MRT of the abdomen in combination with enteroclysis in Crohn disease with oral and intravenous Gd-DTPA

Summary In spite of the improved MR-diagnosis of the abdomen, MRI is not used as a routine method for the diagnosis of inflammatory small bowel disease. The aim of this study was – after optimazation of the bowel opacification – the correlation of the findings obtained with enteroclysis and MRI in patients with known Crohns' disease. 60 patients beween 17 and 72 years of age were investigated. First, an enteroclysis was performed in typical manner. The applicated methylcellulosis was blended with positive oral MR contrast media (Magnevist oral, Schering). After enteroclysis, MRI of the abdomen was performed using T1- and T2-weighted breathhold sequences (Flash 2D pre- and postcontrast and TSE) in axial and coronal planes. The lenght of the affected bowel and the stenosis seen with enteroclysis correlated well with the visible thickening of the small bowel wall and the stenosis seen in MRI. Using MRI, additional findings could be obtained in 28 patients, such as fistulas, abscesses or a hydronephrosis, or a better assessment of the stenosis was possible with MRI, because of the avoidance of overshadowing of the affected bowel loop with MRI. A brilliant MR-tomographic imaging of the small bowel is possible under the condition, that the small bowel contrast is optimal. The main prerequisite is a large filling volume of the small bowel to reach a homogeneous contrast and a good distension of the small bowel lumen.      

Breathhold MRI of the small bowel in Crohn's disease after enteroklysis with oral magnetic particles

Purpose: To evaluate the efficacy of breathhold MRI following enteroclysis with addition of oral magnetic particles to study the extension, detection of stenoses and extraluminal manifestations in Crohn's disease. Material and Methods: 18 patients with Crohn's disease and potential of surgical intervention were studied with enteroclysis with addition of oral magnetic particles. T1-/T2-weighted breathhold MRI w/o spectral fat suppression w/o i. v. Gd-DTPA was applied. Results: Typical findings were marked bowel wall thickening with strong contrast enhancement. 95.8 % of affected small bowel segments and 94.7 % of stenoses were correctly detected by MRI. All four fistulas were detected and important extraluminal findings were seen in 6/18 patients. Additionally, one ileoileal and two ileosigmoidal adhesions, two extraluminal abscesses and affection of the right ureter were delineated. Conclusion: MRI in Crohn's disease offers the potential to avoid radiation exposure in this relatively young patient group. Important additional findings relevant to indication of surgery are seen in approximately one third of cases. The replacement of transduodenal intubation by oral contrast application remains to be further studied.      

MRI in Crohn's disease after transduodenal contrast administration using negative oral MRI contrast media

PURPOSE: To compare the efficacy and quality of conventional and MR enteroclysis with different filling methods regarding the assessment of extension and extraluminal manifestations in Crohn's disease. MATERIAL AND METHODS: 190 patients with known Crohn's disease were studied following small bowel enteroclysis, after oral administration or direct transduodenal filling in the MRI-department.T1- and T2-weighted breathhold MRI-scans w/o spectral fat suppression w/o i.v. Gd-DTPA were applied using negative oral superparamagnetic contrast media. RESULTS: Typical findings were marked bowel wall thickening with laminated wall contrast enhancement. In 135 patients 98,2% of affected bowel segments, 97,5% of stenoses and all 16 fistulas were detected, when conventional enteroclysis was employed as standard of reference. Additional important extraluminal findings such as ileoileal (n = 18), ileosigmoidal adhesions (n = 12), extraluminal abscesses (n = 35) and pseudotumors (n = 8) were visualized in 73/135 patients. Concerning the distension of jejunum and ileum, oral filling was rated significantly inferior to transduodenal filling in all small bowel segments,whereas filling in the MRI-unit was rated superior to fluoroscopic, mostly due to a mean transport time of 20 min to the MRI-unit. CONCLUSION: No clinically important findings of enteroclysis were missed when using MRI. Therefore, in patients with Crohn's disease, conventional enteroclysis can be replaced by MRI. For optimal bowel distension oral contrast administration is inferior to transduodenal filling, if a larger time delay between filling and the MRI-scan can be avoided.     

Imaging of small bowel Crohn's disease: can abdominal CT replace barium radiography?

Objectives: To analyze imaging features and compare the diagnostic information provided by abdominal computed tomography (CT) and enteroclysis to see whether CT can replace barium examinations in the assessment of patients with small bowel Crohn's disease. Methods: The abdominal CT studies and enteroclysis of 33 patients with small bowel Crohn's disease who underwent both examinations within a 2-week time period were retrospectively reviewed and scored for the presence and severity of the following features: mural edema, ulceration, small bowel obstruction, stricture, sinus tract formation, fistula, abscess, extraintestinal manifestations, and total number of intestinal sites involved. The statistical significance of the differences for each variable was calculated. Results: A total of 37 case sets were reviewed. CT demonstrated 10 (27 %) abscesses compared to 7 (19 %) shown by enteroclysis. Three extraintestinal sites shown by CT were not demonstrated by enteroclysis. Enteroclysis detected a larger number of intestinal sites of involvement (54 vs. 47) and more cases of ulceration (78 % vs. 19 %), small bowel obstruction (46 % vs. 16 %), stricture (38 % vs. 11 %), fistula formation (24 % vs. 8 %), and sinus tract formation (27 % vs. 5 %). The two modalities were similar in characterizing and grading the severity of mural thickening (CT showed 41 %, enteroclysis 46 %). Conclusion: Abdominal CT and enteroclysis provide unique and complementary diagnostic information in patients with Crohn's disease of the small bowel. Both methods may be required for the accurate assessment of the severity and extent of Crohn's disease of the small bowel. The choice of initial examination will depend on the clinical issue in question.     

Blueberry juice used per os in upper abdominal MR imaging: composition and initial clinical data

The aim of this study was to evaluate the use of a commercially available blueberry juice (BJ) both as a positive and negative oral contrast agent and to present the exact contents of paramagnetic ions. The concentration of Mn and Fe were determined in tinned myrtilles in syrup (atomic absorption). Nine healthy volunteers and 12 patients (age range 20–65 years) were examined using a 1-T MR scanner before and after per os administration of 430 ml of BJ. A qualitative analysis of signal alterations in the stomach, duodenum, and proximal small intestine was performed. In addition, a quantitative analysis was assessed in terms of signal-to-noise ratio calculation. The mean concentration ( ×± SD) of the ions found in the content of the three cans were 3.3 ± 0.4 μg/g for iron and 20.6 ± 2.6 μg/g for manganese. Based on the qualitative evaluation, signal alteration on T1-weighted images after administration of BJ was statistically significant in the stomach and duodenum, but not in the proximal small bowel. Signal alteration on T2-weighted images was not statistically significant in any part of the gastrointestinal tract. The quantitative analysis of the T1- and T2 shortening showed that BJ is efficient with only T1-weighted sequences, and this applied to the stomach, duodenum, and proximal small bowel. Blueberry juice can be used as an oral contrast agent in upper abdominal MR for T1-weighted imaging. Received: 7 September 1999; Revised: 29 November 1999; Accepted: 16 February 2000     

Multifocal nodular fatty infiltration of the liver mimicking metastatic disease on CT: imaging findings and diagnosis using MR imaging

The aim of this study was to describe the MR appearance of multifocal nodular fatty infiltration of the liver (MNFIL) using T1-weighted in-phase (IP) and opposed-phase (OP) gradient-echo as well as T2-weighted turbo-spin-echo sequences with fat suppression (FSTSE) and without (HASTE). Magnetic resonance imaging examinations at 1.5 T using T1-weighted IP and OP-GRE with fast low angle shot (FLASH) technique, and T2-weighted FSTSE, T2-weighted HASTE of 137 patients undergoing evaluation for focal liver lesions were reviewed. Five patients were identified in whom CT indicated metastatic disease; however, no liver malignancy was finally proven. Diagnosis was confirmed by biopsy (n = 3), additional wedge resection (n = 1) or follow-up MRI 6–12 months later (n = 5). Regarding the identified five patients, the number of focal liver lesions was 2 (n = 2) and more than 20 (n = 3). The MR imaging characteristics were as follows: OP-image: markedly hypointense (n = 5); IP image: isointense (n = 2) or slightly hyperintense (n = 3); T2-weighted FSTSE-image: isointense (n = 5); T2-weighted HASTE image isointense (n = 1); slightly hyperintense (n = 4). On OP images all lesions were sharply demarcated and of almost spherical configuration (n = 5). Further evaluation by histology or follow-up MR imaging did not give evidence of malignancy in any case. Histology revealed fatty infiltration of the liver parenchyma in three patients. Magnetic resonance follow-up showed complete resolution in two patients and no change in three patients. Multifocal nodular fatty infiltration can simulate metastatic disease on both CT and MR imaging. The combination of in-phase (IP) and opposed-phase (OP) gradient-echo imaging can reliably differentiate MNFIL from metastatic disease. Received: 15 September 1999 Revised: 3 February 2000; Accepted: 7 February 2000     

Ultrasound and magnetic resonance imaging in Crohn's disease: a comparison

The objective of this retrospective study was to compare MRI of the abdomen with ultrasound of the abdomen and gastrointestinal tract in patients with Crohn's disease. Forty-six patients were included in the study. We analyzed the localization of Crohn's lesions, the number of affected bowel segments, the number of stenoses, and the presence of abscesses, fistulae, and any additional findings. Findings were verified by means of one or more of the following: enteroclysis; surgical findings; and colonoscopy. The results show that MRI is superior to ultrasound in the localization of affected bowel segments (sensitivity: MRI 97.5%; US 76%) and in recognizing fistulae (sensitivity: MRI 87%; US 31%), stenoses (sensitivity: MRI 100%; US 58%) and abscesses (sensitivity: MRI 100%; US 89%). Magnetic resonance imaging of the abdomen should be obtained to clarify discrepant clinical and sonographic findings. In addition, despite its higher cost, MRI of the abdomen is justified in patients in whom Crohn's lesions are known or suspected in anatomic areas proximal to the terminal or neoterminal ileum and in cases with suspicion of fistulae and abscesses.     

Accuracy of enteroclysis in Crohn's disease of the small bowel: a retrospective study

The aim of this study was to evaluate the accuracy of enteroclysis in the diagnosis of Crohn's disease of the small bowel in a group of consecutive patients. From January 1992 to December 1995, 165 patients with suspected Crohn's disease of the small bowel presented to our institution for enteroclysis. In 14 patients up to three enteroclysis exams were performed. Most patients (78 %) underwent colonoscopy and retrograde ileoscopy. In the remaining patients clinical follow-up was used as gold standard. In 79 patients no radiographic abnormalities were found. Sixty-one patients (40 men and 21 women; mean age 34.2 years) had a radiological diagnosis of Crohn's disease. This involved the terminal ileum in 39 patients (64 %) either alone (n = 25) or in association with the pelvic ileum (n = 14). In 12 of these patients retrograde ileoscopy was not feasible. Twenty-one patients underwent surgery. In 4 patients pathology revealed diseases other than Crohn's. These patients had all ileocecal diseases (tuberculosis = 2; non-Hodgkin's lymphoma = 1; adenocarcinoma = 1). One false-negative result was observed. Overall, enteroclysis showed a sensitivity of 98.2 % and a positive predictive value of 93.4 %. Enteroclysis is a sensitive technique in evaluating both the extent and the severity of small bowel involvement in Crohn's disease, although the overlap of radiographic findings may hamper its accuracy when the disease is confined to the ileocecal area. Received: 25 November 1999; Revised: 4 April 2000; Accepted: 6 April 2000     

MRI in Crohn's disease

Technological developments have extended the role of MRI in the evaluation of the gastrointestinal tract. The potential of MRI to evaluate disease activity in Crohn's disease has been investigated extensively, as MRI has intrinsic advantages over other techniques, including noninvasiveness and the absence of ionizing radiation. For perianal fistulizing disease MRI has become a mainstay in evaluation of disease, as localization and extent of disease can be very well appreciated using both T2-weighted and T1-weighted sequences, fat suppression, and intravenous contrast medium. Imaging of the small bowel and colon in Crohn's disease is more complicated due to bowel peristalsis and respiratory movement. However, using fast breathhold sequences and intravenous spasmolytic medication, images of good diagnostic quality can be acquired. To obtain sufficiently distended bowel, which in our estimation is a prerequisite for evaluation of the bowel, MR enteroclysis can be performed. However, applicability of different oral contrast media has been studied, as a noninvasive method for bowel distension would be preferable. Abdominal MRI is a valuable imaging technique for evaluation of luminal, transmural, and extraintestinal manifestations of Crohn's disease as degree of disease activity, presence of luminal pathology (e.g., stenoses), and extraintestinal manifestations of disease (e.g., abscesses, fistulas) can be accurately assessed.     

MRI of the plantar structures of the foot after falanga torture

Falanga is an ancient form of punishment or torture but is still commonly reported by our refugees. The late result of caning the heel and ball of the foot is a chronic painful condition with few clinical signs. The aim of the present study was to assess, by MRI, possible morphologic characteristics of the heel and ball of the foot, related to falanga and pain in correlation to clinical findings. Magnetic resonance imaging of the foot was obtained in 12 victims exposed to falanga torture and 9 healthy volunteers. Sagittal T1-weighted spin-echo images (TR 616–840 ms, TE 20 ms), T2-weighted spin-echo images (TR 1900 ms, TE 90 ms), and short tau inversion recovery (STIR) images (TR 1200 ms, TE 15 ms, TI 100 ms) were performed. The central portion of the plantar aponeurosis was generally significantly thicker in victims exposed to falanga torture as compared with that of controls (P < 0.05). In all except one of the victims, MRI demonstrated two layers of the thickened plantar aponeurosis: a deeper portion with normal homogeneous low signal intensity (SI) appearance, and a superficial layer with characteristic areas of mixed SI on both T1- and T2-weighted images. There were no signs of chronic muscular compartment syndromes, and the thickness of the plantar pad did not differ between the two groups. Magnetic resonance imaging may demonstrate morphologic characteristics of the plantar aponeurosis which may confirm falanga torture. Further imaging with more specific sequences is warranted to demonstrate the supposed injuries in the compartmental fat tissue chambers and the vascularity of the ball pad of the foot. Received: 25 November 1999; Revised: 8 February 2000; Accepted: 6 April 2000     

Safety of colon contrast material in (helical) CT examination of patients with suspected diverticulitis

Purpose: Today patients with suspected diverticulitis are commonly imaged with a CT scan utilizing a variety of methods of contrast medium administration. Although CT with rectally administered colon contrast has demonstrated a high diagnostic accuracy, concerns have been raised over its safety in patients with diverticulitis. The following retrospective investigation was undertaken to answer this concern. Materials and methods: Between January, 1997, and July, 1999, 308 patients with suspected diverticulitis were examined in the Emergency Radiology Division of the Massachusetts General Hospital by means of a helical CT scan performed with rectally administered colon contrast material. The patients' CT findings were correlated with their clinical courses and/or surgical findings. Results: Of the 308 CT scans, 115 (37.3 %) were positive for diverticulitis. Of 193 patients without diverticulitis, 91 (47.1 % of patients without diverticulitis) had alternative diagnoses made by CT, such as small bowel obstruction, epiploic appendagitis, and urinary tract calculi. No complications due to colon contrast material were noted. Even in cases where there was free extraluminal air, no extravasated contrast material was seen. Rectally administered colon contrast was well tolerated by patients and provided excellent large bowel opacification. This method saved imaging time in an emergency setting, as there was no need to wait for oral contrast to reach the colon. Conclusion: Out of 308 CT scans, there were no complications from the use of rectally administered colon contrast material. No case of traumatic bowel perforation secondary to colon contrast material was observed. Rectally administered colon contrast material is safe in the CT examination of patients with suspected diverticulitis.     

A prospective randomized comparison between two MRI studies of the small bowel in Crohn’s disease,the oral contrast method and MR enteroclysis

The aim was to compare bowel distension and diagnostic properties of magnetic resonance imaging of the small bowel with oral contrast (MRI per OS) with magnetic resonance enteroclysis (MRE). Forty patients with suspected Crohn’s disease (CD) were examined with both MRI methods. MRI per OS was performed with a 6% mannitol solution and MRE with nasojejunal intubation and a polyethylenglycol solution. MRI protocol consisted of balanced fast field echo (B-FFE), T2 and T1 sequences with and without gadolinium. Two experienced radiologists individually evaluated bowel distension and pathological findings including wall thickness (BWT), contrast enhancement (BWE), ulcer (BWU), stenosis (BWS) and edema (EDM). The diameter of the small bowel was smaller with MRI per OS than with MRE (difference jejunum: 0.55 cm, p < 0.001; ileum: 0.35 cm, p < 0.001, terminal ileum: 0.09 cm, p = 0.08). However, CD was diagnosed with high diagnostic accuracy (sensitivity, specificity, positive and negative predictive values: MRI per OS 88%, 89%, 89%, 89%; MRE 88%, 84%, 82%, 89%) and inter-observer agreement (MRI per OS k = 0.95; MRE k = 1). In conclusion, bowel distension was inferior in MRI per OS compared to MRE. However, both methods diagnosed CD with a high diagnostic accuracy and reproducibility.     

Comparative efficacy of and sequence choice for two oral contrast agents used during MR imaging.

OBJECTIVE: Our objective was to compare the efficacy of a positive and a negative oral contrast agent and to determine the optimal sequence choice for use in pelvic MR imaging. SUBJECTS AND METHODS: We undertook a prospective randomized trial of 57 patients with pelvic cancer who were examined with MR imaging after oral administration of a positive contrast agent (27 patients) or a negative contrast agent (30 patients). T1- and T2-weighted breath-hold and non-breath-hold gradient-recalled echo and turbo spin-echo sequences were obtained. Using the hard-copy images, we graded filling and distention of the small bowel, bowel wall conspicuity, delineation of normal and pathologic structures, and artifacts. RESULTS: Good or excellent small-bowel filling and distention was obtained in 17 patients (63%) receiving the positive agent and in 26 patients (87%) receiving the negative agent, and bowel wall conspicuity was graded good or excellent in 19 patients (70%) and 20 patients (67%), respectively. Normal and pathologic structures were better delineated with the negative agent (20 patients [74%] and 27 patients [90%], respectively; p = .02). Breath-hold gradient-recalled echo T1-weighted images were preferred for the positive agent (78%), and breath-hold T2-weighted images were preferred for the negative agent (93%). Contrast artifacts were more frequently seen with the negative agent (11% and 93%, respectively; p = .0001), and such artifacts were eliminated using T2-weighted sequences. CONCLUSION: Both contrast agents were effective in pelvic MR imaging, but delineation of normal and pathologic structures was better with the negative agent. Gradient-recalled echo T1-weighted sequences are recommended for positive contrast agents, and breath-hold T2-weighted sequences are recommended for negative contrast agents.     

Masses of the tongue and floor of the mouth: findings on magnetic resonance imaging

The purpose of this study was to determine the value of MR imaging for the demonstration of masses in the tongue and floor of the mouth. Nine patients were prospectively examined with MR imaging after physical examination. Imaging protocol included T2 and contrast-enhanced and non-contrast-enhanced T1-weighted turbo spin-echo sequences, and the findings were compared with surgical and histopathological results. Histopathological examination revealed four squamous cell carcinomas, one adenoid cystic carcinoma, two tongue abscesses, and one chronic inflammatory change. The other case was diagnosed as hemangioma depending on clinical and imaging findings alone. In cases with squamous cell carcinoma, staging was done on the basis of MR imaging findings, and was found to be T4 in two cases, T3 in one case, and T2 in another. The primary role of MR imaging of the tongue and oropharynx is not to make a tissue diagnosis. Multiple deep biopsies are mandatory for the differentiation of other inflammatory and neoplastic lesions. Magnetic resonance imaging produces coronal and sagittal image planes to assess the volume and spread of the lesion and helps the surgeon determine the direction in which the biopsy should be performed. Received: 4 October 1999; Revised: 31 January 2000; Accepted: 6 April 2000     

Small volume hyperosmolar abdominal CT in acute small bowel obstruction

Purpose: Identification of the degree of small bowel obstruction (SBO) is helpful in patient management decisions since the treatment of partial SBO is nonoperative, whereas the treatment of complete SBO requires surgical intervention. We describe a technique providing the benefits of CT and enteroclysis together, which allows the radiologist to determine the degree of the SBO and thereby to distinguish partial from complete SBO. Methods and materials: Abdominal CT scans were performed on 21 patients. Patients were given 200 ml concentrated oral contrast medium. Serial radiographs followed the progression of contrast at 1, 3, and 5 h in order to optimize the timing of the abdominal CT. A partial SBO was diagnosed if the oral contrast reached the colon at or before the 5-h film, and a complete SBO if the oral contrast did not reach the colon by the 5-h abdominal radiograph. The CT was performed when the contrast reached the colon or after the 5-h film. Results: CT findings demonstrated complete obstruction in eight patients, all of whom required surgical treatment. Partial SBO was demonstrated in 13 patients. Ten of the 13 patients with partial SBO had an uncomplicated transition zone, probably due to adhesions, and were treated medically. Nine of these patients were managed with nasogastric tube decompression and were discharged without surgical intervention. One patient failed to respond to a 12-h course of nasogastric tube decompression and underwent lysis of adhesions. Three of the 13 patients had a complicated transition zone, due to an intussusception and two hernias, which were responsible for the SBO. Two were treated surgically and one medically. Conclusion: The use of small-volume hyperosmolar abdominal CT allows the radiologist to determine the degree of the SBO. This additional information is very useful in patient management because uncomplicated partial SBO is often treated successfully via tube decompression, whereas complete SBO requires surgical intervention.     

Tumors of the cardiac valves: imaging findings in magnetic resonance imaging, electron beam computed tomography, and echocardiography

We describe the findings from various cross-sectional imaging modalities in patients with cardiac valve adherent masses. The techniques are discussed, and imaging findings are compared with the results of cardiac surgery. All three patients had neurological symptoms and/or cardiac murmurs. Transthoracic and/or transesophageal echocardiography revealed the cardiac mass in all three. For differentiation of thrombus and cardiac neoplasm magnetic resonance imaging (MRI) was also performed in all three patients and electron-beam computed tomography (EBCT) in two. Fast segmented cine gradient-echo MRI techniques provided mass depiction in all patients, while T1-weighted spin-echo imaging failed in mass detection in one patient. None of the patients showed evidence of valve regurgitation or stenosis in flow sensitive cine MRI. EBCT excluded mass calcifications in both patients and reliably demonstrated the valve attached lesions. Although echocardiography is the modality of choice in evaluating cardiac masses and especially valve attached masses, MRI and EBCT provide additional information about tissue characteristics and allows an excellent overview of the cardiac and paracardiac morphology. Fast segmented cine gradient-echo MRI is especially able to depict even small tumors attached to rapidly moving cardiac valves, and valve competence can be easily assessed within the same examination. Received: 17 December 1998; Revision received: 1 June 1999; Accepted: 10 August 1999     

Computed tomographic enterography and enteroclysis: pearls and pitfalls

Computed tomographic (CT) enterography and enteroclysis improve visualization of the small bowel mucosa and wall in comparison with traditional CT and fluoroscopic studies by distending the small bowel through enteric hyperhydration with a negative contrast agent. Although CT enterography is performed with oral hyperhydration, CT enteroclysis requires the placement of an enteroclysis tube, often in patients who are unable to orally consume the amount of liquid. When tolerated, CT enterography is often preferred due to its lack of invasiveness. Magnetic resonance enterography and enteroclysis are other modalities that are still being studied and show promise in the imaging of small bowel. Unlike small bowel follow-through, conventional enteroclysis, or capsule endoscopy, extraenteric findings are best assessed on CT enterography. These include findings in the surrounding mesentery, perienteric fat, and the adjacent solid organs that may be associated with the small bowel process and include fistulas or abscesses, mural hyperenhancement, prominent vasa recta, and other inflammatory changes. CT enterography has developed into the first-line modality in the imaging of Crohn's disease and is considered the most appropriate imaging modality in patients with suspected Crohn's disease. It is also increasingly being used in the assessment of small bowel infections, neoplasms, adhesions, and polyps.     

Role of radiology in the diagnosis of neurosarcoidosis

Clinical studies report a rate of 5 % and autopsy results a rate of 25 % of brain involvement in sarcoidosis. The aim of this study was to evaluate the role of radiology in the diagnosis of patients with neurosarcoidosis. The chest radiographs and MRI brain scans of 22 patients with sarcoidosis were retrospectively reviewed, along with the information that was provided in the request form and clinical charts. All patients had neurological signs and symptoms; 21 patients were examined with contrast enhancement. Facial nerve paralysis was the most common clinical manifestation identified in 10 patients. A wide spectrum of MR findings was noted: periventricular high-signal lesions on T2-weighted images (46 %); multiple supratentorial and infratentorial brain lesions (36 %); solitary intra-axial mass (9 %); solitary extra-axial mass (5 %); and leptomeningeal enhancement (36 %). Neurological signs and symptoms can be significant manifestations of sarcoidosis. Magnetic resonance imaging shows a wide spectrum of brain abnormalities associated with neurosarcoidosis. The patient's history and chest X-ray are helpful in arriving at the correct diagnosis, but in selected cases with isolated brain involvement biopsy may still be required. Received: 21 June 1999; Revised: 27 August 1999; Accepted: 15 October 1999     

Discriminatory power of MRI for differentiation of adrenal non-adenomas vs adenomas evaluated by means of ROC analysis: Can biopsy be obviated?

The purpose of our study was to evaluate the discriminatory power of MRI in high-field magnet (1.5 T) for differentiation of adrenal non-adenomas vs adenomas assessing the following parameters separately and in combination: mean diameter of adrenal mass; previously described and new ratios as well as index calculated from signal intensity (SI) on SE T2-weighted images, chemical shift imaging (CSI), and Gd-DTPA-enhanced dynamic studies. One hundred eight adrenal masses (36 non-hyperfunctioning adenomas, 27 pheochromocytomas, 23 aldosterone-secreting adenomas, 20 malignant masses and 2 cortisol-secreting adenomas) in 95 patients were evaluated with SE sequences, CSI and Gd-DTPA dynamic studies. Indices and ratios of SI for all examined MRI methods were calculated and examined retrospectively for significance of differences between the groups with calculation of sensitivity and specificity. Receiver operating characteristics (ROC) analysis of calculated parameters in combination was performed. The multifactorial analysis of all four parameters, including size of the tumor, T2_liver index, CSI ratio reflecting lipid content in the tumor and Wo_max/last ratio reflecting maximal washout of contrast agent from the tumor had 100 % sensitivity and 100 % specificity in characterization of adrenal non-adenoma. The best performance of combination of mean tumor diameter with single MRI SI parameter was achieved in combination with T2_liver index for all adrenal masses (area under ROC 0.987) and CSI ratio for non-hyperfunctioning adrenal masses (area under ROC 0.991). Magnetic resonance imaging enables sensitive and specific diagnosis of adrenal non-adenoma. Received: 18 June 1998; Revised: 11 January 1999; Accepted: 5 May 1999     

MR appearance of parasymphyseal insufficiency fractures of the os pubis

 Objective. To clarify the MRI features of parasymphyseal insufficiency fractures of the os pubis. Design and patients. MRI was performed in four postmenopausal women with parasymphyseal insufficiency fractures. The diagnosis was confirmed with plain films in every patient. T1-weighted and T2-weighted images were obtained in four patients using a 1.5-T unit. Postcontrast T1-weighted imaging was also done in three patients. Results and conclusions. MRI of pubic parasymphyseal insufficiency fracture characteristically demonstrates a hyperintense mass lesion with a hypointense rim on T2-weighted imaging, showing peripheral and septal enhancement after contrast administration. It is important to have this entity in mind in patients with osteoporosis, especially in patients with a history of pelvic irradiation for malignant disease, so as not to misinterpret it as a chondroid tumor or bone metastasis.