Local amiodarone delivery via an extracellular collagen matrix patch to reduce new-onset postoperative atrial fibrillation

Figures

• 

  CardiaMend ECM patch saturated with Amiodarone on epicardial surface.

• 

  Figure 1. Representative hematoxylin and eosin–stained sections from the rat model at 28 days are shown with control patches (without amiodarone, A and C) and with amiodarone-loaded patches (B and D). The patches can be seen at low magnification in panels (A) and (B), conforming to the epicardial surface of the heart without visual evidence of fibrosis or any adverse effects in the adjacent myocardium. In the greater magnification images (C and D, regions indicated by dotted boxes in corresponding images A and B), the interface between the patch and epicardium is highlighted with an absence of classic foreign body response (no giant cells or encapsulation) in either condition. The patches with amiodarone do not show signs of any increased inflammatory response. Scale bars: 2 mm (top images), 200 μm (bottom images).

• 

  Figure 2. Tissue amiodarone concentrations from the rat model. The greatest concentration of amiodarone in atrial tissue was observed at day 1, decreasing rapidly thereafter. Liver and plasma concentrations remained low to undetectable throughout the study period. Data presented as mean ± standard deviation.

• 

  Figure 3. Gross morphology of the patch in the ovine model at necropsy. Representative images show the patch at 3 days (A), before incision into the pericardial space at day 7 (B), after incision into the pericardial space at day 7 (C), and at day 28 (D). Arrows indicate the location of the patch. No adhesions were observed between the patch and the epicardial surface at any time point. In the early postoperative period (A-C), the lateral edges of the device maintained direct apposition to the native epicardium.

• 

  Figure 4. Hematoxylin and eosin (H&E)-stained sections of the left atria (A, C, E, G, I) and Masson's trichrome stained sections of the left ventricles (B, D, F, H, J) in the ovine model. No fibrosis was visualized comparing the normal, untreated heart (A, B), or in experimental conditions including day 3 (C, D), day 7 (E, F), day 28 (G, H), or in the high-dose intravenous control at day 14 (I, J). There is no patch in any image. The epicardium is seen at the top of each atrium image and oriented with the epicardium to the right on each ventricle image. Scale bar = 100 μm.

• 

  Figure 5. Histologic images (hematoxylin and eosin) of the patch in the ovine model. At day 3 (A), the patch exhibited minimal early inflammation and no adverse reaction to amiodarone, with sparse cellularity and a thin superficial fibrin layer (arrows). By day 28 (B), normal host cell repopulation and neovascularization are evident. The collagen fiber architecture of the patch is visible within the middle of the host tissue (bracket) with a new epithelial layer of tissue on the top surface. The greater-magnification image (C and indicated by the dotted rectangle in image B) is at the interface between the patch and the neoepithelial layer and shows a seamless transition between host and implant, no foreign body response, with neovascularization by small capillaries (arrow) to support the cells, predominantly fibroblasts, now populating the patch. Scale bars: 50 μm (A), 300 μm (B), 100 μm (C).

• 

  Figure 6. Temporal distribution of amiodarone concentration in the sheep model as measured by liquid chromatography–mass spectrometry. Data points represent the means of measured values for tissue samples. Amiodarone levels were greatest in adipose-rich tissues, whereas atrial concentrations remained within therapeutically relevant ranges. Plasma, lung, and liver concentrations were consistently below 10 μg/g at all time points and are represented by a single light blue line. The standard deviation is high and not included on this image so as not to obscure the visualization.

• 

  Figure 7. Comparison of tissue amiodarone concentrations between experimental sheep and a maximally loaded intravenous (IV) positive control animal. The amiodarone concentration in atrial tissue was significantly greater in the experimental group compared with the control animal at day 3 (A) and day 7 (B). Conversely, concentrations in lung and liver tissue were 6 to 8 times lower in the experimental group than in controls. Data presented as mean ± standard deviation.

• 

  Figure E1. Quantification of fibrosis was evaluated by measuring total collagen content from color analysis of trichrome stained images in the ovine model. The percentage of total image area occupied by collagen, quantified via image analysis of histology sections, was greater in the atria compared with the ventricles. Experimental conditions with amiodarone-loaded patches did not show increased collagen content relative to positive or negative controls, confirming visual observations suggesting no development of fibrosis. Data presented as mean ± standard deviation. LA, Left atrium; RA, right atrium; LV, left ventricle; RV, right ventricle.