Interventions for treating acute bleeding episodes in people with acquired hemophilia

  • Protocol
  • Intervention



This is the protocol for a review and there is no abstract. The objectives are as follows:

To determine the efficacy and safety of different hemostatic therapies for acute bleeds in AH.


Description of the condition

Acquired hemophilia (AH) is a rare, but severe, hemorrhagic disorder. It is characterized by either a deficiency in the function of coagulation factor VIII (FVIII) or the targeting of specific epitopes (that cause neutralization or accelerated clearance of FVIII from the plasma) by autoantibodies, or a combination of both (Franchini 2008). This condition, which affects both sexes (Buczma 2007), occurs in the general population with a reported incidence of one to four cases per million per year, increasing with age (Collins 2007a; Delgado 2003; Franchini 2008).

The etiology of AH remains unknown. It is commonly associated with a variety of clinical conditions, including autoimmune disorders, solid tumors, lymphoproliferative diseases, and pregnancy; however, in approximately 50% of patients AH occurs apparently in the absence of other concomitant disease, i.e. FVIII autoantibodies are of an idiopathic origin (Baudo 2004; Baudo 2007; Collins 2007a; Delgado 2003; Green 1981; Franchini 2008a; Meiklejohn 2001).

The diagnosis of AH is typically based on:

  1. a negative personal or family history of bleeding events or abnormal coagulation assays until the present episode;

  2. the initial detection of an isolated prolongation of activated partial thromboplastin time (APTT), which cannot be corrected by incubating for two hours at 37°C equal volumes of patient plasma and normal plasma (mixing study) (Verbruggen 1995); and

  3. subsequent identification of a reduced FVIII:C level with evidence of FVIII inhibitor activity (detected and measured by the Bethesda assay or its Nijmegen modification) (Baudo 2010; Coppola 2009; Delgado 2003).

Hemorrhages in AH patients usually occur suddenly and spontaneously, although approximately 25% of cases occur after trauma or invasive procedures (Baudo 2003). While bleeding at presentation is usually severe or life-threatening, requiring hemostatic treatment and transfusion; it can also be mild, with approximately 25% of patients not requiring hemostatic treatment (Baudo 2004; Franchini 2008). The mortality rate resulting from bleeding is high and reported to be between 7.9% and 22%. This is particularly true in elderly patients and during the first weeks after the onset of symptoms for a number of reasons. These include underlying associated diseases, diagnostic delays, inadequate treatment of acute bleeds, bleeding complications during invasive procedures for controlling hemorrhages, or adverse events of treatment (Baudo 2004; Collins 2007a; Green 1981; Franchini 2008).

Observed clinical bleeding does not correlate with FVIII level or inhibitor titer and differs from hereditary hemophilia. Bleeding into the skin, muscles or soft tissues, and mucous membranes (e.g. epistaxis, gastrointestinal and urologic bleeds, retroperitoneal hematomas, postpartum bleeding) are more common, whereas bleeding into joints (hemarthrosis), a typical feature of congenital factor VIII deficiency, is unusual (Boggio 2001; Green 1981).

Description of the intervention

The therapeutic strategy for patients with AH involves two aspects of treatment:

  1. to control actual bleeding episodes (hemostatic agents); and

  2. to eradicate FVIII autoantibodies (immunosuppressive treatment), the cause of coagulation abnormalities and of bleeding (Baudo 2010; Baudo 2012; Collins 2012; Coppola 2009; Franchini 2008; Huth-Kühne 2009; Knoebl 2012; Tengborn 2012).

Available therapeutic interventions to control bleeding in patients with AH are: bypassing agents (recombinant activated FVII (rFVIIa) (NovoSeven®) or activated prothrombin complex concentrate (aPCC) (FEIBA®)); or FVIII replacement therapy with human or porcine FVIII concentrates; or induction of FVIII release using 1-desamino-8-D-arginine-vasopressin (DDAVP) (Baudo 2010; Baudo 2012; Coppola 2009; Franchini 2008; Huth-Kühne 2009; Knoebl 2012; Tengborn 2012).

Bypassing agents are the most commonly used first-line treatment, particularly for severe bleeds in patients with high-titer inhibitors. Both activated prothrombin complex concentrate (aPCC) and recombinant activated factor VII (rFVIIa) have been shown to be effective in these patients (Collins 2007b; Franchini 2008; Sallah 2004). Treatment with rFVIIa is administered as a bolus injection every two to six hours until hemostasis is achieved at average doses of 90 ug/kg (range 40 to 180 ug/kg) (Hay 1997; Huth-Kühne 2009; Franchini 2008; Sallah 2004; Sumner 2007); and aPCC is administered as a bolus injection every 8 to 12 hours to a maximum of 200 IU/kg/day (with the standard dose ranging between 50 and 100 IU/kg) (Franchini 2008; Huth-Kühne 2009; Sallah 2004).

Therapy to replace FVIII can only be advocated for the treatment of acute bleeding in patients with low-titer inhibitors (i.e. less than five Bethesda units (BU)) and non-severe bleeds (Baudo 2010; Huth-Kühne 2009). Patients with low-titer inhibitors can be treated with plasma-derived or recombinant human FVIII concentrates, which should be administered at doses sufficient to overwhelm the inhibitor and thus achieve hemostatic levels of factor VIII. Human FVIII concentrates are usually administered as an intravenous bolus dose of 20 IU/kg for each BU of the inhibitor plus an additional 40 IU/kg, with monitoring of FVIII activity (FVIII:C) levels 10 minutes after the initial dose, and the subsequent administration of an additional bolus dose if the incremental recovery is not adequate (Collins 2007b; Franchini 2008; Huth-Kühne 2009 ). The use of DDAVP releases FVIII/von Willebrand factor from the vascular endothelium; it is administered subcutaneously at a dose of 0.3 g/kg per day for three to five days (Franchini 2008; Mudad 1993).

How the intervention might work

In pharmacological concentrations, rFVIIa can bind to the surface of activated platelets and directly activate factor X (FXa) in the absence of tissue factor (TF). The platelet surface FXa can then, in complex with activate factor V (FVa), lead to a thrombin burst in the absence of FVIII or FIX. As more thrombin is generated, positive feedback loops with FV, FVIII, and factor XI (FXI ) occur. Cross-linked fibrin is formed by thrombin activation of factor XIII (FXIII) to FXIIIa, producing a more stable, covalently linked clot, protected from degradation by the thrombin-activatable fibrinolysis inhibitor (TAFI) (Abshire 2004; Ananyeva 2009).

The aPCC contains the proenzymes of the prothrombin complex factors, prothrombin, FVII, FIX and FX, but with only very small amounts of their activation products; with the exception of FVIIa, which is contained in aPCC in greater amounts. The aPCC controls bleeding by the induction and facilitation of thrombin generation (prothrombin is converted into thrombin), by FXa on a phospholipid surface mediated by calcium, but only if FV or its activated form, FVa, is present (Turecek 2004).

Treatment with DDAVP increases the plasma concentration of factor VIII and vWF through the endogenous release of the patient's own stores (Baudo 2010; Lethagen 1987). By increasing the concentration of the FVIII to values within the normal range, human FVIII concentrates or DDAVP administration normalizes the haemorrhagic risk or stops the bleeding in responsive patients (Baudo 2010; Lethagen 1987).

Why it is important to do this review

Acute bleeding is a common clinical problem in AH patients and the optimal therapy is controversial. At this stage, no systematic review or meta-analysis of hemostatic therapies in AH patients is available. We are aiming to obtain more evidence regarding the clinical benefit  by systematically analysing and assessing the reliability and validity of the data and by considering only randomised controlled trials (RCTs) and quasi-randomised controlled trials for our review.


To determine the efficacy and safety of different hemostatic therapies for acute bleeds in AH.


Criteria for considering studies for this review

Types of studies

Randomised controlled trials (RCTs) and quasi-randomised controlled trials.

Types of participants

We will include patients with AH, with no restrictions on gender, age or ethnicity.

Types of interventions

We will consider the following interventions:

  • Experimental intervention: first-line hemostatic therapy (bypassing agent (recombinant factor VIIa (rFVIIa) or activated prothrombin complex concentrate (aPCC)), replacement therapy (factor VIII (FVIII) or desmopressin (DDAVP));

  • Comparator intervention: aPCC, fresh frozen plasma, a different experimental intervention (e.g. rFVII versus FEIBA, or FEIBA versus DDAVP). 

We will consider the following comparisons:

  1. any hemostatic therapy, alone or in combination, versus no treatment

  2. bypassing agent (rFVIIa or aPCC) versus replacement therapy (FVIII or DDAVP)

  3. rFVIIa versus aPCC

  4. FVIII versus DDAVP

Types of outcome measures

Primary outcomes
  1. Bleeding control (the response to therapy judged by a clinician as bleeding resolved with date or bleeding not resolved)

Secondary outcomes
  1. Number of participants with adverse effects (thromboses; allergic reactions)

  2. Overall survival (OS) (defined as the time interval from randomisation or study entry to death from any cause or to last follow up)

Search methods for identification of studies

We will include both full-text and abstract publications (if sufficient information is available on study design, characteristics of participants, interventions and outcomes).

Electronic searches

Search strategies have been adapted from those suggested in the Cochrane Handbook for Systematic Reviews of Interventions (Lefebvre 2011). No language restriction will be applied.

The following databases of medical literature will be searched:

  • the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library, latest issue);

  • MEDLINE (Ovid) (1948 to present);

  • Embase (1948 to present).

Searching other resources

We will search the following conference proceedings from 2000 to present, if they are not included in CENTRAL:

  • American Society of Hematology;

  • European Hematology Association;

  • International Society on Thrombosis and Haemostasis (ISTH) ;

  • European Association for Haemophilia and Allied Disorders (EAHAD).

We will also search clinical trial registers for ongoing trials:

Data collection and analysis

Selection of studies

Two authors (ZY, ZR) will independently apply the selection criteria to determine the trials to be included in the review. If a disagreement occurs, we will resolve this by discussion with a third person (LD) (Higgins 2011a).

Data extraction and management

Two review authors (ZY, ZR) will independently extract data according to chapter 7 of the Cochrane Handbook for Systematic Reviews of Interventions by using a standardized data extraction form containing the following items (Higgins 2011a).

  • General information: title; authors; journal or source; contact address; country of origin; language; publication type; year of publication; setting of trial.

  • Trial characteristics including: design; sample size; setting; location of trial; randomization method; concealment of allocation; blinding of patients and clinicians; withdrawals; median length of follow up; funding; conflict of interest statement.

  • Study interventions (basic information): disease(s) and stage(s) studied; category of treatment investigated; inclusion criteria; exclusion criteria; experimental intervention; control intervention; type of control; additional treatment; compliance; outcomes assessed; subgroup evaluated; confounders.

  • Baseline characteristics of patients: number of patients; age; ethnicity; gender; diagnosis; definition of diagnosis; extent of disease; organ involvement; additional diagnoses in group; stage; previous treatment; concurrent conditions.

  • Interventions: setting; dose and duration of hemostatic therapy; supportive treatment; additional treatment.

  • Outcomes: OS; bleeding control; adverse events.

We plan to combine data at the following time-points: up to three months of follow up and over three months and up to two years of follow up.

If necessary, we will contact the principal trial investigators to clarify data and obtain any additional information needed.

Assessment of risk of bias in included studies

Two authors (ZY and ZR) will independently assess the risk of bias of each included trial as per the recommendations in chapter 8 of the Cochrane Handbook for Systematic Reviews of Interventions using the following criteria (Higgins 2011b):

  • sequence generation;

  • allocation concealment;

  • blinding;

  • incomplete outcome data;

  • selective outcome reporting;

  • other sources of bias.

We will assess each criterion as having either a low, unclear or high risk of bias. We will resolve any disagreements by discussion with a third author (LD).

Measures of treatment effect

We will use Review Manager 5.2 to conduct the analysis (RevMan 2013). For dichotomous (binary) outcomes, we will use the risk ratio (RR) with 95% confidence intervals (CI) for the effect measure. For continuous outcome data, we will use the mean difference (MD) with 95% CIs, except where continuous data are reported using different units. In such cases we will calculate a standardised mean difference (SMD) and corresponding CIs. For survival data, we will estimate the treatment effects of individual studies as hazards ratios (HR) using the methods described by Parmar and Tierney (Parmar 1998; Tierney 2007).

Unit of analysis issues

We will include data from any eligible cross-over trials in the review; we plan to analyse these using a method recommended by Elbourne (Elbourne 2002).

Dealing with missing data

If data are missing or unavailable, we will contact the corresponding author of trials to obtain further information. This may include missing outcomes (primary or secondary), missing participants due to dropout, and missing statistics such as standard deviations or correlation coefficients. If this is not possible and if we assume that these data are 'missing at random', we will perform an available case analysis (i.e. ignoring the missing data) and discuss the impact of missing data on our results. When we assume the missing outcome data are not 'missing at random', we will conduct an intention-to-treat (ITT) analysis by imputing the missing data with replacement values, and treating these as if they were observed (e.g. last observation carried forward, imputing an assumed outcome such as assuming all were poor outcomes, imputing the mean, imputing based on predicted values from a regression analysis). We will perform a sensitivity analysis by including or excluding trials with significant dropouts to assess how sensitive results are to reasonable changes in the assumptions that are made (Higgins 2011c).

Assessment of heterogeneity

We will assess heterogeneity among trials by inspecting the forest plots and using the Chi2 test and I2 statistic for heterogeneity with a statistical significance level of P < 0.10 and interpret I2 as follows (Higgins 2003):

  • 0% to 40%: might not be important;

  • 30% to 60%: may represent moderate heterogeneity;

  • 50% to 90%: may represent substantial heterogeneity;

  • 75% to 100%: considerable heterogeneity.

Assessment of reporting biases

If we include at least 10 trials in any meta-analysis, we will explore potential publication bias by generating a funnel plot and statistically testing by means of a linear regression test. We will consider P < 0.1 as significant for this test (Sterne 2011). If we detect asymmetry, we will explore causes other than publication bias.

Data synthesis

We will perform analyses according to the recommendations of chapter 9 of the Cochrane Handbook for Systematic Reviews of Interventions (Deeks 2011). We will use aggregated data for analysis. For statistical analysis, we will enter data into Review Manager 5.2 (RevMan 2013). One review author will enter data into the software and a second review author will check it for accuracy. We will perform meta-analyses using a fixed-effect model (e.g. the generic inverse variance method for survival data outcomes, the inverse variance method for continuous outcomes, and the Mantel-Haenszel method for dichotomous data outcomes). We will use the random-effects model in terms of sensitivity analyses. We will explore causes of heterogeneity by subgroup analyses.

Subgroup analysis and investigation of heterogeneity

We will consider performing subgroup analyses on the following characteristics, if appropriate:

  1. severe bleeds versus non-severe bleeds;

  2. high-titer inhibitors versus low-titer inhibitors.

Sensitivity analysis

We plan to undertake the following sensitivity analyses if possible:

  1. quality components, including full text publications or abstracts, preliminary results versus mature results;

  2. random-effects modelling;

  3. imputation of missing values.


Many thanks to Cochrane Cystic Fibrosis and Genetic Disorders Group for their help, and to Nikki Jahnke and Tracey Remmington for their expertise and technical support.

Contributions of authors

Yan Zeng and Ruiqing Zhou have made the same contribution to this protocol.

Roles and responsibilities 
Protocol stage: draft the protocolZeng Yan, Zhou Ruiqing, Duan Xin
Review stage: select which trials to include (2 + 1 arbiter)Zeng Yan, Zhou Ruiqing+ Long Dan
Review stage: extract data from trials (2 people)Zeng Yan, Zhou Ruiqing
Review stage: enter data into RevManZeng Yan
Review stage: carry out the analysisZeng Yan, Zhou Ruiqing
Review stage: interpret the analysisZeng Yan, Zhou Ruiqing, Yang Songtao
Review stage: draft the final reviewZeng Yan, Zhou Ruiqing, Duan Xin
Update stage: update the reviewZeng Yan, Zhou Ruiqing

Declarations of interest

None known.