Search

A new MR technique is proposed to study the renal collecting system and the proximal ureteral portion in patients without obstructive uropathy, using fat-suppressed Turbo Spin Echo sequences, after drug-induced distension of the urinary tract. Nine normal volunteers and 16 patients (11 with renal stones, 1 with bilateral pyeloureteral junction stenosis, 1 with renal ptosis, 1 with ureteral hyperkinesia, 1 with renal tumor and 1 with a symptomatic renal cyst) were submitted to MR-Urography, performed with a 3D non-breath-hold fat-suppressed Turbo SE sequence (TR = 3000 ms, TE = 800 ms, 6 acquisitions, Turbo Factor = 128, Matrix = 128 x 256 or 256 x 256, acquisition time = 5 min 15 sec or 10 min 36 sec) on the coronal plane. With these acquisition parameters, parenchymal signal can be completely suppressed while enhancing fluid signal. These acquisitions were post-processed with the MIP algorithm to obtain very similar images to those of conventional urography. The maximum filling of the renal collecting system was obtained with the i.v. administration of 250 ml saline solution within 2-3 minutes and then the i.v. injection of 20 mg furosemide. The renal collecting system was optimally depicted in all the volunteers and the patients, except for 2 cases because of malfunctioning respiratory gating. Anatomical detailing was really improved after the diuretic administration, especially in the study of the major caliceal systems. To conclude, MR-Urography permits accurate morphological detailing of the renal collecting system also in patients without obstructive uropathy.

The latest MR units are provided with the Turbo-Field-Echo technique which permits gradient-echo imaging with very short TE and TR and is optimized to yield the highest possible image quality within a very short acquisition time--i.e., less than 5 seconds. This dynamic study was aimed at depicting normal laryngeal and oropharyngeal movements during maximal inspiration and the prolonged uttering of vowels, in both a normal and a loud voice. We examined 10 healthy volunteers (7 men and 3 women) with an 0.5-T superconductive unit (Gyroscan T5 III, Philips Medical System). We used Turbo-Field-Echo sequences with the following acquisition parameters: TR 12 ms, TE 6 ms, flip angle 30 degrees, 4 acquisitions, acquisition time: 5 seconds. A single coronal scan was acquired at the larynx, while a midsagittal scan and 2 coronal scans were acquired at the oropharynx. The volunteers were asked to breathe in long and deep, to prolong the emission of the vowel [i] during laryngeal studies and to prolong the emission of the fundamental vowels [a], [i] and [u] in a normal (50 db) and a loud (70 db) voice during oropharyngeal studies. The movements of true and false vocal cords were clearly depicted in all the volunteers, and the activity of the different anatomical structures of the oropharyngeal cavity (lips, tongue, hard and soft palate, pharynx and epiglottis) was also demonstrated. During vowel production in a loud voice, the vocal tract was enlarged at the oral cavity for the vowels [a] and [u] and at the pharynx for the vowel [i]. To conclude, fast MRI with midsagittal scans is the best imaging modality to study different vocal tract patterns during speech and can thus replace midsagittal radiography and xeroradiography in the study of vowel production.