Abstract
Background
Although the phenotype of severe hemophilia has been well studied, there are still knowledge gaps in nonsevere hemophilia.
Objectives
The objective of this study was to characterize the clinical bleeding phenotype in nonsevere hemophilia and its association with different factor VIII/IX assessments.
Methods
This was a cross-sectional, multicenter study to investigate the bleeding phenotype in adults with nonsevere hemophilia by the number of bleeding and joint bleeding in the past 5 years, a joint score, and the International Society on Thrombosis and Haemostasis bleeding assessment tool (ISTH-BAT). Factor levels were analyzed by 1-stage (lowest in history and at study inclusion) and chromogenic assay (at study inclusion). Patients were enrolled between March 2015 and May 2019.
Results
Of the 111 patients (86 with mild and 25 with moderate hemophilia), 57 patients (54.8%) reported any bleeding and 24 (23.1%) any joint bleeding in the past 5 years. A joint score ≥1 was found in 44 patients (41.9%), an ISTH-BAT ≥4 in 100 patients (90.1%), and an ISTH-BAT joint item ≥1 in 50 patients (45.0%). Within the ISTH-BAT, muscle and joint bleeds showed the largest difference between mild and moderate hemophilia. The lowest factor VIII/IX level in patients’ history was best associated with bleeding outcomes. Factor was inversely associated with joint bleeds (incidence rate ratio 0.88; 95% CI, 0.79-0.98), joint score, and ISTH-BAT (odds ratios from proportional odds ordinal logistic regression 0.92; 95% CI, 0.87-0.97; and 0.89; 95% CI, 0.86-0.93, respectively).
Conclusion
The occurrence of joint bleeding differentiated persons with mild and moderate hemophilia. The ISTH-BAT and lowest factor in patients’ history provided valuable information of the bleeding phenotype in nonsevere hemophilia.
Keywords: bleeding, factor VIII, factor IX, hemophilia, phenotype
Essentials
-
•
Data on bleeding, assessed by standardized clinical scores, are scarce in nonsevere hemophilia.
-
•
We evaluated bleeding type using the International Society on Thrombosis and Haemostasis bleeding assessment tool, joint status, and bleeding frequency.
-
•
Joint and muscle bleeds were used to differentiate persons with mild and moderate hemophilia.
-
•
Lowest factor levels in patient history were best associated with bleeding outcomes.
1. Introduction
The most common classification of severity in hemophilia utilizes patients’ residual factor VIII/IX (FVIII/IX) activity levels to categorize the disease into severe (<1 IU/dL), moderate (1-5 IU/dL), and mild (5-40 IU/dL) hemophilia [1]. However, in nonsevere hemophilia, accurate measurement of residual factor activity and adequate classification can be complicated by interassay variation and age-related increase in FVIII/IX levels [[2], [3], [4], [5], [6], [7]].
Although well established, this simple classification of disease severity is unreliable in reflecting the individual bleeding phenotype in hemophilia. About a quarter of persons with moderate hemophilia show a phenotype more similar to the severe form. These persons experience more bleeding and joint impairment and might require intermittent or long-term prophylaxis [8]. On the other hand, most persons with mild and moderate hemophilia remain free of spontaneous bleeding [9]. These observations indicate that factors other than FVIII/IX activity levels might also influence the bleeding phenotype and underline the importance of the clinical evaluation of the individual bleeding phenotype in persons with hemophilia.
Various clinical outcome parameters to evaluate the individual bleeding phenotype in hemophilia are available [[8], [9], [10], [11], [12], [13], [14], [15], [16]]. Among them, the annualized number of (joint) bleeding and the joint score of the World Federation of Hemophilia (WFH)—based on the physical examination of ankles, knees, and elbows [10,11]—are common hemophilia-specific measures but apply mostly to persons with severe hemophilia.
Data characterizing the bleeding burden in moderate and mild hemophilia, however, remain scarce. Recent studies have shown that bleeding episodes also impede the health-related quality of life in persons with nonsevere hemophilia [17]. The bleeding assessment tool of the International Society on Thrombosis and Haemostasis (ISTH-BAT), short ISTH-BAT, has originally been developed not only to identify patients with pathologically increased bleeding on the basis of reported bleeding symptoms but also to allow standardized estimation of bleeding severity [12]. The ISTH-BAT is not a hemophilia-specific tool and assesses the bleeding severity of bleeding diatheses independent of the underlying cause [18].
This study aims to evaluate the clinical bleeding phenotype as assessed by the ISTH-BAT, a modified WFH joint score, and the number of bleeding episodes in the 5 years before study inclusion and its association with FVIII/IX activity levels in persons with nonsevere hemophilia.
2. Methods
2.1. Patients
This was an observational, cross-sectional study conducted as a joint project of 3 Austrian hemophilia treatment centers from Vienna and Lower Austria, Upper Austria, and Tyrol within the framework of the Austrian Hemophilia Registry [19,20]. All study procedures were approved by the local ethic committees (EC Nr. 981/2011, 553/2010, AN2016-0271 369/4.5, F-10-16), based on the Declaration of Helsinki. All patients gave written informed consent before study inclusion. For obtaining patients’ raw data, it is possible to contact the corresponding author (J.G.) for more information.
Adult patients (≥18 years of age) with at least 1 FVIII/IX activity measurement at or above 1 and below 40 IU/dL in their medical history were eligible for study inclusion. The following exclusion criteria applied: platelet count <100 G/L, impaired renal or hepatic function (serum creatinine >2.0 mg/dL or prothrombin time <75% of normal), active malignancy, surgery within the past 6 weeks or overt infection within the past 2 weeks before study inclusion, or inhibitors against FVIII/IX on the time of sample collection.
Molecular analysis of the FVIII- or FIX-encoding gene was performed in all patients. In 4 persons with hemophilia A and 1 person with hemophilia B, no underlying genetic defect was found. In persons with hemophilia A, von Willebrand disease type 2N was excluded.
For the definition of mild and moderate hemophilia, we followed the recommendations of ISTH [1]. Patients with a lowest FVIII/IX measurement >5 but <40 IU/dL were classified as mild, and those with levels ≤5 but at least 1 IU/dL, as persons with moderate hemophilia.
Patient recruitment started in March 2015 and was completed in May 2019. Eligible patients were contacted and asked for study participation within a routine follow-up appointment. One hundred nineteen of 172 eligible patients were screened, and 111 patients were included in the participating centers. All included patients are of European descent.
A STROBE diagram is shown in Figure 1.
Figure 1.
STROBE diagram—Patients included in this study. VWD 2N, von Willebrand disease type 2N; MGUS, Monoclonal gammopathy of undetermined significance
To exclude interobserver variation, data collection through patient interviews, evaluation of patients’ joint status, and recording of the ISTH-BAT score was performed by the same person (J.R.) in all patients included in this analysis.
2.2. Estimation of bleeding severity
On study inclusion, trained medical personnel recorded data on bleeding and joint bleeding within the past 5 years using a standardized questionnaire, measured the joint status, and recorded the ISTH-BAT. All data recorded in the patient interviews were cross-checked with hospital charts and the Austrian Hemophilia Registry.
Because the number of bleeding episodes in most patients with nonsevere hemophilia is typically low, we chose to record all bleeding and joint bleeding that occurred in the past 5 years before study inclusion [6].
The joint status comprised a physical examination and an evaluation of chronic joint pain, based on the recommendation of the WFH [21]. In addition to ankles, knees, and elbows, we also evaluated shoulders and hips. The joint score used in our study ranged from 0 to 130 points, with a score of 0 reflecting joints without pathological finding and 130 reflecting most severe arthropathy of all joints. A description of the evaluation of the score as applied in this study is provided in Supplementary Paragraph 1 and Supplementary Table 1.
Bleeding severity was further assessed using the ISTH-BAT score, which, for men, consists of 12 items for recording bleeding symptoms: 1) epistaxis, 2) cutaneous bleeding, 3) bleeding from minor wounds, 4) hematuria, 5) gastrointestinal bleeding, 6) oral cavity bleeding, 7) bleeding after tooth extraction, 8) bleeding after surgery or major trauma, 9) muscle bleeding, 10) joint bleeding, 11) central nervous system bleeding, and 12) other bleeding. Each bleeding symptom is scored with 0 to 4 points depending on the severity, as described in the study of Rodeghiero et al. [22]. The ISTH-BAT thus spans a range from 0 to 48 points in men with a score of ≥4 points indicating a pathologic bleeding tendency in adult men [14,23,24]. Because joint bleeding is the hallmark of hemophilia, we also chose to analyze the joint item of the ISTH-BAT separately. In detail, no joint bleeds scored 0, posttraumatic untreated joint bleeds scored 1, spontaneous untreated joint bleeds scored 2, joint bleeds treated by desmopressin or factor replacement therapy scored 3, and joint bleeds treated by surgery or transfusion scored 4 [22].
2.3. Determination of FVIII/IX activity
On study inclusion, blood sampling was performed as reported in Supplementary Paragraph 2.
Factor VIII/IX activity was routinely measured at study inclusion by a 1-stage clotting assay (OSA) using the following reagents: FVIII- or FIX-deficient plasma (Technoclone, Vienna, Austria) and Actin FS activator (Siemens, Marburg, Germany) in Vienna and Lower Austria and FVIII-deficient plasma (Siemens) and Pathromtin SL activator (Siemens) in Linz and Innsbruck.
Factor VIII/IX activity by chromogenic substrate assay (CSA, TECHNOCHROM FVIII:C, Technoclone, Vienna, Austria or Biophen Hyphen Biomed, Neuville-sur-Oise, France) was measured in frozen plasma samples after thawing. In one patient from Innsbruck, CSA was performed locally with the COAMATIC Factor VIII assay (Chromogenix, Instrumentation Laboratory SpA, Bedford, USA).
Lowest FVIII/IX OSA levels were defined as the lowest available FVIII/IX activity among measurements at study inclusion and previous measurements recorded in the Austrian Hemophilia Registry after cross-checking with patients’ medical records.
2.4. Statistical analysis
Categorical and continuous variables were expressed as frequencies (percentages) and median (25th-75th percentile), respectively. Comparisons between groups were performed by using the Wilcoxon rank sum test for continuous variables and the chi-squared test for categorical values. To describe the relationship of the 3 different factor level measurements with the bleeding phenotype, we fitted separate regression models for every factor level (FVIII/IX OSA and FVIII/IX CSA at study inclusion and lowest FVIII/IX OSA in history) and outcome measurement (all bleeding, joint bleeding, joint score, and ISTH-BAT total and joint item score). We estimated the number of bleeding episodes in negative binomial regression models, whereas the joint score, the ISTH-BAT total score, and the ISTH-BAT joint item score were modeled using ordinal logistic regression. Because the joint score and the ISTH-BAT reflect the lifelong accumulation of joint damage and bleeding episodes, respectively, we further adjusted the ordinal logistic regression models for age at study inclusion. In an effort to compare the 3 different factor level measurements, we also calculated an adequacy index for every model fitted. This index was defined as the ratio between the likelihood ratio of the model in question (that is, with only 1 of the factor level measurements included as a predictor) compared with the likelihood ratio of a full model, in which all 3 factor level measurements were included as predictors [25]. This adequacy index can be interpreted as the proportion of predictive information provided by all 3 measurements that is already contained in the respective single measurement on its own—that is, the adequacy of the simpler model with only 1 factor level measurement that ignores the other 2 predictors. An adequacy index of 1 would, thus, indicate that a specific measurement already incorporates all of the predictive information found in a full model including all 3 factor level measurements. All statistical analyses were performed using R (4.2.0) [26].
3. Results
3.1. Patients’ characteristics
The study cohort consisted of 111 persons with nonsevere hemophilia, including 86 persons (77.5%) with mild and 25 patients (22.5%) with moderate hemophilia, 5 of whom (20%) received prophylaxis. Eighty-nine persons (80.2%) had hemophilia A (mild: N = 70, 78.7%) and 22 persons (19.8%) had hemophilia B (mild: N = 16, 72.7%). Characteristics of all persons stratified by severity are summarized in Table 1.
Table 1.
Demographic and laboratory characteristics of all patients and stratified by severity.
| Overall,N = 111 |
Mild hemophilia,N = 86 |
Moderate hemophilia,N = 25 |
|
|---|---|---|---|
| n (%) | n (%) | n (%) | |
| Hemophilia A | 89 (80.2) | 70 (81.4) | 19 (76.0) |
| Hemophilia B | 22 (19.8) | 16 (18.6) | 6 (24.0) |
| Blood group O | 63 (56.8) | 46 (53.5) | 17 (68.0) |
| Patients of European descent | 111 (100.0) | 86 (100.0) | 25 (100.0) |
| median (25th-75th percentile) | median (25th-75th percentile) | median (25th-75th percentile) | |
| Age, y | 49.8 (34.6-60.4) | 51.0 (37.1-61.5) | 39.5 (32.1-55.5) |
| BMI, kg/m2 | 25.7 (23.3-28.3) | 25.7 (23.3-28.3)a | 25.3 (23.6-28.3) |
| Time since diagnosis, y | 33.0 (19.6-45.8) | 32.7 (16.5-43.3)c | 34.0 (27.7-48.9)b |
| Hemoglobin, g/dL | 15.2 (14.5-15.7) | 15.2 (14.3-15.7)c | 15.2 (14.8-15.7)b |
| Platelet count, G/L | 234.0 (200.0-265.0) | 237.0 (202.0-264.0)c | 212.5 (187.2-271.5)b |
| aPTT, s | 49.5 (44.9-55.3) | 48.3 (44.0-52.8) | 56.9 (54.5-61.6)b |
| Prothrombin time, % | 100.0 (91.0-115.2) | 100.0 (91.0-117.0)d | 100.0 (92.0-103.0)e |
| Fibrinogen, mg/dL | 308.0 (259.5-353.5) | 308.0 (265.0-352.2) | 293.0 (249.0-352.0) |
| Lowest FVIII/IX OSA, IU/dL | 12.0 (6.5-19.0) | 14.6 (10.0-20.8) | 3.0 (2.0-4.0) |
| FVIII/IX OSA, IU/dL | 15.0 (10.0-23.0) | 21.0 (13.0-24.8) | 5.2 (3.0-9.0) |
| FVIII/IX CSA, IU/dL | 11.0 (6.0-21.5) | 13.6 (8.9-25.0) | 4.0 (2.5-8.0) |
aPTT, activated partial thromboplastin time; FVIII, factor VIII; FIX, factor IX; OSA, one-stage assay; CSA, chromogenic substrate assay.
Data available on 84 patients (98%).
Data available on 24 patients (96%).
Data available on 85 patients (99%).
Data available on 77 patients (90%).
Data available on 21 patients (84%).
3.2. The bleeding phenotype in mild and moderate hemophilia
In total, 57 patients (55%) reported any bleeding and 24 (23%) any joint bleeding in the past 5 years before study inclusion. Forty-four patients (42%) had a joint score ≥1 point, 100 patients (90%) had an ISTH-BAT score equal to or above the cutoff of 4 points, and 50 patients (45%) had an ISTH-BAT joint item of 1 or above (Table 2).
Table 2.
Parameters of the bleeding phenotype in all patients and stratified by severity.
| Overall, N = 111 |
Mild hemophilia, N = 86 |
Moderate hemophilia, N = 25 |
P value | |
|---|---|---|---|---|
| n (%) | n (%) | n (%) | ||
| Any bleeding in past 5 ya | 57 (54.8) | 42 (51.2) | 15 (68.2) | .16 |
| Any joint bleeding in past 5 ya | 24 (23.1) | 13 (15.9) | 11 (50.0) | <.001 |
| Joint score ≥1b | 44 (41.9) | 30 (36.6) | 14 (60.9) | .04 |
| ISTH-BAT ≥ 4 | 100 (90.1) | 76 (88.4) | 24 (96.0) | .45 |
| Joint item of the ISTH-BAT ≥1 | 50 (45.0) | 31 (36.0) | 19 (76.0) | <.001 |
| median (25th-75th percentile) | median (25th-75th percentile) | median (25th-75th percentile) | ||
| Bleeding in past 5 yc | 0.0 (0.0-1.0) | 0.0 (0.0-1.0) | 1.0 (0.0-2.0) | .17 |
| Joint bleeding in past 5 yd | 0.0 (0.0-0.0) | 0.0 (0.0-0.0) | 0.0 (0.0-2.0) | .002 |
| Joint scoreb | 0.0 (0.0-4.0) | 0.0 (0.0-2.0) | 4.0 (0.0-7.5) | .02 |
| ISTH-BAT | 11.0 (7.0-15.0) | 9.0 (6.0-14.0) | 14.0 (12.0-16.0) | <.001 |
| Joint item of the ISTH-BAT | 0.0 (0.0-3.0) | 0.0 (0.0-3.0) | 3.0 (1.0-4.0) | <.001 |
ISTH-BAT, International Society on Thrombosis and Haemostasis bleeding assessment tool.
Data available on 82 (95%) persons with mild and 22 (88%) persons with moderate hemophilia.
Data available on 82 (95%) persons with mild and 23 (92%) persons with moderate hemophilia.
Data available on 75 (87%) persons with mild and 16 (73%) persons with moderate hemophilia.
Data available on 78 (91%) persons with mild and 18 (72%) persons with moderate hemophilia.
The bleeding phenotype differed between persons with mild and moderate hemophilia (Table 2). As expected, persons with moderate hemophilia more often had joint bleeding in the past 5 years, a higher joint score, and a higher ISTH-BAT score compared with those with mild hemophilia (Table 2). All but 1 person with moderate hemophilia (96.0%) and 76 of 86 (88.4%) with mild hemophilia had a total ISTH-BAT score equal to or above the cutoff, indicating a pathologic bleeding tendency.
When the various items of the ISTH-BAT score were evaluated separately, the most frequently reported bleeding symptoms were bleeding after tooth extraction and bleeding after surgery or major trauma, affecting more than half of persons with mild or moderate hemophilia and, most interestingly, with a similar frequency in both groups (Table 3). Joint and muscle bleeding occurred more frequently in moderate than in persons with mild hemophilia, whereas mucocutaneous bleeding, such as epistaxis, hematoma, and bleeding from small wounds, were comparable in both groups (Table 3).
Table 3.
Frequency of bleeding symptoms as assessed by the ISTH-BAT stratified by severity.
| Overall, N = 111 |
Mild hemophilia, N = 86 |
Moderate hemophilia, N = 25 |
P value | ||||
|---|---|---|---|---|---|---|---|
| N | (%) | n | (%) | n | (%) | ||
| Epistaxis | 54 | (48.6) | 42 | (48.8) | 12 | (48.0) | .94 |
| Cutaneous bleeding | 52 | (46.8) | 38 | (44.2) | 14 | (56.0) | .3 |
| Bleeding from minor wounds | 42 | (37.8) | 34 | (39.5) | 8 | (32.0) | .49 |
| Hematuria | 13 | (11.7) | 8 | (9.3) | 5 | (20.0) | .16 |
| Gastrointestinal bleeding | 11 | (9.9) | 8 | (9.3) | 3 | (12.0) | .71 |
| Oral cavity bleeding | 17 | (15.3) | 12 | (14.0) | 5 | (20.0) | .53 |
| Bleeding after tooth extractiona | 63 | (67.0) | 52c | (70.3) | 11c | (55.0) | .2 |
| Bleeding after surgery or major traumab | 71 | (73.2) | 53d | (69.7) | 18d | (85.7) | .14 |
| Muscle bleeding | 51 | (45.9) | 33 | (38.4) | 18 | (72.0) | .003 |
| Joint bleeding | 50 | (45.0) | 31 | (36.0) | 19 | (76.0) | <.001 |
| Central nervous system bleeding | 5 | (4.5) | 3 | (3.5) | 2 | (8.0) | .32 |
| Other bleeding | 4 | (3.6) | 3 | (3.5) | 1 | (4.0) | >1 |
ISTH-BAT, International Society on Thrombosis and Haemostasis bleeding assessment tool.
n, number of patients with a respective bleeding manifestation.
Including 94 patients who underwent tooth extraction (74 mild hemophilia and 20 moderate hemophilia).
Including 97 patients who underwent surgery or experienced major trauma (76 mild hemophilia and 21 moderate hemophilia).
Mild hemophilia: prophylaxis in 18 patients (34.6%), no prophylaxis in 30 (57.7%), infrequent prophylaxis in 2 patients (3.8%), no data in 2 patients (3.8%); moderate hemophilia: prophylaxis in 2 patients (18.2%), no prophylaxis in 8 patients (72.7%), and no data in 1 patient (9.1%).
Mild hemophilia: prophylaxis in 17 patients (32.1%), no prophylaxis in 34 (64.2%), infrequent prophylaxis in 2 patients (3.8%); moderate hemophilia: prophylaxis in 6 patients (33.3%), no prophylaxis in 11 (61.1%), and infrequent prophylaxis in 1 patient (5.6%).
3.3. The bleeding phenotype in hemophilia A and B
Patients’ demographic and laboratory characteristics in hemophilia A and B are described in Supplementary Table 2. Persons with hemophilia A were older and had lower platelet counts than persons with hemophilia B. There was no difference in the other demographic or laboratory characteristics between persons with hemophilia A and B.
Concerning the bleeding phenotype, the ISTH-BAT score was slightly higher in persons with hemophilia A than in persons with hemophilia B; nevertheless, this difference did not persist after adjustment for age (data not shown). As shown in Supplementary Table 3, no difference in the frequency or number of the reported bleeding outcome measurements was seen between persons with hemophilia A or B.
3.4. The bleeding phenotype according to FVIII/FIX levels
The different measurements of FVIII/IX activity correlated well with Pearson’s correlation coefficients of 0.92 (lowest FVIII/IX OSA and OSA at study inclusion), 0.80 (lowest FVIII/IX OSA and CSA at study inclusion), and 0.87 (FVIII/IX OSA and CSA at study inclusion), as depicted in Supplementary Figure 1.
For describing the continuous relationship between the different FVIII/IX measurements and bleeding events within the past 5 years, the joint score, and the ISTH-BAT score, we fitted negative binominal and ordinal logistic regression models, respectively, as summarized in Table 4. Patients with higher factor FVIII/IX activity were more likely to have lower numbers of any bleeding and joint bleeding and scored lower on the joint score, the ISTH-BAT, and the joint item of the ISTH-BAT (Table 4).
Table 4.
Association of FVIII/IX measurements with the bleeding phenotype.
| All bleeds in last 5 yb | IRR | (95% CI) | AIC | pseudo R2 | LR chi-squared | Adequacy |
|---|---|---|---|---|---|---|
| Lowest FVIII/FIX OSA | 0.94 | (0.89-1.00) | 327.49 | 4.67% | 4.23 | 0.347 |
| FVIII/FIX OSA | 0.96 | (0.92-1.00) | 329.76 | 2.19% | 1.96 | 0.161 |
| FVIII/FIX CSA | 1.00 | (0.96-1.05) | 331.68 | 0.05% | 0.04 | 0.004 |
| Joint bleeds in last 5 yc | IRR | (95% CI) | AIC | pseudo R2 | LR chi-squared | Adequacy |
| Lowest FVIII/FIX OSA | 0.88 | (0.79-0.98) | 177.91 | 6.79% | 5.66 | 0.424 |
| FVIII/FIX OSA | 0.91 | (0.83-0.99) | 179.60 | 4.81% | 3.98 | 0.298 |
| FVIII/FIX CSA | 0.96 | (0.88-1.05) | 182.64 | 1.15% | 0.93 | 0.070 |
| Joint scored | ORa | (95% CI) | AIC | pseudo R2 | LR chi-squared | Adequacy |
| Lowest FVIII/FIX OSA | 0.92 | (0.87-0.97) | 332.36 | 10.56% | 11.14 | 0.721 |
| FVIII/FIX OSA | 0.93 | (0.88-0.97) | 330.57 | 12.17% | 12.94 | 0.837 |
| FVIII/FIX CSA | 0.92 | (0.88-0.96) | 328.28 | 14.16% | 15.23 | 0.985 |
| ISTH-BATe | ORa | (95% CI) | AIC | pseudo R2 | LR chi-squared | Adequacy |
| Lowest FVIII/FIX OSA | 0.89 | (0.86-0.93) | 685.37 | 22.85% | 28.72 | 0.990 |
| FVIII/FIX OSA | 0.92 | (0.89-0.96) | 692.12 | 18.00% | 21.97 | 0.757 |
| FVIII/FIX CSA | 0.93 | (0.90-0.96) | 695.21 | 15.68% | 18.88 | 0.651 |
| Joint item of the ISTH-BATe | ORa | (95% CI) | AIC | pseudo R2 | LR chi-squared | Adequacy |
| Lowest FVIII/FIX OSA | 0.89 | (0.84-0.94) | 235.40 | 19.46% | 21.10 | 0.973 |
| FVIII/FIX OSA | 0.91 | (0.87-0.96) | 237.78 | 17.45% | 18.72 | 0.863 |
| FVIII/FIX CSA | 0.92 | (0.88-0.96) | 240.36 | 15.22% | 16.14 | 0.744 |
Separate negative binomial (IRR) and ordinal logistic regression (OR) models fitted per outcome and factor assay. IRR and OR are depicted as per 1-unit increase. The adequacy index compares the respective model to a full model including all 3 factor assays as predictors.
AIC, Akaike information criterion; CSA, chromogenic substrate assay; IRR, incidence rate ratio; LR, likelihood ratio; OR, odds ratio; OSA, one-stage assay; pseudo R2, Nagelkerke pseudo R2.
adjusted for age at study inclusion.
n = 91.
n = 96.
n = 105.
n = 111.
Among the different FVIII/IX measurements, lowest FVIII/IX activity appeared the most adequate measurement for predicting bleeding or joint bleeding in the past 5 years, the ISTH-BAT score, and the joint item of the ISTH-BAT score because it contained the highest proportion of the predictive information found in a full model incorporating all 3 measurements as predictors. Lowest FVIII/IX activity by itself already contained 99.0% and 97.3% of the information for predicting the ISTH-BAT score and ISTH-BAT joint item scale (Table 4). This adequacy index was far lower for the number of any bleeding (34.7%) or joint bleeding (42.4%) in the past 5 years.
A one unit increase in lowest FVIII/IX OSA was associated with a decrease in total and joint bleeds by an incidence rate ratio of 0.94 (95% CI, 0.89-1.00) and 0.88 (0.79-0.98), respectively. The total number of bleeding in the past 5 years amounted to 5.81 (2.56-13.18) at 1 IU/dL lowest OSA, 4.59 (2.39-8.80) at 5 IU/dL lowest OSA, 3.42 (2.08-5.62) at 10 IU/dL lowest OSA, and 2.54 (1.59-4.06) at 15 IU/dL lowest OSA. Estimated numbers of joint bleeding in the past 5 years at the aforementioned levels were 4.94 (1.14-21.50), 2.97 (0.94-9.41), 1.57 (0.66-3.77), and 0.83 (0.35-1.96), respectively. Based on these model results, Figure 2 depicts probabilities for the occurrence of at least one or 5 (joint) bleeding episodes in the past 5 years according to the lowest FVIII/IX OSA levels.
Figure 2.
Probability of the occurrence of (A) any bleeding and (B) joint bleeding within the past 5 years according to lowest factor VIII/IX one-stage clotting assay levels
As depicted in Figure 3A, the probability for a total ISTH-BAT score equal to or above the cutoff for healthy men of 4 [23] was 98.5% (95.9-99.4) for 1 IU/dL lowest FVIII/IX OSA, 97.6% (94.4-99.0) for 5 IU/dL, 95.9% (91.6-98.1) for 10 IU/dL, and 93.1% (87.0-96.4) for 15 IU/dL lowest OSA. Other cutoffs of the ISTH-BAT score (7, 11, and 15) shown in Figure 3A depict the 25th, 50th, and 75th percentiles of the total group. Holding age constant at the cohort median of 49.8 years, the corresponding probabilities of having a joint score ≥1 were 68.0% (50.3-81.7), 60.3% (46.1-73.0), 50.0% (39.3-60.7), and 39.6% (29.9-50.3) at the respective FVIII/IX OSA levels mentioned above (Figure 3B).
Figure 3.
Probability of (A) International Society on Thrombosis and Haemostasis bleeding assessment tool ISTH-BAT) score above or equal to the official cutoff of 4 points, and above the 25th, 50th, and 75th percentiles (7, 11, and 15 points) and (B) joint score above zero according to lowest factor VIII/IX one-stage clotting assay levels, holding age constant at the cohort median of 49.8 years
4. Discussion
In our study, we analyzed the bleeding phenotype in 111 persons with mild or moderate hemophilia A or B and found that joint and muscle bleeds occur more often in persons with moderate than in those with mild hemophilia, whereas there was no clear difference in the occurrence of other bleeding symptoms. Higher FVIII/IX levels were associated with a decrease in severity of all bleeding outcomes with lower numbers of any bleeding and joint bleeding and lower scores in the joint score, the ISTH-BAT, and the joint item of the ISTH-BAT. Among the different FVIII/IX measurements, lowest FVIII/IX OSA activity was the most adequate, especially for predicting the lifelong bleeding symptom burden recorded by the ISTH score and the joint item of the ISTH-BAT.
So far, a limited number of studies have been conducted to evaluate the bleeding phenotype in nonsevere hemophilia [8,15,16,[27], [28], [29], [30], [31], [32]]. Some of these studies have focused only on bleeding frequencies as a marker for the bleeding phenotype [13] or examined only specific aspects such as the onset of bleeding symptoms and joint impairment or the use of prophylaxis [8,28]. Our study evaluated the bleeding phenotype in nonsevere hemophilia on the basis of various measures including the ISTH-BAT score. The ISTH-BAT and especially the joint item of the ISTH-BAT have not yet been used as indicators of the bleeding phenotype in persons with nonsevere hemophilia. In contrast to previous studies and accounting for the generally low bleeding frequency in persons with nonsevere hemophilia, we analyzed bleeding and joint bleeding within the past 5 years before study inclusion instead of the more commonly investigated bleeding rate more than 12 months, as performed in other cohorts [13,15,16,33]. The median bleeding rates of our patients were similar in mild hemophilia and lower in moderate hemophilia compared with the median rates reported in the study by den Uijl et al. [8] with 0.0 bleeding/year in mild and 1.0 bleeding/year in persons with moderate hemophilia, compared with 0.0 bleeding/year in mild and 0.2 bleeding/year in persons with moderate hemophilia in our study. In the recently published Dynamo study, annual bleeding rates were calculated from the number of treated bleeds divided by a follow-up time of median 11 years [34]. The annualized bleeding rates, both in total and when stratified by severity, were higher in the Dynamo study compared with our study (overall: 0.2 vs 0.0, mild hemophilia: 0.2 vs 0.0, moderate hemophilia: 0.6 vs 0.2) and the study by den Uijl. The median annualized joint bleeding rate was 0.0 in persons with mild hemophilia in all 3 studies. In persons with moderate hemophilia, we and den Uijl et al. reported an annualized joint bleeding rate of 0, whereas it was slightly higher in the Dynamo study (median 0.2).
Den Uijl et al. [8] reported on 61% of persons with mild hemophilia and 27% with moderate hemophilia who were free of bleeding the year before study inclusion. Among our patients with mild hemophilia, the rate without bleeding in the previous 5 years was 48.8% and, thus, lower, which can most probably be attributed to the longer observation period. Surprisingly, however, in moderate hemophilia, the proportion of patients who reported no bleeding was higher in our study (31.8%). The Dynamo study also evaluated the lifelong bleeding history and reported 25% of mild and 7% of persons with moderate hemophilia to be free of any treated bleeding. Sixty percent and 22% of persons with mild and moderate hemophilia were free of joint bleeding [34]. The rates of patients free of joint bleeding before study inclusion were higher in the cohort of den Uijl than in our study (91% compared with 84.1% in mild hemophilia and 61% compared with 50.0% in moderate hemophilia), which was also most probably owing to the longer observation period in our analysis.
Comparing the bleeding phenotype between persons with mild and moderate hemophilia, we found a higher total ISTH bleeding score in persons with moderate hemophilia with most pronounced differences in the muscle and joint bleeding items (Table 3). Previous data from the study of Borhany et al. [31] found no significant difference in the ISTH-BAT score between 10 persons with mild and 57 with moderate hemophilia. Nevertheless, the total ISTH-BAT scores of persons with mild and moderate hemophilia were similar to our results; thus, the lack of statistical difference might have resulted from the small cohort they investigated.
Our data showed significant differences between persons with mild and moderate hemophilia, especially in the occurrence and frequency of joint bleeds. Repeated intra-articular bleeding episodes result in hemophilic arthropathy, which has been acknowledged as a frequent complication not only in severe but also in persons with nonsevere hemophilia [35]. It has been previously reported that the age of the first joint bleeding is lower in persons with moderate hemophilia than that in persons with mild hemophilia [28] and that joint damage increases with age [36]. The earlier onset in persons with moderate hemophilia could have contributed to the higher joint damage found in persons with moderate hemophilia in our study. De la Corte-Rodriguez et al. [37] evaluated the presence of arthropathy in 85 mild hemophilia patients and detected arthropathy in 36.5%. Despite the different methods to detect arthropathy, the rate was equal to the 36.6% of mild hemophilia patients with joint impairment in our study. The rates of joint damage might even be higher, as demonstrated in a recent study, which detected joint damages of the ankle by MRI in 71% of patients with nonsevere hemophilia [38]. Because joint damage in hemophilia is one major contributor to health-related quality of life impairment [39], frequent monitoring of joint health and possibly early prophylaxis to avoid joint damage should also be considered in patients with nonsevere hemophilia [40].
Higher FVIII/IX levels were associated with a lower occurrence of joint bleeding, independent of the FVIII/IX measurement in our study. On the other hand, FVIII/IX levels seemed to affect the occurrences of other bleeding, such as mucocutaneous bleeding or bleeding from minor wounds, much less in our cohort. In the Dynamo study, in line with other studies, bleeding rates and joint bleeding rates decreased with increasing FVIII/IX activity in nonsevere hemophilia [13,34,37]. Nevertheless, and in contrary to our study, less severe bleeding manifestations, such as mucocutaneous bleeding, were not systematically recorded in these studies. Regarding joint bleeding, den Uijl et al. [28] calculated a FVIII level of 12.0 IU/dL of basal FVIII activity, above which patients generally did not have joint bleeds. In our cohort, 4 (8.2%) of the 49 patients with a lowest FVIII/IX OSA above 12 IU/dL reported a joint bleeding within the past 5 years. The number of bleeding episodes at a lowest factor activity of 12 IU/dL was estimated to be 1.22 (95% CI, 0.53-2.80) corresponding to a probability of 17.7% (13.5-21.9) of experiencing at least 1 joint bleed in the past 5 years.
Among laboratory measurements of FVIII/IX, lowest FVIII/IX OSA was found to predict most adequately for most bleeding outcomes. This was especially true for the total ISTH-BAT score and the joint item of the ISTH-BAT, reflecting a patient’s lifelong bleeding tendency. In the Dynamo study, the discrepancies between OSA and CSA were specifically studied and seen to be minor (Fijnvandraat K, personal communication). To our knowledge, a comparison between the clinical feature of a relatively large group of patients with nonsevere hemophilia and different assays has not yet been performed.
Our study has some limitations to consider. First, this was an observational, cross-sectional study with outcome data collected retrospectively by structured interviews. Thus, we are not able to provide treatment data for each of the bleeding episodes. Second, we were only able to include a relatively small number of persons with moderate hemophilia and with hemophilia B. Data on the lowest FVIII/IX levels measured by CSA were not available because until recently, FVIII/IX CSA was not routinely performed in the participating centers. In this study, although only ankles, knees, and elbows are included in the original WFH joint score, we chose to use a modified version adding the evaluation of shoulders and hips in an effort to give a more comprehensive and more sensitive representation of the joint health status. This could be regarded as a limitation regarding the external comparability of our study. Patient-reported chronic pain is part of the WFH joint score, with the limitation that pain evaluation is influenced by psychosocial factors [41]. Finally, all our regression models assumed a simple linear relationship between the factor activity measurements and the different outcomes, which might have led to biased results. Unfortunately, as is often the case with orphan diseases, our sample size was too small to fit more complex nonlinear models without risking severe overfitting. Even so, a linear continuous relationship is still more informative and less biased than a simple dichotomization of patients into a mild and moderate group that assumes the relationship to be flat below and above the threshold.
In our patient cohort, factor activity levels were inversely associated with all clinical bleeding outcome measures. Nevertheless, we could not confirm a “safe” FVIII/IX activity level above which no bleeding or joint bleeding or no joint damage occur (Figures 2 and 3). In dichotomized comparison, joint impairment, more than other bleeding manifestations, differentiated patients with mild and moderate hemophilia. Lowest FVIII/IX activity appeared the most adequate measurement for predicting the bleeding outcomes.
5. Conclusions
In conclusion, our data indicate that besides common hemophilia-specific measurements, information on the lifelong bleeding history as obtained with the ISTH-BAT, the joint item of the ISTH-BAT, and lowest FVIII/IX levels in the patients’ history provide valuable information for the estimation of the bleeding phenotype in persons with nonsevere hemophilia.
Acknowledgments
This study was carried out within the framework of the Austrian Hemophilia Registry. We thank Karl Zwiauer and Michael Sohm (Universitätsklinikum St. Pölten) for the recruitment of patients for the current study. We also thank Tanja Altreiter for proof-reading this manuscript. Last but not the least, we want to thank our patients, who are always interested in moving forward research and development in hemophilia in Austria.
Funding
The following resources supported our work. Bayer Hemophilia Clinical Training Award 2011 to J.G., unrestricted grant from Novo Nordisk to the institution, Hans-Egli-Stipendium of the German-Austrian-Swiss Society for Thrombosis and Haemostasis 2018 to J.G.
Author contributions
O.K., I.P., J.G., and J.R. designed the study; C.A., C.F., J.G., I.P., J.R., and G.S. included patients; P.Q., J.R., and R.S.P. performed laboratory analyses; E.G., O.K., D.K., and J.R. performed statistical analyses; C.G., J.G., E.G., O.K., D.K., I.P., and J.R. analyzed the data; J.G., O.K., D.K., I.P., and J.R. interpreted the data; J.R., J.G., I.P., and D.K. wrote the manuscript. All authors reviewed, edited, and approved the manuscript.
Relationship Disclosure
C.A. received personal fees for lectures and/or participation in advisory boards from Bayer, CSL Behring, Biomarin, Sobi, Roche, and LFB. C.F. has received occasional honoraria for lectures and advisory board meetings from Bayer, CSL Behring, Biomarin, Novo Nordisk, Sobi, and Takeda and research grants from Sobi and Biotest. C.G. has no conflict of interest to declare. J.G. received honoraria for lectures and/or participation in advisory boards from CSL Behring, Sobi, Novartis, and Amgen, and research grants to the Medical University of Vienna from CSL Behring, Sobi, and Novartis. E.G. has no conflict of interest to declare. O.K. received honoraria for advisory boards from CSL Behring. D.K. received honoraria for participation in advisory boards from CSL Behring. I.P. received honoraria for lectures and/or participation in advisory boards from Bayer, CSL Behring, Biomarin, Sobi, NovoNordisk, Roche, and Takeda and research grants to the Medical University of Vienna from Biotest, CSL Behring, NovoNordisk, and Sobi. P.Q. has no conflict of interest to declare. J.R. has no conflict of interest to declare. G.S. has no conflict of interest to declare. R.S. has no conflict of interest to declare.
Footnotes
Funding information The following resources supported our work. Bayer Hemophilia Clinical Training Award 2011 to J..G, unrestricted grant from Novo Nordisk to the institution, and Hans-Egli-Stipendium of the German-Austrian-Swiss Society for Thrombosis and Haemostasis 2018 to J.G.
Handling editor: Dr Johnny Mahlangu
The online version contains supplementary material available at https://doi.org/10.1016/j.rpth.2023.100047
Supplementary material
Supplementary figure 1.
References
- 1.Blanchette V.S., Key N.S., Ljung L.R., Manco-Johnson M.J., van den Berg H.M., Srivastava A., et al. Definitions in hemophilia: communication from the SSC of the ISTH. J Thromb Haemost. 2014;12:1935–1939. doi: 10.1111/jth.12672. [DOI] [PubMed] [Google Scholar]
- 2.Trossaërt M., Boisseau P., Quemener A., Sigaud M., Fouassier M., Ternisien C., et al. Prevalence, biological phenotype and genotype in moderate/mild hemophilia A with discrepancy between one-stage and chromogenic factor VIII activity. J Thromb Haemost. 2011;9:524–530. doi: 10.1111/j.1538-7836.2010.04174.x. [DOI] [PubMed] [Google Scholar]
- 3.Peyvandi F., Oldenburg J., Friedman K.D. A critical appraisal of one-stage and chromogenic assays of factor VIII activity. J Thromb Haemost. 2016;14:248–261. doi: 10.1111/jth.13215. [DOI] [PubMed] [Google Scholar]
- 4.Rejtő J., Königsbrügge O., Grilz E., Hofer S., Mauracher L.-M., Gabler C., et al. Influence of blood group, von Willebrand factor levels and age on factor VIII levels in non-severe haemophilia A. J Thromb Haemost. 2020;18:1081–1086. doi: 10.1111/jth.14770. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kurachi K., Kurachi S. Genetic mechanisms of age regulation of blood coagulation: factor IX model. Arterioscler Thromb Vasc Biol. 2000;20:902–906. doi: 10.1161/01.atv.20.4.902. [DOI] [PubMed] [Google Scholar]
- 6.Cid A.R., Calabuig M., Cortina V., Casaña P., Haya S., Moret A., et al. One-stage and chromogenic FVIII:C assay discrepancy in mild haemophilia A and the relationship with the mutation and bleeding phenotype. Haemophilia. 2008;14:1049–1054. doi: 10.1111/j.1365-2516.2008.01781.x. [DOI] [PubMed] [Google Scholar]
- 7.Kihlberg K., Strandberg K., Rosén S., Ljung R., Astermark J. Discrepancies between the one-stage clotting assay and the chromogenic assay in haemophilia B. Haemophilia. 2017;23:620–627. doi: 10.1111/hae.13219. [DOI] [PubMed] [Google Scholar]
- 8.Den Uijl I.E., Fischer K., Van Der Bom J.G., Grobbee D.E., Rosendaal F.R., Plug I. Clinical outcome of moderate haemophilia compared with severe and mild haemophilia. Haemophilia. 2009;15:83–90. doi: 10.1111/j.1365-2516.2008.01837.x. [DOI] [PubMed] [Google Scholar]
- 9.Franchini M., Favaloro E.J., Lippi G. Mild hemophilia A. J Thromb Haemost. 2010;8:421–432. doi: 10.1111/j.1538-7836.2009.03717.x. [DOI] [PubMed] [Google Scholar]
- 10.Gouw S.C., Timmer M.A., Srivastava A., Kleijn P de, Hilliard P., Peters M., et al. Measurement of joint health in persons with haemophilia: A systematic review of the measurement properties of haemophilia-specific instruments. Haemophilia. 2019;25:e1–e10. doi: 10.1111/hae.13631. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gilbert M.S. Prophylaxis: musculoskeletal evaluation. Semin Hematol. 1993;30:3–6. [PubMed] [Google Scholar]
- 12.Fasulo M.R., Biguzzi E., Abbattista M., Stufano F., Pagliari M.T., Mancini I., et al. The ISTH Bleeding Assessment Tool and the risk of future bleeding. J Thromb Haemost. 2018;16:125–130. doi: 10.1111/jth.13883. [DOI] [PubMed] [Google Scholar]
- 13.Den Uijl I.E., Fischer K., Van Der Bom J.G., Grobbee D.E., Rosendaal F.R., Plug I. Analysis of low frequency bleeding data: the association of joint bleeds according to baseline FVIII activity levels. Haemophilia. 2011;17:41–44. doi: 10.1111/j.1365-2516.2010.02383.x. [DOI] [PubMed] [Google Scholar]
- 14.James P.D., Mahlangu J., Bidlingmaier C., Mingot-Castellano M.E., Chitlur M., Fogarty P.F., et al. Evaluation of the utility of the ISTH-BAT in haemophilia carriers: a multinational study. Haemophilia. 2016;22:912–918. doi: 10.1111/hae.13089. [DOI] [PubMed] [Google Scholar]
- 15.Den Uijl I.E., Biesma D., Grobbee D., Fischer K. Outcome in moderate haemophilia. Blood Transfus. 2014;12:s330–s336. doi: 10.2450/2012.0091-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Nissen F., Burke T., Asghar S., Ferri Grazzi E., Shang A., Castro F. An insight into clinical outcomes in mild, moderate, and severe hemophilia A (HA): a preliminary analysis of the CHESS II study [abstract] Res Pract Thromb Haemost. 2020;4 [Google Scholar]
- 17.Chai-Adisaksopha C., Noone D., Curtis R., Frick N., Nichol M.B., Germini F. Non-severe haemophilia: is it benign?—insights from the PROBE study. Haemophilia. 2020;27:17–24. doi: 10.1111/hae.14105. [DOI] [PubMed] [Google Scholar]
- 18.Gebhart J., Hofer S., Panzer S., Quehenberger P., Sunder-Plassmann R., Hoermann G., et al. High proportion of patients with bleeding of unknown cause in persons with a mild-to-moderate bleeding tendency: Results from the Vienna Bleeding Biobank (VIBB) Haemophilia. 2018;24:405–413. doi: 10.1111/hae.13422. [DOI] [PubMed] [Google Scholar]
- 19.Reitter S., Streif W., Schabetsberger T., Wozak F., Hartl H., Male C., et al. Austrian Hemophilia Registry: design, development and set of variables. Wien Klin Wochenschr. 2009;121:196–201. doi: 10.1007/s00508-009-1156-0. [DOI] [PubMed] [Google Scholar]
- 20.Rejtő J., Reitter-Pfoertner S., Kepa S., Feistritzer C., Grundbichler M., Hörbst A., et al. Epidemiology and treatment of patients with haemophilia in Austria-update from the Austrian haemophilia registry. Hamostaseologie. 2019;39:284–293. doi: 10.1055/s-0038-1675354. [DOI] [PubMed] [Google Scholar]
- 21.WFH Physical Examination Score (Gilbert Score)—World Federation of Hemophilia. https://www1.wfh.org/docs/en/Publications/Assessment_Tools/Gilbert_Score.pdf [accessed February 18, 2023].
- 22.Rodeghiero F., Tosetto A., Abshire T., Arnold D.M., Coller B., James P., et al. ISTH/SSC bleeding assessment tool: a standardized questionnaire and a proposal for a new bleeding score for inherited bleeding disorders. J Thromb Haemost. 2010;8:2063–2065. doi: 10.1111/j.1538-7836.2010.03975.x. [DOI] [PubMed] [Google Scholar]
- 23.Elbatarny M., Mollah S., Grabell J., Bae S., Deforest M., Tuttle A., et al. Normal range of bleeding scores for the ISTH-BAT: adult and pediatric data from the merging project. Haemophilia. 2014;20:831–835. doi: 10.1111/hae.12503. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Pathare A., Omrani S.A., Hajri F.A., Obaidani N.A., Balushi B.A., Falahi K.A. Bleeding score in type 1 von Willebrand disease patients using the ISTH-BAT questionnaire. Int J Lab Hematol. 2018;40:175–180. doi: 10.1111/ijlh.12761. [DOI] [PubMed] [Google Scholar]
- 25.Harrell F.E. Springer International Publishing; Cham: 2015. Regression modeling strategies: with applications to linear models, logistic and ordinal regression, and survival analysis. [Google Scholar]
- 26.R Core Team (2022). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/ [accessed February 18, 2023].
- 27.Mancuso M.E., Bidlingmaier C., Mahlangu J.N., Carcao M., Tosetto A. subcommittee on factor viii, factor ix and rare coagulation disorders. The predictive value of factor VIII/factor IX levels to define the severity of hemophilia: communication from the SSC of ISTH. J Thromb Haemost. 2018;16:2106–2110. doi: 10.1111/jth.14257. [DOI] [PubMed] [Google Scholar]
- 28.Den Uijl I.E.M., Mauser Bunschoten E.P., Roosendaal G., Schutgens R.E.G., Biesma D.H., Grobbee D.E., et al. Clinical severity of haemophilia A: does the classification of the 1950s still stand? Haemophilia. 2011;17:849–853. doi: 10.1111/j.1365-2516.2011.02539.x. [DOI] [PubMed] [Google Scholar]
- 29.Sood S.L., Cuker A., Wang C., Metjian A.D., Chiang E.Y., Soucie J.M., et al. Similarity in joint function limitation in Type 3 von Willebrand’s disease and moderate haemophilia A. Haemophilia. 2013;19:595–601. doi: 10.1111/hae.12119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Fischer K., van der Bom J.G., Mauser-Bunschoten E.P., Roosendaal G., Beek F.J., de Kleijn P., et al. Endogenous clotting factor activity and long-term outcome in patients with moderate haemophilia. Thromb Haemost. 2000;84:977–980. [PubMed] [Google Scholar]
- 31.Borhany M., Fatima N., Abid M., Shamsi T., Othman M. Application of the ISTH bleeding score in hemophilia. Transfus Apher Sci. 2018;57:556–560. doi: 10.1016/j.transci.2018.06.003. [DOI] [PubMed] [Google Scholar]
- 32.Trossaert M., Lienhart A., Nougier C., Fretigny M., Sigaud M., Meunier S., et al. Diagnosis and management challenges in patients with mild haemophilia A and discrepant FVIII measurements. Haemophilia. 2014;20:550–558. doi: 10.1111/hae.12381. [DOI] [PubMed] [Google Scholar]
- 33.Den Uijl I.E., Bunschoten E.P., Roosendaal G., Schutgens R.E., Biesma D.H., Grobbee D.E., et al. Clinical severity of haemophilia A: does the classification of the 1950s still stand? Haemophilia. 2011;17:849–853. doi: 10.1111/j.1365-2516.2011.02539.x. [DOI] [PubMed] [Google Scholar]
- 34.Kloosterman F.R., Zwagemaker A.-F., Bagot C., Beckers E.A., Castaman G., Cnossen M.H., et al. The bleeding phenotype in people with non-severe hemophilia. Blood Adv. 2022;6:4256–4265. doi: 10.1182/bloodadvances.2022007620. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Beyan C., Kaptan K., Ifran A. Hemophilic arthropathy in non-severe hemophiliacs. Rheumatol Int. 2007;27:501–502. doi: 10.1007/s00296-006-0246-9. [DOI] [PubMed] [Google Scholar]
- 36.Martinoli C., Alberighi O.D.C., Minno G.D., Graziano E., Molinari A.C., Pasta G., et al. Development and definition of a simplified scanning procedure and scoring method for Haemophilia Early Arthropathy Detection with Ultrasound (HEAD-US) Thromb Haemost. 2013;109:1170–1179. doi: 10.1160/TH12-11-0874. [DOI] [PubMed] [Google Scholar]
- 37.De la Corte-Rodriguez H., Rodriguez-Merchan E.C., Alvarez-Roman M.T., Martin-Salces M., Rivas-Pollmar I., Jimenez-Yuste V. Arthropathy in people with mild haemophilia: exploring risk factors. Thromb Res. 2022;211:19–26. doi: 10.1016/j.thromres.2022.01.010. [DOI] [PubMed] [Google Scholar]
- 38.Zwagemaker A.-F., Kloosterman F.R., Hemke R., Gouw S.C., Coppens M., Romano L.G.R., et al. Joint status of patients with nonsevere hemophilia A. J Thromb Haemost. 2022;20:1126–1137. doi: 10.1111/jth.15676. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Fischer K., van der Bom J.G., Mauser-Bunschoten E.P., Roosendaal G., van den Berg H.M. Effects of haemophilic arthropathy on health-related quality of life and socio-economic parameters. Haemophilia. 2005;11:43–48. doi: 10.1111/j.1365-2516.2005.01065.x. [DOI] [PubMed] [Google Scholar]
- 40.Srivastava A., Santagostino E., Dougall A., Kitchen S., Sutherland M., Pipe S.W. WFH guidelines for the management of hemophilia. 3rd ed. Haemophilia. 2020;26:1–158. doi: 10.1111/hae.14046. [DOI] [PubMed] [Google Scholar]
- 41.Turk D.C., Fillingim R.B., Ohrbach R., Patel K.V. Assessment of psychosocial and functional impact of chronic pain. J Pain. 2016;17:T21–T49. doi: 10.1016/j.jpain.2016.02.006. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.