Basic Tests in Hemorrhagic Disorders
History. A history of easy bleeding or easy bruising should lead to further investigation.
Platelet count. Platelet disorders will be discussed later. Using the direct count (reference values 150,000-400,000/mm3; 150-400 x 109/L), a platelet count less than 100,000/mm3 indicates moderate thrombocytopenia, and one less than 50,000/mm3 indicates severe thrombocytopenia. Platelet number can be estimated with a reasonable degree of reliability from a well-made peripheral blood smear.
Clot retraction. Platelets have a major role in clot retraction. The clot shrinks and pushes out serum that was trapped within as the blood clotted. The shrunken clot is much firmer than it was originally. Normally, clot retraction begins at approximately 1 hour and usually is complete (a firm clot with about 50% serum and 50% clot) by 24 hours. There is deficient clot retraction in all typesof thrombocytopenia and in Glanzmann’s thrombasthenia, but clot retractionis usually normal in other varieties of platelet function abnormality. Clot retraction is rarely used today because of wider availability of accurate plateletcounts and also platelet function tests.
Tourniquet test. This test demonstrates capillary abnormality due either to an intrinsic defect in the capillary walls (capillary fragility) or to some types of thrombocytopenia. The tourniquet test is usually abnormal in immunologic thrombocytopenias such as idiopathic thrombocytopenic purpura (ITP) and drug-induced thrombocytopenia. It produces variable results, more often normal, in thrombocytopenia of other etiologies. It is usually normal in disorders that donot entail increased capillary fragility or thrombocytopenia, but occasionally it may be abnormal in patients with hemophilia or vitamin K disorders. The testis abnormal in hereditary telangiectasia. The tourniquet test is another procedure that is rarely used today since platelet counts are readily available, but it could be helpful in some instances of nonthrombocytopenic purpura if capillary fragility is suspected.
Bleeding time. The template (Mielke) modification of the Ivy technique is now the standard procedure for the bleeding time test. The template procedure, however, requiresthat a 9-mm long, 1-mm deep incision be made, which may leave a scar. Many laboratories (including my own) use commercial modifications of this technique suchas Surgicutt or Simplate, a disposable spring-driven small lancet in a plastic case, which makes a uniform and reproducible incision 5 mm long and 1 mm deep with minimal pain or sequelae. There are some technical factors that influence results; for example, an incision parallel to the antecubital fossa produces a longer bleeding time than one oriented perpendicular to that line. The bleeding time is most helpful as an indicator of platelet abnormality, either in number or function. The bleeding time is usually normal when the platelet count is decreased but still more than 100,000/mm3 (100 Ч 109/L). With platelet counts less than 100,000/mm3, there is a rough correlation between severity of thrombocytopenia and degree of bleeding time prolongation. The bleeding time is usually abnormal in congenital defects of platelet function such as Glanzmann’s thrombasthenia. The bleeding time is frequently abnormal in acquired platelet function abnormality such as that seen in uremia and the myeloproliferative syndromes. In uremia, there frequently aredemonstrable abnormalities in platelet function tests but not sufficient to entirely explain bleeding problems. In addition, up to 50% of uremic patients develop some degree of thrombocytopenia. In one study, about 45% of uremic patients had elevated bleeding times; the occurrence and degree of bleeding time elevation did not correlate well with the blood urea nitrogen (BUN) level since elevated bleeding times occurred at BUN levels as low as 50 mg/100ml (18 mmol/L) (although more often >100 mg/100 ml), whereas patients with normal bleeding times had BUN values as high as 180 mg/100 ml (64 mmol/L). The majority of studies report that actual bleeding in uremia is infrequent unless the bleeding time is elevated, but some studies do not agree. Reports indicate that the uremic bleeding tendency can be corrected by fresh frozen plasma, cryoprecipitate, or a synthetic analog of vasopressin (antidiuretic hormone, ADH) named 1-deamino-8-D-arginine vasopressin, or desmopressin (DDAVP).
Certain drugs interfere with platelet function and can produce a prolonged bleeding time, the most common and important being aspirin. After asingle small dose of aspirin, prolongation of the bleeding time over baseline is present by 2 hours or less, with maximum effect at about 24 hours. Whether the bleeding time exceeds population normal limits depends on whether the preaspirin value was in the lower or the upper half of the reference range and on individual susceptibility to aspirin. In one study, bleeding times that exceeded the reference range upper limit occurred in about 50% of clinically normalpersons even with a relatively small aspirin dose such as 10 grains (two adult tablets); sometimes even with one 5-grain tablet. About 5% of the population appear to be hyperresponders, producing relatively large and prolonged bleeding time elevations. Aspirin permanently affects all platelets in circulation; platelet life span is 7-10 days, and about 10% of the circulating platelets are replaced each day. After aspirin is stopped, it takes 2-3 days (range, 1-8 days) for sufficient production of new (nonaffected) platelets to reduce an elevated bleeding time to normal. Ethyl alcohol ingested with aspirin is reported to increase the magnitude and duration of the bleeding time prolongation. Pretreatment with aspirin (aspirin tolerance test) increases the sensitivity ofthe bleeding time to platelet function defects such as occur in von Willebrand’s disease. Even without aspirin the bleeding time is usually, although notalways, abnormal in von Willebrand’s disease (discussed later). The bleeding time may be abnormal in a variable number of patients with capillary fragility problems. The bleeding time typically is normal in the hemophilias and the vitamin K or fibrinogen deficiencies but may be abnormal in severe (or sometimes even in moderately severe) cases. The bleeding time is elevated in about 20% (range, 0%-65%) of hemophilia A patients. In one study, theIvy bleeding time was consistentlynormal even though the Simplate method was abnormal in a significant number of patients. Heparin can increase the bleeding time.
There is disagreement in the literature as to the clinical significance of aprolonged bleeding time as a predictor of hemorrhage during surgery. There definitely is not a good linear correlation be- tween degree of elevated values andprobability of hemorrhage. In general, however, when the bleeding time is 1.5 times the upper limit of the reference range, the possibility of excessive bleeding during surgery is somewhat increased. When the bleeding time is more than twice the upper reference limit, there is definitely an increased risk of excess bleeding. Others found no correlation between bleeding time elevation (at least, up to moderate elevation) and hemorrhage during surgery, with approximately the same numberof patients bleeding with and without elevated bleeding time. There is also disagreement as to the surgical risk after ingestion of aspirin or aspirin-containing compounds. In the majority of patients there does not seem to be a greatly increased risk. However, some patients develop a greater degree of platelet dysfunction than the average, so that the bleeding time is useful in uncovering these cases.
Prothrombin time (PT). The prothrombin time (PT) is used in three ways: (1) to monitor anticoagulant therapy with Coumadin, (2) as part of a general screen for coagulation systemdisorders, (3) as a liver function test. A “complete” tissue thromboplastin plus calcium is added to the patient’s plasma (complete thromboplastin contains tissue-derived material that activates the extrinsic coagulationsystem plus phospholipid that acts as a platelet substitute). Formation of a fibrin clot is the end point. The PT mainly indicates defects in the extrinsic coagulation system (prothrombin and factors V, VII, and X). If defect in fibrinogen is severe, it also will produce abnormal PT test results, since the test depends on an intact fibrin mechanism to generate the clot end point. However, thefibrinogen level usually must be less than 100 mg/100 ml (reference range, 200-400 mg/100 ml; 2-4 g/L) before hypofibrinogenemia affects the PT. Platelet or intrinsic system factor defects before the prothrombin to thrombin stage do not affect the PT because a complete thromboplastin reagent activates the extrinsic coagulation system and bypasses the intrinsic system.
Because the coagulation theory designates conversion of prothrombin to thrombin as the major reaction directly affected by thromboplastin activity, prothrombin is commonly considered the principal agent measured by the PT. Actually, the test is much more sensitive to factor VII than to prothrombin. Clinically this makes little difference, since both factor VII and prothrombin are altered by the same two major conditions affecting the extrinsic system—liver parenchymal disease (most often cirrhosis) and vitamin K deficiency. Vitamin K is discussed in the section on prothrombin. Interference with vitamin K metabolism is most often drug-induced (coumarin anticoagulation or, less frequently, use of certain cephalosporin antibiotics (cefamandole, cefoperazone, and moxalactam)or secondary to malabsorption. In some cases low vitamin K intake due to anorexia or use of prolonged intravenous (IV) feeding because of serious illness may accentuate previous subclinical degrees of vitamin K deficiency or the action of medications that affect vitamin K metabolism.
The effect of sodium warfarin on the PT can be influenced by the amount of vitamin K in the diet. Some other factors include the effect of previous low tissue vitamin K stores (potentiated by low dietary intake), impaired vitamin K absorption, or action of broad-spectrum antibiotics on gastrointestinal (GI) tract bacteria. In one report, antibiotic therapy superimposed on IV feeding was said to precipitate vitamin K deficiency in as little as 48 hours after admission.
The PT can be affected by heparin if the blood level is high enough. Levels ordinarily associated with continuous IV heparin usually do not prolong the PT significantly. With intermittent bolus (rather than continuous) full-dose IV heparin, heparin blood levels are relatively high (sufficiently high to prolong the PT shortly after administration; they decline thereafter with minimal (1-2 seconds) PT elevation at 2 hours. Subcutaneous heparin peak occurs later than with IV bolus and the heparin effect lasts longer. Coagulation monitoring may notadequately demonstrate this heparin effect since the tests are customarily not done until approximately 1 hour or less before the next heparin dose, thereby not reflecting peak heparin sodium activity. When heparin and warfarin sodium (Coumadin) are given together, heparin (even continuous IV heparin) affects the PT in addition to the usual effect of Coumadin. To obtain a valid PT, one must wait until the heparin effect has mostly disappeared (5 hours after IV bolus injection and 12-24 hours after subcutaneous injection). Heparin blood levels are increased in patients with a high hematocrit level (the same dose in a smaller plasma volume). This is especially important in neonates, who normally have a relatively high hematocrit level.
Also, it should be remembered that neonates and young children have higher reference-range values than older children and adults. The effect of heparin canbe neutralized in vivo and in laboratory specimens with protamine (1 mg of protamine sulfate given intravenously for every 100 mg of heparin in the last dose). Excess protamine should be avoided since protamine itself may elevate the PT. Other ways to eliminate heparin from laboratory specimens are neutralization with hexadimethrine bromide (Polybrene) or absorption onto a special cellulose material (Heparsorb). Warfarin effect can be treated by parenteral administration of vitamin K, after which the PT should return to normal within 6-24 hours. In bleeding emergencies, fresh frozen plasma can be used.
Certain technical problems are discussed after the following section on the activated partial thromboplastin time. Coumadin and heparin anticoagulation is discussed in a later section on Anticoagulant Systems.
Activated partial thromboplastin time. The precursor of the activated partial thromboplastin time (APTT) was the partial thromboplastin time (PTT). An “incomplete” thromboplastin reagent plus calcium is added to patient plasma, and the time necessary to form a fibrin clot is measured. The partial thromboplastin reagent is only a phospholipid platelet substitute without any of the other components of thromboplastin. The PTT was useful in detecting intrinsic factor abnormalities but was relatively insensitive to effects of heparin. Adding certain “contact activators” (usually chemicals or particulate matter, such as kaolin) to the PTT reagent was found to activate factor XII (contact factor) swiftly and uniformly and thus eliminate another variable in the clotting process. In addition, the activated APTT was found to be sensitive to heparin. The APTT is very sensitive tocoagulation factor deficiencies within the intrinsic system before the prothrombin to thrombin stage. It may also be abnormal in prothrombin or fibrinogen deficiencies but only if the defect is relatively severe (prothrombin or fibrinogen/fibrin abnormalities may affect the test because the test depends on fibrin clot formation as the reaction end point). The APTT is not as sensitive to prothrombin abnormalities as the PT because the extrinsic thromboplastin used inthe PT test is more powerful than the intrinsic system prothrombin activator complex generated by the APTT, thus enabling the PT to demonstrate relatively smaller defects in prothrombin. Platelet abnormalities do not influence the APTT.
Advantages of the APTT are adequate reproducibility (<10% variation), speed (reaction time of about 30-50 seconds), ease of performance, and suitability for automation. Disadvantages are the following:
1. Blood levels of heparin that are very much above anticoagulant range cause the APTT to become nonlinear, excessively prolonged, and unreliable.
2. Various techniques and equipment are used for the APTT readout (clot detection); the different machines as well as different companies’ reagents may produce results that deviate significantly on the same specimen. This may cause problems in comparing values from different laboratories.
3. The APTT is not affected by platelets, whereas platelets do influence heparin activity in vivo (platelets contain platelet factor 4 [PF-4], which inhibits the anticoagulant activity of heparin).
4. The APTT is affected by warfarin. When the PT is in the warfarin therapeutic range, the APTT is also prolonged and may even be above the APTT therapeutic range for heparin.
Technical problems that affect interpretation of PT and APTT test results. The labile factors are preserved better in citrate than in oxalate anticoagulant. Once exposed to air, citrated plasma is stable for only 2 hours at room temperature (22°-25°) and 4 hours in the refrigerator at 4°C. If collected in a vacuum tube and if the top has not been opened, the plasma is stable for a longer time (6 hours at room temperature in one study and 14-16 hours at 4°C in another study). Excess anticoagulant relative to the amount of plasma may affect results. Insufficient blood drawn in the tube means that less plasma is available for the amount of anticoagulant present, and the test resultsmay be falsely prolonged. The same thing may occur in blood that has a high hematocrit value, since in that case the excess red blood cells (RBCs) are replacing some of the plasma. When 3.8% sodium citrate (the most commonly used anticoagulant) is used to prepare plasma, both the PT and APTT may be falsely prolonged when the hematocrit value reaches 50%, and prolongation may become marked with a hematocrit value more than 60%. The APTT is affected more than the PT. The hematocrit effect is especially troublesome in neonates, since a newborn normally has a relatively high hematocrit value. A hematocrit value of 20% or less may produce a false decrease in the PT and APTT. If 3.2% sodium citrate is used, the effect of hematocrit level is reported to be considerably less. One of the most frequent causes for falsely elevated APTT (and sometimes PT) arises from attempts to keep IV lines open by heparin flushes. Usually this is not known to the phlebotomist. One report indicates that heparin tends to adhere to the wall of catheters (depending to some extent on the chemical composition of the catheter), which can affect results from specimens drawn from the catheter. In patients on constant-infusion heparin therapy, faulty or improperly calibrated delivery systems may be the reason for otherwiseinexplicable APTT fluctuations. Technical considerations are important even in patients not known to be receiving anticoagulants; according to one report, 40% of abnormal APTT results in these patients were eventually interpreted as being false elevations due to technical factors. Another report found a 2% incidence of false elevations in clinically normal persons and 11% inpatients being evaluated for bleeding. Recollection by an experienced technologist has solved many problems. Other possible causes for unexpected APTT elevation are circulating factor inhibitors or the lupus anticoagulant (discussed later). One study found 3% of routine preoperative APTT results elevated in patients below 12 years of age and 0.5% elevated in patients over age 12. All had inhibitors (acquired anticoagulants), most of which disappeared in less than 7 months (range, 1 day to 7 months). A few persisted. No patients bled excessively in surgery even though no therapy was given. The reference range for the APTT is higher in young children than in adults.
Activated whole blood clotting time. Whole blood is drawn into a vacuum tube containing a contact factor activator. The tube is incubated in a heat block at 37°C and tilted every 5 seconds. The normal upper limit is about 2 minutes. The test is claimed to be very reproducible. It can be used for coagulation defect screening and to monitor heparin therapy. It is said to have approximately the same sensitivity as the APTT in coagulation defects, and in heparin assay it is more linear than APTT at greater heparin effects. Many coagulation experts prefer this test to the APTT whenmonitoring heparin therapy. Disadvantages are the need for a portable heating unit and inability to automate the procedure (that the test is done at the bedside is considered by its proponents to be an advantage). The test is not sensitive to platelet variations.
Venous clotting time. The Lee-White method is preferred for venous clotting time (VCT). Technique is extremely important. When the usual three test tubes are used, number 3 is filled first, since the last blood to enter the syringe is probably least contaminated with tissue juice. If glass syringes are used, the test timing should bestarted as soon as blood enters the syringe. If plastic syringes are used, one can wait until blood enters tube number 3 (the first tube filled) to start the test timing, because clotting time in plastic material is prolonged. The tubes should be incubated at 37°C. The VCT is affected mainly by defects in the intrinsic pathway factors before prothrombin and by defects in fibrinogen/fibrin. It is not sensitive to platelet abnormalities and is relatively insensitive to abnormalities of prothrombin or factor VII (in which severe deficiency isrequired to produce significant VCT abnormality). In deficiencies of the intrinsic pathway factors or fibrinogen, the test is only moderately sensitive, requiring considerable deficiency to cause abnormal test results. The VCT is reasonably sensitive to heparin effect and was the original test used to monitor heparin therapy.
Disadvantages of the VCT are relative lack of reproducibility (>15% variation in most laboratories when the test is repeated on the same patient), necessity for 37°C incubation and careful technique (hardly ever observed by the average technologist), relatively long reaction time (5-15 minutes), the fact that each test must be done separately at the patient’s bedside, and the fact that platelets do not affect the test. The VCT has been replaced in most laboratories by the APTT or some other method that is more reproducible and more easily controlled.
Thrombin time. If commercially supplied thrombin is added to patient plasma, the time required to form a clot can be used to estimate the rate of fibrin formation. Variables include the amount of patient fibrinogen available and whether any inhibitors are present. The inhibitors could either be fibrin split products (which interfere with fibrin monomer polymerization) or antithrombins (e.g., heparin). Some manufacturers market reagents based on certain snake venoms that directly convert fibrinogen to fibrin without being affected by heparin. Fibrinolysins canalso produce abnormality both by destroying fibrin or fibrinogen and by producing fibrin split products. The thrombin time is therefore used as a test for abnormalities in the fibrinogen/fibrin stage of coagulation. The test is not as widely used as some of the other coagulation procedures.
Plasma fibrinogen. There are several methods for measuring plasma fibrinogen levels. The oldestmethods (“clottable protein”) involved precipitating a fibrin clot from plasma either by adding thrombin or by recalcifying the plasma (adding enough calcium to neutralize the chelating effect of the laboratory anticoagulant used to obtain the plasma). The fibrin clot was then assayed chemically by one of several indirect methods, or the change in optical density of the clot was measured. Newer techniques consist of immunologic methods using antibodies against fibrinogen and methods based on a modified thrombin time principle. At present, the most widely used is the thrombin time technique. Most of the techniquesare not reliable when high titers of fibrinolysins are present or when heparin is present. Fibrinogen can be measured using certain snake venom reagents (Reptilase or Ancrod) instead of thrombin; these venoms convert fibrinogen to fibrinand are not affected by heparin. True low plasma fibrinogen levels may be due to high titers of fibrinolysin (when fibrinolysins are present in high titer, fibrinogen may be attacked as well as fibrin) or to the disseminated intravascular coagulation (DIC) syndrome. Disseminated intravascular coagulation is by far the more common cause. A thromboplastin tissue substance or substance with equivalent action is liberated in the bloodstream and causes fibrin deposition (clots) in small blood vessels.
It might be useful to compare the results of some laboratory tests just discussed in the various phases of blood coagulation:
Test result* affected by abnormality in:
* VS = very sensitive; Ins = insensitive; Mod S = moderately sensitive; No = not affected.
†Factors XIII, IX, X, XI, and XII (not including platelets or common pathway.
In hemophilia (factor VIII defect), the various tests have approximately the following sensitivity:
PT—normal at all levels of factor VIII C deficiency.
VCT—normal until factor VIIIC activity levels are less than 2% of normal.
APTT—normal until factor VIII activity levels are less than 30%-35% of normal.
Note that normal persons may have 50%-150% of “normal” levels on factor VIII activity assay.