Scoring system for predicting 90-day mortality of in-hospital liver cirrhosis patients at Cipto Mangunkusumo Hospital

Background: Liver cirrhosis is the final stage of chronic liver disease. Complications due to progression of liver disease may deteriorate the liver function and worsen prognosis. Previous studies have shown that patients with liver cirrhosis are at increased risk of death within 90-day after hospitalization. It is necessary to identify patients who are at higher risk of early mortality. This study aims to develop a scoring system to predict the 90-day mortality among hospitalized patients with liver cirrhosis that could be used for modification of treatment plan according to the scores that have been obtained. By using this scoring system, crucial care of plans can be taken to reduce the risk of mortality. Method: This prospective cohort study was conducted on hospitalized cirrhotic patients at Cipto Mangunkusumo National General Hospital, Jakarta. Demographic, clinical, and laboratory data were recorded. Patients were monitored for up to 90-day after hospitalization to determine their condition. Cox regression analysis was performed to obtain predictor factors contributing to mortality in liver cirrhosis patients. The scoring system that resulted from this study categorized patients into low, moderate, and high-risk categories based on their predicted mortality rates. The sensitivity and specificity of the scoring system were evaluated using the AUC (area under the curve) metric. Result: The study revealed that liver cirrhosis patients who were hospitalized had a 90-day mortality rate of 42.2%, with contributing factors including Child-Pugh, MELD, and leukocyte levels. The combination of these variables had a good discriminative value with an AUC of 0.921 (95% CI: 0.876–0.967). The scoring system resulted in three risk categories: low risk (score of 0–3) with a 4.1-18.4% probability of death, moderate risk (score of 5–6) with a 40.5-54.2% probability of death, and high risk (score of 8–11) with a 78.1-94.9% probability of death. Conclusion: The scoring system has shown great accuracy in predicting 90-day mortality in hospitalized cirrhosis patients, making it a valuable tool for identifying the necessary care and interventions needed for these patients upon admission.