HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH RSS TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Related articles in JNM Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Stegger, L. Articles by Theilmeier, G. Search for Related Content PubMed PubMed Citation Articles by Stegger, L. Articles by Theilmeier, G.

Accurate Noninvasive Measurement of Infarct Size in Mice with High-Resolution PET

¹ Department of Nuclear Medicine, University Hospital Münster, Münster, Germany; ² Interdisciplinary Center for Clinical Research, University Hospital Münster, Münster, Germany; ³ Institute of Anatomy, University Hospital Münster, Münster, Germany; and ⁴ Department of Anesthesiology and Intensive Care Medicine, University Hospital Münster, Münster, Germany

View larger version (44K): [in a new window]   FIGURE 1.  PET of mouse heart at baseline (A) and 7 d after MI (B). For each study, a coronal whole-body slice at the level of LV and bladder with superimposed body contour is shown on left-hand side. Arrow indicates apex of LV (site of infarcted tissue after MI). On right-hand side, short-axis (SA) slices from apex to base, vertical long-axis (VLA) slices from septum to lateral wall, and horizontal long-axis (HLA) slices from anterior to inferior wall show superb "human-like" image quality in mice.

View larger version (51K): [in a new window]   FIGURE 2.  Manual infarct delineation on histologic slices as well as on vertical long-axis PET images. (A and B) Photographs of stained slice of paraffin-embedded LV myocardium without (A) and with (B) superimposed manually drawn midmyocardial contours depicting viable (green) and infarcted (blue) areas. (C) Vertical long-axis PET image with superimposed manually drawn midmyocardial contours depicting viable (green) and infarcted (blue) areas.

View larger version (43K): [in a new window]   FIGURE 3.  Automated contour detection with PET: midmyocardial (solid line), endo- and epicardial (dashed lines) contours superimposed on short-axis (SA; A), VLA (B), and horizontal long-axis (HLA; C) PET images. (D) 3D surface representation of midmyocardial contour of LV; color corresponds to regional 18F-FDG uptake, with infarcted area dark and high 18F-FDG uptake bright. Area of midmyocardial contour is denoted as MCA (MCA3D).

View larger version (11K): [in a new window]   FIGURE 4.  PET-based infarct sizing using manual delineation. Infarct size is defined as the proportion of MCA (MCAslices) of infarcted myocardium, derived by PET and manual delineation, in relation to histomorphometry. (A) Linear regression analysis. (B) Bland–Altman analysis. Solid line in Bland–Altman plot depicts mean difference; dashed lines depict limits of agreement (mean ± 2 SD).

View larger version (16K): [in a new window]   FIGURE 5.  PET-based infarct sizing using automated delineation with varying threshold values. Infarct size is defined as the proportion of MCA of infarcted myocardium, derived by PET and automated delineation, in relation to histomorphometry. Dependency of correlation coefficient (A), slope (B), and y-intercept (C) of linear regression line, as well as mean of wrongly assumed infarct proportion preoperatively (D), on choice of threshold in PET analysis. All data are given for 3D contour calculation (MCA3D, open bars) as well as for stacked slices' calculation adapted to histomorphometry (MCAslices, solid bars).

View larger version (21K): [in a new window]   FIGURE 6.  PET-based infarct sizing using automated delineation. Depicted are linear regression (A) and Bland–Altman analysis (B) of infarct size defined as the proportion of MCA3D of infarcted myocardium as derived by PET and automated segmentation with standard parameters (50% threshold, postoperative (post-OP) scan only, MCA3D) in comparison with histomorphometry. Additionally, nonstandard PET analysis with inclusion of both pre- and post-OP scans for calculation of infarct size (C and D) and nonstandard PET analysis using MCAslices instead of MCA3D (E and F) were compared with PET analysis with standard parameters (50% threshold, postoperative scan only, MCA3D). In both cases, linear regression (C and E) and Bland–Altman analyses (D and F) are shown. Solid line in Bland–Altman plots depicts mean difference; dashed lines depict limits of agreement (mean ± 2 SD).

View larger version (15K): [in a new window]   FIGURE 7.  Illustration of difference between stacked slices and true 3D calculation of MCA. In stacked slices' analysis, the MCA (MCAslices) is calculated as the length of the midmyocardial contour on 2D images, as illustrated on left-hand side, multiplied by slice thickness (d). As illustrated on right-hand side, 3D calculation of MCA (MCA3D) of LV locally differs from MCAslices depending on myocardial curvature perpendicular to slice orientation. The relationship between d and the real extension of the midmyocardial contour in the through-plane direction (L) depends on the angle () enclosed by the through-plane curvature and the sectioned slice. Midventricular contour is approximated by a piecewise linear line to demonstrate this effect.