GRADE evidence summary assessing the effect of normalization of VWF levels with age
Certainty assessment . | Impact . | Certainty . | ||||||
---|---|---|---|---|---|---|---|---|
Studies, n . | Study design . | Risk of bias . | Inconsistency . | Indirectness . | Imprecision . | Other considerations . | ||
Age change in VWF:Ag | ||||||||
5 | Observational | Serious* | Serious† | Serious‡ | Not serious | None | 5 studies with 1142 patients reported the change in VWF levels longitudinally (follow-up between 1 and 10 y). The mean change in VWF was 7.9 IU/dL per decade (range, 3.0-24.0). | ⨁◯◯◯ Very low |
Frequency of normalization of VWF levels | ||||||||
4 | Observational | Serious* | Serious§ | Serious‡ | Not serious | None | 4 studies with 435 patients reported the normalization of VWF levels over a period of 1-10 y. The number of patients with normalized levels ranged from 25-60%, with a weighted average of 43%. | ⨁◯◯◯ Very low |
Bleeding with normalization of levels | ||||||||
1 | Observational | Not serious | Not serious | Not serious | Not serious | None | Binary logistic regression analysis with bleeding in the year prior to inclusion in the WiN study as a dependent variable. After adjusting for age, sex, BMI, and the presence of any relevant comorbidities (hypertension, cancer, diabetes, and thyroid dysfunction), normalization of VWF levels > 0.50 was still not associated with the incidence of bleeding requiring treatment in the year prior to inclusion in the study (odds ratio, 1.26; 95% CI, 0.72-2.21; P = .414). We can conclude that, even after taking other important factors that influence VWF levels and bleeding into account, normalization of VWF levels is not associated with a lower incidence of bleeding episodes requiring hemostatic treatment. 27% of patients with normalized levels had bleeding symptoms at the time of the study, and 21% of patients with abnormal levels had bleeding symptoms. | ⨁⨁◯◯ Low |
BS in patients with normalized levels | ||||||||
2 | Observational | Serious* | Not serious | Serious¶ | Not serious | None | Nummi et al15 showed that the mean BS in patients with a confirmed diagnosis ranged between 10 and 24. Mean BS in patients with a diagnosis of low VWF and those with normal VWF levels was 6. Including all patients with historical VWD, BS showed a weak and negative correlation with VWF:RCo (r = +0.43), VWF:Ag (r = +0.51), VWF:CB (r = +0.54), FVIII (r = +0.44), RIPA, 0.6 mg/mL (r = +0.34), and RIPA, 0.8 mg/ mL (r = +0.54) and a positive correlation with PFA C/EPI (r = +0.45) and C/ADP (r = +0.46) (P ≤ .001 for all). Sanders et al26 showed that BS did not differ between elderly and younger patients. | ⨁◯◯◯ Very low |
Certainty assessment . | Impact . | Certainty . | ||||||
---|---|---|---|---|---|---|---|---|
Studies, n . | Study design . | Risk of bias . | Inconsistency . | Indirectness . | Imprecision . | Other considerations . | ||
Age change in VWF:Ag | ||||||||
5 | Observational | Serious* | Serious† | Serious‡ | Not serious | None | 5 studies with 1142 patients reported the change in VWF levels longitudinally (follow-up between 1 and 10 y). The mean change in VWF was 7.9 IU/dL per decade (range, 3.0-24.0). | ⨁◯◯◯ Very low |
Frequency of normalization of VWF levels | ||||||||
4 | Observational | Serious* | Serious§ | Serious‡ | Not serious | None | 4 studies with 435 patients reported the normalization of VWF levels over a period of 1-10 y. The number of patients with normalized levels ranged from 25-60%, with a weighted average of 43%. | ⨁◯◯◯ Very low |
Bleeding with normalization of levels | ||||||||
1 | Observational | Not serious | Not serious | Not serious | Not serious | None | Binary logistic regression analysis with bleeding in the year prior to inclusion in the WiN study as a dependent variable. After adjusting for age, sex, BMI, and the presence of any relevant comorbidities (hypertension, cancer, diabetes, and thyroid dysfunction), normalization of VWF levels > 0.50 was still not associated with the incidence of bleeding requiring treatment in the year prior to inclusion in the study (odds ratio, 1.26; 95% CI, 0.72-2.21; P = .414). We can conclude that, even after taking other important factors that influence VWF levels and bleeding into account, normalization of VWF levels is not associated with a lower incidence of bleeding episodes requiring hemostatic treatment. 27% of patients with normalized levels had bleeding symptoms at the time of the study, and 21% of patients with abnormal levels had bleeding symptoms. | ⨁⨁◯◯ Low |
BS in patients with normalized levels | ||||||||
2 | Observational | Serious* | Not serious | Serious¶ | Not serious | None | Nummi et al15 showed that the mean BS in patients with a confirmed diagnosis ranged between 10 and 24. Mean BS in patients with a diagnosis of low VWF and those with normal VWF levels was 6. Including all patients with historical VWD, BS showed a weak and negative correlation with VWF:RCo (r = +0.43), VWF:Ag (r = +0.51), VWF:CB (r = +0.54), FVIII (r = +0.44), RIPA, 0.6 mg/mL (r = +0.34), and RIPA, 0.8 mg/ mL (r = +0.54) and a positive correlation with PFA C/EPI (r = +0.45) and C/ADP (r = +0.46) (P ≤ .001 for all). Sanders et al26 showed that BS did not differ between elderly and younger patients. | ⨁◯◯◯ Very low |
BMI, body mass index; BS, bleeding score; C/ADP,Cartridge with collagen and adenosine diphosphate; C/EPI, Cartridge with collagen and epinephrine; PFA, Platelet Function Analyser; RIPA, RIPA, ristocetin-induced platelet aggregation; VWF:CB, VWF collagen binding assay.
Serious study confounding occurred because the investigators did not adjust for comorbidities, with the exception of Sanders et al26 .. In their study, more elderly patients reported ≥1 comorbidity, including diabetes, cancer, cardiovascular disease, and depression, compared with younger patients. Atiq, 2018 showed that comorbidities are associated with higher levels of VWF and FVIII in type 1 VWD and may explain the age-related increase in VWF and FVIII levels.
The change in VWF levels varies between 3.0 and 24 IU/dL per decade, leading to serious inconsistency.
Although the change in VWF levels is presented, the bleeding symptoms of patients with normalized levels is not reported in the studies.
The normalization of VWF levels varies between 25% and 60%, leading to serious inconsistency.
The BS does not predict the bleeding symptoms in patients in normal VWF levels but informs on the bleeding history in those patients.