Diagnosis and Management of Disseminated Intravascular Coagulation

Symposium on Laboratory Diagnosis

Diagnosis and Management of Disseminated Intravascular Coagulation

Margaret Karpatkin, MD.*

The diagnosis of disseminated intravascular coagulation depends upon the clinical evaluation of the patient together with the interpretation of appropriate laboratory tests. Although this paper is concerned primarily with diagnosis and management, a broader discussion of the disorder is necessary to explain the rationale behind laboratory tests and treatment. Disseminated intravascular coagulation is an acquired disorder of hemostasis which may appear during the course of certain diseases or pathologic events. It is not in itself a primary disease process but a response to certain stimuli which accompany underlying disease. The result is widespread coagulation within the vasculature, leading to formation of intravascular thrombi and utilization of coagulation factors and platelets. Hemorrhage occurs when rapid utilization causes platelets and clotting factors to be depleted to levels inadequate for hemostasis. The disorder has also been called consumption coagulopathy 33 or defibrination syndrome. In its severe form it is extremely dangerous, so that prompt recognition and treatment is vital. Plasma of the normal person contains nine coagulation proteins, most of which are present in an inactive form. Certain stimuli, such as injury to a vessel, will trigger the interaction of these factors, resulting in the formation of a fibrin clot at the site of injury. Clot formation is preceded by a platelet plug. Platelets also provide a phospholipid surface which enhances interaction of factors in the coagulation process. This interaction takes place in a sequential fashion. Some factors are converted to active enzymes; others are cofactors. This sequential process is called the coagulation cascade 25 or waterfall,1O and is depicted in Figure 1. As shown in Figure 1, factor XII is converted to an active form, XIIa, by an appropriate stimulus. In vitro, the stimulus is contact with glass, and in vivo it is probably contact with collagen in the wall of the injured vesse1.39 Factor XIIa and factor XI then form a complex which activates *Assistant Professor of PediatriCS, New York University School of Medicine

Pediatric Clinics of North America- Vol. 18, No. I, February, 1971

23

24

MARGARET KARPATKIN

XII----XIIa

1

XI

IX

Tissue factor viI

----+)

IXa

1

VIII, platelet phospholipid

X----Xa

I II

v, platelet phospholipid IIa

I

----Fibrin

Figure 1. The coagulation cascade. Activated factors are designated "a". XII, Hageman factor; XI, plasma thromboplastin antecedent (P.T.A.); IX, plasma thromboplastin component (P.T.C.); VIII, antihemophilic factor (AHF); X, Stuart Prower factor; VII, serum prothrombin conversion accelerator (SPCA); V, proaccelerin; II, prothrombin; I, fibrinogen.

factor IX.15 The activated factor IX complexes with factor VIII and platelet phospholipid to activate factor X. Activated factor X complexes with factor V and phospholipid to convert factor II (prothrombin) to factor IIa (thrombin). Thrombin splits two pairs of peptide chains off the fibrinogen molecule, and the resulting protein (fibrin monomer) then polymerizes to form the insoluble fibrin clot. An alternative pathway of fibrin formation is also shown in Figure 1; here a protein-phospholipid complex which exists in all mammalian tissues complexes with factor VII to activate factor X.28 This pathway is referred to in the older literature as the "extrinsic" coagulation mechanism. There is little evidence that coagulation occurs within the intact blood vessels of the healthy individual. In patients with disseminated intravascular coagulation, the coagulation cascade becomes activated within the intact blood vessels. Evidence for this is provided by the clinical observation of thromboses and by the finding of intravascular thrombi at autopsy. Laboratory data suggest that coagulation factors and platelets become reduced because they are being used up more rapidly than they can be synthesized; hence, the term "consumption coagulopathy." Thus in many cases of intravascular coagulation, one observes thrombosis and hemorrhage occurring at the same time. The combined picture of thrombosis and hemorrhage is not seen in all patients. Frequently, when abnormal bleeding develops and there is definite laboratory evidence of intravascular coagulation, thromboses are not observed clinically, and in some patients who come to autopsy, intravascular fibrin cannot be demonstrated. 26 In such cases, the fibrin formed is thought to be removed rapidly by the fibrinolytic enzyme system. A brief description of this system follows. In the fibrinolytic system, depicted in Figure 2, the inactive protein plasminogen, which is present in normal plasma, can become activated to plasmin. Plasmin is a proteolytic enzyme which is capable of breaking down many proteins, including coagulation factors, but it has a particular avidity for fibrin. Normal plasma does not contain free plasmin, because little or no plasminogen activation takes place. The small amounts of

DISSEMINATED INTRAVASCULAR COAGULATION

25

Activator

1

Plasminogen - - - - - » Plasmin - - - - - » Antiplasmins (inactivation)

1

Fibrin --------» Fibrin degradation products (FDP) Figure 2.

The fibrinolytic system.

plasmin which may be formed become inactive by complexing with the antiplasmins. It is believed that during intravascular coagulation a plasminogen activator is released into the vessels from their walls at the site of fibrin deposition. Release is possibly the result of local anoxia. This activator is adsorbed to the fibrin and activates plasminogen in the vicinity. The resulting plasmin lyses the fibrin, leading to the appearance of fibrin fragments in the blood. The activation of plasminogen also causes the circulating level of this protein to be reduced. This process is called secondary fibrinolysis and might be regarded as protective to the patient, since it lessens the degree of intravascular thrombosis.

CONDITIONS ASSOCIATED WITH DISSEMINATED INTRA VASCULAR COAGULATION Intravascular coagulation has occurred in children in association with the conditions listed in Table 1. It is not an invariable accompaniment of these disorders but has been described on a number of occasions. There are other conditions which are occasionally complicated by intravascular coagulation. It is not known how coagulation is stimulated in all these disorders, but it seems likely that different mechanisms are involved. In some diseases the direct cause is probably release into the circulation of a substance which activates the coagulation system. In malignant disease, tissue from the tumor or its metastases may activate factor X through the extrinsic pathway. Support for this possibility is provided by the observation that injection of tissue extracts into animals will induce disseminated intravascular coagulation.3 ! In gram-negative Table 1.

Conditions Which May Be Complicated by Intravascular Coagulation

Bacterial infections, particularly gram-negative sepsis Viral infections Idiopathic respiratory distress syndrome Leukemia Malignant solid tumors Giant hemangioma Burns Hemolytic uremic syndrome (some cases) Surgery, particularly when extracorporeal circulation is used Trauma

26

MARGARET KARPATKIN

sepsis it is likely that endotoxin is responsible for the coagulopathy, because appropriately spaced injections of gram-negative endotoxin in rabbits brings about the generalized Shwartzmann reaction. The animal dies in shock. Hemorrhage and widespread intravascular deposition of fibrin is found at autopsy. There is evidence that endotoxin activates factor XII34 and also alters platelets. Intravascular coagulation associated with giant hemangiomata, and following extracorporeal circulation, is probably due to mechanical trauma to the blood. Detailed discussion of the possible mechanisms by which disseminated intravascular coagulation is caused is beyond the scope of this article.

CLINICAL PICTURE AND DIAGNOSIS The tremendously variable clinical picture of intravascular coagulation usually depends upon the severity of the syndrome. If the process is acute, the child may be extremely ill, with widespread hemorrhage into the skin, mucous membranes, and viscera. At the other extreme, the process may be so indolent that the patient may have an occasional spontaneous ecchymosis with or without thrombosis. Studies in adults using isotopically labelled fibrinogen have shown that low grade intravascular coagulation can occur without clinical symptoms or signs. 4 The ease of diagnosis is also related mainly to the severity of the disorder. In severe intravascular coagulation, the diagnosis is easy, and can often be made on the basis of the clinical picture and a few simple tests; when the coagulopathy is low grade, on the other hand, diagnosis is far more difficult, and may require procedures not available in many clinical laboratories. In Table 2 are listed the initial procedures which should be undertaken when the diagnosis of intravascular coagulation is suspected. All are simple, can easily be completed within 45 minutes, and can be performed with as little as 2 ml. of blood. A detailed discussion of the rationale for their performance follows. On the basis of the results obtained, intravascular coagulation can be diagnosed when the process is acute. It should be stressed at this point that in the acutely affected patient, speed of diagnosis is of great importance so that therapy can be undertaken to prevent irreversible changes such as renal cortical necrosis due to fibrin deposition. Table 2. Initial Laboratory Tests in the Diagnosis

of Intravascular Coagulation History and physical examination Platelet count Examination of stained blood smear Prothrombin time (P.T.) Partial thromboplastin time (P.T.T.) Thrombin clotting time (T.T.) Fibrinogen estimation

DISSEMINATED INTRAVASCULAR COAGULATION

27

History and Examination As with all laboratory investigations, those for intravascular coagulation are most useful when interpreted in conjunction with the history and clinical findings. It is important to know if the patient has any previous history of spontaneous bleeding, and whether he is presently suffering from a disease known to be associated with intravascular coagulation. Information can also be gained as to the clinical severity of the coagulation defect. Platelet Count The platelet count is usually depressed. The level will depend upon the rate at which platelets are being destroyed and the speed with which the bone marrow is able to replace them. The platelet count may vary from extremely low levels in acute intravascular coagulation up to normal levels. In some instances, when the coagulation process is slow and the marrow response good, platelet counts may be in the normal range. In such cases examination of the bone marrow would show an increase in megakaryocytes. Examination of a Stained Blood Smear Examination of a peripheral blood smear should be routine, as a check of the platelet count. Many of the platelets seen will be large, an indication of rapid production by the bone marrow.H In addition, the appearance of the red cells is frequently abnormal during intravascular coagulation. They may be fragmented and distorted, the appearances described in microangiopathic hemolytic anemia. There is experimental evidence that these changes are due to the red cells being fragmented by passing through meshes of intravascular fibrin.:15 Prothrombin Time This measurement is prolonged if factors II, V, VII, or X are decreased. Gross reductions of fibrinogen will also affect it. In intravascular coagulation, factor V is usually reduced, although it may be normal and in some instances abnormally elevated. Factors II (prothrombin) and X may also be reduced. Partial Thromboplastin Time (P. T. T.) Factor VIII is usually reduced in intravascular coagulation, leading to prolongation of the P.T.T. In addition, reduction of factors V, II, and X (referred to above) will also lengthen the P.T.T. Factors XI and IX are sometimes reduced and will also lengthen the P.T.T. In some cases, in which consumption of coagulation factors is unquestionably occurring, very high levels of factor VIII are recorded and may cause an abnormally short P.T.T. This will be discussed in more detail later. Thrombin Clotting Time (T.T.) The thrombin clotting time is usually prolonged. The test consists of adding thrombin to the patient's plasma and measuring the time taken for a fibrin clot to appear. Thus it measures only the very last part of the

28

MARGARET KARP ATKIN

coagulation reaction-conversion of fibrinogen to fibrin. It is prolonged if fibrinogen is grossly depleted (usually below 80 mg. 100 mI.). The more usual cause for its prolongation in intravascular coagulation is the presence of fibrin degradation products. The mechanism for this will be discussed later. Fibrinogen Estimation As the name defibrination implies, fibrinogen is frequently low in disseminated intravascular coagulation and is always low when the coagulopathy is severe. However, the presence of a normal fibrinogen level does not rule out the diagnosis of intravascular coagulation. Firstly, the patient may have had a very high fibrinogen level before consumption began, since the level is increased in a number of diseases. It may take several days for fibrinogen to fall below the normal range. Secondly, consumption may be so slow that synthesis keeps up with removal. This has been shown by fibrinogen turnover studies using radioactive isotopes. 4 The method by which fibrinogen is measured is extremely important. Ideally it should be measured as clottable protein. While a number of reliable methods are available, all take 2 to 3 hours to perform. If the result is required very quickly, a more rapid method must be used. Merskey, Johnson, Kleiner, and Wohp6 found the technique described by Ellis and Stransky 14 to correlate well with clottable protein measurement. This method, which depends upon the optical density change when calcium and thrombin are added to diluted plasma, takes about 25 minutes to perform. We have found it satisfactory unless the patient is jaundiced. Since bilirubin absorbs at the suggested wavelength, the test becomes grossly inaccurate. This is a disadvantage in pediatrics because the plasma bilirubin level is elevated in many neonates. We have found that if readings are performed at a wavelength of 550 mJL instead of 470 mJL as described in the original method, reasonable results can be obtained, even from severely jaundiced neonates. Methods which depend upon ammonium sulfate precipitation, and . immunological methods such as the Fi-test, are not suitable because fibrin and fibrinogen degradation products are indistinguishable from fibrinogen, and lead to falsely high values. If these tests are completely normal, it is unlikely that intravascular coagulation is occurring, unless it is so low grade that production of clotting factors is keeping pace with consumption, and fibrinogen degradation products are being rapidly removed by the reticuloendothelial system. If results are slightly abnormal but do not permit a clear-cut diagnosis, the following should be done: assay of factors II (prothrombin), V, and VIII, and measurement of fibrin degradation products. Even when a clear-cut diagnosis has been made and treatment started, it is an advantage to carry out these tests for confirmation of the diagnosis. The interpretation of these tests is discussed below. Factor VIII Assay The level of factor VIII is frequently reduced in disseminated intravascular coagulation. Merskey and co-workers 26 found it to be reduced in

DISSEMINATED INTRAVASCULAR COAGULATION

29

all patients with acute intravascular coagulation and in many patients with the subacute syndrome. However, a normal level of this factor does not rule out the diagnosis, and in some instances assay has revealed very high levels, up to 300 or 400 per cent of normal. It is likely that these patients do not really have high levels of factor VIII but that other factors which have been activated, particularly factor X and thrombin, are present in their plasma. In most laboratories the factor VIII assay is based upon the partial thromboplastin time. Activated factors below VIII in the cascade will cause the VIII to appear falsely high. 26 • 29 Factor V Assay Merskey and co-workers 26 found factor V to be reduced in acute defibrination and in many cases of subacute defibrination. It is rarely normal or slightly elevated. Factor II (Prothrombin) Assay Factor II is frequently reduced in intravascular coagulation. Measurement of Fibrin Degradation Products (F.D.P.) As described earlier, intravascular fibrin is broken down by the proteolytic enzyme, plasmin. Degradation products can be detected in blood samples by a variety of methods. Before discussing methodology, a brief description of the different degradation products and their properties will be given. When plasmin breaks down fibrinogen, a high molecular weight fragment first results. This has been designated fragment X. Continued exposure to plasmin results in further breakdown to fragment Y and then to two smaller fragments, D and E, with molecular weights of 80,000 and 30,000 respectively. Similar changes occur when fibrin is exposed to plasmin. All these fragments cause prolongation of the thrombin clotting time, either by interfering with release of fibrinopeptides by thrombin or by inhibiting the polymerization of fibrin monomer.22 The smaller molecular weight fragments are not clotted by thrombin, and are therefore present in serum, where their presence may be demonstrated. The presence of increased degradation products in serum indicates that fibrinolysis is occurring. Since this most commonly occurs secondary to disseminated intravascular coagulation, the presence of degradation products is usually evidence of intravascular coagulation. Very rarely (and the word rarely should be emphasized), acute primary fibrinolysis may occur resulting in formation of FDP. (This will be discussed later under differential diagnosis.) In this laboratory we measure FDP in serum by the tanned red blood cell hemagglutination inhibition technique, as described by Merskey, Lalezari, and JohnsonP The method is extremely sensitive, and as little as 1 microgram of fibrinogen or FDP per ml. of serum may be detected. Normal serum may contain up to 8 micrograms per ml.; any level greater than this suggests that fibrin or fibrinogen is being proteolyzed. In severe intravascular coagulation, we have recorded values of greater than 800 micrograms per ml. Mention should be made at this point of some other investigations

30

MARGARET KARP ATKIN

which we do not routinely carry out in our laboratory but which may be useful occasionally when diagnosis is difficult. Euglobulin Lysis Time The euglobulin lysis time is an indication of the amount of fibrinolytic activator present in the blood. If the level of fibrinolytic activator is increased, the lysis time becomes shorter than normal, and if free plasmin is present it becomes very short. In disseminated intravascular coagulation, although local release of activator occurs, the activator is adsorbed to intravascular thrombi and there is no general increase in the blood level. Thus euglobulin lysis time is usually normal. It is always significantly shortened when primary pathological fibrinolysis occurs, and the test should be done whenever this rare diagnosis is suspected. However, a short lysis time is not diagnostic of primary pathological fibrinolysis, and is only an indication that further tests should be performed in order to confirm or rule out the diagnosis. Primary pathological fibrinolysis is discussed later in this article. Plasminogen Estimation Plasminogen is reduced during intravascular coagulation. Its estimation may occasionally help in the diagnosis when other tests are equivocal. Unfortunately, under these circumstances the plasminogen level is frequently equivocal also. It can be measured by a clot lysis method, using bovine fibrinogen, or by a caseinolytic method. Trial of Heparin Therapy Specific treatment for disseminated intravascular coagulation consists of the administration of heparin in order to halt the coagulation process. If the dosage is adequate, the platelet count should rise slowly, as should any coagulation factors which were reduced. If disseminated intravascular coagulation is suspected but cannot definitely be demonstrated, heparin therapy may be tried. If the levels of platelets and coagulation factors rise, it strongly suggests that the diagnosis is correct.

INTERPRETATION OF RESULTS

It has already been stressed that diagnosis depends upon a careful evaluation of each patient in conjunction with results of laboratory investigations. For example, a previously healthy 15 month old girl was admitted with a diagnosis of meningococcemia. There were widespread petechiae and ecchymoses. The platelet count was 47,000/cu. mm.; prothrombin time was 42 seconds (control 14), partial thromboplastin time was 170 seconds (control 42), thrombin clotting time 52.5 seconds (control 13), and fibrinogen was 60 mg. per 100 ml. (normal range 200 to 450 mg. per 100 mI.). Heparin therapy was started as soon as these results were available. Continuation of the work-up showed factor V 25 per cent, factor VIII 10 per cent, and FDP to be greater than 800 micrograms per mI., thus confirming the diagnosis of disseminated intravascular coagulation.

DISSEMINATED INTRAVASCULAR COAGULATION

31

Thus in a patient who has a disease known to predispose to disseminated intravascular coagulation, no past history of coagulation abnormality, and no evidence of liver disease, the following are sufficient to justify the immediate institution of therapy: low platelet count, prolonged prothrombin time, partial thromboplastin time, and thrombin time, and reduced fibrinogen. These values can easily be available within 45 minutes of drawing blood, thus allowing treatment to be started rapidly. Unfortunately diagnosis is not always so simple. Sometimes there is only slight abnormality of the screening tests, and measurement of plasminogen, coagulation factors, and serum FDP are necessary. These techniques are time-consuming and are not available in many clinical laboratories. In some instances, it is impossible to confirm the diagnosis, even when these methods are available. Replacement of clotting factors may be as rapid as consumption leading to normal levels. In addition, cells of the reticuloendothelial system may remove FDP as rapidly as they are formed, so that none are detected in serum. In such patients the coagulopathy can be demonstrated only by fibrinogen survival studies showing decreased half-life of this protein. A shortened platelet survival time with a normal platelet count is also suggestive of the diagnosis.

DIAGNOSIS OF DISSEMINATED INTRAVASCULAR COAGULATION IN THE NEONATE

Intravascular coagulation is being increasingly recognized in newborn infants, 11.12.19.24 particularly in association with the idiopathic respiratory distress syndrome. The coagulation system in the normal neonate differs considerably from that of the adult and older child, and this must be remembered when interpreting laboratory values. In Table 3 are shown values in normal neonates compared with older children. The figures in parentheses indicate values obtained in our laboratory. It has been known for more than 30 years that the prothrombin time

Table 3. Coagulation Values in Newborn Infants':' Platelet count Prothrombin time Partial thromboplastin time Thrombin time Fibrinogen Factors II, VU, IX, and X Factor V Factor VIII Plasminogen Fibrinogen degradation products Euglobulin lysis time

Same as in adults Prolonged (range 16-31 seconds, adults 14 seconds) Prolonged (range 44-73 seconds, adults 42 seconds) Prolonged (range 19-44 seconds, adults 16.3 seconds) Reduced (range 117-225 mg. per 100 ml., adult range 200-350 mg. per 100 mI.) Reduced Same as in adults for term infants Slightly reduced in prematures Same as adults Reduced Same as in adults (not greater than 6.76 micrograms per m!.) Short

':'Adult samples were taken from healthy children over the age of 2 years or adults.

32

MARGARET KARPATKIN

of the newborn infant is prolonged due to depression of vitamin K dependent factors (II, VII, IX, and X).7' 30 When vitamin K is administered, hemostatic levels of these factors are achieved, but they are still low enough to prolong the prothrombin time and partial thromboplastin time. The neonatal fibrinogen level is also slightly but significantly lower. The thrombin clotting time of cord blood is markedly prolonged when compared with that of adults and older children. 1, 23, 32. 36 While the cause for this difference has not been determined, it may be due to an altered reactivity of neonatal fibrinogen with thrombin. It appears to persist for the first few months of extrauterine life. Levels of factors V and VIII are in the adult range in the normal neonate and therefore their measurement gives a useful indication of whether or not intravascular coagulation is occurring. The neonatal platelet count is also in the adult range.2 , 16 Reports thatthe platelet count in the healthy premature may be significantly lower than in the term baby have recently been shown to be inaccurate, and thrombocytopenia in the premature infant should always be regarded as an indication of disease and investigated. There have been reports in the literature of raised levels of FDP in normal cord blood,6. 37 but other workers, including ourselves,s, 13, 18, 21 have failed tQ confirm this. It appears that the positive findings reported may be due to faulty collection and preparation of serum samples. Euglobulin lysis time of cord blood is usually very rapid.

DIFFERENTIAL DIAGNOSIS OF DISSEMINATED INTRAVASCULAR COAGULATION, COAGULATION DEFECT DUE TO LIVER DISEASE, AND PRIMARY PATHOLOGICAL FIBRINOLYSIS

Hepatocellular damage, whether due to infection, drugs, or chemicals, can cause a marked coagulation defect. Most clotting factors and plasminogen are synthesized exclusively in the liver, and levels of all except factor VIII may decrease in liver disease. In addition, intravascular coagulation or low-grade fibrinolysis 26 is sometimes associated with liver disease, and differential diagnosis is very difficult. Hepatocellular damage is usually very severe before a coagulation defect is evident, and does not usually accompany hepatitis unless transaminase levels are extremely high. The difficulty in diagnosis arises when a patient with severe liver disease has a coagulation defect and it has to be determined whether this is due to lack of synthesis or intravascular coagulation, or to both. A normal platelet count is suggestive, but is not absolute evidence against intravascular coagulation, just as a low platelet count is not proof of intravascular coagulation. An increased level of fibrin degradation products is good evidence that intravascular coagulation is at least partly responsible for the defect. A low level of factor VIII is against the diagnosis of uncomplicated hepatocellular damage. Laboratory findings in the two diseases are summarized in Table 4.

33

DISSEMINATED INTRAVASCULAR COAGULATION

Table 4. Laboratory Differential Diagnosis Between Disseminated Intravascular Coagulation (D.I.C.), Liver Disease Not Accompanied by D.I.C., and Primary Fibrinolysis

D.LC.

Platelet count

Reduced

Prothrombin time Partial thromboplastin time Thrombin time Fibrinogen Factor VIII

Prolonged Prolonged Prolonged Reduced Reduced or elevated Reduced Increased Decreased Normal

Factor V F.D.P. Plasminogen Euglobulin lysis time

LIVER

PRIMARY

DISEASE

FIBRINOL YSIS

Normal or reduced'" Prolonged Prolonged Prolonged Reduced Normal or elevated Reduced Normal Decreased Short or normal

Normal Prolonged Prolonged Prolonged Reduced Reduced Reduced Increased Decreased Markedly shortened

"'Reduced when hypersplenism is associated with liver disease.

Primary pathological fibrinolysis is the term used to describe a condition in which there is sudden massive activation of plasminogen throughout the body, resulting in free circulating plasmin. This plasmin lyses fibrinogen and other clotting factors, resulting in decreased coagulation and hemorrhage. Fibrinogen degradation products which are formed are identical immunologically to fibrin degradation products and are therefore indistinguishable by most methods of measurement. Until 10 years ago, this condition was diagnosed quite frequently, mainly in association with complications of labor. It now appears that the majority of those cases were in fact disseminated intravascular coagulation, and that primary pathological fibrinolysis is an extremely rare syndrome. Merskey and co-workers26 in 1967 had seen only five cases in the preceding 10 years, and none in the preceding 4 years. Laboratory findings in this condition are summarized in Table 4.

TREATMENT OF DISSEMINATED INTRAVASCULAR COAGULATION Treatment of the Underlying Disease This is the most important aspect of treatment of intravascular coagulation. If the condition which underlies it is halted, intravascular coagulation will cease. Thus, appropriate administration of antibiotics, correction of pH, electrolyte imbalance, and shock, when necessary, are of the utmost importance.

Heparin Administration Heparin is a potent anticoagulant. It acts as an antithrombin and also inhibits the coagulation cascade at several stages. With adequate

34 dosage, it will arrest intravascular coagulation and so may be used to prevent consumption of coagulation factors and platelets until the underlying disease has been eradicated and the consumptive process stopped. It is best given intravenously, when its anticoagulant action is immediate. It can be given by deep subcutaneous injection but is not as effective, and there is a tendency to form painful ecchymoses at the injection sites. Dosage is usually based upon response as measured by the Lee White clotting time and varies greatly in different patients. Other methods of controlling heparin therapy, based upon the partial thromboplastin time, are available. 5 , 40 The usual method of administration is to give 1 mg. per kg. of body weight, followed by continuous intravenous infusion of 1 mg. per kg. every 4 hours. The dose is then adjusted to bring the clotting time to the desired level. Opinion varies as to what this level is. Some workers maintain it between two and three times the average normal value, while others are given enough heparin to maintain the clotting time at more than 30 minutes. We feel that it is most important that heparinization should be adequate to completely arrest the coagulopathy, and usually give the larger dose. Excretion is usually complete in 4 to 6 hours, unless there is renal impairment, in which case it may be considerably delayed. The decision to administer heparin is relatively easy when the patient has no site from which catastrophic hemorrhage can occur. If the patient has recently undergone surgery or trauma and has a large wound as a potential site for bleeding, the decision is much more difficult. By the very nature of the properties which halt intravascular coagulation, heparin will prevent normal hemostasis at the site of injury and may induce worse hemorrhage than that already caused by the coagulopathy. Under these circumstances, the decision to administer heparin has to be made with careful consideration of the particular clinical situation. If heparin is given and bleeding becomes more severe, the anticoagulant effect is very rapidly reversed by the intravenous administration of protamine sulfate in a dose equivalent in milligrams to the milligrams of heparin which have been given. When there is suspicion that intracranial bleeding has occurred, as is often the case with neonates in respiratory distress, the decision to give heparin is even more difficult. The effectiveness of heparin therapy can be judged most simply by serial platelet counts. If the consumptive process has halted, the platelet count should remain stable and then rise. Continuing consumption of platelets indicates that too little heparin is being given or that there is an accompanying vasculitis which is causing platelet utilization independent of the coagulation process. The coagulation factors should rise during heparin therapy, but they are difficult to measure because heparin interferes with the tests involved. Even the in vitro addition of such antiheparins as toluidine blue or protamine sulfate does not overcome this problem, because these agents have other effects upon the clotting process and will themselves distort many of the tests. If heparin is temporarily discontinued in a patient with normal renal function and a specimen is drawn 4 hours after the last dose, the heparin level is usually insufficient to interfere with the necessary studies.

DISSEMINATED INTRAVASCULAR COAGULATION

35

Heparin therapy should be continued until the consumptive coagulopathy has stopped. The problem is how to judge when this has occurred. It may be assumed that consumption will continue until the underlying disease has been controlled or eliminated. Once we think this has occurred, heparin is discontinued, and 4 hours later the following tests are performed: platelet count, prothrombin time, partial thromboplastin time, thrombin time, fibrinogen level, factor VIII level and factor V level. If these are in the normal range, the tests are repeated in 6 hours. If there has been no significant change during this time, it can be assumed that the coagulopathy has probably ceased. We usually follow platelet count, prothrombin time, partial thromboplastin time, and thrombin time daily for a few days to be sure that coagulopathy has not recurred. Heparin is not usually administered unless laboratory tests which have been performed are at least suggestive of disseminated intravascular coagulation, but in some instances the drug should be given before laboratory diagnosis has been made. In fulminating meningococcemia, vascular occlusion can occur very rapidly, leading to renal cortical necrosis. In a severely ill child with meningococcemia associated with spontaneous bleeding, blood should be drawn for coagulation studies and heparin given immediately without waiting for the results. Transfusion with Blood Components It is sometimes necessary to replace the platelets and coagulation factors which have been depleted during consumptive coagulopathy. Ideally this should not be done until the patient is fully heparinized, as the transfused factors may feed the consumptive process. The blood component most often necessary is platelet concentrate. If the platelet count is below 40,000 per cu. mm., we usually administer platelets whether or not there is evidence of hemorrhage. This is done particularly in the newborn where the danger of intracerebral hemorrhage is great. If there is obvious bleeding, platelets should be given even when platelet counts are higher than this. We have seen hemorrhage which was controlled after platelet infusion even when pre-infusion levels were 70,000 to 80,000 per cu. mm. The reason for hemorrhage at platelet levels which are normally adequate for hemostasis is not clear, though it is possible that platelet function is impaired during intravascular coagulation. There is evidence that this impairment is brought about by fibrin degradation products. 2o Coagulation factors usually return toward normal levels during heparinization, and it is rarely necessary to replace them unless hepatic function is poor. If replacement is necessary and there is depression of several factors, fresh frozen plasma should be used. If depression of one factor such as fibrinogen or factor VIII predominates, the appropriate concentrate should be given. Steroids The administration of adrenocorticosteroid hormones is sometimes indicated during intravascular coagulation but only as part of the treat-

36

MARGARET KARP ATKIN

ment of the underlying disease (for example, acute lymphoblastic leukemia). Physiciims have been hesitant to use steroids because of a report that they enhance the generalized Shwartzman reaction. 3s Corrigan 9 and co-workers have been unable to reproduce this finding. Some authors believe that when steroids are indicated they should be administered in full doses, particularly when the patient has been fully heparinized. 3 Heparin will block the generalized Shwartzman reaction.

Epsilon-amino-caproic Acid (EACA, Amicar) This compound, when given in therapeutic doses, inhibits the activation of plasminogen. In very large doses it blocks the proteolytic activity of plasmin. It is given in cases of primary pathological fibrinolysis. Its use is contraindicated in disseminated intravascular coagulation and it may be extremely harmful. It blocks the secondary fibrinolysis which occurs, and thus intravascular thrombi will persist, leading to local tissue necrosis.

SUMMARY Disseminated intravascular coagulation (defibrination syndrome, consumption coagulopathy) is a syndrome which may complicate a number of pathologic states. If clotting factors and platelets are "consumed" more rapidly than the patient can produce them, they may fall to levels which are not adequate for hemostasis. Thus, thrombosis with resulting necrosis and hemorrhage may occur simultaneously in the same patient. Local fibrinolysis (secondary fibrinolysis) may remove or partially remove intravascular thrombi, thus protecting against local tissue necrosis. Diagnosis of the fulminating syndrome is usually simple and depends upon a few relatively easy tests. If the syndrome is low grade, diagnosis is very difficult and may be impossible without sophisticated techniques. Differential diagnosis from coagulation defects secondary to liver disease is extremely difficult. Treatment in the fulminating case before intravascular thrombi can cause irreversible tissue damage is urgent. Treatment depends upon therapy of the underlying disease plus adequate heparinization. It may be necessary to administer platelets and blood fractions after heparinization. Corticosteroids are not contraindicated, particularly when the patient is heparinized. Epsilon-amino-caproic acid (EACA, Amicar) is contraindicated. This drug inhibits local fibrinolysis, which is protective to the patient.

REFERENCES 1. Aballi, A. ]., Lopez-Barnus, V., Delamerens, S., and Rozengvaig, S.: The coagulation de-

fect of full-term infants. Pediatria Internazionale, 9 :315, 1959. 2. Aballi, A. J., Puapondh, Y., and Desposito, F.: Platelet counts in thriving premature infants. Pediatrics, 42:685, 1968.

DISSEMINATED INTRAVASCULAR COAGULATION

37

3. Abildgaard, C. F.: Recognition and treatment of intravascular coagulation. J. Pediat., 74:163,1969. 4. Baker, L. R. I., Rubenberg, M. L., Dacie, J. V., and Brain, M. C.: Fibrinogen catabolism in microangiopathic haemolytic anaemia. Brit. J. HaematoI., 14:617,1968. 5. Blakely, J. A.: A rapid bedside method for the control of heparin therapy. Canad. Med. Assoc. J., 99:1072,1968. 6. Boniface, E., Baggio, P., and Gravina, E.: Demonstration of split products of fibrinogen in the blood of normal newborns. Biologie Neonatorum, 12:29, 1968. 7. Brinkhous, K M., Smith, H. P., and Warner, E. D.: Plasma prothrombin level in normal infancy and in hemorrhagic disease of the newborn. Amer. J. Med. Sci., 193:475,1937. 8. Chessells, J. M., and Pitney, W. R.: Letter to the editor. Pediatrics, 45: 155, 1970. 9. Corrigan, J. J., Abildgaard, C. F., Seeler, R. A., and Schulman, I.: Quantitative aspects of blood coagulation in the generalized Shwartzman reaction. II. Effect of cortisone. Pediat. Res., 1 :99, 1967. 10. Davie, E. W., and Ratnoff, O. D.: Water-fall sequence for intrinsic blood clotting. Science, 145:1310,1964. 11. Du, J. N. H., Briggs, J. N., and Young, G.: Disseminated intravascular coagulopathy in hyaline membrane disease. Pediatrics, 45 :287, 1970. 12. Edson, J. R., Blaese, R. M., White, J. G., and Kriuit, W.: Defibrination syndrome in an infant born after abruptio placentae. J. Pediat. 72:342,1968. 13. Ekelund, H., Hedner, U., and Nilsson, I. M.: Letter to the editor. Pediatrics, 45: 156, 1970. 14. Ellis, B. C., and Stransky, A.: A quick and accurate method for the determination of fibrinogen in plasma. J. Lab. Clin. Med., 58 :477, 1961. 15. Esnouf, M. P.: Biochemical aspects of blood coagulation. Proceedings of Twelfth Congress. Internat. Soc. Hemat. New York 1968. Plenary Session Papers, p. 315. 16. Fogel, B. J., Arias, D., and Kung, F.: Platelet counts in healthy premature infants. J. Pediat., 73:108, 1968. 17. Garg, S., Amorosi, E., and Karpatkin, S.: Use of the megathrombocyte as an index of thrombopoiesis. New Eng. J. Med. (in press). 18. Hathaway, W. E.: Letter to the editor. Pediatrics, 45:154, 1970. 19. Hathaway, W. E., Mull, M. M., and Pechet, G. S.: Disseminated intravascular coagulation in the newborn. Pediatrics, 43:233, 1969. 20. Jerushalmy, Z., and Zucker, M. B.: Some effects of fibrinogen degradation products (F. D.P.) on blood platelets. Thromb. et. diath. haemorrh.,15:413, 1966. 21. Karpatkin, M. H.: Letter to the editor. Pediatrics, 45:157,1970. 22. Kowalski, E.: Fibrinogen derivatives and their biologic activities. Seminars in Hematology, 5 :45, 1968. 23. Larrieu, M. J., Soulier, J. P., and Minkowski, A. Le sang du cordon ombilical. Etude complete de sa coagulabilite, comparison avec Ie sang maternale. Et Neonat., 1 :1,1952. 24. Leissring, J. C., and Vorlicky, L. N.: Disseminated intravascular coagulation in a neonate. Amer. J. Dis. Child., 115: 100, 1968. 25. Macfarlane, R. G.: An enzyme cascade in the blood clotting mechanism and its function as a biochemical amplifier. Nature,202:498, 1964. 26. Merskey, C., Johnson, A. J, Kleiner, G. J., and Wohl, H.: The defibrination syndrome: Clinical features and laboratory diagnosis. Brit. J. HaematoI., 13:528,1967. 27. Merskey, C., Lalezari, P., and Johnson, A. J.: A rapid, simple, sensitive method for measuring fibrinolytic split products in human serum. Proc. Soc. Exper. BioI. Med., 131 :871, 1969. 28. Nemerson, Y.: The phospholipid requirement of tissue factor in blood coagulation. J. Clin. Invest., 47:72,1968. 29. Niemetz, J., and Nossel, H. L.: Activated coagulation factors: in vivo and in vitro studies. Brit. J. HaematoI.,16:337, 1969. 30. Nygaard, K K: Prophylactic and curative effect of vitamin K in hemorrhagic disease of the newborn. Acta. Obstet. Gynec. Scandinav., 19 :361, 1939. 31. Penick, G. D., Roberts, H. R., Webster, W. P., and Brinkhous, K M.: Hemorrhagic states secondary to intravascular clotting. Arch. Path., 66:708,1958. 32. Roberts, J. T., Gray, O. P., and Bloom, A. L.: An abnormality of the thrombin fibrinogen reaction in the newborn. Acta Paed. Scandinav., 55:148, 1966. 33. Rodriguez-Erdmann, F.: Bleeding due to increased intravascular blood coagulation; hemorrhagic syndromes caused by consumption of blood-clotting factors (consumptioncoagulopathies). New Eng. J. Med., 273:1370,1965. 34. Rodriguez-Erdmann, F.: Studies on the pathogenesis of the generalized Shwartzman reaction. III. Trigger mechanism for the activation of the prothrombin molecule. Thromb. Diath. Haemorrh., 12 :471, 1964. 35. Rubenberg, M. L., Regaeczi, E., Bull, B. S., Dacie, J. V., and Brain, M. C.: Microangiopathic hemolytic anemia: The experimental production of hemolysis and red-cell fragmentation by defibrination in vivo. Brit. J. HaematoI., 14:627,1968. 36. Salazar de Sousa, C.: Alterations de la coagulation sanguine chez Ie nouveau-ne. Pediatre, 9:784,1954.

4

38

MARGARET KARP ATKIN

37. Stiehm, E. R., and Clatanoff, D. v.: Split products of fibrin in newborns. Pediatrics, 43:770, 1969. 38. Thomas, L., and Good, R. A.: The effect of cortisone on the Shwartzman reaction. The production of lesions resembling the dermal and generalized Shwartzman reactions by a single injection of bacterial toxin in cortisone treated rabbits. J. Exper. Med., 95 :409, 1952. 39. Wilner, G. D., Nossel, H. L., and LeRoy, E. C.: Activation of Hageman factor by collagen. J. Clin. Invest., 47:2608, 1968. 40. Zucker, S., Cathey, M. H., and Wylie, R. L.: Control of heparin therapy. Sensitivity of the activated partial thromboplastin time for monitoring the antithrombic effects of heparin. J. Lab. Clin. Med., 73:320,1969. 550 First Avenue New York, New York 10016