Small animal ex vivo machine perfusion of the liver: A comprehensive literature review

Abstract

BACKGROUND

Liver transplantation is the only treatment for acute and chronic liver failure, but the global organ shortage has increased reliance on extended criteria donor livers, which are more susceptible to ischemia-reperfusion injury. While static cold storage is standard, these grafts often require improved preservation strategies.

AIM

To summarize the current state of small animal liver machine perfusion (MP), highlight variability in protocols, and emphasize the need for standardization to guide future research.

METHODS

A comprehensive literature search of PubMed was conducted to identify studies on small animal (rat and mouse) ex vivo liver MP. Only English-language animal studies were included, with no restrictions on publication date. Relevant full-text articles were reviewed, and reference lists were screened to ensure completeness.

RESULTS

Small animal liver MP provides a cost-effective model to explore dynamic preservation strategies. Rat perfusion studies face challenges including dual-vessel perfusion, maintaining physiological perfusate volumes, and lack of standardized protocols. Open- and closed-circuit setups have distinct advantages and limitations, and experimental designs vary widely across studies.

CONCLUSION

This review illustrates the wide variability in small animal liver MP protocols and underscores the urgent need for standardization. Addressing these inconsistencies will enhance reproducibility, facilitate comparison across studies, and support the development of optimized liver preservation strategies.

Key Words: Liver transplant; Liver transplantation; Machine perfusion; Ex vivo liver machine perfusion; Small animal machine perfusion

Core Tip: Small animal liver machine perfusion models are essential for studying dynamic liver preservation in transplantation research. Rat liver perfusion provides a cost-effective and accessible platform, but currently no standardized protocols exist, limiting reproducibility and progress. This review highlights variations in existing studies, technical challenges, and limitations, emphasizing the urgent need for standardization. By summarizing key developments and system differences, it offers researchers practical insights to optimize perfusion strategies, improve reproducibility, and reduce animal use in future studies.

INTRODUCTION

Liver transplantation (LTx) is the only possible treatment option for acute and chronic liver failure. Recent advances in LTx outcomes, immunosuppressive techniques, and cancer therapies have expanded the eligibility of individuals for inclusion in transplant waiting lists[1]. The global organ shortage obligates centers to use grafts from extended criteria donors (ECDs) including grafts donated after circulatory death (DCD)[2]. ECD livers are susceptible to greater ischemia reperfusion injury, contributing to inferior graft function and outcomes[3]. While static cold storage (SCS) has long been considered the gold standard, ECD do not fare well under this traditional preservation method. Dynamic preservation methods such as normothermic (NMP), subnormothermic (SNMP), and hypothermic machine perfusion (HMP) have emerged to enhance the viability of donor livers and optimize transplantation outcomes[4,5].

In contrast to conventional SCS, ex situ machine perfusion (MP) enhances LTx outcomes, enabling extended preservation times and viability testing[6,7]. The prospective application of MP is anticipated to be even more extensive, with currently investigated therapeutic interventions like defatting cocktails, RNA interference, senolytics, and stem cell therapy showing promise in facilitating the repair and regeneration of injured livers before LTx[8].

Large animal studies on MP are expensive and face feasibility challenges[9]. Therefore, small animal experiments are necessary to clarify potential applications in the future. Adherence to the 3R rule (replacement, reduction, refinement) is crucial for ethical animal experimentation[10]. While the demand for animal experiments in MP is significant, a universally accepted standard protocol for its implementation is currently lacking.

This literature review outlines the steps and diverse possibilities involved in establishing MP for small animal livers, aiming to assist in considering essential elements and ultimately contributing to a reduction in the number of animals used for experiments.

MATERIALS AND METHODS

The comprehensive literature search was conducted using the PubMed database. The following combination of medical subject headings and keywords with the employment of “AND” or “OR” or “NOT” Boolean operators were used: (“HMP” OR “hope” OR “NMP” OR “SNMP” OR “machine perfusion”) AND “liver perfusion” AND (“rat” OR “mice” OR “small animal”) NOT “kidneys”. Only articles written about animals and not humans were included in the search. No restrictions on publication dates were applied, allowing for the inclusion of studies from all available years. Only abstracts written in the English language were reviewed. Full-text articles were retrieved if relevant abstracts were identified. An additional manual search of the reference lists was performed to ensure the comprehensive literature search procedure. The most recent search was performed on June 23^rd, 2024. A preferred reporting items for systematic reviews and meta-analyses flow diagram of the literature search and study selection is presented in Figure 1.

Figure 1 Preferred reporting items for systematic reviews and meta-analyses flow diagram of study selection.

RESULTS

Liver preparation for MP

Several different techniques are described in literature. Before connecting the liver to the ex vivo MP circuit, a standard cold flush in situ is performed to remove blood remnants and reduce warm ischemia (WI) time[11,12]. Clear descriptions of the flushing technique are crucial for achieving favorable outcomes in liver preservation[13]. Table 1 outlines various liver flushing solutions, routes, and volumes. The administration routes for the cold flush vary among authors. Most perform the flush through the portal vein (PV) only[14-66]. Others use a dual vessel approach, flushing through both the PV and the aorta (A)[67-74] or the PV and the hepatic artery (HA)[75-77]. Flushing via the A alone is another option[78-84]. The use of HA exclusively is a rare approach, employed by only a single author[85].

Table 1 Liver graft preparation for small animal liver ex vivo machine perfusion.

Ref.	Via	Liver flush before MP
Kim et al[14]	PV	10 mL of cold UW-G solution
Dutkowski et al[15]	PV	2 mL + 3 mL + 1 mL of 4 °C UW solution (hydroxyethyl starch and glutathione free)
Compagnon et al[67]	A, PV	25 mL of ice-cold Celsior-HES solution with 500 IU of heparin
Lauschke et al[16]	PV	60 mL of HTK or Belzer MPS solutions
Lee et al[90]	NA	NA
Tan et al[91]	NA	NA
Xu et al[92]	NA	NA
Bessems et al[17]	PV	50 mL of ringer lactate
Dutkowski et al[18]	PV	2 mL + 3 mL + 1 mL of 4 °C UW solution (hydroxyethyl starch and glutathione free)
Tolboom et al[19]	PV	5 mL + 5 mL of 4 °C UW solution
Vairetti et al[20]	PV	Oxygenated KH medium
Manekeller et al[21]	PV	HTK or Belzer MPS solutions
Stegemann et al[22]	PV	20 mL of HTK or Custodiol-N solution at 4 °C
Ferrigno et al[23]	PV	Oxygenated KH medium
Lüer et al[24]	PV	20 mL of HTK solution
Olschewski et al[25]	PV	20 mL of Lifor organ preservation solution (at 4 °C, 12 °C and 21 °C)
Minor et al[26]	PV	20 mL of HTK solution
Tolboom et al[27]	PV	10 mL of saline
Giannone et al[28]	PV	20 mL of cold Celsior solution (in situ) + 30 mL of cold Celsior solution
Perk et al[29]	PV	10 mL of saline
Schlegel et al[30]	PV	6 mL of 20 °C heparinized (1 IU/mL) saline
Bruinsma et al[88]	NA	Cold UW solution
Liu et al[31]	PV	10 mL of 4 °C perfusate (control or defatting)
Carnevale et al[93]	NA	NA
Schlegel et al[32]	PV	6 mL of 20 °C heparinized (1 IU/mL) saline
Schlegel et al[33]	PV	6 mL of heparinized (1 IU/mL) saline at room temperature and UW solution at 4 °C
Bae et al[34]	PV	200 mL of 0.9% saline
Niu et al[68]	A, PV	50 mL of ice-cold ross perfusion fluid with 10 IU/mL heparin. 12 mL of cold supplemented UW solution. 20 mL of Hartmann’s solution with heparin (5 U/mL)
Tarantola et al[35]	PV	KH medium
Bruinsma et al[36]	PV	10 mL of 21 °C 3-OMG loading solution + 10 mL of 21 °C 3-OMG loading solution
Ferrigno et al[37]	PV	50 mL of modified KHB
Jia et al[38]	PV	Cooled saline containing 25 IU/mL heparin
Westerkamp et al[94]	NA	NA
Carbonell et al[95]	NA	NA
Op den Dries et al[75]	PV, HA	5 mL of 37 °C 0.9% NaCl via PV + 5 mL of 4 °C 0.9% NaCl via SIVC + 20 mL of 4 °C 0.9% NaCl via PV + 5 mL of 4 °C 0.9% NaCl via HA
Okamura et al[39]	PV	50 mL of preservation solution
Berardo et al[40]	PV	Ringer lactate
Chai et al[89]	NA	Cold UW solution until liver color changed to khaki
Zeng et al[41]	PV	20 mL of 4 °C HTK solution
Beal et al[42]	PV	60 mL of cold 0.9% saline with 1 mL heparin (100 U)
Tabka et al[69]	A, PV	15 mL of 20 °C Ringer’s lactate or Celsior or with Celsior supplemented with 10-9 M of Ang IV
Xue et al[43]	PV	20 mL of 4 °C HTK solution
He et al[78]	A	Approximately 10 mL of 0-4 °C saline
Gassner et al[70]	A, PV	20 mL of 4 °C HTK solution (with/without 12 mmol/L glycine)
Zeng et al[79]	A	2 mL of cold HTK solution with 10 IU/mL heparin
Jia et al[80]	A	Cold saline with 25 IU/mL of heparin
Oldani et al[44]	PV	20 mL of cold IGL-1 solution with 100 IU heparin
Chin et al[45]	PV	50 mL of 4 °C 0.9% NaCl
Gillooly et al[46]	PV	10 CC of cold saline with 100 IU heparin
Martins et al[47]	PV	4 °C Celsior solution
Scheuermann et al[48]	PV	40 mL of cold UW solution
Claussen et al[71]	A, PV	20 mL of 4 °C HTK solution supplemented with 12 mmol/L glycine
Haque et al[49]	PV	50 mL of ice cold Ringer’s lactate with heparin
Schlegel et al[50]	PV	10 mL of cold heparinized IGL-1 solution
Nösser et al[51]	PV	20 mL of Ringer solution
Yamada et al[96]	NA	NA
Hu et al[52]	PV	20 mL of 0-4 °C HTK solution
Von Horn and Minor[53]	PV	60 mL of HTK solution
Raigani et al[54]	PV	50 mL of ice-cold 0.9% saline
Yang et al[55]	PV	2 mL of saline
Westerkamp et al[76]	PV, HA	10 mL of 37 °C 0.9% NaCl via PV. 5 mL of 4 °C HTK solution via PV. 20 mL of 4 °C HTK solution via PV. 5 mL of 4 °C HTK solution via HA
De Vries et al[86]	NA	60 mL of ice-cold saline
Liu et al[85]	HA	Heparin saline 40 mL (50 IU/L)
Lin et al[56]	PV	NA
Rigo et al[57]	PV	10 mL of cold Celsior solution. 30 mL of cold Celsior solution. 10 mL of WEM
Xu et al[97]	NA	NA
Carlson et al[58]	PV	20 mL of NaCl saline
Zhou et al[81]	A	50 mL of 4 °C HTK solution
Cao et al[59]	PV	10 mL of UW solution
De Stefano et al[60]	PV	10 mL of saline or Celsior solution
Sun et al[61]	PV	NA
Jennings et al[62]	PV	20 mL of cold 0.9% saline
Wang et al[72]	A, PV	3 mL of cold HTK solution
Asong-Fontem et al[73]	A, PV	50 mL of preservation solution. 20 mL of preservation solution at 5 °C ± 3 °C
Shi et al[77]	PV, HA	0-4 °C heparinized saline
Von Horn et al[63]	PV	60 mL of 4 °C HTK solution
Zhou et al[82]	A	50 mL of 4 °C HTK solution
Luo et al[83]	A	40 mL of heparinized saline (50 IU/mL)
Ohara et al[64]	PV	Ringer’s solution
Chen et al[65]	PV	5 mL of 37 °C heparinized saline (2500 IU/mL)
Fukai et al[74]	A, PV	60 mL of ice-chilled saline 25 mL of ice-chilled Belzer UW or Belzer MPS solutions
Hughes et al[84]	A	80 mL of 4 °C HTK solution
Bai et al[98]	NA	NA
Von Horn et al[66]	PV	60 mL of 4 °C HTK solution
Li et al[87]	NA	NaCl solution

A: Aorta; NA: Not written or not applicable in the main article; HA: Hepatic artery; UW: University of Wisconsin; UW-G: University of Wisconsin-gluconate; HTK: Histidine Tryptophan Ketoglutarate; HES: Hydroxyethyl starch; Belzer MPS: Belzer machine perfusion solution; KH: Krebs Henseleit; KHB: Krebs Henseleit bicarbonate; 3-OMG: 3-O-methyl-D-glucose; NaCl: Sodium chloride; PV: Portal vein; SIVC: Suprahepatic inferior vena cava; Ang IV: Angiotensin IV; IGL: Institute George Lopez; WEM: William’s E Medium; MP: Machine perfusion.

The solutions used for cold flushes vary significantly, with volumes ranging from 3 mL to 80 mL. Some researchers choose simple heparinized saline, which is a standard choice for removing blood remnants[80,83,85-87]. In contrast, others use more advanced preservation solutions such as University of Wisconsin (UW), Histidine Tryptophan Ketoglutarate (HTK), Krebs Henseleit, Belzer MP Solution (Belzer MPS), 3-O-methyl-D-glucose, Institute George Lopez (IGL)-1, and William’s E Medium (WEM)[66-73,79,81,82,84,88,89]. These more complex solutions are often chosen because they are also used during the ex vivo MP process, potentially offering better preservation properties.

Additionally, the temperature at which these solutions are administered can vary. Many protocols employ solutions cooled to 0-4 °C to minimize metabolic activity. Others use solutions at higher temperatures, such as 20 °C or even 37 °C[30,33,36,65,75]. Some authors perform the flush with a solution at the same temperature that will be used for the subsequent MP[25]. While cold flush has long been standard procedure before MP, controversially, some suggest that a cold flush before NMP subjects grafts to higher WI damage[12]. Some studies also incorporate oxygenated solutions, which might affect oxygen delivery and tissue viability during the flushing process[20,23].

While authors suggest different solutions and volumes for cold flush, it is important that the pressure during the cold flush for small animal livers is not too high and that adequate blood remnant washout is achieved. Some authors described using constant flow for the cold flush[67], others performed the cold flush until the liver color changed to khaki[89], while the majority applied different volumes and performed it multiple times, both in situ and ex vivo[90-98]. However, some studies do not detail any flushing procedure before applying MP, and future studies should not omit the description of this crucial step.

Heparin administration

Heparin, an essential anticoagulant that prevents clotting, is crucial for use during rat liver explant surgery[99,100]. Administering heparin before liver perfusion in the donor and/or into the perfusion solution aids in optimizing liver harvesting[101]. Authors propose diverse routes and dosages for heparin administration, noting its non-liver-toxic effect in rats[102]. Several studies have delved into the administration of heparin via the vena cava (VC), with dosages spanning from 250 IU to 5000 IU[17,20,23,28,31,35,37,40,45,48,54,56]. While other suggested heparinization via iliac vein or abdominal A[41,43,52,78,80]. Alternative routes, including the tibial vein and dorsal penile vein, were explored in other investigations, employing heparin dosages ranging from 100 IU to 500 IU[59,68,75,76]. Moreover, some researchers chose to apply heparin via intraperitoneal administration, dosages ranged from 1000 IU to 1500 IU[60,79]. Various techniques have been employed for administering heparin during liver flushing via the PV, with some studies opting to incorporate heparin into the perfusate solutions[30,42,50,65,77]. These findings show the variability in heparin administration protocols and highlight the importance of further research to establish standardized guidelines for optimal anticoagulation during liver explant surgery and MP.

Cannulation

Anatomically, small animal livers are perfused by proportionally less arterial and more portal blood compared to large animals[103]. Thus, a significant number of authors prefer single-vessel MP via the PV, which necessitates simpler surgery and a more cost-effective MP circuit. Oxygenated MP through the PV provides oxygen and perfusate to the entire graft, including the extrahepatic biliary tree[104]. In contrast, other groups argue that although the PV supplies nutrients to hepatocytes and maintains a higher flow rate compared to the HA, it does not serve as the liver’s primary route for oxygen delivery and does not support the vascularization of the biliary tree as effectively as the HA does[105]. In addition, when selecting a perfusion circuit among the PV, VC, and HA, it’s noted that retrograde perfusion is comparable to PV perfusion, while perfusion via the HA is considered less advantageous[67]. Furthermore, concerns arise regarding the potential direct damage caused by arterial cannulation to the arterial intima, which could compromise vascular anastomosis[106].

However, some authors advocate for dual vessel MP, asserting that it results in superior outcomes compared to single-vessel MP[71]. Although HA cannulation is a complicated procedure, scientists often opt to cannulate via the celiac artery due to its larger diameter[39,84].

To establish a closed MP circuit, the VC must be cannulated, with the other VC outflow either ligated or sutured. Alternatively, the perfusate can flow freely via the infrahepatic and/or suprahepatic VC into the organ chamber, where it is immediately recirculated inside the system[20,23,35].

Bile duct cannulation is essential for measuring bile output. Moreover, it allows to test bile composition which is a great marker of biliary viability[107]. While some authors provide detailed descriptions of the bile duct cannulation process, others may omit it entirely. Nevertheless, it is crucial not to overlook this step. Bile flow is heavily influenced by perfusion temperature and oxygen delivery rate. At 37 °C and adequate oxygenation, bile flow should be at least 1 μL/minute/g liver[108]. Table 2 summarizes which vessels (PV, HA, VC) and common bile duct cannulations were performed by different authors in their MP settings.

Table 2 Cannulation techniques for small animal liver ex vivo machine perfusion.

Ref.	Portal vein	Hepatic artery	Vena cava	Common bile duct
Kim et al[14]; Manekeller et al[21]; Stegemann et al[22]; Giannone et al[28]; Perk et al[29]; Bae et al[34]; Zeng et al[79]; Jia et al[80]; Gillooly et al[46]; Martins et al[47]; Haque et al[49]; Schlegel et al[50]; Cao et al[59]; Wang et al[72]; Asong-Fontem et al[73]; Li et al[87]	Yes	No or NA	No or NA	No or NA
Dutkowski et al[15]; Lee et al[90]; Bessems et al[17]; Olschewski et al[25]; Tolboom et al[27]; Bruinsma et al[88]; Carnevale et al[93]; Niu et al[68]; Bruinsma et al[36]; Zeng et al[41]; Beal et al[42]; Tabka et al[69]; Xue et al[43]; Gassner et al[70]; Oldani et al[44]; Nösser et al[51]; Hu et al[52]; Lin et al[56]; Xu et al[97]; Sun et al[61]	Yes	No or NA	Yes	Yes
Compagnon et al[67]; Westerkamp et al[94]; Op den Dries et al[75]; Okamura et al[39]; Claussen et al[71]	Yes	Yes	Yes	Yes
Lauschke et al[16]; Xu et al[92]; Dutkowski et al[18]; Tolboom et al[19]; Vairetti et al[20]; Ferrigno et al[23]; Lüer et al[24]; Minor et al[26]; Liu et al[31]; Tarantola et al[35]; Ferrigno et al[37]; Berardo et al[40]; He et al[78]; Chin et al[45]; Scheuermann et al[48]; Yamada et al[96]; Von Horn and Minor[53]; Raigani et al[54]; Yang et al[55]; De Vries et al[86]; Rigo et al[57]; Carlson et al[58]; De Stefano et al[60]; Jennings et al[62]; Von Horn et al[63]; Luo et al[83]; Chen et al[65]; Fukai et al[74]; Von Horn et al[66]	Yes	No or NA	No or NA	Yes
Tan et al[91]; Carbonell et al[95]; Chai et al[89]	No or NA	No or NA	No or NA	No or NA
Schlegel et al[30]; Schlegel et al[32]; Zhou et al[81]; Zhou et al[82]	Yes	No or NA	Yes	No or NA
Schlegel et al[33]	Yes	Yes	Yes	No or NA
Jia et al[38]; Shi et al[77]	Yes	Yes	No or NA	No or NA
Westerkamp et al[76]; Liu et al[85]; Ohara et al[64]; Hughes et al[84]; Bai et al[98]	Yes	Yes	No or NA	Yes

This table shows which vessels (vena cava, hepatic artery, portal vein) and the common bile duct were cannulated in various studies, with “Yes” denoting cannulation and “No” or “NA” indicating absence or non-application. NA: Not written or not applicable in the main article.

MP applications

Scientists have studied dynamic preservation methods under different durations and temperatures, with varying warm and cold ischemia times before applying MP. As there are still no standard times for the MP application, various studies demonstrate possible outcomes under different durations ranging from 30 minutes to 120 hours (Table 3). Some authors aim to minimize MP time which results in improved liver function[50,72,81,95,96], while others seek to extend MP time as much as possible to prolong preservation[24,26,36,49,88]. Additionally, some researchers focus on determining the optimal time for MP application[52].

Table 3 Experimental conditions and main findings in small animal liver ex vivo machine perfusion studies.

Ref.	Animal	WI time before MP	Length of SCS before MP	MP type and length of MP
Kim et al[14]	Rats	NA	0 hour	HMP: 48 hours
Dutkowski et al[15]	Rats	11.4 minutes 0.8 minutes or 4.2 minutes 0.4 minutes	NA	HMP: 10 hours
Compagnon et al[67]	Rats	< 20 seconds	NA	HMP: 24 hours or 48 hours; NMP: 2 hours
Lauschke et al[16]	Rats	60 minutes	NA	HMP: 24 hours; NMP (reperfusion): 45 minutes
Lee et al[90]	Rats	30 minutes	NA	HMP: 5 hours
Tan et al[91]	Rats	NA	30 minutes	HMP: 36 hours
Xu et al[92]	Rats	NA	0 hour (before HMP) or 24 hours (before NMP)	HMP: 24 hours; NMP: 1 hour
Bessems et al[17]	Rats	NA	NA	HMP: 24 hours; NMP (reperfusion): 1 hour
Dutkowski et al[18]	Rats	NA	10 hours	HOPE: 3 hours; NMP (reperfusion): 40 minutes
Tolboom et al[19]	Rats	NA	0 hour	NMP: 6 hours
Vairetti et al[20]	Rats	NA	NA	4 °C, 10 °C, 20 °C, 25 °C, 30 °C, or 37 °C MP: 6 hours, NMP (reperfusion): 2 hours
Manekeller et al[21]	Rats	30 minutes	NA	HMP: 18 hours; NMP (reperfusion): 2 hours
Stegemann et al[22]	Rats	30 minutes	NA	HMP: 18 hours; NMP (reperfusion): 2 hours
Ferrigno et al[23]	Rats	NA	NA	HMP or SNMP: 6 hours; NMP (reperfusion): 2 hours
Lüer et al[24]	Rats	NA	NA	HMP: 18 hours; NMP (reperfusion): 2 hours
Olschewski et al[25]	Rats	60 minutes	NA	HMP or SNMP: 6 hours; NMP (reperfusion): 6 hours
Minor et al[26]	Rats	30 minutes	NA	HMP: 18 hours; NMP (reperfusion): 2 hours
Tolboom et al[27]	Rats	1 hour	NA	SNMP: 5 hours; NMP: 5 hours
Giannone et al[28]	Rats	NA	NA	Normobaric or hyperbaric HMP: 24 hours
Perk et al[29]	Rats	60 minutes or 90 minutes	NA	NMP: 6 hours
Schlegel et al[30]	Rats	30 minutes	4 hours	HMP: 1 hour
Bruinsma et al[88]	Rats	NA	0 hour, 24 hours, 48 hours, 72 hours and 120 hours	SNMP: 3 hours
Liu et al[31]	Rats	NA	NA	SNMP: 6 hours
Carnevale et al[93]	Rats	45 minutes	NA	HMP: 24 hours; NMP (reperfusion): 1.5 hours
Schlegel et al[32]	Rats	NA	30 minutes	HMP or HNP: 1 hour
Schlegel et al[33]	Rats	30 minutes or 60 minutes	NMP: 0 minute or 15 minutes; HOPE: 4 hours	NMP: 4 hours; HOPE: 1 hour
Bae et al[34]	Rats	30-40 minutes	NA	HMP: 8 hours; NMP (reperfusion): 90 minutes
Niu et al[68]	Rats	60 minutes	5 hours	NMP: 2 hours
Tarantola et al[35]	Rats	NA	6 hours	SNMP, NMP (reperfusion): 6 hours
Bruinsma et al[36]	Rats	NA	NA	Loading SNMP: 80 minutes; supercooling: 3-4 days; recovery SNMP: 5 hours
Ferrigno et al[37]	Rats	NA	NA	10 °C, 20 °C, 30 °C or 37 °C MP: 6 hours
Jia et al[38]	Rats	NA	0 hour or 6 hours	HMP: 6 hours or 0 hour
Westerkamp et al[94]	Rats	30 minutes	6 hours	HMP or SNMP or rewarming MP: 1 hour; NMP: 2 hours
Carbonell et al[95]	Rats	NA	NA	NMP: 15 minutes to stabilize (all groups); SNMP or NMP: 30 minutes
Op den Dries et al[75]	Rats	0 minute or 30 minutes	0 hour or 3 hours	NMP: 3 hours; NMP (reperfusion): 2 hours
Okamura et al[39]	Rats	NA	NA	SNMP: 4 hours; NMP: 2 hours
Berardo et al[40]	Rats	NA	6 hours	SNMP: 6 hours; NMP: 2 hours
Chai et al[89]	Rats	NA	NA	HMP: 2 hours or 12 hours
Zeng et al[41]	Rats	30 minutes	NA	HMP: 3 hours; NMP (reperfusion): 2 hours
Beal et al[42]	Rats	NA	NA	NMP: 4 hours
Tabka et al[69]	Rats	NA	NA	SNMP: 6 hours; NMP (reperfusion): 2 hours
Xue et al[43]	Rats	30 minutes	NA	HMP: 3 hours; NMP: 1 hour
He et al[78]	Rats	NA	NA	HMP: 3 hours or 6 hours
Gassner et al[70]	Rats	30 minutes	NA	NMP: 6 hours
Zeng et al[79]	Mice	30 minutes	4 hours	HOPE or HNPE: 1 hour; NMP (reperfusion): 2 hours
Jia et al[80]	Rats	NA	NA	HMP: 6 hours
Oldani et al[44]	Rats	1 hour	30 minutes	HOPE or NMP: 2 hours
Chin et al[45]	Rat	NA	NA	NMP: 6 hours
Gillooly et al[46]	Rats	25 minutes	NA	HMP, NMP: 4 hours
Martins et al[47]	Rats	NA	12 hours	NMP: 1 hour
Scheuermann et al[48]	Rats	NA	0 hour	SNMP or NMP: 4 hours; NMP (reperfusion): 2 hours
Claussen et al[71]	Rats	< 15 minutes	< 60 minutes	NMP: 6 hours
Haque et al[49]	Rats	NA	NA	NMP: 24 hours
Schlegel et al[50]	Rats	30 minutes	4 hours	NMP or HOPE: 1 hour
Nösser et al[51]	Rats	NA	81.71 minutes ± 28.44 minutes	SNMP or NMP: 6 hours or 12 hours
Yamada et al[96]	Rats	30 minutes	6 hours	SNMP or NMP: 30 minutes, 60 minutes or 90 minutes
Hu et al[52]	Rats	30 minutes	NA	HMP: 1 hour, 3 hours, 4 hours, 12 hours, 24 hours; NMP: 2 hours
Von Horn and Minor[53]	Rats	20 minutes	18 hours	NMP: 2 hours
Raigani et al[54]	Rats	NA	< 10 minutes	NMP: 6 hours
Yang et al[55]	Rats	30 minutes	NA	NMP: 8 hours
Westerkamp et al[76]	Rats	NA	4 hours	NMP: 3 hours
De Vries et al[86]	Rats	0 h	0 hour or 24 hours or 72 hours	SNMP: 3 hours
Liu et al[85]	Rats	0 minute, 10 minutes or 30 minutes	NA	HMP or NMP: 4 hours
Lin et al[56]	Rats	NA	1 hour	HMP: 3 hours
Rigo et al[57]	Rats	3.10 minutes (0.35) mean (SEM)	NA	NMP: 4 hours
Xu et al[97]	Rats	NA	NA	NMP: 4 hours
Carlson et al[58]	Rats	NA	< 5 minutes	NMP: 4 hours
Zhou et al[81]	Rats	30 minutes	23 hours	HOPE: 1 hour; NMP: 1 hour
Cao et al[59]	Rats	30 minutes	NA	NMP: 4 hours
De Stefano et al[60]	Rats	60 minutes	No or 34 minutes ± 7 minutes	NMP: 6 hours
Sun et al[61]	Rats	30 minutes	NA	NMP: 6 hours
Jennings et al[62]	Rats	NA	< 5 minutes	NMP: 4 hours
Wang et al[72]	Mice	10 minutes	11 hours	HOPE: 1 hour; NMP (reperfusion): 2 hours
Asong-Fontem et al[73]	Rats	NA	24 hours	HOPE: 2 hours; NMP (reperfusion): 2 hours
Shi et al[77]	Rats	30 minutes	8 hours	NMP: 2 hours
Von Horn et al[63]	Rats	20 minutes	17 hours or 18 hours	HMP: 1 hour; NMP (reperfusion): 2 hours
Zhou et al[82]	Rats	30 minutes	0 hour or 23 hours	HOPE: 1 hour; NMP: 1 hour
Luo et al[83]	Rats	30 minutes	3 hours or 4 hours	HMP or HOPE: 1 hour; NMP: 2 hours
Ohara et al[64]	Rats	60 minutes	NA	NMP: 4 hours
Chen et al[65]	Mice	NA	NA	NMP: 12 hours
Fukai et al[74]	Rats	30 minutes	NA	HMP: 3 hours; NMP (reperfusion): 90 minutes
Hughes et al[84]	Rats	30 minutes	During MP priming	NMP: 4 hours
Bai et al[98]	Rats	30 minutes	8 hours	NMP: 2 hours
Von Horn et al[66]	Rats	20 minutes	18 hours	Rewarming MP: 2 hours; NMP: 1 hour
Li et al[87]	Rats	NA	NA	NMP: 3 hours

WI: Warm ischemia; SCS: Static cold storage; MP: Machine perfusion; HMP: Hypothermic machine perfusion; HOPE: Hypothermic oxygenated machine perfusion; NMP: Normothermic machine perfusion; NA: Not written or not applicable in the main article; HNPE: Hypothermic deoxygenated (nitrogenated) perfusion; SNMP: Subnormothermic machine perfusion.

Ex vivo rat liver MP studies are performed for a number of different reasons. They compare dynamic preservation techniques with SCS, especially concerning older or fatty livers[67,75,78,90,93,94]. Comparisons between dynamic preservation methods aim to identify the most effective approach[33,37,50,63,96]. Some authors study the role of oxygenation in liver preservation, providing evidence of its advantageous outcomes in improving liver function[22,24]. Researchers investigate various perfusates[17,21,92], including additives such as tacrolimus[32], α-tocopherol[34], pegylated-catalase[42], dopamine[26], angiotensin IV[69], metamizole[71], metformin[76], oxygen carriers[62], and IGL-2[73], to understand how they impact liver function. Some studies explore the introduction of stem cells to improve liver function[55,59-61], while others aim to treat fatty liver disease during MP[23,31,35,39,54,56,72,87,97]. Additionally, efforts are made to enhance donation after DCD liver function[81-85,94,96]. Moreover, there are promising initial attempts to apply gene therapy during MP[46].

MP system

A standard perfusion system is made of organ chamber, perfusate reservoir, peristaltic pump, heat exchanger, bubble trap, oxygenator and tubing[65] (Figures 2 and 3). Various MP systems have been described, ranging from simpler self-made setups to more complex designs[70-110]. The organ chamber lies at the center of the MP system, where the liver is carefully positioned hilum facing upwards. Some authors suggest using an elastic pillow to aid in positioning the liver[51]. The perfusate reservoir, connected to the chamber, holds the perfusion solution to be pumped through the organ. Alternatively, the organ chamber could also serve as perfusate reservoir[19]. The peristaltic pump regulates the flow rate of the perfusate, ensuring precise delivery to the organ. When perfusion is performed via both PV and HA, two peristaltic pumps are necessary[11,75]. Additionally, controlling temperature is crucial, often achieved through the heat exchanger, thermostat[76], heating water bath with temperature sensor[41,52], combined heat exchanger-oxygenator[54], and other alternatives, maintaining the perfusate at the desired temperature throughout the process. The oxygenator enriches the solution with oxygen to sustain the metabolic needs of the organ. Various types of oxygenators, including membrane[29,47,55,59,61,75,76,84,88], hollow-fiber[32,52,57,77], bubble[93], tubing[15,53,58] and others, are discussed by the authors. The bubble trap removes air bubbles from the perfusate and prevents air embolism. Tubing connects these components, allowing for the seamless flow of the perfusate through the system. The implementation of sample ports is also crucial for obtaining perfusate samples during the MP process[93]. Monitoring MP parameters is also highly significant. Some researchers choose basic manometers to measure pressure[105], while others prefer sophisticated equipment equipped with diverse sensors and data acquisition devices, enabling real-time analysis and display of parameters[75]. Some researchers propose incorporating a dialysis unit into the MP circuit, which could potentially improve preservation outcomes during NMP[27,51].

Figure 2 Dual-vessel closed machine perfusion circuit. Perfusate is stored in the perfusate reservoir where the targeted temperature is achieved through tubing connected to the thermos unit (e.g., water bath). The perfusate is oxygenated via an oxygenator connected to a gas tank. Oxygenated perfusate, maintained at a specific temperature, is pumped through roller pumps controlled by flow and pressure regulators. Upon passing through the pumps, the perfusate goes through a heat exchanger and bubble traps before reaching its final destination: The portal vein and hepatic artery. Exiting the liver via the infrahepatic vena cava, the perfusate returns to the perfusate reservoir through tubing. The common bile duct is cannulated, and the catheter is connected to the tube for bile collection. HA: Hepatic artery; PV: Portal vein.

Figure 3 Single-vessel machine perfusion circuit. The liver is placed inside an organ chamber, which serves as a perfusate reservoir. The perfusate is oxygenated via an oxygenator, and the targeted temperature is achieved via a thermos unit. The perfusate from the organ chamber is then taken up by a roller pump, which is controlled by a flow and pressure controller. After passing through the roller pump, the perfusate goes through a bubble trap and into the portal vein. Upon perfusing the liver, the perfusate exits the liver freely via the vena cava. The common bile duct is cannulated, and the catheter is connected to the tube for bile collection. PV: Portal vein.

Perfusate volume and composition

According to the literature, the volume of perfusate used in small animal MP can range from 2 mL to 500 mL, as detailed in Table 4. Although there are no specific recommendations regarding the total perfusion volume, some authors propose reducing it to 50 mL. Notably, when the perfusate includes red blood cells, reducing the volume has been shown to increase hematocrit levels and decrease the release of transaminases[51].

Table 4 Perfusate compositions in small animal liver ex vivo machine perfusion studies.

Ref.	Animal weight	Liver weight	Perfusate volume	Perfusate composition
Kim et al[14]	200-400 g	NA	200 mL	Cold UW-G solution
Dutkowski et al[15]	250-300 g	10.4 ± 0.3 g	500 mL	Modified UW solution: Starch and glutathione was omitted, supplemented with 80 mg/L gentamycin, and 5000 IU/L heparin
Compagnon et al[67]	300 ± 50 g	NA	150 mL	HMP: Celsior-HES solution. NMP: KHB buffer with 5% bovine serum albumin
Lauschke et al[16]	250-300 g	NA	125 mL	HMP: HTK or Belzer MPS solution supplemented with 6000 IU superoxide dismutase. NMP: KH buffer
Lee et al[90]	200-250 g	NA	100 mL	KH solution, then switched to UW solution (starch omitted)
Tan et al[91]	180-220 g	NA	120 mL	Modified Hoffmann perfusate: Hydroxyethyl starch 50 g/L, calcium gluconate 80 mmol/L, raffinose 10 mmol/L, KH2PO4 25 mmol/L, hydroxyethyl piperazine 10 mmol/L, dexamethasone 12 mg/L, penicillin 2 × 105 units/L, insulin 100 units/L, and with/without 25 mmol/L MgCl and/or 5 mmol/L ATP
Xu et al[92]	200-250 g	NA	NA	HMP: UW solution with/without starch. NMP: KH buffer solution containing 112 μmol/L taurocholic acid, and 150 μg/L hyaluronic acid
Bessems et al[17]	350 ± 50 g	16.5 ± 0.5 g	250 mL	HMP: The UW-G solution or polysol. NMP: KHB without bovine serum albumin
Dutkowski et al[18]	250-300 g	10.8 ± 1.4 g	450 mL	HOPE: Modified starch-free UW solution. NMP: KHB buffer
Tolboom et al[19]	250-300 g	9.74-0.81 g	55-60 mL	NMP: Phenol red-free WEM supplemented with 2 IU/L insulin, 40000 IU/L penicillin, 40000 mg/L streptomycin, 0.292 g/L L-glutamine, 10 mg/L hydrocortisone, 1000 IU/L heparin with 25% (v/v) freshly isolated rat plasma and freshly isolated rat erythrocytes to a hematocrit of 16% to 18%
Vairetti et al[20]	250-300 g	NA	200 mL	KH solution with 1.25 mmol/L CaCl2 or with 0.25 mmol/L CaCl2
Manekeller et al[21]	250-300 g	NA	125 mL	HMP: HTK or Blezer MPS solutions. NMP: KH buffer
Stegemann et al[22]	250-300 g	NA	HMP: 125 mL; NMP: 250 mL	HMP: HTK or modified HTK solution (custodiol-N). NMP: KH buffer with 3 g/100 mL of bovine serum albumin
Ferrigno et al[23]	375 ± 15 g and 300 ± 10 g	NA	200 mL	KH medium
Lüer et al[24]	250-300 g	NA	125 mL	HMP: HTK solution. NMP: WEM solution, supplemented with 3 mg/100 mL of bovine serum albumin
Olschewski et al[25]	250-280 g	NA	NA	HMP, SNMP: Lifor solution. NMP: KH solution
Minor et al[26]	250-300 g	NA	HMP: 125 mL; NMP: 250 mL	HMP: HTK solution supplemented with 0 μmol/L, 10 μmol/L, 50 μmol/L or 100 μmol/L of dopamine. NMP: WEM solution with 3 mg/100 mL of bovine serum albumin
Tolboom et al[27]	250-300 g	NA	55-60 mL	WEM with autologous erythrocytes and plasma. To this were added: Insulin (2 IU/L), penicillin (40000 IU/L)/streptomycin (40000 μg/L), L-glutamine (0.292 g/L), hydrocortisone (10 mg/L), and heparin (1000 IU/L)
Schlegel et al[30]	250-300 g	9.83 ± 0.95 g	50 mL	Modified starch free UW-solution
Giannone et al[28]	250-300 g	NA	NA	Celsior solution
Perk et al[29]	200-300 g	NA	55-60 mL	Phenol red-free WEM supplemented with 2 IU/L insulin (28.85 units/mg), 100000 IU/L penicillin, 100 mg/L streptomycin sulfate, 0.292 g/L L-glutamine, 10 mg/L hydrocortisone, and 1000 IU/L heparin. Fresh rat plasma (25% v/v) and erythrocytes (18%-20% v/v) were collected and added to the perfusate
Bruinsma et al[88]	180-250 g	NA	500 mL	WE supplemented with insulin (2 IU/L), penicillin (40000 IU/L)/streptomycin (40000 μg/L), L-glutamine (0.292 g/L), hydrocortisone (10 mg/L)
Liu et al[31]	NA	21.4 ± 3.6 g (control) 22.9 ± 5.8 g (defatting group)	200 mL	The control perfusate: Minimum essential medium supplemented with 3% wt/vol bovine serum albumin, 1.07 mmol/L lactic acid, and 0.11 mmol/L pyruvic acid. The defatting perfusate: The control perfusate supplemented with the 6 defatting agents (forskolin, GW7647, scoparone, hypericin, visfatin, and GW501516)
Carnevale et al[93]	250-300 g	NA	250 mL	HMP: HTK solution: 100 mmol/L sodium gluconate, 7 mmol/L potassium gluconate, 20 mmol/L sucrose, 30 mmol/L BES, 2.5 mmol/L KH2PO4, and 0.03 mmol/L polyethylene glycol (35 kDa), 5 mmol/L MgSO4, 3 mmol/L glutathione, 5 mmol/L adenosine, and 15 mmol/L glycine, together with 0.25 mg/mL streptomycin and 10 IU/mL penicillin G. NMP: KH buffer with 4% dextran added
Schlegel et al[32]	250-320 g	9.7 ± 1.5 g	50 mL	Modified starch free UW-solution
Schlegel et al[33]	250-320 g	10.14 ± 2.73 g	50 mL	NMP: Diluted full blood or leukocyte and platelet depleted blood perfusate. HOPE: Modified starch-free UW solution
Bae et al[34]	300 ± 25 g	NA	100 mL	HMP: KPS-1 solution (Identical to Belzer’s UW MPS) or KPS-1, enhanced with α-ketoglutarate, L-arginine, N-acetylcysteine, nitroglycerin, and prostaglandin E1 or KPS-1, with 5.4 × 10-2 mmol/L of α-tocopherol diluted in acetone. NMP: KHB buffer
Niu et al[68]	261 ± 4 g	NA	353 ± 11 mL	KH buffer
Tarantola et al[35]	375 ± 15 g and 300 ± 10 g	NA	200 mL	KH medium
Bruinsma et al[36]	250-300 g	Approximately 10 g	100 mL	3-OMG loading solution: 500 mL phenol-red free WEM, 5 mL of 200 mmol/L L-glutamine (0.292 mg/L), 4 mL of penicillin-streptomycin (5000 IU/mL), 5 mg of hydrocortisone, 5000 U of sodium heparin, 375 U of insulin, 19.42 g of 3-O-methyl glucose (0.2 M)
Ferrigno et al[37]	250-300 g	NA	NA	A modified KH buffer
Jia et al[38]	250-300 g	NA	60 mL	UW or HTK solutions
Westerkamp et al[94]	290-320 g	NA	100 mL	HMP, SNMP, rewarming MP: Belzer MPS. NMP: 25 mL of human red blood cell concentrate (final hematocrit 25%), 53.9 mL of WEM solution, 20 mL of human albumin (200 g/L), 1 mL of insulin (100 IU/mL), and 0.1 mL of unfractionated heparin (5000 IU/mL)
Carbonell et al[95]	225-250 g	NA	NA	KH buffer: 118 mmol/L NaCl, 4.7 mmol/L KCl, 1.2 mmol/L MgSO4, 1.2 mmol/L KH2PO4, 2.5 mmol/L CaCl2, 25 mmol/L NaHCO3, 20 mmol/L HEPES
Op den Dries et al[75]	303 ± 4 g (mean ± SEM)		100 mL	20 mL of human red blood cell concentrate (final hematocrit 15%-20%), 59 mL of WEM solution, 20 mL of human albumin (200 g/L), 1 mL of insulin (100 IU/mL), and 0.1 mL of unfractionated heparin (5000 IU/mL)
Okamura et al[39]	250-300 g	NA	300 mL	SNMP: Polysol solution. NMP: KH buffer
Berardo et al[40]	250-300 g	NA	NA	KH buffer
Chai et al[89]	250-300 g and 600-630 g	NA	80 mL	UW solution with/without 0.165 mg/L of metformin
Zeng et al[41]	250-300 g	NA	150 mL	HTK solution
Beal et al[42]	250-350 g	NA	300 mL	86 mL of 25% albumin, 184 mL of WEM, 30 mL of penicillin/streptomycin (10 IU/mL penicillin and 0.01 mg/mL streptomycin), insulin (50 IU/L), heparin (0.01 IU/mL), L-glutamine (0.292 g/L), and hydrocortisone (0.010 g/L). Addition of 625 IU/mL pegylated-catalase
Tabka et al[69]	250-300 g	NA	SNMP: 150 mL; NMP: 140 mL	SNMP: Celsior with/without Ang IV. NMP: KBB enriched with 5% albumin
Xue et al[43]	250 ± 10 g	NA	HMP: 100 mL; NMP: 250 mL	HMP: HTK solution. NMP: KH buffer with 4% dextran
He et al[78]	250-300 g	NA	NA	HTK solution
Gassner et al[70]	280-350 g	13.6 g ± 2.15 g	50 mL	NMP: Low-glucose DMEM supplemented with the rat erythrocyte concentrate and 12.5 mL strain specific rat plasma
Zeng et al[79]	20-24 g	NA	50 mL	HTK solution
Jia et al[80]	250-300 g	NA	60 mL	HTK solution
Oldani et al[44]	176 (155-193) g	NA	50 mL	HOPE: IGL-1 solution with 150 IU heparin. NMP: 25 mL Fischer rat blood with 25 mL normal saline, and with 150 IU heparin
Chin et al[45]	200-250 g	NA	150 mL	DMEM supplemented with 200 mmol/L L-glutamine, 10% v/v FBS, 5% with bovine serum albumin, 8 mg/L dexamethasone, 2000 IU/L heparin, 2 IU/L insulin, and 5 × 106 engineered rat fibroblasts
Gillooly et al[46]	Approximately 300 g	NA	100 mL	99 mL WEM with 10 IU insulin, and lipid nanoparticles (50 nM siRNA)
Martins et al[47]	320-350 g	NA	NA	50% plasma-Lyte and 50% KH solution
Scheuermann et al[48]	296 g ± 8 g (mean ± SEM)	NA	100 mL	SNMP, NMP: 80 mL of WEM supplemented with 5% bovine serum albumin, 20 mL of type O + human RBCs, 0.2 IU insulin, 29.2 mg L-glutamine, 1 mg hydrocortisone, and 500 IU heparin. NMP: KH buffer supplemented with 5% bovine serum albumin
Claussen et al[71]	280-350 g	NA	50 mL	10 mL of the erythrocyte concentrate, 35 mL of DMEM, 5 mL of strain specific rat plasma supplemented with 1000 IU heparin and 12 mmol/L glycine
Haque et al[49]	250-300 g	NA	500 mL	NMP: 950 mL of William’s E media, 20 g of bovine serum albumin, 20 g of polyethylene glycol 35000, 20 mg of dexamethasone, 2 mL of heparin, 1 mL of regular insulin, 10 mL penicillin-streptomycin, 10 mL of antibiotic-antimycotic, and 2.2 g of sodium bicarbonate
Schlegel et al[50]	250-320 g	9.8 0.6 g	100 mL	HOPE: Belzer MPS. NMP: Belzer MPS or diluted heparinized blood
Nösser et al[51]	398.87 ± 133.12 g	14.15 ± 2.66 g	250 mL, 100 mL, 80 mL, 50 mL	SNMP: 500 Ml DMEM supplemented with 100 μg/mL penicillin and streptomycin, 4 mmol/L L-glutamine/L-alanine, 1 μM human insulin, 14 ng/mL glucagon, 1 μM dexamethasone. NMP: The isolated RBCs suspended in DMEM, isolated rat plasma (10% of the total volume), and 500 IU of heparin
Yamada et al[96]	260-350 g	DCD 10.12 ± 0.93 g	NA	KH buffer
Hu et al[52]	250-300 g	NA	150 mL	HTK solution
Von Horn and Minor[53]	250-300 g	NA	200 mL	Aqix RS-I solution
Raigani et al[54]	NA	NA	500 mL	High-glucose DMEM supplemented with 10% v/v FBS, 2% v/v penicillin-streptomycin, and 3% w/v bovine serum albumin. Defatting cocktail agents include 10 μM forskolin, 1 μM GW7647, 1 μM GW501516, 10 μM scoparone, 10 μM hypericin, 0.4 ng/mL visfatin, 0.8 mmol/L L-carnitine, and additional amino acids
Yang et al[55]	200-220 g	NA	NA	60 mL DMEM/F12 (1:1) containing 20% FBS and 1% penicillin-streptomycin solution (penicillin 10000 IU/mL, streptomycin 10000 μg/mL), 20 mL of fresh blood, 5 IU/mL of heparin, 2 IU/L of insulin, and 2.5 μg/mL of dexamethasone
Westerkamp et al[76]	270-300 g	NA	100 mL	25 mL human red blood cell concentrate, 53.9 mL WEM solution, 20 mL human albumin (200 g/L), 1 mL insulin (100 IU/mL), and 0.1 mL unfractionated heparin (5000 IU/mL)
De Vries et al[86]	200-250 g	NA	500 mL	WEM supplemented with sodium bicarbonate (2.2 g/L), dexamethasone (24 mg/L), insulin (5 IU/L), heparin (2000 IU/L), and bovine serum albumin (10 mg/mL)
Liu et al[85]	NA	NA	70 mL	HMP: UW solution with 250 IU heparin, 20000 IU penicillin and 2 mg hydrocortisone. Addition of CIRP competitive inhibitor (C23, 300 ng/mL). NMP: 20 mL rat blood, 50 mL WEM, 250 IU heparin, 20000 IU penicillin, 2 mg hydrocortisone, 0.4 IU insulin, and 0.0292 g glutamine
Lin et al[56]	180 ± 20 g	NA	50 mL	HTK solution with or without various defatting agents (10 mmol/L forskolin; 1 mmol/L GW7647; 10 mmol/L hypericin; 10 mmol/L scoparone; 0.4 ng/mL visfatin; 1 mmol/L GW501516)
Rigo et al[57]	200-250 g	10.35 g (0.41) mean (SEM)	70 mL	NMP: 20 mL of fresh rat blood, 50 mL of complete phenol red-free WEM, supplemented with 11.6 mmol/L glucose, 50 IU/mL penicillin, 50 μg/mL streptomycin, 5 mmol/L L-glutamine, 1 IU/mL insulin, 1 IU/mL heparin, and 2 mEq of sodium bicarbonate
Xu et al[97]	Approximately 500 g	Approximately 15 g	60-110 mL	Whole blood-based perfusate with different defatting components
Carlson et al[58]	331 ± 16 g (mean ± SEM)	NA	95 mL	WEM, 3250 IU each penicillin/streptomycin, 0.65 mmol/L sodium pyruvate, 1.30 mmol/L L-glutamine, 1% human albumin, 500 IU heparin, 15 mg papaverine, 1 mg insulin, 1.25 mg hydrocortisone, and 30 mL of leukoreduced, packed RBCs
Zhou et al[81]	250-300 g	NA	NA	NA
Cao et al[59]	200-220 g	NA	80 mL	DMEM/F12, 20 mL rat blood, 100 IU/mL penicillin, 100 μg/mL streptomycin, and 5 IU/mL heparin
De Stefano et al[60]	200-250 g	Control group: 17.10 (1.93). WI group: 15.17 (0.83)	150 mL	100 mL of phenol red-free WEM, supplemented with 100 IU/mL penicillin, 100 μg/mL streptomycin, 0.292 g/L L-glutamine, 1 IU/mL insulin, 1 IU/mL heparin, and 50 mL of recently expired (max 5 days) human red blood cell
Sun et al[61]	200-220 g	NA	2 mL	Normal saline or single cell suspension containing 1× 107-3 × 107 BMMSCs
Jennings et al[62]	320 ± 11 g (mean ± SEM)	NA	130 mL or 146 mL	WEM (65 mL for pRBC oxygen carriers and 50 mL for oxyglobin) with 3250 U each penicillin/streptomycin, 500 U heparin, 1 mg insulin, 1.25 mg hydrocortisone, and 15 mg papaverine. In addition, sodium pyruvate 0.65 mmol/L, L-glutamine 130 mmol/L, and human albumin 1%. The oxygen carriers added to the perfusate were human pRBC (30 mL), rat pRBC (30 mL), or oxyglobin (46 mL)
Wang et al[72]	NA	NA	HOPE: 50 mL; NMP: 30 mL	HOPE: HTK solution. NMP: 7.5 mL blood, 22.5 mL KHB solution containing 2% FBS, 60 mg glucose, 1 IU insulin, and 150 IU heparin
Asong-Fontem et al[73]	NA	NA	NA	HOPE: IGL-2 or PERF-GEN (Belzer MPS) solutions. NMP: Williams medium E supplemented with insulin 2 U/L, penicillin (40000 U/L)/streptomycin (40000 μg/L), L-glutamine, hydrocortisone 10 mg/L and heparin (1000 U/L)
Shi et al[77]	320-350 g	NA	36 mL	24 mL whole heparinized blood supplemented with 10% sodium citrate, 1% penicillin and streptomycin, and 12 mL circuit priming solution with 45% lactated ringer, 5% sodium bicarbonate and 50% hydroxyethyl starch
Von Horn et al[63]	250-300 g	NA	150 mL	HMP: Aqix RS-I solution. NMP: WEM supplemented with 3 mg/100 mL of bovine serum albumin
Zhou et al[82]	250-300 g	NA	NA	NA
Luo et al[83]	250-300 g	NA	HMP: 150 mL; NMP: 180 mL	HMP: HTK solution. NMP: DMEM/F12, 20% FBS, 1% penicillin streptomycin solution (penicillin 10000 IU/mL, streptomycin 10000 mg/mL), 5 IU/mL of heparin, 2 IU/L of insulin, and 2.5 mg/mL of dexamethasone. Full rat blood (30-45 mL) was reconstituted up to a total volume of 180 mL perfusate
Ohara et al[64]	250-350 g	NA	100 mL	Rat whole blood
Chen et al[65]	34 ± 4 g (mean ± SEM)	Approximately 1 g	300 mL	WEM was supplemented with 20% FBS, 1% penicillin/streptomycin, 5000 IU/L heparin, 50 IU/L insulin, and 0.010 g/L hydrocortisone
Fukai et al[74]	NA	NA	300 mL	HMP: UW-MPS solution. NMP: KHB solution
Hughes et al[84]	250-300 g	14 ± 0.2 g	150 mL	Phenol red-free WEM with addition of 25% bovine albumin, rat RBCs, 2 IU/L insulin, 40000 IU/L penicillin, 40000 μg/L streptomycin, 0.292 g/L L-glutamine, 10 mg/L hydrocortisone, 1000 IU/L heparin, and bicarbonate 75%
Bai et al[98]	360-380 g	NA	36 mL	24 mL whole heparinized blood supplemented with 10% sodium citrate, 1% penicillin, and streptomycin and 12 mL circuit priming solution with 45% lactated ringer, 5% sodium bicarbonate, and 50% hydroxyethyl starch
Von Horn et al[66]	250-300 g	NA	NA	Rewarming MP: Diluted Steen solution or Belzer MPS. NMP: WEM supplemented with 3 mg/100 mL of bovine serum albumin
Li et al[87]	250-350 g	NA	100 mL	90 mL WEM with 10 mL rat blood cells supplemented with 10 mg/L hydrocortisone, 5 IU/mL heparin, 1 IU/mL insulin, 5 mmol/L L-glutamine, 40 IU/mL penicillin, 40 μg/mL streptomycin, and with/without 10 mmol/L epigallocatechin gallate

MP: Machine perfusion; HMP: Hypothermic machine perfusion; NMP: Normothermic machine perfusion; NA: Not written or not applicable in the main article; SNMP: Subnormothermic machine perfusion; DCD: Donated after circulatory death; WI: Warm ischemia; HOPE: Hypothermic oxygenated machine perfusion; UW: University of Wisconsin; UW-G: University of Wisconsin-gluconate; HES: Hydroxyethyl starch; KHB: Krebs Henseleit bicarbonate; HTK: Histidine Tryptophan Ketoglutarate; Belzer MPS: Belzer machine perfusion solution; KH: Krebs Henseleit; KH₂PO₄: Potassium dihydrogen phosphate; MgCl: Magnesium chloride; ATP: Adenosine triphosphate; WEM: William’s E Medium; CaCl₂: Calcium chloride; MgSO₄: Magnesium sulfate; KPS: Kidney perfusion solution; 3-OMG: 3-O-methyl-D-glucose; NaCl: Sodium chloride; KCl: Potassium chloride; NaHCO₃: Sodium bicarbonate; HEPES: 4-(2-Hydroxyethyl)piperazine-1-ethanesulfonic acid; Ang IV: Angiotensin IV; KBB: Krebs bicarbonate buffer; DMEM: Dulbecco’s modified Eagle’s medium; IGL: Institute George Lopez; siRNA: Small interfering RNA; FBS: Fetal bovine serum; RBCs: Red blood cells; CIRP: Cold-inducible RNA-binding protein; BMMSC: Bone marrow mesenchymal stem cell; pRBC: Packed red blood cell.

There is a wide range of suggested perfusate compositions for HMP and SNMP in the literature. Some of the most commonly used perfusates are Krebs Henseleit bicarbonate, UW solution, HTK solution, Belzer MPS, WEM and others. Some researchers use modified perfusates by omitting starch[15,18,30,32,33,90], while others argue that starch-containing solutions enhance endothelial cell function and reduce hepatocellular damage compared to starch-free solutions[92]. In all three types of MP authors usually supplement perfusate with antibiotics and heparin[73,83-85,87,98]. Supplementation with other various medications is less common in HMP and SNMP setting as liver function is suppressed in lower temperatures[109]. However, some authors suggest supplementation in order to improve MP effectiveness and improve liver preservation outcomes. For instance, adding dopamine has been shown to potentially improve HMP effectiveness[26]. Additionally, supplementation with low-dose tacrolimus has resulted in better graft function and survival[32]. Adding α-tocopherol to the perfusate has demonstrated a reduction in inflammatory cytokines[42], while the inclusion of metformin in the perfusion solution has decreased rat liver injury[89]. Furthermore, perfusate supplementation with IGL-2 has been reported to reduce transaminases and significantly lower levels of glycocalyx proteins, CASP3, and HMGB1, indicating its protective role in preserving fatty livers[73]. Supplementation with angiotensin IV has been found to decrease the median effect concentration value and improve endothelium-dependent relaxation of HA rings[69].

For NMP, authors typically use various perfusates mixed with whole fresh blood, artificial blood, red blood cells, or other oxygen carriers. During NMP perfusate supplementation with erythrocytes reduces cell damage and improves liver function[51]. However, some protocols have demonstrated that 12-hour erythrocyte-free NMP in mice has no significant impact on histological structure[65]. Supplementation with additional medications such as insulin, glucose, heparin, hydrocortisone, albumin, and amino acids is more frequent in NMP to better mimic the in vivo liver environment. Some researchers have added pegylated-catalase to the base perfusate, which has reduced liver preservation injury[42]. Glycine treatment has synergistically preserved the integrity of both normal and donation after DCD liver grafts[70]. Supplementation with metamizole has led to higher bile production, lower transaminase levels, and reduced necrosis in liver and bile duct tissue[71]. Several researchers have investigated perfusate supplementation with bone marrow mesenchymal stem cells (BM-MSCs), which have shown promise in enhancing liver quality in rat DCD livers by reducing oxidative stress, improving mitochondrial function, lowering reactive oxygen species and free ferrous ion levels, and repairing the morphology and function of donor livers[55,61]. Additionally, BM-MSCs modified with heme oxygenase 1 have inhibited natural killer cell and cluster of differentiation 8⁺ T cell activation, thus reducing acute graft rejection[59]. Other medications, such as defatting agents, are used when researchers aim not only to bridge the time until LTx but also to treat conditions like fatty liver disease[56,73,87,97].

Table 4 represents various perfusate compositions and volumes proposed by the authors. To highlight the importance of adjusting perfusate volume according to liver weight, we included the animal and liver weights. This provides a more accurate representation of the relationship between organ weight and perfusate volume used.

MP parameters

Ex vivo rat MP studies are published with a variety of temperature, flow, and pressure settings, although detailed descriptions are missing in several studies. Table 5 represents possible MP parameters used in small animal liver MP models.

Table 5 Parameters used in small animal liver ex vivo machine perfusion studies.

Ref.	HMP			SNMP			NMP
	Temperature	Flow	Pressure	Temperature	Flow	Pressure	Temperature	Flow	Pressure
Kim et al[14]	4 °C	0.5 mL/minute/g liver	11.2 ± 0.4 mmHg
Dutkowski et al[15]	3-6 °C	0.43-0.44 mL/minute/g liver	4.48-0.47 mmHg
Compagnon et al[67]	4 °C	0.1-0.4 mL/minute/g liver	PV, VC: ≤ 1 mmHg; HA: ≤ 11 mmHg				37 °C	3 mL/minute/g wet liver	NA
Lauschke et al[16]	4 °C	0.5 mL/minute/g	NA				37 °C	3 mL/g/minute	NA
Lee et al[90]	4-5 °C	0.4 mL/minute/g liver	< 3 mmHg
Tan et al[91]	6-8 °C	0.1 mL/minute/g	NA
Xu et al[92]	4 °C	0.4 mL/minute/g liver	NA				37 °C	NA	NA
Bessems et al[17]	4 °C	< 1 mL/minute/g liver	< 20 cmH2O				37.1 °C ± 0.4 °C	< 3 mL/minute/g liver	< 20 cmH2O
Dutkowski et al[18]	3-5 °C	2.75-3.25 mL/minute	4.4 ± 0.5 mmHg				37 °C	15 mL/minute	NA
Tolboom et al[19]							37.5 °C	1.8 mL/ minute/g ± 0.12 mL/ minute/g wet liver	12-15 cmH2O
Vairetti et al[20]	4 °C, 10 °C	NA	NA	20 °C, 25 °C, 30 °C	NA	NA	37 °C	Initial: 1 mL/minute/g, increased to 4 mL/minute/g	NA
Manekeller et al[21]	4 °C	NA	NA				37 °C	3 mL/g/minute	NA
Stegemann et al[22]	4 °C	0.5 mL/minute/g	NA				37 °C	3 mL/g/minute	NA
Ferrigno et al[23]	4 °C or 8 °C	4 mL/minute/g	7.4 ± 0.6 mmHg (lean, 4 °C), 8.7 ± 2.1 mmHg (fat, 4 °C), 6.9 ± 0.8 mmHg (lean, 8 °C), 7.1 ± 0.9 mmHg (fat, 8 °C)	20 °C	4 mL/minute/g	7.3 ± 0.8 mmHg (lean) 7.5 ± 1.6 mmHg (fat)	37 °C	4 mL/minute/g	NA
Lüer et al[24]	4 °C	0.5 mL/minute/g	NA				37 °C	3 mL/g/minute	NA
Olschewski et al[25]	4 °C or 12 °C	1 mL/minute/g liver	NA	21 °C	1 mL/minute/g liver	NA	37 °C	3 mL/minute/g liver	NA
Minor et al[26]	4 °C	0.5 mL/minute/g	NA				37 °C	3 mL/g/minute	NA
Tolboom et al[27]				20 °C or 30 °C or 37 °C	2 mL/minute/g	10-14 cmH2O	37 °C	2 mL/minute/g	10-14 cmH2O
Giannone et al[28]	4 °C	1 mL/minute/g liver	NA
Perk et al[29]							37 °C	Pressure dependent	10-12 cmH2O
Schlegel et al[30]	4 °C	Pressure dependent	≤ 3 mmHg
Bruinsma et al[88]				Room temperature	NA	NA
Liu et al[31]				20 °C	1 mL/minute/g	NA
Carnevale et al[93]	5.0 °C ± 0.5 °C	0.23 mL/minute/g liver	40 mmH2O (25% of the NMP PV pressure)				37 °C	NA	NA
Schlegel et al[32]	4 °C	Pressure dependent	≤ 3 mmHg
Schlegel et al[33]	4 °C	PV: 1-2 mL/minte	PV: 3 mmHg				37 °C	HA: 6 mL/minute; PV: 15 mL/minute	PV: 8 mmHg
Bae et al[34]	4 °C	3.5 mL/minute/g of liver	NA				37 °C	NA	NA
Niu et al[68]							37 °C	20 mL/minute	NA
Tarantola et al[35]				20 °C	4 mL/minute/g	NA	37 °C	4 mL/minute/g	NA
Bruinsma et al[36]				21°C	8-12 mL/minute	10-15 cmH2O
Ferrigno et al[37]	10 °C	2.6 mL/minute/g	5.8 ± 0.2 mmHg	20 °C or 30 °C	2.6 mL/minute/g	4.9 ± 0.1 mmHg, 4.9 mmHg ± 0.2 mmHg	37 °C	2.6 mL/minute/g	4.2 ± 0.1 mmHg
Jia et al[38]	4 °C	1.4 mL/minute	NA
Westerkamp et al[94]	8 °C	Pressure dependent	PV: 3 mmHg; HA: 25 mmHg	20 °C	Pressure dependent	PV: 4 mmHg; HA: 40 mmHg	37 °C	Pressure dependent	PV: 11 mmHg. HA: 110 mmHg
Carbonell et al[95]				22 °C or 26 °C	3 mL/minute/g	NA	37 °C	3 mL/minute/g	NA
Op den Dries et al[75]							37 °C	PV: 22.6 ± 0.8 mL/minute; HA: 5.3 ± 0.4 mL/minute	PV: < 11 mmHg. HA: < 110 mmHg
Okamura et al[39]				20-24 °C	PV: 1 mL/g-liver/minute; HA: 0.1 mL/g-liver/minute	NA	37 °C	PV: 3 mL/g-liver/minute	NA
Berardo et al[40]				20 °C	NA	Starting: 6-7 mmHg	37 °C	NA	Starting: 6-7 mmHg
Chai et al[89]	4 °C	4 mL/minute	NA
Zeng et al[41]	0-4 °C	0.5 mL/g/minute	NA				36-37 °C	Pressure dependent	10.3 mmHg
Beal et al[42]							37 °C	1-2 mL/minute	10-16 cmH2O
Tabka et al[69]				20 °C	0.5 mL/g/minute	NA	37 °C	NA	NA
Xue et al[43]	0-4 °C	0.23 mL/minute/g	NA				36-37 °C	3 mL/minute/g	NA
He et al[78]	4 °C	1.4 mL/minute	NA
Gassner et al[70]							37 °C	1 mL/minute/g liver weight	4-9 mmHg
Zeng et al[79]	Approximately 4 °C	0.2 mL/minute/g	< 2 mmHg				36.5 °C ± 0.5 °C	2.5 mL/minute/g	NA
Jia et al[80]	4 °C	1.4 mL/minute	4.61 mmHg (mean)
Oldani et al[44]	4 °C	Pressure dependent	4 mmHg				37 °C	Pressure dependent	8 mmHg
Chin et al[45]							37 °C	Initial: 5 mL/minute, then pressure dependent	5 mmHg
Gillooly et al[46]	4-7 °C	Pressure dependent	10 mmHg				37 °C	Pressure dependent	10 mmHg
Martins et al[47]							32 °C or 37 °C	NA	NA
Scheuermann et al[48]				25 ± 0.1 °C or 30 °C	1.80 mL/minute/g liver	NA	37 °C	1.80 mL/minute/g liver. Reperfusion: 2.84 ± 0.04 mL/minute/g liver	NA
Claussen et al[71]							37 °C	PV: 1 mL/minute/g liver. HA: 0.1 mL/minute/g liver	PV: 5.65-9 mmHg. HA: 48.8-110 mmHg
Haque et al[49]							37 °C	25-30 mL/minute	< 12 mmHg
Schlegel et al[50]	10 °C	1-2 mL/minute	≤ 3 mmHg				37 °C	15-18 mL/minute	12 mmHg
Nösser et al[51]				21 °C	1 mL/g wet liver/minute	5.0 mmHg	37 °C	1 mL/g wet liver/minute	5.0 mmHg
Yamada et al[96]				20-25 °C	NA	NA	37 °C	NA	NA
Hu et al[52]	0-4 °C	0.5 mL/g/minute	NA				36.5 °C ± 0.5 °C	NA	10.3 mmHg
Von Horn and Minor[53]							35-42 °C	NA	Initial: 3 mmHg. Later: 5 mmHg
Raigani et al[54]							37 °C	NA	10-12 mmHg
Yang et al[55]							35-38 °C	2 mL/g wet liver/minute	10-12 mmH2O
Westerkamp et al[76]							37 °C	NA	PV: 11 mmHg. HA: 110 mmHg
De Vries et al[86]				21 ± 1 °C	≤ 25 mL/minute	5 mmHg
Liu et al[85]	4 °C	PV: 8 mL/minute	NA				37 °C	PV: 8 mL/minute; HA: 4 mL/minute	NA
Lin et al[56]	4 °C	< 0.15 mL/minute/g	< 3 mmHg
Rigo et al[57]							37 °C	1.1-1.3 mL/minute/g	8-10 mmHg
Xu et al[97]							37 °C	NA	NA
Carlson et al[58]							37 °C	1.8 mL/minute/g	NA
Zhou et al[81]	NA	NA	NA				NA	NA	NA
Cao et al[59]							36-38 °C	1.5 mL/minute/g wet liver	10-14 cmH2O
De Stefano et al[60]							37 °C	NA	12-16 mmHg
Sun et al[61]							NA	NA	NA
Jennings et al[62]							37 °C	1.8 mL/minute/g	NA
Wang et al[72]	4 °C	NA	NA				36.5 ± 0.5 °C	NA	NA
Asong-Fontem et al[73]	5 °C ± 3 °C	NA	NA				37 °C	NA	NA
Shi et al[77]							38 °C	PV: 5-15 mL/minute. PV and HA flow ratio 3:1	PV: 8-10 mmHg. HA: 90-100 mmHg
Von Horn et al[63]	8 °C	NA	5 mmHg				37 °C	3 mL/g minute	NA
Zhou et al[82]	NA	NA	NA				NA	NA	NA
Luo et al[83]	4 °C	1.2 mL/minute	NA				35-37 °C	2 mL/g/minute	NA
Ohara et al[64]							37 °C	PV: 1 mL/minute/g liver. HA: 0.25 mL/minute/g liver	NA
Chen et al[65]							37 °C	1 mL/minute	7-10 mmHg
Fukai et al[74]	7-10 °C	0.5 mL/minute/g liver	4-6 cmH2O				37 °C	NA	8-12 cmH2O
Hughes et al[84]							36-37.5 °C	PV: 1.2 mL/g-liver/minute; HA: 0.2 mL/g-liver/minute	NA
Bai et al[98]							NA	NA	NA
Von Horn et al[66]				Risen from 8 to 35 °C	5.5 ± 1.4 mL/g/minute or 5.2 ± 1.1 mL/g/minute	Risen from 3 mmHg to 6 mmHg	37 °C	3 mL/g/minute	NA
Li et al[87]							37 °C	Pressure dependent	< 4 mmHg

HMP: Hypothermic machine perfusion; NMP: Normothermic machine perfusion; NA: Not written or not applicable in the main article; SNMP: Subnormothermic machine perfusion; PV: Portal vein; VC: Vena cava; HA: Hepatic artery.

Perfusion applied via PV exhibits varying flow settings depending on the type of MP employed: In HMP, the potential flow settings range from 0.1 mL/minute/g to 0.5 mL/minute/g liver; In SNMP, the range extends from 1 mL/minute/g to 12 mL/minute/g liver; And in NMP, the range spans from 1 mL/minute/g to 30 mL/minute/g liver. Another feasible approach is to tailor the flow based on targeted pressure levels[29,44,46,94]. However, HA flow and pressure descriptions are less common, as the majority of authors prefer single-vessel MP. In NMP, HA flow settings range from 4 mL/minute to 6 mL/minute or from 0.1 mL/minute/g to 0.25 mL/minute/g liver weight, while in SNMP, it is suggested to be 0.1 mL/g of liver/minute. Some authors propose setting PV and HA flow rates in a 3:1 ratio[77].

As there are three main types of MP, literature reveals significant inconsistency regarding the specific temperatures associated with these terms[13]. Most authors perform HMP at 4 °C, though some use lower temperatures between 0 °C and 4 °C, and a few employ temperatures as high as 12 °C. NMP is generally conducted at 37 °C, with slight variations of up to 2 degrees in some studies. SNMP exhibits the most diverse temperature settings, ranging from 20 °C to 35 °C. Table 5 presents a range of temperature settings employed in small animal HMP, SNMP, and NMP.

DISCUSSION

Over the past two decades, MP has been considered a significant advancement in the field of transplantation[110]. While it has shown superiority over standard SCS and the potential to enhance suboptimal livers, many of its possible applications remain unexplored. Utilizing small animal liver MP presents an excellent opportunity to investigate these potential roles. Rat liver MP is an ideal model for investigating possible MP applications due to its lower costs compared to large animal studies[49]. In rats, around 10% of the liver’s blood supply is arterial and 90% is venous, compared to larger animals. Most literature suggests that perfusion through the PV alone is sufficient and that LTx can be done without reconstructing the HA. Moreover, rat liver explant surgery and transplantation operating times can be under 1 hour, making the process faster, cheaper, and easier compared to large animals. Additionally, the wide availability of inbred rat strains allows researchers to avoid immunological compatibility issues during MP and transplantation[19].

For liver explant surgeries in experimental animals, ensuring proper anesthesia is crucial for maintaining the welfare and stability of the subjects throughout the procedure. Researchers employ various anesthesia protocols tailored to meet experimental requirements and ethical considerations. One common approach involves isoflurane inhalation for sedation, combined with subcutaneous administration of analgesics[21,44,71,76,90]. Alternatively, some researchers opt for intramuscular injection of ketamine hydrochloride (90 mg/kg) and xylazine (10 mg/kg) to induce anesthesia[16,22,24,26]. Conversely, other authors utilize intraperitoneal administration of anesthetics such as sodium pentobarbital at different dosages ranging from 20 mg/kg to 60 mg/kg[35,43,52,82,89,95], or chloral hydrate at varying concentrations (5% 10 mL/kg or 10% 3 mL/kg or 500 mg/kg)[41,55,93]. Each method aims to ensure adequate sedation and pain management for the animals undergoing liver explant surgery, thereby facilitating a successful experimental procedure with minimal discomfort or distress to the subjects.

The first description of rat orthotopic LTx dates back to 1979 and has since remained the gold standard, despite its complexity and long learning curve[111,112]. Although various alternative techniques were proposed to simplify the surgery and make it easier to learn, liver explant surgery remains complex and demands meticulous training[99,113,114]. In the study by Tolboom et al[19], a surgeon with prior experience of over 100 orthotopic liver transplants in rats, performed all surgeries. Specific expertise in liver surgery, prior training, and a deep understanding of the procedure are crucial for successful outcomes[115]. Unfortunately, many studies lack detailed descriptions of liver explant surgery and information about surgeons’ training backgrounds.

Currently, no studies have definitively analyzed which MP setup closed or open circuit is superior. Both configurations are shown to be effective in preserving liver function according to the articles reviewed in this study. A closed circuit, while requiring additional cannulation and precise surgery to prevent perfusate leakage and ensure continuous perfusion, may offer certain advantages. In contrast, an open circuit simplifies the procedure by not requiring VC cannulation. Dual vessel MP, involving both the PV and HA, is generally superior to PV-only perfusion, as it better supports the vascularization of the biliary tree and ensures adequate oxygen delivery[64]. However, it is also more expensive and necessitates advanced techniques to avoid damage to arterial intima[106]. Some authors suggest that single-vessel PV or retrograde perfusion is sufficient for liver preservation. Perfusion via the HA alone is less advantageous because this artery supplies roughly one-tenth of the liver[67]. Dual vessel MP offers superior outcomes but at a higher cost and complexity, whereas single-vessel PV or retrograde perfusion can be sufficient in many cases. Bile duct cannulation is essential and should always be performed, as it allows for the measurement of bile output and the assessment of bile composition, which are critical markers of biliary viability[107].

The heterogeneity of MP setups makes it difficult to select the best one. Key considerations in setting up the apparatus include: Ensuring the organ chamber accommodates the liver position without bending its edges and remains close to physiological conditions, positioning cannulas connected to the tubing close to the anatomical position of vessels, ensuring adequate oxygenation and temperature control, utilizing pumps and sensitive sensors for necessary flow and pressure adjustments, and incorporating bubble traps into the system to prevent air embolisms. We recommend avoiding excessively large tubing or perfusate reservoirs, as miniaturization is crucial for the successful establishment of small animal MP[70].

The choice of perfusate is as critical as the setup of the perfusion apparatus. Compositions and volumes of perfusates proposed by different authors vary significantly, yet few studies have examined the relative superiority of one perfusate over another[17,21]. Future research should focus on identifying the optimal perfusate composition. Regardless of the chosen perfusate composition, we recommend adjusting the perfusate potential of hydrogen (pH) to the normal range before connecting the liver to the MP circuit. Additionally, mild acidosis of the perfusate is believed to enhance cytoprotection during hypothermia, which can be considered an added advantage[17]. Therefore, it is advisable to maintain the perfusate within the normal pH range and avoid alkalosis.

HMP and SNMP do not require the addition of red blood cells or hourly supplementation of the perfusate because liver metabolic activity is suppressed and the risk of clotting and hemolysis is higher at lower temperatures[109]. During NMP, supplementation with red blood cells or other oxygen carriers is crucial for adequate oxygen delivery to the cells[116]. Additionally, during NMP, consideration should be given to hourly supplementation with heparin, insulin, amino acids, or other additives throughout the entire perfusion process as at 37 °C the liver is metabolically active and should be provided with substances similar to in vivo conditions[19].

CONCLUSION

This review highlights the current state of small animal liver MP systems, emphasizing their benefits and challenges in experimental research. Rat liver MP models offer a cost-effective and accessible alternative to larger animal studies, with simplified perfusion processes and shorter operative times, making them valuable for LTx and preservation research. However, significant challenges remain, particularly in dual-vessel perfusion, which enhances vascularization and oxygenation but requires advanced surgical skills and additional resources. Furthermore, the lack of a standardized protocol limits reproducibility across studies. Maintaining adequate perfusate volume is another challenge, as the small blood volume of rats often falls short of the fluid requirements of the perfusion apparatus. Closed-circuit systems may improve preservation through continuous perfusion but require meticulous handling, while open-circuit setups simplify procedures at the cost of reduced control over perfusion parameters. Likewise, perfusate composition remains an area of ongoing investigation, as factors such as red blood cell content, heparin use, and pH adjustments significantly impact liver viability, particularly in NMP perfusion. In conclusion, while dual-vessel closed-circuit MP offers superior physiological outcomes, single-vessel and open-circuit approaches may still be suitable for many studies, depending on research objectives. Future efforts should focus on optimizing perfusate composition and refining MP setups to improve reproducibility and minimize animal use. Establishing a standardized protocol will be crucial for advancing small animal MP research, facilitating faster implementation, and reducing the number of animals required for experimentation. We hope this review serves as a valuable resource for researchers seeking to streamline their MP protocols and enhance experimental efficiency.