← Back

STOP

Stroke Prevention Trial in Sickle Cell Anemia

Share Share Copied! Compare

Year of Publication: 1998

Authors: Robert J. Adams, MD, Virgil C. McKie, ..., and Myron A. Waclawiw

Journal: The New England Journal of Medicine

Citation: Adams RJ, et al. N Engl J Med 1998;339:5-11.

Link: https://www.nejm.org/doi/full/10.1056/NEJM199807023390102

PDF: https://www.nejm.org/doi/pdf/10.1056/NEJM199807023390102

Clinical Question

In children with Sickle Cell Disease (Hb SS) at high risk for stroke identified by TCD, does chronic blood transfusion therapy reduce the incidence of first-time stroke compared to standard care?

Bottom Line

Chronic blood transfusion therapy reduced stroke risk by 92% in high-risk children with sickle cell disease identified by TCD velocities ≥200 cm/s. The trial was stopped early due to overwhelming benefit and established TCD screening with prophylactic transfusion as standard of care.

        Major Points
        STOP is the landmark trial establishing primary stroke prevention in sickle cell disease — the first demonstration that a screening test (TCD) could identify high-risk children and an intervention (transfusion) could prevent strokes before they occurred.
130 children (age 2–16) with HbSS or Sβ0-thalassemia and abnormal TCD (time-averaged mean velocity ≥200 cm/s in ICA or MCA) randomized at 14 US and Canadian centers.
Stopped early by DSMB: 11 strokes in observation group vs 1 in transfusion group (92% risk reduction, p<0.001). The single stroke in the transfusion group occurred when transfusion was temporarily interrupted.
TCD screening identified ~10% of screened children as high-risk (≥200 cm/s). Without transfusion, stroke rate was ~10%/year in this group — one of the highest stroke rates in any population.
Transfusion target: reduce HbS to <30% with regular simple or exchange transfusions every 3–4 weeks. This became the standard protocol for primary stroke prevention in SCD worldwide.
Established TCD as a standard screening tool — AHA/ASA guidelines now recommend annual TCD screening starting at age 2 for all children with HbSS or Sβ0-thalassemia (Class I, Level A evidence).
Led directly to STOP II (2005), which showed that discontinuing transfusions after TCD normalization led to reversion to abnormal velocities in 39% and strokes — establishing that transfusion must be continued indefinitely.
Raised critical questions about transfusion burden: iron overload requiring chelation therapy, alloimmunization risk, infection risk, and quality of life impact of lifelong transfusions.
Hydroxyurea later emerged as an alternative in the TWiTCH trial (2016) for children with no MRA vasculopathy — but STOP transfusion remains standard for highest-risk patients.
One of the most impactful pediatric neurology trials ever — transformed SCD stroke from a common devastating complication to a largely preventable disease when screening is implemented.

      

Design

Study Type: Prospective, randomized, controlled, multi-center treatment trial.

Randomization: 1

Blinding: Blinded reading of TCDs and blinded adjudication of endpoints.

Enrollment Period: 18 months of TCD screening and randomization.

Follow-up Duration: Observation for stroke from entry through month 54.

Centers: Not specified

Countries:

Sample Size: 130

Analysis: Not specified

Inclusion Criteria

Age, from 24 months to 16 years.
Hemoglobinopathy diagnosis of Hb SS or Sβ⁰ thalassemia.
Willingness and ability to be screened as evidenced by signed parental informed consent and child's assent.
Availability for follow-up for at least 2 years.

Exclusion Criteria

Prior stroke.
An indication for chronic blood transfusion or a contraindication to chronic transfusion.
Participation in any study involving treatments which might confound the interpretation of the results of the proposed work.
Previous bone marrow transplant.
Evidence of HIV infection.
Generalized seizure disorder treated with anticonvulsant medication.
Pregnancy.

Baseline Characteristics

Characteristic	Control	Active

Arms

Field	Control	Chronic Transfusion Therapy
Intervention	Standard supportive care for sickle cell disease without prophylactic transfusion. Patients received usual clinical management including hydroxyurea if already prescribed, but no chronic transfusion program for stroke prevention.	Simple or exchange transfusions every 3–4 weeks, targeting HbS <30%. Initial transfusion to raise hemoglobin and reduce HbS fraction, then maintenance transfusions. Iron chelation with deferoxamine initiated when ferritin exceeded thresholds.
Duration	Median follow-up ~21 months (trial stopped early).	Median follow-up ~21 months (trial stopped early).

Outcomes

Outcome	Type	Control	Intervention	HR / OR / RR	P-value
First cerebral infarction (ischemic or hemorrhagic stroke) confirmed by clinical presentation and CT/MRI.	Primary	11 strokes in 67 patients (16.4%)	1 stroke in 63 patients (1.6%)	14.80%	<0.001
TCD velocity changes during follow-up	Secondary	Velocities remained elevated or increased	Significant reduction in TCD velocities with transfusion
Transient ischemic attacks	Secondary	Additional TIAs occurred in observation group	No TIAs in transfusion group
Iron overload	Adverse	N/A	Expected complication requiring chelation therapy with deferoxamine
Alloimmunization	Adverse	N/A	Transfusion-related antibody formation occurred in a subset of transfused patients

Criticisms

Small sample size (130 patients) — while the 92% risk reduction was overwhelming, the trial was underpowered for subgroup analyses and rare adverse events.
Open-label design — families and clinicians knew treatment allocation, potentially affecting reporting of neurological symptoms and threshold for imaging.
US/Canada only enrollment — generalizability to African, Caribbean, and other populations with high SCD burden but different healthcare infrastructure is uncertain.
No blinding of TCD operators — knowledge of treatment allocation could theoretically bias velocity measurements and endpoint adjudication.
Does not address the DURATION of transfusion — STOP established that transfusion prevents stroke but did not define when/if it could be stopped (STOP II later showed it cannot).
Iron overload burden not fully characterized — the lifelong transfusion commitment creates substantial iron chelation requirements, quality-of-life impact, and healthcare costs not captured in the trial.
Does not address hydroxyurea as an alternative — the trial predated widespread hydroxyurea use in SCD, leaving the question of whether a less burdensome therapy could achieve similar results (later addressed by TWiTCH).
TCD screening requires trained operators and standardized protocols — implementation in resource-limited settings where SCD is most prevalent remains challenging.
Selection of ≥200 cm/s threshold — while validated, some children below this threshold still develop strokes, suggesting the screening misses a proportion of at-risk patients. Conditional (170–199 cm/s) velocities have uncertain management.

Funding

National Institute of Neurological Disorders and Stroke (NINDS)

Based on: STOP (The New England Journal of Medicine, 1998)