Marys Medicine

Ar ¥ vol 10 ¥ nr 5

AR • Vol 10 • Nr 5 15.01.2001 16:11 Uhr Seite 232 Table 1 (left): The correlation between the concentration of the hemo- globin and percent of hema- tocrit and either the ery- throcyte sedimentation rate (ESR) or the variables of the test (VR, AR, EP) in 380 patients with ischemic vas- cular diseases and in the controls (VR = vacuum radius, AR = aggregation radius, EP = erythrocyte per- PATIENTS AND METHODS
Table 2 (right): A significant correlation is Previous studies have shown that there is an present between the three increased red blood cell aggregability in patients PATIENTS AND CONTROLS
variables of the erythrocyte with ischemic heart [1-5] and brain[6-9] disease.
Consent was obtained from all the individuals test (EAAT) (VR, AR, EP) and This augmented aggregation might have detri- who participated in the present study according either fibrinogen concentra- mental rheological consequences that lead to to the approval of the local ethics committee.
tion or the erythrocyte sedi- the need for therapeutic interventions for reduc- Included were two groups of patients with ische- mentation rate (ESR) in 380 ing the degree of these cellular interactions.[10- mic vascular diseases and one of healthy controls.
patients and controls. 16]. There are various reasons for the appearance Group assignment was as follows: Figure 1 (bottom): of increased erythrocyte aggregation in patients A typical picture from a with ischemic vascular diseases, among which 1. Ischemic heart disease: patients with chest patient with increased ery- are the presence of an acute phase response and pain who were admitted for observation and throcyte adhesiveness/ag-gregation (left) and a con- an increased synthesis of adhesive proteins,[17] in whom acute myocardial infarction was trol (right). hyperlipidemia,[18] and changes in the erythro- ruled out. Documentation of ischemic heart cyte's membrane.[19] We have recently adapted disease was by means of a history of myocar- a simple slide technique and image analysis to dial infarction (MI), coronary artery bypass reveal the erythrocyte adhesiveness/aggrega- graft surgery (CABG), present or past positive tion state in the peripheral blood.[20-25] The pre- coronary angiography, or a positive thallium sent study was conducted in patients with docu- mented ischemic heart and brain disease. We 2. Ischemic neurological event: patients with an could note the expected significant correlation acute transient ischemic attack or ischemic between the results of this test and either fib- stroke in whom a hemorrhagic stroke was rinogen concentration or the erythrocyte sedi- ruled out by means of computerized tomog- mentation rate. However, an unexpected and significant increment in the EAA state was found 3. Healthy controls: members of the medical in a subgroup of patients who had normal fib- staff and members of a health club who vol- rinogen concentrations and a normal rate of ery- unteered to participate in the study.
throcyte sedimentation, suggesting a potentialadvantage of the erythrocyte adhesive- Excluded from the present study were individuals ness/aggregation test (EAAT) in unmasking rhe- who had any underlying inflammatory or malig- ological information that is not provided by nant disease as well as those who were on either the measurement of clottable fibrinogen steroidal or non-steroidal anti-inflammatory or the erythrocyte sedimentation rate.
medications (except for aspirin up to a dose of 325mg per day). In addition, we excluded individualswho had any traumatic or infectious conditionincluding angiography, surgery or MI during thethree months preceding study recruitment.
VARIABLES OF THE ACUTE PHASE RESPONSE
The blood count was performed by means of theCoulter STKS electronic cell analyzer, the erythro-cyte sedimentation rate by the method of West-ergren [26] and the concentration of fibrinogenby the method of Clauss [27] and a Sysmex x6000analyzer.
232 Applied Rheology
September/October 2000 AR • Vol 10 • Nr 5 15.01.2001 16:11 Uhr Seite 233 2.2.2 PREPARATION OF PERIPHERAL BLOOD
to Class 1, which we name Erythrocyte Percent- age (EP). This is the proportion of image area cov- The guidelines for the preparation of the periph- ered by erythrocytes. The two-point statistics are eral blood slides were reported elsewhere.[28] described by the probability p1(r) of a pixel
Briefly, blood was drawn into a syringe contain- belonging to Class 1, given that it is a distance r
ing sodium citrate (one volume of 3.8% sodium from a pixel Class 1, and similarly the probability citrate and three volumes of whole blood). Sev- p2(r) of a pixel belonging to Class 2, given that it
eral large drops of blood were placed on a slide is a distance r from a pixel of Class 2. These prob-
that was held for 2 to 3 seconds at an angle of 45° abilities are equal to 1 for r = 0 and decrease as
so that the blood could run down by gravity, leav- functions of r. As r grows without bound they
ing a fine film. The slides were then dried at room tend to EP and 1-EP, respectively. The typical size temperature while in a completely horizontal of erythrocyte aggregates and of plasma spaces position. Automatic staining was performed by may be quantified by the rate of decrease of the means of the HEMA TEK slide stainer (AMES) and functions p1(r) and p2(r). In the following discus-
a HEMA TEK bloc colorant stain pack (Bayer Diag- sion we will deal with p1, where it is understood that the treatment of p2 is similar. If, for exam-ple, the erythrocyte aggregates are very small, 2.2.3 INFLAMMATION METER (INFLAMET™)
the probability p1(r) will decrease very fast as a
For analysis of the slides, we used an image function of r because at a quite a short distance
analysis system (INFLAMET™) [20-25] consisting from a point of Class 1, a pixel has no longer a of a Pentium Win 95 equipped with a Matrox marked preference to belong to Class 1. A typical Meteor colour frame grabber, a colour CCD cam- distance scale for the size of aggregates may be era and a microscope which was operated at x defined by the negative inverse of the derivative 200 magnification, resulting in an image resolu- of p1 at r = 0. The derivative must be computed
tion of 0.4 micron per pixel. Nine images were numerically, which we did using the following taken from each slide. The fields of view were chosen systematically to sample differentregions on the slide. Each image is processed sep- • Choose a threshold somewhat smaller than 1 arately and the outputs are then averaged to • Find r such as p ) = threshold form the final slide outputs. The nine fields of • Estimate p1'(0) as: view cover a total area of 0.6 mm2.
TION TEST (EAAT)
• The desire distance scale is: The erythrocyte adhesiveness/aggregation statein the peripheral blood was determined by usingthe same image analysis system (INFLAMET™).
We have used color characteristics to classifyimage pixels into two classes: In practice we chose threshold = 0.7 since downto this value p1 appears to be linear. We define Class 1: Aggregates of erythrocytes the aggregation radius (AR) as r0 - differing by a Class 2: Everything else (plasma, platelets, constant factor of 0.3 from above. Vacuum Radius (VR) is the analogous quantity defined forclass 2.
A description of one-point and two-point statis-tics for this classification turns out to require very 2.2.5 STATISTICAL ANALYSIS
few parameters. The main reason for this is that Means ± SD, Student's t-test and the Pearson the image statistics are homogenous (position- coefficient of correlation were carried out by independent) and isotropic (direction-indepen- means of the SPSS package. By using the paired dent). The one point statistics are completely t-test, we compared a subgroup of patients who described by the probability of a pixel belonging had either normal concentrations of fibrinogen September/October 2000 AR • Vol 10 • Nr 5 15.01.2001 16:11 Uhr Seite 234 or a normal erythrocyte sedimentation rate to A significant difference in all healthy controls who presented with exactly the three variables of the ery- same values. Thus, a patient with a fibrinogen ness/aggregation test exists concentration of 292 mg/dL was compared to a between patients with healthy individual with the same value, etc.
ischemic heart and brain disease and the controls despite of their being no dif- ference in the concentration In the present study, we examined a total of 134 of fibrinogen between the patients with an ischemic heart disease (age groups (paired t-test; Hb = hemoglobin, HCT = hemat- 62±13 years; 89 males and 45 females), 72 with an ocrit, ESR = erythrocyte sedi- ischemic neurological event (age 70 ± 12 years; 39 mentation rate, Fg = fib- males and 33 females) and 174 healthy individu- rinogen; all values are given in mean ± SD). als (age 39 ± 17 years, 92 males and 82 females).
The correlation between the three variables ofthe erythrocyte adhesiveness/aggregation test,i.e., the VR, AR and EP, and the hemoglobin con-centration and the hematocrit values in 380 indi-viduals (patients and controls) are shown in Table 1. These correlations were also significant A highly significant differ- when they were applied to the erythrocyte sedi- ence in all three variables of the erythrocyte adhesive- mentation rate (Table 1).
There was a highly significant correla- (EAAT) exists between tion between the three variables of the EAAT and patients with ischemic heart the concentration of fibrinogen as well as the ery- and brain disease and the controls despite of there throcyte sedimentation rate (Table 2). However, being no significant differ- when the results obtained in a subgroup of ence in their erythrocyte patients in whom the concentration of fibrino- gen was found to be within normal limits (i.e.,200-400 mg/dL) were compared to the normalvalues in the controls, significant differences inthe variables of the erythrocyte adhesive-ness/aggregation were observed, despite com-parable fibrinogen concentrations (paired t-test,Table 3). Similar differences could also be seen ina subgroup of patients who had not only com-parable fibrinogen concentrations but also exactly the same erythrocyte sedimentation rate A highly significant differ- (Table 4). In an analysis of a smaller group, we ence exists for all three vari- could reproduce the results and show that this is ables of the erythrocyte not an age-related effect (Table 5).
test (EAAT) between patients with ischemic heart and brain disease and the controls despite of there We adapted a simple slide test coupled with a being no significant differ- precise image analysis system to study the ery- ence in their age, erythro- throcyte adhesiveness/aggregation state in the cyte sedimentation rate or peripheral blood of patients with documented (n.s. = not significant). ischemic heart and brain diseases. As expected,a significant correlation was noted between theadhesiveness/aggregation state and the con-centration of fibrinogen in both the patients andcontrols. In effect, fibrinogen has been shown in 234 Applied Rheology
September/October 2000 AR • Vol 10 • Nr 5 15.01.2001 16:11 Uhr Seite 235 the past to be a major determinant in the appear- correlation between the EAAT and fibrinogen ance of aggregated erythrocytes in the peripher- concentrations that was observed when the al blood [29] although other acute phase group was evaluated as a whole is suggestive response proteins could participate as well [17, that this molecule does have some role in the 18]. However, our main intent in conducting this induction and on the maintenance of an study was to determine whether the EAAT is increased cell aggregability. The fact that we capable of disclosing information that is not could increase the aggregability of the cells by available by looking at the results of either clot- using an intravenous infusion of gamma globu- table fibrinogen or the sedimentation rate. To do lin [22] as well as our yet unpublished observa- so, we matched patients and controls who had tions that this aggregability can be attenuated exactly the same within normal limits (200-400 by reducing cholesterol and triglyceride concen- mg/dl) fibrinogen concentrations. It turned out trations in hyperlipidemic individuals, raises the that a clear and significant increment in the EAAT possibility that we are facing a kind of a "sum- values could be seen in the group of patients with marizing" effect of various aggregatory forces ischemic vascular diseases as opposed to the con- that operate on the cells in the peripheral blood trols (Table 3). In a further sub-analysis, we could of a given patient.
show the same results for patients and controls The use of image analysis to detect the who had an identical erythrocyte sedimentation aggregability of red blood cells is not new and has rate: again, a highly significant increment in the been reported in the past [33]. The novel EAAT existed in the patients as opposed to the approach is that we use no dilution or washing controls. Our conclusion is that the direct obser- steps and that the slide test is extremely easy to vation of the aggregated erythrocytes made pos- perform. At the moment we do not know if the sible by using the slide test discloses information aggregation radius and the vacuum radius have that is not demonstrated by either clottable fib- any physiological significance except the idea rinogen concentrations or the erythrocyte sedi- that they give for the typical size of erythrocyte mentation rate in some of the patients.
aggregates and plasma spaces, respectively.
There are multiple causes that can However, we are currently investigating the rela- increase the adhesiveness and aggregation of tions between the morphology of the aggrega- red blood cells in the peripheral blood of patients tion and the inter erythrocytic adhesion forces. It with ischemic vascular disorders, including pro- remains to see whether the most round and cir- teins of the acute phase response [17, 18] hyper- cular aggregates represent the most sticky cells.
lipidemia [30] and changes in the composition of This might have consequences in terms of flow the erythrocyte membrane [19, 31]. The potential in the microcirculation.
additional diagnostic value of our test is in that The present study was performed in while the above-mentioned factors can acceler- patients with both ischemic heart and brain dis- ate the sedimentation rate, our test was clearly ease. We assume that both have similar etiolo- positive in a group of patients in whom the ery- gies in terms of the presence of atherosclerosis throcyte sedimentation rate was observed as smoldering inflammation as well as similar risk being within normal limits. Thus, our slide test factors. However, we could reproduce the results might be a sensitive test for detecting the pres- in a group of patients with ischemic brain disease ence of a smoldering inflammatory response in solely (to be published elsewhere). Our present a group of individuals with relatively normal fib- study favors, therefore, the notion that the find- rinogen concentrations or erythrocyte sedimen- ings are not confined to a single clinical entity tation rates. Although sensitive CRP assays could and might be related to the underlying patho- be used for this purpose, the importance of directly diagnosing the presence of red blood cell Finally, one has to be aware that various hyperaggregability [32] lies in the potential ther- optical methodologies have been used to quan- apeutic applications that might result from this titate the aggregability of red blood cells includ- information [16].
ing the erythroaggregometer (Regules, France), Although the mechanisms of increased Laser-assisted optical rotational (cell analyzer, erythrocyte adhesiveness/aggregation were not Mechatronics, Netherlands) and the fully auto- investigated in the current work, the significant matic erythrocyte aggregometer (Myrene, Ger- September/October 2000 AR • Vol 10 • Nr 5 15.01.2001 16:11 Uhr Seite 236 many) [34]. However, we introduced our slide Tanahashi N, Gotoh F et al.: Enhanced Erythro-cyte Aggregability in Occlusive Cerebrovascular test due to its simplicity, low cost and a potential Disease, Stroke 20 (1989) 1202-1207.
of becoming a real time (almost bed-side) Kee DBJr, Wood JH: Rheology of the Cerebral Cir- methodology. The question was therefore, is it culation, Neurosurgery 15 (1984) 125-131.
superior to the indirect determination that one [10] Atkinson RP: Ancrod in the Treatment of Acute can obtain by using the westergen sedimenta- Ischemic Stroke. A Review of Clinical Data, Cere- tion rate [35].
brovasc. Dis 8 Suppl 1 (1998) 23-28.
In conclusion, we have demonstrated Ehrly AM: Improvement of the Flow Properties that a simple slide test and image analysis of Blood: a New Therapeutical Approach in enables the detection of the presence of aggre- Occlusive Arterial Disease, Angiology 27 (1976) gated red blood cells in the peripheral blood of Walzl M, Lechner H et al.: Improved Neurologi- patients with ischemic heart and brain diseases.
cal Recovery of Cerebral Infarctions After Of interest was the discovery that the test Plasmapheretic Reduction of Lipids and Fibrino- remained positive even when fibrinogen con- gen, Stroke 24 (1993) 1447-1451.
centrations or the erythrocyte sedimentation Tanahashi N, Fukuuchi Y et al.: Ticlopidine rate were comparable to those obtained in a con- Improves the Enhanced Erythrocyte Aggregabil- trol group of healthy volunteers. Thus, the EAAT ity in Patients With Cerebral Infarction, Stroke 24 might be useful for unmasking relevant infor- (1993) 1083-1086.
mation that relates to the acute phase response Beigel Y, Fuchs J et al.: Lovastatin Therapy in that is not always given by either determining Hypercholesterolemia: Effect on Fibrinogen,Hemorrheologic Parameters, Platelet Activity, the concentration of clottable fibrinogen or the and Red Blood Cell Morphology, J Clin. Pharma- erythrocyte sedimentation rate.
col. 31 (1991) 512-517.
Branchi A, Rovellini A et al.: Effect of Three Fibrate Derivatives and of Two HMG-CoA Reduc- We are indebted to Ms. Esther Eshkol for editor- tase Inhibitors on Plasma Fibrinogen Level in ial assistance.
Patients With Primary Hypercholesterolemia,Thromb. Haemost. 70 (1993) 241-243.
[16] Dujovne CA, Harris WS et al.: Effect of Atorvas- tatin on Hemorheologic-Hemostatic Parame- Chien S: Blood Rheology in Myocardial Infarction ters and Serum Fibrinogen Levels in Hyperlipi- and Hypertension, Biorheology 23 (1986) 633- demic Patients, Am. J. Cardiol. 85 (2000) 350-353.
Weng X, Cloutier G et al.: Influence of Acute- Boss N, Wietelmann H et al.: Red Blood Cell Phase Proteins on Erythrocyte Aggregation, Am Aggregation in Men With Coronary Artery Dis- J Physiol. 271 (1996) H2346-H2352.
ease, Eur. J Cardiol. 12 (1980) 47-54.
Weng X, Roederer GO et al.: Contribution of Rainer C, Kawanishi DT et al.: Changes in Blood Acute-Phase Proteins and Cardiovascular Risk Rheology in Patients With Stable Angina Pectoris Factors to Erythrocyte Aggregation in Nor- As a Result of Coronary Artery Disease, Circula- molipidemic and Hyperlipidemic Individuals, tion 76 (1987) 15-20.
Thromb. Haemost. 80 (1998) 903-908.
Arntz HR, Perchalla G et al.: Blood Rheology in [19] Martinez M, Vaya A et al.: Alterations in Ery- Acute Myocardial Infarction: Effects of High- throcyte Aggregability in Diabetics: the Influ- Dose I.V. Streptokinase Compared to Placebo, ence of Plasmatic Fibrinogen and Phospholipids Eur. Heart J. 13 (1992) 275-280.
of the Red Blood Cell Membrane, Clin. Hemorhe- Dintenfass L, Forbes CD: Effect of Fibrinogen on ol. Microcirc. 18 (1998) 253-258.
Aggregation of Red Cells and on Apparent Vis- [20] Berliner S, Shapira I et al.: Combined Leukocyte cosity of Artificial Thrombi in Haemophilia, and Erythrocyte Aggregation in the Peripheral Myocardial Infarction, Thyroid Disease, Cancer Venous Blood During Sepsis. A Clue to the Pres- and Control Systems: Effect of ABO Blood ence of a Commonly Shared Adhesive Protein(s), Groups, Microvasc. Res. 9 (1975) 107-118.
Int. J. Clin. Lab. Res. 30 (2000) 27-31. Grotemeyer KH: Abnormal Hemorheological Rotstein R, Zeltser D et al.: The Inflammation Parameters in Vertebrobasilar-Insufficiency, Meter (INFLAMET). A New Diagnostic Approach Acta Neurol. Scand. 81 (1990) 529-532.
to Reveal the Presence of an Inflammatory Fisher M, Meiselman HJ: Hemorheological Fac- Response and for the Assessment of Its Intensi- tors in Cerebral Ischemia, Stroke 22 (1991) 1164- ty, Isr. Med. Assoc. J. 2 (2000) 476-477. [22] Fusman R, Zeltser D et al.: INFLAMET: an Image 236 Applied Rheology
September/October 2000 AR • Vol 10 • Nr 5 15.01.2001 16:11 Uhr Seite 237 Analyzer to Display Erythrocyte Adhesive- [29] Fabry TL: Mechanism of Erythrocyte Aggrega- ness/Aggregation, Eur. J. Int. Med. (in press tion and Sedimentation, Blood 70 (1987) 1572- [23] Maharshak N, Kassirer M et al.: The Inflamma- [30] Vaya A, Martinez M et al.: Red Blood Cell Aggre- tion Meter: Novel Technique to Detect the Pres- gation and Primary Hyperlipoproteinemia, ence of Infection/Inflammation in Patients Thromb. Res. 72 (1993) 119-126.
Without Leukocytosis but With an Increased Hadengue A, Razavian SM et al.: Influence of Sial- Leukocyte Adhesiveness/Aggregation, Acta ic Acid on Erythrocyte Aggregation in Hypercho- Haematol. (in press). lesterolemia, Thromb. Haemost. 76 (1996) 944- [24] Rotstein R, Zeltser D et al.: An Inflammation Meter to Reveal the Presence and Extent of [32] Weng X, Cloutier G et al.: Contribution of the - Inflammation in Elderly Patients, J. Am. Ger. Soc.
455G/A Polymorphism at the Beta-Fibrinogen (letter to editor, in press).
Gene to Erythrocyte Aggregation in Patients [25] Rogowski O, Zeltser D et al.: Correlated Expres- with Coronary Artery Disease, Throm. Haemost.
sion of Adhesive Properties for Both White and 82 (1999) 1406-1411.
Red Blood Cells During Inflammation. A Clue to [33] Chen S, Eldor A et al.: Enhanced Aggregability of the Presence of Denominating Proteins, Biorhe- Red Blood Cells of Beta-Thalassemia Major ology (in press). Patients, Am. J. Physiol. 270 (1996) H1951-H1956.
[26] International Committee for Standardization in [34] Zhao H, Wang X et al.: Comparison of Three Opti- Hematology. Recommendation of Measure- cal Methods to Study Erythrocyte Aggregation, ment of Erythrocyte Sedimentation Rate of Clin. Hemorheol. Microcirc. 21 (1999) 297-302.
Human Blood., Immunochemistry 2 (1965) 235- [35] Danesh J, Collins R et al.: Haematocrit, Viscosity, Erythrocyte Sedimentation Rate: Meta-Analyses [27] Clauss A: Gerinnungsphysiologische Schnell- of Prospective Studies of Coronary Heart Dis- methode zur Bestimmung des Fibrinogens, Acta ease., Eur. Heart. J. 21 (2000) 515-520.
Haematol. Basel 17 (1957) 237-246.
[28] Berliner S, Fishelson Z et al.: The Phenomenon of Leukergy: Induction and Detection of LeukocyteAggregation in Whole Human Blood., J. Lab. Clin.
Med. 109 (1987) 575-582.
September/October 2000

Source: http://vg09.met.vgwort.de/na/68378abb7e42141a27e1d2598860ba?l=http://www.ar.ethz.ch/ARONLINE_FREE/10_231.pdf