Full robotic liver transplantation in recipients: a narrative review of three competing surgical strategies

Abstract

Robotic liver transplantation (RLT) has emerged as the latest frontier in minimally invasive transplant surgery. Since the first fully robotic living donor liver transplant in 2023, multiple centers worldwide have reported their initial experiences, leading to the development of three distinct technical strategies. This review systematically compares these three competing approaches: the Asian strategy (partial caval clamping, no veno-venous bypass, Pfannenstiel incision) reported by Broering et al., the American strategy (total caval clamping with veno-venous bypass, upper midline incision) reported by Khan et al., and the European strategy (total caval clamping without bypass, upper midline incision, routine machine perfusion) reported by Pinto-Marques, Di Benedetto, and colleagues. Key technical differences, patient selection criteria, and perioperative outcomes are analyzed. Notably, the most recent European case employed partial caval clamping, suggesting a trend toward convergence. While all three strategies have demonstrated feasibility with excellent short-term outcomes, no standardized approach currently exists. This review provides a framework for understanding current RLT techniques and offers guidance for centers considering adoption of robotic liver transplantation.

Introduction

Liver transplantation is the definitive treatment for end-stage liver disease, with over 30,000 procedures performed annually worldwide and continuously improving long-term outcomes [17, 18]. but traditional open surgery requires large incisions (J-shaped or Mercedes incisions) and is associated with significant surgical trauma, prolonged recovery, and considerable morbidity [1, 2, 19]. Robotic surgery, with its 3D visualization, tremor filtering, and enhanced dexterity, has revolutionized minimally invasive surgery across multiple disciplines over the past three decades [3, 4, 20]. These platforms offer improved ergonomics and precision compared to conventional laparoscopy, leading to reduced surgeon fatigue and enhanced operative capabilities [21, 22]. In transplantation, robotic techniques have been increasingly adopted for living donor hepatectomy, with several centers reporting large series demonstrating safety and feasibility [5, 6, 23–26]. The International Laparoscopic Liver Society consensus guidelines have recognized robotic donor hepatectomy as a viable approach in experienced hands [27].

The extension of robotic technology to recipient liver transplantation, however, has lagged due to the technical challenges posed by portal hypertension, coagulopathy, and the complexity of vascular and biliary reconstruction [7]. The first breakthrough came in 2021 when Suh et al. reported a hybrid approach combining laparoscopic hepatectomy with robotic-assisted graft implantation [8, 30]. Subsequent developments included purely laparoscopic explant hepatectomy techniques [29] and laparoscopic-assisted approaches [31]. This was followed by the world’s first fully robotic living donor liver transplant by Broering et al. in 2023, where both donor hepatectomy and recipient implantation were completed robotically [9]. Subsequently, the same group reported a comparative study of 54 robotic versus 453 open cases, demonstrating significantly reduced morbidity and shorter hospital stay with the robotic approach [14]. Khan et al. reported the first fully robotic deceased donor whole liver transplant in North America [10], and Pinto-Marques, Di Benedetto, and colleagues reported the first European series of robotic whole liver transplantation [11, 12]. Laparoscopic-assisted approaches have also been explored as a transitional strategy toward fully minimally invasive recipient surgery [28].

As experience accumulates, three distinct technical strategies have emerged, each with different approaches to inferior vena cava (IVC) clamping, use of Veno-venous bypass, and incision selection [13]. The evolution of robotic liver transplantation since 2021 is illustrated in Fig. 1. However, no systematic comparison of these competing strategies has been published to date. This review aims to provide a comprehensive comparison of the three main RLT techniques, analyzing their technical details, patient selection criteria, perioperative outcomes, and emerging trends. Understanding these differences will help centers considering adoption of RLT to select the most appropriate approach based on their expertise and resources.

Fig. 1.

Timeline of robotic liver transplantation development. Key contributions from Asian (green), American (blue), and European (orange) centers are shown. The first hybrid approach (laparoscopic hepatectomy with robotic-assisted implantation) was reported by Suh et al. in 2021 [8], followed by pure laparoscopic living donor liver transplantation in 2022 [29]. The world’s first fully robotic living donor liver transplant was achieved by Broering et al. in 2023 [9]. In 2024, Khan et al. reported the first fully robotic deceased donor whole liver transplant [10]. The first European series was published by Pinto-Marques et al. in 2025 (6 cases) [11], followed by the Modena single-center series of 10 cases in 2026 [12]. The largest series to date (54 robotic vs. 453 open cases) was reported by Broering et al. in 2026 [14].

Methods

This review aims to systematically compare the three technical strategies for robotic transplantation surgery in terms of technical details, patient selection, perioperative outcomes, and emerging trends. A comprehensive literature search was performed in PubMed, Web of Science, and Scopus databases for studies published between January 2019 and March 2026. The search terms included combinations of “robotic liver transplantation,” “robotic recipient hepatectomy,” “full robotic liver transplant,” “surgical technique,” “IVC clamping,” and “venous bypass.” The reference lists of relevant articles were also manually screened to identify additional studies.

Studies were included if they met the following criteria: (1) original reports of fully robotic recipient hepatectomy and graft implantation; (2) clear description of surgical technique, including incision type, inferior vena cava (IVC) clamping method, and use of veno-venous bypass; (3) reporting of perioperative outcomes; and (4) English language. Review articles, editorials, conference abstracts, and studies focusing only on donor surgery were excluded.

Data extracted from each included study comprised: first author, publication year, center, number of cases, patient selection criteria (including Model for End-Stage Liver Disease [MELD] score and graft type), surgical technique details (incision, IVC clamping, bypass use, anastomotic techniques), and perioperative outcomes (operative time, warm ischemia time, blood loss, transfusion requirements, complications, length of stay, and graft/patient survival).

Due to the heterogeneity of the included studies and the descriptive nature of this review, no meta-analysis was performed. The findings are presented as a narrative synthesis, with key data summarized in comparative tables.

Results

The initial literature search yielded a total of 186 articles. After removal of duplicates and screening of titles and abstracts, 42 full-text articles were assessed for eligibility. Following application of the inclusion and exclusion criteria, 13 studies were included in this review, comprising original reports of fully robotic recipient hepatectomy and graft implantation with detailed technical descriptions.

Among the included studies, three distinct technical strategies were identified, as previously categorized by Liou and Rocca [13]. The Asian strategy is represented by two reports from Broering et al., including the initial description of the technique in three cases [9] and a subsequent comparative study of 54 robotic versus 453 open living donor liver transplants [2]. The American strategy is represented by the first reported case of fully robotic deceased donor whole liver transplantation by Khan et al. [10]. The European strategy is represented by a multicenter report from Pinto-Marques et al. (six cases) [11] and a subsequent single-center series from Magistri et al. (ten cases) [12].

The characteristics of the included studies are summarized in Table 1. Detailed descriptions of each strategy, including patient selection criteria, surgical techniques, and perioperative outcomes, are presented in Sects. 3.2–3.4.

Table 1.

Key publications and basic characteristics of the three robotic liver transplantation strategies

Strategy	Represented by	Key references	Center	No. of cases	Graft type
Asian strategy	Broering et al.	[5, 9]	King Faisal Specialist Hospital, Riyadh, Saudi Arabia	54	Living donor right lobe
American strategy	Khan et al.	[10]	Washington University, St. Louis, USA	1 (first report)	Deceased donor whole liver
European strategy	Pinto-Marques et al. Magistri et al.	[11, 12]	Curry Cabral Hospital, Lisbon, Portugal University of Modena, Italy	6 (multicenter) 10 (single-center)	Deceased donor whole liver

Overview of the three strategies

Since the first fully robotic living donor liver transplant was reported by Broering et al. in 2023 [9], several centers worldwide have initiated robotic liver transplantation (RLT) programs and published their initial experiences [10–12]. As experience accumulates, three distinct technical strategies have emerged, each characterized by different approaches to inferior vena cava (IVC) clamping, use of veno-venous bypass, and incision selection [13]. The inclusion and exclusion criteria for recipients and donors/grafts across the three strategies are summarized in Table 4.

Table 5.

The robotic platforms and specific instruments used across the three strategies

Item	Asian Strategy	American Strategy	European Strategy
Robotic platform	da Vinci Xi [9]	da Vinci Xi [10]	da Vinci Xi [11, 12]
Parenchymal transection	Fenestrated Bipolar, Vessel Sealer Extend, ProGrasp, Maryland Bipolar [9]	Not specified	Not specified
Hepatic vein division	GIA vascular stapler [9]	Vascular stapler [10]	Vascular stapler [11]
Portal vein division	Hem-o-lok clips [9]	Vascular stapler [10]	Vascular stapler [11]
Hepatic artery division	Hem-o-lok clips [9]	Vascular clips [10]	Vascular clips [11]
Bile duct division	Scissors with clips [9]	Scissors with clips [10]	Scissors with clips [11]
Vascular anastomosis suture	Hepatic vein: Gortex 6/0 continuous Portal vein: Gortex 6/0 continuous Hepatic artery: Prolene 7/0–8/0 interrupted [9]	Caval: Gore-Tex 4/0 running (side-to-side) Portal vein: Gore-Tex 6/0 running Hepatic artery: Prolene 6/0–7/0 running [10]	Caval: Gore-Tex 4/0–5/0 running Portal vein: Gore-Tex 5/0–6/0 running Hepatic artery: Prolene 6/0–7/0 running/interrupted [11]
Biliary anastomosis suture	PDS 7/0 interrupted [9]	PDS 5/0 interrupted [10]	Prolene/PDS 5/0–6/0 running [11]
Hand-assist device	GelPort [9]	GelPort [10]	Alexis retractor (Lisbon) / GelPort (Modena) [11]

Note: “Not specified” indicates that the original publication did not provide detailed information on the specific instrument used

Table 1 summarizes the key studies representing each of the three strategies. The Asian strategy, pioneered by Broering’s team at King Faisal Specialist Hospital in Riyadh, Saudi Arabia, is the largest series to date, with 54 fully robotic living donor liver transplants reported in a comparative study against 453 open cases [2]. This strategy is characterized by partial IVC clamping, no veno-venous bypass, and a Pfannenstiel incision for graft extraction and insertion [1, 2].

The American strategy, reported by Khan et al. from Washington University in St. Louis, represents the first fully robotic deceased donor whole liver transplant in North America and globally [10]. This approach employs total IVC clamping with veno-venous bypass to prioritize hemodynamic stability, using an upper midline incision that can be extended for graft removal and insertion [10].

The European strategy, developed collaboratively by the Lisbon and Modena teams and reported by Pinto-Marques et al. and Magistri et al., represents the first European series of robotic whole liver transplantation [11, 12]. This strategy combines total IVC clamping without veno-venous bypass, an upper or periumbilical midline incision, and routine use of hypothermic oxygenated machine perfusion (HOPE) for graft preservation [11, 12]. Notably, the most recent case in the Modena series employed partial caval clamping, suggesting a potential trend toward convergence of techniques [12, 13]. A schematic comparison of the three strategies is presented in Fig. 2. Port placement and incision sites are illustrated in Fig. 3. The robotic platforms and specific instruments used across the three strategies are compared in Table 5.

Fig. 2.

Surgical strategies for robotic liver transplantation. Created in https://BioRender.com Detailed descriptions of each strategy, including patient selection criteria, surgical techniques, and perioperative outcomes, are provided in the following sections

Fig. 3.

Port placement and incision sites for robotic liver transplantation. Created in https://BioRender.com. Four 8-mm robotic ports are placed in the upper abdomen. A GelPort (light gray oval) is placed in the periumbilical midline for graft extraction and hand-assistance. The green horizontal line indicates the Pfannenstiel incision (Asian strategy); the red dashed vertical line indicates the upper midline incision (American and European strategies). Port and incision positions may vary between centers.

Asian strategy: partial caval clamping, no bypass, pfannenstiel incision

Represented center and key publications

The Asian strategy is best represented by Broering’s team at King Faisal Specialist Hospital in Riyadh, Saudi Arabia. This group has the largest global experience with robotic liver surgery, having performed over 1,000 minimally invasive living donor hepatectomies before transitioning to robotic recipient surgery [5, 6, 32, 33].

Importantly, the Broering series exclusively utilizes partial grafts from living donors—predominantly right lobes (81.5% of cases) but also left lobes (16.7%) [14]. The use of partial grafts, which are smaller and more maneuverable than whole grafts, is a key factor enabling the feasibility of partial caval clamping and the specific implantation maneuvers described below.

Their initial report in 2023 described the world’s first fully robotic living donor liver transplant in three cases [9], followed by a comparative study of 54 robotic versus 453 open cases—the largest series to date and the only study providing direct comparison with the open approach [14].

Patient selection criteria

The patient selection criteria for the Asian strategy are detailed in Table 4. Briefly, the initial series favored lower-risk candidates with MELD scores < 30, right lobe grafts, single artery and bile duct, graft weight < 900 g, and GRWR > 0.8% [9, 14]. With accumulated experience, indications were progressively expanded to include left lobe grafts (16.7% of cases) and retransplantation [14].

Surgical technique

The Asian strategy is characterized by three key elements: partial caval clamping, no veno-venous bypass, and a Pfannenstiel incision [9, 14]. Organ procurement was performed via robotic living donor hepatectomy [9]. Back-bench preparation details were not reported [9].

After liver mobilization, an 8–12 cm Pfannenstiel incision is created with GelPort placement. A Glover clamp is manually introduced through the GelPort to achieve partial IVC clamping just below the right hepatic vein stump, maintaining caval flow and hemodynamic stability without bypass [9]. Hepatic veins are divided with staplers, and anastomoses are performed robotically (hepatic vein: Gortex 6/0 continuous; portal vein: Gortex 6/0 continuous; hepatic artery: Prolene 7/0/8/0 interrupted; bile duct: PDS 7/0 interrupted) [9].

Robotic platform and instruments

The da Vinci Xi System is used. Instruments include Fenestrated Bipolar, Vessel Sealer Extend, ProGrasp, and Maryland Bipolar forceps. Vascular staplers (GIA) are used for division of hepatic veins, while Hem-o-lok clips are used for smaller vessels [9].

Distinct technical features for partial graft implantation

The implantation technique differs depending on graft type. For right lobe grafts, the graft is positioned in the native right liver fossa with the hepatic vein oriented toward the recipient’s right hepatic vein stump, allowing standard orientation for vascular anastomoses [9].

For left lobe grafts, a specific “flopping” technique is employed. The graft is rotated 180° anticlockwise on its sagittal axis and placed in the right hepatic fossa rather than the left [44]. The left hepatic vein stump of the graft is then anastomosed to the recipient’s right hepatic vein stump under partial caval clamping (Fig. 2A). This positioning facilitates exposure and optimizes the spatial arrangement for portal vein and hepatic artery anastomoses. After reperfusion, the graft may be temporarily rotated clockwise to the left side for optimal exposure during biliary anastomosis, then rotated back to its final position [44]. This “flopping” technique is a distinctive feature of the Asian strategy that enables safe implantation of left lobe grafts in the robotic setting.

Perioperative outcomes

The comparative study of 54 robotic versus 453 open cases demonstrated significant advantages for the robotic approach despite longer operative times [14]:

Outcome	Robotic (n = 54)	Open (n = 453)	p value
Operative time	572 min	428 min	< 0.001
Warm ischemia time	48 min	32 min	< 0.001
Blood loss	650 ml	2000 ml	< 0.001
Infection rate	9.3%	42.8%	< 0.001
Major complications (≥ 3a)	18.5%	30.9%	0.060
CCI	14.8	20.9	0.045
ICU stay	3 days	5 days	< 0.001
Hospital stay	14 days	20 days	0.030
6-month recipient survival	100%	91.8%	0.040

Multivariable analysis confirmed that the robotic approach was independently associated with a 14.35-point reduction in CCI (p = 0.003) [14]. Conversion to open surgery occurred in 1 of 54 cases (1.9%) due to technical difficulties with native liver explant [14].

Summary

The Asian strategy, with partial caval clamping, no bypass, and a Pfannenstiel incision, represents the largest and most mature RLT experience. The exclusive use of partial grafts from living donors—combined with the distinctive “flopping” technique for left lobe implantation—enables safe and effective robotic transplantation with excellent outcomes. Comparative data demonstrate significantly reduced morbidity, faster recovery, and excellent survival, supporting its broader adoption [13, 14]. Anastomotic times were not reported in the Asian series [9, 14].

American strategy: total caval clamping with bypass, upper midline incision

Represented center and key publication

The American strategy is represented by Khan and colleagues at Washington University in St. Louis, USA. In 2024, they reported the world’s first fully robotic deceased donor whole liver transplant—the first robotic liver transplant performed in North America [10]. Their team had accumulated over 600 robotic cases before attempting this procedure [10].

Patient selection criteria

The patient selection criteria for the American strategy are summarized in Table 4. Key criteria included MELD < 25, donor liver weight < 2000 g, BMI 25–30 kg/m², and absence of portal vein or hepatic artery thrombosis [10]. Abdominal wall laxity (e.g., from prior ascites or pregnancy) was preferred to facilitate pneumoperitoneum [10].

Surgical technique

The American strategy is characterized by total caval clamping, routine veno-venous bypass, and an upper midline incision [10]. After procurement, the deceased donor liver was placed on normothermic machine perfusion (NMP, OrganOx). During back-bench preparation, the liver was taken off NMP, both ends of the donor cava were stapled and tagged with silk sutures, and the hepatic artery and portal vein were prepared for suturing with orientation marked [10].

Incision and ports

A gel port is placed in the periumbilical midline, with four 8-mm robotic ports in the upper abdomen. This location allows hand-assistance if needed and can be extended (to 15 cm) for graft extraction [10].

IVC management and bypass: Total clamping of the IVC is employed—suprahepatic IVC clamped with a Chitwood clamp via flank stab incision, infrahepatic IVC with bulldog clamps. Veno-venous bypass (femoral-to-jugular) is initiated at the start of IVC dissection and continued until after reperfusion. The authors plan to use bypass routinely for the first ten cases before considering transition to partial clamping [10].

Implantation

After intra-abdominal graft placement, caval anastomosis is performed as side-to-side cavocavostomy with 4 − 0 Gore-Tex, followed by portal vein anastomosis with 6 − 0 Gore-Tex. Hepatic artery reconstruction uses donor celiac to recipient common hepatic artery, and biliary reconstruction is duct-to-duct with interrupted 5 − 0 PDS [10].

Perioperative outcomes

Parameter	Value
Total operative time	~ 9 h
Warm ischemia time	77 min
Estimated blood loss	1000 ml
Hospital stay	12 days
6-month follow-up	Alive, well

The patient recovered uneventfully with no early allograft dysfunction or surgical complications and no conversion to open surgery was reported [10].

Technical considerations

Deceased donor whole grafts offer advantages for North American practice (> 90% of transplants), including longer vascular structures and single-team logistics, but require careful selection to avoid livers > 1800 g or those with large left lateral Sect. [10]. Warm ischemia time (77 min) remains longer than open transplantation and requires improvement [10]. Anastomotic times were reported: caval anastomosis 17 min, portal vein 23 min, hepatic artery 30 min, and biliary reconstruction 19 min [10]. Specific caval and portal clamping times were not reported.

Summary

The American strategy, with total caval clamping, routine bypass, and upper midline incision, represents a fundamentally different approach applicable to deceased donor whole grafts. The authors’ plan to use bypass for the first ten cases before considering transition to partial clamping suggests potential evolution toward the Asian strategy as experience accumulates [10, 13].

European strategy: total caval clamping without bypass, upper midline incision, routine machine perfusion

Represented centers and key publications

The European strategy was developed collaboratively by two high-volume centers: Curry Cabral Hospital in Lisbon, Portugal (Pinto-Marques et al.) and the University of Modena, Italy (Di Benedetto, Magistri et al.). Both institutions combine extensive experience in open and robotic liver resection (> 100 robotic cases annually) with active liver transplant programs (> 100 per year) [11]. Their initial multicenter report in 2025 described six cases of fully robotic whole liver transplantation from deceased donors—the first European series of its kind [11]. Subsequently, the Modena group published their single-center experience with ten cases, representing the largest European series to date [12].

Patient selection criteria

The patient selection criteria for the European strategy are presented in Table 4. Notable selection factors included low MELD scores (median 8.5), small caudate lobe anatomy, low-grade portal hypertension, and BMI < 35 [11, 12]. Patients with large caudate lobes, significant portal hypertension with collateralization, or portal vein thrombosis were excluded [11].

Surgical technique

The European strategy is characterized by three key elements: total caval clamping without veno-venous bypass, an upper or periumbilical midline incision, and routine use of hypothermic oxygenated machine perfusion (HOPE/D-HOPE) [11, 12]. All grafts were treated with hypothermic oxygenated perfusion (HOPE or D-HOPE) for at least 2 h before implantation [11]. Back-bench preparation details were not reported [11].

Incision and access

An upper midline or periumbilical incision (8–15 cm) is used for native liver extraction and graft insertion. The Lisbon group employs an Alexis retractor, while Modena uses a GelPort system [11].

IVC management

The technique employs total clamping of the inferior vena cava using laparoscopic vascular bulldog clamps. Notably, no veno-venous bypass is used [11]. However, in the Modena series of ten cases, nine utilized total clamping without bypass, while the most recent case employed partial caval clamping—a significant technical evolution [12].

Implantation

After graft introduction, caval anastomosis is performed side-to-side using 4/0 or 5/0 Gore-Tex running sutures. Portal vein anastomosis uses 5/0 or 6/0 Gore-Tex running sutures. Arterial reconstruction is performed with polypropylene 6/0 or 7/0 (running or interrupted based on vessel size), and biliary reconstruction uses running polypropylene or PDS 5/0/6/0 [11]. Anastomotic times were not detailed in the European reports [11, 12].

Perioperative outcomes

The combined European experience demonstrates excellent early outcomes [11, 12]:

Parameter	Pinto-Marques 2025 (n = 6)	Magistri 2026 (n = 10)
Operative time	440–710 min (median 595)	Not detailed
Warm ischemia time	55–90 min	55–90 min (inferred)
Blood transfusion >1U	2/6 patients	Not detailed
Major complications (≥ 3a)	0% (30-day)	0% (30-day)
Hospital stay	Median 7.5 days (range 6–40)	7.5 days (inferred)
Graft survival (6 months)	100%	100%
Patient survival (6 months)	100%	100%

The single prolonged hospital stay (40 days) occurred in a 69-year-old patient with MELD 23, diabetes, and pulmonary emphysema who received a steatotic graft, highlighting the importance of careful donor-recipient matching [11]. None of the recipients required conversion to open surgery in the Lisbon-Modena series [11]. No conversions were reported in the Modena series [12].

Technical evolution: toward partial clamping

A noteworthy observation from the Modena series is that the most recent case employed partial caval clamping rather than total clamping [12]. This shift, also noted in a recent editorial [13], suggests that the European strategy may be evolving toward the Asian approach. The authors of the editorial commented: “Notably, the Modena group’s last case employed partial caval clamping, hinting at the influence of the Riyadh group, which has the largest global RLT experience” [13].

Summary

The European strategy, combining total caval clamping without bypass, an upper midline incision, and routine machine perfusion, has demonstrated excellent short-term outcomes in the largest European series to date. The routine use of HOPE/D-HOPE may provide a safety margin that helps offset the prolonged warm ischemia times inherent to robotic transplantation. The recent shift toward partial clamping in the Modena series suggests a potential convergence of techniques, with European centers increasingly adopting elements of the more hemodynamically stable Asian approach [12, 13].

Discussion

Comparative analysis of the three strategies

The three robotic liver transplantation (RLT) strategies described in this review differ fundamentally in their approach to inferior vena cava (IVC) management, use of veno-venous bypass, incision selection, and graft type. These technical distinctions reflect the different philosophies and practice patterns that have emerged as centers worldwide have begun to explore this challenging procedure [32, 42, 43].

The key technical differences among the three strategies are compared in Table 2, while patient selection criteria and perioperative outcomes are summarized in Table 3.

Table 2.

Technical comparison of the three robotic liver transplantation strategies

Technical Parameter	Asian Strategy	American Strategy	European Strategy
Incision	Pfannenstiel incision (lower abdominal transverse, 8–12 cm) + GelPort	Upper midline GelPort (extendable to 15 cm superiority)	Upper or periumbilical midline incision (8–15 cm) Lisbon: Alexis retractor Modena: GelPort
IVC Management	Partial clamping Manual introduction of Glover clamp	Total clamping Suprahepatic: Chitwood clamp (via flank stab incision) Infrahepatic: 2 bulldog clamps	9 cases total clamping, most recent case partial clamping Laparoscopic vascular bulldog clamps
Veno-Venous Bypass	None	Routine use (femoral-to-jugular)	None
Machine Perfusion	Not reported	NMP (used in index case)	Routine HOPE/D-HOPE (≥ 2 h)
Vascular Anastomosis	Hepatic vein: Gortex 6/0 running Portal vein: Gortex 6/0 running Hepatic artery: Prolene 7/0/8/0 interrupted Bile duct: PDS 7/0 interrupted	Caval: Gore-Tex 4/0 running (side-to-side) Portal vein: Gore-Tex 6/0 running Hepatic artery: donor celiac to recipient GDA Bile duct: PDS 5/0 interrupted	Caval: Gore-Tex 4/0/5/0 running Portal vein: Gore-Tex 5/0/6/0 running Hepatic artery: Prolene 6/0/7/0 (running/interrupted) Bile duct: Prolene/PDS 5/0/6/0 running
Patient Selection	MELD 16 (14–20) Right lobe, single artery, single bile duct GRWR > 0.8%	MELD < 25 (index case 10) Graft < 2000 g BMI 25–30	MELD 8.5 (6–25) Small caudate lobe (critical) BMI < 35
Technical Evolution	Expanded from 3 to 54 cases; indications progressively broadened	Plan to use bypass for first 10 cases, then consider transition to partial clamping	Most recent case transitioned to partial clamping
Current Trend	Comparative study demonstrates superiority over open approach	Still in early exploration	Converging toward Asian strategy

Table 4.

Inclusion and exclusion criteria for robotic liver transplantation across the three strategies

Strategy	Recipient Criteria	Donor/Graft Criteria
Asian Strategy	Inclusion • MELD < 30 (initial series, median 16) • Single hepatic artery • Single bile duct • No complex venous reconstruction Exclusion • MELD ≥ 30 • Need for segment 5 or 8 hepatic vein reconstruction • Multiple arteries or bile ducts • Portal vein thrombosis	Inclusion • Living donor • Right lobe graft (81.5%) • Graft weight < 900 g • GRWR > 0.8% Exclusion • Left lobe graft (initially excluded, later included) • Graft weight ≥ 900 g • GRWR ≤ 0.8% • Donor with complex vascular anatomy
American Strategy	Inclusion • MELD < 25 (index case 10) • BMI 25–30 kg/m² • Abdominal wall laxity (preferred) • No portal vein or hepatic artery thrombosis Exclusion • MELD ≥ 25 • BMI > 30 or < 25 • Portal vein or hepatic artery thrombosis • Previous upper abdominal surgery	Inclusion • Deceased donor • Whole graft • Donor liver weight < 2000 g • Native liver volume < 2000 cc Exclusion • Donor liver weight ≥ 2000 g • Native liver volume ≥ 2000 cc • Steatotic graft > 30%
European Strategy	Inclusion • MELD < 30 (median 8.5, range 6–25) • Small caudate lobe • Low degree of portal hypertension • BMI < 35 • No porto-mesenteric thrombosis • Child A cirrhosis (80% in Modena series) Exclusion • Large caudate lobe • Significant portal hypertension with collateralization • BMI ≥ 35 • Porto-mesenteric thrombosis •Coagulopathy or thrombocytopenia	Inclusion • Deceased donor • Whole graft • Donor ICU stay 2–4 days • No prolonged vasopressor support Exclusion • Large liver graft • Steatotic graft > 30% • Donor with prolonged ICU stay

Table 2 highlights the core technical distinctions. The Asian strategy employs partial IVC clamping without bypass through a Pfannenstiel incision, primarily using living donor right lobe grafts. The American strategy utilizes total IVC clamping with routine veno-venous bypass via an upper midline incision for deceased donor whole liver transplantation. The European strategy combines total IVC clamping without bypass with an upper midline incision and routine use of hypothermic oxygenated machine perfusion (HOPE/D-HOPE) for deceased donor grafts. Notably, the most recent European case transitioned to partial clamping, suggesting an evolving technique [12, 13].

Table 3 compares patient selection and outcomes. The Asian strategy, with the largest cohort (54 cases), demonstrates excellent results despite longer operative times, with median blood loss of 650 ml, hospital stay of 14 days, and 100% 6-month recipient survival [14]. The American strategy, while based on a single case, achieved good outcomes with 1000 ml blood loss and 12-day hospital stay [10]. The European strategy, with 10 cases in the largest series, shows the shortest hospital stay (7.5 days) and 100% survival, potentially attributable to routine machine perfusion [11, 12].

Table 3.

Comparison of patient selection and perioperative outcomes among the three strategies

Parameter	Asian Strategy	American Strategy	European Strategy
MELD score	16 (14–20) [14]	10 [10]	8.5 (6–25) [12]
Warm ischemia time	48 (41–51) min [14]	77 min [10]	55–90 min [11,12]
Operative time	572 (524–638) min [14]	~ 540 min [10]	595 (440–710) min [11]
Blood loss	650 (500–1000) ml [14]	1000 ml [10]	—
Blood transfusion	1.4 (0.5–1.5) U [14]	2 U [10]	Only 2/6 patients required >1U [11]
Infection rate	9.3% [14]	0% [10]	—
Major complications (≥ 3a)	18.5% [14]	0% [10]	0% (30-day) [12]
CCI ®	14.8 (0.0-20.9) [14]	—	—
ICU stay	3 (2–4) days [14]	—	—
Hospital stay	14 (12–22) days [14]	12 days [10]	7.5 (6–40) days [11,12]
6-month graft survival	98.1% [14]	100% [10]	100% [11,12]
6-month recipient survival	100% [14]	100% [10]	100% [11,12]

Table 4 Inclusion and exclusion criteria for robotic liver transplantation across the three strategies. Criteria are based on the initial experience of each center; indications have since expanded with accumulated experience. For the Asian strategy, left lobe grafts (16.7%) and retransplantation were later included [14]. For the European strategy, patients with refractory ascites were not excluded [11].

Key technical controversies

Historical context: evolution of caval clamping

Caval management has been a central technical challenge in liver transplantation since its inception. In the early era of open transplantation, the classic technique required complete clamping of the suprahepatic and infrahepatic vena cava, with veno-venous bypass routinely used to maintain hemodynamic stability during the anhepatic phase [45]. A major milestone occurred in 1989 when Tzakis, Todo, and Starzl introduced the piggyback technique, which preserved the recipient’s inferior vena cava (IVC) by anastomosing the donor suprahepatic vena cava to a common opening formed by the recipient’s left and middle hepatic veins [46].This technique maintained caval flow throughout the procedure, improved hemodynamic stability, and reduced the need for veno-venous bypass [45].

Further refinement came with the modified piggyback technique using side-to-side cavocaval anastomosis. In a large series of 275 orthotopic liver transplants, Belghiti et al. demonstrated that IVC preservation was feasible in almost all candidates (90% maintained caval flow throughout the procedure). Temporary IVC cross-clamping during hepatectomy (mean duration 11 ± 4 min) was well tolerated, whereas caval clamping after graft reperfusion (mean 23 ± 5 min) was associated with higher graft failure and mortality rates, highlighting the importance of avoiding post-reperfusion caval clamping [46].

In the robotic era, these historical lessons have shaped the three current strategies: the Asian strategy embraces partial clamping from the outset, while the American and European strategies initially adopted total clamping with or without bypass [10, 11]. Notably, the most recent European case transitioned to partial clamping [12], reflecting a broader evolution toward more physiological, less invasive techniques that echoes the historical progression from total to partial clamping in open surgery.

The role of veno-venous bypass

The American strategy routinely uses veno-venous bypass to prioritize hemodynamic stability, with the authors planning to employ bypass for their first ten cases before considering transition to a non-bypass approach [10]. In contrast, both Asian and European strategies have demonstrated that RLT can be performed safely without bypass, even with total caval clamping in the European series [11, 12]. Notably, veno-venous bypass is no longer standard practice in most open liver transplant centers worldwide [48], further supporting the rationale for bypass-free approaches. Therefore, most centers initiating a robotic recipient program may find routine bypass use unnecessary, as it adds complexity to an already challenging procedure without clear benefit in carefully selected patients.

Incision selection: Pfannenstiel versus upper midline

The Pfannenstiel incision offers cosmetic advantages, lower risk of incisional hernia, and the ability to incorporate a GelPort for hand-assistance—providing a valuable safety net during the learning curve [9, 13]. Weill Cornell adopted this approach for their seven RLT cases, finding it “reliable and reproducible” [13]. It should be noted that the Asian strategy, while described as “fully robotic,” involves the placement of an assistant surgeon’s hand through the GelPort to stabilize the partial caval clamp until reperfusion is completed [9]. Strictly speaking, this represents a “robotic-assisted” rather than a “fully robotic” procedure. This distinction is important because the hand-assistance provides a valuable safety net during the learning curve, facilitating graft positioning and offering a means to control unexpected bleeding without full conversion. As dedicated robotic instruments—such as purpose-designed caval clamps—are developed in the future, the procedure may evolve toward true fully robotic surgery without the need for manual assistance [49].

The upper midline incision, favored by the American and European strategies, provides excellent exposure and can be easily extended if conversion to open surgery becomes necessary [10–12]. However, it may be associated with higher hernia risk and is less cosmetic.

The role of machine perfusion

The European strategy’s routine use of HOPE/D-HOPE represents a significant difference from other approaches. By mitigating ischemia-reperfusion injury, machine perfusion may provide a safety margin that offsets the prolonged warm ischemia times inherent to robotic transplantation [11, 12]. This may explain the remarkably short hospital stay (7.5 days) achieved in the European series, despite warm ischemia times comparable to or longer than other strategies.

The American strategy employed normothermic machine perfusion (NMP) selectively [10], while the Asian strategy did not utilize machine perfusion. As evidence accumulates, machine perfusion may become a standard adjunct for RLT, particularly during the learning curve or when extended warm ischemia times are anticipated.

Living donor partial grafts versus deceased donor whole grafts

A fundamental distinction among the three strategies lies in the type of graft used. The Asian strategy exclusively utilizes partial grafts from living donors (right lobe in 81.5%, left lobe in 16.7%) [14], whereas the American and European strategies employ whole grafts from deceased donors [10–12].

Partial grafts are smaller and more maneuverable, facilitating robotic manipulation and enabling partial caval clamping. The smaller size also reduces the time required for graft positioning and may contribute to shorter warm ischemia times (48 min in the Asian series versus 55–90 min in the European series) [11, 14]. Moreover, the Asian strategy’s distinctive “flopping” technique for left lobe grafts—rotating the graft 180° and placing it in the right hepatic fossa—demonstrates the adaptability of robotic platforms to accommodate different graft types [44].

Whole grafts, in contrast, are larger and require more space, contributing to longer warm ischemia times primarily due to graft positioning rather than anastomosis itself [10]. Chaouch et al. noted that larger grafts require more time for intra-abdominal placement and orientation, which is the main driver of prolonged warm ischemia in whole graft RLT [47]. However, whole grafts offer longer vascular pedicles that simplify anastomosis, which may be advantageous for centers without extensive experience in microvascular suturing. The European experience with routine machine perfusion may provide a safety margin that helps offset the challenges associated with whole grafts [11, 12].

For centers initiating RLT programs, the choice of graft type should align with their donor pool and technical expertise. Programs with access to living donors may benefit from the greater maneuverability of partial grafts during the learning curve, while those reliant on deceased donors should consider incorporating machine perfusion to mitigate the risks associated with prolonged warm ischemia.

Recipient selection: current limitations and future expansion

All three strategies have employed relatively strict patient selection criteria. The Asian strategy selected patients with MELD scores < 30, right lobe grafts, single artery and bile duct, and no complex vascular reconstruction [9, 14]. The American strategy required MELD < 25 and donor liver weight < 2000 g [10]. The European strategy emphasized small caudate lobe anatomy and BMI < 35 [11, 12]. These stringent criteria reflect the early stage of RLT development and the priority placed on patient safety.

However, these criteria also highlight a critical limitation: current RLT candidates represent only a small fraction of the overall transplant population. High-MELD patients, those with significant portal hypertension, previous upper abdominal surgery, or complex vascular anatomy are generally excluded. As Cherqui noted, high-MELD patients are too sick to withstand prolonged operative times, and those with major portal hypertension or portal vein thrombosis are poor candidates [48]. Consequently, the generalizability of current RLT outcomes to the broader transplant population remains uncertain.

As experience accumulates, indications are gradually expanding. The Asian strategy has already extended its criteria to include left lobe grafts (16.7% of cases) and retransplantation [14]. Future expansion to higher MELD scores, patients with significant portal hypertension, and marginal grafts will likely follow, supported by technical refinements and growing experience.

Emerging trend: convergence of strategies

A notable observation from this review is the emerging convergence of the three strategies. The European experience provides the clearest evidence, with the Modena team’s most recent case transitioning from total to partial caval clamping [12, 13]. This shift, explicitly noted in a recent editorial, suggests that European centers are increasingly adopting elements of the more hemodynamically stable Asian approach [13].

The American strategy may follow a similar trajectory. Khan et al. explicitly stated their intention to use veno-venous bypass routinely for the first ten cases before “reconsider transitioning to partial caval clamping without bypass” [10]. This planned evolution acknowledges that the optimal long-term approach may align more closely with the Asian strategy.

A potential future standard could therefore combine: partial caval clamping (for hemodynamic stability), no veno-venous bypass (for simplified workflow), Pfannenstiel incision (for cosmetic benefit and hand-assistance option), and routine machine perfusion (for graft protection during prolonged warm ischemia). Such a hybrid approach would incorporate the best elements of all three strategies while minimizing their respective drawbacks.

Learning curve and team requirements

Robotic liver transplantation has been aptly described as the “Everest” of hepatobiliary surgery [13]. The experience of pioneering centers underscores the steep learning curve and the necessity of extensive preparatory training, consistent with published consensus guidelines for minimally invasive liver surgery [39, 40].

Broering’s team performed over 1000 minimally invasive living donor hepatectomies before attempting recipient RLT [5, 6]. This experience is supported by a growing body of literature demonstrating the safety and feasibility of robotic donor hepatectomy [23–27, 32, 33]. Khan’s team completed over 600 robotic cases, including 300 complex HPB procedures such as major hepatectomy and pancreaticoduodenectomy, as well as 150 living donor nephrectomies and 40 kidney transplants, before their first RLT [10]. Both groups emphasized the importance of structured preparation, including cadaver simulations, transitional procedures (robotic hepatectomy with open implantation), and dedicated team training [9–11].

Regarding safety, several world-renowned transplant and laparoscopic surgeons have critically discussed the broader implications of RLT. Cherqui noted that liver transplantation is a life-saving procedure with established safety standards (3- and 12-month mortality < 5% and 10%, respectively), and the primary outcome measures should remain graft and patient survival rather than traditional minimally invasive metrics such as incision size or cosmetic appearance [48]. Pomfret, writing from the perspective of a nonrobotic surgeon, offered a more cautiously optimistic view. She acknowledged the promising early data showing reduced donor blood loss, shorter hospital stays, and fewer complications, but emphasized that the decision to adopt robotics for complex procedures should be grounded in compelling evidence rather than the assumption that all surgical advancements should be robotic by default [49].

Based on these experiences, a recommended pathway for centers considering RLT adoption would include:

1. Mastery of robotic donor hepatectomy [23–27].

2. Proficiency in complex robotic HPB procedures, including major resections and biliary reconstruction [34, 35].

3. Robotic hepatectomy with planned open implantation (transitional cases).

4. Cadaveric simulation of full RLT.

5. Proctored initial RLT cases with experienced mentoring.

As Magistri et al. caution, “proficiency in advanced robotic liver surgery—including vascular resection and reconstruction—is mandatory and remains the major obstacle for teams considering RLT” [12]. Attempting this procedure unprepared “can be unforgiving; one catastrophic complication could jeopardize an entire transplant program” [13].

Potential fatal complications and technical adjustments

Postoperative complications in the reviewed studies were graded according to the validated Clavien-Dindo classification [36], and overall morbidity was quantified using the Comprehensive Complication Index (CCI) where reported [37]. While the reported series demonstrate excellent short-term outcomes, potential fatal complications specific to the robotic approach warrant discussion. Based on the critical commentaries by Cherqui [48], Pomfret [49], and Chaouch [47], several key risks and corresponding technical adjustments can be identified.

Intraoperative bleeding remains the most feared complication. In the European series, two patients experienced significant bleeding (1,900 and 5,000 cc) requiring 5 and 8 units of packed red blood cells, respectively [47]. Notably, these patients had low MELD scores, minimal portal hypertension, and no coagulopathy—factors that would typically predict minimal bleeding in open surgery [47]. As Cherqui emphasized, major bleeding during robotic surgery is more challenging to manage than in open procedures because bleeding obscures the surgical field, may splash the optical system, and both loading needles onto robotic arms and conversion to open surgery are time-consuming [48]. To mitigate this risk, the use of a GelPort for hand-assistance (as in the Asian strategy) provides a valuable safety net for manual compression or rapid clamp placement [9, 13]. Emergency undocking protocols for uncontrolled bleeding must be established and rehearsed before undertaking RLT [50].

Gas embolism is another concern unique to robotic and laparoscopic approaches, as CO₂ pneumoperitoneum can enter disrupted venous vessels. This risk is theoretically increased during low central venous pressure conditions [50]. Intraoperative transesophageal echocardiography (TEE) monitoring is recommended for early detection [50]. Chaouch et al. noted that if a significant CO₂ embolism is detected, insufflation should be paused, 100% oxygen administered, and the patient placed in the left lateral decubitus position while central venous aspiration is attempted [47].

Prolonged warm ischemia time (55–90 min) represents a significant concern for graft function, particularly when using extended criteria donors. Chaouch et al. emphasized that while no graft dysfunction was reported in the selected low-MELD recipients, these observations may not be generalizable to marginal grafts [47]. Cherqui questioned whether prolonged warm ischemia would be acceptable for extended criteria donors [48]. Technical adjustments to mitigate this risk include: (1) optimizing graft positioning workflow (the American series reported 25 min for graft positioning alone [10]); (2) utilizing partial caval clamping to maintain some hepatic flow [14]; (3) employing machine perfusion as a protective strategy [11, 12]; and (4) considering a temporary portocaval shunt [47].

Prolonged portal vein clamping (estimated 2–3 h) can lead to significant bowel congestion and swelling, which reduces the limited pneumoperitoneal working space [47]. Chaouch et al. suggested that a temporary portocaval shunt, which has been shown to improve hemodynamic stability and reduce blood loss in open living donor liver transplantation, may be feasible in the robotic setting, though this has not yet been evaluated [47].

Vascular and biliary anastomotic complications are inherent to any liver transplantation procedure. Pomfret highlighted that robotic platforms offer advantages for managing complex complications, such as hepatic artery dissection, through enhanced magnification and tremor-eliminating instruments that enable the use of ultra-fine suture materials (7 − 0 or 8 − 0 Prolene) [49]. However, she emphasized that open conversion remains the safest default strategy when complications cannot be managed robotically [49].

Cardiopulmonary arrest during RLT, while rare, requires immediate action. Dutta et al. emphasized that emergency undocking protocols for cardiopulmonary arrest must be clearly defined and rehearsed, with distinct roles for the console surgeon, bedside assistant, scrub tech, circulator, and anesthesiologist [50]. These protocols should be reviewed during the surgical time-out before each RLT case [50].

Limitations

This review has several limitations that should be acknowledged. First, all included studies are small case series with inherent selection bias and lack of control groups. The only comparative study is from a single center [14], and no randomized controlled trials have been performed. Second, there is likely publication bias, with successful cases more likely to be reported than unsuccessful ones. Third, follow-up periods are relatively short, limiting assessment of long-term outcomes including chronic complications, late graft function, and infectious complications [38]. Fourth, cost-effectiveness analyses are lacking, which is particularly relevant given the significant expense associated with robotic platforms and disposable instruments.

Future directions

The field of robotic liver transplantation is in its infancy, and several directions for future development can be identified.

International registry

Establishment of a global RLT registry, such as the International Living Donor Liver Transplantation Outcomes Registry [15, 16], would facilitate data collection, enable identification of best practices, and support standardization of techniques. Regional chapters could tailor recommendations to local contexts, such as predominance of deceased versus living donor grafts [13].

Multicenter collaboration

Collaborative efforts among pioneering centers will be essential to accumulate sufficient experience, refine techniques, and develop evidence-based guidelines. The Lisbon-Modena collaboration [11] provides a model for such cooperation, and recent expert consensus statements have emphasized the importance of structured collaboration [39, 40].

Technological integration

Future RLT platforms may integrate advanced technologies including machine perfusion (already routine in Europe), artificial intelligence for intraoperative guidance [22], and enhanced imaging such as indocyanine green fluorescence [10]. Several recent reviews have highlighted the transformative potential of these technologies [41–43].

Expansion of indications

As experience accumulates, selection criteria will likely expand to include higher MELD scores, more complex vascular reconstructions, and potentially retransplantation. The Asian strategy has already demonstrated progressive expansion from low risk to more challenging cases [14]. A landmark development in this regard is the recent report by Rela et al. [51]. describing the world’s first fully robotic pediatric liver transplantation in a 5-year-old child with arginosuccinate lyase deficiency, demonstrating that robotic techniques can be successfully applied to pediatric recipients and complex transplant variants such as auxiliary partial orthotopic liver transplantation.

Training and standardization

Development of structured training programs, simulation curricula, and standardized operative protocols will be crucial for safe dissemination of RLT beyond pioneer centers. International consensus guidelines for minimally invasive liver surgery provide a framework for such efforts [39, 40].

Conclusion

Robotic liver transplantation has emerged as a technically feasible approach for selected recipients, representing a significant milestone in the evolution of minimally invasive transplant surgery. The three strategies described in the literature—each with distinct approaches to caval clamping, bypass use, and incision selection—demonstrate that the robotic platform can successfully address the complex vascular and biliary reconstructions inherent to liver transplantation. The Asian strategy, with its partial caval clamping, no bypass, and Pfannenstiel incision, has provided the largest body of evidence to date, showing reduced morbidity, faster recovery, and excellent short-term outcomes compared to open surgery. The American and European strategies have demonstrated the feasibility of robotic whole-graft transplantation from deceased donors, expanding the potential applicability of this approach.

Looking ahead, the future of robotic liver transplantation hinges on several key developments. Rigorous multicenter collaborations and international registries are needed to establish safety and efficacy in broader patient populations. Standardized training programs and emergency protocols must be developed to ensure safe adoption beyond pioneer centers. Cost-effectiveness analyses will be essential to justify the substantial investment in robotic platforms, and hybrid strategies combining laparoscopic donor surgery with robotic recipient surgery may offer a more economically viable pathway. Most importantly, as Cherqui emphasized, the primary outcomes must remain graft and patient survival, not incision size or cosmetic appearance.

In conclusion, robotic liver transplantation represents a remarkable technical achievement that has extended the frontiers of minimally invasive surgery. Its future, however, will be determined not by what is technically possible, but by what is safe, effective, and sustainable for patients. As the field evolves, the guiding principle must remain the Hippocratic dictum: “first, do no harm.”

Author contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Yuzhen Basang and Hanhua Dong. The first draft of the manuscript was written by Yuzhen Basang and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.