Comparing the predictive efficacy of MELD and ALBI scores in liver cirrhosis patients with acute upper gastrointestinal bleeding

HIGHLIGHTS

  • This research sought to determine the predictive power of the albumin-to-bilirubin ratio (ALBI) and the model for end-stage liver disease (MELD) scores for these patients’ future performance
  • The ALBI and MELD scores demonstrate a reasonable capacity for short-term outcome prediction
  • The MELD and ALBI scores may predict both 1-month mortality and death and rebleeding
  • The findings suggest that the MELD score is a more reliable prognostic factor than the ALBI score in patients with advanced liver cirrhosis.

ABSTRACT

Background –

Acute upper gastrointestinal bleeding (AUGIB) is a critical medical emergency and is a common cause of illness and death in individuals with liver cirrhosis. Objective – The point of this study was to check how well the albumin-to-bilirubin ratio (ALBI) and model for end-stage liver disease (MELD) scores could predict how these patients would do in the future. Methods – The Imam Khomeini Hospital gastroenterology department conducted a retrospective examination. We admitted 102 patients with AUGIB and liver cirrhosis from April 2021 to September 2023. The study included a full medical history and clinical evaluation upon admission, as well as all laboratory test results throughout the hospital stay. We diagnosed liver cirrhosis using clinical, laboratory, and radiologic data. We diagnosed AUGIB as having hematemesis, melena, or hematochezia. We then tested the ALBI, MELD, and liver and kidney function. Some criteria allow continuous variable comparison, whereas others allow discrete variable comparison. Death during hospitalization and rebleeding were the key outcomes, with one-month mortality assessed. We compared ALBI and MELD before establishing their relationship to mortality and rebleeding. Results – Of the 102 patients, 68.5% survived. Upon arrival, we noted a markedly elevated prevalence of edema, ascites, and chilly extremities among patients who did not survive. The MELD and ALBI scoring systems effectively forecast in-hospital mortality. The threshold for MELD is 21 (CI: 0.759–0.930, P=0.00), whereas for ALBI it is -2.3 (CI: 0.865–0.950, P=0.01). Neither party could foresee hospitalization or premature rebleeding. The probability of death may be forecasted using the MELD during the first discharge phase (P<0.05). Conclusion – The MELD and ALBI scores show a suitable ability to predict short-term outcomes and both of them can predict death and rebleeding, as well as 1-month mortality. Nevertheless, we recommend that in individuals with advanced liver cirrhosis, the MELD score is a more accurate prognostic indicator compared to the ALBI score.

AUTORES

Zahra Shokati ESHKIKI, Razieh KHAZAEI, Abazar PARSI and Ali Akbar SHAYESTEH