Research ArticleClinical Investigations

PET/MR Imaging in the Detection and Characterization of Pulmonary Lesions: Technical and Diagnostic Evaluation in Comparison to PET/CT

Isabel Rauscher, Matthias Eiber, Sebastian Fürst, Michael Souvatzoglou, Stephan G. Nekolla, Sibylle I. Ziegler, Ernst J. Rummeny, Markus Schwaiger and Ambros J. Beer
Journal of Nuclear Medicine May 2014, 55 (5) 724-729; DOI: https://doi.org/10.2967/jnumed.113.129247

Article Figures & Data

Figures

  • FIGURE 1.
    FIGURE 1.

    Correlation analysis of tracer uptake between PET/CT and subsequent PET/MR imaging as assessed by SUVmean in lung lesions (A) and normal lung parenchyma (B). x-axis displays quantitative values as obtained by PET/CT, and y-axis displays corresponding values by PET/MR imaging. High correlations as expressed by Spearman correlation coefficient (R) are found for SUVmean (0.9 for normal lung parenchyma, 0.57 for lung lesions) between findings from both modalities. Difference between 2 SUV measurements is shown by Bland–Altman (for SUVmean in normal lung parenchyma [C] and for SUVmean in lung lesions [D]) on which difference between 2 SUV measurements is plotted against their average. For SUVmean-healthy lung (mean SUV in healthy lung), mean difference is 1.3 SUV; 95% CI is +6.4 and −3.9 SUV; for SUVmean_lung lesion (mean SUV of lung lesion), mean difference is −0.12 SUV; 95% CI for SUVmax is +0.14 and −0.37 SUV.

  • FIGURE 2.
    FIGURE 2.

    A 26-y-old patient with testicular cancer. (A, C, and D) Morphologic datasets (axial CT [A], axial VIBE [C], and axial Dixon sequence [D] [in phase]). (B) Corresponding PET of PET/CT. Lung lesion (arrow) was seen in CT and VIBE scan but not in Dixon sequence. Additionally, there was no suspected 18F-FDG uptake.

Tables

  • TABLE 1

    Technical Parameters of Different MR Imaging Sequences Used in This Study

    SequenceT1-weighted VIBE DixonT1-weighted VIBE fat saturated
    Repetition time/echo time (ms)*3.61/1.23 and 2.463.29/1.16
    Slice thickness (mm)3.15
    Gap (%)00
    Matrix79 × 192320 × 240
    Field of view (mm)500460
    Bandwidth (kHz)960540
    Voxel size (mm3)4.1 × 2.6 × 3.11.8 × 1.4 × 5
    Flip angle (°)109
    % phase field of view65.575.0
    Acquisition time (min:sec)0:190:16
    No. of excitations11
    Integrated parallel acquisition technique factor22

    • ↵* Attenuation-correction technique with T1 VIBE Dixon requires 2 echo times.

  • TABLE 2

    Number and Size of All Lung Lesions and PET-Positive and PET-Negative Lesions

    No. of lung lesionsSize of lung lesions (mm ± SD)
    GroupCT (standard of reference)DixonVIBEPET MR imagingPET CTCT (standard of reference)DixonVIBEPET MR imagingPET CT
    All471532222210.0 ± 11.4 (2–60)12.2 ± 5.0 (4–21)11.9 ± 11.7 (2–59)13.5 ± 10.2 (4–49)13.5 ± 10.2 (4–49)
    <1 cm diameter339151010
    >1 cm diameter146171212

    • Data in parentheses are ranges.

  • TABLE 3

    Characterization of Lung Lesions in CT, VIBE, and Dixon Sequences

    Modality1—most likely unspecific/benign2—indeterminate3—suggestive of malignancy
    CT (n = 47)21224
    VIBE (n = 32)9716
    Dixon (n = 15)249
  • TABLE 4

    Image Quality of CT, MR Imaging, and PET

    DatasetSequenceMean ± SD
    MorphologicVIBE2.60 ± 0.63†§
    CT3.00 ± 0†‡
    Dixon2.70 ± 0.46‡§
    PETPET AC-CT2.80 ± 0.41‖
    PET AC-MR imaging2.83 ± 0.38‖
    PET NAC-CT2.98 ± 0.16
    PET NAC-MR imaging2.90 ± 0.30

    • ↵† P = 0.0003*.

    • ↵‡ P = 0.0002*.

    • ↵§ P = 0.33.

    • ↵‖ P = 0.74.

    • ↵ P = 0.18.

    • ↵* Indicates a significant difference between two sequences.

    • AC = attenuation correction; NAC = non–attenuation correction.

Additional Files

Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Bookmark this article

Jump to section

Keywords