Inhibition of Red Cell Aggregation Prevents Spontaneous Echocardiographic Contrast Formation in Human Blood
Background Spontaneous echocardiographic contrast (SEC) is a pattern of blood echogenicity that has been attributed to ultrasonic backscatter from blood cell aggregates that form under low shear conditions. Patients with left atrial SEC have an increased thromboembolic risk. This study examined the role of red cell and platelet aggregates in the pathogenesis of SEC in human blood and the effects on SEC of antithrombotic therapy and red cell disaggregatory agents.
Methods and Results Blood echogenicity was examined with the use of quantitative videodensitometry over a controlled range of flow velocities in an in vitro model characterized by nonlaminar flow conditions. One hundred ninety study samples were prepared from single fresh blood donations (40 to 120 mL) from 24 healthy volunteers and 11 patients. Whole blood echogenicity was unaltered by depletion of platelets, stimulation of platelet aggregation with adenosine diphosphate, or inhibition of platelet aggregation with aspirin. Low flow–related echogenicity increased with increasing hematocrit (P<.001) but was abolished when red cells were lysed selectively with saponin (P<.001). In the presence of red cells, low flow–related echogenicity increased with increasing fibrinogen concentration (P<.001) and with plasma paraproteins. Low flow–related echogenicity in whole blood was unaltered by heparin and warfarin but was reduced in a dose-dependent manner by dextran 40 (40 mg/mL, 70% reduction, P<.001) and poloxamer 188 (8 mg/mL, 47% reduction, P<.001), which inhibited red cell aggregation.
Conclusions These results support protein-mediated red cell aggregation as the mechanism of SEC in human blood. Inhibition of red cell aggregation, indexed by resolution of SEC, may provide an alternative to anticoagulant and antiplatelet therapy to reduce cardiac thromboembolic risk.
Spontaneous echocardiographic contrast (SEC) is a smokelike pattern of blood echogenicity that may be detected in the left atrium during transesophageal echocardiography. The presence of SEC has been associated with blood stasis and increased thromboembolic risk.1 2 3 Increased blood echogenicity at low flow rates has been attributed to increased ultrasonic backscatter from aggregates of blood cells, but the relative role of red cell and platelet aggregates has been controversial.4 5 6 7 8 9 10 11 12 Previous studies of the pathogenesis of SEC have been limited by the use of flow models that did not simulate the nonlaminar flow conditions in the human left atrium,4 5 6 7 10 11 12 the use of nonhuman blood or pooled or stored human blood with potentially different properties than those of fresh human blood,4 7 9 10 11 12 and reliance on qualitative observations.4 5 6 10
To address these limitations, we investigated the pathogenesis of SEC in an in vitro model in which the echogenicity of fresh human blood was determined by quantitative videodensitometry under nonlaminar flow conditions over a controlled range of flow velocities. Given the association between SEC and increased thromboembolic risk, we sought to determine the effects on SEC of various pharmacological agents used in the prevention and treatment of thromboembolism.
In Vitro Model
The in vitro model consisted of a 20-mL plastic cylinder in which nonlaminar flow was generated by a magnetically driven stirring device (Fig 1⇓). Stirring speed was altered in eight graded steps from high flow velocity to stasis. Echocardiographic images and blood velocity data were obtained at each stirring speed. This protocol was repeated three times for each blood sample. At any given stirring speed, the variability in blood velocity measurements was <2%. Blood echogenicity was examined with the use of a Hewlett-Packard Sonos 1000 ultrasonograph with a 5-MHz transesophageal transducer (HP 21362A) that was positioned adjacent to a window in the chamber, which was sealed with parafilm. A constant alignment of the transducer relative to the window was maintained with a partitioned perspex base. Blood velocity was determined by pulsed-wave Doppler interrogation at a fixed point at the chamber periphery, tangential to the direction of flow. Acoustic power output, overall gain, time gain compensation, compression, depth, and preprocessing and postprocessing settings were constant throughout all experiments.
The mechanical effects of the stirring procedure per se on blood cells were determined quantitatively by Coulter count and flow cytometry before and after stirring. No change in the number of red cells, white cells, or platelets or in red cell volume were observed. On microscopic examination, there was no evidence of change in cell shape with stirring.
Echocardiographic images recorded on s-VHS videotape were digitized at a resolution of 512×512 pixels with 256 gray levels (0=black, 255=white) and were transferred to a Sun SPARC Station-2 computer for videodensitometric analysis. The mean gray scale videodensity within a defined region of interest in the blood field was determined for each stirring speed. The variability of videodensitometric data obtained from the three repetitions of the stirring protocol ranged from 1% at high flow velocity to 5% at stasis.
Fresh blood was obtained by antecubital venepuncture from 24 healthy male and female volunteers, aged 23 to 54 years, and was anticoagulated with 15% EDTA. For each experiment, multiple study samples were prepared from single blood donations (40 to 120 mL) from each volunteer. All experiments were repeated six times with blood from separate volunteers. Blood was obtained also from 5 patients with paraproteinemias (2 with myeloma, 3 with Waldenstrom’s macroglobulinemia studied before and after plasmapheresis) and 6 patients before and after initiation of oral warfarin (5 for atrial fibrillation, 1 for dilated cardiomyopathy). A total of 190 blood samples were examined. All volunteers and patients gave written informed consent. The study was approved by the institutional research and ethics committee.
To determine the role of platelets, red cells, and plasma proteins in the pathogenesis of SEC, the following combinations of blood cells and plasma were examined in three groups of experiments: (1) platelet group: whole blood (mean hematocrit, 45%; platelet count, 234×109/L); red cells in platelet-poor plasma; platelet-rich plasma alone (platelet count, 445×109/L); whole blood plus adenosine diphosphate (ADP, 20 μg/mL) to promote platelet aggregation9 ; whole blood examined before and 2 hours after a single oral dose of aspirin (600 mg); (2) red cell group: whole blood; whole blood plus saponin (20 mg/mL) to selectively lyse red cells; variable hematocrit (60%, 45%, 30%, and 15%); (3) plasma protein group: red cells plus saline; red cells plus plasma; red cells plus fibrinogen-depleted plasma plus purified human fibrinogen (6, 3, 1, and 0 g/L).
The hematocrit, platelet count, and fibrinogen concentration of study blood samples were kept constant with the exception of experiments in which individual parameters were examined specifically. In the platelet group experiments, platelet inactivation by aspirin (600 mg) was confirmed in 4 subjects by assessment of platelet aggregation responses to ADP, epinephrine, and collagen immediately before drug administration and at the peak plasma concentration, 2 hours after drug administration.13 Platelet aggregation responses examined before and after the experimental protocol demonstrated that platelets inactivated by aspirin were not reactivated by the mechanical effects of stirring.
To determine the effects of antithrombotic therapy on SEC, whole blood was studied before and after addition of the following agents: (1) heparin sodium (0.3 IU/mL of blood), (2) warfarin orally, which produced an international normalized ratio between 1.5 and 3.2, (3) dextran 40 (D4133, Sigma Chemical Co) dissolved in plasma, after the red cells were separated by centrifugation, to achieve concentrations of 0, 20, and 40 mg/mL (dextran 40 plasma solutions were then centrifuged to remove insoluble material and recombined with red cells), (4) poloxamer 188 (150 mg/mL, RheothRx injection, Glaxo Wellcome) added to whole blood to achieve concentrations of 0, 0.25, 0.5, 1.0, 2.0, 4.0, and 8.0 mg/mL. Effects of poloxamer 188 on red cell aggregation were estimated by the erythrocyte sedimentation rate (Westergren method).
Relations between blood velocity and videodensitometric score were demonstrated on curves derived by piece-wise linear interpolation of mean data from all subjects. Statistical comparisons between curves were made by two-way ANOVA with data from each subject at three points on the curves: stasis (0 cm/s), low velocity (5 cm/s), and high velocity (20 cm/s). These velocity points corresponded with the maximal echogenicity, low-level echogenicity, and absence of echogenicity in whole blood, respectively. When significant differences between curves were identified, data for each of the three velocity points were analyzed separately by one-way ANOVA and the Mann-Whitney U test. Statistical comparisons were performed after data were normalized to control values in whole blood in each subject. Statistical analyses were performed with the SPIDA software package (Statistical Computing Laboratories) Data are expressed as mean±SD. A value of P<.05 was considered significant.
Whole blood was echolucent at high flow velocities. At flow velocities <10 cm/s, whole blood became echogenic. The extent of echogenicity increased with further reduction of blood flow velocity and was maximal at stasis, with a dense, homogeneous appearance (Fig 2⇓).
We hypothesized that the variegated patterns of SEC observed in vivo might be caused by mixing of relatively higher velocity pulmonary venous flow with static blood in the left atrium. To test this hypothesis, 2 mL of blood from the flow chamber was withdrawn into a 5-mL plastic syringe to which was attached a 3-cm length of plastic tubing (internal diameter, 1.5 mm). Blood in the syringe was injected into the top of the flow chamber, which was stirred at a constant low blood velocity (5 cm/s). This procedure reproduced the heterogeneous echodensity and circular flow patterns characteristic of SEC (Fig 3⇓; also see Table 1⇓).
Role of Blood Constituents
Platelets. Platelet-rich plasma without red cells showed no flow-dependent changes in echogenicity. Low flow–related echogenicity in whole blood was not altered by the absence of platelets, stimulation of platelet aggregation with ADP, or inhibition of platelet aggregation after platelet inactivation with aspirin. (See Fig 4⇓ and Table 1⇑.)
Red cells. Low flow–related echogenicity in whole blood was abolished after lysis of red cells with saponin. The extent of low flow–related echogenicity in whole blood increased with increasing hematocrit.
Plasma proteins. Low flow–related echogenicity was observed in suspensions of red cells in plasma but not in suspensions of red cells in saline. Microscopic examination of blood smears demonstrated that red cells in saline were in a dispersed state; red cells in plasma aggregated to form rouleaux (Fig 5⇓). Low flow–related echogenicity increased with increasing fibrinogen concentrations. Smaller increases in echogenicity at low flow velocity were observed in whole blood when fibrinogen was absent, provided that normal concentrations of the other plasma proteins were present. Low flow–related echogenicity in patients with abnormal plasma paraproteins was increased markedly in excess of that observed with high fibrinogen concentrations in the healthy volunteers. In patients with Waldenstrom’s macroglobulinemia, videodensitometric scores at low flow velocities were reduced by plasmapheresis.
Effects of Antithrombotic Therapy
Anticoagulants. Neither heparin nor warfarin significantly altered blood echogenicity at low flow velocity (Fig 6⇓). On microscopic examination, red cells aggregated to form rouleaux in both the heparin and warfarin blood samples (Fig 5⇑).
Dextran 40. Low flow–related echogenicity was reduced by dextran 40 in a dose-dependent manner (Fig 6⇑). On microscopic examination, red cell rouleaux were absent with dextran 40 concentrations of 20 and 40 mg/mL (Fig 5⇑).
Poloxamer 188. Progressive reductions in low flow–related echogenicity were observed with increasing poloxamer 188 concentrations >1.0 mg/mL, achieving statistical significance with concentrations of 4.0 and 8.0 mg/mL (Fig 6⇑). Similarly, reductions in the erythrocyte sedimentation rate were observed with these higher poloxamer 188 concentrations. On microscopic examination, red cells at these concentrations showed no rouleaux formation (Figs 5⇑ and 6⇑; also see Table 2⇓).
Mechanism of SEC
The results of this study are consistent with protein-mediated red cell aggregation as the mechanism of SEC in human blood. Both red cells and plasma proteins were required to produce low flow–related echogenicity. Individual red cells are normally prevented from aggregating by the repulsive electrostatic effects of their negative surface charge. Plasma proteins, particularly fibrinogen, are able to overcome these electrostatic forces and facilitate aggregation of cells at low shear rates by the formation of cross-bridges between the cells.10 14 15 In this study, low flow–related echogenicity was increased markedly in the presence of plasma paraproteins, which promote red cell rouleaux formation,14 and was reduced by plasmapheresis. Low flow–related echogenicity increased with increasing hematocrit. Injection of blood into the flow chamber produced variegated echodensity with circular flow patterns, consistent with our hypothesis that this phenomenon in vivo is produced by the relatively higher velocity pulmonary venous inflow mixing with static, echogenic blood in the left atrium.
Previous Studies of Blood Echogenicity
Sigel et al4 observed that test tubes of static whole blood or red cells in plasma were echogenic, whereas red cells in saline or plasma without red cells were not echogenic. The authors concluded that red cells and fibrinogen were required for blood echogenicity. Stored and fresh human blood pooled from multiple donors was used in these experiments. Similar results were reported by Merino et al,9 who examined porcine blood under laminar and nonlaminar flow conditions. Wang et al10 observed the simultaneous onset of echogenicity and red cell rouleaux when macromolecular polymer was added to static canine and banked human red cells suspended in saline. Hematological data in patients with atrial fibrillation support these findings.16 17 Black et al16 reported significant correlations between the presence of SEC and the hematocrit and fibrinogen concentration. We found that the severity of SEC correlated with the erythrocyte sedimentation rate and low-shear blood viscosity, both of which are indices of red cell aggregation.17
In contrast, Mahony et al11 12 attributed a type of echogenicity observed in canine blood in vitro and in vivo to the formation of platelet aggregates. Erbel et al18 found increased platelet aggregation in patients with SEC and mitral valve disease. Recently, Kearney and Mahony19 reported that aspirin reduced the size of circulating platelet macroaggregates that were detected echocardiographically in the brachial veins of normal subjects.
A unique feature of the present study was the use of a model that permitted the study of fresh unpooled human blood under nonlaminar flow conditions. Ultrasonic backscatter in blood is influenced by the number and size of cells and cellular aggregates, the interaction between cells, and variation in the concentration of cells within a given region.7 20 21 22 23 Consequently, blood composition and flow characteristics are important determinants of echogenicity. Studies of animal blood or human blood that has been stored or pooled thus may not be directly applicable to SEC in fresh human blood. For example, human red cells are smaller than porcine and canine red cells and have different aggregation kinetics.24 25 Stored human red cells become rigid,26 which alters the reflective surfaces of cells, cell deformability in response to variable shear conditions, and aggregation properties. The composition of pooled blood may change from one experiment to another as the result of variation in the hematocrit and plasma protein concentration. Most previous studies have used static blood in test tubes or laminar flow models,4 5 6 7 10 11 12 which differ considerably from the nonlaminar flow conditions within the human left atrium.
Inhibition of SEC
In clinical studies, SEC has been observed in patients receiving anticoagulant therapy.1 2 The effects of anticoagulation on the severity of SEC have not been examined previously. In the present study, low flow–related echogenicity was not altered by either heparin or warfarin. This finding is consistent with the mechanism of SEC because anticoagulants inhibit the coagulation cascade to prevent thrombus formation but do not alter red cell aggregation.
Low flow–related echogenicity was reduced by dextran 40 and poloxamer 188. Dextran 40 is a low-molecular-weight polysaccharide with colloid osmotic properties that expands plasma volume and exerts antithrombotic effects by improvement in blood flow and reduction of red cell aggregation.27 28 29 30 31 Red cell disaggregatory effects of dextran 40 have been attributed to reduction of red cell surface charge, competition for binding sites, increased exclusion volume between cells, and cryoprecipitation of fibrinogen.28 29 30 31 Several investigators have found that concentrations of dextran 40 required to inhibit red cell aggregation in vitro have been in excess of those used clinically.27 29 30 In the present study, a dextran 40 concentration of 20 mg/mL was selected specifically on the basis of steady-state plasma concentrations of dextran 40 reported in vivo.31
Poloxamer 188 is a nonionic block copolymer surfactant that is thought to modify hydrophobic interactions between cells.32 In animal models of cerebral and myocardial ischemia, poloxamer 188 improved blood flow and reduced tissue damage, without hemodilution.33 34 Poloxamer 188 has concentration- and shear-dependent effects on blood viscosity.32 35 Dose-related reductions in the extent and strength of red cell aggregation and prolongation of the time to aggregation have been demonstrated in vitro.32 35
Left atrial thrombus is composed of red cell aggregates within a fibrin meshwork.36 Regional hyperviscosity, promoted by low blood flow and red cell aggregation, is considered to be a prerequisite for “red” thrombus formation.37 38 39 The factors that link these prerequisite conditions to the process of thrombosis have not been identified precisely but may include local generation of thrombin or platelet activation.39 Prevention of the prothrombotic milieu in the left atrium by inhibition of red cell aggregation represents a potential novel therapeutic approach to thromboembolic prophylaxis in cardiac patients. Red cell disaggregatory agents may be useful as alternative or adjunctive treatment to existing anticoagulant or antiplatelet regimens.
Several limitations of current red cell disaggregatory agents need to be noted. Prolonged intravenous administration of dextran 40 (≈4 days) is required to achieve “red cell disaggregatory” plasma concentrations.31 More rapid administration may be complicated by adverse hemodynamic and nephrotoxic effects in euvolemic subjects. In a recent clinical study with poloxamer 188, an unacceptable incidence of adverse renal effects was observed with doses aimed at producing plasma concentrations of 0.5 to 1.0 mg/mL (Glaxo Wellcome, unpublished data, 1996).
Sequential measurements of blood velocity rather than shear rate were used in this study to document reductions in blood flow with decreasing stirring speeds. Although blood shear rate is the major determinant of red cell aggregation at low flow rates,7 9 40 local shear conditions in nonlaminar flow are complex and variable.40 In contrast, blood velocity can be determined rapidly and repeatedly by Doppler measurements. In our experimental model, blood became echogenic at a relatively lower velocity (<10 cm/s) than observed in the human left atrium (<35 cm/s).2 This difference is consistent with the smaller size of the in vitro chamber when compared with the left atrium, which would result in relatively higher shear rates at any given blood velocity level. Red cell aggregation in vitro has been shown to be sensitive to changes in ambient temperature.14 In the present study, all experiments were performed at room temperature (23°C). Although videodensitometry has been regarded as a “gold standard” for quantitative assessment of blood echogenicity in vitro, it does not measure ultrasonic backscatter directly and thus can be difficult to calibrate in an absolute manner. This limitation may be overcome with recently developed acoustic densitometric techniques that are not affected by factors such as compression and gray scale mapping operations performed by the ultrasonic scanner.41 The evidence implicating aggregation of red cells as the mechanism of SEC in this study is indirect. Methods of quantifying cell volume, such as flow cytometry, are unable to demonstrate velocity-dependent rouleaux formation because the shear rates required for application of the cytometric technique exceed the limits at which rouleaux can form. Direct visualization of red cell rouleaux in vivo, in association with the onset of SEC, may be possible with future development of high-frequency ultrasonic transducers.42
The results of this study support plasma protein–mediated red cell aggregation as the mechanism of SEC in human blood. Although anticoagulants may impede progression from red cell aggregation to thrombus formation, these results suggest that directly inhibiting red cell aggregation may provide an alternative method of reducing thromboembolic risk in the future.
Dr Fatkin was supported by a Postgraduate Medical Research Scholarship from the National Health and Medical Research Council of Australia, Canberra, ACT, Australia.
- Received November 4, 1996.
- Revision received February 3, 1997.
- Accepted February 16, 1997.
- Copyright © 1997 by American Heart Association
Daniel WG, Nellessen U, Schroder E, Nonnast-Daniel B, Bednarski P, Nikutta P, Lichtlen PR. Left atrial spontaneous echo contrast in mitral valve disease: an indicator for an increased thromboembolic risk. J Am Coll Cardiol. 1988;11:1204-1211.
Fatkin D, Kelly R, Feneley MP. Relations between left atrial appendage blood flow velocity, spontaneous echocardiographic contrast and thromboembolic risk in vivo. J Am Coll Cardiol. 1994;23:961-969.
Leung DYC, Black IW, Cranney GB, Hopkins AP, Walsh WF. Prognostic implications of left atrial spontaneous echo contrast in nonvalvular atrial fibrillation. J Am Coll Cardiol. 1994;24:755-762.
Sigel B, Coelho JCU, Spigos DG, Preston Flanigan D, Schuler JJ, Kasprisin DO, Nyhus LM, Capek V. Ultrasonography of blood during stasis and coagulation. Invest Radiol. 1981;16:71-76.
Wolverson MK, Nouri S, Joist JH, Sundaram M, Heiberg E. The direct visualisation of blood flow by real-time ultrasound: clinical observations and underlying mechanisms. Radiology. 1981;140:443-448.
Mikell FL, Asinger RW, Elsperger KJ, Anderson WR, Hodge M. Regional stasis of blood in the dysfunctional left ventricle: echocardiographic detection and differentiation from early thrombosis. Circulation. 1982;66:755-763.
Shung KK. Physics of blood echogenicity. J Cardiovasc Ultrasonogr. 1983;2:401-406.
Yuan YW, Shung KK. Ultrasonic backscatter from flowing blood, I: dependence on shear rate and hematocrit. J Acoust Soc Am. 1988;84:52-58.
Merino A, Hauptman P, Badimon L, Badimon JJ, Cohen M, Fuster V, Goldman M. Echocardiographic ‘smoke’ is produced by an interaction of erythrocytes and plasma proteins modulated by shear forces. J Am Coll Cardiol. 1992;20:1661-1668.
Wang X-F, Liu L, Cheng TO, Li Z-A, Deng Y-B, Wang J-E. The relationship between intracardiovascular smoke-like echo and erythrocyte rouleaux formation. Am Heart J. 1992;124:961-965.
Mahony C, Ferguson J, Fischer PLC. Red cell aggregation and the echogenicity of whole blood. Ultrasound Med Biol. 1992;18:579-586.
Mahony CM, Ferguson J. The effect of heparin versus citrate on blood echogenicity in vitro: the role of platelet and platelet-neutrophil aggregates. Ultrasound Med Biol. 1992;18:851-859.
Insel PA. Analgesic-antipyretics and antiinflammatory agents. In: Goodman Gilman A, Rall TW, Nies AS, Taylor P, eds. The Pharmacologic Basis of Therapeutics. 8th ed. New York, NY: Pergamon Press; 1990:649.
Sigel B, Coelho JCU, Schade SG, Justin J, Spigos DG. Effect of plasma proteins and temperature on echogenicity of blood. Invest Radiol. 1982;17:29-33.
Yuan YW, Shung KK. Ultrasonic backscatter from flowing blood, II: dependence on frequency and fibrinogen concentration. J Acoust Soc Am. 1988;84:1195-1200.
Black IW, Chesterman CN, Hopkins AP, Lee LCL, Chong BH, Walsh WF. Hematologic correlates of left atrial spontaneous echo contrast and thromboembolism in nonvalvular atrial fibrillation. J Am Coll Cardiol. 1993;21:451-457.
Fatkin D, Herbert E, Feneley M. Hematologic correlates of spontaneous echo contrast in patients with atrial fibrillation and implications for thromboembolic risk. Am J Cardiol. 1994;73:672-676.
Erbel R, Stern H, Ehrenthal W, Schreiner G, Treese N, Kramer G, Thelen M, Schweizer P, Meyer J. Detection of spontaneous echocardiographic contrast within the left atrium by transesophageal echocardiography: spontaneous echocardiographic contrast. Clin Cardiol. 1986;9:245-252.
Kearney K, Mahony C. Effect of aspirin on spontaneous contrast in the brachial veins of normal subjects. Am J Cardiol. 1995;75:924-928.
Shung KK, Yuan YW, Fei DY. Effect of flow disturbance on ultrasonic backscatter from blood. J Acoust Soc Am. 1984;75:1265-1272.
Twersky V. Low frequency scattering by correlated distributions of randomly oriented particles. J Acoust Soc Am. 1987;81:1609-1618.
Mo LYL, Cobbold RSC. A unified approach to modeling the backscattered Doppler ultrasound from blood. IEEE Trans Biomed Eng. 1992;39:450-461.
Shung KK. In vitro experimental results on ultrasonic scattering in biological tissues. In: Shung KK, ed. Ultrasonic Scattering in Biological Tissues. Boca Raton, Fla: CRC Press; 1993:295.
Usami S, Chien S, Gregersen MI. Viscometric characteristics of blood of the elephant, man, dog, sheep and goat. Am J Physiol. 1969;217:884-890.
Wickham LL, Bauersachs RM, Wenby RB, Sowemimo-Coker S, Heiselman HJ, Elsner R. Red cell aggregation and viscoelasticity of blood from seals, swine and man. Biorheology. 1990;27:191-204.
Haradin AR, Weed RI, Reed CF. Changes in physical properties of stored erythrocytes: relationship to survival in vivo. Transfusion. 1969;9:229-237.
Engeset J, Stalker AL, Matheson NA. Effects of dextran 40 on erythrocyte aggregation. Lancet. 1966;1:1124-1127.
Engeset J, Stalker AL, Matheson NA. Effects of dextran 40 on red cell aggregation in rabbits. Cardiovasc Res. 1967;1:379-384.
Jan K-M, Usami S, Chien S. The disaggregation effect of dextran 40 on red cell aggregation in macromolecular suspensions. Biorheology. 1982;19:543-554.
Maeda N, Shiga T. Inhibition and acceleration of erythrocyte aggregation induced by small macromolecules. Biochim Biophys Acta. 1985;843:128-136.
Kroemer H, Haass A, Muller K, Jager H, Wagner EM, Heimburg P, Klotz U. Haemodilution therapy in ischaemic stroke: plasma concentrations and plasma viscosity during long-term infusion of dextran 40 or hydroxyethyl starch 200/0.5. Eur J Clin Pharmacol. 1987;31:705-710.
Carter C, Fisher TC, Hamai H, Johnson CS, Meiselman HJ, Nash GB, Stuart J. Haemorheological effects of a nonionic copolymer surfactant (poloxamer 188). Clin Hemorheol. 1992;12:109-120.
Colbassani HJ, Barrow DL, Sweeney KM, Bakay RAE, Check IJ, Hunter RL. Modification of acute focal ischemia in rabbits by poloxamer 188. Stroke. 1989;20:1241-1246.
Justicz AG, Farnsworth WV, Soberman MS, Tuvlin MB, Bonner GD, Hunter RL, Martino-Saltzman D, Sink JD, Austin GE. Reduction of myocardial infarct size by poloxamer 188 and mannitol in a canine model. Am Heart J. 1991;122:1671-1680.
Audibert G, Donner M, Lefevre JC, Stoltz JF, Laxenaire MC. Rheologic effects of plasma substitutes used for preoperative hemodilution. Anesth Analg. 1994;78:740-745.
Andersen JR. Disturbances of blood flow and body fluids. In: Muir’s Textbook of Pathology. London, UK: Edward Arnold; 1980:226.
Dintinfass L. The rheology of blood in vascular disease. J R Coll Physicians Lond. 1971;5:231-240.
Dormandy JA. Haemorheological aspects of thrombosis. Br J Haematol. 1980;45:519-522.
Thomas DP. Pathogenesis of venous thrombosis. In: Colman RW, Hirsh J, Marder VJ, Salzman EW, eds. Haemostasis and Thrombosis: Basic Principles and Clinical Practice. Philadelphia, Pa: JB Lippincott; 1982: 820.
Shung KK, Cloutier G, Lim CC. The effects of hematocrit, shear rate and turbulence on ultrasonic Doppler spectrum from blood. IEEE Trans Biomed Eng. 1992;39:462-469.
Klein AL, Murray RD, Black IW, Chandra S, Grimm RA, D’Sa AP, Leung DYC, Miller D, Morehead AJ, Vaughn SE, Thomas JD. Integrated backscatter for quantification of left atrial spontaneous echo contrast. J Am Coll Cardiol. 1996;28:222-231.
Lockwood GR, Ryan LK, Foster FS. A 45 to 55 MHz needle-based ultrasound system for invasive imaging. Ultrason Imaging. 1993;15:1-13.