Interventions for preventing silent cerebral infarcts in people with sickle cell disease

  • Protocol
  • Intervention



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

To evaluate the effectiveness of interventions (blood transfusion therapy, hydroxyurea and HSCT) for primary or secondary prevention of SCI in patients with SCD.


Description of the condition

Sickle cell disease (SCD) is a recessive genetic disorder determined by a nucleotide substitution at the 17th position of the β-globin gene and, consequently, the sixth amino acid in the β-globin chain changes from glutamic acid to the hydrophobic amino acid, valine. The hemoglobin that results from this change is called hemoglobin S (HbS) and, when oxygen concentration is low, it polymerizes and the red blood cells change to a sickle shape. This process results in vaso-occlusion, tissue hypoxia, red blood cell destruction (hemolysis) and chronic inflammation. People with SCD have recurring episodes of pain (due to vaso-occlusion and hypoxia) and are at risk for organ damage over their lifetime (Powars 2005). In developed countries, the survival of children with SCD is no longer the major concern, since neonatal screening programs and the high standard of care provided have increased life expectancy (Almeida 2001; Quinn 2004). Morbidity has now become the greater concern.

Although ischemia reperfusion injury can occur in any organ, one of the most overwhelming is that of cerebral injury, such as silent cerebral infarct (SCI) and overt stroke. Stroke is an acute event expressed by neurological symptoms, which are caused by the restriction of cerebral blood flow, resulting in ischemia (Wang 2007). The Cooperative Study of Sickle Cell Disease (CSSCD) defined SCI in children as lesions of abnormal hyperintensity on cerebral T2-weighted magnetic resonance imaging (MRI) in patients without any history of focal neurological symptoms, or abnormalities on neurologic examination in that particular vascular territory (Moser 1996). Recently, the definition became more precise, considering lesions that are greater than 3 mm in diameter and that appear in at least two views of T2-weighted MRI images (Casella 2010). In the adult population, the definition of SCI must be more restrictive, as adults accumulate T2 hyperintensities with time. The definition in adults considers lesions of abnormal hyperintensity on cerebral T2 weighted MRI that are greater than 5 mm and that show a corresponding hypointensity on the T1-weighted image (Vichinsky 2010).

In people with SCD, SCI is the most common form of neurologic disorder, occurring in 27% of children before six years of age (Kwiatkowski 2009) and in 37% by the time they reach 14 years old (Bernaudin 2011). Males with SCD are more likely to experience SCI, other risk factors include lower baseline hemoglobin concentration and higher baseline systolic blood pressure (DeBaun 2012a). The pathophysiology of stroke in SCD is characterized by the co-existence of macro- and microvascular abnormalities. Multiple mechanisms are thought to be involved, such as proliferation of intima and fibrosis that is amplified by inflammation associated with a hypercoagulable state and vascular tone dysregulation. Although evidence about the pathophysiology of SCI is still lacking, it is thought to be a small vessel disease, since it is not associated with large vessels stenosis (Wang 2007). A SCI occurs preferentially in the border zones of the brain, suggesting the participation of hemodynamic factors in its pathogenesis (DeBaun 2012b).

Patients with SCI do not have any obvious symptoms, but they show significant cognitive deficits when compared with people who have a normal brain MRI scan (Armstrong 1996). Identifying patients with SCI is an important issue, given that the cost of routine cerebral MRI scans means this option is not feasible for most healthcare services and there is an additional risk for children who require sedation. Neuropsycologic testing, especially those tests focused on attention and executive functions, may be useful as a screening tool to identify people with SCI (DeBaun 1998). Patients with altered tests should then undergo an MRI scan of the brain to confirm the diagnosis.

The occurrence of SCI leads to an increased risk for further progression of neurological disease, including new or progressive SCI lesions on MRI scans and a 14-fold increase in the incidence of overt strokes (Miller 2001). Despite the high prevalence, little is known about the causes of SCI and the effectiveness of therapies for its prevention.

Description of the intervention

At present, no standard intervention is available for the primary or secondary prevention of SCI.

In chronic blood transfusion therapy 10 mL to 15 mL per kg or one international unit of packed red blood cells are given per transfusion, to maintain the hemoglobin S at < 30%. Depending on the evaluation of the pre-transfusional hemoglobin, simple transfusion or partial exchange transfusions can be performed.

Treatment with hydroxyurea (hydroxicarbamide, an oral use of the chemotherapeutic agent) is administered once per day for at least four days each week. The dose ranges from 15 mg/kg of body weight per day until the maximal tolerated dose for each individual patient is reached.

Allogeneic hematopoietic stem-cell transplantation (HSCT) consists of the transplantation of a graft from a sibling or unrelated donor with stem cells derived from the bone marrow or umbilical cord blood, after any myeloablative conditioning regimen.

How the intervention might work

Chronic blood transfusion therapy is an attempt to dilute the sickled red blood cells, thus reducing the complications of the disease. The main objective of transfusion therapy is to keep the hemoglobin S at a low level. This in turn increases the oxygen carrying capacity and decreases the sickling of red blood cells, reducing the intravascular hemolysis and the free hemoglobin that consumes nitric oxide (Quattlebaum 1986). The transfusion also raises total hemoglobin levels, which leads to a lower flow velocity, an increase in red cell mass and the reduction of the adhesion to the endothelium. It has long been used to prevent strokes in patients with SCD (Sarnaik 1979). The 'Stroke Prevention Trial in Sickle Cell Anemia (STOP)' studying children with altered transcranial doppler (TCD) tests observed a significant reduction of stroke episodes in children maintained in chronic blood transfusion compared with children with no transfusion treatment. This study led to recommendations for TCD screening and prophylactic chronic blood transfusion therapy for children with abnormal TCD results (Adams 1998).

Data from the 'Optimizing Primary Stroke Prevention in Sickle Cell Anemia (STOP II) Trial' suggest that the discontinuation of transfusions in children who reverted to low risk on TCD after long-term transfusion would lead to progression or development of SCI. At the end of the study new brain MRI lesions were diagnosed in 8.1% of the patients who continued the transfusions versus 27.5% of the patients who stopped the therapy (P = 0.03). This result suggests that transfusion was protective for those patients, with an absolute SCI risk reduction of 19.4% (Abboud 2011). However, when studying a population of children receiving chronic blood transfusion therapy for secondary stroke prevention, new or enlarging cerebral infarcts occurred in 45% of patients and progressive cerebral vasculopathy in 38% (Hulbert 2011).

Although chronic blood transfusion is a potential therapy for preventing SCI, significant side effects must be considered, such as the risks of alloimmunization, the transmission of viral or bacterial infections, allergic reactions and iron overload.

Hydroxyurea is a cytostatic myelosuppressive drug initially used to treat cancer. Since the 1980s it has also been used to treat adult patients with SCD due to its ability to increase the fetal hemoglobin levels and decrease the reticulocyte, white blood cell and platelet count (Charache 1992; Platt 2008). In 1996 pediatric patients began to use this agent with similar laboratory results. The patients using hydroxyurea had a reduction in the number and duration of hospitalizations (Fester 1996). The treatment with hydroxyurea also led to fewer pain episodes and decreased episodes of acute chest syndrome (Charache 1995; Hankins 2005). The Belgian experience also suggests that hydroxyurea may be effective in primary or secondary stroke prevention (Gulbis 2005). Several studies have proven that this drug is safe and well-tolerated in adults and children (Jayabose 1996; Scott 1996; Zimmerman 2004). The most frequent adverse events are mild hematologic toxicities (thrombocytopenia, leukopenia (including neutropenia), reticulocytopenia and pancytopenia), which rarely lead to cessation of treatment (de Montalembert 2006).

Allogeneic HSCT is the only curative treatment for SCD, therefore it is a potential option for the primary or secondary prevention of SCI (Apperley 1993; Bernaudin 2007). It has ben reported that HSCT in SCD is successful in 85% to 90% of patients (Locatelli 2012).

Why it is important to do this review

It remains unclear whether chronic blood transfusion therapy, hydroxyurea or HSCT are effective in preventing primary or secondary SCI. This review aims to identify and appraise the best quality evidence on medical interventions for primary and secondary prevention of SCI.


To evaluate the effectiveness of interventions (blood transfusion therapy, hydroxyurea and HSCT) for primary or secondary prevention of SCI in patients with SCD.


Criteria for considering studies for this review

Types of studies

All randomized controlled trials and quasi-randomized controlled trials. We will only include quasi-randomized controlled trials if there is sufficient evidence that the treatment and control groups were similar at baseline.

Types of participants

Patients of any age or gender with SCD (SS, SC, Sβ+ thalassemia and Sβ0 thalassemia) proven by electrophoresis and sickle solubility test, high performance liquid chromatography (HPLC), with family studies or DNA tests as appropriate.

Types of interventions

We will analyse trials of patients with SCD that include any of the following interventions to prevent SCI. We will compare each intervention with each other or no intervention.

  1. Chronic blood transfusion therapy for at least one year

  2. Hydroxyurea used in the dose range of 15 mg/kg up to the maximal tolerated dose for each patient once per day for at least four days a week for at least one year

  3. HSCT

Types of outcome measures

Primary outcomes
  1. Presence of new SCI

  2. SCI progression (increase of size measured by brain MRI in at least two planes of T2-weighted images – axial and coronal)

  3. Time to clinical overt strokes

Secondary outcomes
  1. Preservation of cognitive function (evaluated, when possible, by a validated test (e.g. the Wechsler Adult Intelligence Scale-third edition (WAIS-III), Wechsler Abbreviated Scale of Intelligence (WASI), Wechsler Preschool and Primary Scale of Intelligence-III (WPPSIII),  Wechsler Memory Scale third edition (WMS-III), Tests of Everyday Attention (TEA), Test of Variables of Attention (TOVA), Wisconsin Card Sorting Test, computer version 4 (WCST-CV4), California Verbal Learning Test for Children (CVLT-C) and Behavior Rating Inventory of Executive Function (BRIEF))

  2. Quality of life (evaluated, when possible, by a validated questionnaire (e.g. SIMS (Sims 1971) and PEDSQL (Varni 1999))

  3. Mortality

Search methods for identification of studies

Electronic searches

We will identify relevant studies from the Cystic Fibrosis and Genetic Disorders Group's Haemoglobinopathies Trials Register using the terms: (sickle cell OR (haemoglobinopathies AND general)) AND stroke.

The Haemoglobinopathies Trials Register is compiled from electronic searches of the Cochrane Central Register of Controlled Trials (CENTRAL) (updated each new issue of The Cochrane Library) and quarterly searches of MEDLINE. Unpublished work is identified by searching the abstract books of five major conferences: the European Haematology Association conference; the American Society of Hematology conference; the British Society for Haematology Annual Scientific Meeting; the Caribbean Health Research Council Meetings; and the National Sickle Cell Disease Program Annual Meeting. For full details of all searching activities for the register, please see the relevant section of the Cochrane Cystic Fibrosis and Genetic Disorders Group Module.

We will also search for relevant studies registered at

Searching other resources

We will check reference lists of articles, reviews and textbooks for possible relevant studies. Whenever necessary, we will also contact other researchers or experts in the field for unpublished studies.

Data collection and analysis

Selection of studies

Two authors (JSRC and GSS) will independently assess and select potentially eligible studies for inclusion in the review. We will resolve any disagreements by discussion and, whenever necessary, discussion with a third review author (JAPB).

Data extraction and management

Two review authors (JSRC and EKMS) will use a piloted data extraction form to independently collect the data. We will resolve any disagreements by discussion with a third review author (JAPB). We will enter data into the Review Manager (RevMan) program (RevMan 2011). When necessary, we will send requests to trial authors for additional information or data. We plan to report outcomes for short-term follow up (one to two years), medium-term follow up (three to four years) and long-term follow up (over five years).

Assessment of risk of bias in included studies

Two review authors (JSRC and EMKS) will independently assess the risk of bias in the included studies. We will resolve any disagreements by discussion with a third review author (MSF). As recommended by the Cochrane Collaboration's 'Risk of bias' tool (Higgins 2011a), we will evaluate selection bias (random sequence generation and allocation concealment), performance bias (blinding of participants and personnel), detection bias (blinding of outcome assessment), attrition bias (incomplete outcome data), selective reporting bias and other sources of bias (including sponsorship bias).

A major risk of another source of bias is the improvement of MRI techniques, which increases the detection of SCI in patients with SCD. Use of a greater magnet strength (3.0 Tesla instead of 1.5 Tesla magnets) can impact on the reported incidence and prevalence of SCI, thus altering the results in clinical trials.

Each domain will be categorized as being at either a 'low', 'high' or 'unclear' (either lack of information or uncertainty over the potential for bias) risk of bias.

Measures of treatment effect

We will analyze dichotomous outcome data using risk ratios (RR) with 95% confidence intervals (CI). We will express continuous outcome data as a mean difference (MD) when two or more studies present data from the same instrument of evaluation. When studies express the same variables through different instruments and different units of measurement, we will use the standardized mean difference (SMD). We will consider 95% CIs for both cases. For time-to-event data (i.e.: time to clinical overt strokes), since the most appropriate way of summarizing it is to use methods of survival analysis and express the intervention effect as a hazard ratio (HR) (with 95% CIs), these data will be taken directly from the results of the studies (Higgins 2011a).

Unit of analysis issues

We anticipate that the unit of randomization in the presented studies is the individual participant. However, since the primary outcome may occur in a participant more than once, we will perform this analysis using a rate ratio, comparing the rate of events in the two groups (Higgins 2011a).

For trials with a cross-over design, we plan to use only first-arm data (before participants have crossed over to the second treatment).

For cluster-randomized trials we will check these for unit of analysis errors based on the advice given in Chapter 16 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011b).

Dealing with missing data

Where possible, we will perform an intention-to-treat analysis including all patients randomized to any intervention. We will attempt to contact authors of included studies if there are key missing data, such as the number of events or patients, means or SDs. If necessary, we will impute any missing data with replacement values. For dichotomous outcomes we will assume that the missing data represent treatment failures and for continuous outcomes we will impute the mean observed value. We will perform sensitivity analyses excluding the participants with missing data to assess the strength of the results. If it is not possible to impute missing data, we will consider the potential impact on the findings of the review.

Assessment of heterogeneity

We will identify heterogeneity by visual inspection of the forest plots and by using a standard Chi2 test with a significance level of α = 0.1. Inconsistency among the pooled estimates will be quantified using I2 ((Q - df)/Q) x 100%, where Q is the Chi2 statistic and df represents the degree of freedom. This illustrates the percentage of the variability in effect estimates resulting from heterogeneity rather than sampling error (Higgins 2011a). The thresholds for the interpretation of I2 will be represented as follows:

  • 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 10 or more studies in the review, we plan to assess reporting biases and small study effects by drawing a funnel plot (study effect versus study size).

Data synthesis

We will use the RevMan software to carry out quantitative analysis, based on the intention-to-treat principle (RevMan 2011). If the pooling of results is clinically appropriate and we do not find substantial statistical heterogeneity, we will use the fixed-effect model; however, if we identify substantial statistical heterogeneity we will use the random-effect model.

Subgroup analysis and investigation of heterogeneity

Where sufficient data are available and we identify statistical heterogeneity, we will conduct subgroup analyses to explore the outcomes of the treatments in patients:

  • who had a previous SCI compared to those who have not;

  • with normal TCD compared to conditional TCD compared to abnormal TCD;

  • with SS and Sβ0 thalassemia compared to patients with SC and Sβ+ thalassemia.

We will investigate whether the results of subgroups are significantly different by inspecting the overlap of CIs, and performing the test for subgroup differences and noting the I² statistic available in RevMan (RevMan 2011).

Sensitivity analysis

We will assess the robustness of results performing sensitivity analyses using fixed-effect versus random-effects assumptions, the inclusion or exclusion of studies with an overall high risk of bias or those studies that were not blinded.

We will perform sensitivity analyses excluding the participants with missing data to assess the strength of the results.

Contributions of authors

 Roles and responsibilities
Protocol stage: draft the protocolJSRC, JAPB, EMKS, PD
Review stage: select which trials to include (2 + 1 arbiter)JSRC, JAPB, GSS
Review stage: extract data from trials (2 people)JSRC, EMKS
Review stage: enter data into RevManJSRC, EMKS, PD
Review stage: carry out the analysisJSRC, EMKS
Review stage: interpret the analysisJSRC, JAPB, MSF, EMKS, GSS
Review stage: draft the final reviewJSRC, JAPB, MSF, EMKS
Update stage: update the reviewJSRC, JAPB, MSF, EMKS, GSS

Declarations of interest

None known