Lenore I. Everson

In vivo quantification of choline compounds in the breast with 1H MR spectroscopy.

This work describes a methodology for quantifying levels of total choline‐containing compounds (tCho) in the breast using in vivo 1H MR spectroscopy (MRS) at high field (4 Tesla). Water is used as an internal reference compound to account for the partial volume of adipose tissue. Peak amplitudes are estimated by fitting one peak at a time over a narrow frequency band to allow measurement of small metabolite resonances in spectra with large lipid peaks. This quantitative method significantly improves previously reported analysis methods by accounting for the variable sensitivity of breast 1H MRS measurements. Using this technique, we detected and quantified a tCho peak in 214 of 500 in vivo spectra. tCho levels were found to be significantly higher in malignancies than in benign abnormalities and normal breast tissues, which suggests that this technique could be used to diagnose suspicious lesions and monitor response to cancer treatments. Magn Reson Med 50:1134–1143, 2003.

Eliminating spurious lipid sidebands in 1H MRS of breast lesions.

Detecting metabolites in breast lesions by in vivo 1H MR spectroscopy can be difficult due to the abundance of mobile lipids in the breast which can produce spurious sidebands that interfere with the metabolite signals. Two‐dimensional J‐resolved spectroscopy has been demonstrated in the brain as a means to eliminate these artifacts from a large water signal; coherent sidebands are resolved at their natural frequencies, leaving the noncoupled metabolite resonances in the zero‐frequency trace of the 2D spectrum. This work demonstrates that using the zero‐frequency trace—or equivalently the average of spectra acquired with different echo times—can be used to separate noncoupled metabolite signals from the lipid‐induced sidebands. This technique is demonstrated with simulations, phantom studies, and in several breast lesions. Compared to the conventional approach using a single echo time, echo time averaging provides increased sensitivity for the study of small and irregularly shaped lesions. Magn Reson Med 48:215–222, 2002.

Bursitis in association with solitary osteochondromas presenting as mass lesions.

Three patients with solitary osteochondromas which were increasing in size have been recently examined. Plain films were available on all patients; two patients had MR studies, and two had CT scans. In all three cases, malignant transformation of the osteochondroma was suspected from the cross-sectional imaging studies, but pathologic examination proved that these patients all had bursa formation without any evidence of malignancy. The incidence of this rare complication of solitary osteochondroma is discussed. Ultrasound is also recommended for the evaluation of enlarging solitary osteochondromas.

Hyperextension strain or “whiplash” injuries to the cervical spine

PurposeTo define “whiplash” radiologically. Material and methods. A full cervical spine radiographic series (including flexion and extension views) was reviewed in 40 patients with clinically proven “whiplash” injuries and compared to the radiographs in 105 normal controls. The level and degree of kinking or kyphosis, subluxation, and the difference in the amount of fanning between spinous processes on flexion and extension films were measured in each patient.ResultsLocalized kinking greater than 10° and over 12 mm of fanning, often occurring at the level below the kinking or kyphosis, occurred mainly in the group of whiplash patients (sensitivity 81%, specificity 76%, accuracy 80%).ConclusionsLocalized kinking greater than 10° and fanning greater than 12 mm are useful measurements by which to separate patients with true whiplash injuries from those with minor ligamentous tears. Flexion and extension views are essential to help define whiplash and other ligamentous injuries of the cervical spine.

Feasibility of single-voxel MRS measurement of apparent diffusion coefficient of water in breast tumors.

We report initial results with single voxel spectroscopy (SVS) using diffusion weighting and localization by adiabatic selective refocusing (LASER) in breast tumors to measure the apparent diffusion coefficient of water (ADCw). This is a quick (30 s) and relatively easy method to implement compared with image‐based diffusion measurements, and is insensitive to lipid signal contamination. The ADCw and concentration of total choline containing compounds [tCho] were evaluated for associations with each other and final pathologic diagnosis in 25 subjects. The average (± SD) ADCw in benign and malignant lesions was 1.96 ± 0.47 mm2/s and 1.26 ± 0.29 × 10−3 mm2/s, respectively, P< 0.001. Receiver operating characteristic curve analysis showed an area under the curve of 0.92. Analysis of the single voxel (SV) ADCw and [tCho] showed significant correlation with a R2 of 0.56, P< 0.001. Compared with more commonly used image‐based methods of measuring water ADC, SV‐ADCw is faster, more robust, insensitive to fat, and potentially easier to implement on standard clinical systems. Magn Reson Med, 2009.

High-Spatial- and High-Temporal-Resolution Dynamic Contrast-enhanced MR Breast Imaging with Sweep Imaging with Fourier Transformation: A Pilot Study

PURPOSE To report the results of sweep imaging with Fourier transformation (SWIFT) magnetic resonance (MR) imaging for diagnostic breast imaging. MATERIALS AND METHODS Informed consent was obtained from all participants under one of two institutional review board-approved, HIPAA-compliant protocols. Twelve female patients (age range, 19-54 years; mean age, 41.2 years) and eight normal control subjects (age range, 22-56 years; mean age, 43.2 years) enrolled and completed the study from January 28, 2011, to March 5, 2013. Patients had previous lesions that were Breast Imaging Reporting and Data System 4 and 5 based on mammography and/or ultrasonographic imaging. Contrast-enhanced SWIFT imaging was completed by using a 4-T research MR imaging system. Noncontrast studies were completed in the normal control subjects. One of two sized single-breast SWIFT-compatible transceiver coils was used for nine patients and five controls. Three patients and five control subjects used a SWIFT-compatible dual breast coil. Temporal resolution was 5.9-7.5 seconds. Spatial resolution was 1.00 mm isotropic, with later examinations at 0.67 mm isotropic, and dual breast at 1.00 mm or 0.75 mm isotropic resolution. RESULTS Two nonblinded breast radiologists reported SWIFT image findings of normal breast tissue, benign fibroadenomas (six of six lesions), and malignant lesions (10 of 12 lesions) concordant with other imaging modalities and pathologic reports. Two lesions in two patients were not visualized because of coil field of view. The images yielded by SWIFT showed the presence and extent of known breast lesions. CONCLUSION The SWIFT technique could become an important addition to breast imaging modalities because it provides high spatial resolution at all points during the dynamic contrast-enhanced examination.

What is new in breast MRI spectroscopy.

A review of breast spectroscopy and its metabolite imaging – MRS is useful for clinical evaluation in neoadjuvant chemotherapy for the treatment of breast cancer. Why has it not been implemented on all clinical breast MRI scanners? Doing MRI/MRS on smaller breast cancers is difficult and cannot be done on breast cancers less than 6mm × 6mm × 6mm. High field magnets are used in single voxel spectroscopy (laser) for beast quantitative results (3T or 4T). The MRI/MRS latest results from I-SPY I (NIH/ACRIN) are not yet published for the multi-centered trial. Also having to shim the magnet and preset sequences for optimizing MRS has been an issue and requires a MRI physicist to do these studies. Choline measurements on 1.5T magnets that are shimmed well can detect choline within a large tumor (>2–3 cm). However, it is difficult to quantitate the choline with the breast cancer using 1.5T magnets. The results from the first MRS trial at the University of Minnesota Center for Magnetic Resonance Research will be published at the San Antonio Breast Meeting (December 2012). The report will be on 60 MRI/MRS cases and 5 year survival data. MRI/MRS neoadjuvant breast cancer imaging can be evaluated by three MRI methods to get to the RECIST evaluation: (1) L/D – longest diameter; (2) MRS – metabolite imaging and (3) Volume imaging of contrast (3D). All these methods have difficulty with non-mass like tumors. Making objective measurements of changes in breast tumor size can be challenging. The spectrum of breast tumor morphology and appearance on MRI creates difficulty generating accurate objective measurements which is used by oncology RECIST criteria to track tumor response in breast cancer. Depending on morphology and enhancement of an individual lesion an objective LD comparison may be problematic as seen in the following abstract. Inter-observer agreement in assessment of breast cancer response to neoadjuvant chemotherapy on MRI: Qualitative versus quantitative evaluation (11/28/11 RSNA, Chicago, IL). Volume Imaging: MRI tumor volume for predicting response to neoadjuvant chemotherapy in locally advanced breast cancer: findings from ACRIN 6657/calgb 15007 (2009 Asco) Noxla Hilton. Tumor response measured volumetrically by MRI is a stronger and earlier predictor of pathologic response after (NACT) than clinical exam or tumor diameter. Volumetric MRI imaging must use accurate volume (3D) placement of a voxel under MRI guidance. This procedure is not yet automated and must be completed by an experienced MRI radiologist. If the clinical trial shows that the pathologic response is very good (in 1–2 weeks) then this test may not be as sensitive for predicting (NACT) response. (Chemotherapy drugs work extremely well and the tumor decreases gadolinium uptake to zero.)