Endometrial cancer: preoperative staging using three-dimensional T2-weighted turbo spin-echo and diffusion-weighted MR imaging at 3.0 T: a prospective comparative study

Objectives

To prospectively assess the efficacy of 3-T magnetic resonance (MR) imaging using the three-dimensional turbo spin-echo T2-weighted and diffusion-weighted technique (3D-TSE/DW) compared with that of conventional imaging using the two-dimensional turbo spin-echo T2-weighted and dynamic contrast-enhanced technique (2D-TSE/DCE) for the preoperative staging of endometrial cancer, with pathological analysis as the reference standard.

Methods

Seventy-one women with endometrial cancer underwent MR imaging using 3D-TSE/DW (b?=?1,000 s/mm²) and 2D-TSE/DCE. Two radiologists independently assessed the two imaging sets. Accuracy, sensitivity, and specificity for staging were analysed with the McNemar test; the areas under the receiver operating characteristic curve (Az) were compared with a univariate z-score test.

Results

The results for assessing deep myometrial invasion, accuracy, sensitivity, specificity and Az, respectively, were as follows: 3D-TSE/DW—observer 1, 87 %, 95 %, 85 % and 0.96; observer 2, 92 %, 84 %, 94 % and 0.95; 2D-TSE/DCE—observer 1, 80 %, 79 %, 81 % and 0.89; observer 2, 86 %, 84 %, 87 % and 0.86. Most of the values were higher with 3D-TSE/DW without significant differences (P?>?0.12). For assessing cervical stromal invasion, there were no significant differences in those values for both observers (P?>?0.6).

Conclusions

Accuracy of 3D-TSE/DW was at least equivalent to that of the conventional technique for the preoperative assessment of endometrial cancer.

Key Points

? New techniques in MR imaging help assess patients with endometrial cancer. ? A 3D T2-weighted TSE sequence seems equally as accurate as conventional techniques. ? Three-dimensional TSE/DW imaging does not require intravenous contrast material and is relatively quick. ? Tumour extent of endometrial cancer can be clearly shown on diffusion-weighted images. ? Junctional zone can be visualised well on 3D-TSE T2-weighted images.     

Uterine cervical carcinoma: a comparison of two- and three-dimensional T2-weighted turbo spin-echo MR imaging at 3.0 T for image quality and local-regional staging

Objective

To compare three-dimensional (3D) T2-weighted turbo spin-echo (TSE) with multiplanar two-dimensional (2D) T2-weighted TSE for the evaluation of invasive cervical carcinoma.

Methods

Seventy-five patients with cervical carcinoma underwent MRI of the pelvis at 3.0 T, using both 5-mm-thick multiplanar 2D (total acquisition time?=?12 min 25 s) and 1-mm-thick coronal 3D T2-weighted TSE sequences (7 min 20 s). Quantitative analysis of signal-to-noise ratio (SNR) and qualitative analysis of image quality were performed. Local-regional staging was performed in 45 patients who underwent radical hysterectomy.

Results

The estimated SNR of cervical carcinoma and the relative tumour contrast were significantly higher on 3D imaging (P?<?0.0001). Tumour conspicuity was better with the 3D sequence, but the sharpness of tumour margin was better with the 2D sequence. No significant difference in overall image quality was noted between the two sequences (P?=?0.38). There were no significant differences in terms of the diagnostic accuracy, sensitivity, and specificity of parametrial invasion, vaginal invasion, and lymph node metastases.

Conclusion

Multiplanar reconstruction 3D T2-weighted imaging is largely equivalent to 2D T2-weighted imaging for overall image quality and staging accuracy of cervical carcinoma with a shorter MR data acquisition, but has limitations with regard to the sharpness of the tumour margin.

Key Points

? 3D T2-weighted MR sequence is equivalent to 2D for cervical carcinoma staging. ? Coronal 3D acquisitions can reduce the examination time. ? SNR and relative tumour conspicuity were significantly higher on 3D sequences. ? Reformatted 3D T2-weighted imaging had limitations in sharpness of tumour margin.     

High-resolution myocardial perfusion imaging at 3 T: comparison to 1.5 T in healthy volunteers

The purpose of this study was to evaluate high-resolution (HR) myocardial first-pass perfusion in healthy volunteers at 3 T compared to a typical clinical imaging protocol at 1.5 T, with respect to overall image quality and the presence of subendocardial dark rim artifacts. Myocardial first-pass rest perfusion studies were performed at both field strengths using a T1-weighted saturation-recovery segmented k-space gradient-echo sequence combined with parallel imaging (Gd-DTPA 0.05 mmol/kg). Twenty-six healthy volunteers underwent (1) a HR perfusion scan at 3 T(pixel size 3.78 mm²) and (2) a standard perfusion approach at 1.5 T(pixel size 9.86 mm²). The contrast enhancement ratio (CER) and overall image quality (4-point grading scale: 4: excellent; 1: non-diagnostic) were assessed, and a semiquantitative analysis of dark rim artifacts was performed for all studies. CER was slightly higher (1.31 ± 0.32 vs. 1.14 ± 0.34; p<0.01), overall image quality was significantly improved (3.03 ± 0.43 vs. 2.37 ± 0.39; p<0.01), and the number of dark rim artifacts (139 ± 2.09 vs. 243 ± 2.33; p<0.01) was significantly reduced for HR perfusion imaging at 3 T compared to the standard approach at 1.5 T. HR myocardial rest perfusion at 3 T is superior to the typical clinical perfusion protocol performed at 1.5 T with respect to the overall image quality and presence of subendocardial dark rim artifacts.     

Breast MRI at 7 Tesla with a bilateral coil and T1-weighted acquisition with robust fat suppression: image evaluation and comparison with 3 Tesla

Objectives

To evaluate the image quality of T1-weighted fat-suppressed breast MRI at 7 T and to compare 7-T and 3-T images.

Methods

Seventeen subjects were imaged using a 7-T bilateral transmit-receive coil and 3D gradient echo sequence with adiabatic inversion-based fat suppression (FS). Images were graded on a five-point scale and quantitatively assessed through signal-to-noise ratio (SNR), fibroglandular/fat contrast and signal uniformity measurements.

Results

Image scores at 7 and 3 T were similar on standard-resolution images (1.1?×?1.1?×?1.1-1.6 mm³), indicating that high-quality breast imaging with clinical parameters can be performed at 7 T. The 7-T SNR advantage was underscored on 0.6-mm isotropic images, where image quality was significantly greater than at 3 T (4.2 versus 3.1, P?≤?0.0001). Fibroglandular/fat contrast was more than two times higher at 7 T than at 3 T, owing to effective adiabatic inversion-based FS and the inherent 7-T signal advantage. Signal uniformity was comparable at 7 and 3 T (P?<?0.05). Similar 7-T image quality was observed in all subjects, indicating robustness against anatomical variation.

Conclusion

The 7-T bilateral transmit-receive coil and adiabatic inversion-based FS technique produce image quality that is as good as or better than at 3 T.

Key Points

? High image quality bilateral breast MRI is achievable with clinical parameters at 7 T. ? 7-T high-resolution imaging improves delineation of subtle soft tissue structures. ? Adiabatic-based fat suppression provides excellent fibroglandular/fat contrast at 7 T. ? 7- and 3-T 3D T1-weighted gradient-echo images have similar signal uniformity. ? The 7-T dual solenoid coil enables bilateral imaging without compromising uniformity.     

Bilateral ce-MR angiography of the hands at 3.0 T and 1.5 T: intraindividual comparison of quantitative and qualitative image parameters in healthy volunteers

The purpose of this study was to determine the benefit of bilateral contrast-enhanced MR angiography (ce-MRA) of the hands at 3.0 Tesla (T) compared with an established 1.5-T technique in healthy volunteers. Intraindividual bilateral ce-MRA of the hands was performed at 1.5 T and 3.0 T in 14 healthy volunteers using a timed ultra-fast GRE sequence featuring parallel acquisition. The evaluation comprised measurement of the vessel signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR), rating of the image quality and the assessment of artefacts and venous contamination. At 3.0 T, SNR improved up to 95% and CNR up to 129%. The image quality of the larger inflow arteries, the palm arches and common digital arteries was good or sufficient at either magnetic field strengths. However, 3.0-T MRA was clearly superior in the depiction of the digital arteries. Ce-MRA of the hand clearly profits from the use of 3.0 T. Compared with 1.5 T, a substantial increase of CNR is found resulting in a significantly better delineation of the small digital arteries. Saturation affects more the SNR of the perivascular tissue than the contrast-enhanced blood, and thus leads to a marked increase of CNR at 3.0.     

Intraindividual comparison of myocardial delayed enhancement MR imaging using gadobenate dimeglumine at 1.5 T and 3 T

For contrast-enhanced imaging techniques relying on strong T1 weighting, 3 T provides increased contrast compared with 1.5 T. The aim of our study was the intraindividual comparison of delayed enhancement MR imaging at 1.5 T and at 3 T. Twenty patients with myocardial infarction were examined at 1.5 T and 3 T. Fifteen minutes after injection of contrast agent (0.1 mmol gadobenate dimeglumine per kg body weight), inversion recovery gradient recalled echo (IR-GRE) sequences were acquired (1.5 T/3 T: TR 11.0/9.9 ms, TE 4.4/4.9 ms, flip 30°/30°, slice thickness 6/6 mm) to assess myocardial viability. Two observers rated image quality (Wilcoxon signed rank test). Quantification of hyperenhanced myocardium and standardized SNR/CNR measurements were performed (Student’s t test). There was no significant difference with respect to image quality (1.5 T/3 T: 3.5/3.3, p = 0.34, reader 1; 2.4/2.7, p = 0.12, reader 2) and infarction size (760 ± 566/828 ± 677 mm² at 1.5 T, 808 ± 639/826 ± 726 mm² at 3 T, reader 1/reader 2, p > 0.05). Mean SNR in hyperenhanced/normal myocardium was 19.2/6.2 at 1.5 T and 29.5/8.8 at 3 T (p < 0.05). Mean CNR was 14.3 at 1.5 T and 26.0 at 3 T (p < 0.05). Delayed enhancement MR imaging at 3 T is a robust procedure yielding superior tissue contrast at 3 T compared with 1.5 T which is, however, not reflected by increased image quality.     

TTC post-processing is beneficial for functional MRI at low magnetic field: a comparative study at 1 T and 3 T

This study aimed to broaden the diagnostic possibilities of low-field MRI systems (i) by examining the feasibility of functional MRI of human brain activation at 1 T, and (ii) by assessing its reliability in comparison with acquisitions at 3 T. Eight subjects were studied at 1 T and 3T using standard echo-planar-imaging sequences at 3-mm isotropic spatial resolution. Paradigms included silent word generation, sequential finger-to-thumb opposition, and passive finger movements. Image post-processing was carried out either with statistical parametric mapping (SPM5, single-subject and group analysis) or with a two-threshold correlation (TTC, single-subject analysis only) analysis. Single-subject analysis with SPM5 resulted in 3–5 times more activated pixels at 3 T than at 1 T in the examined Broca and sensorimotor regions. By comparison, the TTC single-subject analysis yielded the same amount of activated pixels at 3 T and 1 T. Moreover, this number was identical to that obtained with SPM at 3 T. The group analysis with SPM5 resulted in very similar numbers of activated pixels at both field strengths. The present findings suggest that a field strength of 1 T combined with adequate post-processing allows for reliable functional MRI studies of human brain activation. High-field advantages are therefore best invested in higher spatial resolution.     

Anatomical parcellation of the brainstem and cerebellar white matter: a preliminary probabilistic tractography study at 3 T

Introduction The aims of this study were: (1) to test whether higher spatial resolution diffusion tensor images and a higher field strength (3 T) enable a more accurate delineation of the anatomical tract within the brainstem, and, in particular, (2) to try to distinguish the different components of the corticopontocerebellar paths in terms of their cortical origins. Methods The main tracts of the brainstem of four volunteers were studied at 3 T using a probabilistic diffusion tensor imaging (DTI) axonal tracking. The resulting tractograms enabled anatomical well-delineated structures to be identified on the diffusion tensor coloured images. Results We tracked corticopontine, corticospinal, central tegmental, inferior and superior cerebellopeduncular, transverse, medial lemniscal and, possibly, longitudinal medial fibres. Moreover, DTI tracking allowed a broad delineation of the corticopontocerebellar paths. Conclusion Diffusion tensor coloured images allow a rapid and reliable access to the white matter broad parcellation of the brainstem and of the cerebellum, which can be completed by fibre tracking. However, a more accurate and exhaustive depiction of the anatomical connectivity within the brainstem requires the application of more sophisticated techniques and tractography algorithms, such as diffusion spectrum imaging.     

Comparative study of image quality between axial T2-weighted BLADE and turbo spin-echo MRI of the upper abdomen on 3.0 T

Objective

To compare image quality of turbo spin-echo (TSE) with BLADE [which is also named periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER)] on magnetic resonance imaging (MRI) for upper abdomen.

Materials and methods

This study involved the retrospective evaluation of 103 patients (63 males, 40 females; age range 19–76 years; median age 53.8 years) who underwent 3.0 T MRI with both conventional TSE T2-weighted imaging (T2WI) and BLADE TSE T2WI. Two radiologists assessed respiratory motion, gastrointestinal peristalsis, and vascular pulsation artifacts, as well as the sharpness of the liver and pancreas edges. Scores for all magnetic resonance (MR) images were recorded. Wilcoxon’s rank test was used to compare hierarchical data. Cohen’s kappa coefficient was adopted to analyze interobserver consistency.

Results

Compared to TSE T2WI, BLADE TSE T2WI reduced all of the examined motion artifacts and increased the sharpness of the liver and pancreas edges (all P < 0.05). Medium to good interobserver consistency was obtained for evaluating these indicators. The scanning time of BLADE TSE T2WI was 4–16 s shorter than that of conventional TSE T2WI.

Conclusion

Compared to TSE sequence, the BLADE technique can reduce the respiratory motion, gastrointestinal peristalsis, and vascular pulsation artifacts, while decreasing the scanning time and improving the anatomic detail and image quality.

     

Influence of age and sex on signal intensities of the posterior lobe of the pituitary gland on T1-weighted images from 3 T MRI

Purpose

To elucidate the influence of age and sex on the signal intensity (SI) of the posterior lobe of the pituitary gland (PPG) on T1-weighted images (T1WI) from 3 T MRI.

Materials and methods

Sagittal T1WI acquired from three-dimensional fast spoiled gradient recalled acquisition in the steady state in 1,634 subjects without conditions affecting antidiuretic hormone were evaluated retrospectively. The presence or absence of a bright signal in the PPG was assessed qualitatively. The SI ratio of the PPG to the pons (SIR) was obtained from quantitative measurements. We statistically analyzed these data, creating 14 subject groups categorized according to age and sex, and applied a Poisson generalized linear model to the SIR data.

Results

The characteristic bright signal was absent in 47 subjects (2.8 %), with no significant difference in incidence among the groups. The SIR was inversely related to age in both males (r > 0.7) and females (r > 0.9), and was significantly higher in females in the third to the eighth decades (p < 0.05). Analysis of the whole SIR dataset using a generalized linear model showed that the estimated SIR decreased by 1.7 % per decade and is higher in females.

Conclusion

Age and sex influence the SI of the PPG on T1WI. These findings may aid the recognition of PPG signal abnormalities on T1WI.     

In vitro comparison of different carotid artery stents: a pixel-by-pixel analysis using CT angiography and contrast-enhanced MR angiography at 1.5 and 3 T

Introduction  

CT angiography (CTA) and MR angiography (MRA) are increasingly used methods for evaluation of stented vessel segments. The purpose of this study was to compare CTA, contrast-enhanced MRA (CEMRA) at 1.5 T, and CEMRA at 3 T for the visualization of carotid artery stents and to define the best noninvasive imaging technique for each stent.     

(1)H MR spectroscopy with external reference solution at 1.5 T for differentiating malignant and benign breast lesions: comparison using qualitative and quantitative approaches

Objective

To compare the diagnostic capability of proton (¹H) magnetic resonance spectroscopy (MRS) in differentiating benign from malignant breast lesions on the basis of qualitative and quantitative approaches.

Methods

We performed single-voxel ¹H MRS for 208 breast lesions, identified a clear total composite choline compounds (tCho) peak of signal-to-noise of ??2 to represent malignancy (qualitative approach), and regarded tCho concentration equal to or greater than the cut-off value to represent malignancy (quantitative approach). We compared the diagnostic ability of both approaches using the Akaike information criterion (AIC) and McFadden??s R ².

Results

Histologically, 169 lesions were malignant; 39 were benign. The qualitative approach demonstrated 84.6?% sensitivity and 51.3?% specificity for differentiating malignant and benign lesions. The mean tCho concentration was 1.13?mmol/kg for malignancy, 0.43?mmol/kg for benignity. The optimal cut-off point was 0.61?mmol/kg, use of which achieved 68.1?% sensitivity and 79.4?% specificity. Calculated AIC and R ² score suggested the superiority of the quantitative approach for differentiating malignancy.

Conclusions

Quantitative MRS provides higher specificity than qualitative MRS for differentiating malignant from benign lesions and could be more useful as an additional examination in routine breast MR imaging.

Key Points

? Magnetic resonance spectroscopy of the breast helps distinguish malignant from benign lesions. ? Magnetic resonance spectra demonstrate a choline signal even from benign lesions. ? Choline concentration is higher in breast carcinomas than in benign lesions. ? Quantitative magnetic resonance spectroscopy differentiates breast malignancies better than qualitative MRS.     

High resolution myocardial magnetic resonance stress perfusion imaging at 3 T using a 1 M contrast agent

Objective

Stress perfusion magnetic resonance imaging (MSPMRI) is an established technique for the assessment of myocardial perfusion. Shortcomings at 1.5 T are low signal to noise ratio (SNR) and contrast to noise ratio (CNR). One approach to overcome these shortcomings is to increase field strength and contrast concentration. The aim of our study was to investigate the diagnostic capability of high resolution MSPMRI at 3-T field strength using a 1 M contrast agent.

Material and methods

Fifty-seven patients (62.3?±?11.0 years) with symptoms of coronary artery disease (CAD) were examined at 3 T. MMRSPI was assessed using a 2D saturation recovery gradient echo (SR GRE) sequence in short axis orientation (TR 1.9 ms, TE 1.0 ms, flip 12°, 0.1 mmol gadobutrol/kg body weight (bw), 140 µg adenosine/kg bw/min). Perfusion images were assessed visually and semiquantitatively (upslope, peak signal intensity (SI), and myocardial perfusion reserve index (MPRI)). Standard of reference was invasive coronary angiography.

Results

Stress-induced hypoperfusion was found in 43 patients. Sensitivity for hemodynamically relevant CAD (stenoses greater than 70%) was 95%/98%, specificity 80%/87%, diagnostic accuracy 91%/95% (reader 1/reader 2). The MPRI was significantly lower in hypoperfused myocardium (1.3?±?0.2) compared with normal myocardium (2.6?±?0.7).

Conclusion

High resolution MMRSPI at 3 T using 1 M contrast agent under daily routine conditions provides reliable detection of stress-induced myocardial hypoperfusion with higher diagnostic accuracy than 1.5-T conditions.     

Comparative study of imaging at 3.0 T versus 1.5 T of the knee

Purpose  The objectives of the study were to compare MR imaging at 1.5 and 3.0 T in the same patients concerning image quality and visualization of cartilage pathology and to assess diagnostic performance using arthroscopy as a standard of reference. Materials and methods  Twenty-six patients were identified retrospectively as having comparative 1.5 and 3.0 T MR studies of the knee within an average of 102 days. Standard protocols included T1-weighted and fat-saturated intermediate-weighted fast spin-echo sequences in three planes; sequence parameters had been adjusted to account for differences in relaxation at 3.0 T. Arthroscopy was performed in 19 patients. Four radiologists reviewed each study independently, scored image quality, and analyzed pathological findings. Sensitivities, specificities, and accuracies in diagnosing cartilage lesions were calculated in the 19 patients with arthroscopy, and differences between 1.5 and 3.0 T exams were compared using paired Student’s t tests with a significance threshold of p < 0.05. Results  Each radiologist scored the 3.0 T studies higher than those obtained at 1.5 T in visualizing anatomical structures and abnormalities (p < 0.05). Using arthroscopy as a standard of reference, diagnosis of cartilage abnormalities was improved at 3.0 T with higher sensitivity (75.7% versus 70.6%), accuracy (88.2% versus 86.4%), and correct grading of cartilage lesions (51.3% versus 42.9%). Diagnostic confidence scores were higher at 3.0 than 1.5 T (p < 0.05) and signal-to-noise ratio at 3.0 T was approximately twofold higher than at 1.5 T. Conclusion  MRI at 3.0 T improved visualization of anatomical structures and improved diagnostic confidence compared to 1.5 T. This resulted in significantly better sensitivity and grading of cartilage lesions at the knee.     

Comparison of T2-weighted and contrast-enhanced T1-weighted MR imaging at 1.5?T for assessing the local extent of cervical carcinoma

Objective  

To compare two MR sequences at 1.5 T—T2-weighted and contrast-enhanced T1-weighted images—by using macroscopic sections to determine which image type enables the most accurate assessment of cervical carcinoma.     

Sensitivity of T2-weighted FSE sequences towards physiological iron depositions in normal brains at 1.5 and 3.0 T

To evaluate the sensitivity of T2-weighted fast spin-echo (FSE) sequences to physiological iron depositions in normal brains at MR imaging field strengths of 1.5 and 3.0 T. T2-weighted FSE sequences acquired at 1.5 and 3.0 T clinical imaging systems (Gyroscan Intera, Philips Medical Systems, Best, The Netherlands) were compared by means of MRI in phantoms (n=6) and healthy volunteers (n=10). Contrast-to-noise ratios (CNRs) of tubes doped with iron oxides at different concentrations and of brain areas with physiological iron depositions (nucleus ruber, substantia nigra, globus pallidus) were calculated for either field strength. Apparent susceptibility effects of iron-containing brain structures were qualitatively analyzed by comparing the degree of visible hypointensity by a score system at either field strength. The mean CNR of iron oxide tubes and iron-containing brain areas was significantly decreased at 3.0 T. Qualitative analysis confirmed these measurements. Detection and diagnosis of brain disorders with altered iron content such as neurodegenerative parkinsonian disorders (NPD) or intracerebral hemorrhage should benefit from the increased sensitivity of T2-weighted FSE sequences to susceptibility effects at 3.0 T.     

Fat suppression techniques for breast MRI: Dixon versus spectral fat saturation for 3D T1-weighted at 3 T

Objective

To compare two fat suppression techniques used for 3D T1-weighted sequence in breast MRI (magnetic resonance imaging), namely Dixon versus spectral fat saturation (fat sat).

Materials and methods

All breast MRI examinations performed in a Philips 3 T unit between March 2013 and October 2015 including either a Dixon or a fat sat sequence were retrospectively analyzed. The examinations were subjectively evaluated by two independent experienced readers in a scale of 5 for overall quality of fat suppression, homogeneity of fat suppression, definition of anatomic structures and focal lesions, diagnostic confidence for axillary and internal mammary regions and the presence of artifacts, 1 corresponding to excellent and 5 to non-diagnostic quality. Contrast-to-noise-ratio (CNR) measurements for muscle and focal lesions were also performed.

Results

Overall 161 women (mean age 51.6 ± 12.0 years) underwent 189 MR examinations, 113 with the fat saturation and 76 with the Dixon sequence. Interobserver variability was good (kappa = 0.757). In all subjectively evaluated parameters, the Dixon sequence was superior to the fat sat (p < 0.05). Mean values of CNR for muscle and focal lesions were 9.98 (±4.2), 17.9 (±7.53) for the fat sat and 18.3 (±10.4) and 29.3 (±14.1) for the Dixon sequence, respectively (p < 0.001).

Conclusion

3D T1 Dixon sequence is superior to fat sat for dedicated breast MRI at 3 T, in terms of efficiency of fat suppression and image quality with the added advantage of optimal exploration of the axillary areas.