Development and validation of diagnostic risk score for fatty liver grading among rural population visiting tertiary care hospital

Abstract

Background: Fatty liver disease caused by alcoholic or non-alcoholic etiologies is an emerging major public health issue in India as well as worldwide. Non-alcoholic fatty liver disease has a global prevalence of approximately 25%. Aims and Objectives: We aim to develop and validate a risk-scoring system for the diagnosis of fatty liver disease among high-risk individuals (diabetes mellitus and obesity) using basic clinical history, laboratory investigations, and radiological examinations. Materials and Methods: Study design: Cross-sectional study conducted in a tertiary care teaching hospital. A patient case form has been designed for this study which includes all the relevant information under three headings, personal history, laboratory investigation findings, and radiological examination findings. Patients who have a known medical history of diabetes mellitus and/or body mass index (BMI) more than or equal to 25.00 will be included in this study. Results: The total number of patient data included in the study was 164. The total sample under consideration comprised 88 males (53.7%) and 76 females (46.3%). The average height was reported to be 159 ± 10.27 cm. The average weight of participants was 68.4 ± 12.6 kg. The average BMI of the study populace was 27 ± 4.46 kg/m2. Fatty score grading showed a significant positive univariate correlation with the weight, BMI, and triglyceride levels of the patients. Rest all correlations were found to be insignificant. Patients additionally presented with fibrosis grading and LSM score which were found to be positively correlated significantly with fatty liver grading. Fibrosis score and low-density lipoprotein level showed a significant negative correlation. All significant univariate factors were found to produce an insignificant multivariate regression model. The receiver operating characteristic curve and cutoff scores could not be determined. Therefore, the study was not able to develop and validate a multivariable scoring system. Conclusion: Even though the study was not able to fulfill the objectives intended, problems were identified and solutions were recommended for conducting such a study. This study can be taken as a baseline for future studies taking the recommendations into consideration at the time of execution.