Coagulation Testing in the Core Laboratory

William E. Winter, MD; Sherri D. Flax, MD; Neil S. Harris, MD


Lab Med. 2017;48(4):295-313. 

In This Article

The Physiology of Coagulation

Following endothelial injury, a thrombus will normally form (Table 1). A thrombus is a blood coagulum formed in the blood vessel of a living person. The function of the thrombus is to stem the loss of blood to secure vascular integrity, therefore, maintaining blood volume. Elements of the thrombus, such as platelets, can then contribute to healing. It is rather amazing that flowing blood does not readily coagulate; however, upon removal of blood from the body when placed into a glass tube, the blood immediately begins to clot.

In contrast to a thrombus, a clot is a blood coagulum that develops 1) in blood vessels in the postmortem state, 2) when blood is removed from the body and is not anticoagulated, or 3) in a living person outside of their blood vessels (Table 1). Nevertheless, in practice, the terms blood clot and thrombus are often used interchangeably when, indeed, the terms are actually quite distinct. That said, we discuss the biology of blood coagulation.

The first challenge is to recognize that the coagulation events in vivo and in vitro are not identical. In order to properly interpret coagulation testing, the similarities and differences between in vivo and in vitro events must be understood. Ultimately, coagulation in vivo is the physiologically important event, and therefore we first examine in vivo coagulation.

In Vivo Coagulation

Hemostasis is the balance between factors that favor and factors that oppose thrombosis. Factors that favor thrombosis are endothelial injury, hypercoagulability. and stasis. This thrombotic triad was described more than 100 years ago by Virchow (Table 2). Opposing thrombus formation are the following factors: 1) the continuous flow of blood through the blood vessels, 2) the vascular endothelium, 3) prostacyclin production by endothelial cells, 4) heparan sulfate (HS) expression on endothelial surfaces, and 5) circulating plasma proteins that regulate coagulation including plasminogen, antithrombin (AT), protein S (PS), protein C (PC) and alpha-2 antiplasmin (Table 2). Once we examine coagulation in vivo, we return to the actions of plasminogen, AT, PC, PS and alpha-2 antiplasmin, and then discuss blood clotting in vitro.

Coagulation is frequently divided into 3 stages: 1) primary hemostasis (a function of platelets and von Willeband factor [VWF]), 2) secondary hemostasis that concludes with the formation of a fibrin lattice and 3) tertiary hemostasis that begins with fibrin crosslinking and ends with thrombus dissolution (which includes fibrinolysis).

Primary Hemostasis

Coagulation begins with endothelial injury. In terms of primary hemostasis, VWF binds to exposed subendothelial collagen (Figure 1). Similar to factor VIII (FVIII) and unlike other plasma protein clotting factors, VWF is produced by endothelial cells of the body. VWF monomers are 2813 amino acids in size. VWF monomers bind head-to-head and tail-to-tail, producing VWF multimers of 500 to 20,000 kDa. In VWF, amino acids 1 to 272 are involved in FVIII binding; amino acids 449 to 728 bind to the glycoprotein receptor Ib (GPIb) on platelets (Table 3); amino acids 911 to 1365 bind collagen; and amino acids 1744 to 1746 bind to the GPIIb/IIIa receptor on platelets.

Figure 1.

In most capillaries, endothelial cells form a continuous barrier separating blood from the basement membrane, subendothelial space (that includes collagen), and subendothelial cells. A break in this barrier exposes subendothelial collagen (upper panel). The initial event in primary hemostasis is the binding of von Willebrand factor to collagen (lower panel).

Via the GPIb receptor, platelets bind to VWF (bound to collagen) forming a monocellular cover over the site of endothelial injury (Figure 2). This constitutes platelet adhesion. In the venous system where shear stress is lower, platelets can bind directly to exposed collagen via the GPIa/IIa receptor.

Figure 2.

Once VWF has bound to collagen, platelets bind to VWF via the platelet's GPIb receptor. This event defines platelet adhesion. GPIb is part of a complex that also includes GPV and GPIX (ie, GPIb-V-IX).

Once platelets have adhered to the injured endothelium, via the collagen-VWF-GPIb interaction, the platelet becomes activated, leading to three important events (Figure 3): 1) the platelet changes shape, extending interlocking pseudopodia, 2) the platelet granules are discharged via the platelet canalicular system, and thromboxane A2 (TxA2) is synthesized and released, and 3) the GPIIb/IIIa receptor conformation changes, allowing it to bind plasma fibrinogen. The open canalicular system is the surface-connected channels that afford 1) transport into platelets and 2) the release of alpha granule contents from platelets. Functioning similarly to the sarcotubules of skeletal muscle, the dense tubular system regulates platelet activity by releasing or sequestering calcium.

Figure 3.

With platelet binding to VWF, platelets become activated, and 3 events occur: 1) platelets develop pseudopodia that allow platelets to interlock with one another providing strength to the thrombus; 2) platelets degranulate, releasing ADP, and activated platelets synthesize thromboxane A2 (TxA2); and 3) the molecular confirmation of the GPIIb/IIIa receptor changes allowing it to bind to fibrinogen. The release of ADP and TxA2 activates nonadhered platelets in the microenvironment. This will foster their involvement in aggregation.

Regarding platelet granules, the platelet's alpha granules contain various proteins including VWF, factors V and XIII, fibrinogen, fibronectin, P-selectin, platelet-derived growth factor, beta-thromboglobulin, platelet factor-4, and thrombospondin. Individual platelets have 40 to 80 alpha granules of 200 to 500 nm diameter that are round to oval in shape. Table 4 details the functions of these proteins. The platelet's delta (dense) granules contain ADP, ATP, pyrophosphate, calcium and serotonin. There are usually 4 to 8 such electron-dense granules per platelet. These granules contain 60% to 70% of the calcium in the platelet.

During platelet activation, ADP and TxA2 act in autocrine and paracrine fashions. The paracrine effect is to activate non-adhered platelets in the microenvironment. Via their activated GPIIb/IIIa receptors, these platelets can now participate in platelet aggregation as they bind to adhered platelets and other aggregated platelets via fibrinogen (Figure 4). In a cascade-like event, platelets further aggregate one to another via GPIIb/IIIa and fibrinogen to build the mass and strength of the thrombus. This completes primary hemostasis.

Figure 4.

Aggregation occurs when platelets are crosslinked via fibrinogen binding to GPIIb/IIIa on platelets.

The thrombus formed in primary hemostasis can produce an immediate reduction in blood loss, assuming that the size of the blood vessel lumen is not excessive. If a named blood vessel is cut (meaning that the vessel is large enough to be "named" by anatomists), thrombus formation alone will not usually re-establish hemostasis, and external mechanical compression and/or suturing will be required to halt continued blood loss.

Clinically a failure to establish primary hemostasis after injury results in immediate bleeding. Manifestations of primary hemostatic defects involving platelets or VWF include petechiae and mucosal bleeding.

Secondary Hemostasis

Secondary hemostasis is initiated with the binding of factor VII (FVII) to exposed subendothelial tissue factor (TF). TF (CD142) is a transmembrane protein that is expressed on subendothelial cells. However, TF is not normally expressed on vascular endothelial membranes. An archaic name for TF is "coagulation factor III." Once FVII binds to TF, FVII undergoes autocatalysis to FVIIa (a = activated; Figure 5 and Figure 6A). Although the process of the initiation of plasma clotting factor activation is quite complex, it is sufficient for our purposes to understand that TF-FVIIa in vivo ultimately leads to the conversion of factor IX (FIX) to FIXa (Figure 6B). In contrast, we will see that in vitro, the major action of TF-FVIIa is to convert factor X (FX) to FXa.

Figure 5.

Secondary hemostasis in vivo is activated when plasma gains access to TF on subendothelial cells that were exposed as a consequence of injury to the vascular endothelium. FVII binds to TF and undergoes autoactivation to FVIIa.

Figure 6.

As a consequence of formation of the TF-FVIIa complex, a plasma protein cascade ultimately leads to the conversion of fibrinogen to fibrin. The individual steps are described in detail in the text.

Returning to in vivo events, FIXa plus its cofactor factor VIII (FVIII) now acquires the ability to activate FX to FXa (Figure 6C). This action of FIXa-FVIII is that of a "Xase." Just as FIX has FVIII as a cofactor, FX has a cofactor: factor V (FV). Indeed the structure of FV and FVIII are similar. FXa-FV acquires prothrombinase activity converting prothrombin (factor II, FII) to thrombin (FIIa) (Figure 6D). Thrombin has numerous actions. Collectively these actions are sometimes described as the "thrombin explosion." Subsequently thrombin converts fibrinogen to fibrin (Figure 6E).

Fibrinogen is a symmetrical molecule composed of 2 sets of alpha, beta, and gamma chains (one set of the alpha-beta-gamma chains forms one half of fibrinogen). At the ends of the molecule are D domains, while in the middle of the molecule is an E domain. Thrombin converts fibrinogen to fibrin by cleaving fibrinopeptides A (FPA) and B (FPB) respectively from the alpha and beta chains of fibrinogen (Figure 7, top). Subsequently fibrin monomers form a noncovalent, overlapping lattice by associating one with the other (Figure 7, bottom). This lattice provides initial strength to the thrombus; however, fibrin crosslinking does not occur until early tertiary hemostasis, resulting in greater strength of the thrombus.

Figure 7.

With the release of fibrinopeptide A and fibrinopeptide B, fibrinogen is converted to fibrin via the action of thrombin-activated FXIIIa. Fibrin monomers initially stack, noncovalently forming a lattice.

Besides its action in converting fibrinogen to fibrin (Figure 8), thrombin 1) converts FV to FVa (with increased activity over FV), 2) converts FVIII to FVIIIa (with increased activity over FVIII), 3) feeds forward converting factor XI to FXIa, 4) activates factor XIII (FXIII, fibrin stabilizing factor) to FXIIIa and 5) stimulates platelets involved in primary hemostasis.

Figure 8.

The actions of thrombin are detailed in this figure (large arrows). This figure also explains the role of thrombin in the conversion of FXI to FXIa, which explains the role of FXI in normal hemostasis. FVa is more active than FV. FVIIIa is more active than FVIII. FXIII is fibrin-stabilizing factor.

The action of thrombin to convert FXI to FXIa allows thrombin and fibrin formation to continue until otherwise inhibited (eg, inhibited by tissue-factor-pathway-inhibitor [TFPI]). The action of FXIa is to convert FIX to FIXa. Therefore, FIX can be converted in vivo to FIXa in 2 ways: 1) via TF-FVIIa or 2) via FXIa.

If after successful primary hemostasis, no further procoagulant events occur (eg, there is a failure of secondary hemostasis), delayed bleeding is likely. Therefore, the role of secondary hemostasis is to strengthen the thrombus through the development of a fibrin matrix around and among the platelets. It is important to also point out that red blood cells (RBCs) are caught in the process of thrombus formation and contribute to the mass of the thrombus. Therefore, very severe anemia might impair normal hemostasis. Additional clinical manifestations of defects in secondary hemostasis include bruising, soft tissue hematomas, and, in children, hemarthrosis.

Tertiary Hemostasis

Tertiary hemostasis begins with the action of FXIIIa crosslinking D domains of adjacent fibin monomers in the lattice forming a polyfibrin meshwork (Figure 9). This adds to the strength of the thrombus preventing short-term dissolution of the thrombus. The later stages of tertiary hemostasis conclude with the resolution of the thrombus.

Figure 9.

The top image shows the noncrosslinked fibrin lattice. Via FXIIIa, D and E domains become crosslinked, adding strength to the thrombus (bottom image).

Temporally, primary hemostasis and secondary hemostasis actually occur concurrently. Furthermore, the events of primary hemostasis and secondary hemostasis interact in the formation of the thrombus. To understand this interaction, recall that factors II, VII, IX and X are vitamin K-dependent factors. The role of vitamin K is to assist in the addition of gamma carboxylate groups to glutamic acids in the GLA domains of these factors. Gamma carboxylation is necessary for proper synthesis of the factor and proper function of the synthesized factors. Without adequate levels of vitamin K, the synthesis of these factors is reduced and the function of the synthesized factors is also reduced.

It is believed that factors II, VII, IX, and X interact with the surface of the platelet, localizing and concentrating the events of secondary hemostasis (Figure 10). This changes the events from 3 dimensions to 2 dimensions concentrated on the platelet surface. In a sense, the surface of the activated platelet provides a physical "stage" for the formation of fibrin. As more and more fibrin is produced, the fibrin lattice is formed, which contributes to the thrombus. Crosslinking of fibrin via FXIIIa provides increased and necessary strength to maintain hemostasis.

Figure 10.

Primary hemostasis and secondary hemostasis occur concurrently and do interact. This drawing depicts the binding of factors VIIa, IX/IXa, X/Xa, and II/IIa to cell surfaces, localizing secondary hemostasis to a 2-dimensional framework, allowing active clotting factors to become concentrated.

Regulation of Coagulation

The regulation of coagulation is essential to prevent both bleeding and pathologic thrombi. Examples of pathologic thrombi are deep venous thrombi (DVT) in the legs and pelvis, and atrial thrombi that can occur in the setting of atrial fibrillation. The danger posed by DVT is thromboembolism to the lung producing a potentially fatal pulmonary embolism (PE). The danger posed by an atrial thrombus is thromboembolism to the brain producing acute ischemic stroke.

In the modern world, where death from hypovolemic shock is unusual, the most common causes of death involve pathologic or physiologic thrombi. PE and thromboembolic stroke are consequences of pathologic thrombi, whereas thrombosis of a medium-sized artery is the expected (physiologic) consequence of the fracture of an atherosclerotic plaque in a coronary or cerebral artery. Although hypercoagulability can contribute to atherosclerosis and accentuate thrombus formation, the major pathologic process in coronary artery disease, cerebrovascular disease, and peripheral vascular disease is the endothelial pathology of the atherosclerotic plaque. In a sense, thrombus formation following rupture of an atherosclerotic plague is expected and physiologic. Unfortunately, this "physiologic" thrombus formation can lead to acute ischemia or infarction.

To understand the balance between bleeding and thrombosis, we now review the factors that prevent excessive thrombus formation. The continuous flow of blood through arteries, capillaries, and veins modulates the time allowed for the interaction of the blood and the endothelium. Stasis predisposes to thrombosis; flow opposes thrombosis. The vascular endothelium creates a physical barrier between the blood and subendothelial collagen and TF. This prevents thrombus formation in the absence of endothelial injury. Normal endothelium also produces prostacyclin, which inhibits platelets (Figure 11). Therefore prostacyclin provides a counterbalance to the stimulatory effects of TxA2 produced by activated platelets.

Figure 11.

Hemostatic balance is provided by the barrier action of endothelial cells, endothelial cell production of prostacyclin, which inhibits platelets, and heparin sulfate that activates antithrombin.

Expressed on endothelial surfaces, HS activates AT to impair thrombus formation (Figure 11). The action of AT is to impair thrombin and FX, and, to a lesser extent, FIX and FXI. Physicians take advantage of the actions of AT when heparin is therapeutically administered. During ongoing coagulation, at least 3 further processes can prevent excessive thrombosis: 1) the activation of protein S, 2) the release of tissue plasminogen activator (tPA), and 3) the release of TFPI.

While thrombin is an extremely powerful and important procoagulant factor at the beginning of thrombus formation, later in this process, thrombin becomes a relative anticoagulant when it binds to a normal endothelial plasma membrane protein termed "thrombomodulin" (Figure 12). The binding of thrombin to thrombomodulin extinguishes the procoagulant action of thrombin and initiates its anticoagulant action. Thrombin-thrombomodulin activates PC producing activated PC (APC). APC plus protein S (PS) inactivates FVa and FVIIIa. It is worth noting that thrombin stimulates the formation FVa and FVIIIa early in coagulation, whereas, later in coagulation, thrombin triggers inactivation of FVa and FVIIIa.

Figure 12.

In the later stages of coagulation, thrombin (FIIa) binds to thrombomodulin. This quenches the procoagulant actions of thrombin. Furthermore, thrombin now acquires the ability to activate protein C (forming APC). APC plus protein S inactive FVa and FVIIIa inhibiting further thrombin generation. Injured cells release tissue plasminogen activator (tPA). tPA converts plasminogen to plasmin. Plasmin degrades cross-linked fibrin (eg, fibrinolysis) releasing fibrin degradation products such as D-dimers (not shown).

The release of tPA from injured tissues may be interpreted as "tissues requiring perfusion despite the possibility of hemorrhage." tPA catalyzes the conversion of plasminogen to plasmin (Figure 12). Plasmin degrades the crosslinked fibrin strands, producing a variety of fibrin-split (or degradation) products. One of these products is the D-dimer subunit that was earlier formed during FXIIIa-initiated fibrin crosslinking. In early tertiary hemostasis, FXIIIa normally catalyzes the crosslinking of D-domains between adjacent fibrin monomers present in the nascent lattice. An elevation in plasma (or serum) D-dimers is evidence of thrombus formation and subsequent thrombus breakdown. Not shown in Figure 12 is alpha-2 antiplasmin, a normal plasma protein that inhibits plasmin.

Lastly, TFPI is a physiological "brake" on the activity of TF-FVIIa (Figure 13). However even if the activity of the TF-FVIIa complex is extinguished by TFPI, the feed-forward loop of thrombin converting FXI to FXIa, etc, will keep coagulation running until a factor is exhausted or the process is otherwise inhibited (eg, by the actions of AT and APC plus PS).

Figure 13.

An overview of in vivo coagulation and the inhibitor role of tissue-pathway-factor inhibitor.

In Vitro Coagulation

We now explore coagulation in vitro, beginning with a discussion of 3 pathways: 1) intrinsic, 2) common, and 3) extrinsic (Table 5). The intrinsic pathway is initiated when fresh whole blood is placed in a glass tube. The negative charge of the glass initiates the contact pathway, converting factor XII (FXII) to FXIIa. FXIIa then converts prekallikrein (PK; bound to high molecular weight kininogen [HMWK]) to kallikrein (K). Kallikrein in turn converts more FXII to FXIIa, triggering a positive feedback loop for further generation of FXIIa (Figure 14). FXIIa catalyzes the conversion of FXI to FXIa (Figure 15).

Figure 14.

The contact pathway is outlined that initiates the in vitro intrinsic pathway.

Figure 15.

The aPTT includes the intrinsic and common pathways. The PT includes the extrinsic and common pathways.

As in the in vivo pathway, FXIa cleaves FIX to FIXa (Figure 15). Thereafter an Xase develops (FIXa plus FVIII or FVIIIa) cleaving FX to FXa and the process continues as described for the in vivo events leading to the formation of thrombin and fibrin. In summary the intrinsic pathway (named such because blood will intrinsically clot when added to a glass tube) includes PK, HMWK, and factors XII, XI, IX and VIII. The common pathway includes factors X, V, II (prothrombin), and I (fibrinogen), essentially identical to the events involving factors X, V, II, and I that occur in vivo.

The extrinsic pathway is triggered when tissue factor, phospholipid, and calcium are added to plasma anticoagulated with citrate. In essence, to make citrated plasma clot, extrinsic factors must be added. In vitro, FVII is activated to FVIIa and TF-FVIIa preferentially converts FX to Fxa, activating the common pathway. In contrast in vivo, the major action of TF-FVIIa is to convert FIX to FIXa.

In Vitro Pathways and Clotting Factor Tests

As we discusse below, the activated partial thromboplastin time (aPTT; including the intrinsic and common pathways) is triggered when a negatively charged substance (eg, silica, kaolin, celite, or elargic acid), partial thromboplastin (aka phospholipid) and calcium are added to citrated plasma. The aPTT consists of the intrinsic and common pathways.

The prothrombin time (PT; including the extrinsic [TF and FVII] and common pathways) is triggered when complete thromboplastin (aka phospholipid plus tissue factor) and calcium are added to citrated plasma. The PT consists of the extrinsic and common pathways. Both the PT and aPTT are recorded in seconds. These tests detect the generation of fibrin, which causes a change in resistance, light scattering, or viscosity of the sample. Neither the PT nor the aPTT include the events of early tertiary hemostasis: FXIII activity causing fibrin covalent crosslinking is not included in the PT or aPTT measurements. Therefore, people with FXIII deficiency have normal PT and aPTT (and thrombin time [TT]) values.

Upon reviewing the in vivo events, considering the in vitro extrinsic and intrinsic pathways, the in vivo activation of FIX to FIXa by TF-FVIIa constitutes a crossover pathway between the in vitro extrinsic pathway (TF+FVIIa) and the in vitro intrinsic pathway (FIX and FVIII) (Figure 16).

Figure 16.

In vivo, a cross-over pathway normally exists where TF-FVIIa (of the extrinsic pathway) activates FIX to FIXa (of the intrinsic pathway).