activation loss an extracorporeal - from bmj and acp · during extracorporeal circulation of blood,...

J. clin. Path (1963), 16, 558

Activation of coagulation and fibrinogen lossafter using an extracorporeal circulation

A. L. BLOOM

From the Institute ofPathology, Royal Infirmary, Cardiff

SYNOPSIS Alterations of the coagulation potential of heparinized blood after using an extracorporealcirculation have been studied by means of a toluidine blue-calcium chloride reagent. This techniquewas originally used to detect the effect of activation by contact on the coagulation mechanism inheparinized blood. It has been shown that it also detects, in the presence of heparin, the clot-potentiating effect of blood cell contents liberated in vitro by mechanical trauma to blood.

Variable destruction of platelets, red cells, and white cells occurred in heparinized sheep bloodrecirculated in a heart-lung machine in vitro. This was accompanied by increased clotting potential.Complete coagulation was prevented by heparin and fibrinogen levels remained unaltered.

Similar enhancement of the coagulation potential and destruction of blood cells were detectedin the blood of heparinized patients and sheep after perfusion for open-heart surgery. The coagu-

lation changes were usually transient, and impaired coagulation associated with significant fibrinogenloss was detected in most samples taken after the neutralization of heparin.

It is suggested that the coagulation changes are due to activation by contact of the coagulationmechanism during perfusion and to the clot-accelerating effect of blood cell contents. The resultssupport the hypothesis that coagulation defects and fibrinogen loss after using an extracorporealcirculation are due, at least in part, to intravascular coagulation. This is thought to occur, especiallyduring neutralization of heparin, while the coagulation mechanism is hyperactive.

Abnormal bleeding is one of the most importantcomplications of the use of heart-lung machines foropen-heart surgery. Although the use of heparinand its antagonists poses its own special problemsother important defects may occur in the haemo-static mechanisms. These include thrombocytopenia(Perkins, Osborn, and Gerbode, 1958; Sharp, Excell,Salzman, and Thorup, 1961 ; Bloom, 1961), deficiencyof factor V, factor VIII, and prothrombin (Nilssonand Swedberg, 1959; Fantl and Ward, 1960;Holemans, Amery, and Verstraete, 1960; Bloom,1961), and the appearance of a circulating anti-coagulant active in the thrombin-fibrinogen reaction(Von Kaulla and Swan, 1958; Hougie, Garrard,Warren, Dammann, and Muller, 1959). Severalworkers have reported fibrinogen loss after perfusionranging up to 50% or so of the pre-perfusion levels(Rothnie, Norman, Steele, and Kinmonth, 1960;Bloom, 1961), and cases of severe fibrinogen deple-tion with excessive bleeding have been reported (DuBourg, Trarieux, Fontan, Moulinier, Bricaud, andBroustet, 1960; Rothnie et al., 1960; Samama, 1961).Received for publication 26 March 1963.

Two main theories have been put forward toexplain the loss of fibrinogen and other coagulationfactors after perfusion. Douglas (1961) favours theview that activation of the fibrinolytic system leadsto proteolytic digestion of these factors, whilstOsborn, Mackenzie, Shaw, Perkins, Hurt, andGerbode (1956), Wright, Darte, and Mustard (1958),Hoeksema, Mustard, and Mustard (1959), and GansSiegal, Lillehei, and Krivit (1962), amongst others,favour the view that activation of coagulation by theextracorporeal circuit, perhaps accelerated by pro-ducts of destroyed blood cells, results in consumptionof coagulation factors in vivo and defibrination.There is little doubt that increased fibrinolysis

frequently occurs after perfusion, for it has beendetected by several workers (Von Kaulla and Swan,1958; Hougie et al., 1959; Sharp et al., 1961;Bloom, 1961; Gans, Lillehei, and Krivit, 1961). Infact the observation by Bloom (1961) that fibrinolysiswas consistently decreased on the day after operationsuggests that activation of this system with subse-quent depletion of its components occurs in all cases.The demonstration of activation of coagulation

558

copyright. on O

ctober 3, 2020 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.16.6.558 on 1 N

ovember 1963. D

ownloaded from

http://jcp.bmj.com/

Activation of coagulation and fibrinogen loss after using an extracorporeal circulation

after perfusion is more difficult. It is reasonable toassume such activation when fibrin deposits arepresent in the circuit. The tendency for these toform on glass surfaces or at points in the circuitwhere blood is subjected to trauma was noted byseveral earlier workers (Stokes and Gibbon, 1950;Dennis, Karlson, Eder, Nelson, Eddy, andSanderson, 1951; Clowes, Neville, Hopkins, Anzola,and Simeone, 1954). With improved apparatus andtechnique such deposits are now uncommon andhistological evidence of fibrin emboli is seldomobtained. Nevertheless coagulation defects still occurand have been reported even in the absence ofdetectable fibrinogen loss (Fantl and Ward, 1960).The demonstration of activation of coagulation

in these circumstances depends upon laboratorytests. The presence of heparin in samples to be testedmakes orthodox methods impracticable and mostof the support for activation of coagulation afterperfusion is derived from indirect evidence.The object ofthispaper is todescribetheapplication

of a toluidine blue-calcium chloride reagent (Bloom,1962) to the study of coagulation activity in hepari-nized blood. Evidence will be presented to showthe development of increased coagulation potentialin heparinized blood recirculated in a heart-lungmachine in vitro and after experimental and clinicalperfusion for open-heart surgery, and an attemptwill be made to relate the changes to the develop-ment of fibrinogen loss. The evidence supports thehypothesis that activation of the coagulationmechanism during perfusion plays a part in thedevelopment of coagulation defects.

MATERIALS AND STANDARD METHODS

COLLECTION OF SPECIMENS Blood was collected usingsiliconed syringes and containers. For coagulation studiesit was taken into 3-8% w/v trisodium citrate in the ratio ofnine volumes of blood to one of citrate. Platelet-poorplasma was obtained by centrifuging at 3,000 r.p.m. for30 minutes. Blood for cell counts and plasma haemo-globin estimations was taken into the dipotassium salt ofethylene diamine tetra-acetic acid (E.D.T.A.). For plasmafibrinogen estimation blood was taken into dried potas-sium oxalate (2 5 mg./ml.) and plasma obtained bycentrifuging at 3,000 r.p.m. for 10 minutes.

HEPARIN SOLUTION, INTACT PLASMA, CEPHALIN, AND GLASS-WARE These were prepared as described by Bloom(1962).

PLASMA HAEMOGLOBIN This was estimated by the methodof Fielding and Langley (1958).

CELL COUNTS White cell counts were performed as des-cribed by Whitby and Britton (1953). Platelet counts wereperformed by the method of Baar (1948).

PLASMA FIBRINOGEN This was estimated by the methoddescribed by Varley (1960) modified as follows: Plasma,0 1 ml., was added to 0 25 ml. ofnormal saline, and to themixture was added 01 ml. of a solution of polybrene,10 mg.% in normal saline, and 0 1 ml. of a 1 % solution ofepsilon amino-caproic acid (E.A.C.A.). The plasma wasclotted at 37°C. with calcium chloride and subsequentlythe fibrin was estimated as tyrosine as described by Varley.Polybrene was added to neutralize heparin, and E.A.C.A.to minimize possible fibrinolysis during the performanceof the test. Epsilon amino-caproic acid is an inhibitor offibrinolytic activation (Ablondi, Hagan, Philips, and DeRenzo, 1959). Preliminary studies were performed toensure that the addition of these substances to plasma didnot affect the fibrinogen estimation.

Fibrinogen estimations were performed on 0-1 ml.volumes of eight different normal plasma samples adding0 1 ml. of heparin, 4units/ml.,and 0.1 ml. ofpolybrene andE.A.C.A. as described above. Simultaneous estimationswere performed on aliquots of the same plasma substitut-ing equal volumes of saline for the heparin, polybrene,and epsilon amino-caproic acid. The results are shown inTable I. The addition ofheparin, polybrene, and E.A.C.A.to plasma does not affect the fibrinogen estimation.

TABLE IEFFECT OF SIMULTANEOUS ADDMON OF HEPARIN,

POLYBRENE, AND EPSILON AMINO-CAPROIC ACID TOPLASMA ON ESTIMATION OF FIBRINOGEN

Sample Fibrinogen Level Fibrinogen LevelNumber (mg. Y.) Recorded in (mg. °'0) Recorded in

Plasma + Saline Plasma + Heparin,Polybrene, and E.A.C.A.

2345678Mean

273280237469225175320329288

273273245480225180308343291

TOLUIDINE BLUE-CALCIUM CHLORIDE This reagent wasprepared as described by Bloom (1962). It has been foundthat the anticoagulant activity of toluidine blue may varyfrom batch to batch. The reagent was prepared so that 0 1ml. clotted a mixture of 0-2 ml. glass contacted normalplasma (Bloom, 1962) containing 0 05 ml. cephalin in 160to 200 seconds at 37'C. The concentration of toluidineblue needed varies from 80 mg. to 100 mg. per 100 ml. ofM/20 calcium chloride.

TESTS OF COAGULATION ACTIVITY Non-heparinized plas-ma was tested by the silicone calcium clotting timetechnique with and without cephalin. Plasma, 0.2 ml.,was placed in siliconed tubes and 0 05 ml. of cephalin ornormal saline added. The clotting time was determined at37°C. on the addition of 0Q1 ml. of M/20 calcium chloride.

Heparinized plasma was tested similarly except that

559

copyright. on O


http://jcp.bmj.com

/J C


ovember 1963. D

ownloaded from

http://jcp.bmj.com/

A. L. Bloom

the toluidine blue-calcium chloride reagent was usedinstead of M/20 calcium chloride. Experiments to deter-mine if this technique is suitable to detect alterations incoagulation activity of heparinized plasma are describedbelow.

DEMONSTRATION OF INCREASED COAGULATIONPOTENTIAL IN HEPARINIZED BLOOD

The importance of contact with glass and otherforms of silica and adsorbants in the activation ofthe coagulation mechanism has been demonstratedby the work of Margolis (1957), Ratnoff andRosenblum (1958), and Soulier and Prou-Wattelle(1960), and substances active in this respect, such asstainless steel (Rose and Broida, 1954), are used inextracorporeal circuits.

In a previous paper Bloom (1962) described theuse of the toluidine blue-calcium chloride reagent todemonstrate activation ofthe coagulation mechanismin heparinized plasma when in contact with glass.During extracorporeal circulation of blood, coagu-lation activity may be affected not only by contactwith foreign surfaces but also by products liberatedfrom destroyed blood cells. Substances active incoagulation are present in platelets (Van Creveldand Paulssen, 1952; Deutsch, Johnson, and Seegers,1955), red cells (Quick, Georgatsos, and Hussey,1954; Desmet, Verstraete, and Vandenbroucke,1957) and possibly in white cells (Gollub, 1953;O'Brien, 1959). Preliminary to studies with anextracorporeal circulation experiments were there-fore carried out to demonstrate the effects of cellulardamage on coagulation activity of normal bloodand to determine if these could be detected inheparinized blood by using the toluidine blue-calcium chloride reagent.

CHANGES IN NORMAL BLOOD DAMAGED MECHANICALLYCitrated blood, 2 ml., was delivered into each of twotubes. Fifteen siliconed glass beads, 2 mm. in diameter,were placed in one tube which was stoppered with a poly-thene-covered rubber bung and rotated on a Matburnmixer at 28 r.p.m. for about 15 minutes. The second tubewas left unrotated. The blood from both tubes wastransferred to fresh ones and samples removed for cellcounts. The remainder was centrifuged at 3,000 r.p.m. for30 minutes and the cell-poor plasma was removed. This

Sample Tested

was then tested by the silicone calcium clotting time tech-nique with and without cephalin. To control the absenceof contact activation by siliconed beads, intact normalplasma was rotated with such beads at the same time asthe blood and subsequently tested. At no time did unsili-coned glass come into contact with the blood or plasma.

The results are shown in Table II. Trauma toblood by rotation with siliconed glass beads leadsto release of haemoglobin and loss of platelets.The changes in the white cell counts were insignifi-cant. These cellular changes were accompanied byacceleration of the plasma silicone clotting times.This occurred in the absence of contact with glassand was not observed to a significant extent whenintact plasma was rotated with siliconed beads. It isconcluded that the accelerated clotting times aredue to substances liberated from blood cells. Theaddition of cephalin to the test system acceleratedclotting but the effects of cellular products could stillbe detected.

DETECTION OF CHANGES IN HEPARINIZED BLOOD DAMAGEDMECHANICALLY Heparinized blood was prepared byadding 0 1 ml. of a solution of heparin, 40 units/ml., to 4ml. of citrated venous blood. The final concentration ofheparin was thus approximately 1 unit/ml., i.e., about 2units/ml. in plasma. This concentration is completelyneutralized by the toluidine blue-calcium chloride reagent(Bloom, 1962). Non-heparinized blood was prepared bysubstituting normalsaline for the heparin.Samples of eachwere subjected to trauma with siliconed glass beads asdescribed above, and the cell-poor plasma was testedusing both M/20 calcium chloride and the toluidineblue-calcium chloride reagent.

The results are shown in Table III. The acceleratingeffects of trauma to whole blood on the plasma sili-cone clotting times is easily detected with the tolui-dine blue-calcium chloride reagent, although thepresence of the dye prolongs the clotting times. Whenplasma from heparinized blood is tested with thereagent, the results are very similar to those obtainedwith non-heparinized blood. The slight differencesobserved are probably due to variations which occurin the extent of cellular destruction even when thesame number of glass beads is used. It is concludedthat the technique satisfactorily detects the effects of

3LE IICOAGULATION AND CELLULAR CHANGES IN NORMAL BLOOD DAMAGED MECHANICALLY

Uncontacted Rotated Uncontacted Undamaged Uncontacted DamagedPlasma Blood Blood

Plasma CaCi, clotting time (sec.)Plasma CaCl, clotting time (sec.) (cephalin added)Plasma haemoglobin (mg. %)Platelet count (c.mm.)White c,ell count (c.mm.)

509306

6

520313

6231,000

7,700

28824474

147,0007,500

560

copyright. on O


http://jcp.bmj.com

/J C


ovember 1963. D

ownloaded from

http://jcp.bmj.com/

Activation of coagulation andfibrinogen loss after using an extracorporeal circulation

TABLE IIIDETECTION OF COAGULATION CHANGES IN DAMAGED HEPARINIZED BLOOD USING

TOLUIDINE BLUE-CALCIUM CHLORIDE REAGENTUndamaged Blood Damaged Blood

Non-heparinized Heparinized Non-heparinized Heparinized

Plasma CaCl, clotting time (sec.)Plasma toluidine blue-CaCl, clotting time (sec.)Plasma CaCl, clotting time (sec.) (cephalin added)Plasma toluidine blue-CaCl, clotting time (sec.) (cephalin added)

6251,000+336863

2,000+1,000+2,000+766

303672266467

2,000+536

2,000+421

destruction of blood cells on plasma coagulationpotential in the presence of heparin.

ACTIVATION OF THE COAGULATIONMECHANISM DURING EXTRACORPOREAL

CIRCULATION OF BLOOD

RECIRCULATION EXPERIMENTS The object of theseexperiments was to determine the effect of recircu-lation of heparinized blood in a Melrose heart-lungmachine on the cellular elements and on the plasmatoluidine blue-calcium chloride clotting times.

The priming blood was obtained from sheep on themorning of each experiment. In the first four experimentsit was obtained by exsanguination of one animal and col-lected into glass bottles. In the last four experiments 1

pint of blood was taken from each of several donoranimals into siliconed glass bottles. Heparin was uised asanticoagulant at concentrations of 1,000 to 3,000 units to500 ml. of blood. About 3 litres of blood were circulatedintermittently in the closed circuit for 30 to 40 minuteswith continuous oxygenation. This procedure was adopt-ed for the purpose of other experiments in progress.Samples of blood were obtained before and after recircu-lation by aspiration from the oxygenator through wide

o 700- WITHOUT CEPHALINC~~~~~~

w600-

E 5°° ' 37

4U§oos 400

2

300

3

0 ~~~~~~~~~~2200 7 4

0 ~~~~~~~~~6

0o -~

plastic catheters into siliconed syringes and were kept at4°C. until tested.The toluidine blue-calcium chloride clotting times were

determined in neat plasma and in plasma diluted I in 3 innon-heparinized intact plasma in order to exclude incom-plete neutralization of heparin by the reagent. The resultswith neat and diluted plasma always showed identicaltrends so that only those of the former are reported.

RESULTS The results of clotting time determinationsin individual experiments as well as mean values areshown in Fig. 1. A marked decrease in the clottingtimes without cephalin occurred in nearly all samplesafter recirculation. Only in experiment 6 was this lessobvious. The addition of cephalin to the test acceler-ated clotting in all specimens, but increased activityafter recirculation could still be detected. The collec-tion of the priming blood in siliconed (experiments 5

to 8) or unsiliconed (experiments 1 to 4) glass bottlesdid not apparently affect the results.Plasma fibrinogen and haemoglobin levels are

shown in Figure 2. Plasma fibrinogen levels showedno significant alteration. In some experimentsplasma haemoglobin levels were slightly raised atthe start. The reason for this was not apparent butit may have been due to difficulties in obtaining the

WITH CEPHALIN

FIG. 1. Plasma toluidineblue-calcium chloride clottingtimes before and after

5 recirculation ofheparinizedsheep blood in a heart-lung

7 machine. The code number ofeach experiment is shown

indicating the observed data.

Sample Tested

561

copyright. on O


http://jcp.bmj.com

/J C


ovember 1963. D

ownloaded from

http://jcp.bmj.com/

562

400- PLASMA FIBRINOGEN(mg. I 1001.)

300

200 -

100.

0

82 64

3

6 8B42 3

A. L. Bloom

PLASMA HAEMOGLOBINCmg. /100 ml.)

5

FIG. 2. Plasma fibrinogen andhaemoglobin levels before andafter recirculation ofheparinizedsheep blood in a heart-lungmachine. The code number ofeach experiment indicatesthe observed data.

PLATELET COUNTS / C. A.

2

a

4_7

6

1

2

4 63

-7 l.

5

BEFORE AFTER

RECIRCULATION

WHITE CELL COUNTS / c.0.

-12,000

10,000

-S.000

- 6,000

.4.000

2,000

5

S

6 2 3

71 4

3 6

2 74

FIG. 3. Platelet and white cellcounts before and afterrecirculation of heparinizedsheep blood in a heart-lungmachine. The code number ofeach experiment indicatesthe observed data.

REFCRC AFTER

RECIRCULATION

400.000

300.000

2004000

1004000

0 I -1 _--I

--I I 0 --I I a I

------ ML-&10%9%0 ACVCD

I

copyright. on O


http://jcp.bmj.com

/J C


ovember 1963. D

ownloaded from

http://jcp.bmj.com/


blood in some animals. Further increases in plasmahaemoglobin occurred in all experiments after re-circulation. These varied from 10 to 170 mg. %.White cell and platelet counts are shown in

Figure 3. Platelet counts were surprisingly littleaffected although clumping made accurate countingdifficult. In only three experiments (1, 5, and 8) didappreciable loss of platelets occur. White cellcounts were not greatly affected but a loss of over10% was observed in four experiments (1, 2, 5 and7). Acceleration of the clotting times was mostmarked in the two experiments (1 and 5) in whichthe greatest cellular destruction was detected.

ACTIVATION OF THE COAGULATION MECHANISM DURINGPERFUSION The object of these studies was toinvestigate the changes in the coagulation potentialof heparinized blood during clinical and experi-mental perfusion for open-heart surgery.Donor blood for priming the heart-lung machine was

collected for patients on the day before operation intoE.D.T.A. as anticoagulant and was stored overnight at4°C. Before operation each unit of 500 ml.was heparinizedwith 2,500 units of heparin and recalcified with 1-8 ml. of10% calcium chloride. Donor blood for experimentalperfusion of sheep was collected on the morning ofoperation. One pint was collected from each animal andwas taken into heparin (2,500 units/500 ml.) in siliconedglass bottles. The blood for these experiments was notthat used for the recirculation experiments.

Patients and animals received heparin at a dosage of250 units per kilogram body weight.The perfusions were all short, i.e., up to 40 minutes'

duration. After perfusion heparin was neutralized byintravenous infusion during 10 minutes of polybrene in adose ratio of 0-75 to 1 mg. for sheep and 1-25 to 1-5 mg.for patients to each 100 units of heparin administered.Neutralization was checked by the thrombin clotting timeand further graded doses of polybrene given if indicated.This was actually needed in only one instance.

Samples of blood were obtained from patients and

TABIMEAN PLATELET COUNTS, PLASMA HAEMOOPERATIONS USING AN EXTRACORPOREAL

Data Recorded Operative Subject Patients orAnimalsPre-perfusion

animals before perfusion (after heparinization), at the timeperfusion was just ending, and after the administration ofpolybrene. The mixed donor blood in the oxygenator wasalso sampled.Plasma was tested with the toluidine blue-calcium

chloride reagent with and without cephalin. In order toexclude incomplete neutralization of heparin by thereagent it was tested both neat and diluted one in three innon-heparinized intact plasma. As in the recirculationexperiments the clotting times with neat and dilutedplasma showed identical trends and only those of the neatsamples are reported.

RESULTS Samples were obtained during operationson five patients and 10 sheep. The results of theplasma toluidine blue-calcium chloride clottingtimes are shown in Figures 4 and 5. Figure 4 recordsthe clotting times without cephalin and Fig. 5 thosewith cephalin.The clotting times of the mixed donor blood

tended to be shorter than those of the pre-perfusionsamples of the subject. At the end of perfusion,however, clotting with and without cephalin wasalways accelerated, usually considerably so, andclotting times shorter than those of the mixeddonor blood were always obtained. This increasedclotting potential in the heparinized blood wasusually short lived. Fifteen to 20 minutes afterperfusion following the administration of polybreneonly five samples, all from sheep, still showedaccelerated clotting times. The human samplesand those from the other sheep showed a greater orlesser degree of impaired coagulation.

Increased clotting potential in the heparinizedblood at the end of perfusion was thus often re-placed by impaired clotting by the time the circu-lating heparin had been neutralized.The mean values for the platelet counts and plasma

haemoglobin and fibinogen levels are shown inTable IV.

LE IVGLOBIN AND FIBRINOGEN LEVELS DURINGCIRCULATION (5 HUMAN AND 10 SHEEP)

Oxygenator Patients or Patients orPre-perfusion Animals Animals

Post-perfusion Post-polybrene

Plasma haemoglobin (mg. %)

Platelet counts/c.mm. x 103

Plasma fibrinogen (mg. Y.)

HumanSheep'Whole series

HumanSheepWhole series

HumanSheepWhole series

'Estimation performed on seven sheep only

7

1025.97.7

259416364

241271261

13-21413-7

220222221

236243241

18 22723-3

110128122

213229224

18-8309258

109135126

186187187

563

copyright. on O


http://jcp.bmj.com

/J C


ovember 1963. D

ownloaded from

http://jcp.bmj.com/

A. L. Bloom

OPERATIONS No. 1-5 ON HUMAOPERATIONS No. 6-IS ON SHEE

11 8 9 14

15l

13 7 12

6 10

5

4

2

3

4 5 8

7

*6

Is

9

142310

13

12

4

7

5

2

Is

614 12

3

9

10

13

*. _ __ I I

,NS

15

12 7

4 5

2

14 10 13

9

a

- PATIENT OXYGENATOR PATIENT PATIENTI

Ia POST-PERFUSION

PRE- PERFUSION POST - POLYBRENE

FIG. 4. Plasma toluidine blue-calcium chloride clottingtimes (without cephalin) ofpatients and sheep during opera-tions using an extracorporeal circulation. The code numbersof the operations indicate the observed data. The arrowsindicate that the clotting times of the samples exceeded1,000 seconds.

1.100o

1,000 -

900'

c

700'

c

600'

CL

a500

V

x, 400'

._

E 300 -

° 200'

v

0

-100-

._O OPERATIONS No. 1-S ON HUMANSOPERATIONS No. 6-IS ON SHEEP

a9

5 1115

47

23

136

14 1012

4 5 8

7

1 I'S3149

2 10

61213

47

5

3

Is29 6 10

14 11213

3

7

Is

I

4

2

6

12

14

913

10

PAIN OXGNAO PAIN aTINPATIENT OXYGENATOR PATIENT PATIENTI I___,'J POST-PERFUSiONPRE-PERFUSION POST -POLYBRENE

FIG. 5. Plasma toluidine blue-calcium chloride clottingtimes (with cephalin) ofpatients andsheep during operationsusing an extracorporeal circulation. The code numbers ofthe operations indicate the observed data. The arrowsindicate that the clotting times of the samples exceeded1,000 seconds.

Some destruction of red cells occurred duringperfusion but plasma haemoglobin levels above50 mg. % were not observed. The mean plateletloss in the human cases during perfusion was149,000/c.mm. and in the animals 288,000/c.mm.Clumping of platelets was usually present in thepooled donor blood in the animal experiments andmay account for the relatively low counts observedin this blood.The mean loss of plasma fibrinogen in the human

cases was 55 mg. % and in the animals 84 mg. %.This occurred partly during perfusion and partlyduring the administration of polybrene, when themean loss in the patients was 27-4 mg. % (S.D. 5 4)and in the animals 41-4 mg. % (S.D. 11 '7). Thesechanges differ significantly from zero (P<OO1 forpatients and sheep).The changes occurring in the toluidine blue-

calcium chloride clotting times and in the plateletcounts, plasma haemoglobin and fibrinogen levelsare summarized graphically in Figure 6. The changesoccurring during human perfusions were similar to

those observed in the animal experiments and themean values for the combined series are shown in thefigure.

Increased coagulation potential occurring in theheparinized blood during perfusion is associatedwith the destruction of red cells and platelets andwith mild fibrinogen loss. Further fibrinogen lossoccurring during neutralization of heparin isassociated with prolongation of the clotting times.These findings suggest intravascular coagulationwith consumption of coagulation factors. It was notpossible, however, in this small series to correlatethe extent of the coagulation changes in individualsubjects with that of the changes in blood cells andplasma fibrinogen levels.

DISCUSSION

Support for the hypothesis that activation of thecoagulation mechanism occurs during perfusion inthe absence of fibrin deposits in the circuit is based,because of the difficulty in testing heparinizedblood, mainly on indirect evidence.

564

1.100-

1,000-

v 900-v

C£ 800-U

0 700-

i

._S

E600-

C

-500-

. 400-

0

u

.2

OO

w100-

I.-

copyright. on O


http://jcp.bmj.com

/J C


ovember 1963. D

ownloaded from

http://jcp.bmj.com/


FIG. 6. Graphical representationof the mean values for thetoluidine blue-calcium chlorideclotting times (with and withoutcephalin), platelet countsand plasma haemoglobin andfibrinogen levels offivepatients and 10 sheep duringoperations using anextracorporeal circulation.

Osborn et al. (1956) perfused dogs without usingheparin and detected shortened clotting timesfollowed by incoagulability of the blood withfibrinogen depletion. The authors attributed thechanges to activation of coagulation and exhaustionof coagulation factors. Wright et al. (1958) andHoeksema et al. (1959) drew attention to thesimilarity of the changes occurring in coagulationfactors during extracorporeal circulation to thoseoccurring during coagulation of blood in vitro. Ganset al. (1962) demonstrated accelerated clotting andthrombin generation in citrated blood recirculated ina heart-lung machine, and Penick, Averette, Peters,and Brinkhous (1958) have shown that transfusion ofsuch blood in dogs leads to thrombocytopenia andloss of factor VIII. They attributed these changes toactivation of the coagulation mechanism althoughfibrinogen levels were unaltered.The demonstration of such changes in heparinized

blood is more difficult. Perkins, Osborn, and Gerbode(1959) claimed that clotting times were acceleratedand prothrombin consumption increased in samplestaken immediately after perfusion and tested aftercareful neutralization of heparin with protamine.Later samples showed impaired coagulation. Theseresults would support the hypothesis that coagu-

lation is activated during perfusion but details oftheir experiments were not given.

In the present study a toluidine blue-calciumchloride technique has been used to detect changesin the coagulation potential of heparinized blood.This technique not only detects contact activation ofcoagulation in such blood (Bloom, 1962) but also theclot-accelerating effect of blood cell contents. Thiswas more obvious when cephalin was omitted fromthe test system, a finding in keeping with observationsthat substances present in platelets, red cells, andbrain cephalin have similar activity in thrombo-plastin generation (Bell and Alton, 1954; Quick et al.,1954; Troup and Reed, 1958). Some acceleration ofthe plasma clotting times ofdamaged blood occurred,however, when tested even in the presence ofcephalin. This suggests that blood cells also containa different type of activity.

Increased coagulation potential developed inheparinized sheep blood during recirculation in theheart-lung machine in vitro and was associated to agreater or lesser extent with destruction of bloodcells. In contrast to the findings of Nilsson andSwedberg (1959) and of Gans et al. (1962), workingwith dogs' blood, but in agreement with those ofPerkins, Osborn, and Gerbode (1957), the platelet

400 -

o 300 -

200-I

Z x 2100 -

01

ai0 I

IIJD L

p

u

300

250 -

200 -

lSO -

100 -

50S

0

1000 -

BooO

600 -

400 -

200

0

a

0t'o

'13

v

IV_, -

PRE-PERFUSION POST-PERFUSION POST-POLYBRENE

- I 9 9 I

565

copyright. on O


http://jcp.bmj.com

/J C


ovember 1963. D

ownloaded from

http://jcp.bmj.com/

loss after recirculation was generally not striking. Thediffering results may be due to many factors, such asthe type of machine used, the flow rate, and the con-centration of heparin, but may indicate, as suggestedby Perkins and his colleagues, that in some casesmuch of the platelet loss during clinical perfusionoccurs in vivo.

Activation of coagulation by contact is not pre-vented by heparin (Bloom, 1962) and may contributeto the accelerated clotting times. Didisheim, Hattori,and Lewis (1959) failed to detect an effect of contactwith glass on the clotting time of sheep blood. It iswrong to assume, however, that the contact phase isabsent in the blood of these animals. The additionof kaolin to intact sheep plasma results in con-siderable shortening of the silicone calcium clottingtime (personal unpublished observation). The factthat the blood was pooled in the oxygenator beforesampling and thus exposed to stainless steel mayaccount for the observation that the collection ofsheep blood in siliconed or non-siliconed bottlesdid not appreciably affect the results. In any casethe technique used will not distinguish betweenthe effects of contact and those of cellular, products,as the addition of brain cephalin to the testsystem does not entirely exclude the activity ofcellular elements. Whatever the mode of enhance-ment of the coagulation potential, complete coagu-lation during recirculation was prevented by heparin,and fibrinogen levels remained unaltered. Thisconfirms the observation made by Nilsson andSwedberg (1959) and Gans et al. (1962) that theextracorporeal circuit does not denature thisprotein, at least during the time limits of the experi-ments. Fibrinogen loss during and after perfusionthus appears to be a phenomenon in vivo.Enhanced coagulation potential was also observed

in heparinized blood after clinical and experi-mental perfusion for open-heart surgery. Althoughthe donor blood may have contributed to this effectit was not the only factor responsible. The clottingtimes of the post-perfusion samples were usuallyshorter than those of the mixed donor blood.Although the enhanced coagulation potential was

associated with cellular destruction, no relationshipcould be demonstrated in this small series betweenthe extent of the coagulation changes in individualsubjects and that of the platelet counts and plasmahaemoglobin levels. This does not necessarily meanthat cellular products do not contribute to thecoagulation changes. These are no doubt the re-sult of a complicated interplay of factors, includingpossibly the effects of the donor blood and contactactivation of coagulation as well as those of cellularcontents.

Recently similar enhancement of coagulation has

been described by Ollendorff, Storm, Rygg, andArnfred (1961) after extracorporeal circulation ofhuman blood both with and without a patient. Theywere able to detect this by demonstrating anaccelerating effect of the test plasma on thrombo-plastin generation in intact platelet-rich normalplasma. Heparin was neutralized with protamineand toluidine blue (Ollendorff, 1962). The authorsconsidered that activation of factors XII and XI bycontact was responsible for the changes detected, buttheir technique, like that used in the present study,did not exclude the effects of cellular contents, especi-ally as intact platelets were used in their test system.Margolis (1957) has shown that disintegration ofplatelets increases their activity in coagulation.The enhanced coagulation potential after per-

fusion in the present study was usually shortlivedand it had been replaced by impaired coagulationin most of the post-polybrene samples. This maymerely represent an anticoagulant action of excesspolybrene such as that described by Egerton andRobinson (1961) but other explanations are possible.The work of Spaet and his colleagues (Spaet andKropatkin, 1958; Spaet and Cintron, 1960; Spaet,Horowitz, and Zucker-Franklin, 1960) suggests theexistence of a clearing mechanism for plasmathromboplastin and its intermediates which maybe associated with the reticulo-endothelial system.Similar clearance of coagulant material may accountfor the rapid disappearance of activity noted in thepresent study.The impaired coagulation may, however, be the

result, at least in part, ofconsumption of coagulationfactors following neutralization of heparin whilethe coagulation mechanism is hyperactive. Appreci-able fibrinogen loss occurred during the neutraliza-tion of heparin with polybrene and at the same timeimpaired coagulation was detected in the post-polybrene samples. The available evidence is againstprecipitation by polybrene as a cause of fibrinogenloss. Unlike protamine, polybrene does not appar-ently precipitate fibrinogen in plasma (Godal, 1960;Shanberge, Regan, Talarico, and Busiek, 1961).Similar post-perfusion loss of fibrinogen afterneutralization of heparin by polybrene was noted byGans and Krivit (1962). Ollendorff et al. (1961)describe widespread thrombosis occurring at thisstage. These findings invite caution in the neutrali-zation of heparin, a point emphasized by Du Bourget al. (1960). Although it should eventually be com-plete, slow neutralization in stages may be preferableto the more rapid single dose technique. A recentreport of a nephrotoxic effect of polybrene (Haller,Ransdell, Stowens, and Rubel, 1962), if confirmed,may lead to the abandoning of this drug for use afterextracorporeal circulation. Similar considerations of

566 A. L. Bloom

copyright. on O


http://jcp.bmj.com

/J C


ovember 1963. D

ownloaded from

http://jcp.bmj.com/


intravascular coagulation apply equally, however,with the use of the alternative antiheparin agentprotamine.The demonstration of increased coagulation

activity after perfusion does not exclude the possi-bility that fibrinolysin may contribute to the develop-ment of haemostatic defects. Increased fibrinolysis iscommon after using an extracorporeal circulation.This enzyme system can digest not only fibrin, butalso fibrinogen and other coagulation factors(Soulier, Alagille, and Larrieu, 1956; Fletcher,Alkjaersig, and Sherry, 1959; Coon and Duff. 1958).A close relationship may exist between the coagu-lation and fibrinolytic systems. Surface contact andfactor XII activation may play a role in the develop-ment of endogenous plasminogen activator activity(Niewiarowski and Prou-Wartelle, 1959; Iatridis andFerguson, 1961) and intravascular activation ofcoagulation is a potent stimulus to fibrinolysis.Thus the experimental injection of thrombin resultsnot only in intravascular coagulation but also inmarked enhancement of fibrinolysis (Brayton andZucker, 1957; Hardaway, Watson, and Weiss, 1960).Increased fibrinolysis may thus be partly a secondaryand perhaps compensatory phenomenon but maywell contribute to a bleeding tendency.Whatever the role of fibrinolysis, however, the

sequence of events described in the present study,namely, increased coagulation potential after per-fusion, followed, during neutralization of heparin,by impaired coagulation and significant fibrinogenloss, bears a striking resemblance to the more severechanges observed by Osborn et al. (1956) whenheparin is not used. It also resembles the 'positivephase' of hyperactive coagulation and the 'negativephase' of incoagulable blood reported by earlierworkers (Martin, 1894; Mellanby, 1909) after theexperimental injection of coagulant materials.Although other factors, such as fibrinolysis, the useof stored donor blood, and the presence of residualheparin, may be associated with abnormal bleedingafter use ofan extracorporeal circulation, the findingsreported in this paper suggest that activation ofcoagulation during perfusion is not completelyprevented by heparin and together with clot-accelerating substances liberated from blood cellsmay contribute to the development of coagulationdefects.

I would like to thank Mr. H. R. S. Harley and Mr. D.Thomas for permission to study patients under their careat Sully Hospital, Penarth.

I owe a debt of gratitude to Dr. L. R. West, SullyHospital, for supplying me with samples of blood fromhis experimental animals and to the thoracic surgical andcardiology teams for their willing cooperation.

REFERENCES

Ablondi, F. B., Hagan, J. J., Philips, M., and De Renzo, E. C. (1959).Arch. Biochem., 82, 153.

Baar, H. S. (1948). J. clin. Path., 1, 175.Bell, W. N., and Alton, H. G. (1954). Nature (Lond.)., 174, 880.Bloom, A. L. (1961). Brit. med. J., 2, 16.

, (1962). J. clin. Path., 15, 508.Brayton, R. G., and Zucker, M. B. (1957). Proc.Soc.exp.Biol.(N.Y.)

96, 418.Clowes, G. H. A., Neville, W. E., Hopkins, A., Anzola, J., and

Simeone, F. A. (1954). Surgery, 36, 557.Coon, W. W., and Duff, I. F. (1958). J. Lab. clin. Med., 51, 381.Dennis, C., Karlson, K. E., Eder, W. P., Nelson, R. M., Eddy, F. D.,

and Sanderson, D. (1951). Surgery, 29, 697.Desmet, V., Verstraete, M., and Vandenbroucke, J. (1957). Thrombos.

Diathes. haemorrh. (Stuttg.), 1, 459.Deutsch, E., Johnson, S. A., and Seegers, W. H. (1955). Circulat. Res.,

3, 110.Didisheim, P., Hattori, K., and Lewis, J. H. (1959). J. Lab. clin. Med.,

53, 866.Douglas, A. S. (1961). In Thrombosis and Anticoagulant Therapy

(Proceedings of a Symposium, 1960) edited by W. Walker,Discussion of Sharp et al. p. 104. University of St. Andrews,

Du Bourg, M. G., Trarieux, M., Fontan, F., Moulinier, J., Bricaud, H.,and Broustet, P. (1960). Rev. franc. Stud. clin. biol., 5, 691.

Egerton, W. S., and Robinson, C. L. N., (1961). Lancet, 2, 635.Fantl, P., and Ward, H. A. (1960). Thorax, 15, 292.Fielding, H. E., and Langley, P. E. (1958). Amer. J. clin. Path., 30, 528.Fletcher, A. P., Alkjaersig, N., and Sherry, S. (1959). J. cin. Invest.,

38, 1096.Gans, H. and Krivit, W. (1962). Ann. Surg., 155, 353.

, Lillehei, C. W., and Krivit, W. (1961). Ibid, 154, 915.Siegal, D. L., Lillehei, C. W., and Krivit, W. (1962). Ibid., 156,19.

Godal, H. C. (1960). Scand. J. clin. Lab. Invest., 12, 446.Gollub, S. (1953). Fed. Proc., 12, 54.Haller, J. A. Jr., Ransdell, H. T. Jr., Stowens, D., and Rubel, W. F.

(1962). J. thorac. cardiovasc. Surg., 44, 486.Hardaway, R. M., Watson, H. E., and Weiss, F. H. (1960). Arch. Surg.,

81, 983.Hoeksema, T. D. D., Mustard, J. F., and Mustard, W. T. (1959).

Trans. Amer. Soc. Art. int. Orgs., 5, 225.Holemans, R., Amery, A., and Verstraete, M. (1960). Cardiologia

(Basel)., 37, 193.Hougie, C., Garrard, M. E., Warren, W. D., Dammann, F., and

Muller, W. H. (1959). Bibl. haemet. (Basel), 10, 28latridis, S. G., and Ferguson, J. H. (1961). Thrombos. Diathes.

haemorrh. (Stuttg.) 6, 411.Margolis, J. (1957). J. Physiol. (Lond.), 137, 95.Martin, C. J. (1894). Ibid., 15, 380.Mellanby, J. (1909). Ibid., 38, 441.Niewiarowski, S., and Prou-Wartelle, 0. (1959). Thrombos. Diathes.

haemorrh. (Stuttg.) 3, 593.Nilsson, I. M., and Swedberg, J. (1959). Acta chir. scand., 117, 47.O'Brien, J. R. (1959). J. clin. Path., 12, 45.Ollendorff, P. (1962). Scand. J. cin. Lab. Invest., 14, 303.

Storm, O., Rygg, I., and Arnfred, E. (1961). Acta chir. scand.,122, 217.

Osborn, J. J., Mackenzie, R., Shaw, A., Perkins, H., Hurt, R., andGerbode, F. (1956). Surg. Forum. 6, 96.

Penick, G. D., Averette, H. E. Jr., Peters, R. M., and Brinkhous, K. M.(1958). Thrombos. Diathes. haemorrh. (Stuttg.), 2, 218.

Perkins, H. A., Osborn, J. J., and Gerbode, F. (1957). Clin. Res. Proc.,5, 35.

(1958). In Extracorporeal Circulation. edited by J. G.Allen, p. 253. Thomas, Springfield, Illinois.

(1959). Ann. intern. Med., 51, 658.Quick, A. J., Georgatsos, J. G., and Hussey, C. V. (1954). Amer. J.

med. Sci., 228, 207.Ratnoff, 0. D., and Rosenblum, J. M. (1958). Amer. J. Med., 25, 160.Rose, J. C., and Broida, H. P. (1954). Proc. Soc. exp. Biol. (N. Y.), 86,

384.Rothnie, N. G., Norman, A. G., Steele, M., and Kinmonth, J. B.

(1960). Brit. J. Surg. 48, 272.Samama, M. (1961). In Thrombosis and Anticoagulant Therapy

(Proceedings of Symposium, 1960) edited by W. Walker,p. 103. Discussion of Sharp et al. University of St. Andrews,

Shanberge, J. N., Regan, E. E., Talarico, L., and Busiek, B. (1961).J. Lab. clin. Med., 58, 23.

567

copyright. on O


http://jcp.bmj.com

/J C


ovember 1963. D

ownloaded from

http://jcp.bmj.com/

568 A. L. Bloom

Sharp, A. A., Excell, B., Salzman, E., and Thorup, 0. (1961). InThrombosis and Anticoagulant Therapy (Proceedings of aSymposium, 1960) edited by W. Walker, p. 88. University ofSt. Andrews,

Soulier, J. P., Alagille, D., and Larrieu, M. J. (1956). Proc. int. Soc.Haemat., 4, 488.

-, and Prou-Wartelle, 0. (1960). Brit. J. Haemat., 6, 88.Spaet, T. H., and Cintron, J. (1960). Proc. Soc. exp. Biol. (N. Y.), 104,

498., Horowitz, H. I., and Zucker-Franklin, D. (1960). J. clin. Invest.,

39, 1029.

, and Kropatkin, M. (1958). Amer. J. Physiol., 195, 77.Stokes, T. L., and Gibbon, J. H. (1950). Surg. Gynec. Obstet., 91, 138.Troup, S. B., and Reed, C. F. (1958). J. clin. Invest., 37, 937.Van Creveld, S., and Paulssen, M. M. P. (1952). Lancet, 1, 23.Varley, H. (1960). Practical Clinical Biochemistry, 2nd ed., p. 193.

Heinemann, London.Von Kaulla, K. N., and Swan, H. (1958). J. thorac. Surg., 36, 519.Whitby, L. E. H., and Britton, C. J. C. (1953). Disorders of the Blood.,

7th ed., p. 700. Churchill, London.Wright, T. A., Darte, J. M., and Mustard, W. T. (1958). Lancet, 1,

1157.

copyright. on O


http://jcp.bmj.com

/J C


ovember 1963. D

ownloaded from

http://jcp.bmj.com/

activation loss an extracorporeal - from bmj and acp · during extracorporeal circulation of blood,...

Documents