Dual manganese-enhanced and delayed gadolinium-enhanced MRI detects myocardial border zone injury in a pig ischemia-reperfusion model

Abstract

Background: Gadolinium (Gd)-based delayed-enhancement MRI (DEMRI) identifies nonviable myocardium but is nonspecific and may overestimate nonviable territory. Manganese (Mn(2+))-enhanced MRI (MEMRI) denotes specific Mn(2+) uptake into viable cardiomyocytes. We performed a dual-contrast myocardial assessment in a porcine ischemia-reperfusion (IR) model to test the hypothesis that combined DEMRI and MEMRI identifies viable infarct border zone (BZ) myocardium in vivo.

Methods and results: Sixty-minute left anterior descending coronary artery IR injury was induced in 13 adult swine. Twenty-one days post-IR, 3-T cardiac MRI was performed. MEMRI was obtained after injection of 0.7 mL/kg Mn(2+) contrast agent. DEMRI was then acquired after injection of 0.2 mmol/kg Gd. Left ventricular (LV) mass, infarct, and function were analyzed. Subtraction of MEMRI defect from DEMRI signal identified injured BZ myocardium. Explanted hearts were analyzed by 2,3,5-triphenyltetrazolium chloride stain and tissue electron microscopy to compare infarct, BZ, and remote myocardium. Average LV ejection fraction was reduced (30±7%). MEMRI and DEMRI infarct volumes correlated with 2,3,5-triphenyltetrazolium chloride stain analysis (MEMRI, r=0.78; DEMRI, r=0.75; P<0.004). MEMRI infarct volume percentage was significantly lower than that of DEMRI (14±4% versus 23±4%; P<0.05). BZ MEMRI signal-to-noise ratio (SNR) was intermediate to remote and core infarct SNR (7.5±2.8 versus 13.2±3.4 and 2.9±1.6; P<0.0001), and DEMRI BZ SNR tended to be intermediate to remote and core infarct SNR (8.4±5.4 versus 3.3±0.6 and 14.3±6.6; P>0.05). Tissue electron microscopy analysis exhibited preserved cell structure in BZ cardiomyocytes despite transmural DEMRI enhancement.

Conclusions: The dual-contrast MEMRI-DEMRI detects BZ viability within DEMRI infarct zones. This approach may identify injured, at-risk myocardium in ischemic cardiomyopathy.

Figure 1. DEMRI vs. MEMRI vs. TTC

A) Representative slice progressing from the base to the apex of the heart (top row to bottom row), showing short-axis in vivo MRI images and matching TTC sections from an explanted swine heart 21 days post-IR. Note the MEMRI signal defect area (middle column) is visibly smaller than corresponding DEMRI signal area (left column). Tracings of MEMRI defect and DEMRI signal areas are shown in each slice. B) Bland-Altman plots showing the differences between TTC and DEMRI (left) and TTC and MEMRI (right) infarct volumes in grams, plotted against the average infarct volume (between TTC and DEMRI or MEMRI) for each swine heart. The average difference (Bias) between the measurements is shown (dotted line) +/− two Standard Deviations (solid horizontal lines) for both DEMRI and MEMRI. Note that all TTC-MEMRI differences were greater than zero, all TTC-DEMRI differences were less than zero, and all values fell within two Standard Deviations of their respective average differences. N=13.

Figure 2. SNR Heterogeneity within Border Zone Regions

First Row: Representative short-axis images showing the differing infarct patterns of DEMRI signal (left column) and MEMRI defect (middle column). Superimposed tracings (right column) show the pattern mismatch of the two contrast agents. Second Row: the area of DEMRI (yellow tracing) is visibly greater than the MEMRI defect area (red tracing). Colored circles reflect regions of interest (ROI) with respective SNR calculations noted in lower panel. SNR from the border zone of DEMRI region (blue ROI) trends lower than the core infarct zone of DEMRI region (green ROI). The corresponding region on MEMRI is bright (white ROI), indicating Mn²⁺ uptake. However, the SNR of this region is significantly less than the SNR of the remote region on MEMRI (purple ROI). SNR values represent the average SNR’s from ventricular short-axis images, with 2–3 mid-ventricular slices per swine heart. For DEMRI: *p<0.05 versus core infarct SNR; for MEMRI: *p<0.05 vs remote SNR, ^#p<0.05 vs core infarct SNR.

Figure 3. TEM of Infarct, Border, and Remote Myocardium

TEM sections of explanted swine heart, showing a single representative cardiomyocyte from Remote (upper), Infarcted (middle), and Border zone (lower) myocardium, 21 days post-IR. In the Infarct zone, note the nuclear (N) disarray and clumped chromatin, as well as the near absence of intact mitochondria (M) and disorganized remnants of sarcomeric myofibrillar organization (S). Border zone cells, however, exhibited much less disruption of cell architecture and relative preservation of its contractile components. Detailed TEM findings are shown in the Table.