THROMBOSIS RESEARCH 66; 649-656,1992 0049-3848192 $5.00 + .OOPrinted in the USA. Copyright (c) 1992 Pergamon Press Ltd. All rights reserved.
EFFECT OF HEPARIN ON IN VITRO PLATELET REACTIVITY IN CARDIAC SURGICAL PATIENTS - A COMPARATIVE ASSESSMENT BY WHOLE BLOOD PLATELET AGGREGOMETRY AND HAEMOSTATOMETRY
L.C.H.
John,
G.M.
Rees
and
I.B.
Kovac
8
Departments of Cardiothoracic Surgery and the Thrombosis Hospital, London EC1 7BE, England St. Bartholomews
Unit,
(Received 31 .1.1992; accepted in revised form 13.4.1992 by Editor M.F. Scully)
ABSTRACT
The in vitro effect of heparin on platelet reactivity was assessed simultaneously by haemostatometry (response to shear stress) and whole blood platelet aggregometry response to collagen (WBPA). From each blood sample a ratio (H for haemostatometry and M and for WBPA) showiffg platelet reactivity in tffe prese IRce or absence of heparin (5 U/ml) was calculated. A value < 1 represented a proaggregatory effect and > 1 an inhibitory effect. Non-anticoagulated blood samples obtained from 290 cardiac surgical patients were tested by haemostatometry and titrated whole blood samples from 100 patients with aggregometry. Haemostatometry demonstrated a proaggregatory effect of heparin in 8.6% (25) and an inhibitory effect in 91.4% (265). Assessed by WBPA, heparin was proaggregatory in 41-46% and inhibitory in 54-59%. In the 100 patients tested by both methods there was a significant correlation between the findings with the two techniques (1:= 0.46, p < 0.0001). A wide individual variation in the platelet effect of heparin was demonstrated. This variation appeared greater and a higher proportion showed inhibition when blood was tested by haemostatometry. KEY WORDS: Platelet
function tests, Heparin.
649
HEPARIN EFFECT ON PLATELETS
Vol. 66, No. 6
INTRODUCTION The effect of heparin on platelets is controversial. In vitro heparin has been reported to induce platelet aggregation (1,Z) and potentiate the effect of physiologically relevant platelet aggregating agents (3). However there are also reports of heparin inhibiting platelet aggregation (4). In vivo, the skin bleeding time, which is still physiologically the most relevant in vivo platelet function test is prolonged after heparin administration (5). In some patients heparin treatment is followed by thrombocytopenia which may be associated with thrombosis manifesting itself either or both in haemorrhagic and in arterial thrombotic complications (6,7). All previous in vitro platelet function require anticoagulation. In the majority of studies heparin was added to platelets itrated plasma or whole blood. However, by reducing plasma fi2? and also by a direct effect on the platelet membrane, citrate creates an unphysiological situation where the response of platelets to stimuli is distorted (8). The skin bleeding time suffers from serious limitations such as lack of reproducibility and inaccuracy. In an attempt to clarify the overall effect of heparin on platelets, a new technique, haemostatometry was employed which has 2 distinct advantages over the in vitro platelet function tests in common use. Firstly, it uses non-anticoagulated, undiluted whole blood and secondly, aggregation of platelets is brought about by a physiologically relevant stimulus of arterial thrombogenesis, i.e. haemodynamic shear-forces. Beside haemostatometry, whole blood platelet (impedance) aggregometry was used. The study population consisted of cardiac surgical patients and was relatively large in order to examine the possibility of individual variation in the effect of heparin. METHODS 1) Haemostatometry (Platelet reactivity to shear Stress). The instrument (haemostatometer) has been previously described in detail (9,lO). 2.5 mls of non-anticoagulated blood is perfused through polyethylene tubing by paraffin oil displacement a technique that was shown to cause no measurable platelet dambge. Holes are punched through the tubing resulting in "bleedingfit and a pressure drop. The pattern of pressure recovery shows the formation of haemostatic plugs in the holes. The initial recovery of the pressure represents platelet plug formation due to activation of platelets by shear stress and release of adenosine diphosphate from the shear-damaged platelets and red blood cells. The later recovery is due to stabilisation of the platelet plug secondary to thrombin generation, the subsequent increase of platelet aggregation and fibrin formation so that it can resist the increasing pressure. The haemostasis curve is recorded and its area analysed as H (mm Hg.s). The greater the value of H the greater the inhibition of platelet function. From each of 290 subjects 2 haemostasis measurements were made, with or without heparin (MONOPARIN, CP Pharmaceuticals Ltd, Wrexham, UK. 5 U/ml). This heparin concentration is approximately
HEPARINEFFECTONPLATELETS
Vol.66, No.6
651
equal to that achieved during cardiopulmonary bypass. The ratio assessing the effect of heparin on platelets ,was defined as H =H is the H measR hepjHbas' where Hhep
(HR)
ured from the heparinised
from
the non heparinised on platelet
sample and Hbas is the H measured
sample. HR = 1 implies no effect of heparin
function; HR < 1 a proaggregatory
effect and HR
> 1 a platelet inhibitory effect. The greater the HR, the greater the inhibition of platelet function by heparin. 2) Whole blood platelet
aggregometry
(WBPA).
From each of 100 subjects 2 WBPA measurements were made (Chronolog Dual Channel) using titrated blood (1:lO) diluted with phosphate buffered saline (1:l). A heparinised sample (5 U/ml) was compared to a non heparinised, both with added collagen (Chrono-Par; 1.3 pg/ml). Both the maximum rate of impedance changes (M, n/min) and the total change of impedance over 10 minutes (I, n) was measured. The ratio assessing the effect of heparin on platelets was defined as, 'R = Ihep'Ibas' where
Ihep is the I measured
Ibas is the I measured
from the heparinised
from the non heparinised
where M
sample and
sample.
is the M measured from the non-heparinised hep IR or MR = 1 implies no effect of heparin on platelet
tion; IR or MR < 1 a proaggregatory > 1 a platelet
inhibitory
the greater the inhibition
sample. func-
effect and IR or MR
effect. The greater the IR or MR, of platelet
function by heparin.
Blood sampling: All blood samples were taken from the antecubital vein of the subject using a 21 guage "Butterfly set" (VENISYSTEMS, Abbott Laborities Ltd., U.K.). For haemostatometric measurements 2 consecutive 2.5 mls syringes were withdrawn, 1 with added heparin (MONOPARIN) to give a final concentration of 5U/ml, and 1 without. These were placed immediately into the measuring channels of 2 For WBPA measurements blood was withdrawn (at haemostatometers. the same venepuncture as the haemostatometry samples when directly compared) into a 5 ml syringe containing 0.5 mls 3.8% tri-sodium citrate solution. After 30 minutes to allow for reversible platelet dysfunction 0.75 mls of titrated blood was added to each of 2 cuvettes containing 0.75mls of phosphate buffered saline. To one of the cuvettes was added heparin (MONOPARIN) to give a final concentration of 5 U/ml. The cuvettes were placed in the 2 meas-
HEPARIN EFFECT ON PLATELETS
Vol. 66, No. 6
uring channels of the aggregometer and after 10 minutes to allow for equilibriation 2 I.ILcollagen (1.3 pg/ml) was added to each. Study population: All were cardiac surgical patients, either before coronary bypass surgery or within 1 year postoperatively. 79% were male with an age range of 26-78 years (mean 60 years). Medications taken by the 290 patients measured by haemostatometry were aspirin (75%), calcium antagonists (46%), dipyridamole (10%) and by the 100 patients measured by aggregometry, aspirin (70%), calcium antagonists (47%) and dipyridamole (5%). Despite being on antiplatelet medication, cardiac surgical patients were used as the study population as this is a group in which an effect of heparin on platelet function is particularly relevant. All were about to or had previously undergone cardiopulmonary bypass. The results were analysed to assess the possible contribution of antiplatelet medication to the measured effect of heparin on platelets. TABLE 1 Patient Characteristics And Antiplatelet ostatometry Study Group. OVERALL GROUP (n=290) 79 60 75
Male (%). Mean age (yrs). Proportion on aspirin (%). Mean aspirin dose (mg/day). Proportion on calcium antagonists (%). Proportion on dipyridamole (%).
Medications
In The Haem-
PROAGGREGATORY (n=25) 80 60.8 68
INHIBITORY (n=265) 79 59.7 76
179
150
181
46
36
47
10
20
9
TABLE II Patient Characteristics Aggregometry Group.
And Antiplatelet
OVERALL GROUP (n=lOO) Male (%). Mean age (yrs). Proportion on aspirin (%). Mean aspirin dose (mg/d). Proportion on calcium antags.( Proportion on dipyridamole (%) .
Medication
PROAGGREGATORY MR(41) IR(46)
In The
INHIBITORY MR(59) IR(54)
75 56.9 70
73 59.1 76
76 56.3 76
78 55.3 66
76 56.3 65
179
174
172
175
177
47
46
44
47
47
5
5
5
5
5
Vol. 66, No. 6
HEPARIN EFFECT ON PLATELETS
653
There was no significant difference in any characteristic between those with a proaggregatory or an inhibitory result by either method (Chi squared analysis).
RESULTS
The variation of the relevant parameters HR, MR and IR in the study population is shown in Figure 1.
“‘3
MU
28 -
“3
4.2
7.0
3.9
6.5
3.6
6.0
22
3.3
55
20-
3.0
5.0
2624--
!. '! ..
18 -
:.
16
.:I. .::
:’
2.7
3.
2.4
4.0
14..
12. IO8642-
:i: ... .. :f. 1& *... y.g. . .. . . .: p:;
.&: .::txz. .:y.@;. *....
1.2 -9 -6 -3
4. 5
... ......... ...... ............ ........ ..... ....... .............. ......... ...... .... ... ....
3.5 3.0 2.5
..
2.0
.. ?. ...... .......... .............. ...... ........... ....... ....
1.5 1.0 -5
‘::’
Figure 1. Comparison of the variation of the haemostatometric parameter HR (in 290) and WBPA parameters (MR and IR) in 100 cardiac surgical patients.
Using haemostatometry (n = 290): 8.6% (25) demonstrated a proaggregatory effect of heparin and 91.4% (265)either no effect or an inhibition of platelet function. There was a markedly wide 32.8% (95) had values > 10. variation in H With WBPARin = 100) and titrated blood, using the parameter 41% (41) showed a proaggregatory effect with heparin and 59% MR'
654
HEPARIN EFFECT ON PLATELETS
Vol. 66, No. 6
(59) no effect or inhibited platelet function. Using the parameter IR the respective proportions were 46% (46) and 54% (54). The patient characteristics and antiplatelet medications are shown in Tables I and II. No significant difference was shown between the characteristics of the sugroup demonstrating a proaggregatory platelet effect with heparin and that with an inhibitory effect (Chi squared analysis). In the 100 patients assessed by both techniques there was a significant correlation (Spearman Rank test) between the haemostatometric parameter HR and the aggregometric parameters MR and IR (r = 0.46, p < 0.0001). Reproducibility
of the measurements:
The reproducibility of the techniques was assessed by 10 repeated measurements of platelet sensitivity to heparin in a normal subject on no medication using both haemostatometry and WBPA within a 1 week period. All 10 measurements demonstrated an inhibition of platelet function, the mean values and ranges for H and being 3.0 (1.4-5.4), 1.2 (1.0-1.6) and 1.4 (l.lE;.& respeM8 tively. Although all repeated measurements in this individual gave an uniformly inhibitory result it may be expected that in those with a minimal heparin platelet effect the variation in reproducibility will produce a proaggregatory result on one occasion and an inhibitory on another. However this variation should even out and the proportions demonstrating differing heparin-platelet effects should be valid in a relatively large test population. A ratio value of 1.0 was included for convenience as an inhibitory result. This may have resulted in an overestimate of an inhibitory effect by HR, IR and MR by 0.01% (3 cases), 12% (12 cases) and 3% (3 cases) respectively.
DISCUSSION Although heparin is the most widely used anticoagulant and antithrombotic agent, the effect of the drug on platelet function is still not clear. Heparin induced thrombocytopenia and arterial thrombosis secondary to heparin has long been recognised (6,7). One view is that thrombocytopellia and thrombosis following heparin are different manifestations of the same spectrum with which the drug affects the haemostatic system. An alternative view is that there are two distinct conditions. Firstly, a direct effect which manifests itself shortly following the onset of treatment and secondly, an immune mechanism occuring 5-9 days following the first dose of heparin. However, the most important complication associated with the use of heparin is haemorrhage, the reported incidence ranging between 1% (11) and 32% (12). The majority of studies suggest that the mechanism of the bleeding complication is independent from the anticoagulant effect of the drug (11,13) and is most likely to be secondary to a heparin altered platelet function (14). Although both the inhibitory effect on platelet reaction (the basis of the use of heparin in arterial thrombotic conditions) and the stimulant effect causing platelet aggregation have long been recognised, the contradictions could not be resolved. The
Vol. 66, No. 6
HEPARIN EFFECT ON PLATELETS
655
reason for the conflicting reports may have been the differences in methodology (1) or the relatively small numbers tested for a phenomenon that greatly varies between individuals (proaggregatory in some and inhibitory in others). An individual variation in the effect of heparin on platelet function has been speculated on before. Wide individual variation was found in skin bleeding time following heparin administration (5). Our present findings using two different platelet function tests in a relatively large study population of cardiac surgical patients are consistent with wide individual variation in the effect of heparin on platelet reactivity. The proportion of individuals which exhibit a proaggregatory effect of heparin appears to be smaller (9%) as assessed by the haemostatometer than with whole blood aggregometry (41-46%), and also the degree of variation in the inhibition of platelet reaction appears to be greater. The statistically significant corselation between the results obtained by the two platelet function tests suggests that they both measure the same phenomenon. Whole blood aggregometry requires anticoaaylated and diluted blood. However, citrate by reducing plasma Ca and by a direct effect on the platelet membrane distorts the platelet response (8). Haemostatometry by not requiring an anticoagulant may be a more relevant test of platelet function and especially more adequate for studying the effect of heparin on platelets than platelet aggregometry. In conclusion, we found that in a relatively large population of cardiac surgical patients, a relevant concentration of heparin affected platelet function with a wide individual variation. A possible relevance of this variation in heparin response is in cardiac surgery where heparinisation is a necessity. Post operative blood loss is an important cause of morbidity (15) and has been partly attributed to an acquired inhibition of platelet function (16). A varying platelet inhibition to heparin may have important implications as to individual patients bleeding tendency in this clinical situation. These findings may also provide an explanation for contradictions in the earlier in vitro findings and for the diverse clinical complications of the clinical use of heparin. REFERENCES
1) Brace L.D., Fareed J. An objective assessment of the interaction of heparin and its fractions with human platelets. SEM.THROMB. HEMOST. 11,190-8,1985. 2) Chong B.H., Ismail F. The mechanism of heparin-induced aggregation. EUR.J.HAEMATOL. 43,245-51,1989.
platelet
3) Ruggiero M.,Fedi S.,Bianchini P.,Vannucchi S.,Chiarugi V. Molecular events involved in the proaggregating effect of heparin 802,372-377,1984. on human platelets. BIOCHIMICA.BIOPHYS.ACTA 4) Sobel M. ,Fujimura Y .,Ruggeri M.,Zimmerman T. Platelet glycoprotein Ib is a binding site that participates in platelet aggregation in heparin-associated thrombocytopenia. CLINICAL RESEARCH 34(2) (Abstract:449A),1986.
656
HEPARIN EFFECT ON PLATELETS
Vol. 66, No. 6
5) Heiden D.,Mielke C.H.,Rodvien R. Impairment by heparin of primary haemostasis and platelet (C14) 5-hydroxytryptamine release. BR.J.HAEMATOL. 36,427-36,1977. 6) Kelton J.G. Heparin 173-186,1986. 7) Warkentin ANN.REV.MED.
induced thrombocytopenia.
T.E .,Kelton J.G.Heparin 40,31-44,1989.
HAEMOSTASIS
16,
induced thrombocytopenia.
R.L., 8) Packham M.A.,Bryant N.L., Guccione M.A.,Kinlough-Rathbone Mustard J.F. Effect of the concentration of Ca2+ in the suspending medium on the responses of human and rabbit platelets to aggregating agents. THROMBOS.HAEMOSTAS. 62,968-76,1989. 9) Kovacs I.B., Hutton R.A.,Kernoff P.P.A. Haemostatic evaluation in bleeding disorders from native blood. Clinical experience with the haemostatometer. AM.J.CLIN.PATHOL. 91,271-9,1989. 10) Ratnatunga C.,Rees G.,Kovacs I. Preoperative assessment of haemostatic activity to predict excessive blood loss after cardiopulmonary bypass. ANN.THORAC.SURG. 52,250-7,1991. 11) Salzman E.W.,Deykin D.,Shapiro R.M.,Rosenberg R. Management of heparin therapy, controlled prospective trial. N.ENGL.J.MED. 292, 1047-50,1975. 12) Mant M.J., O'Brien B.D.,Thong K.L.,Hammond G.W.,Birtwhistle V.,Grace M.G. Haemorrhagic complications of heparin therapy. LANCET 1133-1135,1977.
R.
13) Ockelford P.A., Carter C.J.,Cerskus A.,Smith C.A.,Hirsh J. Comparison of the in vivo haemorrhagic and antithrombotic effects of a low Antithrombin III affinity heparin fraction. THROMB.RES. 27,679-690,1982. 14) Levine M.N.,Hirsh J. Haemorrhagic complications lant therapy. SEMIN.THROMB.HEMOSTAS. 12,39-57,1986.
of anticoagu-
15) Love T.R.,Hendren W.G. ,O'Keefe D.D.,Daggett W.M. Transfusion of predonated autologous blood in elective cardiac surgery. ANN. THORAC.SURG. 43,508,1987. 16) Harker L.,Malpass T.W.,Branson H.E. ,Hessel E.A.,Slichter S.A. Mechanism of abnormal bleeding in patients undergoing cardiopulmonary bypass: acquired transient platelet dysfunction associated with selective alpha granule release. BLOOD 56,824-34,198O.
EFFECT OF HEPARIN ON IN VITRO PLATELET REACTIVITY IN CARDIAC SURGICAL PATIENTS - A COMPARATIVE ASSESSMENT BY WHOLE BLOOD PLATELET AGGREGOMETRY AND HAEMOSTATOMETRY
L.C.H.
John,
G.M.
Rees
and
I.B.
Kovac
8
Departments of Cardiothoracic Surgery and the Thrombosis Hospital, London EC1 7BE, England St. Bartholomews
Unit,
(Received 31 .1.1992; accepted in revised form 13.4.1992 by Editor M.F. Scully)
ABSTRACT
The in vitro effect of heparin on platelet reactivity was assessed simultaneously by haemostatometry (response to shear stress) and whole blood platelet aggregometry response to collagen (WBPA). From each blood sample a ratio (H for haemostatometry and M and for WBPA) showiffg platelet reactivity in tffe prese IRce or absence of heparin (5 U/ml) was calculated. A value < 1 represented a proaggregatory effect and > 1 an inhibitory effect. Non-anticoagulated blood samples obtained from 290 cardiac surgical patients were tested by haemostatometry and titrated whole blood samples from 100 patients with aggregometry. Haemostatometry demonstrated a proaggregatory effect of heparin in 8.6% (25) and an inhibitory effect in 91.4% (265). Assessed by WBPA, heparin was proaggregatory in 41-46% and inhibitory in 54-59%. In the 100 patients tested by both methods there was a significant correlation between the findings with the two techniques (1:= 0.46, p < 0.0001). A wide individual variation in the platelet effect of heparin was demonstrated. This variation appeared greater and a higher proportion showed inhibition when blood was tested by haemostatometry. KEY WORDS: Platelet
function tests, Heparin.
649
HEPARIN EFFECT ON PLATELETS
Vol. 66, No. 6
INTRODUCTION The effect of heparin on platelets is controversial. In vitro heparin has been reported to induce platelet aggregation (1,Z) and potentiate the effect of physiologically relevant platelet aggregating agents (3). However there are also reports of heparin inhibiting platelet aggregation (4). In vivo, the skin bleeding time, which is still physiologically the most relevant in vivo platelet function test is prolonged after heparin administration (5). In some patients heparin treatment is followed by thrombocytopenia which may be associated with thrombosis manifesting itself either or both in haemorrhagic and in arterial thrombotic complications (6,7). All previous in vitro platelet function require anticoagulation. In the majority of studies heparin was added to platelets itrated plasma or whole blood. However, by reducing plasma fi2? and also by a direct effect on the platelet membrane, citrate creates an unphysiological situation where the response of platelets to stimuli is distorted (8). The skin bleeding time suffers from serious limitations such as lack of reproducibility and inaccuracy. In an attempt to clarify the overall effect of heparin on platelets, a new technique, haemostatometry was employed which has 2 distinct advantages over the in vitro platelet function tests in common use. Firstly, it uses non-anticoagulated, undiluted whole blood and secondly, aggregation of platelets is brought about by a physiologically relevant stimulus of arterial thrombogenesis, i.e. haemodynamic shear-forces. Beside haemostatometry, whole blood platelet (impedance) aggregometry was used. The study population consisted of cardiac surgical patients and was relatively large in order to examine the possibility of individual variation in the effect of heparin. METHODS 1) Haemostatometry (Platelet reactivity to shear Stress). The instrument (haemostatometer) has been previously described in detail (9,lO). 2.5 mls of non-anticoagulated blood is perfused through polyethylene tubing by paraffin oil displacement a technique that was shown to cause no measurable platelet dambge. Holes are punched through the tubing resulting in "bleedingfit and a pressure drop. The pattern of pressure recovery shows the formation of haemostatic plugs in the holes. The initial recovery of the pressure represents platelet plug formation due to activation of platelets by shear stress and release of adenosine diphosphate from the shear-damaged platelets and red blood cells. The later recovery is due to stabilisation of the platelet plug secondary to thrombin generation, the subsequent increase of platelet aggregation and fibrin formation so that it can resist the increasing pressure. The haemostasis curve is recorded and its area analysed as H (mm Hg.s). The greater the value of H the greater the inhibition of platelet function. From each of 290 subjects 2 haemostasis measurements were made, with or without heparin (MONOPARIN, CP Pharmaceuticals Ltd, Wrexham, UK. 5 U/ml). This heparin concentration is approximately
HEPARINEFFECTONPLATELETS
Vol.66, No.6
651
equal to that achieved during cardiopulmonary bypass. The ratio assessing the effect of heparin on platelets ,was defined as H =H is the H measR hepjHbas' where Hhep
(HR)
ured from the heparinised
from
the non heparinised on platelet
sample and Hbas is the H measured
sample. HR = 1 implies no effect of heparin
function; HR < 1 a proaggregatory
effect and HR
> 1 a platelet inhibitory effect. The greater the HR, the greater the inhibition of platelet function by heparin. 2) Whole blood platelet
aggregometry
(WBPA).
From each of 100 subjects 2 WBPA measurements were made (Chronolog Dual Channel) using titrated blood (1:lO) diluted with phosphate buffered saline (1:l). A heparinised sample (5 U/ml) was compared to a non heparinised, both with added collagen (Chrono-Par; 1.3 pg/ml). Both the maximum rate of impedance changes (M, n/min) and the total change of impedance over 10 minutes (I, n) was measured. The ratio assessing the effect of heparin on platelets was defined as, 'R = Ihep'Ibas' where
Ihep is the I measured
Ibas is the I measured
from the heparinised
from the non heparinised
where M
sample and
sample.
is the M measured from the non-heparinised hep IR or MR = 1 implies no effect of heparin on platelet
tion; IR or MR < 1 a proaggregatory > 1 a platelet
inhibitory
the greater the inhibition
sample. func-
effect and IR or MR
effect. The greater the IR or MR, of platelet
function by heparin.
Blood sampling: All blood samples were taken from the antecubital vein of the subject using a 21 guage "Butterfly set" (VENISYSTEMS, Abbott Laborities Ltd., U.K.). For haemostatometric measurements 2 consecutive 2.5 mls syringes were withdrawn, 1 with added heparin (MONOPARIN) to give a final concentration of 5U/ml, and 1 without. These were placed immediately into the measuring channels of 2 For WBPA measurements blood was withdrawn (at haemostatometers. the same venepuncture as the haemostatometry samples when directly compared) into a 5 ml syringe containing 0.5 mls 3.8% tri-sodium citrate solution. After 30 minutes to allow for reversible platelet dysfunction 0.75 mls of titrated blood was added to each of 2 cuvettes containing 0.75mls of phosphate buffered saline. To one of the cuvettes was added heparin (MONOPARIN) to give a final concentration of 5 U/ml. The cuvettes were placed in the 2 meas-
HEPARIN EFFECT ON PLATELETS
Vol. 66, No. 6
uring channels of the aggregometer and after 10 minutes to allow for equilibriation 2 I.ILcollagen (1.3 pg/ml) was added to each. Study population: All were cardiac surgical patients, either before coronary bypass surgery or within 1 year postoperatively. 79% were male with an age range of 26-78 years (mean 60 years). Medications taken by the 290 patients measured by haemostatometry were aspirin (75%), calcium antagonists (46%), dipyridamole (10%) and by the 100 patients measured by aggregometry, aspirin (70%), calcium antagonists (47%) and dipyridamole (5%). Despite being on antiplatelet medication, cardiac surgical patients were used as the study population as this is a group in which an effect of heparin on platelet function is particularly relevant. All were about to or had previously undergone cardiopulmonary bypass. The results were analysed to assess the possible contribution of antiplatelet medication to the measured effect of heparin on platelets. TABLE 1 Patient Characteristics And Antiplatelet ostatometry Study Group. OVERALL GROUP (n=290) 79 60 75
Male (%). Mean age (yrs). Proportion on aspirin (%). Mean aspirin dose (mg/day). Proportion on calcium antagonists (%). Proportion on dipyridamole (%).
Medications
In The Haem-
PROAGGREGATORY (n=25) 80 60.8 68
INHIBITORY (n=265) 79 59.7 76
179
150
181
46
36
47
10
20
9
TABLE II Patient Characteristics Aggregometry Group.
And Antiplatelet
OVERALL GROUP (n=lOO) Male (%). Mean age (yrs). Proportion on aspirin (%). Mean aspirin dose (mg/d). Proportion on calcium antags.( Proportion on dipyridamole (%) .
Medication
PROAGGREGATORY MR(41) IR(46)
In The
INHIBITORY MR(59) IR(54)
75 56.9 70
73 59.1 76
76 56.3 76
78 55.3 66
76 56.3 65
179
174
172
175
177
47
46
44
47
47
5
5
5
5
5
Vol. 66, No. 6
HEPARIN EFFECT ON PLATELETS
653
There was no significant difference in any characteristic between those with a proaggregatory or an inhibitory result by either method (Chi squared analysis).
RESULTS
The variation of the relevant parameters HR, MR and IR in the study population is shown in Figure 1.
“‘3
MU
28 -
“3
4.2
7.0
3.9
6.5
3.6
6.0
22
3.3
55
20-
3.0
5.0
2624--
!. '! ..
18 -
:.
16
.:I. .::
:’
2.7
3.
2.4
4.0
14..
12. IO8642-
:i: ... .. :f. 1& *... y.g. . .. . . .: p:;
.&: .::txz. .:y.@;. *....
1.2 -9 -6 -3
4. 5
... ......... ...... ............ ........ ..... ....... .............. ......... ...... .... ... ....
3.5 3.0 2.5
..
2.0
.. ?. ...... .......... .............. ...... ........... ....... ....
1.5 1.0 -5
‘::’
Figure 1. Comparison of the variation of the haemostatometric parameter HR (in 290) and WBPA parameters (MR and IR) in 100 cardiac surgical patients.
Using haemostatometry (n = 290): 8.6% (25) demonstrated a proaggregatory effect of heparin and 91.4% (265)either no effect or an inhibition of platelet function. There was a markedly wide 32.8% (95) had values > 10. variation in H With WBPARin = 100) and titrated blood, using the parameter 41% (41) showed a proaggregatory effect with heparin and 59% MR'
654
HEPARIN EFFECT ON PLATELETS
Vol. 66, No. 6
(59) no effect or inhibited platelet function. Using the parameter IR the respective proportions were 46% (46) and 54% (54). The patient characteristics and antiplatelet medications are shown in Tables I and II. No significant difference was shown between the characteristics of the sugroup demonstrating a proaggregatory platelet effect with heparin and that with an inhibitory effect (Chi squared analysis). In the 100 patients assessed by both techniques there was a significant correlation (Spearman Rank test) between the haemostatometric parameter HR and the aggregometric parameters MR and IR (r = 0.46, p < 0.0001). Reproducibility
of the measurements:
The reproducibility of the techniques was assessed by 10 repeated measurements of platelet sensitivity to heparin in a normal subject on no medication using both haemostatometry and WBPA within a 1 week period. All 10 measurements demonstrated an inhibition of platelet function, the mean values and ranges for H and being 3.0 (1.4-5.4), 1.2 (1.0-1.6) and 1.4 (l.lE;.& respeM8 tively. Although all repeated measurements in this individual gave an uniformly inhibitory result it may be expected that in those with a minimal heparin platelet effect the variation in reproducibility will produce a proaggregatory result on one occasion and an inhibitory on another. However this variation should even out and the proportions demonstrating differing heparin-platelet effects should be valid in a relatively large test population. A ratio value of 1.0 was included for convenience as an inhibitory result. This may have resulted in an overestimate of an inhibitory effect by HR, IR and MR by 0.01% (3 cases), 12% (12 cases) and 3% (3 cases) respectively.
DISCUSSION Although heparin is the most widely used anticoagulant and antithrombotic agent, the effect of the drug on platelet function is still not clear. Heparin induced thrombocytopenia and arterial thrombosis secondary to heparin has long been recognised (6,7). One view is that thrombocytopellia and thrombosis following heparin are different manifestations of the same spectrum with which the drug affects the haemostatic system. An alternative view is that there are two distinct conditions. Firstly, a direct effect which manifests itself shortly following the onset of treatment and secondly, an immune mechanism occuring 5-9 days following the first dose of heparin. However, the most important complication associated with the use of heparin is haemorrhage, the reported incidence ranging between 1% (11) and 32% (12). The majority of studies suggest that the mechanism of the bleeding complication is independent from the anticoagulant effect of the drug (11,13) and is most likely to be secondary to a heparin altered platelet function (14). Although both the inhibitory effect on platelet reaction (the basis of the use of heparin in arterial thrombotic conditions) and the stimulant effect causing platelet aggregation have long been recognised, the contradictions could not be resolved. The
Vol. 66, No. 6
HEPARIN EFFECT ON PLATELETS
655
reason for the conflicting reports may have been the differences in methodology (1) or the relatively small numbers tested for a phenomenon that greatly varies between individuals (proaggregatory in some and inhibitory in others). An individual variation in the effect of heparin on platelet function has been speculated on before. Wide individual variation was found in skin bleeding time following heparin administration (5). Our present findings using two different platelet function tests in a relatively large study population of cardiac surgical patients are consistent with wide individual variation in the effect of heparin on platelet reactivity. The proportion of individuals which exhibit a proaggregatory effect of heparin appears to be smaller (9%) as assessed by the haemostatometer than with whole blood aggregometry (41-46%), and also the degree of variation in the inhibition of platelet reaction appears to be greater. The statistically significant corselation between the results obtained by the two platelet function tests suggests that they both measure the same phenomenon. Whole blood aggregometry requires anticoaaylated and diluted blood. However, citrate by reducing plasma Ca and by a direct effect on the platelet membrane distorts the platelet response (8). Haemostatometry by not requiring an anticoagulant may be a more relevant test of platelet function and especially more adequate for studying the effect of heparin on platelets than platelet aggregometry. In conclusion, we found that in a relatively large population of cardiac surgical patients, a relevant concentration of heparin affected platelet function with a wide individual variation. A possible relevance of this variation in heparin response is in cardiac surgery where heparinisation is a necessity. Post operative blood loss is an important cause of morbidity (15) and has been partly attributed to an acquired inhibition of platelet function (16). A varying platelet inhibition to heparin may have important implications as to individual patients bleeding tendency in this clinical situation. These findings may also provide an explanation for contradictions in the earlier in vitro findings and for the diverse clinical complications of the clinical use of heparin. REFERENCES
1) Brace L.D., Fareed J. An objective assessment of the interaction of heparin and its fractions with human platelets. SEM.THROMB. HEMOST. 11,190-8,1985. 2) Chong B.H., Ismail F. The mechanism of heparin-induced aggregation. EUR.J.HAEMATOL. 43,245-51,1989.
platelet
3) Ruggiero M.,Fedi S.,Bianchini P.,Vannucchi S.,Chiarugi V. Molecular events involved in the proaggregating effect of heparin 802,372-377,1984. on human platelets. BIOCHIMICA.BIOPHYS.ACTA 4) Sobel M. ,Fujimura Y .,Ruggeri M.,Zimmerman T. Platelet glycoprotein Ib is a binding site that participates in platelet aggregation in heparin-associated thrombocytopenia. CLINICAL RESEARCH 34(2) (Abstract:449A),1986.
656
HEPARIN EFFECT ON PLATELETS
Vol. 66, No. 6
5) Heiden D.,Mielke C.H.,Rodvien R. Impairment by heparin of primary haemostasis and platelet (C14) 5-hydroxytryptamine release. BR.J.HAEMATOL. 36,427-36,1977. 6) Kelton J.G. Heparin 173-186,1986. 7) Warkentin ANN.REV.MED.
induced thrombocytopenia.
T.E .,Kelton J.G.Heparin 40,31-44,1989.
HAEMOSTASIS
16,
induced thrombocytopenia.
R.L., 8) Packham M.A.,Bryant N.L., Guccione M.A.,Kinlough-Rathbone Mustard J.F. Effect of the concentration of Ca2+ in the suspending medium on the responses of human and rabbit platelets to aggregating agents. THROMBOS.HAEMOSTAS. 62,968-76,1989. 9) Kovacs I.B., Hutton R.A.,Kernoff P.P.A. Haemostatic evaluation in bleeding disorders from native blood. Clinical experience with the haemostatometer. AM.J.CLIN.PATHOL. 91,271-9,1989. 10) Ratnatunga C.,Rees G.,Kovacs I. Preoperative assessment of haemostatic activity to predict excessive blood loss after cardiopulmonary bypass. ANN.THORAC.SURG. 52,250-7,1991. 11) Salzman E.W.,Deykin D.,Shapiro R.M.,Rosenberg R. Management of heparin therapy, controlled prospective trial. N.ENGL.J.MED. 292, 1047-50,1975. 12) Mant M.J., O'Brien B.D.,Thong K.L.,Hammond G.W.,Birtwhistle V.,Grace M.G. Haemorrhagic complications of heparin therapy. LANCET 1133-1135,1977.
R.
13) Ockelford P.A., Carter C.J.,Cerskus A.,Smith C.A.,Hirsh J. Comparison of the in vivo haemorrhagic and antithrombotic effects of a low Antithrombin III affinity heparin fraction. THROMB.RES. 27,679-690,1982. 14) Levine M.N.,Hirsh J. Haemorrhagic complications lant therapy. SEMIN.THROMB.HEMOSTAS. 12,39-57,1986.
of anticoagu-
15) Love T.R.,Hendren W.G. ,O'Keefe D.D.,Daggett W.M. Transfusion of predonated autologous blood in elective cardiac surgery. ANN. THORAC.SURG. 43,508,1987. 16) Harker L.,Malpass T.W.,Branson H.E. ,Hessel E.A.,Slichter S.A. Mechanism of abnormal bleeding in patients undergoing cardiopulmonary bypass: acquired transient platelet dysfunction associated with selective alpha granule release. BLOOD 56,824-34,198O.
