|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||
Background Prophylactic transfusion prevents strokes in children with sickle cell anemia who have abnormalities on transcranial Doppler ultrasonographic examination. However, it is not known how long transfusion should be continued in these children.
Methods We studied children with sickle cell disease who had a high risk of stroke on the basis of a transcranial Doppler screening examination and who had received transfusions for 30 months or longer, during which time the Doppler readings became normal. The children were randomly assigned to continued transfusion or no continued transfusion. Children with severe stenotic lesions on cerebral magnetic resonance angiography were excluded. The composite primary end point was stroke or reversion to a result on Doppler examination indicative of a high risk of stroke.
Results The study was stopped after 79 children of a planned enrollment of 100 underwent randomization. Among the 41 children in the transfusion-halted group, high-risk Doppler results developed in 14 and stroke in 2 others within a mean (±SD) of 4.5±2.6 months (range, 2.1 to 10.1) of the last transfusion. Neither of these events of the composite end point occurred in the 38 children who continued to receive transfusions. The average of the last two transcranial Doppler results before transfusion was started was the only predictor of the composite end point (P=0.05).
Conclusions Discontinuation of transfusion for the prevention of stroke in children with sickle cell disease results in a high rate of reversion to abnormal blood-flow velocities on Doppler studies and stroke. (ClinicalTrials.gov number, NCT00006182
[ClinicalTrials.gov]
.)
Transfusion has been used to prevent recurrent stroke in sickle cell disease for more than 20 years.7 However, cessation of transfusions is associated with recurrence of stroke, and there are no clinical or laboratory indicators to guide the duration of prophylaxis.8,9,10 The duration of the use of transfusion for the primary prevention of stroke is also unknown. We undertook a randomized, controlled trial, the Optimizing Primary Stroke Prevention in Sickle Cell Anemia (STOP 2) Trial, to determine whether we could limit prophylaxis by monitoring patients who had transcranial Doppler examinations after transfusions were halted and by resuming transfusions if the examination indicated a high risk of stroke.
Methods
Transcranial Doppler Examination
We conducted a study in which transcranial Doppler ultrasonographic examinations were performed by trained ultrasonographers who used similar equipment and software (2-MHz pulsed-wave Doppler, Nicolet/EME Companion or Nicolet/EME Pioneer). The Doppler studies were transmitted to central readers who were unaware of the treatment assignments. All results were recorded as the time-averaged mean of the maximum velocity in the middle cerebral or internal carotid artery and were classified as normal (all mean velocities of <170 cm per second), conditional (at least one mean velocity of 170 to 199 cm per second but none 
200 cm per second), abnormal (at least one mean velocity of at least 200 cm per second), or inadequate (no information available on one or both middle cerebral arteries).2
Magnetic Resonance Imaging
Magnetic resonance imaging (MRI) of the brain was required before patients underwent randomization, at the end of the study, and at the time of suspected neurologic events. The study protocol included axial T1-weighted spin–echo images (repetition time, 400 to 800 msec, and echo time, 10 to 30 msec). Axial spin–echo and fluid-attenuated inversion recovery (FLAIR) T2-weighted images, spin–echo or fast (turbo) spin–echo images with slices at a thickness of 5 mm, coronal spin–echo and FLAIR images (with the use of the same protocol as axial), and diffusion-weighted imaging (gradient strength, B 1000), with axial images in the x, y, and z planes, were performed. Magnetic resonance angiography (MRA) was standardized according to a protocol of image acquisition with the use of a three-dimensional time-of-flight technique using the smallest feasible voxel size (small field of view, 15 to 20 cm, and matrix, 256 by 256 to 256 by 512) and the shortest obtainable echo time (<5 msec) to minimize the flow-related loss of the intravascular signal. All images were reviewed for the presence, size, and location of ischemic lesions by experts who were unaware of the treatment assignment. Middle cerebral and carotid angiograms were scored as normal or as showing a stenosis that was mild (<25 percent), moderate (25 to 75 percent), or severe (>75 to <100 percent) or an occlusion.
Eligibility and Monitoring of Patients
Figure 1 shows the design of the trial and the eligibility criteria. This trial was an extension of the previous STOP study, in which children with abnormal velocities on transcranial Doppler ultrasonographic examination were administered transfusions to prevent a first stroke. Children whose Doppler studies normalized after 30 or more months of transfusion were eligible for the present trial. In addition, children who had not participated in the previous STOP study whose condition met the criteria for eligibility and treatment were also eligible for the present study. All participants were required to have normal results on two consecutive transcranial Doppler studies performed at least two weeks apart while they were receiving prophylactic transfusions and within four months before randomization. The protocol was approved by the institutional review boards at the participating institutions. Written informed consent was obtained from a parent or guardian of the child in all instances, and the children's assent was obtained, when appropriate.
|
Blood for transfusions was matched for C, D, E and Kell antigens. Chelation therapy with the use of deferoxamine was recommended if serum ferritin levels exceeded 2500 ng per milliliter, but the type of transfusion (simple, manual exchange, or automated erythrocytapheresis) and initiation of chelation treatment were at the discretion of the investigator. Patients who were randomly assigned to the transfusion-halted group could receive transfusions that were indicated to treat complications of sickle cell disease. Initiation of hydroxyurea therapy11 or regular transfusion in a patient assigned to this group was designated as a crossover and prompted censoring of data on the patient as of the date of treatment.
Information on new neurologic symptoms was solicited quarterly, and changes in medication, interim illness, and episodic transfusion were recorded. A complete blood count, reticulocyte count, quantitative hemoglobin electrophoresis, and alloantibody screening were performed before each transfusion and quarterly; and serum ferritin levels were measured at the core laboratory (at the Medical College of Georgia in Augusta). Measurements of serum alanine aminotransferase, 
-glutamyltransferase, lactate dehydrogenase, and bilirubin and screening for hepatitis B and C viruses were performed annually.
Definition and Adjudication of End Points
The primary composite end point was a stroke (cerebral infarction or intracranial hemorrhage) or reversion to abnormal velocity on transcranial Doppler ultrasonography, defined as two consecutive studies with abnormal velocities, three consecutive studies with an average velocity of 200 cm per second or more, or three consecutive inadequate studies plus evidence of severe stenosis on MRA. Suspected strokes were adjudicated by experts unaware of the treatment assignment using clinical data and all available imaging data. Stroke was defined as persistent neurologic abnormalities or transient symptoms accompanied by a new cerebral lesion appropriate to the patient's clinical presentation.
Statistical Analysis
A sample size was calculated that would provide 80 percent power, with a two-tailed type I error rate of 0.05, to detect an absolute difference of 50 percentage points between the two study groups in the proportion of patients in whom a stroke occurred or who reverted to being at high risk for stroke on the basis of transcranial Doppler examination over three years. Patients were stratified at randomization according to the presence or absence of ischemic lesions on MRI; random, permuted blocks of four or six patients were used within each group as defined by MRI. Institutional balancing with a tolerance of two patients per site was imposed to maintain an approximate balance in treatment assignments at each site.12 Power calculations for the log-rank test were performed with the use of SAS software (version 8, SAS Institute),13 with the software program of Lakatos,14 for a 54-month study involving 50 patients in each group, with 60 of the patients enrolled during the first 12 months and 40 during the next 24 months; after recruitment ended, there were 18 months of follow-up. Eligible patients underwent randomization with equal probability of continuing or halting transfusion.
Baseline characteristics of the patients in the two groups were compared with the use of Student's t-tests for continuous variables and chi-square tests (for the presence or absence of lesions on MRI) or Fisher's exact test (for male or female sex) for categorical variables. Laboratory values 6 months (and in some cases, 12 months) after randomization were compared with baseline values by Student's t-test. All reported P values are two-sided and were not adjusted for multiple testing.
Event rates were compared with the use of a log-rank test.15 Potential predictors of the primary composite end point preselected for analysis were sex, age at randomization, presence or absence of lesions on baseline MRI, transcranial Doppler readings before and after transfusion, average hemoglobin S levels before transfusion in patients receiving transfusions, and the number of transfusions received during the 30 months before randomization. Post hoc analyses were also performed to examine whether variables that have been related to cerebral infarction16 were associated with stroke or reversion to abnormal velocities on Doppler examination in our study. Each of the variables (recurrent or proximate acute chest syndrome, transient ischemic attack, low total hemoglobin levels, and elevated blood pressure) was tested in a separate model.
Four interim analyses and one final analysis were planned for the composite end point, with the use of two-tailed tests, and separately for stroke alone, with the use of one-tailed tests. In both cases, the Lan and DeMets spending function that approximates an O'Brien–Fleming boundary was used.17
Results
The trial was stopped by the National Heart, Lung, and Blood Institute on the advice of the data safety and monitoring committee because of concern about safety at the fourth interim analysis. There were no significant differences in baseline characteristics among the patients in the two groups (Table 1). End-point events occurred in 16 patients (age at event, 8.4 to 19.7 years; median age, 11.8 years), all of whom were assigned to no continued transfusion; 14 of the events were reversion to abnormal velocities on transcranial Doppler studies, and 2 were strokes. The median time from randomization to an end-point event was 3.2 months (range, 2.1 to 10.1), and the mean (±SD) was 4.5±2.6 months. As compared with those in the continued-transfusion group, almost half the patients in the transfusion-halted group had a primary event within 10 months after randomization (Figure 2). There was a significant (P<0.001) difference between the two groups in the number of patients having an end-point event. Eleven neurologic events were adjudicated, and of these, two were determined to be strokes. In each of the two children who had a stroke, the stroke occurred after the child had a single Doppler result showing an abnormal velocity but before a confirmatory test was performed. One child had a first abnormal result (a velocity of 210 cm per second) on day 281 after randomization and presented with a symptomatic new right-hemisphere infarction on day 295. The other child had a first Doppler result showing abnormalities on day 136 (a velocity of 231 cm per second) and had a stroke on day 144.
|
|
|
Of 38 patients assigned to continued transfusion, 32 were still receiving transfusions at the end of the study, 5 stopped transfusions, and 1 died of complications of acute chest syndrome. Of 41 patients assigned to no continued transfusion, 9 recommenced transfusion or started hydroxyurea treatment, and 16 were being followed without treatment or end-point events at the end of the study (8 of the 16 for more than 25 months).
At the end of the trial, 35 patients (93 percent) assigned to continued transfusion and 31 (76 percent) assigned to no continued transfusion were receiving chelation. No cases of hepatitis C were identified among the 68 patients who had serologic testing at the end of the study. One new case of alloimmunization in a patient in the continued-transfusion group was identified (anti-Kpa was detected on day 39 after randomization). At baseline, ferritin levels were 3274±1718 ng per deciliter among those assigned to continued transfusion, as compared with 3005±1504 ng per deciliter among those assigned to no continued transfusion (P=0.46). However, after 12 months, the mean levels of ferritin were 3562±1536 ng per milliliter (25 patients) and 1832±916 ng per milliliter (11 patients), respectively (P=0.002).
Discussion
Although chronic transfusion is effective in preventing stroke in sickle cell disease, this therapy carries immediate and cumulative risks, especially with regard to iron loading.18 Our goal was to determine whether we could safely discontinue protective transfusion in selected patients by monitoring them with transcranial Doppler ultrasonography and reinstituting transfusion if there were abnormal velocities on the Doppler study. We investigated this in children who we thought had a low risk of stroke: all originally had abnormal velocities on Doppler studies that normalized during a trial of prophylactic transfusion, and there was no evidence in these patients of clinically significant intracranial arterial stenosis on MRA. However, the difference in the number of primary end-point events exceeded the stopping boundary, and despite frequent Doppler studies, stroke was not prevented in two children.
How transfusion prevents stroke in sickle cell disease is unknown. Although there is a proportional reduction in flow velocity with increased levels of total hemoglobin,19 the increase in red-cell mass may not be the only beneficial effect. Transfusion may reduce red-cell adhesion20 to endothelium, thereby decreasing vascular injury by sickle red cells. Reduction of intravascular hemolysis and the resulting free hemoglobin, which consumes nitric oxide,21 may increase the capacity for cerebral vasodilatation in response to ischemic stress. In the STOP 2 study, plasma free hemoglobin was not measured, but lactate dehydrogenase, which has been used as a surrogate marker for hemolysis in sickle cell disease,22 was measured at baseline and yearly. There were no significant differences between the continued-transfusion group and the transfusion-halted group in baseline levels of lactate dehydrogenase (479±210 U per liter and 444±266 U per liter, respectively). At one year the levels of lactate dehydrogenase had increased from baseline in the transfusion-halted group (616±240 U per liter) but not in the continued-transfusion group (469±164 U per liter; P=0.046 by Student's t-test for the difference between the two groups at one year). This finding suggests that one of the benefits of regular transfusion may be to reduce intravascular hemolysis, but further direct studies measuring plasma free hemoglobin23 in relation to other effects of transfusion, such as reduction of hemoglobin S–containing red cells and increased total hemoglobin, are needed.
The risk of stopping transfusion that is demonstrated in our study highlights the need for alternative therapies to prevent stroke or better ways to manage iron overload. Reduced transfusion intensity, which allows the target value for the percentage of hemoglobin S to rise from 30 percent to 50 or 60 percent after some years of intensive transfusion, has been tried24,25 after sickle cell–related stroke without an apparent increase in the risk of stroke, but randomized trials comparing different intensities of transfusion have not been reported. A regimen of phlebotomy and hydroxyurea was substituted for chronic transfusion for secondary prevention of stroke with encouraging results,26 although a randomized trial is still needed. Transcranial Doppler screening has been used since 1992 by Bernaudin et al.27 Children with abnormal blood-flow velocities on Doppler ultrasonography are offered either hematopoietic stem-cell transplantation or transfusion. Among those electing transfusion, if the Doppler results normalize within three months after transfusion is started, and if severely stenotic arterial lesions are absent, the children are switched from transfusion to hydroxyurea therapy and followed with transcranial Doppler tests.27 This approach should be tested in a randomized, controlled trial.
In the STOP 2 study, eight patients (20 percent of those assigned to no continued transfusion) who were observed for more than 25 months without prophylactic transfusion therapy had neither a stroke nor reversion to abnormal velocities on Doppler studies. Unfortunately, there is no way to identify such patients prospectively. In the 209 patients who underwent randomization in the STOP and STOP 2 studies, there were 20 strokes (18 in STOP and 2 in STOP 2). The last transcranial Doppler examination before the stroke showed abnormal velocities in all cases, confirming that abnormal velocities on transcranial Doppler ultrasonography are a good indicator of the risk of stroke, both before transfusion is initiated and after it is stopped. These results suggest that if stroke is to be prevented after transfusion is stopped, transcranial Doppler examinations must be performed at frequent intervals and transfusion resumed expeditiously. Although morbidity and mortality from stem-cell transplantation are a concern, limited experience suggests that cerebrovascular disease does not progress after stem-cell transplantation.28 Given the transfusion dependence demonstrated in the STOP 2 study, and given the problems associated with long-term transfusion to prevent stroke, stem-cell transplantation should be considered as an option for primary stroke prevention.
Supported by grants (U01 HL 052193 and U01 HL 052016) from the National Heart, Lung, and Blood Institute.
No potential conflict of interest relevant to this article was reported.
The article is dedicated to the memory of Katie Allen, R.N., Charles Pegelow, M.D., and David Ode, M.D.
We are indebted to the patients and their families for their contribution to this research.
Source Information
Robert J. Adams, M.D., and Donald Brambilla, Ph.D., of the STOP 2 investigative team take responsibility for the content of this article.
Address reprint requests to Dr. Adams at the Department of Neurology, Medical College of Georgia, 1429 Harper St., HF 1154, Augusta, GA 30912, or at rjadams{at}mcg.edu.
References
The STOP 2 team of principal investigators is listed according to site (in descending order of the number of patients who underwent randomization): M. Abboud, J. Barredo, C. Brown, Medical University of South Carolina, Charleston; O. Alvarez, C. Pegelow, University of Miami School of Medicine, Miami; V. McKie, K. McKie, Medical College of Georgia, Augusta; E. Vichinsky, K. Quirolo, Children's Hospital of Oakland, Oakland, Calif.; C. Driscoll, Children's National Medical Center, Washington, D.C.; C. Daeschner, East Carolina University, Greenville, N.C.; S. Piomelli, M. Lee, Columbia University, New York; R. Iyer, University of Mississippi Medical Center, Jackson; P. Lane, B. Gee, B. Files, T. Adamkiewicz, C. Davis, Emory University School of Medicine, Grady Health System, Morehouse School of Medicine, and Children's Healthcare of Atlanta, Atlanta; M. Kirby, Hospital for Sick Children, Toronto; N. Olivieri, University Health Network, Toronto; B. Berman, A. Villella, Rainbow Babies and Children's Hospital, Cleveland; G. Woods, Children's Mercy Hospital, Kansas City, Mo.; W. Wang, St. Jude Children's Research Hospital, Memphis, Tenn.; J. Kwiatkowski, The Children's Hospital of Philadelphia, Philadelphia; Baltimore–Washington Sickle Cell Research Consortium (J.F. Casella, Johns Hopkins University School of Medicine, Baltimore; J. Wiley, Sinai Hospital of Baltimore, Baltimore; N. Grossman, University of Maryland, Baltimore; A. Shad, Georgetown University, Washington, D.C.); L. Hilliard, University of Alabama at Birmingham, Birmingham; A. Provisor, Columbus Regional–The Medical Center, Columbus, Ga.; S.T. Miller, SUNY Downstate Medical Center, Kings County Hospital Center, Brooklyn, N.Y.; T. Coates, University of Southern California, Los Angeles; R. Warrier, D. Ode, Louisiana State University, New Orleans; C. Scher, Tulane University Medical School, New Orleans; K. Kalinyak, Children's Hospital Medical Center, Cincinnati; National Heart, Lung, and Blood Institute (NHLBI): D.R. Bonds (program officer), R.B. Moore, M. Mathis, L. Barbosa, M. Waclawiw.
NHLBI-appointed data safety and monitoring board: V. Mankad (chair), A. Dyer, T. Kinney, L. McMahon, S. Pavlakis, P. Roberson, J. Seibert, F.W. Schmidt, Jr.; Data coordinating center at New England Research Institutes: D. Brambilla (principal investigator), S. McKinlay, D. Gallagher, T. Mansolf, S. Granger, S. Harkness, M. Therencial, M. Pouliot, C. Pollari, M. Berkman, S. Della Grotta, L. Enos, R. Glauber, S. Harter, R. Lagos, K. Morales, M. Pare, T. Wiegand; Central administrative center, Medical College of Georgia, Augusta: R.J. Adams (principal investigator), M. Good, N. Odo, D. Ramsingh, E. Rohde, J.S. Schweitzer, R.K. Wright; Transcranial Doppler (TCD) training center, Medical College of Georgia, Augusta: F.T. Nichols (director), A. Jones, M. Sahm; Core laboratory, Medical College of Georgia, Augusta: A. Kutlar (director), J. Harbin, L. Holley; Clinical and data coordinators: K. Allen, L. Ash, S. Bankston, D. Barnes, S. Bergeron, M. Blumenstein, C. Brown, S. Carson, J. Chow, M. D'Angelo, L. Dabbar, E. Dackiw, M. DeBarr, S.M. Dixon, D. Dodge, K. Doig, M. Doyle, E. Eckroth, H. Enninful-Eghan, T. Faircloth, G. Fortner, D. Gordon, B. Gould, H. Gutin, E. Hackney-Stephens, J. Handy, D. Harris, D. Haughey, E. Hirsch, D. Jack, S. Johnson, C. Kendig, J. Luden, H. Machen, J. Marasciulo, A. Marra, B. Martin, M. Merelles-Pulcini, K. Murch, T. Murdock, A. Mynatt-Norman, C. O'Haver, H. Poplick, E. Randall, B. Record, K. Rey, C. Rhoad, G. Roath, J. Routhieux, K. Ruff, S. Somjee, A. Stevens, K. Stewart, G. Taplin, I. Tillman, T. Walker, D. Wright, A. Zaki; TCD examiners: J. Adams, K. Allen, N. Anderson, G. Bell, L. Bowman, I. Campo-Bustillo, M. DeBarr, R. DeJong, K. Doig, G. Fortner, B. Gould, D. Griffith, T. Hogan, A. Jones, A. Lester, J. Luden, A. Mann, L. Mollo, B. Perret, K. Rey, A. Spinks, K. Stewart, S. Trocio, L. Utley, A. Wong; Neurologists: G. Chari, R. Curless, R. Khan, K. Krohn, R. Lopez Alberola, D. MacGregor, J. Murphy, Y. Park, M. Patterson, B. Philbrook, A. Reddy, A. Rose, F. Silver, V. Vedanaraynan, M. Wiznitzer, K. Yohay; Radiologists, neuroradiologists, and ultrasound radiologists: S. Blaser, B. Bowen, R. Figueroa-Ortiz, D. Greer, K. Helton, G. Hotson, A. Khandji, L. Lowe, J. Nath, M. Nelson, B. McCarville, S. Palas, G. Vezina; Stroke adjudication panel: S. Roach (chair), L. Caplan, D. DeWitt; Magnetic resonance reading panel: R. Zimmerman (chair), J. Bello, F. Moser; Manuscript preparation committee: R.J. Adams (chair), D. Brambilla, S. Miller, D. Bonds.
| |||||||||||||||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
0 hemoglobin SS–