A custom preservation solution extends the ex vivo availability of living heart valves for transplantation

Figures

• 

  Normothermic storage enables the ex vivo preservation of living heart valves.

• 

  Figure 1. Graphical Abstract. Normothermic storage in a custom valvular preservation solution can extend ex vivo valvular allograft viability. Created using BioRender.com.

• 

  Figure 2. Valvular tissue viability in hypothermic versus normothermic storage. A, Porcine sample procurement. Valvular leaflet (B) and pulmonary artery (C) tissues show a rapid decrease in viability in static cold storage in VPS. Leaflet (D) and pulmonary artery (E) tissues similarly show a loss of viability by 2 weeks ex vivo in HBSS with Knock Out Serum. F, Whole tissue viability in normothermic HBSS with Knock Out Serum is significantly reduced in the first 3 weeks ex vivo, versus normothermic VPS where viability is maintained (G). Mean ± standard deviation, 2 to 3 technical replicates per datapoint with 2 to 20 biological replicates per group and time point are shown. Statistics are via the Kruskal-Wallis test with the Dunn multiple comparisons test comparing longitudinal values to baseline. Fresh valve samples are the same in (B/D) and (C/E) (also the same as fresh valves shown in Figures 3 and 4). VPS, Valve Preservation Solution; HBSS, Hanks' Balanced Salt Solution.

• 

  Figure 3. Valvular tissue viability in normothermic storage. Valves demonstrate preserved leaflet (A) and pulmonary artery (B) viability for up to 7 weeks ex vivo, with a minimal decrease in lactate release over time (C). D, Supernatant lactate dehydrogenase decreases significantly throughout the first weeks of ex vivo storage. E, Allografts show consistent glucose consumption over time. F, Brightfield imaging of valve interstitial cells isolated from 4-week stored valves. For (A and B), 2 to 3 technical replicates were obtained for each datapoint. For (A and E), mean ± standard deviation are shown, with 2 to 20 individual tissue samples per group and time point. Statistics are via the Kruskal-Wallis test with the Dunn multiple comparisons test comparing each longitudinal time point to the first-available value. Fresh samples are the same as those shown in Figures 2 and 4.

• 

  Figure 4. Parameter-controlled study of biochemical additives. AA (A, D), ATP (B, E), and glutathione (C, F) were added to VPS throughout the culture period to assess their effect on longitudinal valve viability. None of these additives significantly improved the longitudinal viability of leaflet (A-C) or pulmonary artery (D-F) tissue. For (A-F), 2 to 3 technical replicates were obtained for each datapoint, with 2 to 20 biological replicates per group and time point. G-I, No biochemical additives increased glucose consumption throughout storage (marker of metabolic activity). For (G-I), there were 2 to 20 tissue samples per group and timepoint. J, H&E cross-sections through the valvular leaflet for each condition at 5 and 6 weeks' ex vivo (representative images from 2-3 biological replicates per group and time point). For (A-F), mean ± standard deviation are shown. Statistics are via the Kruskal-Wallis test with the Dunn multiple comparisons test comparing each longitudinal time point to the first-available value. Fresh valve samples are the same in (A-C) and (D-F) (shared with Figures 2 and 3). AA, Arachidonic acid; ATP, adenosine 5′-triphosphate; VPS, Valve Preservation Solution; H&E,…

• 

  Figure 5. Histologic analysis of leaflet viability and microarchitecture. A, Representative H&E cross-sections through the leaflet demonstrating preservation of tissue microarchitecture through approximately 3 weeks ex vivo, at which point the leaflet trilaminar architecture begins to degrade. B, Representative leaflet TUNEL stains. C-D, Leaflet thickness and cellularity are preserved throughout storage. E, The percent of live cells decreases in the first 2 weeks ex vivo, with tissue viability preserved at 70% thereafter. Representative images are shown from 1 to 3 biological replicates per group and timepoint. For (C-E), mean ± standard deviation are shown; 1 to 3 biological replicates per timepoint. Statistics are via the Kruskal-Wallis test with Dunn's multiple comparisons test comparing each longitudinal time point with baseline. H&E, Hematoxylin & eosin; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labeling.

• 

  Figure 6. Leaflet extracellular matrix composition throughout storage. A, Movat Pentachrome staining highlighting the relative preservation of collagen (yellow), glycosaminoglycans (blue), and elastin (black) throughout the valvular leaflet up to 3 weeks ex vivo. B, Extracellular matrix composition was quantified at baseline, and 3 and 6 weeks' ex vivo. Between 3 and 6 weeks of storage, glycosaminoglycans are lost and the valvular leaflet becomes primarily collagen-based, with physiologic elastin present in the ventricularis layer. Mean ± standard deviation are shown; 2 biological replicates per time point are shown.

• 

  Figure 7. Leaflet cell phenotype throughout storage. Immunofluorescent staining of the leaflet cross-section showing preservation of leaflet vimentin expression and lack of α-SMA expression, relative to baseline. Valve interstitial cells retain a quiescent phenotype, without significantly upregulated proliferation (KI67). Valve endothelial cells (VWF) are preserved through 2 weeks ex vivo. Representative images are shown from N = 2 biological replicates per group and timepoint. α-SMA, Alpha-smooth muscle actin; VWF, von Willebrand Factor.

• 

  Figure E1. Experimental design. Upon receipt, pulmonary allografts were dissected into discrete samples. All samples were treated with an antibiotic cocktail of vancomycin, lincomycin, and polymyxin in either Valve Preservation Solution (VPS) or Hanks' Balanced Salt Solution (HBSS) with Knock Out Serum (KOSR). After three 24-hour washes in the antibiotic solution, samples were evenly and randomly distributed to their experimental condition for longitudinal storage (either cold or normothermic static storage).

• 

  Figure E2. Short-term valvular allograft preservation in static cold storage. A, Valvular leaflets tend toward preserved viability after 96 hours ex vivo (72 hours of antibiotic treatment post-tissue receipt). B, Pulmonary artery tissue tends towards reduced viability within 72 hours ex vivo (48 hours posttissue receipt), irrespective of whether antibiotics were included in the solution. Mean ± standard deviation are shown, with 3 technical replicates obtained for each datapoint; 2 to 3 biological replicates per group and time point are shown (each donor labeled). Statistics are via the Kruskal-Wallis test with the Dunn multiple comparisons test comparing longitudinal columns to baseline.

• 

  Figure E3. Histologic tissue preservation in normothermic versus static cold storage. Tissue microarchitecture demonstrates interstitial edema by 3 weeks ex vivo. In HBSS at 37 °C (5% CO2), complete loss of the trilaminar architecture is observed by 3 weeks ex vivo. In VPS at 37 °C (5% CO2), the trilaminar architecture begins to degrade at 3 weeks ex vivo. Representative images are shown from 2 to 3 biological replicates per group and timepoint. HBSS, Hanks' Balanced Salt Solution; VPS, Valve Preservation Solution.

• 

  Figure E4. Histologic analysis of pulmonary artery viability and microarchitecture. A and B, Representative full-thickness and high-magnification H&E cross-sections through the pulmonary artery. C, Representative pulmonary artery TUNEL stains. For (A-C), representative images are shown from 2 to 3 biological replicates per group and time point. D and E, Pulmonary artery thickness and cellularity are preserved throughout storage. F, The percent of live cells decreases steadily throughout storage, with 70% of cells alive at 3 weeks and 45% of cells alive at 6 weeks. For (D-F) mean ± standard deviation are shown; 1 to 3 biological replicates per time point are shown. Statistics are via the Kruskal-Wallis test with the Dunn multiple comparisons test comparing longitudinal columns to baseline. H&E, Hematoxylin & eosin; TUNEL, terminal deoxynucleotidyl transferase dUTP nick end labeling.