The aim of this study was to examine the utility of liver function tests (LFTs) in predicting the prognosis of critically ill patients with primary pulmonary hypertension (PPH) with/without liver disease.
We retrieved the Medical Information Mart for Intensive Care III (MIMIC-III) database to acquire clinical data. From the database, we recruited adult patients that were equal to or older than 18 years with primary pulmonary hypertension (PPH) discharge from intensive care unit (ICU). Then, the relationship between LFTs and duration of hospitalization and ICU stays was examined based on the Spearman correlation. The chi-square assessment was conducted to examine the correlation between LFTs and death rates. Survival curves were plotted with the aid of the Kaplan-Meier technique, and the curves were subsequently compared utilizing the log-rank test. The LFTs were identified as independent predictive variables of death according to the results of multivariable logistic regression. The specificity and sensitivity for mortality were calculated utilizing receiver operating characteristic (ROC) curves and the area under the curve (AUC).
In total, 198 patients satisfying the inclusion criteria were recruited, among which there were 23 patients with liver disease. Only ALB was correlated with the length of ICU stay in the total PPH group. ALB independently served as a risk variable for hospital mortality and 90-day mortality and was significantly associated with 90-day and 4-year survival rates in both total PPH and PPH without liver disease. AST was correlated with hospital mortality and 90-day survival curves in both total PPH and PPH without liver disease and independently served as a risk factor for hospital and 90-day mortality only in the total PPH group. ALT independently acted as a risk variable for hospital mortality and total bilirubin was correlated with hospital mortality in the total group. The diagnostic performance of the predictive model combining the LFTs was moderately good for the hospital, 90-day, and 4-year mortality. Both Modelı End-Stage ıLiverı Disease (MELD) score and albumin-bilirubin (ALBI) score were independent risk factors for short- and long-term prognosis. And they were also significantly associated with short- and long-term prognosis.
Among critically ill patients with PPH and with or without liver illness, aberrant LFT was linked to short- and long-term prognoses.