Chapter 124: Inherited Deficiencies of Coagulation Factors II, V, V+VIII, VII, X, XI, and XIII

Rare congenital deficiencies of plasma proteins involved in blood coagulation, such as fibrinogen, prothrombin, and factors V, V+VIII, VII, X, XI, and XIII, generally lead to lifelong bleeding disorders. These disorders have been described in most populations with an incidence varying from one case in 500,000 for factor VII deficiency, to one case in 2 to 3 million for prothrombin and factor XIII deficiency.^1,2 However, their relative frequency varies among populations, being higher in regions where consanguineous or endogamous marriages are common, partly as a result of increased high frequencies of specific mutant genes in these inbred populations.^3,4,5,6,7,8 Two large surveys were made by the World Federation of Hemophilia (WFH; www.wfh.org) and the European Network of the Rare Bleeding Disorders (EN-RBD; www.rbdd.eu), with the aim of collecting epidemiologic data and providing information to hemophilia organizations and treatment centers to reduce and prevent complications of bleeding. Data collected by these surveys showed that factor VII and factor XI deficiencies are the most prevalent rare bleeding disorders (RBDs), each accounting for approximately one-third of all RBDs, while the rarest disorders are factor II (prothrombin) deficiency and combined deficiency of factors V and VIII (Table 124–1). The severity of bleeding manifestations in affected patients is variable. The most typical symptom, common to all RBDs, is bleeding from the mucosal tracts or at the site of invasive procedures; life- and limb-endangering symptoms, such as umbilical cord and central nervous system bleeding, recurrent hemarthroses, and soft tissue hematomas, occur with higher frequency only in some severe deficiencies.^{9,10,11,12,13,14,15} Although heterozygotes for the coagulation factor deficiencies usually do not manifest a bleeding tendency, some cases of postdelivery and post–dental-extraction bleeding in heterozygotes for factor X deficiency have been reported.¹⁶

The complexity of blood coagulation and the large number of proteins and nonprotein substances involved necessitate that a global test be used to simply and reproducibly assess its function: the screening tests such as prothrombin time (PT) and activated partial thromboplastin time (aPTT) are the first step in evaluating patients reporting a clinical and family history of bleeding. The PT interrogates the extrinsic coagulation pathway, its prolongation is indicative of the deficiency of factor VII; a normal aPTT is highly dependent on the intrinsic coagulation pathway, so that its prolongation is indicative of deficiencies of factors XI, VIII, IX, and XII. However, all patients homozygous or compound heterozygous for factor XI deficiency have aPTT values longer than 2 SD above the normal mean,¹⁷ while heterozygotes substantially overlap the normal range.^17,18 Consequently, screening of patients for a hemostatic abnormality prior to surgery (which is recommended for Jewish patients because of the high prevalence of factor XI deficiency) identifies patients with severe factor XI deficiency. The prolongation of both the PT and aPTT indicates the lack of a factor belonging to the common pathway, including prothrombin, factor V or factor X. However, patients with factor X deficiency harboring mutations that cause a defect only in the tissue factor (TF) pathway will only display a prolonged PT and their aPTT will be normal. Other patients who carry mutations that only affect the intrinsic pathway activity of factor X, will exhibit a normal PT and prolonged aPTT.¹⁹ Abnormal results of either of these screening tests should be followed by mixing studies where equal amounts of patient plasma and normal plasma are mixed and retested; the relevant test time is normalized in patients with factor deficiencies, but is not corrected or only minimally corrected in patients with factor inhibitors. In case of correction, specific coagulation assays are then performed to make the diagnosis of the specific factor deficiency. To evaluate fibrinogen deficiency, all coagulation tests that depend on the formation of fibrin as the end point are necessary; hence, beside PT and aPTT, thrombin time (TT) has to be also performed. In factor XIII deficiency, PT and aPTT are normal. Diagnosis of factor XIII deficiency is established by demonstrating increased clot solubility in 5 M urea, dilute monochloroacetic acid, or acetic acid. However, this method, quantitative and not yet standardized, detects only severe factor XIII deficiency (with activity <5 percent), thus leading to a possible underdiagnosis of factor XIII deficiency. The factor XIII deficiency diagnosis protocol requires a number of assays, which test for both activity as well as antigen levels. In the case of estimation of factor XIII activity using quantitative (e.g., photometric assays, which measure the ammonia released during a transglutaminase reaction) or incorporation assays (dansylcadaverine-casein assay which measures the level of incorporation of a labeled amine into a protein substrate) during transglutaminase mediated cross linking,²⁰ the plasma blanking procedure is mandatory to avoid the factor XIIIa-independent ammonia release that could lead to incorrect results in the low-activity range (below 5 to 10 percent).^21,22 Factor XIII A-subunit antigen can be measured by enzyme-linked immunosorbent assay (ELISA).²³ Factor antigen assays are not strictly necessary for diagnosis and treatment but are necessary to distinguish type I from type II deficiencies that become very important in fibrinogen or prothrombin deficiency, where normal antigen levels and reduced coagulant activity (dysfibrinogenemia and dysprothrombinemia) are associated with higher risk of thrombosis. Hereditary factor V deficiency is also a peculiar case that can be confused with combined deficiency of factor V + factor VIII because the two entities have similar manifestations, and are characterized by prolonged PT and aPTT. Consequently, assays of factor V and factor VIII are mandatory for making the distinction.

CLASSIFICATION

The development of guidelines for classification of RBDs has been historically hampered by a lack of sufficient knowledge about epidemiology and clinical outcomes, the difficulty in recognizing affected patients and collecting longitudinal clinical data, the limits of laboratory assays, and a lack of consensus concerning the criteria by which these disorders are classified. Classification of RBDs based on the residual level of plasma coagulant activity of the missing factor has considered for many years all RBDs as a single entity and a mild, moderate, or severe classification as in hemophilia was adopted (except for some disorders such as afibrinogenemia and factor XIII deficiency). In 2012, the Rare Bleeding Disorders Working Group, under the umbrella of the Factor VIII & Factor IX Scientific and Standardisation Committee (SSC) of the International Society on Thrombosis and Haemostasis (ISTH), analyzed the results of data coming from four registries (EN-RBD, the United Kingdom Haemophilia Centre Doctors’ Organization registry, the North American Rare Bleeding Disorders Registry, and the Indian registry) for a total of 4359 patients. Despite the large number of patients evaluated in this overview (both from the literature and the aforementioned registries), there is a large heterogeneity in the preassigned severity definitions for both coagulant activity and bleeding symptoms.²⁴ At the same time the EN-RBD, based on a cross-sectional study using data from 489 patients and involving 13 European treatment centers, for the first time evaluated the correlation between the coagulant residual plasma activity level and clinical bleeding severity in each RBD. Clinical bleeding episodes were classified into four categories of severity based on the location and the potential clinical impact, as well as the trigger of bleeding (spontaneous, after trauma or drug induced). By means of linear regression analysis, this study found a strong association between coagulant activity level and clinical bleeding severity for fibrinogen, combined factors V + VIII, X and XIII deficiencies. A weak association with clinical bleeding severity was present for factors V and VII deficiencies, while coagulation activity level of factor XI did not predict clinical bleeding severity. From the same study it also clear that the minimum level to ensure complete absence of clinical symptoms is different for each disorder, leading to the conclusion that RBDs should not be considered as a single class of disorders, but instead studies should focus on the evaluation of specific aspects of each single RBD and different from hemophilia.²⁵

MOLECULAR ANALYSIS

The molecular diagnosis of RBDs is based on the mutation search in the genes encoding the corresponding coagulation factor. Exceptions are the combined deficiency of coagulation factors V+VIII, caused by mutations in genes encoding proteins involved in the factor V and factor VIII intracellular transport (multiple combined-factor deficiency [MCFD] 2 and mannose-binding lectin [LMAN] 1) and the combined deficiency of vitamin K–dependent proteins (prothrombin and factors VII, IX, and X), caused by mutations in genes that encode enzymes involved in posttranslational modifications²⁶ and in vitamin K metabolism (γ-glutamyl carboxylase [GGCX] and vitamin K epoxide reductase–oxidase complex [VKORC1]).²⁷ Coagulant factors genes are located on different chromosomes except for the genes of factor VII (F7), factor X (F10), fibrinogen (FGA, FGB, FGG), and factor XI (F11) (Table 124–2). In particular F10 lays only 2.8 kb downstream of the F7 thus the combined deficiency of the two factors can be also the result of chromosomal abnormalities of the long arm of chromosome 13.²⁸ The strategy for molecular analysis is generally based on polymerase chain reaction amplification followed by Sanger sequencing of all exons, flanking intronic sequence and 5′ and 3′ untranslated regions. In contrast with hemophilia A, caused in approximately half of the patients by an inversion mutation involving introns 1 or 22 of the factor VIII gene, RBDs are often caused by mutations unique for each kindred and scattered throughout the genes. Information on already identified mutations causing RBDs is traceable from the mutation database on the ISTH website (http://www.isth.org/?MutationsRareBleedin). Missense mutations are the most frequent gene abnormalities, representing 50 to 80 percent of all identified mutations, except for LMAN1 variants where the most frequent mutations are insertions/deletions (50 percent). Insertion/deletion mutations represent 20 to 30 percent of the gene variations of the fibrinogen, factor V (F5), MCFD2, and factor XIII (F13A) genes, and less than 15 percent of the remaining coagulation factor gene mutations. Splicing and nonsense mutations comprise 5 to 15 percent of all identified mutations in all coagulation factors, with a maximum rate of 20 percent in the LMAN1 gene. Variants located in the 3′ and 5′ untranslated regions of the genes are the least-frequent types of mutation (<5 percent) found only at the fibrinogen, factor VII, factor XI, and factor XIII loci. The combined presence of more than one recessively transmitted coagulation factor defect may also rarely occur resulting in combined deficiency of factors VII and X^31,32,33 and combined deficiency of factors VII and V, VIII, X, or XI.³⁴ Despite significant advances in our knowledge of the genetic basis of the RBDs, in 5 to 10 percent of patients affected with severe clotting factor deficiencies, no genetic defect can be found. In these patients, the use of next-generation sequencing might help to identify novel pathways in coagulation disorders.

Treatment of RBDs is a difficult task because the absence of longitudinal clinical data and the limitations of available laboratory assays make it difficult to develop evidenced-based guidelines for the diagnosis and treatment of RBDs. A patient’s personal and family history of bleeding are important guides for management. Dosages and frequency of treatment depend on the minimal hemostatic level of the deficient factor, its plasma half-life (see Table 124–3) and the type of bleeding episode. At variance with patients affected with hemophilia A or B who have vastly improved the quality of life from advances in the manufacture of safe and effective products,³⁵ patients with RBDs have seen less progress. The main treatments in RBDs are represented by replacement therapy of the deficient coagulation factor and nontransfusional adjuvant therapies (antifibrinolytic amino acids, estrogen/progestin). Fresh-frozen plasma (FFP) and cryoprecipitate are the backbone of RBD treatment, particularly in those countries with low economic resources. However, specific plasma-derived concentrates are currently available only for fibrinogen and factors VII, XI, and XIII, and they are licensed only in some European countries; replacement therapy of coagulation factors may require the prescription of unlicensed products that are not readily available.

Prothrombin and factor X deficiencies are often treated with prothrombin complex concentrates (PCCs), which often also contain uncontrolled amounts of factor II, factor VII, and factor X. Products to cover the need for a dedicated therapy of patients with factor V deficiency and to facilitate the prophylaxis scheme in patients with factor X deficiency are of recent production. Finally, only two recombinant products are currently available for treatment of RBDs: recombinant factor VIIa (rFVIIa; see factor VII deficiency paragraph) and rFXIII (see “Factor XIII Deficiency” below). Although there are a number of reports available in the literature reporting on treatment on demand and by prophylaxis in RBDs,^36,37 no clear cut guidelines are yet available apart from those of the United Kingdom Haemophilia Centre Doctors’ Organization.³⁸ Table 124–3 shows available treatment for each deficiency and suggested dosages.

Women with RBDs require specific attention and care because in addition to experiencing the common associated bleeding symptoms, they may also experience bleeding complications from regular hemostatic challenges during menstruation, pregnancy, and childbirth, as well as from other gynecologic conditions, such as hemorrhagic ovarian cysts, endometriosis, hyperplasia, polyps, and fibroids. Menorrhagia, defined as blood loss of more than 80 mL per menstruation, is reported to be one of the most important symptoms in women with RBDs.^39,40 Menstruation may be quite problematic for women with coagulation disorders who have excessive blood loss, which can have a major impact on their quality of life and employment.

Pregnancy and childbirth pose particular clinical challenges to women with RBDs, as apart from factor XI deficiency, detailed information about these issues and their management are very scarce and limited to just a few case reports.^41,42 Pregnancy is accompanied by increased concentrations of fibrinogen, factor VII, factor VIII, factor X, and von Willebrand factor, particularly marked in the third trimester.^{43,44,45,46,47} In contrast, prothrombin, factor V, factor IX, and factor XIII are relatively unchanged.⁴³ All of these changes contribute to the hypercoagulable state of pregnancy and, in women with RBDs, contribute to improved hemostasis. Despite improved hemostasis, however, women with factor deficiencies do not achieve the same factor levels as those of women without factor deficiencies,³⁹ increasing the possibility of pregnancy loss or bleeding complications, especially if the defect is severe.

DEFINITION

Inherited prothrombin deficiency is one of the rarest coagulation factor deficiencies. It presents in two forms: type I, true deficiency (hypoprothrombinemia), and type II, in which a dysfunctional prothrombin is produced (dysprothrombinemia). These autosomal recessive disorders are genetically heterogeneous, and characterized by a mild to moderate bleeding tendency. Both types of prothrombin deficiency impair the generation or function of thrombin, the central enzyme of the blood coagulation system.

PROTEIN

Prothrombin, approximate Mr 72,000, is structurally homologous with other members of the vitamin K–dependent proteins, factors VII, IX, and X, proteins C, S, and Z, and bone γ-carboxyglutamic acid (Gla) protein. Prothrombin is synthesized in the liver as a prepropeptide of 622 amino acids and its plasma concentration is 100 to 150 mcg/mL. The circulating protein in its mature form is a single chain glycoprotein of 579 residues, composed of the Gla domain (residues 1 to 37) and the catalytic domain (residues 272 to 579), where a light A chain is disulfide-bonded to the heavy B chain containing the catalytic triad. In the zymogen molecule there are several exodomains, such as two kringle domains—kringle 1 domain (F1; residues 38 to 155), kringle 2 domain (F2; residues 156 to 271)—and the prepropeptide regions.^48,49 The prepropeptide domain is responsible for protein processing, targeting, and carboxylation, and it is removed prior to secretion from the cell. The Gla domain constitutes the aminoterminus of the mature prothrombin molecule and contains the 10 glutamic acid residues that are posttranslationally modified through action of vitamin K–dependent carboxylase to Gla. As a result of this modification, prothrombin acquires the capacity to bind calcium and membranes containing acidic phospholipids. The kringle domain contains two extensively folded, disulfide-bonded “kringle” motifs. They are present in diverse proteins and are thought to mediate protein–protein interactions. For example, the second kringle mediates interaction of prothrombin with activated factor V.⁵⁰ Notably, previous data shows that the kringle 2 domain, generated from the precursor prothrombin that is endogenously expressed in human, mouse, and rat brains, including in dopaminergic neurons in the nervous system,^51,52 is able to activate in vitro microglia cells and may be involved in the neuropathologic processes of dopaminergic neuronal death occurring in Parkinson disease.⁵² However, the real clinical significance of these in vitro findings is still to be unraveled. The catalytic domain contains the enzyme active site, which is responsible for fibrinogen cleavage. The residues characteristic for the serine protease family, His363, Asp419, and Ser525, constitute a charge relay system responsible for bond cleavage. The crystal structure of prothrombin has not been determined, but the crystal structure of human α-thrombin complexed with D-Phe-Pro-Arg chloromethylketone (an inhibitor that is a transition state analogue covalently bound to the enzyme) has been determined.⁵³

Prothrombin plays a central role in coagulation, functioning in both TF and contact activation pathways. Prothrombin is converted to its proteolytically active form thrombin by the prothrombinase complex consisting of activated factor X, factor Va, and phospholipid surface of platelets and other cells. Two forms of thrombin are generated: meizothrombin, if prothrombin is cleaved at residue 320, and α-thrombin, if cleavage occurs first at residue 271, removing prothrombin fragment 1.2, and subsequently cleaved at residue 320. The α-thrombin A-chain (residues 272 to 320) formed by factor Xa cleavage is encoded by exons 8 and 9. The B-chain (residues 321 to 579) containing the catalytic site and regulatory elements is encoded by exons 9 to 14. Thrombin is a multifunctional serine protease. In addition to converting fibrinogen to fibrin, thrombin also exerts functions in the coagulation cascade, consisting of both pro- and anticoagulant effects⁵⁴ and activates platelets by cleavage of the protease-activated receptor (PAR)-1 and PAR-4, initiating signals leading to platelet adhesion and aggregation.^55,56 Thrombin also stimulates wound healing through its action as a growth factor and its proangiogenic activity.⁵⁷

GENETICS

The prothrombin gene is located on the short (p) arm of chromosome 11.⁵⁸ It is 20-kb long and consists of 14 exons separated by 13 introns. Fifty-four mutations that cause prothrombin deficiency have been identified, of which 42 are missense, three nonsense, seven deletions/insertions, and two splicing mutations (see mutation database on the ISTH website, http://www.isth.org/?MutationsRareBleedin and Ref. 9). Type II deficiency (dysprothrombinemias) results from missense mutations that are located throughout the gene. As expected, many mutations are in the catalytic domain, imparting catalytic dysfunction on thrombin. Other mutations give rise to abnormally slow activation of prothrombin. Only about 10 mutations were identified in patients with type I deficiency, of which five were present in homozygotes. Globally, there is a clear prevalence of patients with a Latin/Hispanic origin, as nearly 70 percent of all patients with thrombin gene defects come from such areas (Barcelona, Padua, Segovia, and Puerto Rico).⁹

A number of polymorphisms have been identified in the prothrombin gene. One of these polymorphisms, a G>A change at nucleotide 20210 in the 3′ untranslated region of the prothrombin gene, is associated with increased plasma levels of prothrombin and an increased tendency to venous thrombosis (Chap. 130).⁵⁹

CLINICAL MANIFESTATIONS

According to a recent classification by the SSC of the ISTH, prothrombin deficiency may be classified as severe, moderate, and mild corresponding to blood levels of less than 5 percent, 5 to 10 percent, and greater than 10 percent, respectively.²⁴ In severely prothrombin-deficient patients, bleeding may be marked, including spontaneous hemarthroses; less-severe patients may show mild to moderate mucocutaneous and soft-tissue bleeding that usually correlates with the degree of functional prothrombin deficiency. Heterozygous subjects, having plasma prothrombin levels between 30 and 60 percent of normal, are usually asymptomatic; however, occasionally, excessive bleeding after moderate-intensity trauma, tooth extractions, or after surgical procedures may occur. Patients with dysprothrombinemias show a variable bleeding tendency that is usually less severe than in type I deficiency. Women with prothrombin deficiency may suffer from menorrhagia. Because of the extreme rarity of such deficiency, reports on event during pregnancy/delivery are very scarce, being only one describing in four of eight pregnancies in a hypoprothrombinemic woman.⁶⁰ In the same report, one postpartum hemorrhage (PPH) episode on the four-term pregnancies was reported, despite administration of clotting factor concentrate⁶⁰; however, this data was not confirmed by a following Iranian series, including a total of 14 patients with the same deficiency (coagulant activity levels 4 to 10 percent).⁶¹

Undetectable plasma prothrombin probably is incompatible with life, as inferred from the partial embryonic and neonatal lethality of prothrombin knockout mice, which do not survive to adulthood.^62,63

THERAPY

Replacement therapy is needed only in severe patients, in case of bleeding or to ensure adequate prophylaxis before surgical procedures. In severe clinical settings, higher levels of prothrombin may be achieved with FFP, or with PCCs, which avoids the risk of volume overload sometimes associated with the use of FFP.⁶⁴ However, PCCs contain other vitamin K–dependent coagulation factors (VII, IX, and X) and small amounts of their activated forms, which could potentially induce thrombotic complications; those containing an amount of factor VII below 10 percent are commonly known as three-factor PCCs. These concentrates are heated or treated with solvent–detergent, processes that remove HIV, hepatitis B, hepatitis C, and other viruses, but which do not remove parvovirus B19 or hepatitis A virus^65,66,67; the latter viruses can be effectively removed by dry heat and nanofiltration.⁶⁸ However, transmission of other possible bloodborne agents, such as prions causing Creutzfeldt-Jakob disease and its new variant, have not been totally eliminated. Bruises and mild superficial bleeding generally do not require replacement therapy. Antifibrinolytic agents (tranexamic acid and gabexate mesylate) have also been used for minor surgical procedures. The oral contraceptives have been shown to exert beneficial effects on menometrorrhagia in women characterized by prothrombin coagulant levels less than 3 percent.⁹ Thromboprophylaxis in dysprothrombinemic patients considered at high risk for a thrombotic event (e.g., orthopedic surgery) is a controversial issue. It is likely that administering low-molecular-weight heparin prophylactically to surgical patients at the same doses and schedules as those recommended for nondefect patients having similar procedures may be a valuable and safe procedure after correction of factor II deficiency by FFP or PCC infusion.

DEFINITION AND HISTORY

Hereditary factor V deficiency was initially termed parahemophilia because of its similarities with classical hemophilia.⁶⁹ In most of the affected individuals the phenotype is characterized by the concomitant deficiency of factor V activity and antigen (type I deficiency); however, approximately 25 percent of the patients have normal antigen levels (type II deficiency), thus indicating the presence of a dysfunctional protein.⁷⁰

PROTEIN

Factor V is synthesized by the liver⁷¹ and its plasma concentration is approximately 20 nM (7 mcg/mL).^72,73,74 Factor V is a high-molecular-weight (Mr ~330,000), single-chain, large glycoprotein that consists of 2196 amino acids that bears significant, regional sequence homology to factor VIII. Analysis of the approximately 7-kb factor V complementary DNA showed that the protein is organized according to the following domain structure: A₁-A₂-B-A₃-C₁-C₂. The A and C domains have approximately 40 percent homology with analogous domains in factor VIII.^74,75 The large B domain shows no homology with the corresponding B domain of factor VIII. Factor V is converted to its activated form following several proteolytic cleavages by thrombin⁷⁶ or factor Xa.⁷⁷ These cleavages remove the B domain and yield factor Va, which consists of a heavy chain (A₁-A₂ domains) associated by Ca²⁺ with a light chain (A₃-C₁-C₂ domains). The light chain contains the binding sites for membrane phospholipids, prothrombin, and activated protein C; both light and heavy chains probably are necessary for factor Xa binding. Assembly of factors Va and Xa on the phospholipid membrane of platelets in the presence of calcium ions forms the prothrombinase complex, which catalyzes the conversion of prothrombin to thrombin. The contribution of factor Xa in the absence of factor Va to overall thrombin generation is relatively minor. Importantly, incorporation of the cofactor into the macromolecular enzyme complex enhances prothrombin activation by several orders of magnitude.⁷⁸

In addition to hepatocytes, the primary site of factor V secretion, approximately 20 percent of the protein in whole blood is localized in the α granules of platelets, where it is complexed with an extremely large protein, multimerin.⁷⁹ Megakaryocytes do not synthesize factor V; rather, endocytosis of plasma-derived factor V accounts for the platelet factor V pool.⁸⁰ Following endocytosis factor V is modified intracellularly; these changes to platelet factor V appear to provide the cofactor with unique physical and functional characteristics, which render it more procoagulant compared with its plasma counterpart.⁸¹ Platelet degranulation and release of platelet factor V at the site of vascular injury is thought to be a critical contributor to the local factor V concentration. Furthermore, there is some evidence that, because platelet factor V is locally released in high concentrations, it is less susceptible to inhibition and may function normally in hemostasis. Factor Va is inactivated by activated protein C through limited proteolysis at Arg506, Arg306, and Arg679 in the presence of protein S, calcium ions, and either platelet or endothelial cell membrane phospholipids.⁸² Partial protection from this cleavage is provided by factor Xa when bound to factor Va on the surface of platelets.⁸³

GENETICS

The factor V gene maps to chromosome 1q21–25.⁸⁴ It is greater than 80 kb in length and the coding sequence is divided into 25 exons, ranging in size from 72 to 2820 base pairs (bp), and 24 introns, varying between 0.4 kb and 11 kb.⁸⁵ The sequence encoding the large B domain is contained within exon 13.

A total of 132 distinct mutations of the factor V gene have been identified, of which 64 are missense, 36 are insertions/deletions, 17 are nonsense, 15 are splice site mutations, and one is a deletion of the whole gene (see http://www.isth.org/?MutationsRareBleedin and Ref. 10). Most mutations cause truncations and are localized throughout the gene. Several mutations have interesting features. One, a Tyr1702Cys transition, was identified in eight unrelated families, of whom six were Italian. The frequency of this mutant allele in Italy is 0.002.⁸⁶ Another mutation, an Ala221Val (New Brunswick) alteration, characterized in the homozygous state by activity and antigen levels of 29 and 39 percent of normal, respectively displays decreased stability of the expressed protein and was the first genetic defect reported to be associated with type II deficiency.⁸⁷ Additional mutations exhibit decreased secretion of the protein from producing cells.^88,89 Remarkably, the Gln773ter and Arg1133ter mutations and a 4-bp deletion mutation, all present in exon 13 and predicted to result in partial truncation of the B-domain and complete truncation of the A3-, C1-, and C2-domains, cause no bleeding or only a mild bleeding tendency in affected patients having factor V antigen and activity levels 1 percent of normal.^90,91,92

Factor V Leiden (Arg506Gln) is a highly prevalent (up to 5 percent in some populations) polymorphism in the factor V gene that decreases the efficiency of factor Va inactivation by activated protein C.⁹³ Patients with factor V Leiden are at increased risk of unprovoked thrombosis, with homozygotes at greater risk than heterozygotes. The trans association of factor V Leiden and a mutation in factor V that causes factor V deficiency results in a prothrombotic state comparable to factor V Leiden homozygosity. This is sometimes termed “pseudohomozygous” activated protein C resistance and does not cause bleeding despite low factor V antigen levels.⁹⁴ Among several polymorphisms detected in the factor V gene, His1299Arg in exon 13 is particularly interesting because it is associated with a reduced plasma factor V level and mild activated protein C resistance.⁹⁵ His1299Arg co-segregates with several other polymorphisms encoding several amino acid changes, together named R2 haplotype.⁹⁶ In two heterozygotes for factor V Arg506Gln mutation who presented with venous thrombosis, reduced factor V activity resulting from the His1299Arg polymorphism harbored by the non-Leiden chromosome, imparted a pseudohomozygous phenotype for activated protein C resistance.⁹⁷ Additional polymorphisms or mutations in the factor V gene have been observed to increase the risk of venous thrombosis.⁹⁸

In addition, there are at least two examples in which platelet factor V is reduced. In the Quebec platelet disorder, initially described as an autosomal dominant disorder with severe bleeding manifestations, platelet factor V levels are reduced because of enhanced proteolysis resulting from overexpression of urokinase-type plasminogen activator,⁹⁹ as they are in factor V New York.¹⁰⁰

CLINICAL MANIFESTATIONS

Factor V deficiency is inherited as an autosomal recessive trait. Heterozygotes, whose plasma factor V activity ranges between 25 and 60 percent of normal, usually are asymptomatic, although an American registry recorded mild bleeding in 50 percent of the cases.¹⁰¹ According to a recent classification by the SSC of the ISTH, factor V deficiency may be classified as severe, moderate, and mild when factor V levels are undetectable, less than 10 percent, and 10 percent or greater, respectively.²⁴

Common manifestations include ecchymoses, epistaxis, gingival bleeding, hemorrhage following minor lacerations, and menorrhagia.^101,102,103 Severe deficiency typically presents at birth or in early childhood, but depending on factor levels some patients remain asymptomatic. Bleeding from other sites is less common, but instances of hemarthroses unrelated to trauma and intracerebral hemorrhage have been reported.¹⁰² Trauma, dental extractions, and surgery confer a high risk of excessive bleeding.

PPH occurs in more than 50 percent of pregnancies in women with factor V deficiency,^104,105 especially those with low factor V activity levels. Venous and arterial thromboses have been described in patients with factor V levels ranging between 2 and 14 percent of normal.¹⁰⁶ Factor V deficiency deprives activated protein C of one of its essential substrates, thereby downregulating the inhibitory function of the protein C system.

Development of a functional factor V inhibitor after receiving plasma transfusions was reported in only two patients with hereditary deficiency; the inhibitor disappeared in one patient, but a low titer of the inhibitor persisted in the other patient.^107,108 Factor V is indispensable for life, as was demonstrated by experimental knockout mice lacking the factor V gene, which die either in utero at embryonic day 9 or 10 because of defects in yolk-sac vasculature and somite formation; the remaining half develop to term but die of massive hemorrhage within hours of birth.¹⁰⁹ The expression of a minimal factor V activity because of the introduction of a liver-specific transgene, below the sensitivity threshold of the detection assay (<0.1 percent), leads to the survival of mice.¹¹⁰

THERAPY

Patients with epistaxis and gingival bleeding may respond to tranexamic acid (1 g four times daily), and local hemostatic measures may suffice for minor lacerations. Menorrhagia can also be managed directly using oral contraceptives, progestin-containing intrauterine devices, endometrial ablation, or hysterectomy. If these measures fail, severe spontaneous bleeding occurs, or surgery is performed, treatment option is limited to FFP replacement as no specific factor V concentrate is yet available on the market and factor V is not present in cryoprecipitate or PCCs. However, a new factor V concentrate has been developed for clinical use in patients deficient in factor V and preclinical studies are currently being performed for the orphan drug designation application to the European Medicine Agency (EMA) and the Food and Drug Administration (FDA) so as to make it available on the market as soon as possible.

DEFINITION AND HISTORY

Combined deficiency of factors V and VIII (F5F8D) is completely separate from factor V deficiency and factor VIII deficiency. The latter two are transmitted with different patterns of inheritance (autosomal recessive for factor V, X-linked for factor VIII) and involve proteins encoded by two different genes (F5 gene and F8 gene). F5F8D was first described in 1954¹¹¹; however, the molecular mechanism of the association of the combined factor deficiency was not understood until late 1990s,^112,113 when null mutations in the endoplasmic reticulum–Golgi intermediate compartment (ERGIC)-53 gene, now called the LMAN1 gene, were determined to be causative. In 2003, a second locus associated with the deficiency in approximately 15 percent of affected families with no mutation in LMAN1 was identified²⁶: the MCFD2 gene encoding for a cofactor for LMAN1. Even if a debate were carried out on the possible existence of other loci involved in the intracellular transport of factors V and VIII and associated with the disease, until now previous biochemical studies failed to identify additional components of the LMAN1–MCFD2 receptor complex,¹¹⁴ supporting the idea that F5F8D might be limited to the LMAN1 and MCFD2 genes.¹¹⁵ The disorder has been detected in many populations, but a relatively high frequency occurs among Tunisian and Middle Eastern Jews residing in Israel¹¹⁶ and among Iranians.¹¹⁷

PROTEIN

Factors V and VIII are essential coagulation factors that circulate in plasma as precursors. Upon limited proteolysis by thrombin or factor Xa and in concert with negatively charged phospholipid surfaces, factors VIIIa and Va exhibit profound cofactor activities for activation of factor X by factor IXa and for activation of prothrombin by factor Xa, respectively. Inactivation of factors Va and VIIIa is accomplished by activated protein C in the presence of protein S and phospholipids through several proteolytic cleavages at distinct sites. Factor V and factor VIII have similar domain organizations with partial homology (see “Factor V Deficiency” above).

The pathogenesis of combined deficiency of factors V and VIII puzzled investigators for more than 40 years. The enigma was resolved by the finding that the disease stems from the deficiency of either one of two interacting proteins, LMAN1 and MCFD2, which play a role in the intracellular transport of factors V and VIII. LMAN1 is a 53-kDa type 1 transmembrane nonglycosylated protein with homology to leguminous lectin proteins.¹¹⁸ It displays different oligomerization states—monomer, dimer, and hexamer—which have been implicated in its exit/retention within the endoplasmic reticulum (ER), and is thought to bind correctly folded glycosylated cargo proteins, including factors V and VIII in the ER, recruiting the cargo for package into coat protein complex II (COPII)–coated vesicles and to transport them first to the ERGIC and then to the Golgi. MCFD2 is a small (146 residues) soluble protein of 16 kDa with a signal sequence mediating translocation into the ER and two EF-hand motifs that may bind Ca²⁺ ions in the C-terminal region.¹¹⁹ MCFD2 forms a Ca²⁺-dependent 1:1 stoichiometric complex with LMAN1, which works as a cargo receptor for efficient ER–Golgi transfer of coagulation factors V and VIII during their secretion. Although several proteins have been identified as cargo of LMAN1 (factor V, factor VIII, cathepsin C, cathepsin Z, nicastrin, and α₁-antitrypsin),^{120,121,122,123} MCFD2 is only known to be required for transport of the blood coagulation factors, suggesting a possible role for MCFD2 as a specific recruitment factor for this subset of LMAN1 cargo proteins.¹²⁴ The three-dimensional structure of the complex between MCFD2 and the carbohydrate recognition domain (CRD) of LMAN1 was determined and a model of functional coordination between the two proteins was proposed: MCFD2 is converted into the active form upon complex formation with LMAN1, thereby becoming able to capture polypeptide segments of factors V and VIII. The coagulation factors bind the LMAN1 oligomer in the ER, but are released upon arrival in the acidic post-ER compartments because the sugar-binding of ERGIC-53 is pH-dependent.¹²⁵

GENETICS

Homozygosity mapping and positional cloning in nine unrelated Jewish families demonstrated that the LMAN1, composed of 13 exons, localizes on the long arm of chromosome 18.^112,113,126 Using a similar approach in other families with the combined factors V and VIII deficiency identified the short MCFD2, made up of four exons on the short arm of chromosome 2.²⁶ Thirty-four mutations identified in LMAN1 predicted either a truncated protein product or no protein at all, being more than 90 percent deletion/insertion, null, or splicing mutations. In contrast, of the 22 mutations identified in the MCFD2, 11 are missense and 11 are null mutations. Missense mutations are located at the EF-2 domains, giving rise to defective binding to LMAN1.¹²⁷ A distinct founder haplotype was found in patients belonging to six unrelated families of Tunisian-Jewish origin bearing a donor splice-site mutation in intron 9 of LMAN1.^112,127 All six families originated from an ancient Jewish community that has resided on the island of Djerba for more than 2 millennia. A survey of this community, which presently lives in Israel, disclosed that the mutation is prevalent at an allele frequency of 0.0107.¹²⁸ Another founder effect for a G insertion in exon 1 of LMAN1 was observed in eight unrelated Jewish families of Middle Eastern origin.^112,127 A Met to Thr mutation in LMAN1 has been detected in several unrelated Italian families, implying another founder effect.¹²⁷

CLINICAL MANIFESTATIONS

Symptoms of F5F8D are generally mild. Comparison of relatively large cohorts of F5F8D in India, Iran, and Israel indicates that bleeding from trauma/surgery is the most frequently reported clinical manifestation.^{116,117,129,130} This observation likely reflects the fact that often F5F8D is brought to the attention of physicians following excessive bleeding during and after trauma, surgery, and labor. Homozygous patients exhibit spontaneous and posttraumatic bleeding. Menorrhagia, epistaxis, easy bruising, hemarthrosis and gingival hemorrhage are commonly observed, and is unrelated to trauma in approximately 20 percent of cases.^117,131 Hematuria, GI, and spontaneous CNS bleeding is less common.¹¹⁷ There is insufficient data on the incidence of bleeding during pregnancy and PPH in women with combined factor V+VIII.

Heterozygotes exhibit slight but significantly reduced mean levels of factors V and VIII.¹¹⁶ In a literature survey of 161 heterozygotes, 22 reported having significant bleeding manifestations.¹³² However, no correlation between the factor V or factor VIII levels and bleeding tendency was noted.²⁵

LMAN1 gene knockout mice duplicate the F5F8-deficient phenotype in humans, albeit with a milder presentation, resulting from a lesser reduction in plasma levels of factors V and VIII.¹³³ The partial perinatal lethality observed in LMAN1-deficient mice on some genetic backgrounds was unexpected and has been explained as the result of a further drop in the level of LMAN1-dependent protein(s) below a critical threshold, or because of a strain-specific difference in another cargo receptor whose function overlaps with LMAN1.

THERAPY

Because of the mild-to-moderate bleeding symptoms, treatment is on demand, depending on the severity of bleeding. According to a recent result from the EN-RBD project, however, the level of F5F8 to ensure the absence of bleeding symptoms should be greater than 40 percent.²⁵ The recommended therapy includes FFP, which provides factor V, and factor VIII concentrate, which compensates for the shorter half-life of plasma factor VIII. An antifibrinolytic agent such as tranexamic acid or ε-aminocaproic acid can be helpful in patients exhibiting menorrhagia, epistaxis, or gingival bleeding. DDAVP (1-deamino-8-D-arginine vasopressin, desmopressin) could be administered for less-severe bleeding. Patients with severe bleeding episodes or patients undergoing surgical procedures, including dental extractions, should receive FFP as replacement for factor V and cryoprecipitate or factor VIII concentrate as a source of factor VIII. DDAVP can be used to increase factor VIII level, but this treatment sometimes fails.¹³⁴

DEFINITION AND HISTORY

Factor VII was first identified as serum prothrombin conversion accelerator or proconvertin and its hereditary deficiency described by Alexander and colleagues in 1951.¹³⁵ Among the rare clotting factor deficiencies, the relative frequency of factor VII deficiency is high (see Table 124–1).^101,102 A presumptive diagnosis can be easily made because, except for very rare cases of factor X deficiency only affecting the TF pathway of coagulation (see “Laboratory Diagnosis” below), factor VII deficiency is the only coagulation disorder that produces a prolonged PT and a normal aPTT.

PROTEIN

Human factor VII is a single-chain glycoprotein (Mr ~50,000) that is secreted from the liver parenchymal cells as a zymogen. The mature protein consists of 406 amino acids organized in three main domains: a Gla domain at the N-terminus containing 10 Gla residues, an epidermal growth factor (EGF) domain in the center, and a serine protease domain at the C-terminus.¹³⁶ Factor VII zymogen circulates in blood at an extremely low concentration (~500 ng/mL)¹³⁷ and has the shortest half-life of all coagulation factors (4 to 6 hours; see Table 124–3). Factor VII is converted to the activated form, factor VIIa, by cleavage of an Arg152-Ile153 bond, resulting in a two-chain molecule held together by a disulfide bond. Factor VII can be converted to factor VIIa by factor Xa,¹³⁸ factor IXa,¹³⁹ factor XIIa,¹⁴⁰ thrombin,¹³⁸ and factor VIIa in the presence of TF, in an autoactivation reaction.¹⁴¹ Binding of factor VII to TF strikingly enhances these reactions.^{142,143,144,145,146}

The initial generation of thrombin that heralds blood coagulation occurs when blood is exposed to TF present in the subendothelium in tissues, or on the surface of stimulated monocytes or microparticles. The exposed TF forms a complex with circulating factor VIIa and supports the initiation of coagulation by converting factors IX and X into their active forms (factor IXa and factor Xa).^147,148 Following the generation of trace amounts of factor VIIa there is a feedback amplification of the signal, as factor VII bound to TF: factor VIIa is both self-activated, and activated by factor IXa and factor Xa. Hence, the TF–factor VIIa complex has two roles: to increase the conversion of factor VII to factor VIIa and to increase the proteolytic activity of factor VIIa toward its substrates, factors IX and X. Factors IXa and Xa may remain associated with cells that display the TF, or disseminate in the blood and bind to the surface of activated platelets, which form the initial platelet plug.¹⁴⁹

GENETICS

The factor VII gene (F7) spans approximately 12.8 kb¹⁵⁰ and is located on chromosome 13q34,^30,151 2.8 kb upstream from the factor X gene.¹⁵² The gene contains a prepro leader sequence and eight exons that encode the mature protein. Promoter and silencer elements of the 5′ flanking region have been characterized.^153,154 The disorder manifests in homozygotes or compound heterozygotes, some of which are also homozygotes for polymorphisms associated with reduced factor VII levels.^155,156

More than 240 mutations have been reported (see http://www.isth.org/?MutationsRareBleedin and Ref. 12). The mutations are distributed throughout the gene, and most are missense mutations (62.2 percent); other type of mutations are equally present (ranging from approximately 6.2 percent of mutations in 3′-5′ untranslated region [UTR] to 12.3 percent of deletions/insertions [del/ins]). Most mutations causing factor VII deficiency have been observed in individual patients. However, one missense mutation (Ala244Val) was detected in 102 (84 percent) of 121 independent mutant alleles discerned in 88 unrelated patients in Israel.¹⁵⁷ Most subjects were of Iranian and Moroccan-Jewish origin and shared an identical haplotype, consistent with a founder effect. In the general Iranian-Jewish and Moroccan-Jewish populations, the prevalence of the Ala244Val allele are 0.023 and 0.025, respectively.¹⁵⁶ Several additional clusters of patients with a specific mutation were reported: (1) Ala294Val, with or without a deletion of nt C, at position 11128, prevails in patients from Poland and Germany but has also been identified in other Europeans^158,159; (2) 12 unrelated families from Norway who carry Gln100Arg¹⁶⁰; (3) IVS75G>A, which was detected in six unrelated patients from the Lazio region in Italy, all of whom bear the same haplotype, suggesting a founder effect¹⁶¹; and (4) Gly331Ser, which was identified in 10 Italian and four German patients on one haplotype.¹⁶² The widely distributed and common Arg304Gln mutation probably is a recurrent mutation.¹⁶³

Three polymorphisms in the factor VII gene are also associated with reduced plasma levels of the factor. The first polymorphism, an Arg353Gln substitution, results in impaired secretion of factor VII from cells¹⁶⁴ and gives rise to a 20- to 25-percent decrease in plasma factor VII level in heterozygotes and a 40- to 50-percent decrease in homozygotes.^165,166 The second polymorphism associated with a diminished factor VII level is a decanucleotide insertion upstream from the 5′ end of the gene at −323, which confers a 33 percent decrease in the promoter activity.¹⁵⁴ A third polymorphism associated with factor VII level is a hypervariable region 4 polymorphism (HVR4) in intron 7.¹⁶⁷ The variable number of tandem repeats (five to eight copies of 37 bp) apparently influences the splicing efficiency. The effect of the variable repeats on factor VII level is less conspicuous than the decanucleotide insertion at the promoter region and the Arg353Gln polymorphism.

CLINICAL FEATURES

Bleeding manifestations occur in homozygotes and in compound heterozygotes for factor VII deficiency. However, a typical feature of this disease is its clinical heterogeneity: some patients do not bleed at all after major hemostatic challenge, while others with similar levels report frequent bleeding episodes. Life- or limb-endangering bleeding manifestations are relatively rare, the most frequent symptoms being epistaxis and menorrhagia. However, CNS bleeding was also reported to have high incidence (16 percent) in a series of 75 infants,¹⁶⁸ the authors concluding that the greatest risk factor for this development was trauma related to the birth process. In addition, heterozygotes who have partial factor VII deficiency may present with bleeding; a recent survey of 499 heterozygotes revealed that 19 percent reported pathologic bleeding.¹⁵⁵ Dental extractions, tonsillectomy, and surgical procedures involving the urogenital tracts frequently are accompanied by bleeding when no prior therapy is instituted. Normal pregnancy is accompanied by increased concentrations of fibrinogen and factor VII, nonetheless, cases of miscarriages and PPH, albeit at relatively low rates, have been observed in patients with factor VII deficiency.^169,170

Thrombotic episodes have also been reported in 3 to 4 percent of patients with factor VII deficiency, particularly in the presence of surgery and replacement treatment, but spontaneous thrombosis may also occur. A survey of 514 cases with severe or partial factor VII deficiency recorded seven patients with venous thrombosis and one patient with arterial thrombosis.¹⁷¹ Most of the cases presented with associated risk factors, mainly surgery, prolonged immobilization, and treatment with PCCs.¹⁷²

When factor VII is completely lacking, as in knockout mice, there is no embryonic lethality; however, fatal hemorrhage occurs perinatally.^173,174

THERAPY

As for all the other congenital bleeding disorders, replacement therapy is essential in patients who present with severe hemorrhage, such as hemarthrosis or intracerebral bleeding, surgical hemostasis, and for individuals with a bleeding history. Factor replacement therapy may also be used for prophylaxis in children with severe factor VII deficiency.¹⁷⁵ The EN-RBD study suggests a trough factor VII activity level of 25 percent is needed for patients to remain asymptomatic²⁵; prophylactic treatment is usually recommended for patients with major bleeding episodes such as CNS and gastrointestinal bleeding and hemarthroses. A number of replacement therapeutic options have been administered to patients with factor VII deficiency, including FFP, PCCs, plasma-derived factor VII concentrates (volume overload should be expected if plasma is used as the replacement material) and rFVIIa. rFVIIa has been successful in managing patients with hemarthroses and during surgery,^176,177 and is the only treatment supported by substantial literature^12,176 (see Table 124–3). PCCs containing activated clotting factors can be used, but they confer a risk of thrombosis,¹⁷⁵ while specific factor VII concentrates have been used successfully in series of patients.¹⁷⁵ The treatment of factor VII deficiency is sometimes challenging, because of the short in vivo half-life of factor VII, its low recovery, and its rapid clearance, which is more evident in children.¹⁷⁸ Because of these features, replacement regimens require frequent dosing. A significant rise in the factor VII level is observed during pregnancy in women with mild/moderate forms of factor VII deficiency (heterozygotes), but not in women with severe deficiency.^{179,180,181,182} Therefore, in women with mild/moderate deficiency, replacement therapy may not be required during labor and delivery, while it would be required in women with low factor VII coagulant activity levels or a positive bleeding history who are more likely to be at risk of PPH.^{181,183,184,185} A recent review of the literature noted that hemorrhage rates were equivalent in women with and without prophylaxis, thus concluding that use of hemostatic prophylaxis should not be considered mandatory, but as part of an individualized discussion taking into consideration response to previous hemostatic challenges and mode of delivery.¹⁸⁶ Replacement therapy is unnecessary for minor bleeding episodes. Local hemostasis for skin lacerations and administration of an antifibrinolytic agent for menorrhagia, epistaxis, and gingival hemorrhage usually are sufficient to arrest bleeding. Asymptomatic patients undergoing minimally invasive surgery, such as dental procedures, can be successfully treated with tranexamic acid given both orally or intravenously at the usual dosages.

DEFINITION AND HISTORY

Inherited factor X deficiency was identified by two independent groups, each of which described a patient with a bleeding diathesis that could not be attributed to deficiencies in other known coagulation factors. The factor in both patients was subsequently named factor X.^187,188,189

PROTEIN

Factor X is mainly synthesized by the liver as a 488-amino-acid protein and circulates in plasma at a concentration of 8 to 10 mcg/mL.¹⁹⁰ Its primary structure is homologous to that of other vitamin K–dependent proteins, such as prothrombin, factor VII, factor IX, protein C, and protein S.¹⁹¹ The first 40-amino-acid residues, the prepropeptide, contain the hydrophobic signal sequences targeting the protein for secretion.¹⁹² The Gla domain forms the N-terminus of the mature protein and contains 11 Gla residues that are responsible for calcium and phospholipid binding.¹⁹³ Adjacent to the Gla domain is a short aromatic amino acid stack of predominantly hydrophobic amino acids, followed by the EGF domain, believed to mediate protein–protein interactions. The heavily glycosylated 52-amino-acid activation peptide of factor X separates the EGF domain from the C-terminal catalytic domain. Factor X undergoes proteolytic processing in the ER so that circulating factor is a two-chain, disulfide-linked protein consisting of a 17-kDa light chain made up of the Gla and EGF domains and a 40-kDa heavy chain made up of the activation and catalytic domains.¹⁹⁴ The heavy chain contains the activation peptide (residues 143 to 195) and the catalytic serine protease domain, structurally homologous to that of other coagulation serine proteases containing the catalytic site formed by residues His236, Asp282, and Ser379. The 52-residue activation peptide is released after factor X is converted to its active form factor Xa by the cleavage between residues Arg194 and Ile195. Physiologically factor X is activated by TF/factor VIIa (extrinsic pathway) and factor IXa/factor VIIIa (intrinsic pathway),¹⁹⁵ but it can also be activated in vitro by Russel viper venom.¹⁹⁶ In turn, factor Xa catalyses thrombin formation. In presence of factor Va, Ca²⁺, and phospholipid membrane, factor Xa forms the prothrombinase complex that accelerates to 280,000-fold thrombin formation.¹⁹⁷

GENETICS

The factor X gene spans approximately 25 kb and is made up of eight exons.¹⁹² The factor X gene shows significant homology with the genes of other vitamin K–dependent serine proteases, which suggests all of these multidomain genes evolved from a common ancestral gene.¹⁹⁸

The currently described 105 mutations that cause factor X deficiency include large deletions, small frameshift deletions, nonsense mutation, and missense mutations; the missense mutations group is the largest (80 percent), while mutations in the 3′- and 5′-UTRs are completely absent (see http://www.isth.org/?MutationsRareBleedin and Ref. 13). Activation through the TF pathway may be affected when the mutations are located, for example, in the Gla domain, as in Glu7Gly (St. Louis II) or Glu19Ala.^19,199,200 Activation through factor IXa is affected by, for example, Thr318Met (Roma).²⁰¹ Activation of factor X through Russell viper venom is almost intact in the Pro343Ser (Friuli) mutation.²⁰² Two interesting clusters of unrelated families were described in Algeria, with Phe31Ser mutation,²⁰³ and in the border region between Turkey and Iran, with the Gly222Asp mutation.²⁰⁴

CLINICAL MANIFESTATIONS

The clinical manifestations of factor X deficiency are related to the functional levels of the protein. According to the recent results of the EN-RBD study, strong associations between clinical bleeding severity and coagulation factor activity level were shown in factor X deficiency; consequently, patients with factor X activity levels less than 10 percent of normal have a higher occurrence of spontaneous major bleeding.²⁵ Bleeding occurs primarily into joints and soft tissues, from the umbilical cord and mucous membranes.²⁰⁵ The bleeding tendency may appear at any age, although patients with factor X activity less than 2 percent present early in life with, for instance, umbilical-stump or CNS hemorrhage.^13,205,206

In an analysis of 102 patients from Europe and Latin America, three mutations were associated with intracerebral hemorrhage (Gly380Arg, IVS7–1G>A, and Tyr163delAT) and Gly⁻²⁰Arg mutation was associated with severe hemarthrosis.²⁰⁵ The most common bleeding symptom reported at all levels of severity of the deficiency is epistaxis. Patients with severe deficiencies commonly experience hemarthrosis and hematomas, but gastrointestinal and umbilical cord bleeding, hematuria, and CNS bleeding also occur. In a small group of patients with factor X deficiency, one-third of heterozygous patients who had dental extraction, surgery, or delivery without prophylactic replacement therapy showed postoperative bleeding which required treatment.¹⁶ Menorrhagia is a common symptom affecting women with all degrees of severity and PPH was also reported in 4 of 14 pregnancies.²⁰⁷ Two successful pregnancies out of four pregnancies were reported in a woman only when receiving regular prophylaxis; the other two pregnancies, without regular prophylaxis, resulted in preterm labor (both babies died in the neonatal period).²⁰⁸ Other case reports, however, have described successful term pregnancies in women with severe factor X deficiency without antenatal prophylaxis.^207,209 Knockout mice show partial embryonic lethality (E11.5–E12.5); complete absence of factor X is incompatible with murine survival to adulthood, but minimal factor X activity (range: 1 to 3 percent) is sufficient to rescue the lethal phenotype.^210,211,212

THERAPY

Therapy for inherited factor X deficiency usually is administration of heated and solvent-detergent–treated PCCs containing factor X, in addition to factors II, VII, and IX. Use of these concentrates carries a low risk of transmission of bloodborne viruses. However, a risk of thrombosis, including venous thromboembolism, diffuse intravascular coagulation, and myocardial infarction has been reported.

For soft-tissue, mucous membrane, and joint hemorrhage, the aim of treatment should be maintaining a factor X level that is at least 10 to 20 percent of normal. For more serious hemorrhage, a factor X level that is greater than 40 percent of normal should be the goal.²⁵ In patients with particularly severe bleeding manifestations, prophylactic therapy should be considered. FFP also can be used to treat patients with factor X deficiency, however, the administration of FFP can be associated with complications, particularly in children and elderly patients with cardiac disease, because of fluid overload.³⁸ The arrival on the market of a new freeze-dried human factor X concentrate has facilitated prophylaxis in patients with factor X deficiency (Factor X P Behring)²¹³; however, factor X P Behring also contains factor IX, although in a known amount. A clinical trial investigating the pharmacokinetics of a new high-purity factor X concentrate has been completed (ClinicalTrials.gov identifier: 00930176).

DEFINITION AND HISTORY

Factor XI deficiency initially was described as a “new hemophilia” in two sisters and their maternal uncle by Rosenthal and colleagues in 1953.²¹⁴ Because it manifested in both sexes—two sisters and their maternal uncle—the clinical features were not consistent with hemophilia A or B and was called hemophilia C.²¹⁵ The deficiency was erroneously thought to be transmitted as an autosomal dominant disorder with variable expressivity. Later studies clearly established that, in most cases, the mode of transmission of factor XI deficiency is autosomal recessive.²¹⁵ Affected subjects have been described in most populations, but the disorder is common in Jews, particularly those of Ashkenazi origin.²¹⁶

Factor XI deficiency as a result of a dysfunctional protein is rare, as only a few patients with deficiency of factor XI activity and seemingly normal antigen levels have been described thus far.^{217,218,219,220}

PROTEIN

Factor XI is a glycoprotein that consists of two identical 80-kDa polypeptide chains linked by a disulfide bond.²²¹ Each subunit contains 607 amino acids with a serine protease domain at the C-terminus and four tandem repeats of 90 or 91 amino acids, designated “apple domains,” at the N-terminus. The described crystal structure of factor XI dimer²²² defined the interface of the monomers in apple 4 domains in which three residues—Leu284, Ile290, and Tyr329—are essential for noncovalent binding between the monomers. This binding enables the formation of a disulfide bond between Cys321 residues in the fourth apple domain of each monomer.^223,224

Although factor XI is synthesized by the liver, very low levels of factor XI transcript can also be detected in megakaryocytes and platelets, renal tubules, and pancreatic islet cells.²²⁵ Factor XI circulates in blood as an equimolar complex with high-molecular-weight kininogen (HK)²²⁶ at a concentration of 3 to 7 mcg/mL. The importance of the factor XI–HK interaction is not fully understood. Activation of factor XI involves cleavage of an Arg369-Ile370 bond, yielding a heavy chain containing the four apple domains linked by a disulfide bond to a light chain that contains the catalytic domain.²²¹ The physiologic activator of factor XI during hemostasis has long been debated. The original scheme of the coagulation cascade—according to which factor XI is activated by factor XIIa through the intrinsic pathway (the “contact phase”)—was challenged by the observation that deficiencies of factor XII as well as of the other contact factors (HK and prekallikrein) are not associated with a bleeding diathesis.¹⁴

The major activator of factor XI in vivo is thrombin.^227,228 Factor XI binds through its apple 3 domain to lipid rafts on platelets containing glycoprotein Ib–IX–V complex. This glycoprotein complex also binds thrombin; thus, both substrate and enzyme are colocalized at the same site.²²⁹ Factor XI activation also can occur on the fibrin surface after a clot forms.²³⁰ Factor XIa, once generated, activates factor IX by limited proteolysis of two peptide bonds in the presence of calcium ions.²³¹ The presence of factor XI contributes to the activation of thrombin-activatable fibrinolysis inhibitor (TAFI), that once activated, removes terminal lysine residues from fibrin, which impairs binding of certain forms of plasminogen to fibrin and disrupts tissue plasminogen activator-induced plasmin generation in the blood clot.²³² Large amounts of thrombin are necessary for TAFI activation, but the reaction is substantially augmented when thrombin is bound to thrombomodulin.²³³ It follows that impaired generation of thrombin, for example, in inherited deficiency of factor VIII, IX, or XI, not only delays clot formation but also enhances premature lysis of clots.²³⁴ These data fit well with clinical observations in factor XI–deficient patients who are particularly susceptible to bleeding following injury at sites exhibiting local fibrinolytic activity.¹⁸

GENETICS

The 23-kb gene encoding for factor XI consists of 15 exons and 14 introns and is located on chromosome 4q34–35.^235,236

Three mutations, designated types I, II, and III, were first described in six Ashkenazi-Jewish patients with severe factor XI deficiency.²³⁷ The types II and III, a change in exon 5 at Glu117 leading to a stop codon and a change in exon 9 that results in a substitution of Phe283 by Leu, respectively, account for 95 percent of cases in Ashkenazi Jewish patients.²³⁷ Most patients with factor XI deficiency are Jewish.^18,216,238 A recent study indicated that both type II and type III mutations are also prevalent in the Italian population, although at a much lower rate.²³⁹ In patients belonging to different ethnic groups, a significantly higher level of allelic heterogeneity has been reported. Remarkable exceptions are represented by some “closed populations” harboring mutations compatible with a founder effect: Cys38Arg in French Basques,²⁴⁰ Gln88Stop in French families from Nantes,²¹⁹ Cys128Stop in Britons,⁶ Ile436Lys in Northeastern Italy,²⁴¹ and Q263X in Korean patients.²⁴²

As of this writing, 220 mutations have been reported in non-Jewish and Jewish patients of various origins, including 154 missense, 23 nonsense, and 23 del/ins mutations, with the remaining being splice site (18 mutations) and 5′- and 3′-UTR (2 mutations). Inheritance of factor XI deficiency is usually autosomal recessive, as other RBDs, although some missense mutations exert a dominant negative effect through heterodimer formation between the mutant and wild-type polypeptides, resulting in a pattern of dominant transmission.²⁴³

CLINICAL FEATURES

Factor XI deficiency is the only RBD in which the EN-RBD study showed no association between clinical bleeding severity and coagulation factor activity level.²⁵ This disorder manifests as a mild to moderate bleeding manifestations and most bleeding episodes of patients with severe FXI deficiency are injury-related. Most bleeding manifestations of patients with severe FXI deficiency are injury-related. Some patients with severe factor XI deficiency may not bleed at all following trauma.²⁴⁴ The phenotype of bleeding is not correlated with the genotype but frequently with site of injury.^18,238,244 Surgical procedures involving tissues with high fibrinolytic activity (urinary tract, tonsils, nose, tooth sockets) frequently are associated with excessive bleeding in patients with severe factor XI deficiency, irrespective of the genotype.²⁴⁵

Most women with severe factor XI deficiency are asymptomatic or minimally so. During 93 deliveries (85 vaginal, eight cesarean), 43 of 62 women did not experience PPH despite no prophylactic treatment,⁴¹ which was confirmed not to be mandatory for these women, by a subsequent study of 33 women who had approximately 70 uneventful pregnancies out of 105 pregnancies. In the subsequent study, only three women had factor XI activity less than 15 IU/dL, and none of them had PPH.⁴²

Whether heterozygotes exhibit a bleeding tendency (except for those bearing mutations causing a dominant negative effect) is controversial because there are reports on both heterozygotes with no bleeding complications following a variety of surgical procedures, and reports that 20 to 48 percent of heterozygotes do bleed.^{215,238,243,246,247}

In regard to venous thrombosis, it was reported that in five patients, two developed pulmonary embolism following infusion of factor XI concentrate^248,249,250 and a third patient developed thrombus in the inferior vena cava following cryptococcal infection. In contrast, severe factor XI deficiency was shown to confer protection against ischemic stroke.²⁵¹

Mice homozygous for a knockout factor XI allele show a tendency for slightly prolonged tail transection bleeding times and are protected from vessel-occluding fibrin formation after transient ischemic brain injury.^252,253

THERAPY

Available treatments for patients with the severe form of factor XI deficiency are FFP and factor XI concentrate. Factor XI concentrates currently available (Bio Products Laboratory, United Kingdom, and LFB Biomedicaments, France) are associated with thrombosis even after adding heparin to the antithrombin in the Bio Products Laboratory product, and antithrombin and heparin to the C1 esterase in the LFB Biomedicaments product.^14,254 Therefore, it is advisable to monitor patients for clinical and laboratory signs of coagulation activation, in particular in elderly patients, in those with cardiovasclar disease and in those undergoing surgery with thrombotic potential, especially when factor XI concentrate²⁵⁵ or rFVIIa is considered.^256,257 In addition, the presence of an inhibitor, particularly in patients with less than 1 percent of factor XI activity and previously exposed to plasma, factor XI concentrates, or immunoglobulins, should be evaluated. Low doses of rFVIIa along with tranexamic acid seem promising in the treatment of patients with inhibitors. When procedures are planned at tissues exhibiting fibrinolytic activity, which is associated with higher risk of bleeding in comparison to sites without fibrinolytic activity, the use of antifibrinolytic agents alone or in combination with other treatments is recommended. Patients undergoing dental extractions do not require replacement therapy. No plasma replacement therapy is necessary during or after labor unless excessive bleeding occurs. Patients with factor XI deficiency and inhibitor do not bleed spontaneously. Acquired inhibitors that neutralize the activity of factor XI have been described in patients with severe factor XI deficiency and baseline activity of less than 1 IU/dL after being exposed to plasma,²⁵⁸ after injections of Rh immunoglobulin and without previous exposure to blood products,²⁵⁹ or after exposure to factor XI concentrates. Use of PCCs^262,263 and rFVIIa¹⁷⁸ have been successful for major surgical procedures, and an in vitro study revealed that abnormal thrombin generation in the plasma of patients with an inhibitor was corrected by adding moderate amounts of rFVIIa.²⁶²

DEFINITION AND HISTORY

The first clinical report of factor XIII deficiency was in 1960²⁶⁵; since then, more than 500 cases of factor XIII deficiency have been identified worldwide, with an incidence of one individual in 1 to 3 million population.^22,264 Congenital factor XIII deficiency is characterized by severe delayed spontaneous bleeding and recurrent abortion with normal coagulation screening tests.

PROTEIN

Factor XIII (fibrin-stabilizing factor) is a plasma transglutaminase that crosslinks γ-glutamyl–ε-lysine residues of fibrinogen chains, thereby stabilizing the fibrin clot. Plasma factor XIII is an Mr 340,000 heterotetramer composed of two catalytic A subunits and two carrier B subunits linked by noncovalent bonds. The average concentration of the A₂B₂ tetramer in plasma is approximately 22 mcg/mL, and its half-life is 9 to 14 days.²⁶⁵ Intracellularly, factor XIII is found as a homodimer composed of two A subunits (A₂).^266,267 Factor XIII-A subunit is mainly synthesized in macrophages and megakaryocytes.^266,267 Because factor XIII-A subunit lacks a signal sequence, it cannot be released by the classic secretory pathway through the Golgi apparatus. Conceivably, factor XIII-A subunit is released into the circulation from cells as a consequence of cell injury.²⁶⁸ Structurally, each A monomeric subunit (Mr ~82,000) is composed of an activation peptide, that is removed by thrombin cleavage of an Arg37-Gly38 bond in the presence of calcium ions, and four distinct domains: β-sandwich, central core, barrel 1, and barrel 2 regions. The central core domain contains a catalytic triad (common to the transglutaminase family) formed through hydrogen bond interactions between Cys314, His373, and Asp396.^269,270,271 It is structurally homologous with the α chain of tissue transglutaminase,²⁷² the α chain of keratinocyte transglutaminase,²⁷³ and band 4.2 of erythrocytes,²⁷⁴ although the latter lacks transglutaminase activity.

The site of synthesis for factor XIII-B subunit has been suggested to be the liver.²⁷⁵ The B subunit (Mr 76,500) is composed of 10 tandem repeats of complement control protein (CCP) modules designated as Sushi domains, which are also observed in proteins of the complement system.^276,277 The two B subunits of factor XIII function as carrier proteins for the A subunits,^278,279 stabilizing them in the circulation and regulating the calcium-dependent activation of factor XIII.

On activation by thrombin and Ca²⁺ the A and B subunits dissociate. Proteolytic activation by thrombin involves the cleavage of a N-terminal 37-residue activation peptide. The cleavage and the calcium binding both serve to induce structural changes that open up the catalytic triad to substrate access.²⁸⁰ This process is accelerated by fibrin.^281,282,283 The clot stabilizing effect of factor XIII is achieved by the crosslinking of fibrinogen chains, between the γ-carbonyl group of glutamine and the ε-amino group of lysine. In fibrin, this amide bond is located between Aα-chain sequences and between γ-chain sequences^{284,285,286,287,288}; factor XIIIA also crosslinks α₂ antiplasmin to the α-chain fibrin,²⁸⁹ thereby increasing the resistance of fibrin to plasmin degradation, and crosslinks fibronectin to the α-chain of fibrin,²⁹⁰ thereby affecting the mechanical properties of the clot and increasing cell adhesion.²⁹¹

In addition to fibrinogen and α₂-antiplasmin, factor XIII has many other substrates, including fibronectin, vitronectin, collagen, factor V, von Willebrand factor, α₂-antiplasmin, actin, myosin, vinculin, thrombospondin, plasminogen-activating inhibitor (PAI), TAFI 2, and AT1 receptor dimers of monocytes, implicating multiple and different roles for factor XIII in various systems other than coagulation.²⁹²

GENETICS

The gene for the factor XIII A-subunit is located on chromosome 6p24-p25.^293,294 It spans more than 170 kb and is composed of 15 exons.²⁹⁵ The B-subunit gene is located on chromosome 1q31-q32.1.^296,297 The gene for the B subunit spans 28 kb and is composed of 12 exons.²⁹⁷

One hundred and twenty-one mutations causing factor XIII A-subunit deficiency have been reported as of this writing, of which only one maps to the promoter region, 57 are missense, 11 are nonsense, 17 are splice-site, and 35 are del/ins mutants (http://www.isth.org/?MutationsRareBleedin and Ref. 298). A homozygous four-bases insertion in exon 14 (c.2116insAAGA) introducing a frameshift that after seven altered amino acids results a stop codon and a protein with AQ3 truncated second β-barrel domain (p.Pro675TyrfsX7)²⁹⁹ has been reported to cause an extremely rare type II variant. The mutant protein lost its activity, but the plasma factor XIII antigen level was at the lower limit of the reference interval. This finding suggests that the C-terminal part of β-barrel 2 is essential for the expression of factor XIII activity.

Splice-site mutation in intron 5 (IVS5–1 G>A) seems to be the most common mutation as it has already been reported in six unrelated families from six different European countries, whereas the Arg660Pro was found in Palestinian Arabs, consistently with founder effects.^264,300 It is likely that the Arg661stop mutation in Finnish patients and the Arg77Cys mutation in Swiss patients are also a result of founder effects, although both are at CpG dinucleotides and therefore can be considered recurrent mutations.^264,301,302 Another mutation, Ser295Arg, was identified in six Pakistani families and may also stem from a common founder, but this remains to be established.³⁰³ Six nonsynonymous/coding polymorphisms in the factor XIIIA1 (F13A1) gene,²⁵ Val34Leu in exon 2, Tyr204Phe in exon 5, Pro(CCA)331(CCC)Pro in exon 8, Glu(GAA)567Glu(GAG) and Pro564Leu in exon 12, and Val650Ile and Glu651Gln in exon 14 have been analyzed in an association study. The study showed that only the Val34Leu is a true functional polymorphism and the rest are in linkage disequilibrium with this polymorphism. In this study, only haplotypes containing the “34L” allele affected factor XIII function.^298,304 However, a larger number of synonymous/noncoding polymorphisms (>500) are known for the F13A1 gene.³⁰⁴ Only 16 different mutations have been reported so far for the FXIIIB gene.²⁹⁸

CLINICAL MANIFESTATIONS

Factor XIII deficiency causes formation of blood clots that are unstable and susceptible to fibrinolytic degradation by plasmin. As a result, affected individuals have an increased tendency to bleed and rebleeding. Delayed umbilical cord bleeding reported in 80 percent of patients with factor XIII deficiency can be considered as diagnostic symptom of the deficiency. CNS bleeding is reported in approximately 30 percent of cases,^305,306 making primary prophylaxis mandatory in patients affected with severe factor XIII deficiency. Ecchymoses, intramuscular and subcutaneous hematomas, oral cavity, mouth and gingival bleeding, and prolonged bleeding following trauma are also characteristic symptoms.³⁰⁵

Delayed wound healing occurs in approximately 15 percent of patients deficient in factor XIII. The exact mechanism by which factor XIII, or factor XIIIa, exerts its beneficial effect on wound healing is unknown. A proangiogenic effect of factor XIIIa was described, suggesting that decreased vascularization of wounds results in improper repair.³⁰⁶

In a review of the literature on 121 women with factor XIII deficiency, menorrhagia and ovulation bleeding were found to be common gynecologic problems, affecting 26 and 8 percent of women, respectively.³⁰⁷ Of 192 pregnancies, 127 (66 percent) resulted in a miscarriage and 65 (34 percent) reached viability stage, whereas of 136 pregnancies without prophylactic therapy, 124 (91 percent) resulted in a miscarriage and 12 (9 percent) progressed to viability stage. In affected women, formation of the cytotrophoblastic shell is impaired.³⁰⁸ Conceivably, factor XIII A-subunit deficiency at the implantation site abrogates fibrin/fibronectin crosslinking, which is essential for attachment of the placenta to the uterus.³⁰⁹

Placental abruption, preterm delivery, and PPH could be also problem if not adequately treated.³⁰⁹

No large clinical reports on heterozygous patients with factor XIII deficiency are available, thus not allowing to draw evidence-based conclusion on the prevalence of clinical symptoms in this group of patients. Recently, a subset of 28 heterozygotes for factor XIII deficiency among 350 carriers of an autosomal recessive inherited coagulation disorders show an association with prolonged or massive bleeding after minor trauma.³¹⁰ However, these data need to be confirmed in other cohorts of patients.

Factor XIII A-subunit knockout mice manifest no excess embryonic lethality or bleeding into the thoracic cavity, peritoneum, or skin, compatible with survival to adulthood. However, the survival rate of knockout males was markedly lower than that of the wild-types.³¹¹ Female factor XIII knockout mice show intrauterine bleeding during pregnancy, similar to women with severe factor XIII A-subunit deficiency who experience the same problem, as well as recurrent abortions. Factor XIIIB knockout mice show a prolonged bleeding time at variance with patients with a complete factor XIII B-subunit deficiency, who report only mild bleeding symptoms and display normal bleeding times.^312,313

THERAPY

According to the EN-RBD results, blood levels of factor XIII that are 30 percent of normal are necessary to assure an asymptomatic state. This goal may be reached via a number of options. Many case reports show improved bleeding symptoms in patients on prophylactic therapy.³¹⁴ Plasma replacement therapy is highly satisfactory because of the long half-life of factor XIII (9 to 12 days). Plasma-derived, virus-inactivated concentrates of factor XIII are available³¹⁵ and are the treatment of choice. The development of adverse events after treatment is rare. The most dreaded adverse event is the development of inhibitors, although its incidence is rare.³¹⁶ A RBD registry created in North America discovered that 3 percent of factor XIII–deficient patients who received FFP or factor XIII concentrate treatment developed inhibitors.¹⁰² A new rFXIIIA₂ concentrate has become available and a phase III clinical trial (ClinicalTrials.gov identifier: 00713648) has been completed, establishing that rFXIII is safe and effective in preventing bleeding episodes in patients with congenital factor XIII A-subunit deficiency. The rFXIII was recently approved for the treatment of factor XIIIA deficiency in Australia, Canada, the European Union, Switzerland, and the United States.³¹⁷

Acquired coagulation factor deficiencies may occur in patients with liver disease, amyloidosis (specifically factor X),^318,319 autoimmune disorders, patients on oral anticoagulant therapy, and, rarely, in patients who develop nonneutralizing antibodies that remove the protein from the circulation. Such antibodies directed against prothrombin³²⁰ and factor VII³²¹ have been described in patients with lupus anticoagulants. Rare instances of an acquired factor V inhibitor as a result of exposure to bovine thrombin preparations and drugs, or because of an unknown cause, should also be considered.³²² Acquired isolated factor X deficiency with severe bleeding manifestations occurs rarely because of the formation of specific antibodies with no underlying autoimmune disorder³²³ or in association with upper respiratory tract infections, burns, and leprosy.^324,325,326 Inhibitors of factor X with no known precipitating factors have also been described.³²⁷ Acquired factor XIII deficiency with significant reductions in factor XIII levels (down to as low as 20 percent of normal) as a result of decreased synthesis or increased consumption, has been reported in several medical conditions, including pulmonary embolism, Crohn disease, ulcerative colitis, Henoch-Schönlein purpura, liver cirrhosis, and sepsis. There are several case reports of an autoimmune bleeding disorder, designated as autoimmune/acquired hemorrhaphilia, being caused by anti–factor XIII inhibitors.³²⁸ The anti–factor XIII inhibitors tend to be more severe than regular hemorrhagic-acquired factor XIII deficiency and requires both immunosuppressive therapy to eradicate autoantibodies and factor XIII replacement therapy to stop the bleeding.³²⁹