Detection of Graft Nonfunction After Liver Transplantation by Assessment of Indocyanine Green Kinetics

Abstract

Indocyanine green dye (ICG) has been used in tests for a global measure of liver perfusion and excretory function for more than three decades [1] and has shown good correlation with the severity of hepatic disease as well as with outcome in liver transplant recipients [2-5]. We report a case of the early detection of graft nonfunction after orthotopic liver transplantation (OLT) using a bedside monitoring device for fiberoptic assessment of ICG plasma disappearance rate (ICG (PDR)) (normal value 20%-30%) [2] in a 43-yr-old woman who underwent OLT because of cholangiocellular carcinoma. Retransplantation was performed, and the patient was dismissed from the hospital 3 wk after the first OLT. Case Report A 43-yr-old, 59-kg female patient was postoperatively admitted to the intensive care unit (ICU) after she had undergone uncomplicated elective OLT for the treatment of malignancy. On admission, the patient was endotracheally intubated, and the lungs were mechanically ventilated. She was hemodynamically stable. Arterial blood gas analysis revealed no metabolic disorders. Laboratory samples showed reduced synthesis of clotting factors and an increase of serum transaminases. Measurement of gastric intramucosal PCO2 and calculation of intramucosal pH revealed no abnormalities. Postreperfusion ICGPDR (24.8%) and ICGPDR on admission (20.3%) indicated good functioning of the graft initially. The ICGPDR was evaluated after a single bolus injection, via a central venous line into the right atrium, of ICG (0.3 mg/kg body-weight), diluted in 10 mL of dextrose 5%. Exponential decay time for the dye indicator over a period of 240 s was recorded using a 3F fiberoptic thermistor-tipped catheter advanced via the right femoral artery into the descending aorta, and ICG (PDR) was calculated using the COLD-Z021[trade mark sign] system (Pulsion Medical Systems, Munich, Germany). ICGPDR performed routinely three times daily demonstrated a decrease to 5% approximately 8 h after admission and a value of 0% 2 h later. Doppler ultrasound examination was conducted by an experienced investigator. The results, however, were inconclusive, mainly because of anatomic difficulties/bowel movement and intestinal gas. The patient was immediately scheduled for angiography, which showed a complete stop of blood flow in the hepatic artery, as well as in the portal vein. Because no anatomical abnormalities of the anastomoses had been reported during transplantation and clotting times were still prolonged, a thrombosis was extremely unlikely. All other diagnostic tests or clinical signs were inconclusive, not abnormal, or within the range of impairment expected in a patients who had undergone uncomplicated OLT. The time course of all variables is shown in Figure 1 and Table 1.Figure 1: Time course of indocyanine green dye plasma disappearance rate (ICGPDR), gastric intramucosal pH (pHi), and gastric intramucosal PCO2 (PrCO2) in the immediate postoperative period. adm. = admission, p.o. = postoperative.Table 1: Laboratory FindingsThe patient was scheduled for a high-urgency retransplantation. Forty-six hours after OLT, she underwent uncomplicated retransplantation with immediate excellent graft function (ICGPDR 26.4%). Her recovery from the procedure was uneventful, and ICGPDR of the new graft remained stable on follow-up (ICGPDR 19.8% on admission, 21% on Day 4 postoperatively). On Day 21, the patient was discharged. Histological analysis of the primary graft revealed complete ischemic necrosis, despite the fact that no thrombus could be detected during relaparatomy. Anamnesis of donor risk factors revealed short-term noradrenaline support at the end of her 3-day ICU stay due to brain injury. However, the functioning of other organs harvested and transplanted was excellent. Discussion Assessment of graft function and differential diagnosis of graft dysfunction in the immediate postoperative course after OLT remains especially difficult because of etiological complexity. Differential diagnosis includes primary nonfunction of the graft, rejection, infection, drug toxicity, and thrombosis of the hepatic blood vessels. Routine laboratory investigations, including bilirubin, serum transaminases and, clotting tests, are of little diagnostic value because the results are difficult to interpret and are often nonspecific in the early postoperative period [6,7]. Furthermore, invasive techniques increase the risk of serious complications. Early detection of graft dysfunction or nonfunction depends largely on the experience of the intensivist and the correct interpretation of laboratory tests or other more conclusive investigations. In our patient, there were no findings suggesting early graft nonfunction, with the exception of the bedside ICGPDR assessment. In particular, there were neither clinical signs nor alterations in the laboratory tests except those expected. In contrast to previous reports, gastric intramucosal PCO2 and intramucosal pH could not be verified as sensitive markers of liver dysfunction after OLT in this case [8,9]. Those variables reflect gastric mucosal hemodynamics and determine total splanchnic perfusion. Because systemic hemodynamics were stable in our patient, we hypothesize that splanchnic perfusion was also stable and that the problem was solely hepatic. Hepatic dysfunction might not be detected by this method if it does not result from compromised splanchnic hemodynamics or if it does not lead to changes in the acid-base balance [10]. Even Doppler ultrasound examination failed to detect hepatic ischemia. Besides the difficulties in examination because of anatomic causes, this investigation might be flawed by great interindividual variation in the interpretation of the results, depending on the expertise of the investigator. Because of the total stop in hepatic blood flow, the patient presented with functional anhepacy within a few hours. She did not develop hemodynamic instability and metabolic derangement. This might have been a prerequisite for successful retransplantation and outcome, together with the fact that the patient was otherwise healthy, with the exception of the underlying cholangiocellular carcinoma. Routine monitoring with dynamic liver function tests, e.g., ICGPDR or monoethylglycinexylidide, has rarely been reported because the tests used to require specialized laboratories. Investigations of this nature have also been time consuming. The ICGPDR, as assessed using a bedside monitoring device Pulsion COLD-System[trade mark sign], is based on the principle of a once-applied ICG dye indicator bolus. Consecutive measurement of the exponential decay time of the dye via a fiberoptic catheter provides quick and reliable information concerning the global performance of the graft [5]. We do not rely on certain cutoff points of ICGPDR as a prognostic value of outcome. Nevertheless, changes observed in ICGPDR after OLT indicate the necessity to follow-up with more invasive techniques and should influence treatment decisions.