Introduction
In pediatric patients with sickle cell anemia (SCA), stroke is a major
cause of morbidity and mortality with an estimated 1.9-11% risk of
stroke by the age of 20 (1-7). Specific screening and treatment
guidelines aimed at both primary and secondary prevention have
successfully reduced the incidence of overt strokes in this high-risk
population (4). The Stroke Prevention Trial in Sickle Cell Anemia (STOP)
trial showed significant primary stroke prevention by reducing the risk
of first stroke by 90% with chronic transfusions in those patients at
increased risk as determined by Transcranial Doppler (TCD) (4). This
study lead to a national recommendation to conduct annual TCD screening
for all children with SCA between the ages of 2-16 years (8).
Patients with SCA who have conditional or abnormal TCDs or neurologic
symptoms may undergo regular neuroimaging with Magnetic Resonance
Imaging (MRI) and Magnetic Resonance Angiogram (MRA) to confirm cerebral
vasculopathy and stratify stroke risk. Risk stratification using MRA is
recommended by the American Society of Hematology in selection of a
stroke prevention treatment plan (9). Current recommendations based on
the multicenter, randomized controlled trial TWiTCH (TCD with
Transfusions Changing to Hydroxyurea) suggest that a child aged 2-16
with abnormal TCD results who has been receiving transfusion therapy for
at least 1 year with low-risk MRA imaging can be transitioned off
transfusions and onto hydroxyurea treatment (9, 10). Determination of
low-risk or high-risk requires MRA grading of vasculopathy, highlighting
the importance of accurate MRA assessment.
Standard non-contrast MRA is accomplished with a 3-dimensional time of
flight technique and allows assessment of the patency and caliber of
intracerebral arterial vessels (11). Technical artifacts can influence
the appearance of arterial structures on MRA (12), especially in regions
of arterial tortuosity or bifurcation, sites commonly affected in
patients with SCA. These technical artifacts can lead to the erroneous
interpretation of arterial stenosis, and thereby lead to inappropriate
risk stratification and treatment selection for primary stroke
prevention, such as initiation or continuation of chronic transfusion
therapy or transition to hydroxyurea. While chronic transfusion therapy
can significantly lower stroke risk (13), it has notable risks including
iron overload, alloimmunization, and infection (14-16).
Investigators in the Stroke with Transfusions Changing to Hydroxyurea
(SWiTCH) trial developed standardized scanning protocols for their entry
and exit brain MRI, MRA, and TCD studies. The SWiTCH MRA scanning
protocol specified an echo time of <5 msec (12). Echo time or
TE, describes the time between application of radiofrequency excitation
pulse and peak of the signal induced in the coil, measured in
milliseconds (msec). While SWiTCH primary objective was to compare
standard transfusion therapy to hydroxyurea in preventing recurrent
stroke and quantitative liver iron content (12), the standardized
imaging scanning protocols developed provide an opportunity for further
study of flow artifact and its clinical importance. These MRA scanning
protocols were also utilized in the TWiTCH trial to grade vasculopathy
as low and high risk and guide treatment decisions. The generalizability
and real-world application of the results of the TWiTCH trial could
potentially depend on close adherence to study MRA protocols. Few
reports have mentioned minimizing TE to <5 msec to limit
flow-related artifact (5, 6, 17), however, no prior studies have
reported the clinical implications of these protocols.
The primary objective of our study was to document any change in stroke
prevention therapy, particularly discontinuation of chronic
transfusions, that could be attributed to the implementation of the
standardized MRA scanning protocol based on the SWiTCH protocol. We
hypothesized that the implementation of a standardized MRA protocol
would lead to re-classification of the level of cerebral stenosis and
impact treatment plan for a subset of patients.