New online model to predict transplantation need during paracetamol-induced acute liver failure
For patients with acute liver failure, it is crucial to accurately predict who will need an emergency liver transplant, but the most widely used prediction tool internationally - the King’s College Criteria - is outdated. Now, a study led by researchers at King’s College Hospital and King’s College London, has updated the old static criteria into a dynamic online model which can predict individual outcomes more accurately.
Acute liver failure is a rare, but very serious, illness that is most commonly caused by paracetamol toxicity. Acute liver failure from paracetamol poisoning occurs rapidly and patients can progress to deep coma and severe multiple organ failure within hours.
The key to managing acute liver failure is early and accurate evaluation of the patient’s expected prognosis. This enables successful liver transplantation in the narrow window of opportunity for those who will benefit from it, and avoids unnecessary surgery in those who will survive with medical therapy alone.
The new research, published in The Lancet Gastroenterology & Hepatology, updates the King’s College Criteria, first developed in the 1980s, because over time survival has increased for patients managed with medical care alone.
The updated criteria for paracetamol-related acute liver failure include, for the first time, an online model that gives an individualised and updatable prognosis, which can change as the patient’s condition changes. The new model was externally validated against data from 412 patients with paracetamol-induced acute liver failure managed at intensive care units in the UK (London, Birmingham, and Edinburgh) and Denmark (Copenhagen), and showed excellent accuracy.
Dr Yanzhong Wang, co-lead author from the Unit for Medical Statistics in the Division of Health and Social Care Research at King's College London, said: ‘Previous models were static, so what is so novel about this model is that it’s dynamic and updating, so the prognosis changes if a patient’s condition changes for worse or better. This better reflects the real world of acute emergency care for paracetamol-induced acute liver failure. The new model shows excellent performance in predicting survival and need for transplantation with over 90% precision. It’s exciting that our new online app will also be able to collect the data entered into it by liver specialists world-wide and use this to further validate and fine-tune the model to reflect what’s happening on the ground. This study represents the way forward for collaborations between statisticians and clinicians.’
Dr William Bernal, co-lead author and Consultant in Intensive Care Medicine from the Institute of Liver Studies at King’s College Hospital, said: ‘Emergency liver transplantation may be life-saving for some people with severe acute liver failure from paracetamol poisoning, but the existing selection criteria are more than 25 years old and do not reflect the current outcomes of the illness. We need better tools to identify those patients who need transplantation early in their disease course and so we have developed and tested a practical high performance statistical model to predict who will benefit from transplantation - enabling us for the first time to make individual and updated survival predictions. We expect this model will be adopted internationally.’
This study was funded by the Foundation for Liver Research.
The paper ‘Development and validation of a dynamic outcome prediction model for paracetamol-induced acute liver failure: a cohort study’ by William Bernal, Yanzhong Wang and colleagues is published in The Lancet Gastroenterology & Hepatology. DOI: http://dx.doi.org/10.1016/S2468-1253(16)30007-3
The model will be available online later in 2016.
King’s College London and King’s College Hospital NHS Foundation Trust are part of King’s Health Partners Academic Health Science Centre.