Magnetic resonance imaging of the thorax. Past, present, and future

Cardiothoracic MRI in the ICU: A 10-Year Experience

RATIONALE AND OBJECTIVE

The objective of this study was to identify the feasibility and pitfalls of cardiothoracic magnetic resonance imaging (MRI) in intensive care unit (ICU) patients.

MATERIALS AND METHODS

This retrospective study identified adult ICU patients scheduled for cardiothoracic MRIs during a 10-year study period. ICU patients scheduled for brain MRIs served as a comparison group. A chart review was performed to identify factors impacting a patient's ability to undergo an MRI. Differences between completed and canceled examinations for both cardiothoracic and brain MRIs were evaluated. For the cardiothoracic group, clinical indications and the diagnostic value of the study performed were also identified.

RESULTS

A total of 143 cardiothoracic MRIs and 1011 brain MRIs were requested. Cardiothoracic MRI patients were less frequently completed (52% vs 62%), more frequently men (64% vs 43%), younger (55 vs 63 years), less likely mechanically ventilated (8% vs 29%), more likely to require intravenous contrast (83% vs 23%), and had longer examination times compared to brain MRI patients (64 vs 21 minutes). Successful completion of cardiothoracic MRI was associated with lower serum creatinine, higher glomerular filtration rate, and the absence of mechanical ventilation; significant differences were not seen with regard to gender and use of vasoactive agents. Cardiothoracic MRI results were diagnostic in 69% of examinations, most frequently when performed for myocardial disease (84%) and aortic disease (33%), and less frequently for viability (33%).

CONCLUSIONS

In an ICU population, successful completion of cardiothoracic MRI is challenging but feasible in patients with intact renal function and the absence of mechanical ventilation. Examinations were most frequently diagnostic for myocardial and aortic disease indications.

Comparison between high-resolution CT and MRI using a very short echo time in patients with cystic fibrosis with extra focus on mosaic attenuation

BACKGROUND

It would be beneficial to establish pulmonary MRI as a complementary approach to CT for direct visualization of mosaic perfusion, bullae, and emphysema in patients with cystic fibrosis.

OBJECTIVES

The purpose of this study was to compare both modalities, CT and MRI, using the Helbich-Bhalla score with a special focus on reliable detection of a mosaic pattern.

METHODS

Out of 51 patients examined by MRI on a 1.5-Tesla system during a period of 2 years, 19 patients were scheduled for additional low-dose CT in a clinical context. The MRI protocol comprised a gradient echo (GRE) sequence with a very short echo time (TE = 0.8 ms) in inspiration and expiration, a 3-D GRE sequence in breath hold, and a fast spin echo sequence with respiration and ECG triggering. MDCT was carried out in inspiration and adapted to body weight using 100 or 120 kV, 30-60 mA, 1- and 3-mm slice thicknesses, as well as low and high kernels. Additionally incremental slices in 3 positions were recorded in expiration for distinct detection of air trapping. CT and MRI analyses were performed by two radiologic readers in consensus unaware of the clinical parameters. The Helbich-Bhalla score of both examinations was correlated. Mean difference and accordance were assessed in each category.

RESULTS

There was a strong correlation between CT and MRI (R = 0.87, p < 0.01). The mean Helbich-Bhalla score for CT was 12.2 (range 1-18) and for MRI it was 11.7 (range 2-19). The mean difference was 0.5 points. Besides this strong correlation for findings (bronchiectasis, mucus plugging, peribronchial thickening, and consolidation) with a prolonged T2 TE in MRI, we could also state a qualitative agreement of 95-100% in the categories with short T2 and low signal intensity in MRI as emphysema, bullae, and mosaic perfusion.

CONCLUSIONS

These results suggest that in our patient group none of the relevant findings were missed by MR imaging and reading.

Gravity-dependent signal gradients on MR images of the lung in supine and prone positions: a comparison with isogravitational signal variability

PURPOSE

To investigate the tendency of proton MR signal intensity (SI) gradients to be steeper in the supine than in the prone body position, and to quantify the relation between gravity-related and isogravitational changes of SI on proton MR images of the lung.

MATERIALS AND METHODS

In eight healthy volunteers, MR images were obtained in the supine and prone positions using a multiple inversion recovery turbo spin-echo (TSE) sequence. The variation in SI along the gravity-dependent direction and within isogravitational planes was measured on a pixel-by-pixel basis. Ratios of slopes were calculated for comparisons among volunteers. Comparisons of ratios were made using Fisher's exact test. Isogravitational variability was compared with the mean SI, the signal-to-noise ratio (SNR), and the image noise.

RESULTS

The average ratios of slopes showed that the overall SI gradient was steeper in the supine than the prone position, with a substantial difference in the supine/prone ratios between inspiration (1.21) and expiration (1.72). In both the supine and prone positions, gravity-dependent gradients were steeper in expiration than in inspiration (P = 0.001). The SI variability along the gravitational direction was larger than the isogravitational variability. The isogravitational variability in turn was larger than the image noise but smaller than the mean SI of the MR images.

CONCLUSION

Gravity-dependent gradients in proton MR SI are steeper in the supine than in the prone position. The magnitudes of these gradients were larger than the isogravitational signal variability, showing that MRI is sensitive to gravitationally induced effects.

Magnetic resonance imaging of Klebsiella pneumoniae-induced pneumonia in mice

In vivo imaging of small animals is a rapidly developing field. However, the potential of global imaging of infectious processes in animal models remains poorly explored. We used magnetic resonance imaging (MRI) to follow the development and regression of inflammatory lesions caused by infection by Klebsiella pneumoniae in mouse lungs. A virulent strain caused an intense inflammation within 2 days in the whole lungs, while an avirulent strain did not show significant changes. Mice infected with the virulent strain and subsequently treated with antibiotics presented a severe inflammation localized mainly in the left lung that disappeared after a week. The lesions observed by MRI correlated with the damage seen by histological analysis and a 3D representation of the tissue allowed better visualization of the development and healing of inflammatory lesions. MRI thus represents a powerful technique to study in vivo the interactions between a pathogen and its host in real time.

Detection of Pulmonary Nodules Using a 2D HASTE MR Sequence: Comparison with MDCT

OBJECTIVE

The objective of our study was to determine the diagnostic performance of MRI based on a HASTE sequence for the detection of pulmonary nodules in comparison with MDCT.

MATERIALS AND METHODS

Thirty patients with known pulmonary nodules underwent both MRI and CT. CT of the lung served as the standard of reference and was performed on a 4-MDCT scanner using a routine protocol. MRI was performed with axial and coronal HASTE sequences using a high-performance 1.5-T MR scanner. Image data were analyzed in three steps after completion of all data acquisition. Step 1 was the analysis of all the CT image data. Step 2 was the analysis of all the MR image data while blinded to the results of the CT findings. Step 3 closed with a simultaneous review of all corresponding CT and MRI data, including a one-to-one correlation of the size and location of all the nodules that were detected.

RESULTS

Compared with the sensitivity of CT, the sensitivity values for the HASTE MR sequence were as follows: 73% for lesions less than 3 mm, 86.3% for lesions between 3 and 5 mm, 95.7% for lesions between 6 and 10 mm, and 100% for lesions larger than 10 mm. The overall sensitivity of the HASTE sequence for the detection of all pulmonary lesions was 85.4%.

CONCLUSION

An MRI examination that consists of a HASTE sequence allows one to detect, exclude, or monitor pulmonary lesions that are 5 mm and bigger. Suspicious lesions smaller than 5 mm still need to be validated using CT.

Impact of lung volume on MR signal intensity changes of the lung parenchyma

PURPOSE

To test the hypothesis that, in magnetic resonance (MR) imaging of healthy individuals, equal relative changes in lung volume cause equal relative changes in MR signal intensity of the lung parenchyma.

MATERIALS AND METHODS

In two experimental runs, 10 volunteers underwent spirometrically monitored MR imaging of the lungs, with MR images acquired at 10 incremental lung volumes ranging from total lung capacity to 10% above residual volume. Average signal intensity, signal variability, and signal intensity integrals were calculated for each volunteer and for each lung volume. The effect of lung volume on signal intensity was quantified using linear regression analysis complemented by the runs test. Slopes and intercepts of regression lines were compared with an analysis of covariance. Slopes of the lines of best fit for lung volumes and signal intensities from the two runs were compared to the slope of the line of identity. Comparisons between the two runs were visualized using Bland and Altman plots.

RESULTS

The slopes of the 10 individual regression lines yielded no significant differences (F = 1.703, P = 0.101; F = 1.321, P = 0.239). The common slopes were -0.556 +/- 0.027 (P = 0.0001) for the first and -0.597 +/- 0.0031 (P = 0.0001) for the second experimental run. Both slopes displayed no significant nonlinearity (P = 0.419 and P = 0.067). There was a strong association between changes in lung volumes (rs = 0.991, P = 0.0001) and changes in signal intensity (rs = 0.889, P = 0.0001) in the two experimental runs. Lines of best fit for lung volume and signal intensities were not significantly different from the slope of the line of identity (P = 0.321 and P = 0.212, respectively).

CONCLUSION

Equal changes in lung volume cause equal changes in MR signal intensity of the lung parenchyma. This linear and reproducible phenomenon could be helpful in comparing pulmonary MR signal intensity between individuals.

Variable diagnostic approach to suspected pulmonary embolism in the ED of a major academic tertiary care center

The purpose of this study is to retrospectively examine the patterns of diagnostic modality utilization in the setting of suspected acute pulmonary embolism (PE) by physicians in the emergency department (ED) of a major academic tertiary care center. All patients (n=180) with a hospital discharge diagnosis of PE that were admitted through the ED during 1997 to 1999 were included. Overall test frequency and distribution, ancillary testing used when the initial ventilation-perfusion (V/Q) scan or spiral computed tomography (sCT) angiogram was nondiagnostic, concordance of V/Q and sCT, and sensitivities of D-dimer assays, arterial blood gas analysis, and deep venous imaging were assessed. There was a wide discrepancy in practice, most pronounced when first-line radiological studies were nondiagnostic. There was a poor correlation of PE detection between sCT and V/Q. D-dimer assay, arterial blood gas analysis, and deep venous imaging are insensitive tests for PE, and "gold standard" pulmonary angiography is rarely used.

Virtual tools for imaging of the thorax

Helical computed tomography (HCT) allows for volume acquisition of the entire thorax during a single apnoea. Combination of HCT acquisition with synchronous vascular enhancement gives rise to HCT angiography (HCTA). In the last decade, HCT and HCTA have revolutionized the diagnosis of thoracic diseases, modifying many diagnostic algorithms. Because HCT provides for a true volume acquisition free of respiratory misregistration, three-dimensional (3D) rendering techniques can be applied to HCT acquisitions. As these 3D rendering techniques present the HCT information in a different format to the conventional transaxial CT slices, they can be summarized as virtual tools. The purpose of this review is to give the readers the most important technical aspects of virtual tools, to report their application to the thorax, to answer clinical and scientific questions, and to stress their importance for patient management, clinical decision making, and research.