Gong Pei-You ( Department of Radiology, Yantai Yuhuangding Hospital, Yantai 264000, Shandong Province, PR China. )
Li Jun-Xia ( Department of Oncology, Yantai Yuhuangding Hospital, Yantai 264000, Shandong Province, PR China. )
Liu Feng-Li ( Department of Radiology, Yantai Yuhuangding Hospital, Yantai 264000, Shandong Province, PR China. )
Zhang Liang-Ming ( Department of Oncology, Yantai Yuhuangding Hospital, Yantai 264000, Shandong Province, PR China. )
Xie Hai-Zhu ( Department of Radiology, Yantai Yuhuangding Hospital, Yantai 264000, Shandong Province, PR China. )
Yan-Bin ( Department of Radiology, Yantai Yuhuangding Hospital, Yantai 264000, Shandong Province, PR China. )
Objective: To compare the efficacy of computed tomography enterography and magnetic resonance enterography in diagnosing small intestinal diseases.
Methods: The retrospective study comparing computed tomography enterography and magnetic resonance enterography for diagnosing diseases related to small intestine was conducted at Department of Radiology, Yantai Yuhuangding Hospital, Shandong, China, from July 2012 to February 2014. The efficacy of computed tomography enterography and magnetic resonance enterography results were evaluated for randomly-selected cases to compare the location and characteristics of small intestinal diseases together with small bowel endoscopy and clinical pathology observations.
Results: Of the 30 patients in the study, 19(63.3%) were males and 11 (36.7%) were females with an overall mean age of 33.6±19.2 years (range: 24-67 years). the clinical diagnostic accuracy of computed tomography enterography and magnetic resonance enterography was 24(80%) and 21(70%) cases respectively (p>0.05).
Conclusions: Computed tomography enterography and magnetic resonance enterography are two techniques that complement each other for diagnostic purposes.
Keywords: Computed tomography enterography, Magnetic resonance enterography, Small intestinal diseases. (JPMA 65: 710; 2015).
The small intestine is a difficult part of the gastrointestinal tract (GIT) for clinical diagnostic radiology because of its length and complex loops. New imaging methods for diagnosing small intestine diseases, including computed tomography enterography (CTE) and magnetic resonance enterography (MRE), are becoming increasingly popular.1,2 CTE is performed after the intestinal lumen is filled with contrast agent. Not only can CTE identify intestinal pathologies such as bowel wall thickening, mucosal enhancement, bowel wall stratification, luminal stenosis, and mesenteric vasodilation, but it also has unique advantages in diagnosing extra-intestinal diseases.3-5 MRE involves filling the small intestine with isotonic mannitol.5 It boosts high soft-tissue resolution and displays tissue and structures both inside and outside the intestine, which is especially useful for diagnosing intestinal obstruction caused by cancer.5-7 The three-dimensional dynamic contrast-enhanced (DCE) scans clearly show bowel wall and the intensified regions of intestinal lesions.8,9 In addition, it was reported that coronal scans can display the mesenteric vessels.10 Diffusion-weighted imaging is currently the only available clinical method for imaging water molecules and measuring their diffusion. To date, many publications3-5 have reported the utility of CTE and MRE in assessing small intestinal diseases, but very few studies have actually compared the efficacy of the two techniques and indicated whether or not they can complement each other for better diagnosis.3-5
The current study was planned to retrospectively compare the imaging utility of CTE and MRE in diagnosing small intestinal diseases.
Patients and Methods
The retrospective study was conducted from July 2012 to February 2014 and comprised randomly selected data related to patients who were diagnosed with small bowel disease in the Department of Radiology, Yantai Yuhuangding Hospital, Shandong, China. Data included was related to only those who had undergone both CTE and MRE for clinical reasons. CTE and MRE has different advantages in gastrointestinal stromal tumour (GIST) or cancer and Crohn\'s disease, and they are non-invasive in nature. Gastrointestinal stromal data of patients who had acute bleeding was excluded. Pathological diagnosis of postoperative results by operation or biopsy results by small intestinal endoscopy were used as the gold standard. Written informed consent was obtained from all participants before the examination.
For CTE, patients were required to fast for 12 hours before the examination.. Intestinal tract cleaning was not needed for the study. Prior to the examination, patients had to drink 2,000ml of 2.5% isotonic mannitol solution four times within 45 minutes. Mannitol acts as a negative contrast agent and fully dilates the intestine and tastes slightly sweet. The CT value of isotonic mannitol is close to that of water, and mannitol is not easily absorbed in the intestine after oral administration. Previous studies demonstrated that isotonic mannitol effectively dilated the intestine, allowing visualisation of intestinal mucosal lesions, without adverse reactions.3,4 Non-enhanced, arterial-phase and portal-phase scans were conducted on each patient using a GE64-slice CT (GE Healthcare, USA). Scans ranged from the diaphragm level to the pubic symphysis level and included the entire small intestine. One millimetre-thick reconstruction slices, spaced at 0.7mm, were transmitted to an Advanced Workstation 4.3 (GE Healthcare, USA) for reconstruction with post-processing software.
For MRE examination, to enhance imaging contrast, isotonic mannitol (1,500-2,000 ml) was orally administrated and allowed to transit to the caecum within 30 minutes. At this point, 20mg anisodamine (654-2) was injected intravenously (IV) to inhibit intestinal peristalsis, and multi-axis MRE scanning was performed. Each sequence, with fat suppression, included: (1) coronal T2-weighted image of the single-shot fast-spin echo sequence, and required an approximately 25-second breath-hold to complete the scan; (2) coronal T1-weighted image fast spoiled-gradient echo sequence; and (3) an enhanced MR imaging via IV injection of gadopentetic acid, obtaining breath-hold scans of coronal and cross-sectional fast spoiled-gradient echo sequences.
Two experienced radiologists, blinded to the imaging methods, made the evaluation on both CTE and MRE imaging performance of the small intestine lumen, intestinal wall, mesentery, blood vessels, and other morphological changes. Based on these observations, the radiologists assessed the relative diagnostic value of CTE and MRE for imaging small bowel lesions. If they agreed with each other, diagnosis was considered to have been arrived at. If there was a difference, the diagnosis was reached after consultation or discussion at the group meeting.
For statistical analysis, Fisher c2 test was performed for comparison between CTE and MRE imaging. P <0.05 was taken as statistically significant.
Of the 30 patients, 9(30%) had undergone MRE which had missed the diagnosis and they had to have CTE done to have a definite further diagnosis. The remaining 21(70%) patients then underwent both CTE and MRE. Overall, 19(63.3%) patients were males, while 11(36.7%) were females. The mean age of the patients was 33.6±19.2 years (range: 24-67 years). Besides, 22(73.3%) patients had active disease with intestinal symptoms, including abdominal pain, diarrhoea and haemafaecia. However, 8(26.6%) haemafaecia patients displayed no additional clinical symptoms. The CTE and MRE imaging results displayed no significant difference in diagnosis (p>0.05), with an accuracy of 80% for CTE and 70% for MRE (Table-1).
The sensitivity, specificty, positive predictive value (PPV) and negative predictive value (NPV) for the two methods were worked out (Tables 2-4).
Further, 11(36.6%) cases were diagnosed with small bowel tumour lesions by both CTE and MRE, with a consistent, accurate diagnosis. Using CTE, the tumours were identified by rich blood supply and some intestinal obstruction. Eight cases of the primary tumours were located in the extra-intestine, and there were calcification spots in the tumours. In three cases, tumours showed significantly enlarged feeding arteries and draining veins, and there was significant intestinal obstruction (Figure-1A and B).
When intestinal obstruction images were merged, MRE made use of the fluid retention to facilitate imaging, which indicated a clear dilation of the intestinal canal (Figure-1C). The obstruction point and collapse of the intestinal canal distal to the obstruction could be determined (Figure-1C and D). Metastatic lesions were identified far from the intestine.
Diagnosis of intestinal vascular malformations was missed in 3(50%) of the 6(20%) cases when MRE alone was used to evaluate intestinal vascular malformations. This might be due to poor visualisation of vascular malformations by MRE. There were abnormally thickened blood vessels on the arterial wall, and revascularisation showed the blood vessels\' arterial sources (Figure-2A-C).
Because of the long imaging times required, MRE was ineffective for diagnosing vascular malformations.
Further, 7(23.3%) patients were diagnosed with Crohn\'s disease both by CTE and MRE. The two techniques were found to be equally accurate for Crohn\'s diagnosis. Multiple small intestine segments and the colon were affected. All affected small intestine segments had bowel wall-thickening, obvious small bowel enhancement, fuzzy serosa, dirty mesentery, and increased numbers of tortuous, dilated small blood vessels around the intestine to varying degrees (Figure-3A-C).
Lumen narrowing and slight dilation of the proximal bowel were also observed. MRE showed thickening of multiple small bowel wall segments. Low signal was seen in T1-weighted images, whereas high signal was observed in T2-weighted image and hydrography (Figure-4A).
In fat suppression sequences, bowel wall oedema was evident (Figure-4B). After the scan was enhanced, an even signal with a thickened bowel wall was represented. Bowel stenosis was observed to varying degrees. Occasionally, stenosis and dilation alternated, and the intestinal rim was irregular. Where fistulas of Crohn\'s disease had been formed, high signal was seen at the fistula, which was surrounded by relatively low signal tissue.
Both CTE and MRE missed diagnosis of 7(23.3%) cases of small intestinal diverticulosis, telangiectasia, and simple small intestine congestion and oedema. This was considered the blind spot of imaging diagnosis.
The results of the study make it evident that CTE and MRE were very accurate in examining tumour lesions in the intestine. Most primary tumours were indicated with the manifestation of rich blood supply and some intestinal obstruction clearly identified by CTE. One the other hand, MRE showed homogeneous signal mass, rim finishing and clear boundaries; and with enhanced scanning, the signal was obviously and uniformly enhanced. With malignant tumours, the mass showed an image of intermediate signal soft issue, and the signal was uneven after enhanced scanning. In line with some previous studies,5-7 our study demonstrated that both CTE and MRE provided a panoramic view of small intestine cavity, wall, mesentery, lymph nodes, blood vessels, and adjacent organs.
For Crohn\'s disease, the lesions often caused inflammation of lymph nodes at the mesentery root, and small vessel accumulation in peripheral lesions. CTE imaging revealed the typical "comb sign." MRE could also evaluate the activity of Crohn\'s disease through analysis of parameters such as the degree of enhancement and signal strength of the intestine wall, and steatohepatic fibrotic proliferation on T2-weighted fat-suppressed images. Significantly enhanced, high signal images of the intestinal wall and steatohepatic fibrosis proliferation are indicative of active Crohn\'s. This assessment provides guidance for the clinical treatment and prognostic evaluation of Crohn\'s disease.6,7 CTE also has advantages in diagnosing intestinal polyps.8 For intestinal obstruction caused by lesions, tumours, or unknown causes, CTE shows an accurate diagnosis rate of upper intestinal obstruction, whereas CTE provided an accurate diagnosis rate of only 48% for lower intestinal obstruction9,10 MRE can make use of fluid retention for imaging to diagnose intestinal obstruction without increasing patient pain. Earlier studies11,12 reported that MRE demonstrates a higher diagnostic accuracy rate for intestinal obstruction than does CTE. For diagnosing vascular malformations of the small intestine, CTE is superior to MRE because of high time-resolution and powerful post-processing software technology based on our experiences and consensus in the practice (Gong et al, unpublished data). However, MRE has advantages in locating and quantitatively and qualitatively determining small intestine disease.10 MRE is especially suitable for examining pregnant women, children and adolescents because of its inherent multi-planar imaging and high-resolution capabilities in soft tissue contrast, and because of its non-invasiveness and non-ionising radiation improves patient tolerance.7,13 One study14 proposed that MRE should be the preferred method in diagnosing paediatric Crohn\'s disease. Both CTE and MRE missed diagnosis of 7 cases of small intestinal diverticulosis, telangiectasia and simple small intestine congestion and oedema. The reason for missing the diagnosis of small intestinal diverticulosis might be that the small intestine is tortuous and diverticulosis appears similar to the structure of the normal small intestine. The missed telangiectasia, and simple small intestine congestion and oedema diagnoses might have been due to the small size of the dilated capillaries and the relatively minor simple small intestine congestion and oedema, which were not discernible using CTE or MRE.
Both CTE and MRE can be valuable in diagnosing small intestinal diseases. However, both procedures have advantages and disadvantages. Evolving CTE and MRE technologies are likely to play increasingly important roles in diagnosing small bowel disease. The accuracy of diagnosis will be improved when the results of both techniques can be compared. A larger trial is under way to confirm the findings of this study, which was limited by a small sample size.
We are grateful to the Science and Technology Development Project of Yantai for assistance and to the participating subjects. Thanks are also due to Xie Haizhu, Liu Fengli and other colleagues in the Gastroenterology Department of Yantai Yuhuangding Hospital, Yantai, Shandong, China.
This research was supported by Science and Technology Development Project of Yantai (NO.2012091).
1. Wu XW, Liu B, Zhao H.The effect of window setting technique on measuring colon disease in CT virtual colonoscopy. Chin J Radiol 2007; 41:316-318.
2. Furukawa A, Saotome T, Yamasak iM. et al. Cross-sectional imaging in Crohn disease. Radiographics 2004;24:689-702.
3. Zhang LH, Zhang SZ, Hu HJ. Multi-detector CT enterography with isoosmotic mannitol as oral contrast for detecting small bowel disease. World J Gastroenterol 2005;11:2324-9.
4. Fletcher JG. CT enterography technique: theme and variations. Abdom Imaging 2009;34:283-8.
5. Lee SS, Kim AY, Yang SK. Crohn disease of the small bowel: comparison of CT enterography, MR enterography, and small-Bowel follow-through as diagnostic techniques. Radiology 2009;251:751-61.
6. Maccioni F, Bruni A, Viscido A. MR imaging in patients with Crohn disease: value of T2- versus T1-weighted gadolinium-enhanced MR sequences with use of an oral superparamagnetic contrast agent. Radiology 2005;238:517-30.
7. Sempere GA1, Martinez Sanjuan V, Medina Chulia E, Benages A, Tome Toyosato A, Canelles P, etal. MRI evaluation of inflammatory activity in Crohn\'s disease. AJR Am J Roentegenol 2005;184:1829-35.
8. PJ Pickhardt, SM Wise, DH Kim. Positive predictive value for polyps detected at screening CT colonography. Eur Radiol 2010;20:1651-6.
9. Maglinte DD, Gage SN, Harmon BH. Obstruction of the small intestine: accuracy and role of CT in diagnosis. Radiology 1993;186:61-4.
10. Maglinte DD, Reyes BL, Harmon BH. Reliability and role of plain film radiography and CT in the diagnosis of small bowel obstruction. Am J Roentgenol 1996;167:1451-5.
11. Beal DP, Regan F. MRI of bowel obstruction using the HASTE sequence. J Comput Assist Tomogr 1996;20:823-5.
12. Regan F, Beall DP, Bohlaman ME. Fast MR imaging and the detection of small bowel obstruction. Am J Roentgenol 1998;170:1465-9.
13. Maccioni F, Bruni A, Viscido A. MR imaging in patients with Crohn disease: value of T2-versus T1-weighted gadolinium-enhanced MR sequences with use of an oral superpara magnetic contrast agent. Radiology 2005;238:517-30.
14. Casciani E, Masselli G, Di Nardo G, Polettini E, Bertini L, Oliva S, et al. MR enterography versus capsule endoscopy in paediatric patients with suspected Crohn\'s disease. Eur Radiol 2011;21:823-31.