Skip to main content
NCBI home page
npj Cardiovascular Health logoLink to npj Cardiovascular Health
. 2026 Jun 11;3:49. doi: 10.1038/s44325-026-00137-2

Ex vivo heart perfusion induces time dependent inflammatory activation in human donor hearts

Selena S Li 1, Bill Michaud 1, Asishana A Osho 2, David A D’Alessandro 2, Shannon N Tessier 3, Anthony Rosenzweig 4, S Alireza Rabi 5, James S Guseh 6,✉
PMCID: PMC13639011  PMID: 42277431

Abstract

Ex vivo heart perfusion (EVHP) has enabled the use of hearts donated after circulatory death (DCD), expanding the donor pool but with persistently higher rates of rejection and graft dysfunction. The molecular consequences of EVHP, however, remain incompletely defined. We quantified 11,083 proteins in perfusate from 20 human DCD hearts undergoing EVHP using an aptamer-based proteomic platform. Among the most strongly upregulated proteins were inflammatory cytokines IL-6 (log₂FC = 3.5, FDR = 8.9 × 10−6), IL-8 (CXCL8; log₂FC = 2.8, FDR = 4.7 × 10−6), and CCL2/MCP-1 (log₂FC = 2.8, FDR = 2.0 × 10−6). Levels of these cytokines correlated strongly with perfusion duration (Spearman ρ = 0.77–0.78, FDR < 5 × 10−5). Pathway analysis demonstrated enrichment of innate immune signaling, including interleukin-1 signaling (normalized enrichment score (NES) = 2.14, adjusted p = 0.0016) and leukocyte chemotaxis pathways. These cytokine findings were validated in an independent human cohort using multiplex immunoassay and reproduced in a swine DCD perfusion model. Together, these data identify a conserved, time-dependent innate immune response during EVHP and suggest that inflammatory activation may represent one mechanism linking prolonged perfusion to graft dysfunction after DCD heart transplantation.

Subject terms: Biomarkers, Cardiology, Diseases, Immunology, Medical research

Introduction

Heart transplantation remains the definitive therapy for end-stage heart failure but is constrained by a limited donor pool. Traditionally, only hearts from brain-dead donors were used for transplantation, but innovative techniques for evaluation and transportation of donor hearts, including ex-vivo heart perfusion (EVHP), have enabled expansion of the donor pool through use of hearts donated after circulatory death (DCD)1–3. The current standard for EVHP is the Organ Care System (OCS), a transport device that received FDA approval following a multicenter trial and has shown promise in DCD heart recovery2–4 in single-center and national studies1–3,5,6.

EVHP is used in approximately 70% of DCD heart donations7 but its direct effects on organ quality remain poorly understood. Remarkably, ex-vivo perfused hearts have been found to have increased rates of acute rejection8, and longer duration of EVHP is associated with reduced survival among DCD heart recipients, raising questions about the effect of EVHP itself on the donor organ9. The OCS perfuses the heart in an unloaded Langendorff mode10, and potential contributors to injury include non-physiologic coronary perfusion, edema formation, contact with artificial surfaces, hemolysis, and other consequences of prolonged perfusion.

Our group previously reported metabolomic changes during EVHP using the OCS platform11. Building on this work, we now apply high-throughput proteomics to characterize the “secretome” of the DCD heart and its perfusate during EVHP, as reflected by circulating protein levels.

Results

Ex-vivo heart perfusion (EVHP) of 20 human DCD donor hearts led to widespread proteomic changes. Among 11,083 proteins quantified by SomaScan, 2347 were significantly increased and 2323 decreased at end-perfusion compared to initial perfusion (FDR < 0.05) (Fig. 1A, Suppl. Table 1). Eight proteins were significantly upregulated (FDR < 0.05) with FC > 1.2: IL-6, CCL2/MCP-1, CXCL8/IL-8, ABHD5, ASB9, MANEAL, TNNT2/TNNC1, MYH7, and LDB3. The most upregulated proteins were IL-6 (log₂FC = 3.5, FDR = 8.9 × 10⁻⁶), CXCL8/IL-8 (log₂FC = 2.8, FDR = 4.7 × 10−6), and CCL2/MCP-1 (log₂FC = 2.8, FDR = 2.0 × 10−6). Gene set enrichment analysis (GSEA) of upregulated proteins identified predominant activation of inflammatory pathways, including interleukin-1 signaling (NES = 2.14, adj. p = 0.0016), neutrophil chemotaxis (NES = 2.02, adj. p = 0.0045), and monocyte chemotaxis (NES = 1.97, adj. p = 0.021) (Fig. 1, Supplementary Table 2). Fold change thresholds were used for visualization (Fig. 1A), and not for interpretation or pathway analysis, to avoid discarding biologically meaningful signals.

Fig. 1. Inflammatory proteomic changes during ex vivo heart perfusion (EVHP).

Fig. 1

Open in a new tab

A Volcano plot showing protein-level changes from initial to end perfusion across 11,083 quantified proteins. Proteins meeting significance thresholds for false discovery rate (FDR < 0.05) are shown in blue; those meeting both FDR and fold change thresholds are highlighted in red (FC thresholds shown for visualization purposes, not used for interpretation or pathway analysis). IL-6, IL-8 (CXCL8), and CCL2/MCP-1 were among the most significantly upregulated cytokines. B Box plots of IL-6, IL-8, and CCL2/MCP-1 protein levels at initial and end timepoints on EVHP, showing consistent and significant increases in each cytokine by end perfusion (relative fluorescence units, RFU). C Pathway enrichment analysis of proteins significantly increased during EVHP reveals upregulation of multiple inflammatory pathways, including interleukin-1 signaling, neutrophil and monocyte chemotaxis, and leukocyte migration. Circle size reflects statistical significance (–log₁₀ adjusted p-value); normalized enrichment scores (NES) are shown on the x-axis.

Correlation analysis using Spearman’s rank test across perfusion time (range 0–6 h) identified 166 proteins significantly associated with duration of EVHP (FDR < 0.05), including 118 that increased and 48 that decreased (Suppl. Table 3). IL-6, IL-8, and CCL2 again emerged as the top correlates with time on EVHP (ρ = 0.77–0.78, FDR < 5 × 10−5). Notably, we saw no statistically significant correlation between cardiac troponin and time on perfusion, although we observed modest statistically significant increases before and after perfusion (Suppl. Fig. 1). Overrepresentation analysis of these proteins revealed significant enrichment in innate immune and leukocyte recruitment pathways (Fig. 2, Suppl. Table 4).

Fig. 2. Inflammatory protein correlations with time on EVHP.

Fig. 2

Open in a new tab

A Volcano plot displaying the relationship between protein abundance and perfusion duration using Spearman correlation across 11,083 proteins. Proteins significantly correlated with time on perfusion (FDR < 0.05) are shown in blue; those meeting both FDR and correlation strength thresholds are highlighted in red. IL-6, IL-8, and CCL2/MCP-1 were among the top positively correlated proteins. B Loess plots showing time-dependent increases in IL-6, IL-8, and CCL2/MCP-1 protein levels across increasing hours on EVHP, confirming detection of a progressive inflammatory response during EVHP (RFU = relative fluorescence units). Cytokine levels significantly correlate with time (Spearman rho and FDR reported). C Overrepresentation analysis of proteins significantly correlated with time on EVHP reveals strong enrichment in immune regulatory pathways, including up- and down-regulation of NK cell recruitment, upregulated lymphocyte recruitment and activation, and Th2-predominant cytokine signaling and differentiation. Circle size reflects –log₁₀ adjusted FDR; color indicates degree of fold enrichment.

These findings were validated in an independent cohort of nine donor hearts (3 DBD, 6 DCD) perfused on OCS and analyzed by Luminex immunoassay. Unlike DBD hearts, DCD hearts undergo an obligatory period of warm ischemia time (WIT). Next, we investigated if the increase in inflammatory marker is unique to DCD. IL-6, IL-8, and CCL2 levels significantly increased during EVHP in all hearts with similar trends between DCD and DBD hearts (Fig. 3). There were no statistically significant differences between changes in cytokines in DCD and DBD hearts (IL-6: P = 0.286, IL-8: P = 0.643, CCL2: P = 0.095).

Fig. 3. Technical and biologic validation of inflammatory response in independent cohort.

Fig. 3

Open in a new tab

Inflammatory cytokines measured by 15-Plex Luminex immunoassay in an independent cohort of 3 hearts donated after brain death (DBD) and 6 additional hearts after circulatory death. Cytokine levels were compared to time on ex-vivo perfusion assessed by linear regression. IL-6, IL-8, and CCL2 demonstrated the strongest correlation with time on ex-vivo perfusion. Inflammatory profiles were similar between DBD and DCD hearts.

To assess whether cytokine induction was attributable to DCD-associated warm ischemia, we performed linear regression using warm ischemic time as the predictor and cytokine levels as outcomes. No significant associations were observed at either perfusion timepoint, suggesting that EVHP—rather than ischemic exposure—was the primary driver of the observed inflammatory signal within the protocol-constrained WIT range ( < 30 min) (Suppl. Fig. 2).

Finally, the robust cytokine response was validated in a swine DCD model using our in-house EVHP system, demonstrating that key cytokine responses are reproducible in an independent experimental system. While CCL2 is not represented in the porcine Luminex panel, the most pronounced changes were observed for IL-6 (fold change [FC] 377.04, P = 1.45 × 10−6) and IL-8 (CXCL8; FC 855.25, P = 2.65 × 10−5), both of which increased by nearly three orders of magnitude from initial to end-of-perfusion samples (Fig. 4).

Fig. 4. Validation of cytokine increase in a swine DCD model.

Fig. 4

Open in a new tab

Cytokine increases were validated using 6 DCD swine hearts on our in-house perfusion system, measured with the porcine 13-Plex Luminex immunoassay. While CCL2 is not represented in the porcine panel, IL-6 and IL-8 demonstrated the most profound increases after EVHP.

Discussion

Our study identifies three key findings regarding the immunologic effects of ex-vivo heart perfusion (EVHP) in human DCD donor hearts. First, EVHP induces a robust inflammatory response, marked by significant increases in IL-6, IL-8, and CCL2/MCP-1. Second, the induction of these cytokines is time-dependent and correlates strongly with perfusion duration, suggesting that EVHP itself, independent of warm ischemia, drives activation of proinflammatory networks. Third, pathway-level analyses reveal upregulation of innate immune responses—including monocyte recruitment, neutrophil chemotaxis, and interleukin-1 signaling.

Despite corticosteroids present in the perfusate, we observed approximately threefold increases in IL-6, IL-8, and CCL2. These cytokines are known mediators of leukocyte activation and trafficking: IL-8 promotes neutrophil recruitment and monocyte adhesion12; IL-6 and CCL2 are produced by both endothelial and immune cells and are involved in monocyte/macrophage mobilization13,14. Consistent with these roles, gene set enrichment analysis highlighted IL-1 signaling and leukocyte chemotaxis as dominant pathways upregulated during perfusion. These observations were replicated in an independent human cohort that included both DCD and brain-dead donor hearts, where similar inflammatory cytokine increases were observed using a different assay platform, and while CCL2 was not included in the swine panel, IL-6 and IL-8 increases were validated across species and in an independent experimental system using a swine DCD model on our in-house perfusion system. This consistency across technical and biological replication supports the robustness and generalizability of our findings.

Our findings are consistent with prior data from ex vivo lung and kidney perfusion, where similar inflammatory cytokines—including IL-6, IL-8, and CCL2—were elevated in perfusate and tissue samples. Transcriptomic profiling of human lungs on ex-vivo perfusion revealed marked upregulation of CCL2 (MCP-1) and CXCL8 (IL-8) in endothelial and immune cells15, with protein-level confirmation of IL-8 elevation16. Similarly, ex vivo kidney perfusion in a porcine model led to increased perfusate concentrations of IL-6 and IL-8, along with other cytokines17. These parallels suggest that innate immune activation may be a shared response across organ types during normothermic perfusion.

Importantly, we found no correlation between cytokine levels and warm ischemic time, arguing against warm ischemia as the primary driver of the observed immune response. While ischemia-reperfusion injury is known to activate similar pathways, the proportional relationship between cytokine levels and perfusion duration suggests a distinct contribution from the EVHP process itself. This was further supported by increases in cardiac troponin and IL-1 signaling that did not correlate with perfusion time, suggesting different underlying mechanisms.

These findings may help explain the higher rates of acute rejection and graft dysfunction observed in DCD heart recipients. Whether these cytokine responses are adaptive or maladaptive remains unknown, but they raise the possibility that immune modulation during EVHP could improve post-transplant outcomes. Indeed, prior work in a porcine model demonstrated that cytokine adsorption during EVHP improved contractility of DCD hearts18. Clinical use of an adsorption filter during ex-vivo lung perfusion demonstrated decreased cytokines (IL-6, IL-10, CCL2/MCP-1, and GCSF) with improved in-hospital and 1-year mortality after transplantation19. The clinical impact of cytokine accumulation during ex-vivo perfusion and its potential modulation require further investigation.

This study has limitations. Our cohort lacked a non-perfused DCD control group, as static cold storage does not permit comparable sampling. Similarly, in the swine experiments, there is no sham group or non-perfused group, for the same reason. In comparing DCD and DBD hearts, we are limited by the small sample size of our DBD cohort, as it is less common for DBD hearts to undergo EVHP in the clinical setting. In our targeted validation, the swine Luminex panel does not include CCL2, our top protein, though it does include our other top cytokines (IL-6 and IL-8). Sample size was limited, and future studies should incorporate mechanistic animal models and examine whether pharmacologic modulation of inflammation during EVHP can improve graft performance.

In conclusion, EVHP represents an important advancement to preserve, reanimate, and evaluate donor organs, allowing for expansion of the donor pool and utilization of otherwise undesirable organs. However, the effect of EVHP itself on the heart remains poorly understood, and early studies demonstrate potentially increased morbidity including acute rejection and primary graft dysfunction. Our study demonstrates that EVHP leads to a robust inflammatory response, characterized by significant increases in IL-6, IL-8, and CCL2, along with proinflammatory pathways involving innate immune activation and monocyte/macrophage recruitment. Further work is needed to delineate adaptive versus maladaptive immune responses and to understand the role of the inflammatory response on transplanted organ function.

Methods

Direct procurement of organs

The study includes human donor hearts accepted by our institution for transplantation and procured via direct procurement and perfusion (DPP). Our institution’s full DCD procurement protocol has previously been published20. Under DPP protocol, 30,000 IU of heparin is administered prior to extubation. After extubation, determination of circulatory death, and a “no touch” interval of 2–5 min, sternotomy is performed followed by rapid drainage of 1–1.5 L of donor blood from the right atrium. The donor heart is flushed with cold del Nido preservation solution, and cardiectomy is performed. The heart is then instrumented and perfused on the FDA-approved Organ Care System (OCS) machine and reperfused using sub-normothermic (34 °C), leukocyte-reduced whole donor blood1. A Priming solution, including 250 mg of methylprednisolone and 10,000 IU of heparin, as well as Maintenance solution is also added to the blood to make up the perfusate21. All human donor hearts were successfully transplanted.

Study design

This study was approved by the Massachusetts General Hospital Institutional Review Board (protocol #2022P000405, approved on March 1, 2022) and complies with all relevant ethical regulations in accordance with the Declaration of Helsinki. Informed consent for recipient data was obtained from all recipients. Informed consent for the collection of blood samples from donor hearts on OCS was obtained from donor families. In association with EVHP of 20 human DCD hearts on the OCS system, perfusate samples were drawn from the OCS circuit at two timepoints: an early timepoint (ranging 0–2 h) and at end-perfusion (ranging 3–6 h). Donor and recipient clinical characteristics (demographics, comorbidities, infection status, etc.) were recorded in a deidentified intramural heart transplant database (IRB protocol #2017P001969, approved on September 29, 2017).

Sample processing

Samples were stored in room air and transported using 10 mL BD Vacutainer CPT Sodium Citrate tubes. Plasma was separated from peripheral blood cells using centrifugation at 1500 relative centrifugal force for 15 min at room temperature, and stored at −80 °C.

Proteomic Profiling

The SOMAScan aptamer-based assay has previously been described with clinical validity and analytic accuracy in recent studies for high-throughput proteomic analysis22–25. SOMAScan uses synthetic DNA aptamers that bind specific protein targets; the aptamers are the quantified by microarray, and signal intensity is used as a proxy for protein abundance. Using this platform, perfusate samples were sent for proteomic profiling using the 11 K SOMAScan assay (SOMAlogic, Boulder, CO), which includes 11,037 protein analytes corresponding with 10,070 distinct human proteins. Protein abundance is reported in SOMAmer reagent relative fluorescence units (RFU). Assay data are normalized according to standard SOMAScan protocol26.

Statistical Analysis

Statistical analysis was conducted using R software27 and SOMAScan Data Delve Statistics Tool. Protein RFU values were log10-transformed and Z-score calculated for statistical testing. Paired student t-test was performed comparing end-timepoint values to initial-timepoint values to determine significant protein level changes as a result of EVHP. Fold change (ratio of RFU values between end and initial timepoints) was calculated and log2-transformed. Spearman’s test was performed as a nonparametric measure of correlation between individual protein levels and time on EVHP (range 0–6 h). False discovery rate (FDR) was calculated from p-values using the R package qvalue, with FDR ≤ 0.05 considered statistically significant.

Gene set enrichment analysis (GSEA), using the R fgsea package, and Panther overrepresentation analysis were performed for Gene Ontology biological processes28,29. Ranking was determined by -log10(p-value) multiplied by log2(fold change) for each protein, using statistically significant proteins for GSEA and overrepresentation analysis.

Targeted Technical Validation (Human Hearts)

Perfusate samples from an additional 9 donor hearts (3 brain dead donors, and 6 DCD) were obtained and processed as described above. Serum samples were sent for analysis using the Luminex Human Focused 15-Plex Discovery Assay (Eve Technologies, Calgary, Canada) for targeted validation of key cytokine responses using a different assay platform. Cytokine levels were analyzed in duplicates, averaged, and compared to time on EVHP using linear regression. To assess potential differences in cytokine responses between DCD and DBD hearts, we compared protein changes from initial to end perfusion in each cohort and used Wilcoxon rank sum testing to determine statistically significant differences.

Targeted Biological Validation (Swine Hearts)

Targeted validation was performed in swine hearts as a separate mammalian model system, providing biological replication. Hearts from six Yorkshire swine (30–35 kg, age 3–4 months, mixed sex) were procured using our previously described DCD protocol30. Yorkshire swine were chosen for the anatomical and physiological similarity to the human heart for ex-vivo perfusion31,32. Under our previously described protocol, sternotomy was performed and 300 IU of heparin was administered. Following extubation, 19 minutes of functional warm ischemic time was observed (MAP < 30 mmHg). The heart was vented and flushed with 1 L of del Nido cardioplegia. Cardiotomy was performed in standard fashion, hearts were instrumented and placed on the normothermic machine perfusion system. Hearts were perfused under Langendorff conditions using a base perfusate containing 0.96% Krebs-Henseleit Buffer, 0.02% hydrocortisone (50 mg/mL), 0.5% heparin (1000 U/mL), and 1 L of isolated red blood cells (1:2) among other additives30. Perfusate samples were collected after 30 minutes and 6 hours of perfusion. Pro‑inflammatory cytokines were measured using the porcine Luminex 13‑Plex Discovery Assay (Eve Technologies, Calgary, Canada) for targeted validation of cytokine results. Cytokine concentrations at 30 min versus 6 h were compared using paired t-test. An alpha level of 0.0045 was considered significant, using Bonferroni correction for multiple hypothesis testing.

Supplementary information

Supplementary Information (679.3KB, pdf)
Supplementary Tables (1.4MB, xlsx)

Acknowledgements

Dr. Li is funded by the NIH/NIAID R25 AI 147393 award (Pathways to Mentorship and Research: Training the Next Generation Physician Scientist) and the Thoracic Surgery Foundation Nina Starr Braunwald Research Fellowship. Dr. Guseh is supported by the American Heart Association Harold Amos Program (19AMFDP34990046), the Presidents and Fellows of Harvard College (5KL2TR002542-04), and philanthropic gifts from Frederick and Ines Yeatts. He is also supported by American Heart Association Strategically Focused Research Network grants 24SFRNPCN1284389 and 24SFRNCCN1276092.

Author contributions

S.L. and J.G. wrote the main manuscript text. B.M. prepared samples for the experiments, which were collected from transplants performed by A.O., D.D. and S.R. Swine experiments were performed by S.R. and S.T. Data analyses were performed by S.L., which were then reviewed by A.R., S.R. and J.G. All authors reviewed and approved the manuscript.

Data availability

The data supporting the findings of this study are available within the Article and its Supplementary Information. Additional raw data is not publicly available due to patient privacy considerations and informed consent but are available from the corresponding author on reasonable request.

Code availability

Analyses were performed using publicly available packages in R and using the Somascan platform software. Additional questions about code may be addressed to the corresponding author.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s44325-026-00137-2.

References

  • 1.Scheuer, S. E., Jansz, P. C. & Macdonald, P. S. Heart transplantation following donation after circulatory death: Expanding the donor pool. J. Heart Lung Transpl.40, 882–889 (2021). [DOI] [PubMed] [Google Scholar]
  • 2.Dhital, K., Ludhani, P., Scheuer, S., Connellan, M. & Macdonald, P. DCD donations and outcomes of heart transplantation: the Australian experience. Indian J. Thorac. Cardiovasc Surg.36, 224–232 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Messer, S. et al. A 5-year single-center early experience of heart transplantation from donation after circulatory-determined death donors. J. Heart Lung Transpl.39, 1463–1475 (2020). [DOI] [PubMed] [Google Scholar]
  • 4.Tchana-Sato, V. et al. Successful clinical transplantation of hearts donated after circulatory death using normothermic regional perfusion. J. Heart Lung Transpl.38, 593–598 (2019). [DOI] [PubMed] [Google Scholar]
  • 5.Hoffman, J. R. H. et al. Early US experience with cardiac donation after circulatory death (DCD) using normothermic regional perfusion. J. Heart Lung Transpl.40, 1408–1418 (2021). [DOI] [PubMed] [Google Scholar]
  • 6.D’Alessandro, D. A. et al. Hemodynamic and Clinical Performance of Hearts Donated After Circulatory Death. J. Am. Coll. Cardiol.80, 1314–1326 (2022). [DOI] [PubMed] [Google Scholar]
  • 7.Benkert, A. R. et al. Early U.S. Heart Transplant Experience With Normothermic Regional Perfusion Following Donation After Circulatory Death. JACC Heart Fail. 10.1016/j.jchf.2024.06.007 (2024). [DOI] [PubMed]
  • 8.Li, S. S. et al. Acute rejection in donation after circulatory death (DCD) heart transplants. J. Heart Lung Transpl.43, 148–157 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Singh, R. et al. Extended duration of ex-vivo perfusion is associated with worse survival in donation after circulatory death heart recipients: A national database analysis. J. Heart Lung Transpl. 10.1016/j.healun.2024.11.018 (2024). [DOI] [PubMed]
  • 10.Bryner, B. S., Schroder, J. N. & Milano, C. A. Heart transplant advances: Ex vivo organ-preservation systems. JTCVS Open8, 123–127 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Trimigno, A. et al. Metabolic Choreography of Energy Substrates During DCD Heart Perfusion. Transpl. Direct10, e1704 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Gerszten, R. E. et al. MCP-1 and IL-8 trigger firm adhesion of monocytes to vascular endothelium under flow conditions. Nature398, 718–723 (1999). [DOI] [PubMed] [Google Scholar]
  • 13.Mantovani, A. & Garlanda, C. Humoral Innate Immunity and Acute-Phase Proteins. N. Engl. J. Med388, 439–452 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Roca, H. et al. CCL2 and interleukin-6 promote survival of human CD11b+ peripheral blood mononuclear cells and induce M2-type macrophage polarization. J. Biol. Chem.284, 34342–34354 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Gouin, C. et al. Cell type- and time-dependent biological responses in ex vivo perfused lung grafts. Front Immunol.14, 1142228 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Machuca, T. N. et al. Protein expression profiling predicts graft performance in clinical ex vivo lung perfusion. Ann. Surg.261, 591–597 (2015). [DOI] [PubMed] [Google Scholar]
  • 17.Stone, J. P. et al. Ex Vivo Normothermic Perfusion Induces Donor-Derived Leukocyte Mobilization and Removal Prior to Renal Transplantation. Kidney Int Rep.1, 230–239 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Saemann, L. et al. Cytokine Adsorption During Ex Vivo Blood Perfusion Improves Contractility of Donation After Circulatory Death Hearts. J. Am. Heart Assoc.13, e036872 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Boffini, M. et al. Cytokines Removal During Ex-Vivo Lung Perfusion: Initial Clinical Experience. Transpl. Int.36, 10777 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Funamoto, M., Pierson, R. N., Nguyen, J. H. & D’Alessandro, D. A. Surgical and logistical concerns for ex vivo-based perfusion strategies for “donation after circulatory death” multiorgan recovery. JTCVS Tech.11, 49–56 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Ardehali, A. et al. Ex-vivo perfusion of donor hearts for human heart transplantation (PROCEED II): a prospective, open-label, multicentre, randomised non-inferiority trial. Lancet385, 2577–2584 (2015). [DOI] [PubMed] [Google Scholar]
  • 22.Emilsson, V. et al. Co-regulatory networks of human serum proteins link genetics to disease. Science361, 769–773 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Ngo, D. et al. Aptamer-Based Proteomic Profiling Reveals Novel Candidate Biomarkers and Pathways in Cardiovascular Disease. Circulation134, 270–285 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Rhee, J. et al. Serum Proteomics of Older Patients Undergoing Major Cardiac Surgery: Identification of Biomarkers Associated With Postoperative Delirium. Front Aging Neurosci.13, 699763 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Candia, J. et al. Assessment of Variability in the SOMAscan Assay. Sci. Rep.7, 14248 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Somalogic. SomaScan v4 Data Standardization and File Specification Technical Note, https://somalogic.com/wp-content/uploads/2022/07/SL00000048_Rev-3_2022-01_-Data-Standardization-and-File-Specification-Technical-Note-v2.pdf.
  • 27.R Core Team. R: A Language and environment for statistical computing., https://www.R-project.org (2020).
  • 28.Thomas, P. D. et al. PANTHER: Making genome-scale phylogenetics accessible to all. Protein Sci.31, 8–22 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Mi, H. et al. Protocol Update for large-scale genome and gene function analysis with the PANTHER classification system (v.14.0). Nat. Protoc.14, 703–721 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Higuita, M. L. et al. Alternative preservation strategies of DCD hearts: Effect of cold ischemia time before normothermic machine perfusion. medRxiv, 10.1101/2025.06.04.25329001 (2025). [DOI] [PMC free article] [PubMed]
  • 31.Kumar, R., Chung, W. Y., Dennison, A. R. & Garcea, G. Ex Vivo Porcine Organ Perfusion Models as a Suitable Platform for Translational Transplant Research. Artif. Organs41, E69–E79 (2017). [DOI] [PubMed] [Google Scholar]
  • 32.Hatami, S. et al. Normothermic Ex Situ Heart Perfusion in Working Mode: Assessment of Cardiac Function and Metabolism. J. Vis. Exp. 10.3791/58430 (2019). [DOI] [PubMed]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information (679.3KB, pdf)
Supplementary Tables (1.4MB, xlsx)

Data Availability Statement

The data supporting the findings of this study are available within the Article and its Supplementary Information. Additional raw data is not publicly available due to patient privacy considerations and informed consent but are available from the corresponding author on reasonable request.

Analyses were performed using publicly available packages in R and using the Somascan platform software. Additional questions about code may be addressed to the corresponding author.

Articles from npj Cardiovascular Health are provided here courtesy of Nature Publishing Group

RESOURCES