One serious side effect of liver resection is called post-hepatectomy liver failure (PHLF), which is highly correlated with the amount of liver resection and occurs approximately 10% of the time following major liver resection [1]. Severe PHLF is associated with a 90-day mortality rate post-surgery as high as 60%; in nearly half of the cases, mortality occurs within 30 postoperative days [1]. Rescue liver transplantation (LT) has been suggested as a potentially life-saving procedure for individuals with terminal PHLF [2].

Severe PHLF can result in cerebral edema and intracranial hypertension (ICH), which can frequently cause brainstem herniation and death [1]. Pupil size and response are observed in progressive encephalopathy because fixed dilated pupils are a clinical indicator of brain death. The first deceased donor liver transplantation (DDLT) for a patient with permanent dilated pupils who has PHLF is reported here.

A 48-year-old male presented to the emergency room with an altered mental status. He had diabetes, dyslipidemia, and liver cirrhosis due to hepatitis B virus, for which he had received tenofovir and undergone several endoscopic varix ligations due to esophageal varix at another hospital 2 months ago. At the present institute, he was diagnosed with multiple hepatocellular carcinoma in the right lobe and underwent a laparoscopic right hepatectomy. Total operation time was 173 minutes, and blood loss during the surgical procedure was 100 mL. Pathology reported three 1.4 cm tumors in the right lobe. Postoperative platelet counts and liver function values were stable or decreased after the operation. The patient was discharged on postoperative day 7 (POD #7) without complications (Fig. 1).

Figure 1. (A) Multiple tumors in preoperative liver images (arrow), (B) specimen after laparoscopic right hepatectomy, (C) change of platelet count after surgery, (D) change of international normalized ratio after surgery, and (E) change of total bilirubin after surgery.

However, the patient visited the emergency room because of hematemesis and altered mentality at POD #11. His lab parameters were as follows: hemoglobin, 9.9 mg/dL; interntional normalized ratio, 3.04; total bilirubin, 3.2 mg/dL; aspartate transaminase, 291 IU/L; alanine transaminase, 108 IU/L; lactic aicd, 13.05 mmol/L; ammonia, 96.8 µmol/L. The hepatic encephalopathy was grade III with bilateral equal mid-dilated pupils reacting to light. Liver computed tomography (CT) showed multiple low attenuated nodular lesions in the remnant liver, thrombosis in the main portal vein, hypertensive enterocolopathy, and a moderate amount of ascites (Fig. 2A). He underwent an emergency endoscopy to detect or diagnose bleeding sites in the upper gastrointestinal tract, and multiple active bleeding sites at esophageal varices were ligated (Fig. 2B). Invasive hemodynamic monitoring aided in directing fluid management. Low-dose noradrenaline preserved hemodynamic stability. A core temperature of 35℃–36℃ was maintained for the patient. Blood sugar, electrolytes, and arterial blood gases were all tracked and controlled. Because of postoperative liver failure brought on by ischemic liver damage from massive blood loss in esophageal varices, we advised his family members that he would receive a LT family members were advised to either find a living liver donor for an extremely urgent living donor liver transplantation or to permit DDLT.

Figure 2. (A) Multiple hypoattenuating lesions in the remnant liver, (B) bleeding focus on endoscopy in the emergency room, (C) blood oozing in the gastroesophageal junction in the intensive care unit, (D) explant liver, and (E) harvested liver graft from a deceased donor.

However, the patient’s blood pressure decreased to 45/38 mmHg, and he required massive red blood cell transfusions, fresh frozen plasma, platelet concentrates, and inotropes to maintain his blood pressure. He also received vasopressin and epinephrine to increase blood pressure. We urgently consulted an endoscopist for investigation of another bleeding focus in the stomach or esophagus, along with corrective action and treatment. Additional bleeding sites were found and ligated endoscopically (Fig. 2C). A Sengstaken-Blakemore tube was inserted because of severe bloody oozing in the esophagogastric junction.

Around 24 hours post-admission, both pupils demonstrated sluggish reactions to light, which were managed with intra-venous mannitol (0.5 g/kg). A head CT conducted at this time ruled out significant cerebral edema, cerebral bleeding, and brain stem herniation. Fortunately, he was authorized for DDLT at 1.5 days post-admission, but his pupils at that point were fixed on the light reflex examination. Electroencephalogram suggested severe diffuse cerebral dysfunction, and the Glasgow Coma Scale score was 3 points for eye-opening, verbal response, and motor response. We confirmed blood flow via transcranial doppler and positive light reflex on the pupilometer. DDLT was planned as the last option to prevent mortality. In preparation, he received 32 pints of red blood cells, 54 pints of fresh frozen plasma, and six pints of platelet concentrate before entering the operation room.

The abdominal incision revealed a blood-filled and severely distended stomach and small intestine. To secure a visual field, an incision was made in the stomach and jejunum, all blood was removed, and the bleeding site in the stomach was sutured. The amount of blood in the gastrointestinal tract was about 4,000 mL. After the entire liver was removed, a portocaval shunt was created to reduce portal pressure. The total operation time was 120 minutes (Fig. 2D, E).

To oversee standard post-transplant treatment, the patient was sent to the intensive care unit. He received ongoing renal replacement therapy and mechanical ventilation as treatment. Inotropes were tapered off and discontinued gradually. The first signs of a slow pupillary response to light appeared about 6 hours after transplant. After two days, he started to show signs of eye opening and his mental state shifted to alert. Over the course of the following few days, the patient gradually recovered. After surgery, his pupils reacted to light response within 6 to 8 hours. The graft function satisfied the laboratory parameters for evaluation (Fig. 3). Feeding via nasogastric tube was initiated and increased based on tolerance. Tenofovir was added to prevent hepatitis B virus recurrence. Over the next week, the patient's muscle power returned. At POD #5, ventilatory assistance was stopped, and extubation was carried out. The patient's subsequent course was unremarkable; on POD #7, they were moved to the general ward, and on POD #28, they were released from the hospital without experiencing any neurological aftereffects. The patient is currently receiving routine outpatient follow-up care and is doing well.

Figure 3. Post-transplant laboratory changes. (A) AST, (B) ALT, (C) total bilirubin, (D) platelet count, and (E) serum creatinine. AST, aspartate transaminase; ALT, alanine transaminase; CRRT, continuous renal replacement therapy.

Brainstem herniation may result from progressive cerebral edema brought on by an initial worsening of liver function in PHLF patients [2]. This could reduce the survival of comatose patients awaiting LT until a deceased donor is assigned or an appropriate donor is found. Grade III/IV encephalopathy is present in some liver failure patients, and they may also have ICH, which calls for continuous pupil size and response monitoring. Because of the increase in intracranial pressure and irreparable brain injury in patients with acute liver failure, the persistent development of fixed dilated pupils is usually interpreted as an indication of brainstem herniation [3,4]. Prior to halting the transplant, it was essential to verify the brain’s condition when fixed dilated pupils developed. Transcranial Doppler ultrasonography and pupillometry were used to rule out brain death and brain stem herniation. The unpredictable nature of the relationship between fixed dilated pupils and brain death is demonstrated by a number of cases in the literature [5,6].

After receiving large dosages of sympathomimetics, such as dopamine, noradrenaline, or parasympathetic medications, pupils may dilate and become non-reactive. Other diverse causes of anoxia, such as cardiac arrest, hypothermia, severe anesthesia, and cyanide, methanol, propranolol, or barbiturate overdose, have been linked to fixed dilated pupils [5]. The acute liver failure protocol required our patient to be hypothermic, and bi-spectral index was used to track the level of anesthesia.

Normally, medications that prevent neuromuscular depolarization cannot pass through the intact blood-brain barrier [5]. The blood-brain barrier may be disrupted in acute liver failure patients even though it is mainly physically intact due to increased pinocytotic activity, endothelial edema, oxidative stress, and inflammation [5]. However, a pupillary reaction was noticed when our patient was using muscle relaxants. The steady and gradual alteration in pupil size and response was ascribed to an increasing intracranial pressure. However, the use of vasopressors following DDLT provides another explanation. Our brain stem herniation was ruled out by pupillometry and transcranial doppler, which also revealed a positive light reflex and normal cerebral blood flow.

In summary, this scenario emphasizes the ambiguous consequences of fixed dilated pupils while choosing DDLT. The transplantation team was advised to consider brain activity by a positive light reflex test using a pupilometer, even if the cause of the patient's fixed dilated pupils is still unknown. While we do not advocate routine LT for patients with acute liver failure who have fixed dilated pupils, we do encourage cerebral imaging to facilitate accurate diagnosis.

This research was supported by the Basic Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT (NRF-2023R1A2C2005946). The Korean NRF had no influence on study design, data analysis, data interpretation, or drafting of the manuscript.

Jongman Kim is a chief editor of the journal but was not involved in the review process of this manuscript. Any other authors have no conflict of interest.

Conceptualization: JK. Data curation: SA, JK, SJ, NO, EG, GK. Formal analysis: SA, JK. Funding acquisition: JK. Investigation: SA, JK. Methodology: JK. Project administration: JK. Resources: JK. Software: NO, EG. Supervision: EG, GK. Validation: EG, GK. Visualization: SA. Writing – original draft: SA, JK. Writing – review & editing: JK.