Maaike van den Boomen1,2, Gert Jan H. Snel2, Christopher Nguyen1,3,4, Mary Kate Manhard1, David Sosnovik1,3,4, Rudi A.J.O. Dierckx2,5, Ciprian Catana1,6, David Izquerdo-Garcia1,6, Bruce R. Rosen1,6, Niek H.J. Prakken2, Ronald J.H. Borra5,7, and Kawin Setsompop1,6,8
1A.A. Martinos Center for Biomedical Imaging, Boston, MA, United States, 2Department of Radiology, University Medical Center Groningen, Groningen, Netherlands, 3Cardiovascular Research Center, Massachusetts General Hospital, Boston, MA, United States, 4Medicine, Harvard Medical School, Boston, MA, United States, 5Department of Nuclear Medicine and Molecular Imaging, University Medical Center Groningen, Groningen, Netherlands, 6Department of Radiology, Harvard Medical School, Boston, MA, United States, 7Medical Imaging Centre of Southwest Finland, Turku University Hospital, Turku, Finland, 8Division of Health Sciences and Technology, Harvard-MIT, Cambridge, MA, United States
Synopsis
Dynamic Cardiac BOLD imaging techniques can suffer
from signal variabilities due to changes in heartrate during a breath-hold. We
demonstrate that a 2-echo-GESE-EPI sequence can provide T2-
and T2*-weighted images simultaneously per heartbeat, but the dynamic changes in these images represent
physiological effects mixed with heartrate changes. We then proposed a 5-echo-GESE-EPI
for dynamic T2- and T2*-mapping per heartbeat as a readout for
myocardial oxygenation. These dynamically acquired T2- and T2*-values were demonstrated to increase over the time of a
breath-hold for healthy volunteers, while shown to remain constant and even
reverse in hypertension patients.
Introduction
Cardiac magnetic resonance offers several tools
to determine tissue characteristics and can assess tissue oxygenation by making
use of the different paramagnetic properties of hemoglobin[1]. Measure of myocardial
oxygenation could help determine the presence of microvascular dysfunction
which is believed to be associated with increased risks in cardiovascular
diseases, for example in hypertension[2]. The challenge for this
disease is that the oxygenation is not strongly altered as in vascular stenosis[3] and only a dynamic
imaging approach might show the alterations. A number of T2 and T2*
relaxation-based sequences can measure oxygenation in the myocardium[4], however there remains
a trade-off between quantitative or dynamic measurements. This work shows that
a dynamic signal intensity (SI) based sequence could be prone to heartrate
changes and proposes a 5-echoes modified gradient-echo-spin-echo-EPI (GESE-EPI)
sequence to mitigate this issue by providing T2- and T2*-maps
per heartbeat. The sequence is used to acquire breath-hold data on healthy
subjects and hypertension patients to determine oxygenation.
Methods
GESE-EPI sequences were applied to 7 healthy
volunteers (one with a BMI>25) on a 3T Skyra and 3 hypertension patients on
a 3T Prisma Siemens MRI with a Body-18
and Spine-32 coil
using: ECG triggering at end-diastole, acceleration RinplanexRzoom=2x3,
res=2.77x2.77x7mm, two asymmetric saturation pulses were used and FOVPE was
decreased to 38.1% (FOV=127x350x92mm). Two separate acquisitions were performed:
1)dual-echo GESE-EPI with TEs=9.8/38ms, TR=120ms, and 2)5-echo GESE with
TEs=9.8/21.61/34.24/46.86/59.48ms, TR=120ms. The resulting images from the
5-echo-GESE-EPI were used to create T2- and T2*-mapping
per heartbeat, using an iterative parameter fitting technique[5]. The images and
heartrates from the 2-echo-GESE-EPI acquisitions were compare with a Bloch
simulated Mxy-variation based on the heartrate and a MOLLI(5(3)3)-based T1-value.
A T2prep-FLASH series (TE=0,25,55ms) were
acquired to generate control T2-maps. For the healthy subjects FLASH series with 9
different TEs (3-21ms) were also acquired to provide control T2*-maps.
Semi-automatically drawn ROIs in the septum were used to compare the GE and SE
signal intensities and T2- and T2*-values from
the GESE maps with the control T2- and T2*-maps.
To determine dynamic changes of BOLD during a breath-hold a linear fit through
the T2- and T2*-values of the septum was
performed. Furthermore, the dynamically acquired GESE-EPI was performed twice
in the same healthy to confirm reproducibility of the oxygenation rates.
Results
GE- and SE-images were acquired per heartbeat
during a breath-hold using the 2-echo and 5-echo-GESE-EPI sequence. Fig1A shows
the increase in RR-interval time of 21.5% observed in one subject, which translate
to a predicted Mxy increase of 11.0% solely due to the heartrate.
The acquired GE and SE images showed an increase in SI over time(Fig1B), which is
a mixture of heartrate and oxygenation during the breath-hold. The
5-echo-GESE-EPI provides heartrate independent T2 and T2*-maps
per heartbeat that shows a reduced increase of the T2- and T2*-values(Fig1C).
Analysis of the septal ROI shows good correlation with the control mapping
approaches in both healthy subjects and hypertension patients(Fig2). Furthermore,
the change of the T2 and T2*-values during a breath-hold
shows a significant positive slope in all healthy subjects which was also reproducible
in a rescan(Fig3A). Most of the segments showed the same increase, except for
one near the air-, liver-, lung-interface and one of the coronaries(Fig3B). Different T2- and T2*-changes
were observed in the hypertension patients, where one subject showed a negative
slope(Fig4A) and two others show approximate no change(Fig4B). This difference
in slopes compared to healthy subjects might indicate an alteration of
myocardial oxygenation. However, while all healthy controls show positive T2-
and T2*-slopes, one subject with a BMI>25 showed a
smaller increase(Fig5). Discussion
Where a signal intensity based cardiac dynamic
BOLD sequence such as the 2-echo-GESE-EPI suffers from heartrate changes, the
5-echo-GESE-EPI shows T2 and T2*-maps per
heartbeat over the time of a breath-hold that not influences by heartrates. The
choice of slice position should not influence conclusions since microvascular dysfunction
is expected to be diffuse[6]. The mechanism behind
the increasing T2- and T2*-values is based on the increase
in CO2 levels of the blood during a breath-hold, which triggers
vasodilation[7]. Vasodilation could be
compromised in hypertension patients[8] and this non-invasive
oxygenation imaging approach might be able to assess this. Further research is
needed to determine the correlation of these oxygenation measures and potential
alterations in the oxygenation and vasculature.Conclusion
The 2-echo- and 5-echo-GESE-EPI were successfully
applied to the heart and enabled dynamic myocardial T2-
and T2*-imaging and mapping per heartbeat. The heartbeat-to-heartbeat
T2- and T2*-maps give a direct measure of the myocardial
oxygenation during a breath-hold. This approach could potentially offer a new MR
technique for the analysis of microvasculature diseases. Acknowledgements
This work was supported in part by NIH research
grants: R01EB020613, R01EB019437, P41EB015896, and the shared
instrumentation grants: S10RR023401, S10RR019307,
S10RR019254, S10RR023043
References
1. Pauling L, Coryell CD. The
Magnetic Properties and Structure of Hemoglobin, Oxyhemoglobin and
Carbonmonoxyhemoglobin. Proc Natl Acad Sci. 1936;22:210–6
2. Camici PG, Crea F. Coronary
Microvascular Dysfunction. N Engl J Med. 2007;:830–40
3. Fischer K, Yamaji K,
Luescher S, Ueki Y, Jung B, Tengg-kobligk H Von, et al. Feasibility of
cardiovascular magnetic resonance to detect oxygenation deficits in patients
with multi-vessel coronary artery disease triggered by breathing maneuvers.
2018;:1–11.
4. Friedrich MG, Karamitsos
TD. Oxygenation-sensitive cardiovascular magnetic resonance. J Cardiovasc Magn
Reson. 2013;15:1
5. Schmiedeskamp H, Straka M,
Newbould RD, Zaharchuk G, Andre JB, Olivot JM, et al. Combined spin- and
gradient-echo perfusion-weighted imaging. Magn Reson Med. 2012;68:30–40.
6. Salerno M, Kramer CM.
Advances in Parametric Mapping with Cardiac Magnetic Resonance Imaging. JACC
Cardiovasc Imaging. 2014;6:806–22.
7. Fischer K, Guensch DP,
Friedrich MG. Response of myocardial oxygenation to breathing manoeuvres and
adenosine infusion. Eur Heart J Cardiovasc Imaging. 2015;16:395–401.
8.
Arnold JR, Karamitsos TD, Bhamra-Ariza P, Francis JM, Searle N, Robson MD, et al.
Myocardial oxygenation in coronary artery disease: Insights from blood oxygen
level-dependent magnetic resonance imaging at 3 Tesla. J Am Coll Cardiol.
2012;59:1954–64