Dual versus single vessel normothermic ex vivo perfusion of rat liver grafts using metamizole for vasodilatation

Abstract

Background: Normothermic ex vivo liver perfusion (NEVLP) is a promising strategy to increase the donor pool in liver transplantation. Small animal models are essential to further investigate questions regarding organ preservation and reconditioning by NEVLP. A dual vessel small animal NEVLP (dNEVLP) model was developed using metamizole as a vasodilator and compared to conventional portovenous single vessel NEVLP (sNEVLP).

Methods: Livers of male Wistar rats were perfused with erythrocyte-supplemented culture medium for six hours by either dNEVLP via hepatic artery and portal vein or portovenous sNEVLP. dNEVLP was performed either with or without metamizole treatment. Perfusion pressure and flow rates were constantly monitored. Transaminase levels were determined in the perfusate at the start and after three and six hours of perfusion. Bile secretion was monitored and bile LDH and GGT levels were measured hourly. Histopathological analysis was performed using liver and bile duct tissue samples after perfusion.

Results: Hepatic artery pressure was significantly lower in dNEVLP with metamizole administration. Compared to sNEVLP, dNEVLP with metamizole treatment showed higher bile production, lower levels of transaminases during and after perfusion as well as significantly lower necrosis in liver and bile duct tissue. Biochemical markers of bile duct injury showed the same trend.

Conclusion: Our miniaturized dNEVLP system enables normothermic dual vessel rat liver perfusion. The administration of metamizole effectively ameliorates arterial vasospasm allowing for six hours of dNEVLP, with superior outcome compared to sNEVLP.

Fig 1. Perfusion setups.

(A) Technical drawing of the sNEVLP setup consisting of the perfusion chamber with inlets for the portal vein (PV) and outlets for the vena cava (VC) and bile duct (BD), an oxygenator, a bubble trap (BT), a dialysis circuit and four roller pumps (P1, P3-5). Several outlets within the perfusion circuit allowed for sample collection of the arterial (A) and venous (V) perfusate and dialysate (d) and for measurement of the portovenous pressure (PV). (B) Technical drawing of the dNEVLP setup with an additional inflow for the hepatic artery (HA), a fifth roller pump (P2) and another two outlets for metamizole bolus application (B) and measurement of the arterial pressure (AP). (C) Liver in the perfusion chamber (dNEVLP) with cannulated bile duct, hepatic artery, portal vein and infrahepatic vena cava (top to bottom). (D) Close-up on cannulated hepatic artery (yellow cannula) witch patch from abdominal aorta.

Fig 2. Comparison of dNEVLP groups 1.

Comparison of the three dNEVLP groups: (A) portovenous pressure, (B) arterial pressure, (C) perfusate pH, (D) perfusate urea, (E) perfusate alanine aminotranferase, (F) perfusate aspartate aminotransferase. *** indicates p ≤ 0.001. Data shown as median and interquartile range.

Fig 3. Comparison of dNEVLP groups 2.

Comparison of the three dNEVLP groups: (A) amount of bile production, (B) bile gamma-glutamyl transferase, (C) lactate dehydrogenase within bile, (D) liver parenchyma necrosis, (E) liver parenchyma sinusoidal dilatation, (F) bile duct necrosis in TUNEL staining. * indicates p ≤ 0.05, ** p ≤ 0.01 and *** p = 0.001. Data shown as median and interquartile range.

Fig 4. Histopathology.

(A-D) HE staining of liver parenchyma, (E-H) HE staining of bile duct, (I-L) TUNEL & DAPI staining of bile duct, (A, E, I) dNEVLP^-P, (B, F, J) dNEVLP^MH, (C, G, K) dNEVLP^MP, (D, H, L) sNEVLP.

Fig 5. Comparison of dNEVLP^MP and sNEVLP groups 1.

Comparison of dNEVLP^MP and sNEVLP: (A) portovenous pressure; (B) pH of perfusate, (C) urea within perfusate, (D) alanine aminotranferase within perfusate, (E) aspartate aminotransferase within perfusate. * indicates p ≤ 0.05. Data shown as median and interquartile range.

Fig 6. Comparison of dNEVLP^MP and sNEVLP groups 2.

Comparison of dNEVLP^MP and sNEVLP: (A) amount of bile production, (B) gamma-glutamyl transferase within bile, (C) lactate dehydrogenase within bile, (D) liver parenchyma necrosis, (E) liver parenchyma sinusoidal dilatation, (F) bile duct necrosis in TUNEL staining. * indicates p ≤ 0.05 and *** p = 0.001. Data shown as median and interquartile range.