Published online September 3, 2007
PEDIATRICS Vol. 120 No. 4 October 2007, pp. e1107-e1116 (doi:10.1542/peds.2006-2002)
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Mazumdar, M.
Right arrow Articles by Lieu, T. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Mazumdar, M.
Right arrow Articles by Lieu, T. A.
Related Collections
Right arrow Blood
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Facebook   Add to Reddit   Add to Technorati   Add to Twitter  
What's this?

SPECIAL ARTICLE

Preventing Stroke Among Children With Sickle Cell Anemia: An Analysis of Strategies That Involve Transcranial Doppler Testing and Chronic Transfusion

Maitreyi Mazumdar, MD, MPHa,b, Matthew M. Heeney, MDc,d, Colin M. Sox, MD, MSe and Tracy A. Lieu, MD, MPHe,f

a Department of Neurology
c Divisions of Hematology/Oncology
f General Pediatrics, Children's Hospital Boston, Boston, Massachusetts
b Harvard Pediatric Health Services Research Fellowship Program, Boston, Massachusetts
d Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts
e Department of Ambulatory Care and Prevention, Center for Child Health Care Studies, Harvard Medical School and Harvard Pilgrim Health Care, Boston, Massachusetts


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
BACKGROUND. Transcranial Doppler ultrasonography can identify children with sickle cell anemia who are at elevated risk of stroke and may benefit from chronic transfusions. Uncertainty about the risk/benefit trade-offs of chronic transfusion has led some clinicians to decide not to offer transcranial Doppler ultrasonography screening.

OBJECTIVES. Our goals were to (1) compare the projected benefits and risks of 6 primary stroke-prevention strategies, (2) estimate the optimal frequency of screening, and (3) identify key assumptions that influence the risk/benefit relationship.

METHODS. We designed a decision model to compare 6 primary stroke-prevention strategies: (1) annual transcranial Doppler ultrasonography screening until age 16 with children at high risk of stroke receiving monthly transfusion for life; (2) annual transcranial Doppler ultrasonography until age 16 with transfusions until age 18; (3) biannual transcranial Doppler ultrasonography until age 16 with transfusions until age 18; (4) annual transcranial Doppler ultrasonography until age 10 with transfusion until age 18; (5) 1-time screening at age 2 with transfusion until age 18; and (6) no intervention. Assumptions were derived from the published literature.

RESULTS. For a hypothetical cohort of 2-year-old children, the optimal strategy was transcranial Doppler ultrasonography screening annually until age 10 with children at high risk receiving monthly transfusions until age 18. The optimal strategy would prevent 32% of strokes predicted to occur without intervention. The optimal strategy led to benefits similar to more intensive screening and transfusion strategies but resulted in fewer adverse events. All the intervention strategies resulted in net losses in life expectancy, because the projected mortality averted by stroke prevention was outweighed by the projected increase in mortality from transfusion. Results were sensitive to adherence rates to iron-chelation therapy.

CONCLUSIONS. The optimal stroke-prevention strategy was projected to be annual transcranial Doppler ultrasonography screening until age 10 with transfusion for children at high risk until age 18. Better adherence to chelation therapy would improve life expectancy in all intervention strategies.

Key Words: sickle cell anemia • stroke • transcranial Doppler • transfusions • health services research

Abbreviations: TCD—transcranial Doppler ultrasonography • NHLBI—National Heart, Lung, and Blood Institute • STOP—Stroke Prevention Trial in Sickle Cell Anemia

Stroke occurs in 7% to 13% of children with sickle cell anemia (hemoglobin SS) and can lead to motor disability, neuropsychological impairment, and death.1 Transcranial Doppler ultrasonography (TCD) predicts which children are at highest risk for stroke by using measurements of mean blood velocity in the internal carotid and middle cerebral arteries, and chronic transfusions are 90% effective in preventing stroke in children at high risk.2–4 The National Heart, Lung, and Blood Institute (NHLBI) has recommended that all children with sickle cell anemia aged 2 to 16 years undergo TCD screening but has not universally endorsed chronic transfusions. Instead, the NHLBI recommends that families of children at high risk discuss the risks and benefits of transfusions with knowledgeable clinicians.5

To advise patients most effectively, clinicians need more specific information on the risks and benefits of screening and transfusion. Chronic blood transfusion is associated with increased risk of iron overload, alloimmunization, and exposure to infectious diseases.6 The uncertainty about the risks and benefits of screening and transfusion is cited as the most common reason for not offering TCD testing.7,8 Professional societies vary in their recommendations; the American Academy of Pediatrics, for example, decided to "withhold an unequivocal recommendation" for the use of TCD until more data about the long-term consequences are available.9,10

For those who do use TCD, the optimal frequency of screening and duration of transfusion remain unknown.5,11,12 The incidence of a primary stroke is highest in early childhood,1,13 which suggests that there is a period of elevated risk after which transfusions might be stopped. Yet the risk/benefit threshold is difficult to determine, because reports of case series have documented differing rates of stroke recurrence after discontinuation of transfusions.14–16 The Stroke Prevention Trial in Sickle Cell Anemia (STOP) II trial, designed to determine if it is safe to discontinue transfusions after 30 months, was halted because children in the no-transfusion group were found to have TCD results that had reverted to high risk. In addition, there were 2 strokes in the no-transfusion group compared with none in the continued-transfusion group.17

To assess the relative risks and benefits of TCD screening and chronic transfusion for primary stroke prevention in sickle cell anemia, a number of factors must be explicitly considered. These factors include the age-specific incidence of stroke, the feasibility of screening, the acceptability of transfusion, and the concomitant risks of iron overload, alloimmunization, and infection. No single clinical trial or longitudinal cohort study will be able to consider all of these components. A decision-analytic approach that uses mathematical modeling is needed to incorporate data from multiple sources to extrapolate outcomes beyond the time horizon of a clinical study and to evaluate more strategies than are possible in a single trial.

Our objectives for the decision analysis were to (1) compare the projected benefits and risks of 6 clinical strategies for primary stroke prevention by using a decision model, (2) estimate the optimal frequency of TCD screening, and (3) identify key assumptions that influence the risk/benefit relationship. We sought to achieve these objectives by applying a hypothetical cohort of asymptomatic children with sickle cell anemia who are undergoing TCD screening for primary stroke prevention to a decision model. Using data from the published literature, we developed a computer-based model to project the clinical benefits and adverse events of various TCD screening and treatment strategies for primary stroke prevention in children with sickle cell anemia in the setting of clinical practice in the United States.


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Decision-Analytic Model
We synthesized the best available data and used a computer-based Markov model to simulate the natural history of stroke in hypothetical cohorts of children with sickle cell anemia following 6 primary stroke-prevention strategies. We assessed the projected efficacy of alternative screening strategies by determining the incremental benefits, measured as the number of ischemic strokes prevented, cases of major disability prevented, and years of gained life expectancy. We discounted benefits by an annual rate of 3%, which is consistent with current guidelines for economic analysis.18

We designed the model to compare 6 different primary stroke-prevention strategies recommended by professional societies or reported to be currently in use. The NHLBI, American Academy of Neurology, and American Stroke Association each recommend that children with sickle cell anemia undergo TCD screening at least once between the ages of 2 and 16 years.11,12,19 The New England Pediatric Sickle Cell Consortium recommends screening every 6 to 12 months between the ages of 3 and 16 years.20 The American Academy of Pediatrics recommends discussing "screening with TCD ultrasonography, if available."9 Our communication with other sickle cell centers revealed that some screen annually until 10 years of age only. Although none of these groups were able to recommend a clear length of time or age at which to stop transfusion, many centers have historically used the age of 18 years16 as an age to stop transfusions for secondary stroke prevention.

Specifically, the decision model included the following 6 stroke-prevention strategies for 2-year-old children with sickle cell anemia (Fig 1): (1) annual TCD screening until age 16 and lifelong transfusion for children at high risk; (2) annual TCD screening until age 16 and transfusion until age 18 for children at high risk; (3) biannual TCD screening until age 16 and transfusion until age 18 for children at high risk; (4) annual TCD screening until age 10 and transfusion until age 18 for children at high risk; (5) 1-time screening at age 2 and transfusion until age 18 for children at high risk; and (6) no TCD screening.


Figure 1
View larger version (19K):
[in this window]
[in a new window]

  		FIGURE 1 The decision-analysis model. Circles represent chance events; the square represents the choice between 6 different primary stroke-prevention strategies; and the circled M's represent Markov models.

 
In our model, the natural history of stroke was represented as a sequence of yearly transitions between 5 mutually exclusive health states (stroke free, hemorrhagic stroke, ischemic stroke, poststroke, and dead). Hemorrhagic strokes were defined as clinical events with brain imaging or autopsy that documented intracranial bleeding. Ischemic strokes were also defined as clinical events with radiologic evidence of infarction without significant hemorrhage. Hemorrhagic conversion of ischemic stroke, if ascertainable, was included as ischemic stroke. Individuals entered the "poststroke" state after 1 year of surviving the initial neurologic event. Because the ability of TCD to predict subclinical stroke or transient ischemic attack and the benefit of prophylactic transfusions in these populations has not been determined, subclinical or "silent" strokes and transient ischemic attack were not included in this model; these patients were categorized as being in a stroke-free state.

A model that accurately describes the effect of screening and chronic transfusions on overall health mandates including additional health states for adverse effects of transfusions that accumulate over time, such as iron overload, alloimmunization, and infection from blood-borne organisms (Fig 2).


Figure 2
View larger version (17K):
[in this window]
[in a new window]

  		FIGURE 2 Markov model for children with sickle cell anemia who are at high risk of ischemic stroke and undergoing transfusions for primary stroke prevention.

 
Patients in the hypothetical cohort made transitions from one health state to another on a yearly basis. For example, in any given year, a child receiving transfusions could remain event free, develop iron overload, acquire a blood-borne infection, become alloimmunized, experience a stroke, or die. The probability of such transitions depended on age and stroke history and varied with the receipt of prophylactic transfusions. In our model, all children who suffered strokes but survived were subsequently treated with a lifelong chronic transfusion regimen for secondary stroke prevention. In addition to stroke-related deaths, deaths occurred as a result of other events such as other complications of sickle cell anemia. The time horizon of the analysis incorporated the lifetime of individuals in the hypothetical cohort.

Probabilities and Assumptions
The decision model's key probabilities are listed in Table 1 and are reviewed in detail below.


View this table:
[in this window]
[in a new window]

  		TABLE 1 Selected Disease and Screening Rates and Probabilities Used in the Decision Model

 
Natural History
Background mortality rates and ischemic stroke rates for children and adolescents were obtained from a population-based study of children with sickle cell disease in Dallas. The cohort included children born in or after 1983 who were recruited from a large urban and rural catchment area.13 Data for children with sickle cell anemia were presented separately for this cohort. Transition probabilities for death and stroke were generated from incidence rates.21 Mortality rates for adults with sickle cell anemia were generated by using US vital statistics data for the adult black population in the year 200222 in conjunction with survival data from an older study of adults with sickle cell anemia.23 The highest annual mortality rates were seen before age 10 and after age 50.

In the Dallas cohort, the risk of ischemic stroke varied with age. The risk of ischemic stroke was highest between ages 2 and 8 years (1.08–1.67 events per 100 patient-years). This age-specific risk was seen in older cohort studies as well.1 The risk of ischemic stroke in adults with sickle cell anemia was derived from older work published by Ohene-Frempong et al,1 who used a longitudinal cohort of patients enrolled in the Cooperative Study of Sickle Cell Disease. In this cohort, adult risk of ischemic stroke was highest in those who were >50 years old (1.28 events per 100 patient-years). The risk of death immediately after ischemic stroke was assumed to be 5%, which is consistent with multiple case series.24–26

The risk of hemorrhagic stroke was derived from the Cooperative Study of Sickle Cell Disease cohort and also varied by age (0–0.44 events per 100 patient-years). The highest rates were in children aged 6 to 9 years. The risk of death immediately after hemorrhagic stroke was assumed to be 33% by using data from the same study and an older natural history study that used a similar population of patients.24

Transcranial Doppler Ultrasonography
We assumed that 5% of the cohort would not receive TCD screening. We defined being "at high risk for stroke" as having unilateral or bilateral TCD results of ≥200 cm/second on 2 repeated examinations. Our estimates of screening feasibility and results were based on the studies that established TCD as an effective screening strategy3,4,27 and the initial STOP trial.2 We based our prediction of the probability of a child transitioning from showing a normal to an abnormal TCD result on data collected for the STOP trial.27,28

Chronic Blood Transfusion
Our model assumed that all children with abnormal TCD results would be offered chronic transfusion. Our definition of chronic transfusion therapy was monthly packed red blood cell transfusions designed to maintain the hemoglobin S level at <30%. On the basis of the acceptance of randomization into the STOP study, our model assumed that 80% of the families approached would agree and be adherent to chronic transfusions.2 This was a conservative assumption, because the acceptance rates might be higher now after the STOP and STOP II trials' results have been widely disseminated. The probabilities of children with abnormal, normal, or conditional TCD results suffering an ischemic stroke were drawn from Adams'3,4 seminal studies as well as published expert opinion.29

Our assumptions about the magnitude of stroke and mortality risk reduction that would be achieved by chronic transfusions were based on the STOP trial and published expert opinion. For children at high risk who received transfusions, the risk of ischemic stroke was 90% less than what it would have been without transfusion, as seen in the STOP trial. The reduction in stroke risk achieved by chronic transfusions was applied only to ischemic strokes, because it is generally believed that this treatment does not affect the risk of hemorrhagic stroke.11 The risk of ischemic stroke peaks twice over time in people with sickle cell anemia: once before adolescence and again after 30 years of life, with the latter peak assumed to be affected by acquired risk factors such as hypertension and coronary artery disease. We subtracted the stroke incidence rates of the black US population without sickle cell disease30 from stroke seen in adults with sickle cell anemia to estimate how many strokes could be attributed to sickle cell anemia and, thus, likely to be affected by transfusion.

Patients receiving chronic blood transfusions inevitably develop iron overload, whereas alloimmunization and transfusion-related infections are less-common complications. If untreated, iron overload can result in organ failure and death.31 Our model assumed that iron overload occurs after 2 years of transfusions. Once children suffered iron overload, the model assumed that they would all be offered iron-chelating therapy. In our model, we assumed chelation would be achieved through the use of deferasirox, a recently approved oral chelating agent. Our model assumed that 80% of the patients with iron overload would be adherent to chelation treatment and that iron levels would normalize within 1 year on chelation therapy. These estimates were taken from studies of adherence to daily oral medications in a variety of populations.32 We assumed that the patients' risk of death increased while they were iron overloaded and returned to baseline after having received iron-chelation therapy. Our estimate of the increased mortality attributable to iron overload was drawn from studies of adults that suggested that adults with sickle cell and iron overload have an 8-times-higher risk of death than those who are not iron overloaded.33

We based the model's risk of blood-borne infection on the risk of transmission of hepatitis B and C and HIV from the US blood supply.6 Alloimmunization rates were taken from the STOP trial's results, which had used blood matched to a variety of antigens.34 Once children suffered transfusion-related alloimmunization, we assumed that transfusions were stopped and the risk of stroke returned to what it would have been without any transfusions within 1 year.

Major disability in this model was defined as hemiplegia or long-lasting cognitive or expressive deficit such as acquired aphasia from a left-hemisphere stroke. We estimated that 80% of the children who suffered a clinically overt ischemic stroke would be left with neurologic sequelae leading to major disability; this estimate was taken from trials and natural history studies of ischemic stroke.3,24

Base-Case Scenario and Assumptions
We used a hypothetical cohort of 2000 2-year-old children with sickle cell anemia, equally distributed between boys and girls. This is approximately the number of infants with sickle cell anemia who are identified annually by newborn screening programs in the United States.12 These children were considered to be those who would be identified by newborn screening and who, at age 2, would have no previous history of neurologic symptoms or other requirements for multiple or exchange transfusions. We estimated that screening would not be performed successfully in 5% of the children. We estimated that 80% of the families offered transfusion would both accept and be adherent to a chronic transfusion regimen. Once the children had suffered iron overload, 80% were assumed to be adherent to oral iron-chelating therapy.

Other assumptions inherent in the model related to the risk of stroke and death under different circumstances. These estimates were taken from expert opinion compiled in a previous decision analysis and included the relative risk of death while receiving transfusions.29 Once children suffered transfusion-related alloimmunization, we assumed that transfusions were stopped and the risk of stroke returned to what it would have been without any transfusions within 1 year.

Sensitivity Analyses
To identify key assumptions that influence the model's prediction of the risk/benefit relationship, we conducted sensitivity analyses by varying the event probabilities within reasonable ranges. Because widespread chronic transfusion of patients at high risk in the United States would not necessarily include such careful matching of blood antigens, we conducted sensitivity analyses on alloimmunization risk. Similarly, we varied the probabilities of iron overload and adherence with iron-chelating treatment in the sensitivity analyses, because iron overload is the adverse event that most strongly affects mortality. All estimates of the effect of transfusion on mortality were drawn from expert opinion29; thus, these estimates were tested with wide ranges in sensitivity analyses. Sensitivity analyses were also conducted to test the effect of assumptions about the proportion of the cohort that was screened and the proportion that accepted and adhered to the chronic transfusion regimen.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Projected Effects of 6 Stroke-Prevention Strategies on Rate of Ischemic Strokes, Major Disability, and Life Expectancy
In the absence of screening, our model predicted that 295 ischemic strokes and 236 cases of major disability would occur in our hypothetical cohort of 2000 children with sickle cell anemia. The addition of different TCD screening and chronic transfusion strategies to the model resulted in the prevention of 2% to 39% of ischemic strokes and cases of disability (Table 2). The strategies that involved either screening more frequently or over longer periods of time identified more children at high risk and, therefore, more children who received chronic transfusion therapy. In our model, more-intensive strategies would prevent more cases of stroke and disability than less-intensive strategies but resulted in more transfusion-related iron overload. Screening children between 2 and 10 years of age annually would prevent >15 times as many ischemic strokes and cases of disability than would screening at age 2 only. Screening every other year was not as effective as annual screening, because a proportion of children at high risk were not identified; some experienced a stroke in the time between screenings.


View this table:
[in this window]
[in a new window]

  		TABLE 2 Predicted Number of Disease Events and Life Expectancy for the 6 Primary Stroke-Prevention Strategies Modeled

 
Although it is effective in preventing ischemic stroke, the addition of children aged 11 to 16 years to the population receiving TCD screening resulted in many more children receiving transfusions and, subsequently, transfusion-related adverse events. As the number of children receiving transfusions increased, the incremental benefit in stroke reduction achieved by transfusions was counterbalanced by transfusion-related adverse events (Fig 3). Our analysis suggests that children identified as high risk at earlier ages were more likely to suffer strokes, and therefore benefit more from chronic transfusions, than children identified after the age of 10 years.


Figure 3
View larger version (11K):
[in this window]
[in a new window]

  		FIGURE 3 Risk/benefit analysis of primary stroke-prevention strategies. A indicates 1-time screening at age 2, transfusion until 18; B, annual screening at ages 2 to 10, transfusion until 18 years; C, biannual screening ages 2 to 16, transfusion until 18 years; D, annual screening at ages 2 to 16, transfusion until 18 years; E, annual screening at ages 2 to 16, lifelong transfusion. (Annual screening leads to more transfusions being administered, which yields many more strokes prevented than a 1-time screening. Biannual screening holds no advantages over other strategies.)

 
Increasing the number of children who received transfusions in the model resulted in a cohort with a shorter average life expectancy that was attributable to transfusion-related iron overload. The gain in life expectancy provided by receiving chronic transfusions for stroke prevention was outweighed by the corresponding decreased life expectancy caused by iron overload.

No intervention strategy resulted in a significant number of blood-borne infections or cases of alloimmunization (results not shown).

Sensitivity Analyses
The decision model was most sensitive to 2 factors: (1) assumptions about iron overload's effect on mortality and (2) the probability of subjects' adherence to iron-chelation therapy. As shown in Fig 4, the benefits of transfusion outweighed its risks when iron overload only caused a threefold increase in mortality during a given year rather than the eightfold increase that was assumed in the base case. Published literature has suggested that that the mortality attributable to iron overload actually may be much higher, in the range of an 8- to 12-fold increased risk of death.29,33 The negative effect of iron overload on mortality in our model can be overcome if subjects have higher rates of iron-chelating therapy use. If iron-overloaded children in the cohort were assumed to be 100% adherent with chelating therapy, the increased mortality caused by transfusion-related iron overload was offset completely (Fig 5).


Figure 4
View larger version (16K):
[in this window]
[in a new window]

  		FIGURE 4 One-way sensitivity analysis showing the average life expectancy that would be predicted by the different screening strategies by varying the factor by which iron overload affects mortality (life expectancy is presented without discounting).

 

Figure 5
View larger version (20K):
[in this window]
[in a new window]

  		FIGURE 5 One-way sensitivity analysis showing the average life expectancy that would be predicted by the different strategies by varying the probability of adherence to iron-chelation therapy (life expectancy is presented without discounting).

 
When we tested assumptions about the annual probability of alloimmunization, the percentage of children who were able to be screened successfully, and the percentage of those who would accept and be adherent to a chronic transfusion regimen, we did not find any change in the patterns or conclusions seen in our base-case analysis. Results of sensitivity analyses on the probability of death within the year after suffering an ischemic stroke (assumed to be 5% in our base case) showed that ischemic stroke would have to be a fatal event >60% of the time to change the basic findings of our analysis.


   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Our decision model predicts that screening children with sickle cell anemia annually with TCD from 2 to 10 years of age and discontinuing transfusions after they are 18 years old provides similar benefits to those achieved with annual screening until age 16 and lifelong transfusion and would cause significantly fewer adverse events. The strategy of screening annually from 2 to 10 years of age and transfusing children at high risk until age 18 would prevent almost 40% of ischemic strokes and stroke-related disability and result in fewer deaths from iron overload than more-intensive strategies.

We designed the model to delineate explicitly the trade-offs between strokes prevented by chronic transfusion and complications of transfusion, the most significant of which is transfusion-related iron overload. Decision analysis is a useful approach to addressing the risk/benefit trade-offs, because it would be neither feasible nor ethical to conduct a randomized clinical trial to determine the long-term risks and benefits of intervention strategies. To inform patients' and providers' decisions and the development of clinical guidelines, our decision model synthesized available data drawn from published natural history studies, case series, vital statistics, clinical trials, and expert opinion.

Since the publication of the STOP trial in 1998, clinicians and families have had to face the difficult decision of whether to conduct TCD screening on children with sickle cell anemia and, if so, how often to screen and for how long to give the transfusions. Clinical trials have demonstrated remarkable benefits when children at high risk receive prophylactic transfusions. The decreased rate of hospitalizations for first-time stroke has been attributed to widespread adoption of prophylactic transfusion in children at high risk.35 Yet, despite the observed benefits of transfusion and the potentially catastrophic nature of ischemic stroke, it is difficult for many practitioners and families to give monthly blood transfusions to children for primary prevention.

Our analysis has a number of important potential limitations. The efficacies used in the base case were taken from randomized, controlled clinical trials in which adherence, monitoring, and follow-up rates were better than those seen in general clinical practice. Therefore, we may have overestimated the benefits of screening, transfusion, and chelation, especially among members of the population with poorer access to specialized medical care such as that found in comprehensive sickle cell centers. Our sensitivity analyses provide insight into the effect of adherence to medication and suggest that under all intervention strategies, life expectancy can be lengthened if adherence to iron chelation is improved, a finding that can be translated into clinical practice. In addition, we based assumptions about the effect of iron overload on mortality on studies of adults. If iron overload is less dangerous in children, continued screening at older ages would be more favorable.

Our assumptions about the risk of death and stroke in patients at high risk who were not receiving transfusions were based on published expert opinion. We do not explicitly model the effect of prophylactic transfusion on the risk of death from other sickle cell complications such as acute chest crisis, because data are limited on this potential benefit of prophylactic transfusion. We made the most conservative estimates when evidence was not available. Our assumptions about stroke and mortality risk in these children were not highly influenced in the extremes in the sensitivity analyses.

Whether the benefits of screening and transfusion outweigh the risks depends on adherence to iron-chelation therapy and patient preference. A striking finding of our analysis is that all strategies reduced average life expectancy, but emphasizing this finding ignores the potential trade-offs between life expectancy and major disability in our cohort. Our decision model would be greatly strengthened if the utilities, such as quality-adjusted life-years, for these health states were known for children with sickle cell anemia. There is some evidence that transfusion may improve the quality of life of patients with sickle cell disease who experience frequent pain crises,36 but the effect of chronic transfusions on the quality of life for patients without pain crises is unknown. Our analysis did not incorporate these preferences explicitly but, instead, offers the information to enable clinicians to judge the risk/benefit trade-offs on the basis of their patients' preferences.

Our model did not include primary stroke-prevention strategies that are currently under investigation. These new strategies treat with bone marrow transplantation,37–39 hydroxyurea,40–43 citrulline, arginine, aspirin, and overnight oxygen supplementation.44 If any of these strategies have adverse effects that are less than those of chronic transfusion, the benefit/cost balance of screening and prevention strategies could increase markedly. In addition, our model did not assume the use of hydroxyurea for primary stroke prevention in patients at high risk who become intolerant of blood transfusion, a practice that is becoming more common in sickle cell centers. As ongoing trials45 provide more data about the use of hydroxyurea for stroke prevention, it will become possible to incorporate this practice into future analyses.


   CONCLUSIONS
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
Our decision model predicts that annually screening children with sickle cell anemia from 2 to 10 years of age and providing monthly transfusions for children at high risk until 18 years of age has a better risk profile than the 5 other stroke-prevention strategies that are recommended and/or followed in clinical practice. Until the studies of alternative stroke-prevention strategies are completed, chronic prophylactic transfusion remains the main intervention for children who are identified as being at high risk by TCD. Adopting this primary prevention strategy may prevent strokes as effectively as more-intensive strategies while causing fewer deaths from transfusion-related iron overload.


   ACKNOWLEDGMENTS
 
Dr Mazumdar was supported by the Agency for Healthcare Research and Quality, National Institutes of Health (grant T32 HP-100063), and Dr Lieu was supported by a Mid-Career Investigator Award in Patient-Oriented Research from the National Institute of Child Health and Human Development (K24 HD47667).


   FOOTNOTES
 
Address correspondence to Maitreyi Mazumdar, MD, MPH, Department of Neurology, Children's Hospital Boston, 300 Longwood Ave, Boston, MA 02115. E-mail: maitreyi.mazumdar{at}childrens.harvard.edu

The authors have indicated they have no financial relationships relevant to this article to disclose.


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 

  1. Ohene-Frempong K, Weiner SJ, Sleeper LA, et al. Cerebrovascular accidents in sickle cell disease: rates and risk factors. Blood. 1998;91 :288 –294[Abstract/Free Full Text]
  2. Adams RJ, McKie VC, Hsu L, et al. Prevention of a first stroke by transfusions in children with sickle cell anemia and abnormal results on transcranial Doppler ultrasonography. N Engl J Med. 1998;339 :5 –11[Abstract/Free Full Text]
  3. Adams R, McKie V, Nichols F, et al. The use of transcranial ultrasonography to predict stroke in sickle cell disease. N Engl J Med. 1992;326 :605 –610[Abstract]
  4. Adams RJ, McKie VC, Carl EM, et al. Long-term stroke risk in children with sickle cell disease screened with transcranial Doppler. Ann Neurol. 1997;42 :699 –704[CrossRef][Web of Science][Medline]
  5. National Institutes of Health, National Library of Medicine. Clinical alert: periodic transfusions lower stroke risk in children with sickle cell anemia. Available at: www.nlm.nih.gov/databases/alerts/sickle97.html. Accessed October 5, 2005
  6. Vichinsky EP. Current issues with blood transfusions in sickle cell disease. Semin Hematol. 2001;38(1 suppl 1) :14 –22
  7. Lane PA, Buchanan GE, Ware RE. Variable approaches to therapeutic options for children with sickle cell disease (SCD): a practice survey of the American Society of pediatric Hematology/Oncology (ASPH/O) [abstract]. Pediatr Res. 2002;51(4 pt 2) :238A
  8. Fullerton HJ, Gardner M, Adams RJ, Lo LC, Johnston SC. Obstacles to primary stroke prevention in children with sickle cell disease. Neurology. 2006;67 :1098 –1099[Free Full Text]
  9. American Academy of Pediatrics, Section on Hematology/Oncology Committee on Genetics. Health supervision for children with sickle cell disease. Pediatrics. 2002;109 :526 –535[Abstract/Free Full Text]
  10. Sox C. Health supervision for children with sickle cell disease. Pediatrics. 2003;111 :710 –711[Free Full Text]
  11. Sloan MA, Alexandrov AV, Tegeler CH, et al. Assessment: transcranial Doppler ultrasonography—report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology. Neurology. 2004;62 :1468 –1481[Abstract/Free Full Text]
  12. National Heart, Lung, and Blood Institute, Division of Blood Diseases and Resources. The Management of Sickle Cell Disease. 4th ed. Bethesda, MD: National Institutes of Health; 2002
  13. Quinn CT, Rogers ZR, Buchanan GR. Survival of children with sickle cell disease. Blood. 2004;103 :4023 –4027[Abstract/Free Full Text]
  14. Wilimas J, Goff JR, Anderson HR Jr, Langston JW, Thompson E. Efficacy of transfusion therapy for one to two years in patients with sickle cell disease and cerebrovascular accidents. J Pediatr. 1980;96 :205 –208[CrossRef][Web of Science][Medline]
  15. Rana S, Houston PE, Surana N, Shalaby-Rana EI, Castro OL. Discontinuation of long-term transfusion therapy in patients with sickle cell disease and stroke. J Pediatr. 1997;131 :757 –760[CrossRef][Web of Science][Medline]
  16. Wang WC, Kovnar EH, Tonkin IL, 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;118 :377 –382[CrossRef][Web of Science][Medline]
  17. Adams RJ, Brambilla D. Discontinuing prophylactic transfusions used to prevent stroke in sickle cell disease. N Engl J Med. 2005;353 :2769 –2778[Abstract/Free Full Text]
  18. Gold MR. Cost-effectiveness in Health and Medicine. New York, NY: Oxford University Press; 1996
  19. Goldstein LB, Adams R, Alberts MJ, et al. Primary prevention of ischemic stroke: a guideline from the American Heart Association/American Stroke Association Stroke Council: cosponsored by the Atherosclerotic Peripheral Vascular Disease Interdisciplinary Working Group; Cardiovascular Nursing Council; Clinical Cardiology Council; Nutrition, Physical Activity, and Metabolism Council; and the Quality of Care and Outcomes Research Interdisciplinary Working Group: the American Academy of Neurology affirms the value of this guideline [published correction appears in Stroke. 2007;38:207]. Stroke. 2006;37 :1583 –1633[Abstract/Free Full Text]
  20. New England Pediatric Sickle Cell Consortium. Prevention and treatment of stroke for pediatric patients with sickle cell disease. Available at: www.nepscc.org/Completed%20CPGs/CVA%20CPG%20NEPSCC%20Dec03.pdf. Accessed November 14, 2005
  21. Kuntz K, Weinstein MC. Modelling in economic evaluation. In: Drummond MF, McGuire A, eds. Economic Evaluation in Health Care: Merging Theory With Practice. New York, NY: Oxford University Press; 2001:141–171
  22. Arias E. United States Life Tables, 2002. Hyattsville, MD: National Center for Health Statistics; 2004
  23. Platt OS, Brambilla DJ, Rosse WF, et al. Mortality in sickle cell disease: life expectancy and risk factors for early death. N Engl J Med. 1994;330 :1639 –1644[Abstract/Free Full Text]
  24. Powars D, Wilson B, Imbus C, Pegelow C, Allen J. The natural history of stroke in sickle cell disease. Am J Med. 1978;65 :461 –471[CrossRef][Web of Science][Medline]
  25. Balkaran B, Char G, Morris JS, Thomas PW, Serjeant BE, Serjeant GR. Stroke in a cohort of patients with homozygous sickle cell disease. J Pediatr. 1992;120 :360 –366[CrossRef][Web of Science][Medline]
  26. Neonato MG, Guilloud-Bataille M, Beauvais P, et al. Acute clinical events in 299 homozygous sickle cell patients living in France. French Study Group on Sickle Cell Disease. Eur J Haematol. 2000;65 :155 –164[CrossRef][Web of Science][Medline]
  27. Adams RJ, Brambilla DJ, Granger S, et al. Stroke and conversion to high risk in children screened with transcranial Doppler ultrasound during the STOP study. Blood. 2004;103 :3689 –3694[Abstract/Free Full Text]
  28. Lee MT, Piomelli S, Granger S, et al. Stroke Prevention Trial in Sickle Cell Anemia (STOP): extended follow-up and final results. Blood. 2006;108 :847 –852[Abstract/Free Full Text]
  29. Nietert PJ, Abboud MR, Silverstein MD, Jackson SM. Bone marrow transplantation versus periodic prophylactic blood transfusion in sickle cell patients at high risk of ischemic stroke: a decision analysis. Blood. 2000;95 :3057 –3064[Abstract/Free Full Text]
  30. Schneider AT, Kissela B, Woo D, et al. Ischemic stroke subtypes: a population-based study of incidence rates among blacks and whites. Stroke. 2004;35 :1552 –1556[Abstract/Free Full Text]
  31. Wayne AS, Kevy SV, Nathan DG. Transfusion management of sickle cell disease. Blood. 1993;81 :1109 –1123[Free Full Text]
  32. DiMatteo MR, Giordani PJ, Lepper HS, Croghan TW. Patient adherence and medical treatment outcomes: a meta-analysis. Med Care. 2002;40 :794 –811[CrossRef][Web of Science][Medline]
  33. Ballas SK. Iron overload is a determinant of morbidity and mortality in adult patients with sickle cell disease. Semin Hematol. 2001;38(1 suppl 1) :30 –36
  34. Vichinsky EP, Earles A, Johnson RA, Hoag MS, Williams A, Lubin B. Alloimmunization in sickle cell anemia and transfusion of racially unmatched blood. N Engl J Med. 1990;322 :1617 –1621[Abstract]
  35. Fullerton HJ, Adams RJ, Zhao S, Johnston SC. Declining stroke rates in Californian children with sickle cell disease. Blood. 2004;104 :336 –339[Abstract/Free Full Text]
  36. Barakat LP, Lutz M, Smith-Whitley K, Ohene-Frempong K. Is treatment adherence associated with better quality of life in children with sickle cell disease? Qual Life Res. 2005;14 :407 –414[CrossRef][Web of Science][Medline]
  37. Steen RG, Helton KJ, Horwitz EM, et al. Improved cerebrovascular patency following therapy in patients with sickle cell disease: initial results in 4 patients who received HLA-identical hematopoietic stem cell allografts. Ann Neurol. 2001;49 :222 –229[CrossRef][Web of Science][Medline]
  38. Walters MC, Patience M, Leisenring W, et al. Bone marrow transplantation for sickle cell disease. N Engl J Med. 1996;335 :369 –376[Abstract/Free Full Text]
  39. Walters MC, Storb R, Patience M, et al. Impact of bone marrow transplantation for symptomatic sickle cell disease: an interim report—multicenter investigation of bone marrow transplantation for sickle cell disease. Blood. 2000;95 :1918 –1924[Abstract/Free Full Text]
  40. Gulbis B, Haberman D, Dufour D, et al. Hydroxyurea for sickle cell disease in children and for prevention of cerebrovascular events: the Belgian experience. Blood. 2005;105 :2685 –2690[Abstract/Free Full Text]
  41. Ware RE, Zimmerman SA, Schultz WH. Hydroxyurea as an alternative to blood transfusions for the prevention of recurrent stroke in children with sickle cell disease. Blood. 1999;94 :3022 –3026[Abstract/Free Full Text]
  42. Ware RE, Zimmerman SA, Sylvestre PB, 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;145 :346 –352[CrossRef][Web of Science][Medline]
  43. Sumoza A, de Bisotti R, Sumoza D, Fairbanks V. Hydroxyurea (HU) for prevention of recurrent stroke in sickle cell anemia (SCA). Am J Hematol. 2002;71 :161 –165[CrossRef][Web of Science][Medline]
  44. Kirkham FJ, DeBaun MR. Stroke in children with sickle cell disease. Curr Treat Options Neurol. 2004;6 :357 –375[CrossRef][Medline]
  45. National Institutes of Health, ClinicalTrials.gov. Stroke With Transfusions Changing to Hydroxyurea (SWiTCH). Available at: www.clinicaltrials.gov/ct/show/NCT00122980. Accessed December 21, 2006
PEDIATRICS (ISSN 1098-4275). ©2007 by the American Academy of Pediatrics

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Facebook Facebook   Add to Reddit Reddit   Add to Technorati Technorati   Add to Twitter Twitter    What's this?


This article has been cited by other articles:


Home page
 Arch. Dis. Child.Home page
J Pappachan and F J Kirkham
Cerebrovascular disease and stroke
Arch. Dis. Child., October 1, 2008; 93(10): 890 - 898.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Mazumdar, M.
Right arrow Articles by Lieu, T. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Mazumdar, M.
Right arrow Articles by Lieu, T. A.
Related Collections
Right arrow Blood
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Facebook   Add to Reddit   Add to Technorati   Add to Twitter  
What's this?