Contents lists available at SciVerse ScienceDirect
Magnetic Resonance Imaging
j ourna l homepage: www.mr i journa l .com
7T μMRI of mesenteric venous ischemia in a rat model: Timing of the appearanceof findings
Francesco Somma a, Daniela Berritto a, Francesca Iacobellis a,⁎, Nicola Landi a, Carlo Cavaliere a,Marco Corona b, Serena Russo c, Roberto Di Mizio d, Antonio Rotondo a, Roberto Grassi a
a Institute of Radiology, Second University of Naples, P.za Miraglia, 2; 80138 Naples, Italyb Biotechnology Center, A.O.R.N. Cardarelli, Via A. Cardarelli, 9; 80131 Naples, Italyc Department of Pathology, “Maresca” Hospital, Via Montedoro, 80059 Torre del Greco (NA), Italyd Institute of Radiology, San Massimo Hospital, Via Battaglione Alpini l'Aquila, 1; 65017 Penne (PE), Italy
Objectives: The aim of this study is to analyze the chronological development of macroscopic, microscopicand magnetic resonance imaging (MRI) findings in a rat model of Superior Mesenteric Venous (SMV)ligation, and to evaluate the role of MRI in the diagnosis of mesenteric venous thrombosis.Methods: Thirty adult Sprague–Dawley ratswere used and divided in two different groups that underwent adifferent surgical model and a different monitoring of ischemic damage. Group I underwent macroscopicaland histological observation; Group II underwent 7T μMRI evaluation and histological analysis.Results: The first alterations occurred 30min after SMV ligation and progressively worsened until the eighthhour. The morphological and MRI findings showed the same course.Conclusions: This study provides a systematic evaluation of early anatomopathological and MRI findings
following the SMV ligation. MRI allows to identify the early pathological findings of venous mesentericischemia and allows to correlate those to the histopathological features. Our data suggest a relevant role ofMRI in the diagnostic management of mesenteric venous thrombosis, allowing to non-invasively identifyand characterize the histopathologic findings. So, thanks to these skills, its future application in earlydiagnosis of human mesenteric venous ischemia is supposable.
Mesenteric ischemia encompasses a broad spectrum of diseasesdue to acute or chronic decrease of small bowel blood supply thatmay present occlusive (arterial/venous) or non-occlusive etiopatho-genesis [1]. It has been estimated that the majority of cases (65%) arecaused by arterial embolism or thrombosis, 25% by non-occlusiveaetiology and the remaining 10% from venous thrombotic aetiology[2]. Despite the advances in imaging techniques and, consequently, intherapeutic approaches, the overall mortality rate of this condition,still ranges from 60% to 100% [3]. An early diagnosis is important for acorrect therapeutic approach able to reduce the mortality [4–6].
Till now the best imaging method in diagnosis of mesentericischemia/infarction is enhanced CT (computed tomography) [7–9]even if some authors are considering a role for MRI (magneticresonance imaging) in the evaluation of these disorders [10,11].Moreover, in available literature, there exists no systematicevaluation of the radiological findings following superior mesenteric
+39 0815665200.bellis).
l rights reserved.
vein (SMV) occlusion. The aim of this study is to identify the onset ofmacroscopic, microscopic and MRI findings and their chronologicaldevelopment, in a rat model of SMV ligation, and to evaluate the roleof MRI in the diagnosis of mesenteric venous thrombosis.
2. Material and methods
2.1. Animal preparation
The study was carried out from March 2010 to July 2010. Allprocedures performed on animals were approved by the InstitutionalAnimal Care and Use Committee. Adult (average weight 300g) maleSprague–Dawley rats (n=30) were used in this study. Rats weremaintained on a 12/12h light/dark cycle and were allowed free accessto food and water. They were anesthetized with Ketamine hydrochlo-ride 100mg/kg im (CU Chemie Uetikon GmbH, Lahr, Germany) andDomitor 0.25mg/kg im (Medetomidine hydrochloride, Pfizer, Canada)injections. Dolorex 0.1mg/kg sc (Butorphanol, Intervet InternationalB.V., Boxmeer, the Netherlands) was used immediately beforeintervention in order to ensure intra-operative analgesia. Furtherinjections of these drugs were provided during the whole time of
409F. Somma et al. / Magnetic Resonance Imaging 31 (2013) 408–413
intervention to maintain a sufficient state of anesthesia. Each rat wasallowed to breathe spontaneously. During all the procedures, bodytemperaturewasmonitoredbya rectal probeandmaintained at37.0±0.5°C with a heating blanket regulated by a homeothermic blanketcontrol unit (Harvard Apparatus Limited, Edenbridge, Kent, UK). Afterdrug injection, rats were prepared for surgery through chest, abdomenand back depilation. These areas were then washed using povidoneiodine and alcohol. These animalswere divided in two different groups(Group I and Group II), of fifteen rats each. For both groups thefollowing time-points were considered: basal (t0), 30min (t1), 45min(t2), 1h (t3), 2h (t4), 4h (t5), 6h (t6), 8h (t7), as described above.
2.2. Surgical procedures
Surgical model and drug administration were performed by aboard-certified veterinarian with 5years of experience in microsurgi-cal and vascular techniques. Aftermidline laparotomy, the small bowelwas exposed out of the abdominal cavity, displaced to the left and theSMV was identified. The two groups of rats underwent a differentsurgical model and a different monitoring of ischemic damage:
1) in Group I rats (n=15), the mesenteric ischemia and infarctionwere realized by a tight ligation of SMV at its origin. Snapshotsof exposed bowel and mesentery were taken at t0m (macro-scopic), t1m, t2m, t3m, t4m, t5m, using a digital camera (NikonCoolpix S210, 8.0 Megapixels of resolution, ISO 2000, Japan).From t1m, at each time-point three rats were euthanizedby intrapulmonary injection of Tanax 0.5ml (Embutramide+Mebenzonium, Iodide+Tetracaine, Intervet/Schering-PloughAnimal Health, Boxmeer; the Netherlands) and the intestineexcised for histological analysis.
2) in Group II rats (n=15), a silk thread 3/0 was used to bind aloop around the SMV at its origin. The loop was not tight, sothat venous outflow was allowed. The tips of the thread werebrought into a silicon pipe and, through it, carried out to theback of the animal, between its shoulders. The thread tipswere fixed to the pipe using medical plaster. Muscles and skinwere closed in two layers using a Vicryl 2/0 thread. 10mg/kg ofBaytril 10% (Enrofloxacin 2,5%, Bayer AG, Leverkusen, Germa-ny) was topically applied to the wounds to prevent infections.The animals were returned to their cages after awakening, andwater and food were allowed ad libitum. Then, mortality ratewas recorded. Three days after surgery, rats underwent asecond anesthesia according to the same drug protocol, andmesenteric ischemia and infarction were induced by pullingthe threads out of the pipe. In this way, the loop around theSMV got squeezed and the venous outflow through this veinwas stopped. Then the animals underwent 7T μMRI.
2.3. 7T μMRI
Two abdominal radiologists, with 10 years of experience at least,assessed all 7 Tesla micro MRI (7T μMRI) images by consensus,blinded to histological findings. The rats were maintained underanesthesia till the end of MRI monitoring (8h). The 7T μMRIsequences were acquired in prone position.
In all Group II rats, just before pulling the threads and occludingthe vessel, a localizer scan along the three orthogonal planes (TR=6.0ms, TE=100.0ms, FOV=8, Averages=1, Flip Angle 30°,Matrix=128, Slice thickness=2.00mm, Scan time=12s 800ms)and a T2-weighted TurboRare sequence (TR=4428.7ms, TE=36ms; 180° flip angle; 38 slices; slice thickness 1mm, interslicedistance 1mm, field of view, 6cm; acquisition matrix, 256 × 256,scan time=7 min 5 s 156 ms) were performed as basal imaging: t0i(imaging). After SMV occlusion T2-weighted TurboRare sequences
were performed at the following time-points: t1i, t2i, t3i, t4i, t5i.From t1i, at each time-point three rats were euthanized (by thesame procedure used in Group I) in order to assess the histologicalanalysis. MR images were assessed for the following parameters: 1)presence of free fluid and/or gas in the abdomen; 2) dilated loopsonly gas filled, known as hypotonic reflex ileus (HRI); 3) dilatedloops with gas–fluid mixed stasis; 4) mucosal thickness; 5) wallsignal intensity; 6) mesenteric engorgement; 7) wall pneumatosis.
2.4. Histological analysis
Histological tissue processing and analysis were performed by apathologist with 5years of experience in experimental studies. Forlight microscopy, the entire intestine (from duodenum to ascendingcolon and includingmesentery) of all rats of both groupswas excisedand stored in 10% buffered formalin acetate for at least 2days. Beforesample embedding, the different regions were identified and theirlengthmeasured. The extension of evident lesions was also recorded.The samples were then divided into four pieces including theduodenum and the jejunum (1 segment — 1s), ileum (2 segment —2s), terminal ileum (3 segment— 3s), and cecumwith the ascendingcolon (4 segment — 4s). From each piece, 3mm sections wereobtained at 10mm intervals and embedded in paraffin. Transversesections 3 μm thick were cut and stained with hematoxylin–eosin(H&E). Sections were mounted on chromealume-gelatin-coatedslides, dehydrated and coverslipped. Slides were imaged with aZeiss Axioskop 2 light microscope (Zeiss Axioskop microscope)equipped with high-resolution digital camera (C4742-95, Hamama-tsu Photonics K.K., Hamamatsu City, Japan).
Thedegree of histological damagewas evaluatedby theChiu scoreofmucosal injury [12]. The scale consists of values from 0 to 5, where:
• Degree 0: normal mucosal villi.• Degree1: developmentof subepithelial Gruenhagen's space,well-formed villosity, absence of cell lysis or inflammatory process;
• Degree 2: cell lysis, extension of subepithelial Gruenhage'sspace with moderate lifting of epithelial layer from the laminapropria and increased spacing between the villi;
• Degree 3: massive epithelial lifting down the sides of villi,destruction of the free portion of the villi, presence of dilatedcapillaries and inflammatory cells;
• Degree 4: denuded villi with lamina propria and dilatedcapillaries exposed. Increased cellularity of lamina propria maybe noted.
• Degree 5: destruction of the entire tunica mucosa, no longerbeing observed any glandular structure, but only amorphousmaterial deposited on the submucosal tissue with hemorrhageand basal glandular ulceration;
In order to assess the validity and the reliability of the Group IIsurgical model, the histological features of Group I and Group IIwere compared.
2.5. Statistical analysis
Statistical analysis was performed independently and in a blindmanner by a physician with 3years experience in the managementof experimental data.
Data are reported as means±standard deviation or medians(range). Histological data were compared between each group usingthe Unpaired t-test, when indicated. A probability value less than0.05 was considered significant.
3. Results
A 0% mortality rate (0/15) was observed in Group II rats.
410 F. Somma et al. / Magnetic Resonance Imaging 31 (2013) 408–413
3.1. Macroscopic analysis
Fig. 1A shows the rats’ intestine appearance immediately beforethe SMV ligation (t0m): the bowel loops presented an averagediameter of 1.5mm, uniform serosa and rose-coloredmesentery. The
Fig. 1. (A) Physiological appearance of the rat small bowel: the bowel loops haveaverage diameter of 1.5mm, uniform serosa and rose-colored mesentery; (Btortuous appearance of mesenteric vessels (white arrow) with chromatic change(from pink to red) of the bowel wall 30 min after SMV ligation; (C) at 45min oobservation the bowel wall changed the color to dark red and ulcerations of themesentery (arrowhead) were detected; (D) some mild stenotic tracts depicting“sausage aspect” (black arrows) were associated with previous findings at 1hexamination; (E) increasing of previous findings 2h after SMV: (F) 4h after SMVligation a new chromatic change of the ischemic bowel (from dark red to charcoablack) became evident (black arrow); it appears more evident 6h after SMV ligation(black arrows) (G); (H) stenotic tracts were associated with the persistent chromaticchange, vascular congestion of mesenteric vessels and ulcerations of the mesentery
)
f
l
.
early findings, after SMV ligation, were already evident at t1m,consisting of: vascular congestion of mesenteric vessels withtortuous appearance, chromatic change (from pink to red) of thebowel wall (Fig. 1B).
Similar features were evident at t2m, when the bowel wallchanged the color to dark red and ulceration of the mesentery wasdetected (Fig. 1C). At t3m some mild stenotic tracts, depicting a“sausage aspect”, were associated with previous findings (Fig. 1D),which appeared more evident at t4m (Fig. 1E).
At t5m, the damaged areas underwent a new chromatic change(from dark red to charcoal black) and this finding got more apparentat t6m (Fig. 1F–G).
At t7m, some mild stenotic tracts were associated with thepersistent chromatic change, the vascular congestion of mesentericvessels and the ulcerations of the mesentery (Fig. 1H). In all theanimals examined for each time point, the same macroscopicchanges were observed.
3.2. 7T μMRI
During the 7T μMRI study, the animals underwent cardio-respiratory and temperature monitoring by a dedicated device. Att0i no free gas in the abdominal cavity and no sign of bowel ormesenteric/peritoneal alterations were detected. None of bowelloops appeared thicker than 0.5mm and/or had diameter greaterthan 2.8mm (Fig. 2A).
At t1i, a hyperintense signal of intestinal wall in some loops and avery thin layer of peritoneal fluid were detected (Fig. 2B). Thesefindings were mild increased at t2i and t3i (Fig. 2C–D). At t4i,peritoneal free fluid and loops wall thickening (average thickness of1.5mm) were observed beyond hyperintensity of intestinal wall(Fig. 2E). After, at t5i, a further amount of the free fluid wasassociated with previous findings (Fig. 2F).
At t6i, a gas–fluid mixed stasis with a larger amount of peritonealfluid appeared andmesenteric engorgement became evident beyonda hyperintense signal of intestinal wall in many loops (Fig. 2G). Atthe end of the observation (t7i), wall pneumatosis was detected inonly one case, adding to the previously reported findings (Fig. 2H). Inall the animals examined for each time point, the same 7T μMRIchanges were observed, except for wall pneumatosis, detected inonly one case.
3.3. Histological analysis
The macroscopic and histopathologic analysis revealed similarfeatures in the two groups. Table 1 shows the mean Chiu score in thetwo groups rats at each time-point. The t test didn't show anysignificant differences. At t1p (pathologist), as in t2p, histologicalanalysis showed a patchy inflammatory pattern and vascularcongestion with the development of subepithelial edema (Grunha-gen's space) (Fig. 3A–B).
At t3p, extension of the Grunhagen's space, cell lysis and spacingbetween the villi were observed in many samples (Fig. 3C). At t4psome sections obtained by the middle segment of ileum showed anearly loss of the normal shape of villi with the destruction of the freeportion and spared glandular structures; dilated and congestedvessels were also present (Fig. 3D).
At t5p, hemorrhage and congestion were evident both in thebasal glandular area and in the mesentery (Fig. 3E).
At t6p, an intense structural wrecking of the villi led to thereplacement of the free portion of the villi with typical aspect ofthe “phantom villi” (or “shadow villi”) when these structures losethe epithelial layer and their central axis is filled with amorphousand necrotic material; some slides showed haemorrhage in theentire bowel wall and mesenteric congestion is evident.
Fig. 2. (A) 7Tμ-MR abdominal scan before SMV ligation: no free gas in the abdominalcavity and no signs of bowel or mesenteric/peritoneal alterations were detected; (B)7Tμ-MR abdominal scan 30min after SMV ligation: a hyperintense signal of intestinalwall in some loops (white arrow) and a very thin layer of peritoneal fluid (star) weredetected; (C) 7Tμ-MR abdominal scan 45min after SMV ligation: hyperintense signalof intestinal wall was evident in many loops (white arrow), while free peritoneal fluidwas mild increased (star); (D) 7Tμ-MR abdominal scan 1h after SMV ligation: nosignificant changes were detected; (E) 7Tμ-MR abdominal scan 2h after SMV ligation:peritoneal free fluid (star) and hyperintensity of intestinal wall with loops wallthickening (white arrow) were observed; (F) 7Tμ-MR abdominal scan 4h after SMVligation: some loops only gas filled (white arrow) and a further amount of the freefluid (white star) were associated with previous findings; (G) 7Tμ-MR abdominalscan 6h after SMV ligation: gas–fluid mixed stasis (black arrow) and hyperintensesignal of intestinal wall (white arrow) with a larger amount of peritoneal fluid (star);(H) 7Tμ-MR abdominal scan 8h after SMV ligation: wall pneumatosis (white arrow).
Table 1Mean Chiu score in each group rats at each time-point.
411F. Somma et al. / Magnetic Resonance Imaging 31 (2013) 408–413
At t7p, histological findings were similar to t6p time-point(Fig. 3F–G). In all the animals examined for each time point, the samemicroscopic features were observed.
4. Discussion
The availability to reproduce human diseases in animal modelsmakes it possible to increase our knowledge about the physiopa-thology of these disorders and to study new diagnostic andtherapeutic approaches [13].
Concerning this, a variety of dedicated devices have beendeveloped, among which are micro CT, US (ultrasound) and MRI.This study, based on an animalmodel of SMVocclusion,was designedto define exactly the onset and the development of the ischemiclesions, to compare the anatomo-pathological features with the 7TμMRIfindings and to investigate the possible advantages ofMRI in thediagnostic approach of the mesenteric venous thrombosis.
At present, in the available literature a systematic evaluation stilldoes not exist of the radiological findings following SMV occlusion,and the role of MRI is debated. Furthermore, we hypothesize thatvenous occlusion could lead to different findings than otheretiologies (arterial non occlusive).
Basing on our results, the early 7T μMRI findings are the intestinalwall hyperintensity associated with a thin layer of peritoneal fluid,observed 30min after SMV occlusion. At the same time themacroscopical analysis revealed mesenteric vascular congestionwith bowel wall chromatic change (from pink to red). Histologicalfindings were represented by the presence of subepithelial (Grun-hagen's) space due to edema development; this feature clarifies thebowel wall hyperintensity saw at MRI [14]. The 45min findings weresimilar to 30min examination consisting of a hyperintense signal ofintestinal wall in many loops at MRI examination while peritonealfree fluid was mild increased. The macroscopic findings wererepresented by the chromatic change of the bowel wall to dark redand ulcerations of the mesentery. Microscopically, the inflammatorypattern appeared more evident.
At 1h, MRI examination, beyond previous reported findings,revealed an increase of peritoneal free fluid with bowel wallthickening; the latter was macroscopically evident as stenotic tractdevelopment. These lesions appeared microscopically like an exten-sion of Grunhagen's space related to an increase of vascular congestion.At 2h observation the MRI and macroscopic findings became moreevident and widespread. Microscopically loss of normal shape of villiwith destruction of free portion of villi was established and sparedglandular structures dilated and congested vessels were also present.
Four hours after SMV ligation, the main MRI findings were themesenteric engorgement, also evident at macroscopic analysis.Histologically, hemorrhage and mesenteric engorgement were evi-dent. Loops with fluid–gas mixed stasis represent the result of theimpaired peristaltic activity. This finding was noted at 6h MRI scans.Macroscopic and histological analysis showed a progressive andparallel worsening of the injury in the subsequent observations points.
The last significant finding was wall pneumatosis, indicator ofbowel necrosis [15], that 7T μMRI revealed in only one rat at the endof observation (8h).
Therefore, our experience suggests that the bowel wall hyper-intensity and thickening are the earliest MRI findings, differently
Fig. 3. Histological analysis. (A) (200×) 30min after SMV ligation showed a patchy inflammatory pattern and vascular congestion with the development of sub-epithelial edema(Grunhagen's space) (star); (B) (400×) 45min after SMV ligation the histological pattern didn't show any significant differences; (C) 1h after SMV ligation: extension ofGrunhagen's space (star), cell lysis, dilated and congested vessels (arrows); (D) 2h after SMV ligation early loss of the normal shape of villi with destruction of the free portion andspared glandular structures dilated and congested vessels were also present; (E) 4h after SMV ligation, hemorrhage and congestion were evident both in the basal glandular areaand in the mesentery (arrows); (F) 6h after SMV ligation, hemorrhage and congestion in the whole bowel wall (arrows), thin mucosal layer with “phantom villi” and fewglandular structures; lumen filled with amorphous and necrotic material and inflammatory cells (arrowheads); (G) no relevant changes were observed after 8h.
412 F. Somma et al. / Magnetic Resonance Imaging 31 (2013) 408–413
from arterial forms of infarction in which the “paper thin” wall wasdescribed [16,17]. Our results are in line with those reported inliterature, where the bowel wall thickening is described as the mostcommon CT finding strictly related to venous infarction [9,11,17–22].In addition, we displayed that it is an early finding, since it has beendetected already 2h after SMV ligation.
We have never observed the spastic reflex ileus and the HRI,differently from arterial forms of infarction in which these reactionsare subsequent [16], probably masked by bowel wall thickening dueto intramural edema.
According to other authors, the early detection of peritoneal freefluid and mesenteric engorgement could indicate the greaterseverity of occlusion, requiring a well-timed treatment [19,20]. Ourresults conflict with those obtained in another series [23] that statesthat the bowel wall dilation reflects an advanced stage of disease,since dilated loops have been never observed after SMV occlusion.
Moreover, even if in the clinical setting of intestinal ischemiapneumatosis has been considered to be a sign of advanced diseaseusually indicating irreversible injury and transmural necrosis[24–28], some AA. hypothesized it could be observed in patientswith intestinal ischemia before the development of transmuralnecrosis as early sign of bowel ischemia/infarction [29,30].
Therefore, the progression of the ischemic disease, observed inthis experimental model, appears in line with the human pathology.
The ratmodel used in this study shows the advantage of employinga 3-day gap to induce a delayed acute SMV ischemia that avoids thebias in imaging findings related to the manipulation of the abdominalcavity that has been shown to determine intra abdominal air trappingand morphological and functional alterations of the peritoneum andsmall intestine [31]. Limits are related: 1) to the type of vascularocclusion, resulted to be sudden and complete, while partial occlusionin human is possible; 2) to the SMV ligation, performed at theemergency of vessel, whereas in clinical practice distal occlusions arealso observed; 3) to the time window analysis that was limited to
eight hours, while it is quite common for these patients to reach thehospital after several hours away from the setting-occlusive event.
However, despite the limitations of the animal model, this studyseems to clearly show the possible advantages of MRI in themesenteric venous thrombosis diagnosis.
Moreover, even if in this study the MR scans were performedwith a 7T MRmachine, actually not yet adopted for clinical use, bothour results and several studies prospect 7T MR machines for clinicalresearch on humans [32,33].
In conclusion, a parallelism between the experimental bowelischemic damage in this animal model and the humans, issupposable. Compared to histological analysis and macroscopicevidences, the MRI allowed a correct and non-invasive evaluationof the bowel ischemia, in particular, MRI succeeded to early identifythe signs of ischemia already 1h after SMV occlusion and to evaluatethe specific findings due to venous occlusion.
Furthermore our study provides a chronological progression ofthe radiological findings related to macroscopic and histologicalfeatures that is the prerequisite to obtain an early diagnosis andtreatment with a better prognosis.
The further usefulness of MRI for the diagnosis and monitoring ofischemic bowel disease is due to the lack of contrast medium andionizing radiations; it has been also demonstrated that the MRIsensitivity in the detection of bowel ischemia is comparable to thatof CT [34].
Thanks to these skills, its future application in early diagnosis ofhuman mesenteric venous ischemia is highly reasonable.
References
[1] Baden JG, Racy DJ, Grist TM. Contrast-enhanced three-dimensional magneticresonance angiography of the mesenteric vasculature. J Magn Reson Imaging1999;10:369–75.
[2] Gore RM, Yaghmai V, Thakrar KH, Berlin JW, Mehta UK, Newmark GM, et al.Radiol Clin N Am 2008;46:845–75.
[4] Schoots IG, Koffeman GI, Legemate DA, Levi M, van Gulik TM. Systematic reviewof survival after acute mesenteric ischemia according to disease aetiology. Br JSugr 2004;91:17–27.
[5] Oldenburg WA, Lau LL, Rodenberg TJ, Edmonds HJ, Burger CD. Acute mesentericischemia: a clinical review. Arch Intern Med 2004;164:1054–62.
[6] Sreenarasimhaiah J. Diagnosis and management of intestinal ischemic disorders.BMJ 2003;326:1372–6.
[7] Menke J. Diagnostic accuracy of multidetector CT in acute mesenteric ischemia:systematic review and meta-analysis. Radiology 2010;256:93–101.
[9] Wiesner W, Khurana B, Ji H, Ros PR. CT of acute bowel ischemia. Radiology2003;226:635–50.
[10] Chow LC, Chan FP, Li KCP. A comprehensive approach to MR imaging ofmesenteric ischemia. Abdom Imaging 2002;27:507–16.
[11] Kim JH, Ha HK, Sohn MJ, Shin BS, Lee YS, Chung SY, et al. Usefulness of MRimaging for diseases of the small intestine: comparison with CT. Korean J Radiol2000;1:43–50.
[12] Chiu CJ, McArdle AH, Brown R, Scott HJ, Gurd FN. Intestinal mucosal lesion inlow-flow states. A morphological, hemodynamic, and metabolic reappraisal.Arch Surg 1970;101:478–83.
[13] Grassi R, Cavaliere C, Cozzolino S, Manzi L, Cirillo S, Tedeschi G, et al. Smallanimal imaging facility: new perspectives for the radiologist. Radiol Med2009;114:152–67.
[14] Kim MY, Suh CH, Kim ST, Lee JH, Kong K, Lim TH, et al. Magnetic resonanceimaging of bowel ischemia induced by ligation of superior mesenteric artery andvein in a cat model. J Comput Assist Tomogr 2004;28:187–92.
[15] Smerud MJ, Johnson CD, Stephens DH. Diagnosis of bowel infarction: acomparison of plain films and CT scans in 23 cases. AJR Am J Roentgenol1990;154(1):99–103.
[16] Berritto D, Somma F, Landi N, Cavaliere C, Corona M, Russo S, et al. 7T micro-MRIearly detection of arterial acute mesenteric ischemia: evolution of findings in anin vivo rat model. Radiol Med 2011;116:829–41.
[17] Türkbey B, Akpinar E, Cil B, Karçaaltincaba M, Akhan O. Utility of multidetectorCT in an emergency setting in acute mesenteric ischemia. Diagn Interv Radiol2009;15:256–61.
[18] Acosta S, Alhadad A, Ekberg O. Findings in multi-detector row CT with portalphase enhancement in patients with mesenteric venous thrombosis. EmergRadiol 2009;16:477–82.
[19] Bradbury MS, Kavanagh PV, Bechtold RE, Chen MY, Ott DJ, Regan JD, et al.Mesenteric venous thrombosis: diagnosis and noninvasive imaging. Radio-graphics 2002;22:527–41.
[20] Romano S, Lassandro F, Scaglione M, Romano L, Rotondo A, Grassi R. Ischemiaand infarction of the small bowel and colon: spectrum of imaging findings.Abdom Imaging 2006;31:277–92.
[21] Furukawa A, Kanasaki S, Kono N, Wakamiya M, Tanaka T, Takahashi M, et al. CTdiagnosis of acutemesenteric ischemia from various causes. AJR Am J Roentgenol2009;192:408–16.
[22] Macari M, Balthazar EJ. CT of bowel wall thickening: significance and pitfalls ofinterpretation. AJR Am J Roentgenol 2001;176(5):1105–16.
[23] Moschetta M, Stabile Ianora AA, Pedote P, Scardapane A, Angelelli G.Prognosticvalueof multidetector computed tomography in bowel infarction.Radiol Med 2009;114:780–91.
[24] Federle MP, Chun G, Jeffrey RB, Rayor R. Computed tomographic findings inbowel infarction. AJR 1984;142:91–5.
[25] Alpern MB, Glazer GM, Francis IR. Ischemic or infarcted bowel: CT findings.Radiology 1988;166:149–52.
[26] Bartnicke BJ, Balfe DM. CT appearance of intestinal ischemia and intramuralhemorrhage. Radiol Clin North Am 1994;32:845–60.
[27] Taourel PG, Deneuville M, Pradel JA, Régent D, Bruel JM. Acute mesentericischemia: diagnosis with contrast-enhanced CT. Radiology 1996;199:632–6.
[28] Chou CK, Mak CW, Tzeng WS, Chang JM. CT of small bowel ischemia. AbdomImaging 2004;29:18–22.
[29] Wiesner W, Mortele KJ, Glickman JN, Ji H, Ros PR. Pneumatosis intestinalis andportomesenteric venous gas in intestinal ischemia: correlation of CT findingswith severity of ischemia and clinical outcome. AJR 2001;177:1319–23.
[30] Kernagis LY, Levine MS, Jacobs JE. Pneumatosis intestinalis in patients withischemia: correlation of CT findings with viability of the bowel. AJR Am JRoentgenol 2003;180(3):733–6.
[31] Ballabeni V, Barocelli E, Bertoni S, Impicciatore M. Alterations of intestinal motorresponsiveness in a model of mild mesenteric ischemia/reperfusion in rats. LifeSci 2002;71:2025–35.
[32] Kraff O, Thevsohn JM, Maderwald S, Kokulinsky PC, Dogan Z, Kerem A, et al. Highresolution MRI of the human parotid gland and duct at 7 Tesla. Invest Radiol2009;44:518–24.
[33] Kraff O, Bitz AK, Kruszona S, Orzada S, Schaefer LC, Theysohn JM, et al. An eight-channel phased array RF coil for spine MR imaging at 7T. Invest Radiol 2009;44:734–40.
[34] Stamatakos M, Douzinas E, Stefanaki C, Petropoulou C, Arampatzi H, Safioleas C,et al. Ischemic colitis: surgingwaves of update. Tohoku J ExpMed 2009;218:83–92.
