Liver Transplantation for the Surgeon — Select by Biology, Preserve the Cava, Watch Artery and Duct

The transplanted liver wants three verdicts before the knife — whose cancer biology earns an organ, whose failure cannot wait, and whose vessels and ducts will hold the graft — because size criteria select but biology decides, allocation follows death risk not waiting time, and the artery feeds the bile duct whose stricture then strangles the graft. Select HCC by Milan with AFP dynamics, list fulminant failure by King's thresholds, allocate by MELD, preserve the cava by habit with flushing by preference, reconstruct one good artery, stent the leaking or strictured duct endoscopically, grade the portal thrombus, size the graft to its inflow, perfuse the marginal, protect the donor, and retransplant with young whole organs — with every number taken from the papers named beside it.^{[1][6][3][4][22][11]}

A 58-year-old man with cirrhosis and two small HCCs; a 24-year-old woman with paracetamol fulminant failure and a pH of 7.25; and a 60-year-old recipient on day 8 with no Doppler arterial signal. One needs biology-tested listing, one needs a graft tonight, and one needs revascularisation or relisting today. The examiner will watch you quote Milan against UCSF with AFP reclassification, apply King's thresholds from memory of the paper, defend MELD with its mortality curve, concede piggyback and flushing equipoise, price HAT rescue with graft-survival arithmetic, and manage the strictured duct endoscopically — with every number taken from the papers named beside it.^{[1][2][6][3][4][10][11]}

Selection is three separate lotteries — cancer, failure and waiting-list death — each with its own validated ticket: Milan size for HCC with AFP dynamics reclassifying the outsized, King's thresholds for fulminant failure, and MELD mortality for allocation with HCC priority studied, not assumed.^[1][3][4][5] The strategic arc fits one sentence: prove biology, list by validated urgency, implant with caval preservation and one artery, reperfuse by preference, rescue the clotted artery fast, drain and stent the duct endoscopically, grade thrombus and size grafts haemodynamically, perfuse the marginal and DCD, split for two, guard the donor, and retransplant the failing graft with better organs.^{[6][7][22][11][15][13][19][20]}

Select cancer by size, then by biology

Milan still opens listing: a single tumour 5 cm or less, or up to three nodules each 3 cm or less, delivered 75% survival with 83% recurrence-free at four years and 8% recurrence across 48 patients — with criteria-meeters at 85 and 92% against 50 and 59% for exceeders.^[1] UCSF modestly widened the door: a solitary tumour 6.5 cm or less, or three nodules with the largest 4.5 cm or less totalling 8 cm or less, held 90% at one year and 75.2% at five — against 50% at one year beyond — with AFP above 1,000, total diameter above 8 cm, age and grade as univariate warners.^[2] Biology now re-sorts both: across 2,236 transplants the AFP-response score split five-year recurrence into 9.5, 20.5 and 40.5% risk bands, beat Milan on discrimination, and reclassified four-fifths of Milan-out patients into low-or-acceptable risk — so AFP dynamics, not centimetres alone, decide the borderline.^[6] Priority for HCC has been tested, not assumed: cutting exception scores left national transplant rates, waits and survivals unchanged — though regions varied, pointing reform at geography rather than generosity.^[5]

List failure by thresholds, allocate by death risk

Paracetamol failure declares itself in numbers: pH below 7.30, prothrombin time above 100 seconds and creatinine above 300 micromol per litre mark the dying — while non-paracetamol failure needs aetiology, age extremes, a week of jaundice before encephalopathy, bilirubin above 300 and prothrombin time above 50 seconds.^[3] Allocation follows predicted death, not patience: across 3,437 candidates MELD mortality ran 1.9% below 9 to 71.3% at 40 and above, discriminating three-month death at 0.83 against 0.76 for Child-Pugh — so the sickest, not the longest-waiting, is first.^[4]

Implant by habit, reperfuse by preference

Caval preservation is habit, not evidence: two randomised trials in 106 patients found piggyback and conventional with bypass tied on death, non-function, vascular, renal, transfusion and stay — with only warm ischaemia shorter and chest complications higher — so no method is recommended or refuted.^[7] Flushing and reperfusion share the equipoise: six trials in 418 patients found no mortality, survival or severe-morbidity difference across any flush or reperfusion order — absence of evidence, explicitly not evidence of absence.^[8] Aberrant arteries obey one rule: reconstruct the largest, Doppler both lobes, ligate the rest — with biliary, non-function, dysfunction, arterial and survival outcomes tied across 618 grafts, sparing needless complex reconstructions.^[22] Reperfusion still drops pressure: a 39-patient randomised trial of 1500 mg anhepatic ascorbic acid trended 30% to 10.5% without significance — while renal failure and retransplantation clustered harmfully in the treated arm, so antioxidants stay experimental.^[9]

Rescue the artery, stent the duct

Thrombosis is the graft's second killer: 5% of grafts clot, second only to primary non-function as graft loss, killing up to three-fifths untreated — so absent Doppler signal means angiography or theatre now, with endovascular retrieval where organs are scarce.^[23] Failed revascularisation is quantified despair: across ten centres, occlusion at a median 8 days brought 56% biliary and 76% infectious complications, graft survival of 48, 31 and 23% at one, three and five years, and retransplantation in seven — with failure past 30 days costing lives.^[10] The duct strictures in one-tenth: 9.8% across 717 grafts over 15 years, driven by leak sixfold, arterial thrombosis twofold and rejection twofold — with metal-stent ERCP an effective low-complication answer for benign anastomotic disease.^[11] Leaks bridge, strictures dilate: leak-bridging stents succeed at 91 against 53%, anastomotic leaks resolve at 90% endoscopically, and covered metal resolves four-fifths of strictures in 2.5 procedures — with main-duct leaks and leak-plus-stricture pairs conceding lower odds.^[24][25]

Grade the thrombus, size to the inflow

Thrombus grade is prognosis: post-transplant clot rises 14.3, 52.4, 71.4 and 85.7% across grades 0 to III, with grade III carrying sixfold hazard and three-year survival splitting 76.2, 88.1 and 95.2% by risk — so dual-modality imaging grades every portal vein pre-operatively with anticoagulation stratified after.^[15] Weight is stratification, not a wall: below 0.80% graft-to-recipient weight neither small-for-size syndrome nor hospital death rose, with long survival equal when inflow is controlled — though over half still meet small-for-size criteria, tamed by flow below 5 ml per minute per gram and pressure below 20 mmHg with arterial inflow preserved.^[16]

Split for two, perfuse the marginal, expand DCD

Splitting serves under 5% of American grafts yet rewards matching: right lobes beat left at 91.7 against 79.6% one-year survival, adequate weight and volume match whole-organ results, and splits trail whole grafts only through left lobes, small grafts and older recipients.^[19] Perfusion earns its place on bile ducts first: nine randomised trials cut non-anastomotic strictures by three-fifths with both hypothermic and normothermic techniques, while anastomotic disease stands unchanged.^[12] Seven first trials in 1,017 patients agree: both techniques halve early dysfunction, hypothermic perfusion cuts major complications, retransplantation and graft loss — with one-year horizons and longer follow-up still owed.^[13] Normothermic perfusion rescues DCD boldly: across 586 grafts including futile strata, two-year graft survival ran 91 against 80% and patient 96 against 89%, best where risk was highest — making perfusion essential, not optional, for the marginal.^[14] Living donation now ties perfused DCD nationally: across 6,105 recipients, unperfused DCD nearly doubled graft-failure hazard while living donation matched perfused DCD on failure and death — with living donation static and DCD-perfusion spreading.^[21]

Guard the donor above all

Donor death stays below 0.4% with quality of life at or above population norms and most donors recovered within months — yet physical, psychological and financial scars persist in some, with small remnants, female sex, slow recovery and thin support predicting worse.^[17] Complications cluster predictably — biliary, bleeding, respiratory, gastrointestinal, infectious — with mortality very low for either lobe and laparoscopy recommended where volume lives.^[18]

Retransplant with better organs and open eyes

Up to one in ten grafts is a re-graft, most within a year: primary non-function has fallen from 8% to below 2%, yet non-function and arterial thrombosis still lead early emergency relisting with severe biliary, vascular and refractory rejection behind.^[20] Survival always trails the first graft, and the dying declare themselves — high MELD, intensive care, renal failure and marginal organs — so centres answer with young whole brain-dead grafts, increasingly machine-perfused for these hardest cases.^[20]

The cohorts behind the numbers run Mazzaferro 48-patient Milan, Yao 70-patient UCSF, O'Grady 588-patient King's derivation with 175-patient validation, Wiesner 3,437-candidate MELD allocation, Sharma UNOS HCC-priority analysis, Halazun 2,236-patient AFP validation, Gurusamy piggyback and flushing Cochranes, Gajate 39-patient ascorbic trial, Hara 25-patient revascularisation-failure cohort, Eslami 717-graft biliary series, Wang nine-trial stricture meta, Parente seven-trial perfusion meta with 1,017 patients, Ohara 586-graft DCD cohort, Liu 126-patient PVT grading, Raptis 202-graft sizing review, Taj donor synthesis, Magyar donor-safety review, Sierra 840-split registry analysis, Likhitsup retransplant review, Shirini 6,105-patient graft-source comparison, Pravisani 618-graft arterial policy, Meek thrombectomy report, Quintini 65-patient leak series, and Canakis 43-patient covered-metal series — randomised where randomisable, pooled where small, registry where allocation lives.^{[1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25]}

Biology, artery and inflow close the traps: the oversized HCC is listed on AFP response, not centimetres, since dynamics reclassify; the fulminant failure is listed on thresholds tonight, since windows slam shut; the silent artery is imaged and reopened or relisted today, since failed flow halves grafts yearly; the small graft lives on controlled inflow, not weight alone; the DCD graft is perfused as routine, never as rescue; the donor is consented on real numerators below 0.4%; and the failing graft is replaced with young whole perfused organs, since retransplantation forgives nothing.^{[6][3][10][16][14][17][20]}

Bridge leaks exactly: leak-bridging 91 vs 53% with anastomotic 90%; covered metal resolves 81% in 2.5 ERCPs (PMIDs 39384654/37162469).^[25] Grade portal thrombus exactly: 126 patients at 14.3/52.4/71.4/85.7% with grade III x6.2 and survival 76/88/95% (PMID 41913805).^[7] Size haemodynamically exactly: 0.80% not tied to syndrome or death when flow under 5 and pressure under 20 with arterial inflow kept; 56.9% still meet syndrome (PMID 42034478).^[16]

References25ShowHide

1. [1]Mazzaferro V, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med, 1996.PMID 8594428
2. [2]Yao FY, et al. Liver transplantation for hepatocellular carcinoma: expansion of the tumor size limits does not adversely impact survival. Hepatology, 2001.PMID 11391528
3. [3]O'Grady JG, et al. Early indicators of prognosis in fulminant hepatic failure. Gastroenterology, 1989.PMID 2490426
4. [4]Wiesner R, et al. Model for end-stage liver disease (MELD) and allocation of donor livers. Gastroenterology, 2003.PMID 12512033
5. [5]Sharma P, et al. Reduced priority MELD score for hepatocellular carcinoma does not adversely impact candidate survival awaiting liver transplantation. Am J Transplant, 2006.PMID 16771808
6. [6]Halazun KJ, et al. Dynamic α-Fetoprotein Response and Outcomes After Liver Transplant for Hepatocellular Carcinoma. JAMA Surg, 2021.PMID 33950167
7. [7]Gurusamy KS, et al. Piggy-back graft for liver transplantation. Cochrane Database Syst Rev, 2011.PMID 21249703
8. [8]Gurusamy KS, et al. Techniques of flushing and reperfusion for liver transplantation. Cochrane Database Syst Rev, 2012.PMID 22419324
9. [9]Gajate Martín L, et al. A randomized trial of ascorbic acid for the prevention of post-reperfusion syndrome during liver transplantation. Hepatol Commun, 2025.PMID 40719561
10. [10]Hara T, et al. Outcomes After Unsuccessful Revascularization Following Hepatic Artery Occlusion in Living Donor Liver Transplantation: Results From an International Multicenter Study. Clin Transplant, 2026.PMID 42138888
11. [11]Eslami O, et al. Anastomotic biliary stricture following liver transplantation and management analysis: 15 years of experience at a high-volume transplant center. Indian J Gastroenterol, 2022.PMID 35838868
12. [12]Wang X, et al. Machine Perfusion Ameliorates Nonanastomotic Strictures After Liver Transplantation: A Systematic Review and Meta-analysis of Randomized Controlled Trials. Clin Transl Gastroenterol, 2026.PMID 41604272
13. [13]Parente A, et al. Machine perfusion techniques for liver transplantation - A meta-analysis of the first seven randomized-controlled trials. J Hepatol, 2023.PMID 37302578
14. [14]Ohara SY, et al. Revisiting Futility: Normothermic Perfusion Improves Outcomes Across all UK DCD Risk Strata. Clin Transplant, 2026.PMID 42697564
15. [15]Liu Y, et al. Preoperative Portal Vein Thrombosis Grading Predicts Post-Transplant Thrombosis and Survival. Ther Clin Risk Manag, 2026.PMID 41913805
16. [16]Raptis DA, et al. Is the 0.80% graft-to-recipient weight threshold still relevant in modern adult living donor liver transplantation? Hepatobiliary Pancreat Dis Int, 2026.PMID 42034478
17. [17]Taj R, et al. Long-Term Outcomes of Living Liver Donors. Gastroenterol Clin North Am, 2026.PMID 42203400
18. [18]Magyar CTJ, et al. The aim of donor safety: surgical approaches and current results. Updates Surg, 2025.PMID 38916620
19. [19]Sierra L, et al. Demystifying Split Liver Transplantation in Adults: Factors Associated with Early Graft Survival in the New Era. J Gastrointest Surg, 2026.PMID 41962895
20. [20]Likhitsup A, et al. Indications and Outcomes with Liver Retransplantation in 2025. Dig Dis Sci, 2025.PMID 39576429
21. [21]Shirini K, et al. Living donor liver transplantation versus donation after circulatory death in the era of normothermic machine perfusion: Insights from a national cohort. Surgery, 2026.PMID 42705178
22. [22]Pravisani R, et al. Graft aberrant hepatic arteries in deceased donor liver transplantation: The "one liver, one artery" approach. Liver Transpl, 2022.PMID 35751148
23. [23]Meek JC, et al. Use of a mechanical thrombectomy device to treat early hepatic artery thrombosis after orthotopic liver transplant. Radiol Case Rep, 2018.PMID 29904504
24. [24]Quintini D, et al. Endoscopic or combined management of post-surgical biliary leaks: a two-center recent experience. Surg Endosc, 2024.PMID 39384654
25. [25]Canakis A, et al. Management of biliary complications in liver transplant recipients using a fully covered self-expandable metal stent with antimigration features. Minerva Gastroenterol (Torino), 2024.PMID 37162469