Introduction: Most of the lactate in the body is cleared in the liver. Tissue hypoxia results in increased production of lactate and decreased utility of it. Hepatic insult results not only an increase in the blood lactate levels but also is an independent prognostic marker in critically ill cirrhotic patients. Alteration of liver function is indicated by rise in serum bilirubin. The aim of this study was to therefore to assess the utility of blood lactate to Bilirubin index (LBi) in predicting mortality in patients with acute on chronic liver failure (ACLF).
 Methods: This prospective observational study was conducted from January 2019 to June 2021 at the Department of Hepatogastroenterology, Sindh Institute of Urology and Transplantation. Patients aged > 12 years and presenting ACLF were included and their baseline characteristics were recorded. Primary outcome was observed in terms of 30-days mortality and secondary outcome was six months mortality. These indices were then used to calculate the lactate to bilirubin index (LBi) as [1000 × lactate (mmol/L) × bilirubin (µmol/L)]/2. Area under Receiver operating curves (AUROC) for LBi, Child Turcotte Pugh score(CTP) and Model for End-stage Liver Disease score (MELD) were obtained in predicting both 30 days and six months mortality and at an optimal cutoff sensitivity, specificity and diagnostic accuracy for these scores were calculated.
 Results: A total number 159 patients with ACLF were included in the study. Most of the patients were young with mean age of 35.1 ±16.8 years. Males were 97(61%). Hepatitis C was the most common cause of chronic liver disease followed by hepatitis B and autoimmune hepatitis seen in 41 (25.8%), 39(24.5%) and 36 (22.6%) respectively. Hepatitis E was the most common cause of acute injury noted in 60 (37.7%) patients. The baseline characteristics showed mean serum lactate levels of 0.93±1.33 mmol/L, bilirubin levels of 258.5 ± 155.3 µmol/L, CTP score of 10.7 ± 1.8 and MELD score of 26 ±7.6. Out of 159 patients, 26 (16.4%) patients died within 30 days due to ACLF related complicaitons while 133 (83.6%) were discharged. AUROC obtained for LBi, CTP score and MELD score in predicting 30-day mortality in ACLF was 0.98, 0.79 and 0.78 respectively. A cut off of ≥11.8 for LBi Index, ≥30 for MELD score and ≥13 for CTP score were significantly associated with increased risk of 30-day mortality in ACLF patients in our population. However, the sensitivity, specificity, PPV, NPV and diagnostic accuracy of LBi in predicting 30-day mortality was significantly higher than that of CTP and MELD score. The diagnostic accuracy of LBi in predicting 30 days mortality was 87.5%. Similarly, AUROC obtained for LBi, CTP score and MELD score in predicting 6-month mortality in ACLF was 0.89, 0.72 and 0.66 respectively and the diagnostic accuracy of LBi dropped down to 76.6% with a sensitivity of 49.28%, specificity of 97.28%, PPV of 94.4% and NPV of 71.54%. 
 Conclusion: Our results showed that LBi score of >11.8 had an excellent sensitivity and specificity in predicting mortality in ACLF with an excellent diagnostic accuracy in predicting one month mortality as compared to the other scores. However, its utility in predicting long term mortality is yet to be proven. Further studies are needed to validate this index.
Read full abstract