Elsevier

Pediatric Neurology

Volume 34, Issue 6, June 2006, Pages 450-458
Pediatric Neurology

Symposium on pediatric stroke
Trials in Sickle Cell Disease

https://doi.org/10.1016/j.pediatrneurol.2005.10.017Get rights and content

Children with sickle cell disease are at risk of developing neurologic complications, including stroke, transient ischemic attack, seizures, coma, and a progressive reduction in cognitive function. Transcranial Doppler ultrasound, magnetic resonance imaging, and overnight pulse oximetry appear to predict, making prevention an achievable goal so that there is now a focus on randomized controlled trials. The Stroke Prevention Trial in Sickle Cell Anemia (STOP) reported a reduction in the number of overt clinical strokes experienced by those children with critically high transcranial Doppler velocities (>200 centimeters per second) who were chronically transfused. Two additional Phase III studies and two pilot trials have been funded. STOP II focused on whether it is safe to discontinue blood in prophylactically transfused children when their velocities had remained normal for at least 30 months. The Silent Infarct Transfusion trial is designed to determine whether children with sickle cell anemia and silent cerebal infarcts, approximately 20% of the population, will have a decrease in the progressive neurologic complications after receiving regular blood transfusion therapy. Pilot safety and feasibility trials of low-dose aspirin and overnight respiratory support are also beginning. The collaboration provides an infrastructure for future clinical trials in this vulnerable group of children.

Introduction

Stroke is a common occurrence in children with sickle cell disease. In the longitudinal Co-operative Study of Sickle Cell Disease, 25% of patients with homozygous sickle cell anemia and 10% of those with hemoglobin SC-disease had had a cerebrovascular accident by the age of 45 years [1], [2]. Although this burden is obviously large, these data are likely to underrepresent the importance of stroke because some patients with stroke early in childhood do not survive to adulthood. Transient ischemic attacks, seizures, and acute coma are also common in this population [3]. Arterial disease mainly affects the distal internal carotid and proximal middle and anterior cerebral arteries [4], [5]. Sinovenous thrombosis [6], [7], posterior leukencephalopathy [8], and necrotizing encephalitis [9] have also been described. In addition, by adolescence approximately 20% of children with homozygous sickle cell anemia have covert or “silent” infarction on magnetic resonance imaging [10], [11], [12] without having had a clinical stroke. Covert cerebral infarcts are associated with cognitive difficulties [13], [14], [15] which may be progressive [16]. Ischemic infarction peaks in the first two decades of life and from the fourth decade onwards. Hemorrhagic stroke, secondary to aneurysm, sinovenous thrombosis, or moyamoya disease, occurs more commonly in the third decade [2].

Approximately 20% of the children with sickle cell anemia who have had a stroke and receive regular blood transfusion therapy will have a second stroke [17]. Among the group with a second stroke, approximately 30% will have a third stroke despite regular blood transfusion with documented hemoglobins S levels less than 30% [17]. The risk of subsequent strokes appears to be related to at least two factors working either independently or synergistically: not presenting initially in the context of an acute illness [17] and moyamoya collaterals on angiography [18]. There has never been a controlled trial of secondary prevention, but the observational cohort studies in which monthly transfusion has been used suggested a reduction in recurrent stroke from 66-90% to 10-27% [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28]. It would therefore be unethical to conduct a trial of secondary prevention with an untreated arm. There is currently some interest in comparing long-term blood transfusion with changing to hydroxyurea after several years of blood transfusion (Stroke With Transfusions Changing to Hydroxyurea [SWITCH]) [29] in those who have already had a stroke. However, looking at whether this drug has a role in primary prevention in the trial of hydroxyurea in infants (Pediatric Hydroxyurea in Sickle Cell Anemia [BABYHUGS]) [30] has proved difficult because of the requirement to sedate young children for magnetic resonance imaging.

Recent studies suggest that neurologic complications may be predicted using transcranial Doppler ultrasound [31], magnetic resonance imaging [32], or overnight pulse oximetry [3], making prevention an achievable goal for many patients. There is therefore considerable interest in randomized controlled trials for primary prevention of the neurologic complications of sickle cell disease. One has been published [33], and currently several are commencing or are planned, the focus of this review.

The Stroke Prevention Trial in Sickle Cell Anemia (STOP) Trial (principal investigator Dr. Robert Adams of the Medical College of Georgia) recruited from sickle cell centers across the United States. Children aged 2-16 years with sickle cell anemia or sickle beta zero thalassemia and no history of stroke were screened for abnormal transcranial Doppler ultrasound, defined as internal carotid/middle cerebral artery velocity >200 centimeters per second. Three other groups identified were not eligible for randomization: (i) those with normal transcranial Doppler, defined as internal carotid/middle cerebral artery velocity <170 centimeters per second; (ii) those with conditionally abnormal internal carotid/middle cerebral artery velocities, i.e., between 170 centimeters and 200 centimeters per second; and (iii) those with velocity elevation exclusively in the anterior cerebral, posterior cerebral, or basilar arteries. One hundred thirty patients had an abnormal transcranial Doppler ultrasound and were randomly allocated to either blood transfusion or observation. The end point was clinical stroke, defined as a new focal neurologic deficit lasting more than 24 hours. The study was discontinued early and a Clinical Alert issued (http://www.nlm.nih.gov/databases/alerts/sickle97.html) because of the benefit of transfusion in preventing first stroke in children with sickle cell disease [33]. All children with sickle cell anemia or sickle beta zero thalassemia should have transcranial Doppler screening and should be transfused if their velocities are >200 centimeters per second (recommendations of the National Heart, Lung, and Blood Institute) [34]. Preliminary epidemiologic evidence suggests that implementation of transcranial Doppler assessment in routine medical care of children with sickle cell anemia is associated with a reduction in stroke in California [35].

STOP II (www.mcg.edu/neurology/Research/sicklecell/STOPIIresearch.pdf) has followed on from the STOP trial, taking children with sickle cell disease and abnormal transcranial Doppler on screening who have had 30 months or more of regular transfusion on the STOP protocol and have converted from a high-risk to a low-risk transcranial Doppler. These children were screened with magnetic resonance angiography to ensure that they did not have severe lesions and were then randomly allocated to continue transfusion or to discontinue transfusion. They had transcranial Doppler every 3 months and an annual brain magnetic resonance imaging and magnetic resonance angiography. The children were monitored for 2-4 years. There were three likely outcomes for those children. First, they could have developed a stroke, in which case they would have resumed transfusion; second, they might have remained with a low-risk transcranial Doppler ultrasound in which case they would have continued in the study; third, the transcranial Doppler might have reverted to abnormal, in which case they would have resumed transfusion. The data have recently been analyzed and a manuscript published [36]; with a Clinical Alert again issued (http://www.nhlbi.nih.gov/health/prof/blood/sickle/clinical-alert-scd.htm). STOP II was halted early because there were too many reversions to high risk (14 transcranial Doppler end points and 2 others with stroke), all in the halt transfusion arm.

Current policy arising from STOP I and STOP II concludes that children should be screened with transcranial Doppler to find those with velocities over 200 centimeters per second, and those children should then be transfused indefinitely.

Dr. Adams shared with the participants at the conference his experience of the important factors for a successful trial and of using a clinical trials review group (Table 1). Some of the trial design and implementation features of STOP and STOP II, while not unique, may be of interest to triallists in this area. Dr. Adams emphasized the importance of keeping things simple and making absolutely sure that the statistics are appropriate for the study design, as well as the role of clear leadership enabling democratic input. It took some time to learn about the information that was required for STOP and STOP II and how to track it. The investigators had to design new tools, performance summaries, report cards for adverse events, and the like, to ensure fairness and increase acceptance of direction and criticism. It is essential to avoid any split between the clinicians and the data management team. Those involved must therefore spend a lot of time learning each other’s viewpoints and dividing tasks.

Dr. Michael Noetzel, a neurologist from Washington University School of Medicine, outlined the issues involved in designing the Silent Infarct Transfusion (SIT trial) for which Dr. Michael DeBaun, a pediatric hematologist at the same institution, is the Principal Investigator. The definition of a silent infarction is an abnormal magnetic resonance imaging of the brain with increased signal intensity in multiple T2-weighted images and no history or physical findings suggesting a focal neurologic deficit lasting more than 24 hours. Data from three studies in children with sickle cell anemia in the United States, France, and the United Kingdom suggested a prevalence of 15-25% [11], [12], [32]. Evidence has recently been published demonstrating that children with sickle cell disease and silent infarction are at increased risk of both overt stroke [32] and new magnetic resonance imaging lesions [37]. In addition, patients with silent infarction have a lower intelligence quotient when compared with patients with normal magnetic resonance imaging examinations [11], [38], [39], [40], particularly if the lesions are large [40]. The pathophysiology of silent infarction is poorly understood, but most likely is multifactorial. Most lesions occur in the nonmotor areas of the brain, typically the frontal lobes [38], [40]. Pathology and neuroimaging correlation has been limited; however, lesions seem to correlate with focal white matter gliosis, as well as micro-infarcts, with focal perivascular hemorrhage and microscopic evidence that the small precapillary arterioles are dilated and occluded with clumped sickle red cells [41].

The use of blood transfusion as the therapeutic intervention for children with silent cerebral infarcts was selected for several reasons. First, treatment with blood has been demonstrated to be effective for both primary and secondary prevention of strokes, with a relative risk reduction of approximately 85% for both strategies. Second, in a feasibility study, DeBaun’s research group approached 14 children with silent infarcts; 10 were successfully enrolled and received blood transfusion therapy. One patient was taken off-study due to noncompliance, and three patients refused enrollment. All patients enrolled in the study have been transfused on a regular basis to keep the hemoglobin S percentage less than 30%, and magnetic resonance imaging of the brain was performed annually. No progression of silent infarction occurred in the eight patients monitored for more than 24 months of regular transfusion treatment; one moved to a nonparticipating institution (DeBaun MR, personal communication, Washington University, St. Louis, MO, 2003). Another compelling reason as to why blood was chosen as a therapeutic intervention was based on the data from the STOP I trial [42]. In this trial none of the 18 patients receiving transfusions who had a baseline silent lesion on magnetic resonance imaging developed new magnetic resonance imaging lesions or overt strokes over a 36-month follow-up period, whereas of the 29 untreated individuals who had baseline silent infarction on magnetic resonance imaging and were randomized to observation, 52% subsequently developed new magnetic resonance imaging lesions (n = 6) or suffered overt strokes (n = 9). Taken together, these studies strongly suggest that regular blood transfusion therapy will be an effective strategy in preventing progression of silent cerebral infarcts.

The primary hypothesis for the Silent Infarct Transfusion trial is that transfusion therapy in children with silent infarction will result in at least 86% reduction in new overt and silent strokes; there are also two secondary hypotheses: that transfusion therapy will limit the further decline in general intellectual abilities when compared with the observation arm, and that the overall benefit of transfusion therapy for silent infarction outweighs the risks associated with the treatment. This is a multicenter National Institutes of Health–National Institute of Neurological Diseases and Stroke funded clinical study with 24 sites in the United States, Canada, the United Kingdom, and France which will take 6.5 years, with approximately 1800 children screened for silent infarction with magnetic resonance imaging and just over 200 patients randomized to either blood transfusion or observation. Children who have transcranial Doppler velocities >200 centimeters per second will be excluded and offered blood transfusion as standard of care. Neurologic examination will be undertaken after screening and before randomization and then annually and at study exit 36 months after treatment or after any suspected neurologic event. Cognitive assessment using the Wechsler abbreviated scales and the Behavior Rating Inventory of Executive Function measure of executive function will be undertaken postscreening and prerandomization, at 12-18 months and at study exit. Magnetic resonance imaging will be obtained in 1800 patients at screening and also will be repeated at study entry and exit in the randomized patients. If there is an acute neurologic event, there will also be a magnetic resonance imaging; magnetic resonance angiography, and magnetic resonance venography. A unique feature of the study is the collection of biologic samples of 1800 children with sickle cell anemia coupled with complete assessment of neurologic phenotype with magnetic resonance imaging assessment and clinical history. In addition to the Clinical Coordinator Center and the Statistical Coordinator Center, the trial includes a Neurology Imaging Core and a Biological Repository where the neuroimages and deoxyribonucleic acid, respectively, from 1800 study participants will be available for future evaluations.

Although two small trials with a pain end point were negative [43], [44], studies of aspirin for the prevention of neurologic complications in sickle cell disease have been on the agenda for a number of years. Norma Lerner has been developing the Sickle cell Two center Aspirin Response Trial (START) and Sickle cell Multicenter Aspirin Response Trial (SMART) studies from her base at the University of Rochester Medical Center.

Aspirin is an inexpensive, widely used agent that has been used effectively in adult prevention trials. When given to adults early after stroke, aspirin reduces stroke recurrence and mortality [45], [46]. A large trial in women reported a reduction in first stroke [47], although part of the benefit in terms of the reduction in ischemic stroke was counterbalanced by an increase in hemorrhage.

Aspirin impairs platelet function and exerts anti-inflammatory and other ameliorative effects that might combat the processes underlying the cerebrovascular damage associated with sickle cell disease. Thrombocytosis is common in sickle cell disease and in one series was associated with lower intelligence quotient [11]. Furthermore, platelet activation is associated with hypoxia, hypercoagulability, and inflammation [48]. Aspirin may have a beneficial effect on these and other mechanisms [49]. Aspirin compromises platelet function by irreversibly inhibiting cyclooxygenase and blocking thromboxane A2 (Cox 1 inhibition). Recent evidence indicates that low-dose aspirin has anti-inflammatory capabilities that are unrelated to the Cox 2 inhibition that is promoted by higher doses. Aspirin can be administered safely in low dose and has predictable side effects.

Furthermore, it is cheap and might be used in both in the developing as well as the developed world.

The hypothesis for START and SMART is that daily low-dose aspirin therapy will safely diminish the predisposition to overt and silent stroke as well as the progression of cognitive deficit in children with homozygous sickle cell disease and hemoglobin sickle β0- thalassemia. The primary objective of the National Institute of Neurological Diseases and Stroke funded pilot (START) is to evaluate the safety and tolerability of daily low-dose aspirin in children 4 to 10 years of age with sickle cell disease. Secondary objectives include establishing the most useful assessments in a battery of age-appropriate neurocognitive tests and the feasibility of magnetic resonance imaging and magnetic resonance angiography studies. The utility of classification systems and the validity of using trends for group comparisons for magnetic resonance angiography and transcranial Doppler velocities over time will be assessed.

As originally proposed, Sickle Cell Multicenter Response Trial (SMART) was a Phase III randomized, placebo-controlled, double-blind study of low dose aspirin in children from 4 to 10 years with homozygous sickle cell disease or hemoglobin sickle β0- thalassemia. The study’s primary objective was to evaluate the effect of aspirin prophylaxis on the change in IQ score from baseline to month 36. Subsequent to the stroke conference, and in response to reviewer concerns, the investigators have reconsidered the age range as well as the primary outcome variable. They have contemplated alternative study designs and plan to randomize younger children. Investigators have also considered pursuing a larger internal pilot study to better assess effect size and the incidence of neurocognitive and imaging abnormalities in this population. Whatever design and primary outcome are ultimately deemed reasonable, final SMART study objectives will likely include an assessment of aspirin’s effect upon MRI/MRA and TCD imaging studies, neurocognitive test performance, and the incidence of clinical complications (stroke, acute chest crises, painful crises). Aspirin’s long-term safety and tolerability will also be evaluated.

The initially suggested aspirin dosage for START was 1-2.5 mg/kg/day to the nearest 25-mg capsule. The dose was subsequently changed to a half or whole 81-mg tablet, 2.5 to 5.1 mg/kg/day. This chewable orange flavored tablet was preferred in that it has a candy-like taste and an established track record in pediatric patients; an identical inactive pill has been prepared by Bayer for the placebo arm. The literature suggests that the revised dose will be a safe one. In terms of the schedule of activities, the pilot study (START) is designed as a 3-year trial with the drug administered over 18 months. The SMART Phase III trial of aspirin efficacy would require a longer study period.

As START and SMART stand now, the trials pose some special design concerns. For example, chronic Ibuprofen use has been associated with diminished aspirin efficacy in adults. Should the use of this medication be restricted in a population that uses it frequently to combat vaso-occlusive crisis pain? What further safeguards beyond surveillance laboratory testing, dose adjustment, appropriate immunization, and available medical attention should be provided to prevent Reye’s syndrome and/or significant bleeding? Finally, if SMART is ultimately approved, difficulties are anticipated in coordinating the study processes and standardizing laboratory, imaging, and neurocognitive test procedures at as many as 25 geographically separate institutions.

The aim of the Prevention of Morbidity in Sickle Cell Disease (POMS) study, a pilot randomized, double-blind, controlled trial (Principal Investigator Dr. Fenella Kirkham, pediatric neurologist at the Institute of Child Health, University College London) is to explore the safety and feasibility of using prophylactic nocturnal respiratory support, initially for 6 weeks with an attention endpoint, and, unless the trial is stopped, subsequently for 1 year, with a pain endpoint. For the 6 week pilot, patients will be randomized to no treatment or to continuous positive airways pressure, automatically triggered when there is obstruction to breathing (autoCPAP). After 2 weeks, overnight pulse oximetry will be undertaken and if oxyhaemoglobin saturation remains 94% nocturnal continuous positive airways pressure (CPAP) oxygen supplementation, delivered via an oxygen concentrator will be added for those randomized to treatment. The longer term aim is to test whether this intervention will reduce morbidity in sickle cell disease without undue complications. For the 1 year trial, patients will be randomized to autoCPAP +/− oxygen if required or to sham autoCPAP/sham oxygen concentrator and pain frequency will be documented in the two arms. Preliminary evidence suggests that nocturnal oxygen desaturation is a predictor of central nervous system events, including stroke [3], raising the possibility that interventions designed to reverse nocturnal oxyhemoglobin desaturation would prevent some of the neurologic morbidity associated with sickle cell disease. However, approximately 440 patients would be required for a clinical central nervous system end point, and such a full Phase III trial, particularly if magnetic resonance scanning were to be included at entry and exit, would be difficult to justify on the evidence currently available. Nocturnal hypoxemia was also associated with increased frequency of painful episodes in the East London cohort [50]. In the initial phase of this study involving 70 patients initially, and funded by the United Kingdom Stroke Association, the primary hypothesis is that nocturnal desaturation, measured as reduction in tissue oxyhemoglobin saturation from the maximum of 100%, is associated with frequent pain [50] and that this morbidity can be reduced with overnight CPAP respiratory support. Cognitive dysfunction, particularly in the domain of attention, may also be related to chronic oxyhemoglobin desaturation or poor sleep [50], and although there are few published data involving the direct measurement of oxyhemoglobin saturation in sickle cell disease, intelligence quotient does appear to be lower in those with more severe anemia [11], [52]. The possibility that overnight oxygen CPAP supplementation improves sleep quality and daytime attention, measured as omissions on the Conners Continuous Performance Test [53] and by assessment with the Chervin sleep questionnaire [54], will be explored in the pilot phase, which will also explore feasibility, acceptability, compliance, dosage, and toxicity. Neurologic events (strokes [overt and silent], transient ischemic episodes, seizures, and coma), poor growth, and other complications (e.g., priapism, skin ulceration), which may affect quality of life, may also be related to chronic hypoxemia. The possibility that overnight respiratory support reduces these morbidities will be explored in this phase of the trial, but because of the sample size only very large effects on these other features of sickle cell disease can be detected. All patients will have brain magnetic resonance at baseline so that if further funding can be secured, change in magnetic resonance may be used as a surrogate neurologic end point.

The study has been designed as a blinded placebo-controlled, randomized trial. There will be an initial start-up phase designed to demonstrate the feasibility of the project and the safety of the intervention and to obtain preliminary data on patient acceptability and any effect on attention. Funding has been secured for a further 2 years to complete a blinded randomized trial of CPAP respiratory support with an end point of the number of painful episodes requiring opioid treatment. Unless the trial is halted by the Data and Safety monitoring committee, additional funding will be sought to allow the trial to answer the question as to whether overnight CPAP respiratory support reduces neurologic morbidity.

An encrypted computerized data entry system will be set up using the National Health Service Web with an external link to foreign sites. Consenting patients with hemoglobin SS, who are aged >4 years, and are not in a transfusion program or on hydroxyurea, have no other neurologic diagnosis and are not pregnant, are eligible.

A nurse will calibrate all measurement apparatus and will make sure that procedures are standardized in the United Kingdom and the United States. Baseline sleep study will be performed and the encrypted data will be sent via a secure link to a computer where it will be analyzed blind to the clinical data. Randomization will be stratified by center and minimized by adenotonsillectomy status and presence of silent infarction on magnetic resonance imaging. At baseline, hematology including oxyhemoglobin affinity, arterialized capillary gases, oxyhemoglobin saturation, blood pressure, anthropometry, and neurologic examination will be documented. Parents will be asked to fill in the Chervin Sleep Questionnaire [54] and the Strengths and Difficulties Questionnaire [55]. The physiologic measurements will be repeated at each clinic visit and questionnaires will be administered again at 6 months and 1 year after initiating treatment. Magnetic resonance imaging, magnetic resonance angiography, and magnetic resonance venography will be obtained at baseline.

In the start-up phase over the first year, we aim to recruit 20 patients in London, U.K. and 10 in St. Louis, Missouri, USA. We will focus on assessing the acceptability to the children and their families of randomization and of using the autoCPAP/oxygen concentrator for at least 5 hours every night. In addition, we will look at the feasibility of obtaining baseline data, side effects of CPAP respiratory support, e.g. suppression of erythropoiesis, and frequency of primary and secondary outcome events. After randomization to receive overnight autoCPAP/oxygen or shamCPAP/air via specially designed equipment labeled as a Trial machine, baseline data on hemoglobin, white cell count, and all previous complications will be recorded and the sleep questionnaire will be administered by a psychologist with no knowledge of the treatment arm. Overnight sleep studies will be performed after randomization to document nocturnal oxygen desaturation and any improvement in desaturation with respiratory support. It will also be important to determine that reversal of oxyhemoglobin desaturation with respiratory support is not associated with dangerous hypoventilation, evidenced by unchanged transcutaneous carbon dioxide tension. The default oxygen flow rate is 0.5 liters per minute (administered via pediatric nasal prongs attached to nonkink tubing), unless there is unacceptable hypercapnia (transcutaneous carbon dioxide rising by more than 10 mm Hg) when 0.25 liters per minute will be administered. If oxyhemoglobin saturation does not rise to 94-96% on 0.5 liter per minute, then the administered flow rate will be increased in 0.5 liter per minute increments until oxyhemoglobin saturation rises to 94-96%, provided transcutaneous carbon dioxide tension does not rise by more than 10 mm Hg. Adverse events, for example, nausea, nasal discomfort, and bone marrow suppression leading to a decrease in blood components [56], [57] (red cells, reticulocytes, white cells, platelets) will be documented regularly. Erythropoietin and soluble transferrin receptors will be measured serially. Event rates for complications, particularly painful episodes requiring opioids, will be established from health care records, pain diaries, and direct patient contact via mobile telephone text messaging. After 1 year, the sleep study will be repeated without CPAP oxygen to determine whether regular nocturnal CPAP respiratory support alters overnight ventilation or ventilation perfusion-matching in the lung by looking for changes in baseline oxyhemoglobin saturation, transcutaneous carbon dioxide tension, the overnight oxyhemoglobin saturation profile, and arterialized capillary blood gases. Oxyhemoglobin affinity will also be measured again. The Conners Continuous Performance Test will be administered again at 6 weeks (in the pilot), 6 months and 1 year by the blinded psychologist. In the first 30 patients the effect of respiratory support on the number of Conners Continuous Performance Test omissions will be explored as the primary end point for the pilot.

Assuming that it is justifiable to continue the trial for a further 2 years, there is an intention to establish “proof of principle” by recruiting an additional 70 patients (50 in the United Kingdom and 20 in the USA) to assess the effect of respiratory support on pain and to examine the size of any treatment effect. Evidence of an effect on pain would not necessarily be sufficient to halt recruitment (and might be a spur to recruitment in additional centers in the United Kingdom, United States, and Europe). The primary end point will be the number of separate episodes that the patient has pain requiring opioid treatment over a 1-year period. In our cohort study, for patients with sickle cell anemia and mean overnight oxyhemoglobin saturation <94%, the mean number of painful episodes per year for which medical attention was sought was 1.41 (S.D. 1.05), compared with 0.67 (S.D. 0.61) for patients with mean overnight oxyhemoglobin saturation >94% [51]. Using these summary values, and assuming a single mean comparison as the primary analysis, we would need 28 participants per group to detect a mean difference of 0.75 (1.35 minus 0.6) at 5% significance, and with 90% power. Assuming a fairly strict 20% dropout for this activity intensive trial would increase the required sample size to 34 participants per group. Using a bootstrapping procedure, the 95th percentile sample size value (with 20% dropout inflation) would increase to 87 per group. Sleep studies, physiologic measurements, behavioral and sleep questionnaires, and repeat Conners Continuous Performance Test will be available at 1 year. Secondary end points will include change in blood pressure, day and night oxyhemoglobin saturation, growth velocity, attention and behavior, and other complications, for example neurologic events.

Recruitment and follow-up beyond 3 years would require additional funding. Any decision on continuing the trial will be taken by the Trial Steering Committee (TSC) in the light of a report from the independent Data Monitoring and Safety Committee. As discussed above, it may be necessary to increase the number of patients in each arm to 87 to answer the question about the effect of respiratory support on pain. If ethically justifiable, the data from the start-up phase will be used to inform the sample size and therefore the resources required for a larger trial or continuation of the follow-up for a longer period. The specific aim would be to look at the effect of respiratory support on neurologic events, which is likely to require at least 220 patients in each arm. Funding for repeat magnetic resonance will be sought, to look at progression of silent infarction and vascular disease. Further analysis of data from the London, Jamaican, and the Co-operative Study of Sickle Cell Disease cohorts may also inform the design of the trial.

In terms of practicalities, the steering committee should provide a wide body of additional expertise and will need at least weekly telephone conferences. The Data Safety Monitoring Board will be independently appointed. The laboratory should be centralized in order to provide reliable uniform results. Guidelines and procedures should be established with respect to sample preparation and shipping, generation of site reports, transmission of the laboratory data, blinding strategies for selected laboratory tests, and transmission of needed information to the Data Safety Monitoring Board. Neuroimaging, such as magnetic resonance imaging and transcranial Doppler studies, will usually be performed at the local site but will be transmitted and read centrally by blinded neuroradiologists; electronic transmission of deidentified neuroimaging in Dicom format is now a practical alternative to sending hard copy films or media. Neuropsychology testing will usually be undertaken locally by personnel appropriately trained to administer the tests. Pilot trials may be managed by a general clinical research center equipped to handle a small one- or two-site study. Multiple sites will need to commit for a Phase III study, and a full Clinical Trials Coordination Center will be required for data management. A typical budget will be developed as a per patient fee with sequential payouts based upon a site accomplishing set tasks. There may be some benchmarking to balance costs in larger and smaller cities.

Section snippets

Summary

In summary, the successful STOP and STOP II studies have pointed the direction for trials of primary stroke prevention in sickle cell disease. The Silent Infarct Transfusion trial has been funded as a Phase III randomized controlled trial, and pilot studies of low dose aspirin and overnight respiratory support are being developed. The camaraderic among the teams, which crosses oceans as well as borders, is likely to lead to better understanding of the pathophysiology of the neurologic

References (57)

  • J. Wilimas et al.

    Efficacy of transfusion therapy for one to two years in patients with sickle cell disease and cerebrovascular accidents

    J Pediatr

    (1980)
  • M.O. Russell et al.

    Effect of transfusion therapy on arteriographic abnormalities and on recurrence of stroke in sickle cell disease

    Blood

    (1984)
  • W.C. Wang et al.

    High risk of recurrent stroke after discontinuance of five to twelve years of transfusion therapy in patients with sickle cell disease

    J Pediatr

    (1991)
  • S. Rana et al.

    Discontinuation of long-term transfusion therapy in patients with sickle cell disease and stroke

    J Pediatr

    (1997)
  • C.H. Pegelow et al.

    Risk of recurrent stroke in patients with sickle cell disease treated with erythrocyte transfusions

    J Pediatr

    (1995)
  • R.E. Ware et al.

    Prevention of secondary stroke and resolution of transfusional iron overload in children with sickle cell anemia using hydroxyurea and phlebotomy

    J Pediatr

    (2004)
  • W.C. Wang et al.

    A two-year pilot trial of hydroxyurea in very young children with sickle-cell anemia

    J Pediatr

    (2001)
  • S.T. Miller et al.

    Cooperative Study of Sickle Cell Disease. Silent infarction as a risk factor for overt stroke in children with sickle cell anemia: A report from the Cooperative Study of Sickle Cell Disease

    J Pediatr

    (2001)
  • H.J. Fullerton et al.

    Declining stroke rates in Californian children with sickle cell disease

    Blood

    (2004)
  • C.H. Pegelow et al.

    Longitudinal changes in brain magnetic resonance imaging findings in children with sickle cell disease

    Blood

    (2002)
  • J. Greenberg et al.

    Trial of low doses of aspirin as prophylaxis in sickle cell disease

    J Pediatr

    (1983)
  • D.R. Hargrave et al.

    Nocturnal hypoxemia and painful sickle cell crises in children

    Blood

    (2003)
  • C.J. Earley et al.

    Stroke in children and sickle-cell disease

    Neurology

    (1998)
  • J.A. Stockman et al.

    Occlusion of the large cerebral vessels in sickle cell anemia

    N Engl J Med

    (1972)
  • G. Sébire et al.

    Cerebral venous sinus thrombosis in childrenRisk factors, presentation, diagnosis and outcome

    Brain

    (2005)
  • K.H. Lee et al.

    Unusual encephalopathy after acute chest syndromeAcute necrotizing encephalitis

    J Pediatr Hematol/Oncol

    (2002)
  • F. Bernaudin et al.

    Multicenter prospective study of children with sickle cell diseaseRadiographic and psychometric correlation

    J Child Neurol

    (2000)
  • D.E. Saunders et al.

    MRI in children with sickle cell disease without overt stroke

    Dev Med Child Neurol

    (2001)
  • Cited by (37)

    • Sickle cell disease

      2020, Rosenberg’s Molecular and Genetic Basis of Neurological and Psychiatric Disease: Volume 2
    • Sickle Cell Disease

      2014, Rosenberg's Molecular and Genetic Basis of Neurological and Psychiatric Disease: Fifth Edition
    • Acute silent cerebral ischemia and infarction during acute anemia in children with and without sickle cell disease

      2012, Blood
      Citation Excerpt :

      Risk factors, recurrence rates, and optimal treatment approaches for the 2 groups may differ. The term silent cerebral infarction is a misnomer, as previously noted,1,2,13-16 but is the current standard in the field to describe these small lesions with no obvious focal findings. Our observations demonstrate that 舠silent舡 may be more a matter of the depth of the investigation.

    • Treatment of "other" stroke etiologies

      2011, Stroke: Pathophysiology, Diagnosis, and Management
    View all citing articles on Scopus
    View full text