Triglycidyl amine crosslinking combined with ethanol inhibits bioprosthetic heart valve calcification

Abstract

Background: One of the most important factors responsible for the calcific failure of bioprosthetic heart valves is glutaraldehyde crosslinking. Ethanol (EtOH) incubation after glutaraldehyde crosslinking has previously been reported to confer anticalcification efficacy for bioprostheses. The present studies investigated the anticalcification efficacy in vivo of the novel crosslinking agent, triglycidyl amine (TGA), with or without EtOH incubation, in comparison with glutaraldehyde.

Methods: The TGA crosslinking (±EtOH) was used to prepare porcine aortic valves for both rat subdermal implants and sheep mitral valve replacements, for comparisons with glutaraldehyde-fixed controls. Thermal denaturation temperature, an index of crosslinking, cholesterol extraction, and hydrodynamic properties were quantified. Explant endpoints included quantitative and morphologic assessment of calcification.

Results: Thermal denaturation temperatures after TGA were intermediate between unfixed and glutaraldehyde-fixed. EtOH incubation resulted in almost complete extraction of cholesterol from TGA or glutaraldehyde-fixed cusps. Rat subdermal explants (90 days) demonstrated that TGA-EtOH resulted in a significantly greater level of inhibition of calcification than other conditions. Thus, TGA-ethanol stent mounted porcine aortic valve bioprostheses were fabricated for comparisons with glutaraldehyde-pretreated controls. In hydrodynamic studies, TGA-EtOH bioprostheses had lower pressure gradients than glutaraldehyde-fixed. The TGA-ethanol bioprostheses used as mitral valve replacements in juvenile sheep (150 days) demonstrated significantly lower calcium levels in both explanted porcine aortic cusp and aortic wall samples compared with glutaraldehyde-fixed controls. However, TGA-EtOH sheep explants also demonstrated isolated calcific nodules and intracuspal hematomas.

Conclusions: The TGA-EtOH pretreatment of porcine aortic valves confers significant calcification resistance in both rat subdermal and sheep circulatory implants, but with associated structural instability.

Figure 1

Thermal denaturation temperatures (Ts) of porcine aortic leaflets. After crosslinking with Glut, Glut-EtOH, TGA, or TGA-EtOH, Ts is significantly higher than that of untreated (native) tissue. §p<0.001 vs. native, *p<0.001 vs Glut). [mean± S.D.]

Figure 2

Cholesterol analysis of porcine aortic valve leaflets. EtOH treatment significantly reduces cholesterol content of porcine aortic leaftlets, whether the leaflet is first crosslinked with either Glut or TGA, or if initially treated with EtOH. (**p<0.001 vs untreated tissue; NS between EtOH groups.)

Figure 3

90 day rat subdermal explants. (A) Calcium (Ca) and (B) Phosphorus (P) analysis of explanted porcine aortic valve leaflets, showing significant suppression of calcification by all treatments compared to Glut (§p<0.001 by ANOVA). Further significant differences exist between TGA and TGA-ETOH (**p=0.001), and between TGA-EtOH and Glut-EtOH (*p=0.016). Differences between P levels were statistically significant, and showed a difference between Glut and all other treatments by ANOVA (§p<0.001 by ANOVA).

Figure 4

Hydrodynamic testing results, using steady flow conditions. TGA-EtOH valves exhibit lower pressure drops than Glut valves at flow rates ranging from 5 to 30 liters per minute (lpm), an average difference which increases as the flow rate increases. (*p≤0.05 vs Glut; **p<0.01 vs. Glut)

Figure 5

Morphologic findings in 150 day sheep mitral bioprosthetic valve explants. Gross photographs A-C; photomicrographs D-G, with original magnification 100x. A. Typical TGA-EtOH valve, showing no gross abnormalities, and (E) cuspal microscopic qualities comparable to an unimplanted leaflet cusp (D). B. Atypical TGA-EtOH valve explant showing small calcific nodule (black arrow) and hematoma (red arrow), in addition to a tear believed to be the result of necropsy error in upper right quadrant. F and G show von Kossa stain of the calcific nodule and H&E of hematoma respectively. C. Typical Glut150 day sheep mitral bioprosthetic valve explant, with large gross calcifications on all three leaflets, one of which is specifically indicated (black arrow). D, E, and F stained with hematoxylin and eosin, F with von Kossa stain. Micrograph scalebars = 100μM.

Figure 6

Quantification of Ca and P of 150 day sheep mitral valve explants. Both leaflet and wall segments of the TGA-EtOH trileaflet bioprostheses had significantly lower calcification than Glut-prepared materials. A: Ca analysis. B: P analysis. (**p<0.001, *p=0.006 TGA-EtOH vs. Glut leaflet and wall respectively)