Microsoft word - ismrm2005-004043.doc

MRI to assess the contribution of gastric peristaltic activity and tone to the rate of liquid gastric emptying in health
A. Steingoetter1, M. A. Kwiatek2, A. Pal3, G. Hebbard4, M. Thumshirn2, M. Fried2, J. Brasseur3, W. Schwizer2, P. Boesiger1 1University and ETH Zurich, Institute for Biomedical Engineering, Zurich, Switzerland, 2University Hospital, Division of Gastroenterology, Zurich, Switzerland, 3Pennsylvania State University, Department of Mechanical and Bio Engineering, University Park, Pennsylvania, United States, 4Royal Melbourne Hospital, Department of Gastroenterology, Parkville, Victoria, Australia Introduction: Gastric emptying requires coordinated motor activity of gastric fundus, antrum, pylorus and proximal duodenum. However, the pattern of motor and
pressure events, their coordination with pyloric opening, and their relative contributions to gastric emptying is still incompletely understood (1). In this study, effect of
gastric tone, antral peristalsis and pyloric contractility on the rate of gastric emptying was studied using concurrent MRI and high-resolution manometry (HRM).
Pharmacological relaxation of the fundus was used to clarify whether gastroduodenal pressure differences caused by fundal tonic contractions ("pressure pump") or high
antral pressure waves ("peristaltic pump") is the primary force for gastric emptying of liquid. We are presenting the physiological results provided by the placebo group
through MRI and HRM.
Methods: Nine healthy subjects underwent a randomized single-blind, placebo
controlled study with clonidine (2) (α2-adrenergic receptor agonist, fundal relaxant
that does not affect gastric emptying) and sumatriptan (3,4) (5-HT1-receptor
agonist, fundal relaxant that delays gastric emptying). Each volunteer was studied
on three separate occasions with randomized drug regimens. Each day a multi-
lumen nasogastric tube assembly (Dentsleeve Pty Ltd, Parkside, Australia) was
placed across the pylorus and then the subject lied in right decubitus position in the
MRI system (1.5T Intera, Philips Medical Systems, Best, The Netherlands). On each
study day, participants received either: a) clonidine i.v. and placebo s.c.; b)
sumatriptan s.c. or placebo i.v. or c) placebo i.v and s.c. After drug administration,
500 ml 10 % glucose (200 kcal) marked with 500 µM Gd-DOTA (DOTAREM®)
was ingested. During concurrent and simultaneous HRM measurements, MRI
"volume scans" of the complete gastric region were performed every 5 min after the Fig. 1: Sagittal MRI image of the antro- Fig. 2: Enlarged antroduodenal region
meal ingestion until 90 min. A steady state free precession (SSFP) sequence with 40 duodenal region. Manometry tube and of a sagittal MRI image. The
sagittal slices, TR/TE = 4.6/2.3 ms, FOV = 350x205 mm2, ∆z = 10 mm, matrix = pylorus are clearly visible. Defined detection of pylorus and occlusion
256x164, Tscan = 15 s was used. Each volume scan was followed by a 155s long gastric axis is indicated.
dynamic scan sequence ("motility scan") using parallel imaging method SENSE to
increase image acquisition rate. Three parallel oblique image slices (120 dynamic
images each) positioned along the distal gastric antrum were acquired during free
breathing using a SSFP sequence (TR/TE = 4.0/1.8 ms, FOV = 350x205 mm2, ∆z =
8 mm, matrix = 256x164, SENSE factor = 1.6). Five rectangular parallel imaging
surface coils (h = 20 cm, w = 10 cm) were placed around the abdomen for signal
detection. Gastric emptying rates and accommodation were derived from MRI
"volume scans" by segmenting stomach and meal volumes. Average peristaltic
frequency (f), velocity (v), occlusion amplitude (α) and width (w) of antral
contraction waves (ACWs) were derived from detailed analysis of MRI motility
scans. F and v were calculated using a manually defined distal gastric axis (Fig. 1).
The position of the pylorus was indicated in the images to relate the characteristics
of each ACW to the distance from the pylorus (Fig. 2). The propagation of each Fig. 3: Manometry recodings (color
ACW as detected on MRI was incorporated into the manometry data to assess Fig. 4: Rate of gastric emptying over
relationship of ACWs to changes of intragastric pressure and pyloric resistance (Fig. 90 min. Data is given as mean ± SE. The 3). Correlations between the above gastric parameters and gastric emptying rates large inter-individual variation led to the were assessed to investigate their contribution to the gastric emptying process. Results: Stomach and meal volume as well as antral peristaltic activity were detected over the complete study
period in all volunteers. Gastric emptying rates showed large variation between individuals and also during the
emptying process (Fig. 4). All detected ACWs were non lumen occlusive. 80 % of all contractile events were
propagating ACWs of constant speed, which terminated within 1.2 ± 0.6 cm of the pylorus. During first 20
min, f = 1.9/min and v = 2.4 mm/s and thereafter f = 2.6/min and v = 2.9 mm/s. α decreased as ACWs
propagated towards the pylorus (Fig. 5). W also decreased as the ACWs approached the pylorus (Fig. 5). No
correlation was found between peristaltic frequency, velocity, or occlusion amplitude of ACW and the gastric
emptying rate. Greater w was associated with increased gastric emptying rates. Manometry could only detect
31 % of all propagating ACWs detected with MRI. However, pyloric contractile activity could clearly be
determined from manometry recordings.
Discussion: This study demonstrates that MRI allows the comprehensive assessment of human gastric motor
function. Physiological findings demonstrate that gastric peristalsis does not regulate the rate of gastric
emptying of a caloric liquid meal under normal physiological conditions. The use of the SSFP sequence
together with multiple abdominal surface coils and parallel imaging technique SENSE, dynamic high-
resolution image data sets with excellent quality were acquired, allowing detailed analysis of the characteristics
of antral contractility. With the dynamic measurement of three slices at a rate of 2.3 per second the continuous
detection of the antral region and the pylorus was assured during free breathing motility scans. Image analysis
Fig. 5: Average magnitude of occlusion amplitude
of motility data was very time consuming, as sophisticated and robust segmentation and feature detection (α) and width (w) contraction wave plotted over its methods for this kind of MRI data are not yet available. A major limitation of MRI for the investigation of distance XCW from the pylorus (XCW = 0). gastric function is its inability to quantify the force of intragastric and pyloric pressure events. These are
important for understanding physiology of gastric emptying and currently are only obtainable with intraluminal manometry. Hence, concurrent manometry recording, as
performed here, complement functional information based on detailed anatomy visible with MRI. Analysis of image and pressure data under influence of fundal
relaxants will provide further insight into the interplay between gastric fundus, antrum and pylorus during the emptying process.
References: (1) Indireshkumar K et al, Am J Physiol Gastrointest Liver Physiol 278: G604-616, 2000; (2) Thumshirn M. et al, Gastroenterology 1999;116; (3) Tack J et
al, Gastroenterology 1994; 108:A696; (4) Coulie B et al. Am J Physiol 1997; 272:G902-908.
Acknowledgements: This study was supported by the Swiss National Science Foundation, SNF Grant 31.55'932.98.
Proc. Intl. Soc. Mag. Reson. Med. 13 (2005)

Source: http://www.mr.ethz.ch/abstracts/files/ismrm05_00426.pdf