Predicting Liver Injury in Sepsis Patients
Author Information
Author(s): He Yong, Wang Chi, He Wan, Zhang He, Ding Fei, Liu Ying, He He, Ying Binwu, Nie Xin
Primary Institution: West China Hospital, Sichuan University
Hypothesis
This study aimed to investigate the predictive value of serum indices for the occurrence of sepsis-related liver injury (SRLI) in adults.
Conclusion
The study developed a highly accurate predictive model for SRLI using clinically accessible serum indicators, which could aid in early detection and intervention, potentially reducing mortality rates.
Supporting Evidence
- ALT, GGT, CO2-CP, RDW-CV, AT III, and FDP were identified as independent predictors of SRLI.
- The predictive model had an AUC of 0.890, indicating excellent diagnostic value.
- Early detection of SRLI can potentially reduce mortality rates in sepsis patients.
Takeaway
Doctors can use certain blood tests to tell if a patient with sepsis might have liver problems, which can help them treat the patient better and faster.
Methodology
The study was a retrospective analysis of data from 1,573 sepsis patients, using stepwise logistic regression to identify risk factors and construct a predictive model assessed by ROC curve analysis.
Potential Biases
Selection bias due to the retrospective nature of the study and incomplete clinical data.
Limitations
The study is retrospective, which may introduce selection bias, and it has a relatively small sample size with potential confounding variables.
Participant Demographics
Median age was 53 years, with approximately 65.4% male participants.
Statistical Information
P-Value
<0.001
Confidence Interval
95% CI: 1.001–1.002
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website