BMH Medical Journal

Abstract

Metabolic dysfunction-associated fatty liver disease (MAFLD) is now a leading cause of cirrhosis and an increasingly common indication for liver transplantation [1]. Decompensated disease is often accompanied by hepatic hydrothorax, for which liver transplantation is the definitive treatment [1]. Hepatopulmonary syndrome, which affects an estimated 10-30% of transplant candidates and generally resolves after transplantation [2]. Both conditions complicate anaesthetic management of living donor liver transplantation (LDLT).

We present a middle aged man with MAFLD-related Child-Pugh class C cirrhosis (MELD-Na 19), massive right hepatic hydrothorax and hepatopulmonary syndrome who underwent daughter-to-father successful LDLT. Intraoperative events included a diaphragmatic rent with hydropneumothorax, portal vein thrombectomy and 4700 mL blood loss. The postoperative course was complicated by persistent hydrothorax, an acute neuropsychiatric syndrome, catheter-related urinary Enterococcus faecalis infection, and partial thrombosis of the reconstructed middle hepatic vein graft. Everything was resolved due to timely intervention and he was discharged on day 16.

Key words: liver transplantation; hepatic hydrothorax; hepatopulmonary syndrome; thromboelastography; postoperative delirium.

Case Description

A middle aged man (89 kg) with decompensated chronic liver disease secondary to MAFLD was evaluated for living donor liver transplantation, with his daughter as the intended donor. He was classified as Child-Pugh class C (score 11) with a MELD-Na score of 19 and ASA physical status III. Manifestations of portal hypertension included hepatic hydrothorax with recurrent massive right pleural effusion, moderate ascites, small oesophageal varices treated with endoscopic variceal ligation after a prior upper gastrointestinal bleed, and controlled hepatic encephalopathy. He was jaundiced and had pancytopenia secondary to hypersplenism. His past history included vitiligo with a probable overlapping autoimmune/connective tissue disorder and polyarthralgia, previously treated with weekly methotrexate. He had undergone umbilical hernia repair and L4-L5 microdiscectomy for disc extrusion with cauda equina symptoms under general anaesthesia. There was no history of diabetes mellitus, hypertension, coronary artery disease, chronic kidney disease, alcohol use or smoking.

On examination his airway was adequate (Mallampati grade II). He had pedal oedema, room air saturation was 94% with markedly reduced air entry over the right mammary, infra-axillary and infrascapular regions, with dullness to percussion and tracheal deviation to the left, consistent with a massive right pleural effusion.

Blood parameters revealed platelet count 75-92 x 10³/micro litre, PT 16.8 s with INR 1.5-1.73, albumin 2.8 g/dL. His electrocardiography showed sinus rhythm with prolonged QT interval. Transthoracic echocardiography showed a preserved left ventricular ejection fraction (69%), no regional wall motion abnormality, mild mitral regurgitation, normal biventricular function and no intracardiac thrombus or vegetation. Contrast (bubble) echocardiography was consistent with hepatopulmonary syndrome. Dobutamine stress echocardiography showed no inducible wall motion abnormality, reaching a heart rate of 168 bpm with an adequate blood pressure response, and coronary angiography was normal. Pulmonary function testing could not be completed because of persistent cough and pleural effusion (single-breath oxygen count 8). Doppler ultrasound and contrast-enhanced CT of the abdomen showed a cirrhotic liver with an irregular nodular surface, a dilated portal vein with hepatopetal flow, splenomegaly, multiple portosystemic collaterals and moderate ascites, with no focal lesion suspicious for hepatocellular carcinoma. Chest radiography and CT of the thorax confirmed massive right hepatic hydrothorax with collapse of the right lower lobe and the medial segment of the right middle lobe, without pneumothorax. CT of the brain, performed as part of the pretransplant work-up, showed only mild age-related atrophy.

The operation was an orthotopic LDLT from his daughter, a first-degree related donor. The recipient was blood group AB negative and the donor B positive, with a graft-to-recipient weight ratio (GRWR) of 0.8. Biliary reconstruction was by hepaticojejunostomy. Cold ischaemia time was approximately 2 h 30 min and warm ischaemia time 3 min.

Vascular access comprised peripheral cannulae (20G right hand, 18G left forearm), 7Fr and 8.5 Fr central venous catheters and a 20G left radial arterial line. Monitoring included five-lead electrocardiography, non-invasive and invasive blood pressure, pulse oximetry, capnography, central venous pressure, cardiac output (FloTrac), entropy, temperature and urine output, with serial arterial blood gas analysis and viscoelastic testing (Thromboelastogram - TEG) at frequent intervals to guide metabolic correction and transfusion. After pre-oxygenation, general anaesthesia was induced with propofol in titrated dose, fentanyl and cisatracurium with neuromuscular monitoring. The trachea was intubated with an 8.0 mm cuffed tube (fixed at 11 cm at the lip) and controlled ventilation was started. Anaesthesia was maintained with sevoflurane (average end-tidal concentration about 2%) in an oxygen-air mixture, with a remifentanil infusion (about 100 micro gram/h) and a cisatracurium infusion. Total anaesthesia duration was 12 hours 20 minutes.

Intraoperative events included massive blood loss (4700 mL), a diaphragmatic rent with hydropneumothorax and multiple air leaks, repaired intraoperatively with intercostal drain placement, and a portal vein thrombus extending to the confluence, which required eversion thrombectomy before the anastomosis. Fluid management comprised of balanced crystalloids and of 20% albumin; 1250 mL of ascitic fluid was drained and urine output was about 2500 mL. He received 5 units of packed red blood cells and 5 units of cryoprecipitate.

The anhepatic phase lasted 2 h 16 min. Noradrenaline, vasopressin and adrenaline infusions were used throughout the anhepatic and reperfusion phases, titrated to mean arterial pressure targets, together with bicarbonate buffering and calcium replacement for the evolving metabolic acidosis. Reperfusion was uneventful, with no significant post-reperfusion syndrome, arrhythmia or hyperkalaemia.

Postoperatively, the patient was managed in the liver transplant ICU with invasive haemodynamic monitoring and a urinary catheter, with mean arterial pressure kept above 65 mmHg. Doppler on postoperative day 1 showed the hepatic artery with initially reduced diastolic flow, which normalised with temporary portal vein clamping. This was attributed to a transient hepatic arterial buffer response or graft-oedema-related change, and serial Doppler follow-up was advised. Early ultrasound showed small perigraft fluid loculations and mild bilateral pleural effusion. Immunosuppression comprised of tacrolimus (dose adjusted to target trough levels), tapering dose of methylprednisolone and mycophenolate mofetil. Antimicrobial prophylaxis, enoxaparin for deep vein thrombosis prophylaxis, a proton pump inhibitor and thiamine were also given.

He was extubated on day 2 and required supplemental oxygen for the persistent right pleural collection, initially by high-flow nasal cannula, together with incentive spirometry, chest physiotherapy and early mobilisation. The persistent right hepatic hydrothorax was managed with an intercostal drain, with cumulative drainage of about 1950 mL. Drainage decreased progressively, and serial chest radiographs and ultrasound confirmed pleural fluid clearance and lung re-expansion. Serial biochemistry showed improving AST/ALT, a progressive fall in bilirubin and stable INR (about 1.5), with persistent thrombocytopenia requiring monitoring.

Around postoperative day 4 he developed an acute neuropsychiatric syndrome with restlessness, agitation, confusion, disorientation, sleep disturbance, visual hallucinations, delusional behaviour, poor attention and a fluctuating sensorium, which gradually resolved with multidisciplinary input and psychotropic titration. The course was further complicated by catheter-related and urinary infection with Enterococcus faecalis. As his respiratory status improved, the central venous catheter and arterial line were removed, drains were changed to colostomy-type bags, and the intercostal drain was removed after satisfactory lung re-expansion. He was shifted to the ward on postoperative day 6. Contrast-enhanced CT of the abdomen and pelvis on postoperative day 11 showed partial thrombosis of the reconstructed middle hepatic vein (MHV) graft, reflecting venous congestion from impaired hepatic venous outflow rather than infarction. The hepatic arterial, portal venous, hepatic venous and IVC anastomoses were patent, with no parenchymal necrosis or biliary dilatation. A small hepatic hilar fluid collection with subsegmental atelectasis, splenomegaly and small bilateral pleural effusions (right greater than left) were also noted.

He was weaned to oral feeds, mobilised progressively and became ambulant with assistance. He was discharged on postoperative day 16 in stable condition, with satisfactory graft function, resolving hydrothorax and improving neuropsychiatric status. The partial MHV graft thrombosis was managed with escalated low-molecular-weight heparin and later antiplatelet therapy. Explant histology of the native liver showed stage IV (MAFLD-related) cirrhosis. Total hospital stay was 17 days, including 6 days of ICU care.

Discussion

Osman et al. [1] found that liver transplantation for refractory hepatic hydrothorax is associated with a greater survival benefit than transplantation for other complications of cirrhosis. Raevens et al. [2], reviewing hepatopulmonary syndrome, noted that it affects an estimated 10-30% of transplant candidates and usually resolves after transplantation. In our patient, massive hepatic hydrothorax with hepatopulmonary syndrome posed the principal anaesthetic challenge. Baseline hypoxaemia limited the margin for preoxygenation and shaped a cautious, staged respiratory weaning (mechanical ventilation, then HFNC, then intercostal drain, then room air) rather than early extubation.

Lai et al. [3] studied hepatopulmonary syndrome in a large LDLT cohort and found that it was associated with more frequent delayed extubation and pleural drainage, but that overall survival was not adversely affected irrespective of severity. Neither condition should therefore be regarded as a contraindication to transplantation, and our patient's need for an intercostal drain and staged respiratory support is in keeping with this pattern.

Elsarawy et al. [4] reported a high-volume single-centre experience of LDLT in patients with portal vein thrombosis, in which complete thrombectomy was achieved in most patients and post-transplant complication rates did not differ from those without thrombosis. Portal vein thrombosis is thus no longer considered an absolute contraindication to LDLT. In our case, the intraoperative portal vein thrombectomy added technical complexity to an anhepatic phase of 2 h 16 min that already required bicarbonate buffering, calcium replacement and combined noradrenaline, vasopressin and adrenaline support.

Aceto et al. [5], in a systematic review and meta-analysis, found that viscoelastic testing was associated with reduced red cell and plasma consumption in liver transplantation compared with conventional coagulation tests, without an effect on mortality. In our patient, transfusion was guided by TEG rather than fixed ratios, and viscoelastic parameters directed cryoprecipitate use for the 4700 mL blood loss. Reperfusion was uncomplicated, which likely reflects effective pre-reperfusion electrolyte correction and vasopressor titration.

Hou et al. [6], in a meta-analysis of postoperative delirium in organ transplant patients, reported an overall incidence of about 20%, with a history of hepatic encephalopathy and a higher MELD score among the recognised risk factors. Both were present in our patient, who developed an acute neuropsychiatric syndrome around postoperative day 4 that resolved with multidisciplinary management. The catheter-related urinary infection and the partial hepatic venous graft thrombosis were likewise managed by the transplant team.

Cekmen and Uslu [7], in a narrative review of anaesthesia for liver transplantation, emphasised anticipating rapidly changing haemodynamic, metabolic and coagulation status through each phase of the operation, and a multidisciplinary approach. Our case followed this phase-by-phase approach, with invasive haemodynamic and viscoelastic-guided monitoring through the anhepatic and reperfusion phases.

Conclusion

This case highlights the anaesthetic management of LDLT in a patient with Child-Pugh class C cirrhosis, massive hepatic hydrothorax and hepatopulmonary syndrome. It was managed with invasive haemodynamic and viscoelastic-guided monitoring, staged respiratory support, and close attention to the anhepatic and reperfusion phases, despite added technical complexity from portal vein thrombosis. The patient was discharged on postoperative day 16 with satisfactory graft function after a course complicated by persistent hydrothorax, postoperative delirium, infection and partial hepatic venous graft thrombosis, which underlines the need for vigilance for pulmonary, infectious, vascular and neuropsychiatric complications.

References

1. Osman KT, Naritsin A, Amuchi B, Qamar AA. Liver transplantation for refractory hepatic hydrothorax is associated with a greater survival benefit compared to other complications of cirrhosis. Liver Transpl. 2024;30(2):142-150. doi:10.1097/LVT.0000000000000218

2. Raevens S, Boret M, Fallon MB. Hepatopulmonary syndrome. JHEP Rep. 2022;4(9):100527. doi:10.1016/j.jhepr.2022.100527

3. Lai Y, Hung HC, Lee JC, Wang YC, Cheng CH, Wu TH, et al. Hepatopulmonary syndrome and liver transplantation: impact on survival and postoperative complications. Ann Transplant. 2024;29:e945297. doi:10.12659/AOT.945297

4. Elsarawy A, Akbulut S, Aktas S, Kilercik H, Alkara U, Sevmis S. Living donor liver transplantation for patients with portal vein thrombosis: high-volume single center experience. Eur Rev Med Pharmacol Sci. 2024;28(11):3752-3760. doi:10.26355/eurrev_202406_36380

5. Aceto P, Punzo G, Di Franco V, Teofili L, Gaspari R, Avolio AW, et al. Viscoelastic versus conventional coagulation tests to reduce blood product transfusion in patients undergoing liver transplantation: a systematic review and meta-analysis. Eur J Anaesthesiol. 2023;40(1):39-53. doi:10.1097/EJA.0000000000001780

6. Hou SS, Liu J, Qiao PF, Yang DG, Huang LF, Liu F, et al. Meta-analysis of the incidence and risk factors of postoperative delirium in organ transplant patients. World J Psychiatry. 2025;15:104812.

7. Cekmen N, Uslu A. Anaesthesia management for liver transplantation: a narrative review. J Perioper Pract. 2024;34(7-8):226-240. doi:10.1177/17504589231193551

Perioperative journey of a living donor liver transplantation in a patient with Child-Pugh class C cirrhosis and massive hepatic hydrothorax