Larghat - BSCMR 2010 - Reproducibility asessment

Document Sample
Larghat - BSCMR 2010 - Reproducibility asessment Powered By Docstoc
					Reproducibility assessment of First Pass Perfusion CMR in rest and stress for estimation
of myocardial perfusion
Abdulghani M. Larghat, MD1, Aleksandra Radjenovic1, PhD1, Neil Maredia, MRCP1, John Biglands1, John P.Greenwood,
PhD1, Sven Plein, MD, PhD1 .
 University of Leeds, United Kingdom,
Corresponding Author: Dr S Plein. Email:
     1.   To assess intraobserver, interobserver and interstudy reproducibility of myocardial blood flow
          estimation from first pass CMR in normal healthy volunteers at rest and stress.
   2.     To compare of semi quantitative and quantitative measurements .
 Cardiovascular magnetic resonance (CMR) has demonstrated excellent results in terms of accuracy and
interstudy reproducibility for the assessment of left ventricular (LV) and right ventricle (RV) function, mass
and volume parameters. However, reproducibility data of first pass CMR perfusion for estimation of
transmyocardial blood flow in rest and stress in normal heart are still sparse. Assessment of CMR perfusion
reproducibility is crucial to monitor response of the therapies which affect myocardial blood flow.
11 volunteers (6males,Mean age 33±7 years) underwent CMR perfusion on 2 separate days (Mean inter
study duration=84±111 days, Median duration = 7 days) at the same time of the day for each volunteer. were
studied on a 1.5T Philips Intera system during adenosine stress (140 mcg/kg/min for 3 minutes, 0.05
mmol/kg Gd-DTPA) and at rest. A pulse sequence optimised for acquisition of a single midventricular slice at
the time of minimal systolic cardiac motion was used (saturation recovery segmented gradient echo, 2x
SENSE TR/TE/flip 2.7ms/1.0/15°, typical FOV 380x380mm, matrix 160x160, slice thickness 10mm,
preparation pulse delay 150ms, shot duration 130ms). Endo and epicardial contours were drawn and the
slice segmented into 6 equidistant sectors.
Based on LV and blood pool time intensity curves, an in-house mathematical flow Fermi Function
deconvolution algorithm was implemented to estimate the absolute myocardial blood flow (1). All
measurements were performed by one observer, who repeated analysis of the first CMR scan after 4 weeks.
In addition, a second observer performed separate analysis of the first study .
     1.   Reproducibility was highest for intra observer followed by interobserver and lowest for interstudy.
     2.   Semi quantitative was generally more reproducible than fully quantitative analysis. See table (1).
1.   First pass perfusion CMR has demonstrated reasonably good reproducibility in comparison with imaging
     modalities (2).
2.   The highest variability of perfusion CMR was at the inter study perfusion assessment especially in the
     quantitative analysis. This might be mainly attributed to inaccurate transfer of the LV and blood pool
     input function through our implemented deconvolution algorithm. Also due to the phenomenon of
     myocardial blood flow heterogeneity rather than to a methodological error alone.

1. Jerosch-Herold et al. Med Phys 25: 73-84, January, 1998.
2. Kaufmann, P.A., et al, Journal of Nuclear Medicine, 1999. 40(11): p. 1848-56.