In March’s blog we focus on three papers published in the last month, all concerning liver disease. Two of these papers focus on hepatic complications in pregnancy, with an emphasis on practical management and avoiding complications in patient with non-alcohol fatty liver disease. The third paper discusses the use of the Bristol Prognostic Score (BPS) to determine mortality from cirrhosis in a Scottish cohort of patients. Additionally, we discuss a publication in Gut, reclassifying hepatocellular carcinoma based on immunogenomic features that may help determine optimal treatment strategies.
In the first paper Koralegedara and colleagues look at the predictive value of the presence of NAFLD for some specific complications of pregnancy. NAFLD is an established risk factor for a number of complications in pregnancy but as the authors point out, prospective evidence of the effect of NAFLD in low or middle-income countries is lacking. The authors assessed the impact of NAFLD on 632 pregnant women from Anuradhapura district, Sri Lanka. Of these women all were screened with an ultrasound during the 1st trimester- 90 (14%) and 234 (37%) were diagnosed as having fatty liver grade 2 and 1, respectively. Incidence of GDM increased with increasing fatty liver grade, from 11/100,000 in grade 0, 44/100,000 in grade 1 and 162/100,000 in grade 2. Those with grade 2 were 12.5x more likely to develop GDM compared to grade 0 after adjusting for confounders. The authors also determined that fatty liver grade 2 was the only independent risk factor that remained significant for miscarriage after adjustment for age, parity, body mass index, blood sugar, blood pressure, and haemoglobin level, with an odds ratio of 4.2. NAFLD is a major risk factor for adverse outcomes in pregnancy and should be actively treated with weight loss strategies prior to conception.
In the next paper Morrison et al review the evidence and provide practical strategies for managing hepatic complications in pregnancy. Whilst these complications are rare, prompt management can avoid the potentially devastating consequences for both mother and fetus. Diagnosis of these disorders can be determined by biochemical and clinical features. The authors specifically mention acute fatty liver of pregnancy, intrahepatic cholestasis of pregnancy and haemolysis, elevated liver enzymes, low platelets (HELLP) syndrome. Whilst the management of these disorders is largely supportive, it should have impact on the timing of delivery and some specific therapy can be utilised in certain conditions- such as ursodeoxycholic acid in cholestasis. Overall early diagnosis and better supportive therapy has reduced the mortality and morbidity related to this challenging group of disorders but I thoroughly recommend reading the full practical advice in the article.
The final article from BMJOG Bower’s and colleagues present data on the Bristol prognostic score for mortality in cirrhosis. Here the authors apply this score, which has the aim of identification of patients with life expectancy of less than 12 months, to 3 cohorts of patients from Scottish hospitals over two 90-day periods. In total 276 patients were included. Those included were more likely to be middle-aged men and have alcohol-related liver disease. Bristol prognostic score >3 indicated a positive result predicting mortality within 1 year. Compared to the original publication, patients from the Scottish cohorts had more hospital admissions, and higher ongoing alcohol use. The score performed relatively poorly in these cohorts, sensitivity 54.9% (72.2% in original study), specificity 58% (83.8%) and positive predictive value 43.4% (81.3%). The authors conclude that the score was not useful in this patient group and further validation is needed for these predictive scoring mechanisms.
Finally, we mention a ground-breaking publication in Gut, concerned with revising the classification of HCC based on immune and genetic features. The authors took a multi-omic approach through RNA and exome sequencing, T-cell receptor (TCR)-sequencing, multiplex immunofluorescence and immunohistochemistry in a cohort of 240 HCC patients. The classification was compared to response to immunotherapy and the results were validated in a separate cohort of 660 patients. 37% of patients had an inflamed class of HCC, characterised by a 20-gene expression signature (~90% captured by this RNA-based biomarker). Secondly the intermediate class of HCC was characterised by TP53 mutations. Finally, the excluded class were enriched for CTNNB1 variation and PTK2 overexpression. The authors hope that this improved molecular classification system can help to predict the immune response in HCC.