VHrare study: Prevalence, clinical features and management of severe rare bleeding disorders in a large cohort

Abstract Introduction: Rare bleeding disorders (RBD) constitute 5% of total hereditary bleeding disorders, although the number could be higher, due to the presence of undiagnosed asymptomatic patients. The objective of this study was to analyze the prevalence and characteristics of patients with severe RBDs in our area. Material and methods: We analyzed the patients with RBD followed at a tertiary‐level hospital between January 2014 and December 2021. Results: A total of 101 patients were analyzed, with a median age at diagnosis of 27.67 years (range 0–89), of which 52.47% were male. The most frequent RBD in our population was FVII deficiency. Regarding the diagnostic reason, the most frequent cause was a preoperative test and only 14.8% reported bleeding symptoms at the time of diagnosis. A genetic study was carried out in 63.36% of patients and the most frequent mutation type found was finding a missense mutation. Conclusions: The distribution of RBDs in our centre is similar to the one reported in the literature. The majority of RBDs were diagnosed from a preoperative test and this allowed preventive treatment prior to invasive procedures to avoid bleeding complications. 83% of patients did not have a pathological bleeding phenotype according to ISTH‐BAT

asymptomatic patients and it may vary by country or region [1,2].
Symptoms of RBDs may vary depending on the type of deficiency and patient. These symptoms are usually related to the type of mutation causing the disease and the plasma levels of the deficient factor, which may cause more or less bleeding. [2]. Usually, with a factor level >20%, haemostasis is ensured to lead a normal daily life, though it should be individualised according to the factor deficiency and patientt's characteristics.
The diagnosis is initiated in patients with haemorrhagic symptoms and/or abnormal results in basic coagulation tests including activated partial thromboplastin time (aPTT), prothrombin time (PT), and thrombin time (TT). [13]. If abnormal results are found in these tests, a mixing test (50:50) is performed to rule out the presence of an inhibitor. When the mixing test corrects the coagulation tests, levels of clotting factors are determined. Although some hereditary bleeding disorders can be easily diagnosed through clinical evaluation and laboratory studies, genetic studies allow a more accurate and complete diagnosis for some of these conditions. [14] Nowadays, with Next Generation Sequencing techniques, different gene panels have been developed to approach the genetic study of all bleeding disorders with one common strategy which is faster, easier and cheaper than the individual study of each gene using traditional Sanger sequencing technology. [15] The objective of this study was to analyze the prevalence and characteristics of patients with RBDs, with levels < 20% followed at a tertiary-level hospital and to investigate whether there are correlations between factor levels and bleeding phenotype, as well as with the genetic mutation responsible for the disorder if available. The secondary endpoint was to compare the data reported in the literature.

MATERIALS AND METHODS
Due to the variability between the bleeding phenotype and factor levels, we conducted an observational retrospective study where we collected data from patients with RBDs followed at a tertiary-level hospital.
A complete review was made of the medical charts of the patients diagnosed with RBDs with factor levels < 20%, less than 1 g/dl in case of fibrinogen deficiency, who attended a reference centre between January 2014 and December 2021. The cut-off levels that we considered to classify patients as affected with a severe deficiency were < 20% following the CSUR (Reference Centres, Departments and Units) criteria.

What is the NEW aspect of your work?
This study is the first study in which clinical data from south of Europe RBD patients is described and shows the highest rate of genetic testing compared with what was published until now.
What is the CENTRAL finding of your work?
The distribution of RBDs in our centre is similar to the one reported in the literature and It is striking that most patients with pathological bleeding scores are women.
What is (or could be) the SPECIFIC clinical relevance of your work?
The majority of RBDs were diagnosed from a preoperative test and this allowed prophylactic treatment prior to invasive procedures to avoid bleeding complications. This shows the importance of preoperative coagulation tests and good anamnesis.
Patients with factor levels > 20% were excluded, as well as patients with fibrinogen levels > 1 g/dl and those whose bleeding disorders were caused by anticoagulant therapy or other acquired bleeding disorders.

Clotting assays
In all cases, the one-stage clotting assay was used in order to determine factor levels and for fibrinogen, prothrombin time-derived fibrinogen tests and Clauss assay were performed. Samples were collected into tubes containing sodium citrate.

Bleeding phenotype
The bleeding phenotype was assessed by the ISTH-BAT. The cut-off value for an abnormal ISTH-BAT score was ≥4 in adult men, ≥6 in adult women, and ≥3 in children [16]. The score was performed using data collected from the medical charts.

Data analysis and mutation validation
Barcoded sequences were de-multiplexed and analyzed individually.
After variant calling, the resulting files were used as input for Vari-

RESULTS
In total, 101 patients followed in our centre between January 2014 and December 2021 had severe RBDs based on our definition, with factor deficiency levels < 20% or fibrinogen < 1 g/dl. The minimum follow-up time was 6 months. The overall characteristics of the series are summarized in Table 1. The most prevalent deficiency is FVII deficiency (36.63%). Data on the reason why the study was performed were collected being an abnormal basic coagulation test in a preoperative period the most common diagnostic reason (37.62%).
Only 18 patients (17.82%) reported abnormal bleeding at the time of diagnosis (regardless of the diagnostic reasons) and 48 (47.52%) reported it in their lifetime. Twenty-three per cent of our population had a pathological ISTH-BAT score, and of these, 70% were female patients.
In contrast, 64 patients (63.36%) received replacement therapy in their lifetime due to haemorrhage or as prophylaxis before invasive procedures, the latter being the reason in 82% of patients.
Most patients (63.36%) had undergone a genetic study to find the mutation responsible for the deficiency. From those studies, 46% of the patients had either a homozygous mutation or 2 or more heterozygous mutations in the same gene. The most common mutation type in our severe patients was a missense mutation (70%).
The characteristics of the patients with each specific deficiency are summarised in Table 2 and their treatment is in Table 3. Cases with hypofibrinogenaemia and dysfibrinogenaemia are usually asymptomatic as they have fibrinogen levels of around 1 g/L, but they may have pathological bleeding secondary to trauma or surgery and in cases of dysfibrinogenemia, thrombotic events can also occur [19]. In this regard, patients in our series with levels below 0.8-1 g/dl did not experience spontaneous bleeding or postoperative complications as they received replacement therapy with fibrinogen concentrate prior to invasive procedures with no thrombotic events reported. Every mutation studied was missense and the most found was a substitution in exon 8 of the FGG gene (66.67%), although not all patients were related. As recommended in the literature, in our series treatment and prophylaxis were performed with fibrinogen concentrates. In no case was necessary to use fresh frozen plasma (FFP) or cryoprecipitate [13,18].
FII was the most uncommon RBD as reported in the literature, therefore we did not find any patient with levels < 20% in our series [1,3].
Regarding FV deficiency, severe bleeding symptoms occur in patients with FV levels < 5%, while patients with FV between 5%-20% have usually only mild symptoms [21]. Thus, none of the patients with levels between 10-15% in this series presented any spontaneous bleeding, and only one of them required FFP before surgery. According to the literature, in our series, the highest ISTH-BAT score was observed on two female patients with levels < 1% who reported musculoskeletal and intracranial bleeding. It is striking that all patients who underwent the genetic study showed at least 2 heterozygous mutations involving more than one exon in the F5 gene. More than 60 reported mutations were found in the literature, although it is true that no complete deletions of the gene were found, so this is thought to be incompatible with life [21]. Since there is no FV concentrate available, treatment was based on administering FFP. Huang et al. have reported that in the event of serious bleeding, platelet concentrates may be administered, since they are an FV reservoir, and in refractory cases, recombinant activated factor VII (rFVIIa) may be considered as a bypassing agent, when necessary. [21] In our series, platelet transfusion was required in one patient due to a bleeding complication after brain surgery despite FFP transfusions.
In our series, FVII deficiency was the most common disorder as Severe FX deficiency has been one of the most uncommon disorders in our series a somewhat lower incidence than reported in the literature [1]. Usually, patients with levels between 10% and 40% experience mild mucocutaneous bleeding symptoms, like our patients with levels of 20%, who never required replacement therapy. [24] Auerswald recommended continued prophylaxis only in patients with FX < 1% [25]. However, in our series, two patients with levels > 1% (8.4% and 4% respectively) received regular prophylaxis due to frequent musculoskeletal haemorrhages and HMB.
FXI deficiency is characterized because FXI levels correlate very poorly with the bleeding phenotype. The literature shows that bleed-ing usually occurs after trauma or surgery in areas with high fibrinolytic activity such as the oral and nasal cavity, and urinary tract, and in the case of women, HMB and partum-related haemorrhages [26].
In our series, only 35% of patients had experienced bleeding symptoms, mainly after surgery, although one patient also reported HMB.
On the other hand, 56.5% received prophylaxis before surgery, half with FFP and half with FXI. Although cases of thrombosis have been reported during FXI replacement, none of our patients experienced this complication [27].
Regarding the genetic study, patients had different types of mutations, as described in the literature when dealing with a population other than Ashkenazi Jews. However, most of our patients with an available genetic study had either a homozygous mutation or 2 or more heterozygous mutations, as would be expected in this type of rare bleeding disorder due to their autosomal recessive inheritance or variable penetrance [28]. Eighty-one per cent of the mutations found were missense type.
According to the latest studies on FXIII deficiency, patients with levels < 10% will generally present bleeding symptoms. Patients with < 1% have a high risk of spontaneous and early bleeding such as umbilical cord and CNS bleeding, and also delayed cicatrization [29]. In our series, three patients with FXIII deficiency were diagnosed due to bleeding symptoms and their levels ranged from 0.9-3.8%. Twenty-three per cent of our population had a pathological ISTH-BAT score, and of these, 70% were female patients, which suggests that women are more prone to bleeding, so it is particularly important to make an appropriate diagnosis and follow-up in these patients.