia, which has a greater effect on bulk greater effects on red cell aggregation (see viscosity than impaired red cell deformability. below). On the other hand, paraproteinaemia Haematocrit associated with myeloma and Waldenstrom's disease can have very marked effects on The relationship between blood viscosity and plasma and blood viscosity and can directly haematocrit is logarithmic. For example, at lead to hyperviscosity syndromes.7,8 Macro­ low shear rate (1s-I) blood viscosity varies globulinaemia, for example, can raise plasma from about 13 to 45 mPa.s for haematocrit viscosity by several hundred percent. 6 These varying from 0.3 to 0.5, and at a higher shear rheological changes lead directly to ocular rate (100s-l) from 3.6 to 6.4 mPa.s. The dependence on shear rate arises mainly from problems (dilation of retinal veins, constric­ the effectsof red cell aggregation (see below), tion at arterio-venous crossings and retinal but alignment, deformation and close packing haemorrhages) which are reversible by vis­ of red cells also reduces viscosity as the shear cosity lowering therapy such as 8 rate increases. Increasing haematocrit plasmapherisis.4, increases the oxygen carrying capacity of the Red Cell Aggregation blood, and the quotient of haematocrit/vis­ Blood viscosity rises sharply as the shear rate cosity may be used as a parameter to describe is lowered, for example increasing about five­ rate of oxygen transport.I This parameter fold in normal blood when the shear rate gives a bell shape curve when plotted against -I decreases from 100 to 1 S . This phenomenon haematocrit, with optimal transport in the haematocrit region of 0.3 to 0.5. is chiefly caused by reversible red cell aggre­ The normal range of haematocrit is quite gation. Deformation and alignment of red wide, so that the range of viscosities is even cells with increasing shear rate makes a lesser wider. It seems that for changes in viscosity contribution to viscosity reduction. Red cell per se to cause ischaemic problems in the eye, aggregation can be most easily observed at haematocrit must rise above 0.5.3 In this poly­ Table II. Plasma Viscosity: determinants and values cythaemic region there are typical pathologi­ in health and disease cal manifestations of retinal vein dilation, central retinal vein thrombosis and conges­ Normal 1.5 to 1.72 mPa.s (at 25°C) Albumin : Globulin : Fibrinogen tion of conjunctival venules.4 The mechanism Plasma content 4.0 : 2.5 : 0.3 seems to be directly rheological, with Contribution to viscosity 36 : 41 : 22 increased viscosity causing increased flow Acute and Chronic Disease resistance, and symptoms which respond to Malignancy 1.7 to 3.3 mPa.s (at 25°C) treatment which lowers haematocrit. An­ Paraproteinaemia aemia is also associated with abnormalities in Myeloma Typically>3 mPa.s (at 25°C) the ocular circulation.s This is probably not of rheological origin, except possibly when it is Waldenstrom's Macroglobulinaemia Up to 10-20 mPa.s (at 25°C) associated with abnormalities in the red cells themselves (see below). Data from Harkness.' BLOOD RHEOLOGY AND ISCHAEMIA 153

stasis where cells are seen to form long stacks depleted layer near the wall.15 It is probably of cells or rouleaux. Shearing of the blood dis­ true to say that any contribution of aggrega­ rupts these large scale structures, although tion to ischaemia will be most marked when low shear rates may actually accelerate initial other pre-disposing factors are present, such formation of aggregates.9 Aggregation is as vascular occlusion and reduced perfusion generally attributed to the ability of absorbed pressure. With respect to the eye, a number of large proteins to form cross-bridges between systemic illnesses associated with retinal cell membranes. Recent theories have sug­ ischaemia (e.g. diabetes and rheumatic gested that there may also be colloid osmotic disease) will be accompanied by elevated forces resulting from a layer depleted of pro­ fibrinogen and hence aggregation. More tein between adjacent cellsurfaces.lO What­ specifically, retinal vein occlusion has been ever the mechanism, aggregation in whole shown to be accompanied by increased aggre­ blood is caused mainly by fibrinogen, with gation and blood viscosity, only partially . lesser contributions from y-macroglobulin explainable by elevation of fibrinogen.1 6 17 II and immunoglobulins. Thus, aggregation Cellular Properties will be elevated in most conditions in which plasma viscosity rises, and this is the basis for Red Cell Deformability the commonly used erythrocyte sedimenta­ The red blood cell is able to undergo tion rate (ESR) test. The effects on viscosity repeated, reversible shape changes in the cir­ may be quite pronounced at low shear rates. culation, 'and indeed this is an essential Two contributions to ischaemia can be envi­ requirement if it is to pass through narrow saged. First, increased aggregation may capillaries. This deformability depends on the directly elevate blood viscosity in diseases cell's structure: its geometry, membrane vis­ associated with raised fibrinogen. Second, if coelasticity and cytoplasmic viscosity. The red flow rate is reduced for some other reason cell has excess surface area over that required (e.g. vascular obstruction) then viscosity of to encapsulate its volume, and this is essential otherwise normal blood will rise locally. to allow shape transformation without change Because of the dependence on shear rate, in surface area. The membrane in fact will aggregation will have most effect on viscosity allow very little area dilation before rupture, in the venous circulation where shear rates are and so deformation occurs at essentially con­ at their lowest. Flow into microvessels could stant area. Resistance to deformation and also be impeded by aggregates if they were rate of deformation are primarily determined present on the arterial side. Estimates of by the membrane's intrinsic rigidity (shear shear rates in vivo in animal circulations and bending elastic moduli) and viscosity. -I generally do not fall below 100 S in any Internal cytoplasmic viscosity will also affect region.12 However, in any large vessel the rate of deformation under certain circum­ shear rate is low in the centre and high at the stances (particularly during bending rather periphery so that aggregates may form where than shearing), although this viscosity can average shear rates are quite high. For become so elevated upon dehydration or hae­ example, by ultrasound echogenicity, aggre­ moglobin polymerisation that it becomes a gation has been demonstrated to exist in large dominant factor. In principle, abnormalities veins in humans. 13 in any of the above-mentioned factors can In low flow states, viscosity will rise sharply. compromise ability to circulate and could con­ If stasis occurs intermittently, red cell aggre­ tribute to ischaemia. gation may give rise to a yield stress which In practice, marked abnormalities in red must be overcome before flow can be re­ cell deformability are quite rare. In sickle cell started.1 4Nevertheless, the effect of aggrega­ disease the oxygenated cells include a sub­ tion on blood flow remains controversial, and population of dense, viscous and rigid cells. 18 it has been demonstrated that flow in tubes Upon deoxygenation, polymerisation of hae­ can actually be facilitated by aggregate forma­ moglobin S occurs and the internal viscosity tion and migration of cells to the centre line, rises by several orders of magnitude, with because of formation of a lubricating, cell- consequent dramatic further loss of deform- 154 G. B. NASH

ability.1 9,20 Ischaemie complications occur, deformation than red cells (see Table III for gradually leading to damage to the major some comparative properties). White cells organs, including the eye, In no other dis­ have sufficient excess surface area in the form order is there convincing evidence of a red cell of surface folds, and probably a quite flexible mechanical abnormality being a primary membrane, but their internal viscosity is sev­ cause of ischaemia, Genetic defects of mem­ eral orders of magnitude higher than that of brane structure can lead to loss of membrane red cells.33,34 Although outnumbered by about stability and deformability, loss of membrane 700:1 in the normal circulation, they probably area and changes in cell shape; these are contribute appreciably to flow resistance at poorly tolerated and can lead to haemolytic the capillary leveL In the normally function­ anaemia,21 Although haemolytic anaemia is ing human microcirculation, for example, associated with ischaemic optical complica­ white cells have been observed to cause inter­ tions, the mechanism may not be rheologi­ mittent capillary blockage.37 In experimental caL22 Ovalocytosis is quite common in animal studies, an induced drop in perfusion Melanesia, and this genetic variant is associ­ pressure has been found greatly to exaggerate ated with a several fold increase in membrane this phenomenon and lead to impaired flow rigidity.23 No ischaemic complications have even upon reperfusion.38,39 been described, Abnormal filterability of red The most important factor in ischaemia is cells has been found in diabetics (see, e.g. probably the reactive nature of the white Stuart and Juhan-Vague24 for review). The cells, and particualrly neutrophils. Activation. defect is quite mild and apparently only is associated with marked increases in rigidity occurs in those with poorly controlled disease. and adhesiveness.34,40,41 A proportion of Its contribution to ocular complications is unstimulated neutrophils are typically mar­ "doubtfuL ginated (rolling slowly along venular walls) but this margination increases and cells Red Cell Adhesiveness become firmly attached if activated. This In vitro, normal red cells show a slight tend­ adhesion can lead directly to vascular obstruc­ ency to adhere to cultured endothelium. This tion and increase in flow resistance.42 phenomenon has not been demonstrated in There may thus be a rheologically based vivo, except for sickle cells transfused into contribution of neutrophils to ischaemia. animal models25 and in fa1ciparum malaria.26 Acute or chronic reduction of perfusion pres­ In malaria, the adhesiveness develops at a cer­ sure could lead to vascular obstruction if the tain stage of parasite maturation, is mediated driving force is insufficient to force these cells by specific endothelial cell receptors, and is through capillaries. Trapped cells may thought to play a key role in ischaemic compli­ become activated and initiate a process of tis­ cations.27 Sickle cells have increased adhe­ sue damage by releasing reactive oxygen siveness judged by in vitro assays, and plasma metabolites and lytic enzymes.43 This would factors as well as cellular ones may be cause further release of activating factors and involved.28,29 Increased adhesion has also newly arriving cells would become rigidified been observed in diabetes,30 after myocardial and adhesive. infarction,3! and in systemic sclerosis.32 In This presupposes a vascular initiating fac­ these disorders, plasma proteins increased tor, although another possibility is that adhesion, and it might be that the acute phase inflammatory changes or infection could lead response is quite generally associated with to neutrophil activation. In general, changes increased plasma-mediated adhesiveness. induced by activation in one region, e.g. an This hypothesis remains to be tested. Outside ischaemic limb, could lead to neutrophil of malaria, the effect of adhesion on circula­ entrapment in other organs, mediated by tory pathology is uncertain. plasma born activating factors. The involvement of white cells in ischaemia White Blood Cell Mechanics, Adhesiveness in the eye has received little attention, except and Activation in leukaemic states.44,45 In general, hyperleu­ White blood cells are much more resistant to kotie states have compensatory anaemia, so BLOOD RHEOLOGY AND ISCHAEMIA 155

Table III. Comparative Mechanical properties of red and white blood cells

Geometry Red Cells White Cells

Shape Discoidal Spherical Diameter -8um -8um Volume -90ft 200-300ft Surface Area -135um2 300-400um2 Excess Surface Area -40% . -100% Membrane viscoelasticity Shear rigidity _ 10-5 N.m-1 Bending Rigidity _1O-19 N.m Unknown Viscosity -1O-6 N.s.m-1 Cytoplasmic State Viscosity -0.01 Pa.s -10 Pa.s Structure Liquid Liquid/Gel with Organelles Passive Responds to Activation

that bulk viscosity is not usually elevated.45 -Is the abnormality a cause or consequence Although described as hyperviscosity syn­ of the disease? dromes, symptoms of leukaemia therefore -Could rheological factors predispose to probably arise in the microcirculation, ischaemia or accelerate other causative induced by vascular obstruction at the cellular factors? level. The circulatory problems are likely to -Do other pre-existing conditions magnify be worse when poorly deformable blast cells the effect of rheological abnormalities-or are present.46 indeed, lead quite normal rheological phenomena to excaberate flow reduction? Blood/Vessel Wall Interactions -Even if a rheological abnormality is a Adhesion of both red and white cells to vas­ marker for disease rather than a causative cular endothelium has been referred to factor, might it have prognostic or diagnos­ above. Their contribution to ischaemia may tic value, or aid evaluation of treatment? not be simply obstructive. As pointed out, The normal range of values for some rheo- neutrophils can cause local tissue damage logical properties is wide (e.g. blood vis­ when activated. Adhesion of red cells to cosity). For others (e.g. red cell endothelium has also been suggested as a deformability), quite marked changes can be source of mechanical damage.47 Damage to accommodated without ischaemic complica­ endothelium could cause platelet deposition. tions. Thus, only severe abnormalities are There is thus the possibility that flow-induced likely to be primary causes of ischaemia (e.g. damage by adhesive or colliding blood cells in hyperviscosity syndromes, polycythaemia could indirectly lead to thrombosis and vas­ and sickle cell disease). On the other hand, cular obstruction. rheological abnormalities may influence the progress of ischaemia. Here it is hard to separ­ Relation Between Rheological Factors and ate a contributing factor from a coincidental Ischaemia change in rheology. Increases in plasma vis­ It is true that changes in the rheological cosity, red cell aggregation and whole blood properties of the blood can affect circulatory viscosity may commonly occur in conjunction flow. The question remains: what degree of with an acute phase response. These changes change is necessary to cause or contribute to may or may not significantly affect oxygen ischaemic disease? When evaluating the role deliver.y. Another way in which rheological of rheological factors, one might consider the abnormalities could promote ischaemia is by following questions: pre-disposing to venous thrombosis.48 Here, it -What is the magnitude of any abnormality has been suggested that increases in red cell or change in the disease group compared to aggregation and rigidity, haematocrit and the normal range of variation in rheological fibrinogen all have the capability to promote properties? thromosis. 156 G. B. NASH

The greatest potential for a secondary rheo­ is a specialised organ with an atypical cir­ logical contribution to ischaemia is in low flow culatory arrangement, particularly in its states, where even normal rheological supply to the retina. 49 The arterial system is at phenomena could excaberate the condition. a high pressure. The pressure in the veins is Examples might be following a vaso-occlusive also elevated, because of the ambient intra­ event, in venous insufficiency and in shock. ocular pressure and a high venous flow resist­ Two suggestions are made in Figure 2: ance. Veins and arteries exist in close (1) Reduction of flow rate leads to increasing juxtaposition, and veins can become com­ red cell aggregation and increasing blood vis­ pressed at crossing points. Moreover, the cosity. These parameters may or may not have metabolism and respiration of the retina are been within the normal range initially. Flow extremely high. The retinal circulation is thus resistance and local vascular pressure then considered to have a limited ability to adapt to increases. Fluid loss may occur, with local rheological abnormalities, and may be haemoconcentration, elevation of haemato­ particularly susceptible to rheological crit and plasma proteins. Further aggregation changes. 50 and viscosity rises follow, with increased Haemorheological aspects of ocular disease resistance. and ischaemia have recently been reviewed by (2) Reduction of perfusion pressure may pro­ Foulds. 51 Distinction is drawn between over­ mote vascular obstruction by white cells. tly abnormal haemorheology (hyperviscosity Trapped cells may become activated, causing syndromes, polycythaemia, leukaemia, sickle vascular damage, release of stimulating fac­ cell disease), and a range of conditions where tors and changes in the properties of newly evidence of abnormal blood rheology exists, arriving cells. This could result in further vas­ but where the role in ocular diseas� is uncer­ cular occlusion and increased resistance. The tain (ischaemic optic neuropathy, venous above processes are not mutually exclusive insufficiency retinopathy, hyperlipidaemia, and could lead to vicious cycles of increasing diabetes, open angle glaucoma). Lowe48 has vascular resistance. pointed out that 'after the veins of the lower limb and pelvis, spontaneous venous occlu­ Blood Rheology and the Eye sion occurs most commonly in the retinal Inferences regarding the role of rheological vein', and suggests that rheological as well as changes in ischaemia must be drawn from vascular factors may be involved. In the con­ diverse clinical conditions. The eye, however, text of retinal vein occlusion, Shilling notes

ACTIVATION AND /'A&UAR�� WHITE CELL MARGINATION CAPILLARY PLUGGING

FLUID EXUDATION RAISED\�HAEMATOCRIT noo REDUCEDSHEAR RATE __ ACTIVATIONOF NEWLY

If\CREASEDAGGR EGATKlN � 7- FURTHER CAPILLARY INCREASED RESISTAf\CE PLUGGING

VASCULAR INSUFFICIENCY AND OCCLUSION; SHOCK

Fig.I. Cycles of rheological changes initiated by flow reduction (adapted from Nash & Dormand/). BLOOD RHEOLOGY AND ISCHAEMIA 157 that investigations of blood viscosity have Variable ultrasound echogenicity in flowing blood. Science1982, 218: 1321-3. been disappointing in trying to understand the 14 Kiesewetter H: The yield shear stress of blood in cause of occlusion on· a patient-by-patient branched models of the microcirculation: Effect basis.52 In deciding an aetiological role for of haematocrit and plasma macro-molecules. Bibl Haematol1981, 47: 14-20. rheological factors, perhaps the acid test is I S whether rheological treatment improves the Reinke W, Johnson PC, Gaehtgens P: Effect of shear rate on apparent viscosity of human blood clinical state of the patient. Plasmapherisis, in tubes of 29 to 94 [1m diameter. Circ Res 1986, leukapherisis and venesection have accepted 59: 124-32. 16 therapeutic roles in hyperviscosity syn­ Trope GE, Lowe GDO, McArdle BW, Douglas JT, dromes, leukaemia and polycythaemia,4,44,51 Forbes CD, Prentice CRM, Foulds WS: Abnormal blood viscosity and haemostasis in but definitive studies remain to be done in long-standing retinal vein occlusion. Br J Oph­ other putative rheological syndromes, thalmol1983, 67: 137-42. 17 Chabanel A, Glacet-Bernard A, Lelong F, Taccoen Keywords: Blood Rheology, Circulation, Erythro­ A, Coscas G, Samama MM: Increased red blood cyte, Eye, Ischaemia, Leukocyte. cell aggregation in retinal vein occlusion. Br J Haematol1990, 75: 127-31. References 18 Nash GB, Johnson CS, Meiselman HJ: Mechanical 1 Chien S: Physiological and pathophysiological sig­ properties of oxygenated red blood cells in sickle nificance of hemorheology. In Chien S, Dor­ cell (HbSS) disease. Blood1984, 73-82. mandy J, Ernst E, Matrai A, eds. Clinical 19 Chien S, King RG, Kaperonis AA, Usami S: Vis­ haemorheology. Dordrecht: Martinus Nijhoff coelastic properties of sickle cells and hemoglo­ 1987: 125-164. bin. Blood Cells1982, 8: 53-65. 2 Nash GB and Dormandy JA: The involvement of 20 Nash GB, Johnson CS, Meiselman HJ: Influence of red cell aggregation and blood cell rigidity in oxygen tension on the viscoelastic behaviour of impaired microcirculatory efficiency and oxygen red blood cells in sickle cell disease. Blood 1986, delivery. In Fleming JS, ed. Drugs and the 67: 110--8. delivery of oxygen to tissue. Boca Raton: CRC 2 1 Stuart J and Nash GB: Red cell deformability and Press, 1990; 227-252. haematological disorders. Blood Reviews1990, 4: 3 Laurence JH: Polcythaemia. New York: Grune and 141-7. Stratton, 1955. 22 Brain MC: Hematologic and reticuloendothelial dis­ 4 Luxenberg MN: Hematologic and reticuloendothe­ eases. In Mausolf FA, ed. The eye and systemic lial diseases and their relation to the eye. Mau­ In disease. St. Louis: C. V. Mosby, 1980; 294-304. 2 solf FA, ed. The eye and systemic disease. St. 3 Saul A, Lamont G, Sawyer WH, Kidson C: Louis: C. V. Mosby, 1980: 305-324. Decreased membrane deformability in mel an­ 5 Holt JM and Gordon-Smith EC: Retinal abnormal­ esian ovalocytes from Papua New Guinea. J Cell ities in diseases of the blood. Br J Ophthalmol Bioi1984, 98: 1348-54. 1969, 203: 145-60. 6 24 Stuart J and Juhan-Vague I: Erythrocyte rheology in Harkness J: The viscosity of human blood plasma; diabetes mellitus. Clinical Haemorheology 1987, its measurement in health and disease. Biorhe­ 7: 239-45. ology 1971,8: 171-93 25 Kaul DK, Fabry ME, Nagel RL: Microvascular sites 7 McGrath MA and Penny R: Paraproteinaemia: and characteristics of sickle cell adhesion to vas­ Blood hyperviscosity and clinical manifestations. cular endothelium in shear flow conditions. Proc J Clin Invest 1976, 58: 1155-62. Nat Acad Sci (USA) 1989,86: 3356-60. 8 26 Somer T: Rheology of paraproteinaemias and the Luse SA and Miller LH: Plasmodium falciparum plasma hyperviscosity syndrome. Balliere's Clini­ malaria: Ultrastructures of parasitized erythro­ cal Haematology1987, 3: 695-723. cytes in cardiac vessels. Am J Trop Med Hyg1971, 9 Volger E, Schmid-Schonbein H, v Gosen J, Klose 20: 655-60. HJ, Kline KA: Microrheology and light transmis­ 27 Howard RJ and Gilladoga AD: Molecular studies sion of blood. Pfiugers Arch1975, 354: 319-37. related to the pathogenesis of cerebral malaria. 0 1 Janzen J and Brooks DE: Do plasma proteins absorb Blood1989, 74: 2603-18. to red cells? Clinical Haemorheology 1989, 9: 28 Hebbel RB, Moldow CF, Steinberg MH: Modu­ 695-714. lation of erythrocyte-endothelial interactions and II Chien S, Usami S, Dellenback RJ, Gregersen MI: the vasocclusive severity of sickling disorders. Shear dependent interaction of plasma proteins Blood1981, 58: 947-52. with erythrocytes in blood rheology. Am J Physiol 29 Mohandas N and Evans EA: Adherence of sickle 1970, 219: 143-53. erythrocytes to vascular endothelial cells: 12 Lipowsky HH, Kovalchek S, Zweifach BW: The dis­ Requirement for both cell membrane changes tribution of blood rheological parameters in the and plasma factors. Blood1984, 64: 282-7. 0 microcirculation of cat mesentery. Circ Res1978, 3 Wautier JL, Paton RC, Wautier MP, Pintigny D, 43: 738-49. Abadie E, Passa P, Caen JP: Increased adhesion 13 Sigel B, Machi J, Beitler JC, Justin JR, Coelho JC: of erythrocytes to endothelial cells in diabetes 158 G. B. NASH

mellitus and its relation to vascular complications. 40 Nash GB, Jones JG, Mikita J, Christopher B, Dor­ New Eng J Med1981, 305: 237-42. mandy JA: Effects of preparative procedures and 31 Smith BD, Thomas JL, Gillespie JA: Abnormal of cell activation on flow of white cells through erythrocyte endothelial adherence in ischemic tnicropore filters. Br J Haematol1988,70: 171-6. heart disease. Clinical Haemorheology 1990, 10: 41 Harlan JM: Leukocyte-endothelial interactions. 241-53. Blood1985, 65: 513-25. 32 Kovacs IB, Rustin MH, Thomas RH, Ridler C, 42 Schmid-Schonbein GW: Mechanisms of granulo­ Sowemimo-Coker SO, Kirby JD: Plasma or cyte-capillary-plugging. Prog Appl Microcirc serum from patients with systemic sclerosis alters 1987,12: 223-30. behaviour of normal erythrocytes. Ann Rheum 43 Lucchesi BR and Mullane KM: Leukocytes and Dis1985, 44: 395-8. ischaemia-induced myocardial injury. Ann Rev 33 Schmid-Schonbein GW, Shih YY, Chien S: Mor­ Pharmacol1986, 26: 201-24. 44 phometry of human leukocytes. Blood 1980, 56: Mehta B, Goldman JM, Kohner EM: Hyperleuko­ 866-75. cytic retinopathy in chronic granulocytic leuka­ Br J 34 Chien S, Schmid-Schonbein GW, Sung KLP, emia: The role of intensive leukapherisis. Haematol1984,56: 661-7. Schmalzer EA, Skalak R: Viscoleastic properties "' Lichtman MA and Rowe JM: Hyperleukocytic leu­ of leukocytes. In Meiselman HJ, Lichtman MA, kemias: Rheological, clinical and therapeutic LaCelle PL, eds. White cell mechanics: basic considerations. Blood1982, 60: 179-83. science and clinical aspects. New York: Alan R. .\6 Lichtman MA: Rheology of leukocytes, leukocyte Liss 1984: 19-51. suspensions and blood in leukemia: possible 35 Nash GB and Meiselman HJ: Rheological proper­ relationship to clinical manifestations. J Clin ties of individual polymorphonuclear granulo­ Invest1973, 52: 350-8. cytes and lymphocytes. Clinical Haemorheology 47Wautier JL, Pintigny D, Maclouf J, Wautier MP, 1986,6: 87-97. Corvazier E, Caen JP: Release of prostacyclin 36 Brooks DE and Evans EA: Rheology of blood cells. after erythrocyte adhesion to cultured vascular In Chien S, Dormandy J, Ernst E, Matrai A, eds. endothelium. J Lab Clin Med1986, 107: 210-5. Clinical haemorheology. Dordrecht: Martinus 4R Lowe GDO: Blood rheology and venous throm­ Nijhoff1987: 73-96. bosis. Clinical Haemorheology 1984,4: 571-88. 37Bagge U and Branemark P-I: White blood cell rhe­ 49 Trevor-Roper PD and Curran PV: The eye and its ology. An intravital study in man. Adv Microcirc disorders. 2nd ed. Oxford: Blackwell. 1984: 43. 1977,7: 1-17. 50 McGrath MA, Wechsler F, Hunyor ABL, Penny R: 3R Bagge U: Leukocytes and capillary perfusion in Systemic factors contributing to retinal vein shock. In Meiselman HJ, Lichtman MA, LaCelle occlusion. Arch Intern Med1978, 138: 216-20. PL, eds. White cell mechanics: basic science and 51 Foulds WS: 'Blood is thicker than water' Some hae­ clinical aspects. New York: Alan R. Liss 1984: morheological aspects of ocular disease. Eye 285-94. 1987,1: 343-63. 39 Engler RL, Schmid-Schonbein GW, Pavelec RS: 52 Shilling JS: Retinal vein occlusion. [n Kanski JJ, Leukocyte plugging in myocardial ischaemia and Morse PH, eds. Disorders of the vitreous, retina reperfusion in the dog. Am J Pathol 1983, Ill: and choroid. London: Butterworths 1983: 98-111. 115-121.