Improving outcomes for transplantation of critically ill patients with cirrhosis?

Watch a video presentation of this article Watch the interview with the author Patients with end-stage chronic liver disease (CLD) often deteriorate with development of extrahepatic organ failures. These include the development of high-grade encephalopathy, renal dysfunction or failure, and hemodynamic instability, and they are associated with a high mortality. Consequently, in this patient group there is a frequent requirement for hospitalization and critical care organ support, but rescue liver transplantation (LT) is infrequently used. In a recent review of more than 50,000 US patients undergoing first LT for CLD between 2002 and 2014, only 8% of the cohort was hospitalized and requiring organ support at the time of transplantation.1 A number of factors underlie the infrequent use of LT in critically ill patients with CLD. It is recognized that hospitalization and requirement for critical care organ support impact negatively upon patient survival; survival outcomes are consistently worse than those of nonhospitalized recipients1 (Fig. 1). Postoperative length of stay, resource use, and expense are also very substantially increased. Further, in critically ill cirrhotic patients, the practical difficulties in urgent transplant assessment, active substance dependence, paucity of organ donors, and presence of sepsis and circulatory failure often preclude emergency LT.2 In a recent Austrian series of critically ill cirrhotic patients who underwent expedited LT assessment and prioritized transplantation, with median waiting time to LT of 24 days, fewer than a fourth underwent transplantation.3 In the CANONIC (CLIF Acute-on-Chronic Liver Failure in Cirrhosis) study, only 9% of 415 patients hospitalized with acute-on-chronic liver failure (ACLF) were transplanted within 28 days following admission, and 15% within 90 days.4 Transplantation is usually possible only in those assessed and listed for transplantation prior to becoming critically ill, and then only in a proportion because deterioration and death while on the wait list is common. Of 221 Canadian patients wait-listed for LT who deteriorated and required intensive care unit (ICU) care, nearly 50% died while awaiting surgery.5 Although illness severity, wait-list deterioration, and preexisting medical and psychiatric issues may preclude the use of transplantation in the majority of patients, in the minority who do undergo surgery, survival may be remarkably good. In the Canadian and Austrian series, 1-year post-LT patient survival rates were 74% and 87%, respectively, and 90-day survival in the CANONIC study was more than 90% as compared with 12% in those with equivalent severity of ACLF who did not undergo LT (Fig. 2). A clinical parallel is in the use of LT in patients with acute liver failure. This, too, is an uncommon indication for LT, where recipients are often critically ill with frequent presence of organ systems failure and requirement for organ support and a historically high mortality after LT. However, there is strong evidence of an “era effect” in this patient group, with markedly improved post-LT survival over time, reflecting advances in critical care support, recipient identification and selection, and a better understanding of graft and recipient matching in ALF.6, 7 We investigated whether a similar era effect had also occurred in critically ill patients with CLD, and using a large national cohort, we determined whether similar improvements in post-transplant patient survival had also occurred over time in hospitalized CLD patients.8 From the UK national database of NHS Blood and Transplant for the period 1994 to 2016, we studied 7782 patients with cirrhosis undergoing first elective transplantation. These were classified into three groups according to their hospitalization status at the time of transplantation: not hospitalized (NH); hospitalized but without organs systems failure (HNOF); and lastly those hospitalized with organ failure or the requirement for organ support (HOFS), defining this group by the presence of high-grade encephalopathy, use of renal replacement therapy, or mechanical ventilation. This latter group represented less than 5% of the total cohort and had laboratory findings and scores consistent with increasing severity of liver disease. Looking at the whole cohort, there was evidence of an era effect with a progressive improvement in 1-year patient survival rate increasing from 85% in 1994 to 1997 to 97% in 2011 to 2016 (Fig. 3). When divided by hospitalization status, we found patient survival to be greatest in those NH at the time of transplantation, lower in those HNOF, and lowest in those HOFS. One-year survival rates were 91%, 87%, and 75%, respectively (Fig. 4). Improved patient survival over time was seen in all hospitalization groups, but was most marked in HOFS patients. These patients showed a progressive incremental improvement in survival over time, with 1-year patient survival increasing from 56% in 1994 to 1997 to 90% in 2011 to 2016 (Fig. 5). Comparing the HOFS patients from the earliest part of the series with those transplanted in 2011 to 2016, there were no significant differences in age or gender, but recent patients had lower MELD score and significantly higher serum sodium (Table 1). The recent patients showed no difference in the proportion ventilated at time of LT, but more were receiving renal replacement therapy and markedly less had high-grade encephalopathy. Striking differences were also apparent in comparison of the grafts, with more recent grafts derived from older donors with higher body mass index, and an increased proportion of grafts were from cardiac death donors. In our cohort we found that hospitalization at the time of transplant impacted negatively on patient survival, most markedly in those with organ failure or requirement for organ support. There was global improvement in patient survival over time in all hospitalization groups, but most markedly in those with organ failure or requirement for support. These changes in patient survival were associated with alteration in donor and recipient characteristics, with the latter transplanted with less severe liver failure and encephalopathy, but with comparable or even increased levels of organ support. Interestingly, these improvements in survival occurred despite the increasing use of suboptimal grafts. Organ support per se would seem not to be a contraindication for transplantation, particularly in relation to renal replacement therapy. Good survival outcomes can occur if recipients are transplanted earlier in their disease course and prior to the onset of high-grade encephalopathy, arguing perhaps for a lower threshold for critical care admission, provision of organ support, and even wait-list prioritization.2, 9 And, strikingly, these data would suggest that these apparently high-risk recipients tolerated suboptimal grafts well, and that expedited transplantation with such suboptimal grafts may be associated with good transplant outcomes.10 Even so, these outcomes remain inferior to those in recipients who are less unwell at the time of surgery; the success of a transplant program adopting such an approach could not be judged on overall crude survival alone, but rather on outcomes stratified by illness severity. Further, the impact upon graft availability and resource utilization in already constrained transplantation services from increased use of LT in critically ill patients with cirrhosis is yet to be determined.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.