Clinical Model for the Prediction of Severe Liver Fibrosis in Adult Patients with Type II Diabetes Mellitus

The fibrosis investigating navigator in diabetes (FIND): A tool to predict liver fibrosis risk in subjects with diabetes

BACKGROUND

Type 2 diabetes increases the risk of cirrhosis and liver cancer. Noninvasive and early assessment of liver fibrosis is essential. We aimed to develop a score to aid in the initial assessment of liver fibrosis in the diabetic population.

METHODS

A fibrosis investigating navigator in diabetes (FIND) score was developed and validated in the NHANES dataset (2017-2020). Fibrosis was defined as a liver stiffness measurement (LSM) ≥8.0 kPa. The diagnostic accuracies of FIB-4, NFS, LiverRisk, steatosis-associated fibrosis estimator (SAFE) and metabolic dysfunction-associated fibrosis (MAF-5) were compared. FIND was also externally validated in various liver diseases via biopsy as a reference in an Asian centre between 2016 and 2020. Finally, we examined the prognostic implications of the FIND index utilizing data from the UK Biobank cohort (2006-2010).

RESULTS

The FIND score model yielded an AUROC of 0.781 for the prediction of an LSM ≥8 kPa in the validation set, which was consistently greater than that of other available models (all p < 0.05). In the whole NHANES dataset, the 85% sensitivity cut-off of 0.16 corresponded to a NPV of 91.9%, whereas the 85% specificity cut-off of 0.31 corresponded to a PPV of 50.6%. FIND displayed overall accuracies similar to those of the other models in staging fibrosis stages, with biopsy used as a reference. In the UK Biobank cohort, FIND >0.31 was associated with an increased risk of all-cause and liver-related mortality in the diabetic population in adjusted models (HR, 1.75; 95% CI, 1.62-1.89; HR, 23.59; 95% CI, 13.67-40.69).

CONCLUSIONS

In diabetes patients, the novel FIND score performs well in identifying subjects at risk of liver fibrosis and predicting all-cause and liver-related mortality.

Machine learning models including insulin resistance indexes for predicting liver stiffness in United States population: Data from NHANES

Background

Prevention and treatment of liver fibrosis at an early stage is of great prognostic importance, whereas changes in liver stiffness are often overlooked in patients before the onset of obvious clinical symptoms. Recognition of liver fibrosis at an early stage is therefore essential.

Objective

An XGBoost machine learning model was constructed to predict participants' liver stiffness measures (LSM) from general characteristic information, blood test metrics and insulin resistance-related indexes, and to compare the fit efficacy of different datasets for LSM.

Methods

All data were obtained from the National Health and Nutrition Examination Survey (NHANES) for the time interval January 2017 to March 2020. Participants' general characteristics, Liver Ultrasound Transient Elastography (LUTE) information, indicators of blood tests and insulin resistance-related indexes were collected, including homeostasis model assessment of insulin resistance (HOMA-IR) and metabolic score for insulin resistance (METS-IR). Three datasets were generated based on the above information, respectively named dataset A (without the insulin resistance-related indexes as predictor variables), dataset B (with METS-IR as a predictor variable) and dataset C (with HOMA-IR as a predictor variable). XGBoost regression was used in the three datasets to construct machine learning models to predict LSM in participants. A random split was used to divide all participants included in the study into training and validation cohorts in a 3:1 ratio, and models were developed in the training cohort and validated with the validation cohort.

Results

A total of 3,564 participants were included in this study, 2,376 in the training cohort and 1,188 in the validation cohort, and all information was not statistically significantly different between the two cohorts (p > 0.05). In the training cohort, datasets A and B both had better predictive efficacy than dataset C for participants' LSM, with dataset B having the best fitting efficacy [±1.96 standard error (SD), (-1.49,1.48) kPa], which was similarly validated in the validation cohort [±1.96 SD, (-1.56,1.56) kPa].

Conclusions

XGBoost machine learning models built from general characteristic information and clinically accessible blood test indicators are practicable for predicting LSM in participants, and a dataset that included METS-IR as a predictor variable would improve the accuracy and stability of the models.