Colon diseases: MR evaluation using combined T2-weighted single-shot echo train spin-echo and gadolinium-enhanced spoiled gradient-echo sequences

This study demonstrates the appearance of large bowel diseases on magnetic resonance (MR) images using breath-hold T2-weighted half-Fourier acquisition snapshot turbo spin-echo (HASTE), breath-hold T1-weighted spoiled gradient-echo (SGE), and breath-hold gadolinium-enhanced T1-weighted SGE with and without fat-suppression sequences. The study represents a collective experience using a generalized combined abdominal-pelvic imaging protocol. Of 29 patients, 27 had surgical, endoscopic, microbiological, and/or histopathological correlation, and 2 had a diagnosis based on characteristic imaging findings. Fifteen patients had neoplastic disease including colon adenocarcinoma (n = 11), rectosigmoid carcinoid (n = 1), familial adenomatous polyposis (n = 2), and cecal lipoma (n = 1). Fourteen patients had non-neoplastic disease including diverticulosis (n = 6), ischemic colitis (n = 2), pseudomembranous colitis (n = 2), acute appendicitis with periappendiceal abscess (n = 2), Mycobacterium avium intracellulare (MAI) colitis (n = 1), and Crohn's proctocolitis (n = 1). In all 15 patients with neoplastic diseases, MR imaging depicted the primary lesions and demonstrated local extent. Mass lesions were best shown on T2-weighted HASTE and gadolinium-enhanced fat-suppressed SGE images. Of 14 patients with non-neoplastic diseases, inflammatory changes were best shown on gadolinium-enhanced fat-suppressed T1-weighted SGE images in all cases. MR imaging with fast scanning breath-hold techniques and intravenous gadolinium enhancement provided good depiction and characterization of large bowel diseases.     

Magnetization transfer effects in MR-detected multiple sclerosis lesions: comparison with gadolinium-enhanced spin-echo images and nonenhanced T1-weighted images.

PURPOSETo define the relationship between magnetization transfer and blood-brain-barrier breakdown in multiple sclerosis lesions using gadolinium enhancement as an index of the latter.METHODSTwo hundred twenty lesions (high-signal abnormalities on T2-weighted images) in 35 multiple sclerosis patients were studied with gadolinium-enhanced spin-echo imaging and magnetization transfer. Lesions were divided into groups having nodular or uniform enhancement, ring enhancement, or no enhancement after gadolinium administration. For 133 lesions, T1-weighted images without contrast enhancement were also analyzed. These lesions were categorized as isointense or hypointense based on their appearance on the unenhanced T1-weighted images.RESULTSThere was no difference between the magnetization transfer ratio (MTR) of lesions as a function of enhancement. MTR of hypointense lesions on unenhanced T1-weighted images was, however, lower than the MTR of isointense lesions.CONCLUSIONWe speculate that diminished MTR may reflect diminished myelin content and that hypointensity on T1-weighted images corresponds to demyelination. Central regions of ring-enhancing lesions had a lower MTR than the periphery, suggesting that demyelination in multiple sclerosis lesions occurs centrifugally. In addition, the short-repetition-time pulse sequence seems useful in the evaluation of myelin loss in patients with multiple sclerosis.     

Analysis of artefacts and detail resolution of lung MRI with breath-hold T1-weighted gradient-echo and T2-weighted fast spin-echo sequences with respiratory triggering

The aim of this study was to evaluate feasibility and limitations of two MR sequences for imaging of the lung using a semi-quantitative rating scale. Ten healthy volunteers were assessed with a breath-hold T1-weighted gradient-recalled-echo (TR/TE=129/2.2 ms, matrix 173 x 256) and a T2-weighted turbo spin-echo (TSE) sequence with respiratory triggering (TR/TE=3000-5000/120 ms, matrix 270 x 512) in axial 6-mm slices. The T1-weighted GRE protocol included a pre-saturation pulse over the mediastinal structures. Artefacts and resolution of vessel/airway structures in each lung segment were evaluated by two observers (10 volunteers, 180 segments). Cardiac and vessel pulsation artefacts predominated on T1-weighted GRE, respiration artefacts on T2-weighted TSE (lingula and middle lobe). Pre-saturation of the mediastinum reduced pulsation artefacts on T1-weighted GRE. T1-weighted GRE images were improved by bright flow signal of vessels, whereas image quality of T2-weighted TSE was reduced by black-blood effects in central parts of the lung. Delineation of lung periphery and the mediastinum was superior with T2-weighted TSE. Segmental/sub-segmental vessels (up to fourth/fifth order) and bronchi (up to third order) were identified. All 180 lung segments were imaged in diagnostic quality with at least one of the two sequences (T1-weighted GRE not diagnostic in 9 of 180, T2-weighted TSE in 4 of 180). Both sequences were found to be complementary: superior identification of gross lung anatomy with T1-weighted GRE and higher detail resolution in the periphery and the mediastinum with T2--weighted TSE.     

Comparison of gradient-recalled-echo and T2-weighted spin-echo pulse sequences in intramedullary spinal lesions

Nineteen consecutive patients with spinal intramedullary lesions were studied on a 1.5-T system to compare the quality of T2-weighted spin-echo and gradient-recalled-echo (GRE) pulse sequences. Direct comparisons were made in the sagittal and/or axial planes. Twenty-four studies were performed in the 19 patients. The gradient echoes were usually performed at 300/14 (TR/TE) with a flip angle of 10 degrees. Although no significant diagnostic differences were noted in the sagittal plane, there was a distinct anatomic advantage for GRE imaging over spin-echo imaging in the axial plane. This is believed to be the result of CSF time-of-flight effects in the slice-select direction, which are not compensated for by flow-compensating gradients on the spin-echo images, but which are insignificant in the GRE sequence used in this study. Pathology was seen equally well or better on GRE in 79% (19/24) of the sequences. In the other five cases, the spin-echo image showed a brighter intramedullary signal than that seen on GRE, although GRE showed the lesion in all cases. Our results indicate that properly optimized GRE imaging on a high-field-strength system can replace spin-echo imaging in the spine when intramedullary disease is suspected and that the benefits of GRE are most striking in the axial plane.     

Magnetic resonance imaging of the brain with gadopentetate dimeglumine-DTPA: Comparison of T1-weighted spin-echo and 3D gradient-echo sequences

Short TR, short TE, high resolution, 3D gradient-recalled echo (GRE) imaging was evaluated for lesion detection in the brain. High resolution 3D GRE data acquisition was used to reduce partial volume effects and flow artifacts, to better visualize smaller structures, to minimize signal losses caused by field inhomogeneities, and to allow better image reformatting. Spin-echo (SE) and 3D GRE approaches were compared for lesion detection after the administration of an MR contrast agent, gadopentetate dimeglumine. Preliminary clinical studies demonstrated that the signal-to-noise ratio (SNR) in each slice of the GRE scan was worse than that of the SE scan because of the much thicker slices acquired with the SE technique. However, by averaging two adjacent 3D slices, the SNR of the two methods was essentially equivalent. In the averaged GRE slices, large lesions were seen just as well as in the SE images. More importantly, small lesions were better visualized in the thin 3D GRE images than in the thick SE images for the lesions studied in this work and the protocols used. These observations were confirmed by theoretical simulations.     

Pyogenic hepatic abscesses: MRI findings on T1- and T2-weighted and serial gadolinium-enhanced gradient-echo images

The purpose of this study was to determine the magnetic resonance imaging (MRI) features of pyogenic hepatic abscesses on T1-weighted, T2-weighted, and serial gadolinium (Gd)-enhanced T1-weighted spoiled gradient-echo (SGE) images including images acquired in the immediate, intermediate, and late phases of enhancement. The MRI studies of 20 patients with pyogenic liver abscesses were retrospectively reviewed. All patients were examined on 1.5 (n = 19) and 1.0 (n = 1) T MR scanners. MR studies included T1-weighted, T2-weighted, and serial Gd-enhanced SGE images. The following determinations were made: signal intensity of the abscess cavity and perilesional liver tissue, and the presence of internal septations, layering material, or air in the abscess cavity. The pattern of enhancement of the abscess wall, internal septae and peri-abscess liver were evaluated on serial Gd-enhanced SGE images. A total of 53 abscesses were observed in the 20 patients. Fortyeight abscesses were hypointense on T1-weighted and hyperintense on T2-weighted images. Internal septations were present in four abscesses. Lower signal intensity material was observed in a dependent location on T2-weighted images in one abscess. Signal void foci of air located on the nondependent surface was observed in two abscesses. Two other abscesses contained signal void air that occupied the entire abscess cavity, observed on all imaging sequences. On serial gadolinium-enhanced images, all abscesses revealed early enhancement of the wall, which persisted with negligible change in degree of enhancement or thickness on delayed images. Abscess walls ranged in thickness from 2 to 5 mm. Internal septations ranged in thickness from 2 to 3 mm. Abscess walls and septations were relatively uniform in thickness with no evidence of focal nodularity. Periabscess liver tissue was mildly hypointense on T1-weighted and mildly hyperintense on T2-weighted images in 20 lesions, which were either circumferential (n = 12) or wedge-shaped (n = 8). All these regions enhanced more than the remainder of the liver on immediate post-gadolinium images and remained relatively hyperintense on late phase images. Periabscess liver parenchyma was isointense on both T1- and T2-weighted images in 18 lesions, and in these lesions wedge-shaped subsegmental (n = 6) or segmental (n = 12) enhancement was observed on immediate gadolinium-enhanced images, which faded to isointensity on intermediate phase images. No perilesional signal changes and enhancement difference was observed in 15 lesions. Characteristic features of abscesses include: intense mural enhancement on early gadolinium-enhanced images, which persists with negligible change in thickness and intensity on later post-gadolinium images, and the presence of periabscess increased enhancement on immediate post-gadolinium images. These MRI features may help to distinguish abscesses from other focal liver lesions during differential diagnosis.     

Optimal 3-T MRI for depiction of the finger A2 pulley: comparison between T1-weighted, fat-saturated T2-weighted and gadolinium-enhanced fat-saturated T1-weighted sequences

Objective To compare three spin-echo sequences, transverse T1-weighted (T1WI), transverse fat-saturated (FS) T2-weighted (T2WI), and transverse gadolinium-enhanced (Gd) FS T1WI, for the visualisation of normal and abnormal finger A2 pulley with magnetic resonance (MR) imaging at 3 tesla (T). Materials and methods Sixty-three fingers from 21 patients were consecutively investigated. Two musculoskeletal radiologists retrospectively compared all sequences to assess the visibility of normal and abnormal A2 pulleys and the presence of motion or ghost artefacts. Results Normal and abnormal A2 pulleys were visible in 94% (59/63) and 95% (60/63) on T1WI sequences, in 63% (40/63) and 60% (38/63) on FS T2WI sequences, and in 87% (55/63) and 73% (46/63) on Gd FS T1WI sequences when read by the first and second observer, respectively. Motion and ghost artefacts were higher on FS T2WI sequences. Seven among eight abnormal A2 pulleys were detected, and were best depicted with Gd FS T1WI sequences in 71% (5/7) and 86% (6/7) by the first and the second observer, respectively. Conclusion In 3-T MRI, the comparison between transverse T1WI, FS T2WI, and Gd FS T1WI sequences shows that transverse T1WI allows excellent depiction of the A2 pulley, that FS T2WI suffers from a higher rate of motion and ghost artefacts, and transverse Gd FS T1WI is the best sequence for the depiction of abnormal A2 pulley.     

Stomach diseases: MR evaluation using combined t2-weighted single-shot echo train spin-echo and gadolinium-enhanced spoiled gradient-echo sequences

The aim of this study was to describe the magnetic resonance (MR) appearances of diseases of the stomach using combined T1-weighted spoiled gradient-echo, T2-weighted single-shot echo train spin-echo and gadolinium-enhanced spoiled gradient-echo sequences. All patients with gastric diseases who underwent combined T2-weighted single-shot echo train spin-echo and gadolinium-enhanced conventional and fat-suppressed spoiled gradient-echo imaging between October 1, 1996 and March 1, 1999, and who had histological or other imaging proof of disease, were included in this study. This patient population was comprised of 40 patients with subsequently proven gastric abnormalities, including malignant tumors (25) or benign disease (15). The MRI sequences included T1-weighted, T2-weighted, and early and late gadolinium-enhanced spoiled gradient-echo (SGE) images. Evaluation was made of the following parameters: a) the ability to detect the disease process on MRI, by comparing the original prospective MR reports with the records of the pathology department; b) the MR appearance of a variety of gastric diseases; and c) the sequences that most clearly demonstrated abnormalities by retrospective review of the MR studies. MR images demonstrated 22 of 25 malignant tumors. Evaluation of the extent of the tumor was correctly shown in 22 of 25 tumors. Small-volume tumor (one patient with gastric adenocarcinoma, and one patient with lymphoma) and coexistent infiltrative adenocarcinoma and gastritis (one patient) rendered demonstration of tumor poor on MR images in three patients. Tumors were mildly hypointense on T1-weighted images and mildly hyperintense on T2-weighted images. Tumors enhanced in a heterogeneous fashion compared with background stomach wall, but they ranged from hypointense to hyperintense on early and late post-gadolinium SGE images. Regarding benign diseases, the changes of gastritis were evident in three of four cases. Gastritis appeared as increased mucosal enhancement that ranged from moderate to intense on early and late post-gadolinium SGE images. Imaging findings of the various entities are described in greater detail in the text. MR findings in a variety of neoplastic and non-neoplastic diseases of the stomach are described. Neoplastic diseases were consistently observed in most cases; however, small tumors and tumors coexistent with inflammatory changes were poorly evaluated. The changes of gastritis were demonstrated as increased enhancement of the gastric wall. J. Magn. Reson. Imaging 10:950-960, 1999.     

Segmentation of gadolinium-enhanced lesions on MRI in multiple sclerosis

PURPOSE: To develop and implement a method for identification and quantification of gadolinium (Gd) enhancements with minimal human intervention. MATERIALS AND METHODS: Dual fast spin echo (FSE), fluid attenuation inversion recovery (FLAIR), and pre- and postcontrast T1-weighted spin echo were acquired on 22 subjects. The enhancements were identified on the postcontrast T1-weighted images based on morphological operations. A single threshold based on the ratio of the difference of postcontrast and precontrast T1 images with that of precontrast T1 images is applied to the reconstructed images to reduce the false classifications. False classification of enhancements arising from enhancing vasculature and structures such as the choroid plexus that lack a blood-brain barrier were reduced by assuming that the true enhancements are always associated with hyperintense lesions on T2-weighted images (T2 lesions). The enhanced lesions were further delineated based on fuzzy connectivity. RESULTS: The segmented Gd enhancements were evaluated quantitatively with manually identified enhancements based on similarity measures. The average similarity index (SI) of 0.76 suggests excellent performance of the proposed methodology. The Bland-Altman plot shows a close agreement between the results obtained manually and those based on the proposed methodology. CONCLUSION: The proposed algorithm identifies and quantifies Gd enhancements accurately with minimal human intervention.     

Endometrial carcinoma in adenomyosis: assessment of myometrial invasion on T2-weighted spin-echo and gadolinium-enhanced T1-weighted images

OBJECTIVE: The aim of our study was to compare T2-weighted and contrast-enhanced dynamic T1-weighted images with histologic findings in assessing the depth of myometrial invasion by endometrial carcinoma in adenomyosis. MATERIALS AND METHODS: We retrospectively reviewed the MRIs of 11 patients who had a total of 12 lesions of endometrial carcinoma within adenomyosis. T2-weighted and contrast-enhanced dynamic T1-weighted images were compared with the histologic findings separately. We assessed the extent of myometrial invasion by endometrial carcinomas. The depth of myometrial invasion seen on MRI was classified as stage S (superficial invasion), stage D (deep invasion), or undetectable. The staging accuracies of each sequence were assessed. The tumor-myometrium contrast-to-noise ratios were calculated for each sequence. RESULTS: The histologic specimens revealed that myometrial invasion was deep in seven of 12 lesions and superficial in five. On T2-weighted images the depth of invasion was underestimated in two lesions and impossible to determine in five lesions. On dynamic T1-weighted images the depth of invasion was overestimated in one lesion and underestimated in one lesion. The staging accuracy on dynamic T1-weighted images (83%) was significantly higher than that on T2-weighted images (42%). The contrast-to-noise ratio was significantly higher on dynamic T1-weighted studies during the early phase (mean +/- SD, 2.68 +/- 0.94) than it was on T2-weighted studies (1.74 +/- 1.05) and during the delayed phase (2.01 +/- 0.86). CONCLUSION: When adenomyosis coexists with endometrial cancer at the same site on T2-weighted images, contrast-enhanced dynamic T1-weighted imaging improves the accuracy of staging.     

Theoretical analysis of gadopentetate dimeglumine enhancement in T1-weighted imaging of the brain: Comparison of two-dimensional spin-echo and three-dimensional gradient-echo sequences

By using a theoretical model, the signal difference-to noise ratios between simulated lesions and normal white matter and gray matter were calculated as a function of lesion concentration of gadopentetate dimeglumine (GD) for two-dimensional (2D) T1-weighted spin-echo (SE), three-dimensional (3D) steady-state spoiled gradient-echo (GRE) (FLASH [fast low-angle shot]), and 3D magnetization-prepared rapid gradient echo (MP-RAGE) pulse sequences. The 3D GRE sequences provided greater contrast enhancement at relatively high [GD], and the 2D SE sequence demonstrated greater enhancement and a higher rate of enhancement at low [GD]. The results predict that at low [GD], certain lesions could probably be detected with the 2D SE sequence but possibly not with one or both of the 3D GRE sequences. At high [GD], certain lesions could probably be detected with one or both of the 3D GRE sequences but possibly not with the 2D SE sequence. This provides a potential explanation for the clinical observation that certain contrast agent enhanced lesions appear less conspicuous on 3D GRE images than on 2D SE images and vice versa. Modified parameter values were derived for the 3D FLASH and 3D MP-RAGE sequences that are predicted to produce contrast enhancement behavior equivalent or superior to that of a conventional 2D SE sequence.     

3D T2-weighted fast spin-echo MRI sialography of the parotid gland

The diagnostic value of 3D T2-weighted MRI sialography and 2D T2-weighted fast spin-echo (FSE) images for delineation of the normal duct system and characterisation of parotid gland duct pathology was compared in a prospective study. We studied eight healthy volunteers and 18 patients with pathology of the parotid gland (tumours in 3, sialolithiasis in 6, Sjögren's disease in 4, recurrent or chronic parotitis in 4, post-traumatic stricture of the main parotid duct in 1). A heavily T2-weighted 3D FSE sequence was compared with a conventional 2D T2-weighted FSE sequence. The normal main parotid duct was always visible on 3D sialography and seen in 68 % of the 2D T2-weighted FSE studies. The diagnostic reliability of both sequences for diagnosis of luminal concretions in sialolithiasis and dilatation of the duct in duct stricture or chronic parotitis was equal, although slight intraglandular dilatation was appreciated only on 3D sialography. Extraductal pathology resulting in obstruction or displacement of ducts was better characterised on 2D T2-weighted images. However, 3D MRI sialography offered the advantage of postprocessing with overview images and multiple maximum-intensity projection images in any plane.     

T2-weighted MR imaging of focal hepatic lesions: Comparison of various RARE and fat-suppressed spin-echo sequences

The authors prospectively compared four T2-weighted magnetic resonance (MR) sequences, including high-resolution 512 × 512 (matrix size) RARE (rapid acquisition with relaxation enhancement), 256 × 256 RARE, 128 × 256 breath-hold RARE, and 192 × 256 fat-suppressed spin-echo (T2FS) sequences, in the evaluation of 16 patients with focal hepatic masses. MR images were evaluated by quantitative lesion-liver signal difference-to-noise ratios (SDNRs) and subjective evaluation of image artifact and image quality. No significant differences were observed between RARE sequences in SDNR values. The T2FS sequence had a significantly higher SDNR than the 512 × 512 RARE sequence (24.6 ± 15.0 vs 14.5 ± 9.7) (P =.008). Image quality was rated highest for the 512 × 512 RARE and T2FS sequences (P =.006). The inherent advantage of high spatial resolution suggests that the 512 × 512 RARE sequence may be of value in detecting hepatic lesions.     

MRI in characterization of focal liver lesions: Comparison of T2 weighting by conventional spin-echo and turbo spin-echo sequences

Forty-one patients with 61 proved focal liver lesions underwent MRI of the liver at 1.0 T, with the aim of evaluating the usefulness of turbo spin-echo (TSE) sequences in characterizing focal liver lesions, by comparing them with conventional spin-echo (CSE) sequences. Two different TSE protocols were employed, with constant echo time and varying repetition time: TSE-S (3000 msec) and TSE-L (5100 msec). All images were evaluated quantitatively (signal-to-noise ratio ‘SNR’) and qualitatively: because benign lesions were all liquid (12 cysts and 10 hemangiomas), they were well characterized morphologically on the basis of signal intensity. Mean SNR was significantly different between metastases and benign lesions (P < .0001) with all T2 sequences. Among the single T2 sequences tested, logistic regression analysis showed TSE-L to have the best predictive ability of the nature of focal lesions, with a G value of 42.02, compared to 29.87 of TSE-S and 25.55 of CSE second echo (SE II). The combination of TSE-L with TSE-S did not modify these results, whereas the combination of TSE-L with CSE only resulted in slight improvement (G = 46.95). Comparison of the receiver operating characteristic (ROC) curves showed only SE II (area under the ROC curve of .8312) to be significantly inferior to the best single sequence, or TSE-L (area under the ROC curve of .9176; P = .027). All sequences were equivalent in qualitative evaluation, with good reproducibility, sensitivity ranging from .94 to 1.0, and specificity ranging from .86 to .93. This study confirms the value of TSE sequences in characterization of focal liver lesions. Time of acquisition is strongly reduced with these sequences, whereas results are fairly similar to those obtained with CSE. TSE sequences could therefore replace CSE for the study of focal liver lesions.     

Detection of lesions in multiple sclerosis by 2D FLAIR and single-slab 3D FLAIR sequences at 3.0 T: initial results

The aim of this study was to compare conventional 2D FLAIR and single-slab 3D FLAIR sequences in the detection of lesions in patients with multiple sclerosis. Eight patients with MS were examined at 3.0 T by using a 2D FLAIR sequence and a single-slab 3D FLAIR sequence. A comparison of lesion detectability was performed for the following regions: periventricular, nonperiventricular/juxtacortical and infratentorial. The contrast-to-noise ratios (CNRs) between lesions and brain tissue and CSF were calculated for each sequence. A total of 424 lesions were found using the 2D FLAIR sequence, while with the 3D FLAIR sequence 719 lesions were found. With the 2D FLAIR sequence, 41% fewer lesions were detected than with the 3D FLAIR sequence. Further, 40% fewer supratentorial and 62.5% fewer infratentorial lesions were found with the 2D FLAIR sequence. In images acquired with the 3D FLAIR sequence, the lesions had significantly higher CNRs than in images acquired with the 2D FLAIR sequence. These are the first results using a single-slab 3D FLAIR sequence at 3.0 T for detection of lesions in patients with MS. With the 3D FLAIR sequence significantly higher CNRs were achieved and significantly more lesions in patients with MS were detected.     

Fat-suppressed MRI of musculoskeletal infection: fast T2-weighted techniques versus gadolinium-enhanced T1-weighted images

 Purpose. To investigate gadolinium’s role in imaging musculoskeletal infection by comparing the conspicuity and extent of inflammatory changes demonstrated on gadolinium-enhanced fat-suppressed T1-weighted images versus fat-suppressed fast T2-weighted sequences. Design. Eighteen patients with infection were imaged in a 1.5-T unit, using frequency-selective and/or inversion recovery fat-suppressed fast T2-weighted images (T2WI) and gadolinium-enhanced frequency-selective fat-suppressed T1-weighted images (T1WI). Thirty-four imaging planes with both a fat-suppressed gadolinium-enhanced T1-weighted sequence and a fat-suppressed T2-weighted sequence were obtained. Comparison of the extent and conspicuity of signal intensity changes was made for both bone and soft tissue in each plane. Results. In bone, inflammatory change was equal in extent and conspicuity on fat-suppressed T2WI and fat-suppressed T1WI with gadolinium in 19 planes, more extensive or conspicuous on T2WI in three planes, and less so on T2WI in two planes. Marrow was normal on all three sequences in 10 cases. In soft tissue, inflammatory change was seen equally well in 20 instances, more extensively or conspicuously on the T2WI in 11 instances, and less so on T2WI in 2 instances. One case had no soft tissue involvement on any of the sequences. Five abscesses and three joint effusions were present, all more conspicuously delineated from surrounding inflammatory change on the fat-saturated T1WI with gadolinium. The average imaging time for the fat-saturated T1WI with gadolinium was 6.75 min, while that of the T2-weighted sequences was 5.75 min. Conclusion. Routine use of gadolinium is not warranted. Instead, gadolinium should be reserved for clinically suspected infection in or around a joint, and in cases refractory to medical or surgical treatment due to possible abscess formation.