HPB O04 - Human Splenic Ex-Vivo Perfusion Model: A Platform for Translational Basic Science Research

Abstract

Abstract Background The function of the spleen has always been shrouded in mystery and since the Renaissance leading figures have attempted to understand its function. Vesalius questioned Galen and Marcello Malpighi (1628-1694) gave the first accurate description of the histology and the physiology. The spleen still poses unique challenges as invasive approaches may produce intraperitoneal bleeding, differences between human and animal spleens complicates comparative studies, and post-mortem examinations reveal only late-stage disease manifestations and functional studies are impossible. This study describes a human splenic ex-vivo perfusion model for translational research, using microbiological studies as markers of splenic physiological function. Method Patients with left-sided pancreatic cancers scheduled for distal pancreatic and splenic resection were identified from the Hepato-Pancreatic and Biliary MDT at the University Hospitals of Leicester NHS Trust. Patients were informed, consented, and the surgical approach remained standard, with splenic artery ligation deferred until specimen extraction. The artery was cannulated, and residual blood clots were cleared with Custodiol® HTK solution. The perfusion technique employs an extracorporeal circuit incorporateing anoxygenator and a heated water chamber maintaining a 37°C perfusate. The perfused spleen was infected with Streptococcus Pneumoniae strains, with regular biopsies and perfusate gas analysis performed. Results Since February 2023, nine spleens have been successfully perfused, with an average weight of 169.5 grams, cold ischemic time of 116 minutes, and warm ischemic time of 12.2 minutes. Pressure was maintained at 80 mmHg and hourly perfusate gas analysis demonstrated a mean pH of 7.1 and a gradual increase in lactate. Histopathological examination revealed clear distinctions between the lymphoid tissue and splenic red pulp, with intact central arterioles and healthy cell morphology at time 0 and after 6 hours of perfusion. Additionally, the perfused spleens cleared S. pneumoniae strains to below-detection levels. Conclusion Our work demonstrates that a stable human splenic ex-vivoperfusion model which maintains structural and physiological integrity can be achieved from tissue that would otherwise be discarded. The continued physiological function is confirmed by the clearance of S. pneumoniae. While currently employed for microbiological studies, this model has the potential to facilitate a wide range of different translational science studieswhile reducing the need for animal experimentation. The perfusion times employed for the present study were sufficient to demonstrate bacterial clearance which was the aim but there is capacity in the system to significantly prolong the duration.