Coumarin anticoagulants

Liver cells produce certain coagulation factors that require vitamin K for synthesis into a form that can be activated. Drugs of the coumarin family inhibit vitamin K utilization by liver cells. The vitamin K–dependent factors, listed in order of decreasing sensitivity to coumarin, are factor VII, factor IX, factor X, and prothrombin. Factor VII has the shortest half-life (4-6 hours)and is decreased first and most severely, with activity levels less than 20% of baseline by 24 hours after coumarin intake. The other factor activitiesdecrease somewhat more slowly and less profoundly, reaching their lowest level in 4-5 days. Therefore, some anticoagulation effect is seen in 1.5-2.0 days, with maximum effect in 4-5 days. The early effect can be achieved a few hours earlier, and the degree of effect made greater, by use of a larger (“loading”) initial dose. However, use of a loading dose in some patients may result in complications due to excess anticoagulation. As noted while discussing the PT test earlier in this chapter, vitamin K metabolism may be affected by intake (low intake potentiates the effect of coumarins and unusually high intake antagonizes that effect), absorption (availability of bile salts and necessity for an intact duodenal mucosal cell absorption mechanism), and utilization by hepatic cells (number and health of liver cells).

There are several members of the coumarin family derived from the original drug dicumarol; the most commonly used in the United States is warfarin sodium (Coumadin). Warfarin is water soluble, and oral doses are nearly completely absorbed by the small intestine. Maximal plasma concentration is reached 1-9 hours after ingestion. It is 95%-97% bound to albumin after absorption. Because of the very high percentage bound to albumin, substantial decrease in the albumin level increases free warfarin and increases the PT. The warfarin dose may have to be decreased to compensate for this. Warfarin is metabolized by the hepatic cell microsome system. The half-life of warfarin varies greatly among individuals (mean value 35-45 hours; range, 15-60 hours). With an initial dosage of 10 mg once each day, it takes 5-10 days (five half-lives) to reach equilibrium. Coumarin drugs usually take about 36-48 hours (usual range, 24-72 hours) to develop a significant effect on the PT, and between 3 and 5 days to develop maximal anticoagulant effect. Duration of action is 4-5 days. Warfarin effect can be terminated by parenteral vitamin K. The dose needed has some relationship to the degree of PT elevation. Normalization of the PT ordinarily occurs in 6-24 hours.

Monitoring coumarin anticoagulants. Anticoagulation by coumarin-type drugs is best monitored by the PT, although the PT is not affected by factor IX, one of the four vitamin K–dependent coagulation factors. Until recently, the most commonly accepted PT therapeutic range was 2.0-2.5 times a control value derived either from the midpoint of the reference range or from normal fresh pooled plasma. A recent National Conference on Antithrombotic Therapy recommended that PT therapeutic ranges of 1.3-1.5 times control value be used rather than 2.0-2.5 times control if the PT reagent is derived from rabbit brain or is a rabbit brain-lung combination (these arethe most common PT reagents in the United States). With the usual reference range of 11-13 seconds, this would be roughly equivalent to an anticoagulant rangeof 15-20 seconds. An anticoagulant range of 1.5-2.0 times control was recommended for prosthetic valve prophylaxis against thrombosis. When one considers the large number of patients anticoagulated with coumarin drugs, it is surprising how little information is available from human studies regarding clinical evaluation of different therapeutic levels. Reporting PT results in terms of seconds compared with control is preferred to results in terms of percentage of normal, because percentage must be based on dilution curves, which are frequently inaccurate. Various prestigious organizations and researchers have recommended that PT results be reported as the international normalized ratio (INR). This is thepatient PT divided by the midpoint of the laboratory PT reference range; the result is then adjusted by being multiplied by an exponent derived by comparison of the thromboplastin reagent being used with an international standard. A calculator is needed for this mathematical task. This standardizes all thromboplastin reagents and theoretically permits a patient to have a PT performed with a comparable result in any laboratory using the INR system. However, there are problems regarding use of t he INR. When the PT is performed on an automated instrument, different instruments alter the response of the thromboplastin to some degree, potentially negating some of the uniformity that the INR was designed to created. Most importantly, the INR was designed for—and should only be used for—one condition: long-term warfarin therapy after the patient has been stabilized. Stabilization usually takes at least 1 week, sometimes longer. The INR therefore may produce inaccurate and potentially misleading results when patients are beginning warfarin therapy, ending warfarin therapy, being given heparin in addition to warfarin, or when there is interference by various technical conditions that affect the patient sample, the equipment used, or the patient’s coagulation factors (such as effect of medications or severe liver disease). The INR is not applicable when the PT is being used as a coagulation test rather than a monitor for stabilized warfarin therapy. These other applications are best interpreted by comparing test results to the individual laboratory’s PT reference range (in seconds). In addition, it should be remembered that most laboratories, even if they supply INR values, do not know if individual patients are or are not stabilized on warfarin and in many (or most) cases do not know for what reason the PT is being ordered.

As of 1994, the usual recommendation for warfarin therapy PT levels on high-risk patients with mechanical heart valves is a target INR range of 2.5-3.5; for all other indications, the target range is 2.0-3.0.

Warfarin presents increased risk in first trimester pregnancy due to increased congenital malformations and undesirable effects on early fetal bone development and in the last trimester due to increased possibility of fetal or maternal hemorrhage at or near birth. Various medications affecting warfarin are listed.

Heparin/antithrombin III system

Antithrombin III. Antithrombin III (AT-III) is a serine protease that requires heparin as an activator to produce effective anticoagulation. It is now believed that AT-III provides most of the anticoagulant effect, whereas heparin serves primarily as the catalyst for AT-III activation. It is thought that AT-III is synthesized by the liver. Antithrombin III concentrations are decreased in the first 6 months of life and in hereditary AT-III deficiency. Concentrations may be secondarily decreased in severe liver disease, extensive malignancy, the nephrotic syndrome, deep venous thrombosis, malnutrition, septicemia, after major surgical operations, and in DIC. Antithrombin III is also decreased 5%-30% by pregnancy or use of estrogen-containing oral contraceptives. Although there is not a good linear correlation between AT-III levels and demonstrated risk of venous thrombosis, in general a considerably decreased AT-III level tends to predispose toward venous thrombosis. This is most evident in hereditary AT-III deficiency, which is uncommon, is transmitted as an autosomal dominant, and is associated with a very high risk of venous thrombosis. A decreased AT-III level also decreases the apparent effect of standard doses of heparin as measured by coagulation tests and thus produces the appearance of “heparin resistance.” In general, patients respond to heparin when AT-III activity levels are more than 60% of normal and frequently do not respond adequately when AT-III activity levels are less than 40% of normal. Warfarin increases AT-III levels somewhat.

Antithrombin III assay. AT-III can be assayed by several techniques. In general, these can be divided into two groups: those that estimate AT-III activity (“functional” assay) using coagulation factor or synthetic AT-III substrates; and those that measure the quantity of AT-III molecules present (“immunologic” assay) by means of antibody against AT-III. The results of the two assays do notalways correlate in clinical situations, since AT-III normally has little inhibitory effect until it is activated (as with heparin) and since the conditions of AT-III activation and AT-III binding to affected coagulation factors may vary. AT-III assay may not be reliable when thrombosis is present since AT-III may be temporarily depleted. It also may not be reliable when the patient is receiving heparin or warfarin (warfarin tends to increase AT-III levels).

Heparin. Heparin is an anticoagulant that currently is a biologic substance extractedfrom either porcine gastric mucosa or bovine lung tissue. As a biological extract, heparin contains molecules with molecular weight ranging from 5,000 to 30,000. As noted previously, heparin contains a carbohydrate group that binds to antithrombin III and considerably increases its anticoagulation effect. Although antithrombin III is the actual inhibitor, heparin was discovered before antithrombin III and the actions of this heparin-antithrombin III complex are conventionally worded as though the actions are due to heparin. When standard heparin (extract) activates antithrombin III, the major effect is that of an antithrombin, although it also inhibits most of the other coagulation factors to some degree, with greatest effect on activated factors IX and X.

Heparin is not consumed during its coagulation activity. It is degraded primarily in reticuloendothelial (RE) cells and blood vessel endothelial cells. About 80% of plasma heparin is metabolized by the RE cell system and 20% is excreted by the kidneys into the urine. Heparin’s half-life averages 90 minutes (range, 30-360 minutes). The half-life is increased with increase indose and shortened in deep vein thrombosis or pulmonary embolization (possibly due to increased platelet factor 4). Alpha-2 acid glycoprotein (an acute reaction protein) and low density lipoprotein (LDL) can bind heparin and decrease sensitivity to heparin’s action. Protamine and polybrene are specific neutralizing agents for heparin (1 mg of protamine for each 100 units of heparin). Platelets contain a heparineutralizing factor, PF-4. Severe thrombocytopenia may increase sensitivity to heparin.

Heparin is mainly distributed in plasma. Therefore, high hematocrit levels (red cell mass) would tend to increase heparin concentration (in the diminished plasma volume). The anticoagulant effect of heparin is influenced by the antithrombin III level and to a lesser and variable degree by large increases or decreases in platelets (due to PF-4), presence of fibrin (which can bind to thrombin and prevent antithrombin III binding), and by dosage and route of administration. Patient response to heparin is variable, with differences sometimes as much as 300% between individuals within a group of patients. Obese patientsoften have decreased rate of heparin clearance, which may produce elevated plasma heparin levels. Heparin’s major side effect is bleeding; this is influenced by dose, route of administration, presence of renal failure, presence of severe illness, heavy and chronic alcohol intake, use of aspirin, and severe thrombocytopenia. Heparin can itself induce thrombocytopenia in about 10%-15% of patients (range, 1%-31%). Thrombocytopenia generally occurs 5-7 days (range, 2-15 days) after onset of therapy, is usually (but not always) not severe, and is usually reversible a few days after end of therapy. About 2% of patients develop sustained platelet decrease to less than 100,000/mm3 (almost always in patients receiving IV heparin). In about 0.5% of patients there is occurrence of the “white clot” syndrome. In these patients the platelet count falls below 100,000/mm3 6-14 days after heparin is begun, with onset of arterial thrombosis having a partial white color due to masses of platelets. This is due to antibody against platelet membranes. In this type of thrombosis, heparin must be stopped instead of increased.

More recently, the heparin extract has been fractionated with recovery of molecules having anticoagulant effect and a molecular weight of approximately 5,000. These “low molecular weight” heparins induce antithrombin III action predominantly on activated factor X rather than thrombin. Low molecular weight heparin (LMWH) appears to have about 50% greater effectiveness than standard heparin for prophylaxis against clotting, has a more predictable response to standard doses, and can be given once a day with only initial laboratory monitoring in most cases rather than continued monitoring. In one LMWH study, 10%-15% of patients were hyperresponders (relative to most patients) and 10%-15% were hyporesponders.

Monitoring of heparin therapy. Anticoagulation by heparin is monitored by the APTT in the majority of U.S. hospitals, although the activated clotting time or other methods can be used satisfactorily. The Lee-White whole blood clotting time was originally used but proved too cumbersome, insensitive, and nonreproducible. The heparin therapeutic range is usually stated as 1.5-2.5 times the APTT upper reference limit, but it is better to perform a pretherapy APTT test and use that value as the reference point for determining the therapeutic range. Heparin is poorly absorbed from the GI tract, so it is usually administered either by continuous IV infusion or by intermittent IV or subcutaneous injection. With constant IV infusion, heparin anticoagulant response is theoretically uniform, and the APTT test theoretically could be performed at any time. However, several studies have reported a heparin circadian rhythm, even with constant IV infusion; the peak effect occurring between 7 P.M. and 7 A.M. with maximum about 4 A.M., and the trough effect occurring between 7 A.M. and 7 P.M. with the maximum effect about 8 A.M. However, other studies have not found a circadian rhythm. When intermittent injection is used, peak heparin response is obtained 30 minutes to 1 hour after IV injection or 3-4 hours after subcutaneous injection. In the average person, using a single small or moderate-sized heparin dose, anticoagulant effect decreases below the threshold of measurement by 3-4 hours after IV injection and by 6 hours after subcutaneous injection. Most believe that the optimum time to monitor intermittent IV therapy is one-half hour before the next dose; and for subcutaneous injection, halfway between doses (i.e., at 6 hours after injection using 12-hour doses). At this time the APTT should be at least 1.5 times the upper normal limit or pretherapy value. There is controversy as to whether the peak of the response should also be measured. If low-dose therapy is used, many clinicians do not monitor blood levels at all. However, in such cases it may be desirable to perform an AT-III assay since there would be no way to know if an unsuspected low AT-III level were present that would prevent any heparin effect.

So-called low dose heparin has become popular for prophylaxis against thrombosis. This is usually 5,000 units (although the literature range is 3,000-8,000 units) given subcutaneously. In one study, 5,000 units given subcutaneously elevated APTT values above preheparin levels in 80%-85% of patients; 10%-15% of all patients had increase over twice baseline. The frequency and degree of these elevations would depend to some extent on the sensitivity to heparin of the particular thromboplastin used.

Different commercial APTT reagents differ in their sensitivity and linearity of response to heparin effect. Some APTT reagents have shown rather poor sensitivity, and most are not very linear in response once the upper limit of anticoagulant range is exceeded. Variation in reagent sensitivity is a problem because APTT therapeutic limits obtained with one manufacturer’s reagent might not be applicable to another manufacturer’s reagent and could lead to overdosage or underdosage of heparin. Results of various comparative studies have not unanimously shown any single manufacturer’s product to have clear superiority. Results of comparative studies have also been difficult to interpret because of changes in different manufacturer’s reagents over the years. Various other plasma or whole blood clotting tests are claimed to provide better results in monitoring heparin therapy than the APTT, of which the most frequently used are the thrombin time (TT) or the activated whole blood clotting time (ACT). When large heparin doses are used, as in bypass surgery, the APTT and TT are very nonlinear, prolonged, and poorly sensitive to heparin level changes, while the ACT has better performance characteristics. However, all these tests are influenced to varying degree by many factors that are involved with clotting, not the specific action of heparin. This is shown by the elevation of the APTT by warfarin.

WHOLE BLOOD RECALCIFICATION TIME. Whole blood is drawn into a tube with citrate anticoagulant. Calcium chloride is added to an aliquot of the whole blood to begin the clotting process. The reference range upper limit for clotting is about 2 minutes at 37°C or 5 minutes at room temperature. Test results are said to be adequately reproducible. The whole blood recalcification time (WBRT) is used primarily to monitor heparin therapy. Its advantage is that whole blood clotting may be a more truthful representation of the varied effects of heparin than tests that circumvent part of the normal clotting process. Only one very cheap reagent is needed. The WBRT is sensitive to platelet variations and is more linear than the APTT at greater heparin effects. The major disadvantage is that the test cannot be automated. One report states that the test must be performed within 1 hour after venipuncture. Equipment is commercially available that enables the test to be performed at the patient’s bedside.

Thrombin time. One aliquot of diluted thrombin is added to 1 aliquot of patient plasma. Clotting time upper limit is approximately 20 seconds. The test is simple and said to be very sensitive to heparin effect. It is not affected by sodium warfarin (Coumadin). As originally performed, the method was nonlinear and, like the APTT, was greatly prolonged in higher degrees of heparin effect. Certain modifications may have partially corrected this problem. Reagent source and composition may affect results considerably.

There are, however, some specific assays of heparin activity. One of these is the activated factor X neutralization assay (anti-factor Xa assay) using a synthetic chromogenic substrate to show heparin neutralization of factor Xa action on the substrate. Another is titration with protamine sulfate or polybrene, both of which are specific inhibitors of heparin. The plasma heparin assays are necessary to monitor LMWH therapy, because LMWH has little effect on thrombin (whereas inactivation of thrombin is important in the APTT).

Protein C/protein S anticoagulant system

Protein C. Protein C (sometimes called Factor XIV) is a proteolytic enzyme of the serine protease type that produces anticoagulant action by inactivating previously activated factor V and factor VIII. The plasma half-life of protein C is 6-8 hours. Protein C is produced in the liver, is vitamin K–dependent, and circulates as an inactive preenzyme (zymogen). Activation is dependent on a protein called thrombomodulin, which is located on the cell membrane of blood vessel endothelial cells. Thrombomodulin binds thrombin and induces thrombin to attack inactive protein C (rather than the usual thrombin target fibrinogen), converting inactive protein C to active protein C. Activated protein C then attacks both activated factor VIII from the intrinsic factor pathway and activated factor V from the extrinsic factor pathway. Both actions inhibit conversion of prothrombin to thrombin (this end result is similar to the end result of warfarin action, although the mechanism is different). To reach maximum effect, protein C needs a cofactor, called protein S. Protein S is also synthesized in the liver and is also vitamin K dependent. Since protein C is a naturally occurring anticoagulant and protein S is necessary for protein C action, sufficient decrease in either protein C or S may result in a tendency for increased or recurrent vascular thrombosis. Protein C deficiency can be congenital or acquired. The congenital form is transmitted as an autosomal dominant trait. Individuals genetically heterozygous for protein C deficiency have protein C levels about 50% of normal (range, 25%-70%). Some, but not all, of these patients develop recurrent thrombophlebitis or pulmonary embolism. This usually does not occur before adolescence and is often associated with a high-risk condition such as surgery, trauma, or pregnancy. Homozygous protein C deficiency has been reported mostly in newborns, in which case protein C levels are close to zero and symptoms occur soon after birth and are much more severe.

Acquired protein C deficiency can be associated with parenchymal liver disease, vitamin K deficiency, DIC (especially when DIC occurs with neoplasia), and with certain medications, particularly coumarin anticoagulants. Warfarin reduces protein C activity levels about 40% (and also protein S) in addition to its action on the other vitamin K–dependent coagulation factors. Since protein C has a short plasma half-life, if protein C is already decreased before warfarin is administered, protein C decrease sufficient to induce thrombosis may occur before maximal decrease of factor VII, resulting in thrombosis before the anticoagulant effect of warfarin is manifest.

Protein C has an additional anticoagulation effect by increasing production of an enzyme called tissue plasminogen activator (t-PA) that initiates activity of the plasmin fibrinolytic system (discussed later).

It has been estimated that venous thrombosis without disease or drug-induced predisposition in persons less than age 45 can be attributed to protein C, protein S, or antithrombin III deficiency in about 5% of cases for each. About 20%-65% of thrombosis in this type of patient has been attributed to “activated protein C resistance” (APC resistance) that has a hereditary component. It is thought that factor V is responsible due to a genetic alteration.

Assay for protein C. Protein C can be assayed by quantitative immunologic or immunoelectrophoretic methods or by somewhat complex functional methods. Immunologic methods quantitate total amount of protein C. However, in some cases protein C is normal in quantity but does not function normally. Therefore, some investigators recommend functional assay rather than immunologic assay for screening purposes. If it is desirable to assay protein C (and protein S) quantity or function, it should be assayed before coumarin therapy is started because coumarin drugs interfere with vitamin K and decrease both protein C and S quantity and activity values.

Protein S. As noted previously, protein S is a vitamin K–dependent cofactor for protein C. Protein S exists about 40% in active form and about 60% bound to a protein belonging to the complement group. Total protein S is measured using quantitative immunologic or immunoelectrophoretic methods similar to those used for protein C.


Plasmin/fibrinolysin anticoagulants. Fibrinolysins are naturally occurring or acquired enzymes that attack and destroy either fibrinogen or fibrin or both. Fibrinolysins may be either primary or secondary. Primary fibrinolysins are acquired (not normally present), are rare, attack fibrinogen as their primary target, and are most often seen in association with malignancy or after extensive and lengthy surgery (most often lung operations). Secondary fibrinolysins are a part of normal body response to intravascular clotting, which consists of an attempt to dissolve the clot. The fibrinolysin produced is called plasmin, which is generated from an inactive precursor called plasminogen. Certain enzymes, such as streptokinase or urokinase, catalyze the reaction. Plasminogen also can be activated by tissue plasminogen activator (t-PA) which is regulated by another enzyme called t-PA inhibitor. Protein C exerts its action on this system by inhibiting t-PA inhibitor, which, in turn, releases t-PA to activate more plasminogen to become plasmin. Plasmin attacks the fibrin clot and splits the fibrin into various fragments (“split products,” or “fibrin degradation products” [FDPs]). The FDP can interfere with polymerization of fibrin monomer (which ordinarily becomes fibrin polymer) and thus complement the fibrin-dissolving action of plasmin by helping to retard further clotting. Secondary fibrinolysin usually does not attack fibrinogen, although it may do so if the concentration of fibrinolysin (plasmin) is high enough. Primary fibrinolysin seems to be very similar, if not identical, to plasmin. Primary fibrinolysins circulate without clotting being present, so that primary fibrinolysins ordinarily attack fibrinogen rather than fibrin.

In addition to plasminogen activator inhibitor, there is an enzyme that directly inhibits plasmin called alpha-2 antiplasmin. This substance prevents plasmin from prematurely dissolving vessel thrombi that are preventing injured vessels from bleeding. Alpha-2 antiplasmin forms a covalent complex with plasmin that inactivates plasmin and thereby prevents fibrin clot dissolution and formation of FDP.

Tests for fibrinolysins

Various tests available to screen for fibrinolysins include the clot lysis test, fibrinogen assay, thrombin time, PT, APTT, and tests for FDPs. The clot lysis test is the oldest. Patient plasma is added to the blood clot of a normal person, and the clot is observed for 6-24 hours for signs of clot breakdown. This procedure is slow and fails to detect weak fibrinolysins. Thrombin time, PT, and PTT depend on formation of fibrin from fibrinogen as the test end point and thus will display abnormal results if there is sufficient decrease in fibrinogen or sufficient interference with the transformation of fibrinogen to fibrin. Fibrinogen assay was discussed previously; fibrinogen can be measured either directly, by immunologic methods, or indirectly, by measuring fibrin after inducing transformation of fibrinogen to fibrin. These methods are all moderately sensitive to fibrinolysins but still fail to detect some low-grade circulating anticoagulants. The most sensitive tests currently available are those that detect FDP; these will be discussed in detail next.

Fibrinogen/fibrin split products (FDPs). Normally, thrombin catalyzes conversion of fibrinogen to fibrin by splitting two fibrinopeptide molecules, known as fibrinopeptide A and B, from the central portion of fibrinogen. This action exposes polymerization sites on the remaining portion of the fibrinogen molecule, which is now called fibrin monomer. Fibrin monomers spontaneously aggregate together in side-to-side and end-to-end configurations to form a fibrin gel. Thrombin also activates factor XIII, which helps introduce cross-linking isopeptide bonds between the fibrin monomers to form stabilized insoluble fibrin polymer. Fibrin polymer forms the scaffolding for blood clots. Fibrinolysins may attack either fibrinogen or fibrin, splitting off FDPs, which, in turn, are broken into smaller pieces. These split products may form a complex with fibrin monomers and interfere with polymerization. Fibrinogen degradation products can be produced by action of primary fibrinolysin on fibrinogen and also by action of plasmin on fibrinogen and fibrin monomers or fibrin clots formed in a variety of conditions, normal and abnormal. Plasmin attacks intravascular blood clots formed as part of normal hemostasis (e.g., trauma or surgery) as well as blood clots that produce disease (e.g., thrombosis or embolization). In addition, FDPs can be associated with intravascular clots composed of fibrin (e.g., DIC) and some partially extravascular conditions such as extensive malignancy, tissue necrosis and infarcts, infection, and inflammation.

Protamine sulfate test. Protamine sulfate in low concentration is thought to release fibrin monomers from the split-product complex and allow the monomers to polymerize. The test result is positive in conditions producing secondary fibrinolysin and is negative with primary fibrinolysin. Primary fibrinolysin will not induce a protamine sulfate reaction because the end point of this test depends on the presence of fibrin monomers produced by the action of thrombin. Primary fibrinolysin is not ordinarily associated with activation of the thrombin clotting mechanism.

Since DIC is the major condition associated with substantial production of secondary fibrinolysin, various studies endorse protamine sulfate as a good screening test for DIC. Negative results are strong evidence against DIC. Various modifications of the protamine sulfate method have been introduced. These vary in sensitivity, so that it is difficult to compare results in diseases other than DIC. Any condition leading to intravascular clotting may conceivably produce secondary fibrinolysin; these conditions include pulmonary embolization, venous thrombosis, infarcts, and either normal or abnormal postoperative blood vessel clots in the operative area. Occasional positive results have been reported without good explanation.

Immunologic fibrin degradation product assay. When fibrinogen or fibrin monomers are attacked by plasmin, a large fragment called X is broken off. In turn, X is split into a larger fragment Y, a smaller fragment D, and the smallest fragment of all, E. Therefore, intermediate products are fragments X and Y, and final products are two fragment Ds and one fragment E. The split products retain some antigenic determinants of the parent fibrinogen molecule. Antibody to certain of these fragments can be produced and, as part of an immunologic test, can detect and quantitate these fragments. Since the antibodies will react with fibrinogen, fibrinogen must be removed before the test, usually by clotting the blood with thrombin. Heparin may interfere with the action of thrombin, which, in turn, leads to residual fibrinogen, producing false apparent FDP elevation. The earliest immunologic FDP test in general use was the “Fi test,” a slide latex agglutination procedure detecting mainly intermediate fragments X and Y, which was shown to be insufficiently sensitive. Another test, the tanned RBC hemagglutination-inhibition test (TRCHII), detected fragments X, Y, and D and was sensitive enough but too complicated for most laboratories. A newer 2-minute latex agglutination slide procedure (Thrombo-Wellcotest) has immunologic reactivity against fragments D and E and seems to have adequate sensitivity and reliability. The Thrombo-Wellcotest is replacing protamine sulfate in many laboratories. All of the FDP tests are usually abnormal in the various conditions that affect the protamine sulfate test. Titers of 1:10 or less on the Thrombo-Wellcotest are considered normal. Occasional clinically normal persons have titers between 1:10 and 1:40. Patients with venous thrombosis develop a secondary fibrinolysin body response that can produce an abnormal Thrombo-Wellcotest result. In most of these cases the titer is greater than 1:10 but less than 1:40, but a minority of patients may have a titer of 1:40 or greater. Classic DIC induces a strong secondary fibrinolysin response producing large quantities of split products with a Thrombo-Wellcotest titer over 1:40. However, mild cases of DIC may have a titer in the equivocal zone between 1:10 and 1:40. Thus, there is overlap between normal and abnormal between 1:10 and 1:40. In addition, it is necessary to add a plasmin inhibitor (e.g., aprotinin [Trasylol]) to the patient sample soon after collection to prevent continued breakdown of fibrinogen or fibrin, which could result in falsely elevated FDP levels.

D-dimer test. Another type of FDP test detects breakdown products of plasmin action on fibrin clots. Since cross-linkage has already occurred, the degradation fragments also have some degree of residual cross-linkage and are called cross-linked FDP, or D-dimer (since fragment D is the major constituent). D-dimer assay is abnormal in nearly all of the same conditions as protamine sulfate or Thrombo-Wellcotest. However, D-dimer assay (like protamine sulfate) is normal with primary fibrinolysins, since D-dimers can be produced only after fibrin formation and cross-linking.

Disseminated intravascular coagulation (DIC)

The most common serious condition associated with fibrinolysin is DIC. Tissue thromboplastin or unknown substances with similar effect are liberated into the bloodstream and result in deposition of fibrin and fibrin clots in many small blood vessels, thus depleting plasma of the fibrin precursor substance fibrinogen (defibrination syndrome). Originally considered an obstetric disease associated with premature separation of the placenta and amniotic fluid embolism, DIC is now attributed to a growing list of etiologies, among which are septicemia, surgery complicated by intraoperative or postoperative shock, severe burns or trauma, extensive cancer, and newborn respiratory distress syndrome; and it is occasionally seen in many other conditions such as virus infection. Shock is the best common denominator but is not always present, nor has it been definitely proved to be either a cause or an effect.

Clinically, DIC is manifested by coagulation problems, which may be severe. Blood oozing from tissues during surgery is a frequent warning sign. Shock, acute renal failure, and acute respiratory failure are common in severe cases. Vascular thrombi and focal tissue infarction are also relatively frequent. Purpura fulminans refers to the rare cases in which bleeding into the skin dominates the clinical picture. Hemoglobin levels may be within reference range initially or may be decreased. Peripheral smear in classic cases displays many schistocytes, producing the picture of microangiopathic hemolytic anemia. Platelets are usually, although not always, decreased. Differential diagnosis is discussed later in the section on thrombotic thrombocytopenic purpura.

Laboratory diagnosis. Various laboratory tests have been advocated for the diagnosis of DIC, many of which have been discarded as newer tests are announced or older ones reevaluated. At present, hypofibrinogenemia and thrombocytopenia are the two best-established findings, which, occurring together, are considered strongly suggestive of DIC. The PT and APTT are usually both abnormal. Milder forms of DIC have been reported, however, in which one or all of these tests may be normal. The most sensitive and widely used tests for DIC screening are the protamine sulfate test and the immunologic FDP tests (e.g., the Thrombo-Wellcotest and D-dimer). The Thrombo-Wellcotest has been more reliable than protamine sulfate in my experience, but there are not sufficient comparison studies in the literature to permit definitive judgment between the two procedures. Since the two procedures give similar information, there is no need to use both for the diagnosis of DIC. D-dimer assay should give results similar to the Thrombo-Wellcotest.

Primary vs. secondary fibrinolysin. Occasionally it is necessary to distinguish between primary and secondary etiologies for a fibrinolysin. The protamine sulfate test and D-dimer assay are normal with primary fibrinolysins and abnormal with secondary fibrinolysins. This occurs because the protamine sulfate test needs fibrin monomers (and the D-dimer assay needs cross-linked fibrin) to produce a positive reaction, whereas with primary fibrinolysins, since the clotting (thrombin) mechanism is not ordinarily activated, no fibrin monomers or cross-linked fibrin molecules are present and FDPs are generated from fibrinogen rather than fibrin. Primary fibrinolysins can be treated with e-aminocaproic acid, or EACA (Amicar), but EACA is contraindicated in DIC, since inhibition of secondary fibrinolysin generation would aggravate the clotting disorder.

Immunologic-type coagulation factor inhibitors

There are several antibody-type inhibitors of coagulation factors. The most common are antiphospholipid antibodies (APA) and the antibodies that may develop against factor VIII:C. Nomenclature of the APAs is confusing because they were originally discovered in patients with systemic lupus erythematosis (SLE) and therefore were called “lupus anticoagulants.” Later, it was found that APAs did not act as anticoagulants, were found more often without than with SLE, and often did not react in the screening procedure traditionally used to detect the “lupus anticoagulant” antibodies. There are also some other differences between these antibodies that suggest two subdivisions of APAs, currently being named “lupus anticoagulants” (LAC) and “anticardiolipin antibodies” (ACA) and differentiated on the basis of certain laboratory tests, discussed later. In about 40% of patients with APAs, only one or the other type is present; in the other 60% of cases, both are present concurrently. APAs may be IgG, IgM, or IgA type. IgG is the type most often associated with complications; of these, most are thrombosis, either venous (70%) or arterial (30%). Thrombocytopenia also occurs and possibly hemolytic anemia. Increased spontaneous abortion or fetal loss has been reported in some studies but not in others. The antibodies vary considerably in titer, activity, and frequency. A considerable number are transient, and whether they are transient or longer-lasting seem to be a response to infection, acquired immunodeficiency syndrome (AIDS), inflammation, autoimmune diseases, malignancy, certain medications (especially chlorpromazine and procainamide; also dilantin, penicillin, and various antibiotics; hydralazine, and quinidine), and various or unknown antigens. The category known as LAC are IgG or IgM antibodies originally found in SLE (about 10%-15% of cases; range, 0.4%-65%), but have subsequently been reported in some patients with various conditions previously mentioned. In some cases, no cause is found. Clinical bleeding is uncommon unless some other coagulation abnormality is present (e.g., thrombocytopenia, present in about 25% of cases [range, 14%-63%]; and in about 50%-60% of cases when LAC occurs in SLE). Instead, vascular thrombosis is associated with LACs in about 30%-40% of cases (range, 14%-60%).

Laboratory Diagnosis.—The trademark of LACs is interference with coagulation factor assays that use phospholipid reagents or phospholipid-dependent procedures such as the APTT as part of the test method, even though the antiphospholipid LAC does not directly affect the actual coagulation factor that the test is supposed to be measuring. The most commonly used screening test is the APTT, since it is fast, inexpensive, and easily automated. The kaolin clotting time and the dilute Russell viper venom time have also been used; they are claimed to be a little more sensitive than the APTT but are more expensive and not automated. The APTT is elevated in about 90% of patients with LACs. In fact, unexpected APTT elevation is the usual way the LAC is detected (of course, other etiologies of elevated APTT must be considered, especially heparin contamination, which several reports found to be the most common cause for unexpected APTT elevation). Factor assays that use the APTT as part of the assay method are also abnormal. Different APTT reagents have varying phospholipid composition and quantity and therefore do not all have the same sensitivity for the LAC. This partially accounts for reports of APTT detection of the LAC ranging from 45% to 90%. A technical factor of importance is necessity for platelet-poor plasma when testing for LAC. Sensitivity for LAC improves as the number of platelets in the plasma specimen decreases. When LACs are present the PT is typically normal, although the PT can be elevated in about 20% (range 10%-27%) of the patients. The thrombin time is normal. The cardiolipin tests for syphilis (rapid plasma reagin, VDRL) are positive in about 20%-40% (range 0%-87%) of cases. The easiest diagnostic test for the lupus inhibitor consists of diluting the patient’s plasma with an equal portion of fresh normal plasma. The mixture should correct the APTT back to reference range (or nearly so) when the problem is a factor deficiency or antifactor antibody, but the APTT should not decrease substantially in the presence of a lupus inhibitor. Unfortunately, there are varying criteria in the literature as to what constitutes substantial correction. Also, there may be interpretive difficulty when the APTT elevation is less than 8-10 seconds. In this situation, some investigators prefer 1 part fresh normal plasma to 4 parts patient plasma, which is considered to be more sensitive than a 1:1 mixture. To further complicate matters, about 15% of LAC cases (range, 10%-40%) are reported to demonstrate substantial or complete correction. In many of these cases there may be time-dependent inhibition; that is, the APTT corrects but slowly becomes abnormal again 1-2 hours later during incubation at 37°C (this behavior is more typical of a factor VIII inhibitor). Equivocal or time-dependent results should probably be confirmed by a more specific test. Whether clear-cut (noncorrected) results need to be confirmed is not clear. Tests reported to be more sensitive and reliable for definitive diagnosis, although not widely available, include the tissue thromboplastin inhibition test and the platelet neutralization procedure. The platelet neutralization procedure appears to be more reliable and is commercially available in kit form. A phospholipid neutralization test has been reported that is very promising.

Tests for the ACA category include a solid-phase radioimmunoassay (the first reliable procedure developed) and nonradioactive immunoassays like enzyme-linked immunosorbent assay (ELISA). These generally would only be available in large commercial laboratories or university medical centers. Most are now using the ELISA method. These tests are necessary to detect non-LAC antiphospholipid antibodies (i.e., ACAs) and demonstrate that mixtures of LAC (shown by positive LAC tests discussed previously) and non-LAC ACAs are present.

Factor VIII antibody. Antibodies against factor VIII develop in approximately 15% of patients with hemophilia A (literature range, 5%-25%). Similar antibodies occasionally have been reported in nonhemophiliacs, where the presence and severity of clinical disease resembling hemophilia depends on the strength or titer of its antibody. This subject is discussed in the section on hemophilia A. The most commonly used screening test is the same as that for anticardiolipin antibodies.

Miscellaneous anticoagulants

Medications. Various drugs may affect coagulation. Aspirin and certain others have direct effects on clotting. Aspirin inhibits platelet adhesion and thus may potentiate a tendency to hemorrhage when a patient is receiving anticoagulant therapy. Other medications may act indirectly by enhancing or inhibiting the effects of anticoagulants.

Dysproteinemia. Some persons with abnormal serum proteins, most frequently persons with myeloma or one of the macroglobulinemias, have interference with conversion of fibrinogen to fibrin despite normal fibrinogen levels. This is manifested by poor clot stability and failure of clot retraction and may cause a hemorrhagic tendency or purpura.

Thrombolytic therapy. Various conditions predispose to thrombosis (see box on this page). Antithrombin III, protein C and S, and t-PA were discussed earlier. Many of the other conditions are discussed elsewhere. Stasis, vessel narrowing, and hyperviscosity are the most common etiologic problems. Cancer is a surprisingly important cause of thrombosis and embolization, especially in persons without evidence of the usual predisposing conditions. According to several studies about 5%-15% (range, 1%-34%) of cases of pulmonary embolization or deep vein thrombosis are associated with cancer, the cancer usually becoming manifest in less than 2 years. GI tract adenocarcinoma (especially pancreas), lung, and ovary are the most common primary tumors.

Various forms of therapy have been used to dissolve thrombi. Heparin has been used the longest, but heparin therapy has several problems (discussed earlier) and is better in preventing clot propagation than in dissolving the clot. Fibrinolysin (plasmin) can be used therapeutically to lyse intravascular thrombi and has produced better results than heparin. The most common indications are coronary artery thrombi, deep vein thrombosis (femoral, iliac, or popliteal veins), and large pulmonary emboli. A plasminogen activator can be administered to generate endogenous plasmin. To date, these activators include streptokinase (SK), urokinase (UK), tissue plasminogen activator (t-PA), anisoylated plasminogen-streptokinase activator complex (AP-SAC), and pro-urokinase (P-UK). SK is a protein that is produced by beta-hemolytic streptococci. It forms a complex with plasminogen; this complex then activates noncomplexed plasminogen (either circulating or fibrin bound) to form plasmin. However, this process has the potential of overactivating the system, depleting plasminogen, and producing hemorrhage. Also, there is a tendency to form anti-SK antibodies. UK produces therapeutic results at least as good as those of SK and is less antigenic. It directly activates plasminogen. However, it is very expensive. t-PA activator acts on plasminogen by forming a complex with plasminogen that is bound to fibrin. Therefore, t-PA preferentially acts on plasminogen at the site of clotting rather than circulating plasminogen, which is a major advantage. Disadvantages of t-PA include a very short half-life (2-5 minutes), which necessitates substantial doses for best results; an extremely high cost per dose at present; and reports of some systemic fibrinolytic effect when administered in large doses. Better therapeutic results are being reported for t-PA (at least for coronary artery thrombus dissolution) than results for SK and UK. AP-SAC is a modified SK-t-PA complex that can circulate without activating plasminogen. However, when the complex meets fibrin, such as a blood clot, the complex binds to the fibrin and then activates plasminogen that is also bound to the fibrin. P-UK is the inactive precursor of urokinase and circulates in the blood. When clotting occurs, it attaches to fibrin and forms (activated) urokinase there. Action is specific for fibrin, and the half-life of P-UK is considerably longer than t-PA, so that a smaller dose is needed. Once the decision is made to use a thrombolytic agent, current practice is to obtain a baseline blood specimen for either the APTT or the thrombin time. A second specimen is obtained 3-4 hours following administration of the thrombolytic agent. The desired result is a value sufficiently elevated above baseline so that there is no question of a true increase. The reason for testing is to prove that a thrombolytic effect has been obtained; other than this, the degree of test abnormality is not helpful since it does not correlate well with the actual amount of fibrinolysis taking place.

Some Conditions That Predispose to Thrombosis
Vessel lumen narrowing (e.g., atherosclerosis)
Vessel wall damage (e.g., vasculitis)
Stasis (e.g., congestive heart failure, immobile extremity)
Increased platelets (thrombocythemia)
Heparin-induced thrombocytopenia
Serum hyperviscosity (e.g., Waldenstrцm’s macroglobulinemia)
Deficiency of AT-III, protein C, or protein S
Oral (estrogen-containing) contraceptives
Antiphospholipid antibodies
Fibrin thrombi (e.g., DIC)
Platelet thrombi (thrombotic thrombocytopenic purpura [TTP])
Paroxysmal nocturnal hemoglobinuria
Malignancy (e.g., pancreatic carcinoma)