Development and validation of a clinical and laboratory-based nomogram to predict nonalcoholic fatty liver disease

LASSO-derived model for the prediction of lean-non-alcoholic fatty liver disease in examinees attending a routine health check-up

BACKGROUND

Lean individuals with non-alcohol fatty liver disease (NAFLD) often have normal body size but abnormal visceral fat. Therefore, an alternative to body mass index should be considered for prediction of lean-NAFLD. This study aimed to use representative visceral fat links with other laboratory parameters using the least absolute shrinkage and selection operator (LASSO) method to construct a predictive model for lean-NAFLD.

METHODS

This retrospective cross-sectional analysis enrolled 2325 subjects with BMI < 24 kg/m2 from medical records of 51,271 examinees who underwent a routine health check-up. They were randomly divided into training and validation cohorts at a ratio of 1:1. The LASSO-derived prediction model used LASSO regression to select 23 clinical and laboratory factors. The discrimination and calibration abilities were evaluated using the Hosmer-Lemeshow test and calibration curves. The performance of the LASSO model was compared with the fatty liver index (FLI) model.

RESULTS

The LASSO-derived model included four variables-visceral fat, triglyceride levels, HDL-C-C levels, and waist hip ratio-and demonstrated superior performance in predicting lean-NAFLD with high discriminatory ability (AUC, 0.8416; 95% CI: 0.811-0.872) that was comparable with the FLI model. Using a cut-off of 0.1484, moderate sensitivity (75.69%) and specificity (79.86%), as well as high negative predictive value (95.9%), were achieved in the LASSO model. In addition, with normal WC subgroup analysis, the LASSO model exhibits a trend of higher accuracy compared to FLI (cut-off 15.45).

CONCLUSIONS

We developed a LASSO-derived predictive model with the potential for use as an alternative tool for predicting lean-NAFLD in clinical settings.

Development and validation of a nomogram for nonalcoholic fatty liver disease in Western Xinjiang, China

OBJECTIVE

The aim of this study was to establish a simple, nonalcoholic fatty liver disease (NAFLD) screening model using readily available variables to identify high-risk individuals in Western Xinjiang, China.

METHODS

A total of 40 033 patients from the National Health Examination were divided into a training group (70%) and a validation group (30%). Univariate regression and least absolute shrinkage and selection operator models optimized feature selection, while a multivariate logistic regression analysis constructed the prediction model. The model's performance was evaluated using the area under the receiver operating characteristic curve, and its clinical utility was assessed through decision curve analysis.

RESULTS

The nomogram assessed NAFLD risk based on factors such as sex, age, diastolic blood pressure, waist circumference, BMI, fasting plasma glucose, alanine aminotransferase, platelet count, total cholesterol, triglycerides, low-density lipoprotein-cholesterol, and high-density lipoprotein-cholesterol. The area under the receiver operating characteristic curves were 0.829 for men and 0.859 for women in the development group, and 0.817 for men and 0.865 for women in the validation group. The decision curve analysis confirmed the nomogram's clinical usefulness, with consistent findings in the validation set.

CONCLUSION

A user-friendly nomogram prediction model for NAFLD risk was successfully developed and validated for Western Xinjiang, China.

Development and Validation of a Risk Prediction Model for NAFLD: A Study Based on a Physical Examination Population

Purpose

To construct and validate a precise and personalized predictive model for non-alcoholic fatty liver disease (NAFLD) to enhance NAFLD screening and healthcare administration.

Patients and Methods

A total of 730 participants' clinical information and outcome measurements were gathered and randomly divided into training and validation sets in a ratio of 3:7. Using the least absolute shrinkage and selection operator (LASSO) regression and multiple logistic regression, a nomogram was established to select risk predictor variables. The NAFLD prediction model was validated through the receiver operating characteristic (ROC) curve, calibration plot, and decision curve analysis (DCA).

Results

After random grouping, the cohort comprised 517 in the training set and 213 in the validation set. The prediction model employed nine of the 20 selected variables, namely gender, hypertension, waist circumference, body mass index, blood platelet, triglycerides, high-density lipoprotein cholesterol, plasma glucose, and alanine aminotransferase. ROC curve analysis yielded an area under the curve values of 0.877 (95% Confidence Interval [CI]: 0.848-0.907) for the training set and 0.871 (95% CI: 0.825-0.917) for the validation set. Optimal critical values were determined as 0.472 (0.786, 0.825) in the training set and 0.457 (0.743, 0.839) in the validation set. Calibration curves for both sets showed proximity to the ideal diagonal, with P-values of 0.972 and 0.370 for the training and validation sets, respectively (P > 0.05). DCA indicated favorable clinical applicability of the model.

Conclusion

We constructed a nomogram model that could complement traditional NAFLD detection methods, aiding in individualized risk assessment for NAFLD.

Non-alcoholic fatty liver disease risk prediction model and health management strategies for older Chinese adults: a cross-sectional study

BACKGROUND

Non-alcoholic fatty liver disease (NAFLD) is a common chronic liver condition that affects a quarter of the global adult population. To date, only a few NAFLD risk prediction models have been developed for Chinese older adults aged ≥ 60 years. This study presented the development of a risk prediction model for NAFLD in Chinese individuals aged ≥ 60 years and proposed personalised health interventions based on key risk factors to reduce NAFLD incidence among the population.

METHODS

A cross-sectional survey was carried out among 9,041 community residents in Shanghai. Three NAFLD risk prediction models (I, II, and III) were constructed using multivariate logistic regression analysis based on the least absolute shrinkage and selection operator regression analysis, and random forest model to select individual characteristics, respectively. To determine the optimal model, the three models' discrimination, calibration, clinical application, and prediction capability were evaluated using the receiver operating characteristic (ROC) curve, calibration plot, decision curve analysis, and net reclassification index (NRI), respectively. To evaluate the optimal model's effectiveness, the previously published NAFLD risk prediction models (Hepatic steatosis index [HSI] and ZJU index) were evaluated using the following five indicators: accuracy, precision, recall, F1-score, and balanced accuracy. A dynamic nomogram was constructed for the optimal model, and a Bayesian network model for predicting NAFLD risk in older adults was visually displayed using Netica software.

RESULTS

The area under the ROC curve of Models I, II, and III in the training dataset was 0.810, 0.826, and 0.825, respectively, and that of the testing data was 0.777, 0.797, and 0.790, respectively. No significant difference was found in the accuracy or NRI between the models; therefore, Model III with the fewest variables was determined as the optimal model. Compared with the HSI and ZJU index, Model III had the highest accuracy (0.716), precision (0.808), recall (0.605), F1 score (0.692), and balanced accuracy (0.723). The risk threshold for Model III was 20%-80%. Model III included body mass index, alanine aminotransferase level, triglyceride level, and lymphocyte count.

CONCLUSIONS

A dynamic nomogram and Bayesian network model were developed to identify NAFLD risk in older Chinese adults, providing personalized health management strategies and reducing NAFLD incidence.

Development and validation of a nomogram for predicting metabolic-associated fatty liver disease in the Chinese physical examination population

AIM

We aim to develop and validate a nomogram including readily available clinical and laboratory indicators to predict the risk of metabolic-associated fatty liver disease (MAFLD) in the Chinese physical examination population.

METHODS

The annual physical examination data of Chinese adults from 2016 to 2020 were retrospectively analyzed. We extracted the clinical data of 138 664 subjects and randomized participants to the development and validation groups (7:3). Significant predictors associated with MAFLD were identified by using univariate and random forest analyses, and a nomogram was constructed to predict the risk of MAFLD based on a Lasso logistic model. Receiver operating characteristic curve analysis, calibration curves, and decision curve analysis were used to verify the discrimination, calibration, and clinical practicability of the nomogram, respectively.

RESULTS

Ten variables were selected to establish the nomogram for predicting MAFLD risk: sex, age, waist circumference (WC), uric acid (UA), body mass index (BMI), waist-to-hip ratio (WHR), systolic blood pressure (SBP), fasting plasma glucose (FPG), triglycerides (TG), and alanine aminotransferase (ALT). The nomogram built on the nonoverfitting multivariable model showed good prediction of discrimination (AUC 0.914, 95% CI: 0.911-0.917), calibration, and clinical utility.

CONCLUSIONS

This nomogram can be used as a quick screening tool to assess MAFLD risk and identify individuals at high risk of MAFLD, thus contributing to the improved management of MAFLD.

SFI, a sex hormone binding globulin based nomogram for predicting non-alcoholic fatty liver disease in the Chinese population

Background

The purpose of this study is to establish a novel nomogram model for accurate detection of non-alcoholic fatty liver disease (NAFLD) in the Chinese population based on sex hormone binding globulin (SHBG) and other routine laboratory tests.

Methods

A total of 1417 participants (1003 testing and 414 validations) were enrolled into the study. Risk factors independently associated with NAFLD were identified and incorporated in the new nomogram, SFI. The performance of nomogram was assessed by analysis of receiver operating characteristic (ROC) curve, calibration curve, and decision curve.

Results

We formulated a new nomogram incorporating four independent factors: SHBG, body mass index (BMI), ALT/AST, and triglycerides (TG). The nomogram achieved good indexes of area under ROC 0.898 (95% confidence interval 0.865-0.926) in predicting NAFLD, which was significantly superior to previously reported models of FLI, HSI, LFS, and LAP. The calibration curve and decision curve demonstrated high performance and clinical utility of the nomogram in predicting NAFLD.

Conclusion

The nomogram SFI has high performance in predicting NAFLD in Chinese population and may be used as a cost-effective screening model to assess NAFLD in the general population.

The roles of hepatokine and osteokine in liver-bone crosstalk: Advance in basic and clinical aspects

Both the liver and bone are important secretory organs in the endocrine system. By secreting organ factors (hepatokines), the liver regulates the activity of other organs. Similarly, bone-derived factors, osteokines, are created during bone metabolism and act in an endocrine manner. Generally, the dysregulation of hepatokines is frequently accompanied by changes in bone mass, and osteokines can also disrupt liver metabolism. The crosstalk between the liver and bone, particularly the function and mechanism of hepatokines and osteokines, has increasingly gained notoriety as a topic of interest in recent years. Here, based on preclinical and clinical evidence, we summarize the potential roles of hepatokines and osteokines in liver-bone interaction, discuss the current shortcomings and contradictions, and make recommendations for future research.

Assessing the clinical utility of major indices for nonalcoholic fatty liver disease in East Asian populations

Nonalcoholic fatty liver disease (NAFLD) is currently the most common form of chronic liver disease. The growing prevalence of NAFLD is strongly associated with the high incidence of metabolic syndrome. NAFLD affects as much as 19% of the US population with a disproportionate impact on minority racial groups such as Asian Americans. If not promptly managed, NAFLD may progress to more feared complications. Liver indices for NAFLD screening have been proposed but were often developed using study populations with different anthropometrics than patients of East Asian descent. This review compares the accuracy of five indices for NAFLD screening in Asian cohorts. The Fatty Liver Index performed well in multiple large-scale community studies, although other indices may be more suited for specific patient cohorts. This is important, as the utilization of liver indices could accelerate screening for NAFLD for early management and to reduce liver disease-related health disparities among Asian Americans.

Development and validation of a new nomogram to screen for MAFLD

BACKGROUND AND AIM

Metabolic dysfunction-associated fatty liver disease (MAFLD) poses significant health and economic burdens on all nations. Thus, identifying patients at risk early and managing them appropriately is essential. This study's goal was to develop a new predictive model for MAFLD. Additionally, to improve the new model's clinical utility, researchers limited the variables to readily available simple clinical and laboratory measures.

METHODS

Based on the National Health and Nutrition Examination Survey (NHANES) cycle 2017-2020.3, the study was a retrospective cross-sectional study involving 7300 participants. By least absolute shrinkage and selection operator (LASSO) regression, significant indicators independently associated with MAFLD were identified, and a predictive model called the MAFLD prediction nomogram (MPN) was developed. The study then compared the MPN with six existing predictive models for MAFLD. The model was evaluated by measuring the area under receiver operating characteristic curve (AUC), net reclassification index (NRI), integrated discrimination improvement (IDI), calibration curve, and decision curve analysis (DCA) curve.

RESULTS

In this study, researchers identified nine predictors from 33 variables, including age, race, arm circumference (AC), waist circumference (WC), body mass index (BMI), alanine aminotransferase (ALT)-to-aspartate aminotransferase (AST) ratio, triglyceride-glucose index (TyG), hypertension, and diabetes. The diagnostic accuracy of the MPN for MAFLD was significantly better than that of the other six existing models in both the training and validation cohorts (AUC 0.868, 95% confidence interval (CI) 0.858-0.877, and AUC 0.863, 95% CI 0.848-0.878, respectively). The MPN showed a higher net benefit than the other existing models.

CONCLUSIONS

This nonimaging-assisted nomogram based on demographics, laboratory factors, anthropometrics, and comorbidities better predicted MAFLD than the other six existing predictive models. Using this model, the general population with MAFLD can be assessed rapidly.

The Diagnostic and Prognostic Value of the Triglyceride-Glucose Index in Metabolic Dysfunction-Associated Fatty Liver Disease (MAFLD): A Systematic Review and Meta-Analysis

Metabolic dysfunction-associated fatty liver disease (MAFLD) has been related to a series of harmful health consequences. The triglyceride-glucose index (TyG index) appears to be associated with MAFLD. However, no consistent conclusions about the TyG index and incident MAFLD have been reached. PubMed, MEDLINE, Web of Science, EMBASE and the Cochrane Library were searched. Sensitivities, specificities and the area under the receiver operating characteristic (AUC) with a random-effects model were used to assess the diagnostic performance of the TyG index in NAFLD/MAFLD participants. Potential threshold effects and publication bias were evaluated by Spearman’s correlation and Deeks’ asymmetry test, respectively. A total of 20 studies with 165725 MAFLD participants were included. The summary receiver operator characteristic (SROC) curve showed that the sensitivity, specificity and AUC were 0.73 (0.69−0.76), 0.67 (0.65, 0.70) and 0.75 (0.71−0.79), respectively. Threshold effects (r = 0.490, p < 0.05) were confirmed to exist. Subgroup analyses and meta-regression showed that some factors including country, number of samples, age and disease situation were the sources of heterogeneity (p < 0.05). Our meta-analysis suggests that the TyG index can diagnose and predict MAFLD patients with good accuracy. The number of studies remains limited, and prospective studies are needed.

Comparison of the diagnostic value between triglyceride-glucose index and triglyceride to high-density lipoprotein cholesterol ratio in metabolic-associated fatty liver disease patients: a retrospective cross-sectional study

Background

The triglyceride and glucose index (TyG) and triglyceride to high-density lipoprotein cholesterol ratio (TG/HDL-C) are substitute markers of insulin resistance (IR). In a retrospective cross-sectional study, the authors aimed to compare the efficacy of the two indicators in diagnosing metabolic-associated fatty liver disease (MAFLD) to construct a novel disease diagnosis model.

Methods

Overall, 229 patients (97 MAFLD and 132 Non-MAFLD at West China Hospital of Sichuan University were included. MAFLD was diagnosed using ultrasonography. Biochemical indexes were collected and analyzed by logistic regression to screen out indicators that were expressed differently in MAFLD patients and healthy controls, which were incorporated into a diagnostic model.

Results

After adjusting for age, sex, and body mass index (BMI), serum alanine transaminase (ALT), aspartate transaminase (AST), AST/ALT (A/A), fasting plasma glucose (FPG), cystatin C (Cys-C), uric acid (URIC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), alkaline phosphatase (ALP), gamma-glutamyl transferase (GGT), non-HDL-C, LDL-C/HDL-C, non-HDL-C/HDL-C, TG/HDL-C, TC/HDL-C, TyG, and TyG-BMI were risk factors for MAFLD. The odds ratio of TG/HDL-C and TyG were 5.629 (95%CI: 3.039–10.424) and 182.474 (95%CI: 33.518–993.407), respectively. In identifying MAFLD, TyG, TyG-BMI, TG, and TG/HDL-C were found to be the most vital indexes based on the random forest method, with the area under the curve (AUC) greater than 0.9. In addition, the combination of BMI, ALT, and TyG had a high diagnostic efficiency for MAFLD.

Conclusions

TyG and TG/HDL-C were potential risk factors for MAFLD, and the former performed better in diagnosing MAFLD. The combination of BMI, ALT, and TyG improved the diagnostic capability for MAFLD.

Diagnostic accuracy of the visceral adiposity index in patients with metabolic-associated fatty liver disease: a meta-analysis

Background

Conflicting results on the prognostic value of the visceral adiposity index (VAI) in patients with metabolic-associated fatty liver disease (MAFLD) have been reported. This study aimed to assess the diagnostic value of the VAI in MAFLD patients.

Methods

The Cochrane Library, PubMed, Embase, and other databases were searched to collect all documents that met the inclusion criteria from the establishment of the database to September 2021. The methodological quality of the included studies was assessed using the Newcastle–Ottawa Scale. The heterogeneity among the studies was analysed by the Cochran Q test and I² test, and the appropriate model was selected according to the heterogeneity results. The diagnostic efficacy of the VAI was evaluated by sensitivity, specificity, and area under the curve, and a Fagan diagram was generated to evaluate the diagnostic ability of the VAI.

Results

A total of 9 studies were included. The overall quality of the included studies was good. Meta-analysis showed that the combined sensitivity of the VAI for the diagnosis of MAFLD was 0.70 [95% CI (0.69–0.71)], the combined specificity was 0.67 [95% CI (0.67–0.68)], the combined positive likelihood ratio was 2.08 [95% CI (1.87–2.31)], the combined negative likelihood ratio was 0.39 [95% CI (0.34–0.44)], and the combined diagnostic odds ratio was 5.81 [95% CI (4.73–7.14)]. The corresponding area under the curve was 0.79 [95% CI (0.75–0.82)]. Meta-regression analysis showed that the diagnostic method was a potential source of heterogeneity (P < 0.05). The Fagan diagram showed that the precision of MAFLD diagnosis was 70% when the pretest probability was set to 50% and then supplemented by the VAI.

Conclusions

The VAI is an independent predictor in the diagnosis of MAFLD and may be helpful in the detection of MAFLD. A VAI > 2.33 suggests that patients have a high probability of having MAFLD.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12944-022-01636-8.

The triglyceride and glucose index and risk of nonalcoholic fatty liver disease: A dose-response meta-analysis

BACKGROUND

The triglyceride and glucose (TyG) index is associated with the risk of nonalcoholic fatty liver disease (NAFLD), but the dose-response relationship between them is still unknown. We conducted a comprehensive meta-analysis to study the dose-response association between the TyG index and the risk of NAFLD.

METHODS

We systematically searched the Cochrane Library, PubMed, and Embase databases until July 2022 for relevant studies. The robust error meta-regression method was used to investigate the dose-response association between the TyG index and NAFLD. Summary relative risks (ORs) and 95% CIs were estimated by using a random-effects model.

RESULTS

A total of 4 cohort and 8 cross-sectional studies were included, with 28,788 NAFLD cases among the 105,365 participants. A positive association for the risk of NAFLD was observed for each additional unit of the TyG index with a linear association (p=0.82), and the summary OR was 2.84 (95% CI, 2.01-4.01). In the subgroup analyses, a stronger association of the TyG index with NAFLD was shown in females than in males (men: OR=2.97, 95% CI 2.55-3.46, women: OR=4.80, 95% CI 3.90-5.90, P_subgroup<0.001).

CONCLUSION

The TyG index may be a novel independent risk factor for NAFLD beyond traditional risk factors.

SYSTEMATIC REVIEW REGISTRATION

https://www.crd.york.ac.uk/prospero, identifier (CRD42022347813).

The Visceral Adiposity Index in Non-Alcoholic Fatty Liver Disease and Liver Fibrosis-Systematic Review and Meta-Analysis

(1) Background: In order to avoid a liver biopsy in non-alcoholic fatty liver disease (NAFLD), several noninvasive biomarkers have been studied lately. Therefore, we aimed to evaluate the visceral adiposity index (VAI) in NAFLD and liver fibrosis, in addition to its accuracy in predicting NAFLD and NASH. (2) Methods: We searched PubMed, Embase, Scopus, and Cochrane Library, identifying observational studies assessing the VAI in NAFLD and liver fibrosis. QUADAS-2 was used to evaluate the quality of included studies. The principal summary outcomes were mean difference (MD) and area under the curve (AUC). (3) Results: A total of 24 studies were included in our review. VAI levels were significantly increased in NAFLD (biopsy-proven and ultrasound-diagnosed), simple steatosis vs. controls, and severe steatosis vs. simple steatosis. However, no significant MD was found according to sex, liver fibrosis severity, simple vs. moderate and moderate vs. severe steatosis, pediatric NAFLD, and NASH patients. The VAI predicted NAFLD (AUC 0.767) and NASH (AUC 0.732). (4) Conclusions: The VAI has a predictive value in diagnosing NAFLD and NASH, with significantly increased values in adult NAFLD patients, simple steatosis compared to controls, and severe steatosis compared to simple steatosis.

Comparison and development of advanced machine learning tools to predict nonalcoholic fatty liver disease: An extended study

BACKGROUND

Nonalcoholic fatty liver disease (NAFLD) is a public health challenge and significant cause of morbidity and mortality worldwide. Early identification is crucial for disease intervention. We recently proposed a nomogram-based NAFLD prediction model from a large population cohort. We aimed to explore machine learning tools in predicting NAFLD.

METHODS

A retrospective cross-sectional study was performed on 15 315 Chinese subjects (10 373 training and 4942 testing sets). Selected clinical and biochemical factors were evaluated by different types of machine learning algorithms to develop and validate seven predictive models. Nine evaluation indicators including area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC), accuracy, positive predictive value, sensitivity, F1 score, Matthews correlation coefficient (MCC), specificity and negative prognostic value were applied to compare the performance among the models. The selected clinical and biochemical factors were ranked according to the importance in prediction ability.

RESULTS

Totally 4018/10 373 (38.74%) and 1860/4942 (37.64%) subjects had ultrasound-proven NAFLD in the training and testing sets, respectively. Seven machine learning based models were developed and demonstrated good performance in predicting NAFLD. Among these models, the XGBoost model revealed the highest AUROC (0.873), AUPRC (0.810), accuracy (0.795), positive predictive value (0.806), F1 score (0.695), MCC (0.557), specificity (0.909), demonstrating the best prediction ability among the built models. Body mass index was the most valuable indicator to predict NAFLD according to the feature ranking scores.

CONCLUSIONS

The XGBoost model has the best overall prediction ability for diagnosing NAFLD. The novel machine learning tools provide considerable beneficial potential in NAFLD screening.

Development of a preoperative index-based nomogram for the prediction of hypokalemia in patients with pituitary adenoma: a retrospective cohort study

Objective

To develop and validate a preoperative index-based nomogram for the prediction of hypokalemia in patients with pituitary adenoma (PA).

Methods

This retrospective cohort study included 205 patients with PAs between January 2013 and April 2020 in the Sun Yat-sen Memorial Hospital, Guangzhou, China. The patients were randomly classified into either a training set (N = 143 patients) and a validation set (N = 62 patients) at a ratio of 7:3. Variables, which were identified by using the LASSO regression model were included for the construction of a nomogram, and a logistic regression analysis was used to calculate odds ratios (ORs) and 95% confidence intervals (CIs) in the training set. The area under the curve (AUC) was used to evaluate the performance of the nomogram for predicting hypokalemia. Multivariate logistic regression analysis with a restricted cubic spline analysis was conducted to identify a potential nonlinear association between the preoperative index and hypokalemia.

Results

The incidence of hypokalemia was 38.05%. Seven preoperative indices were identified for the construction of the nomogram: age, type of PA, weight, activated partial thromboplastin time, urea, eosinophil percentage, and plateletocrit. The AUCs of the nomogram for predicting hypokalemia were 0.856 (95% CI [0.796–0.915]) and 0.652 (95% CI [0.514–0.790]) in the training and validation sets, respectively. Restricted cubic splines demonstrated that there was no nonlinear association between hypokalemia and the selected variables.

Conclusion

In this study, we constructed a preoperative indices-based nomogram that can assess the risk of hypokalemia after the surgical treatment of pituitary adenomas. This nomogram may also help to identify high risk patients who require close monitoring of serum potassium.

Application of a Novel Prediction Model for Predicting 2-Year Risk of Non-Alcoholic Fatty Liver Disease in the Non-Obese Population with Normal Blood Lipid Levels: A Large Prospective Cohort Study from China

Purpose

The purpose of this study was to develop and validate a nomogram to better assess the 2-year risk of non-alcoholic fatty liver disease (NAFLD) in non-obese population with normal blood lipid levels.

Patients and Methods

This study was a secondary analysis of a prospective study. We included 3659 non-obese adults with normal blood lipid levels without NAFLD at baseline. A total of 2744 participants were included in the development cohort and 915 participants were included in the validation cohort. The least absolute contraction selection operator (LASSO) regression model was used to identify the best risk factors. Multivariate Cox regression analysis was used to construct the prediction model. The performance of the prediction model was assessed using Harrell’s consistency index (C-index), area under the receiver operating characteristic (AUROC) curve and calibration curve. Decision curve analysis was applied to evaluate the clinical usefulness of the prediction model.

Results

After LASSO regression analysis and multivariate Cox regression analysis on the development cohort, BMI, TG, DBIL, ALT and GGT were found to be risk predictors and were integrated into the nomogram. The C-index of development cohort and validation cohort was 0.819 (95% CI, 0.798 to 0.840) and 0.815 (95% CI, 0.781 to 0.849), respectively. The AUROC of 2-year NAFLD risk in the development cohort and validation cohort was 0.831 (95% CI, 0.811 to 0.851) and 0.797 (95% CI, 0.765 to 0.829), respectively. From calibration curves, the nomogram showed a good agreement between predicted and actual probabilities. The decision curve analysis indicated that application of the nomogram is more effective than the intervention-for-all-patients scheme.

Conclusion

We developed and validated a nomogram for predicting 2-year risk of NAFLD in the non-obese population with normal blood lipid levels.