CLINICAL AND BIOCHEMICAL CHARACTERISTICS IN NONALCOHOLIC FATTY LIVER IN LEAN AND NON-LEAN MALE AND FEMALE PATIENTS ACCORDING TO ASIA-PACIFIC BASAL METABOLIC INDEX CRITERIA IN A TERTIARY CARE HOSPITAL, WEST BENGAL– A CROSS SECTIONAL OBSERVATIONAL STUDY

CLINICAL AND BIOCHEMICAL CHARACTERISTICS IN NONALCOHOLIC FATTY LIVER IN LEAN AND NON-LEAN MALE AND FEMALE PATIENTS ACCORDING TO ASIA-PACIFIC BASAL METABOLIC INDEX CRITERIA IN A TERTIARY CARE HOSPITAL, WEST BENGAL– A CROSS SECTIONAL OBSERVATIONAL STUDY

Ashis Kumar Saha *, Koushik Mal and Puja Mahato

Jagannath Gupta Institute of Medical Sciences & Hospital, Budge Budge, Kolkata, West Bengal, India.

ABSTRACT: Introduction: Nonalcoholic fatty liver disease (NAFLD) represents a metabolic spectrum ranging from simple steatosis to cirrhosis and hepatocellular carcinoma. Although classically associated with obesity, lean individuals particularly in Asian populations are increasingly affected due to higher visceral adiposity at lower body mass index (BMI). The Asia-Pacific guideline defines lean as BMI <23 kg/m² and non-lean as ≥23 kg/m², reflecting region-specific metabolic risk. Aims and Objectives: To compare clinical, biochemical, and fibrosis-related characteristics including triglyceride-glucose (TyG) index, TyG-BMI, and triglyceride/HDL ratio—among lean and non-lean male and female NAFLD patients, and to evaluate the diagnostic accuracy of non-invasive fibrosis indices (FIB-4, APRI, TyG). Materials and Methods: A cross-sectional retrospective analysis was performed on 497 NAFLD patients (271 males, 226 females) attending a tertiary care hospital in West Bengal. Demographic data, liver enzymes, lipid and glucose profiles, and FibroScan liver stiffness were collected. Derived indices (FIB-4, APRI, TyG, TyG-BMI) were calculated. Statistical analysis included Welch’s t-test, Mann–Whitney U test, chi-square test, effect size estimation, and ROC analysis. Results: Non-lean males exhibited higher triglycerides, LDL cholesterol, and TyG indices, while non-lean females showed higher FIB-4 and BARD scores. Liver stiffness ≥8 kPa was more prevalent in non-lean patients of both sexes. ROC analysis demonstrated TyG (AUC ≈0.85) and FIB-4 (AUC ≈0.80) as the best predictors of significant fibrosis, each with a negative predictive value of approximately 88%. Conclusion: BMI ≥23 kg/m² effectively stratifies metabolic and fibrotic risk in Indian NAFLD patients. TyG and FIB-4 are simple, cost-effective tools for early fibrosis screening, supporting the clinical relevance of the Asia-Pacific BMI cut-off.

Keywords: Clinical and biochemical characteristics, Nonalcoholic fatty liver disease, Lean and nonlean individuals, Asia-Pacific Basal metabolic criteria, West Bengal

INTRODUCTION: Nonalcoholic fatty liver disease (NAFLD) is a composite metabolic disorder that relate to insulin resistance and genetic susceptibility.

Fatty tissue gradually accumulates in the liver and also in the liver in lean and non-lean subjects and gradually increases when the metabolic capacity of the body does not match side by side with the dietary fat, as a result, hepatic fatty tissue accumulation increases leading to development of inflammation resulting steatohepatitis, that may reverse with the treatment and life style changes or progresses to hepatic fibrosis, cirrhosis and may develop ultimately hepatocellular carcinoma ^{1, 2, 3}. As the age advances, the prevalence of histologically proved NAFLD increases from 20% under 20 years of age to over 60% to 80% in more than 60 years of age and it is aggravated by obesity and diabetes ^{4, 5, 6}.

So, it is becoming a global burden with serious risk to the public health ^{7, 8}. It is seen that NAFLD is directly related to body mass index (BMI) ⁹. As in Asian country, there is tendency of the fat to accumulate in the viscera, at same BMI, hence the definition of lean and nonlean BMI has been reduced to less than 23 in case of lean and more than equal to 23 in case of nonlean individual ^{5, 10, 11}. According to Global Registry, 8% of NAFLD were lean as compared to nonlean i.e. overweight and obese patients but equal risk for hepatic fibrosis or cirrhosis and also higher all-cause mortality than nonlean patients ^{12, 13, 14, 15, 16}.

As NAFLD is associated with certain morbidities like diabetes mellitus, obesity that may lead to cardiovascular or renal disease – this cannot be characterized by NAFLD ¹⁷. Hence, in 2020, International Fatty Liver Disease Expert panel recommended to change the name of NAFLD to Metabolic dysfunction associated steatotic liver disease (MASLD) and NASH to Metabolic dysfunction associated steatohepatitis (MASH) which was endorsed by 3 major international liver associations, later on, one study demonstrated the 99% similarity in the diagnostic criteria between NAFLD with MASLD ^{18, 19}.

As insulin resistance is one important criteria in metabolic syndrome and this is mostly due to insulin resistance which can be measured gold standard method, by hyper-insulinemic Euglycemic clamp, it is usually less frequently utilized in clinical practice, in the normal usual settings, triglyceride-glucose index, triglyceride to HDL ratio, TyG--BMI index can be used to estimate degree of insulin resistance ^{20, 21}. As, very few studies have been done to compare the above values in between lean and non-lean male and female, the aim of this study was to compare the demographic, biochemical statistics including the TyG index, TyG-BMI and Ti-HDL ratio in male and female lean and nonlean patients with NAFLD with fibroscan less than 8 and equal to more than 8 kPa.

MATERIALS AND METHODS: Patients were divided into male and female. Again, male and female patients were subdivided into lean i.e. BMI less than 23 and nonlean i.e. BMI more than equal to 23 were considered according to Asia-Pacific criteria.

Collection of Laboratory Data: From the records, all the data of liver function tests (bilirubin, AST, ALT, alkaline phosphatase, albumin, globulin), lipid profile (cholesterol, LDL, HDL, VLDL, triglyceride), fasting plasma glucose, HbA1C, stiffness score from transient elastography of liver were collected. Transient elastography was used to detect liver fibrosis by measuring liver stiffness (LSM). Significant liver fibrosis was diagnosed when LSM ≥ 8 kPa and <8 kPa was diagnosed as nonsignificant fibrosis.

Following Scores Were Used:

FIB-4 index can be calculated using following formula:

[Age in years x Serum AST level in U/L] / [platelet count/cc x (√serum ALT in U/L)]

AST/ALT ratio = (Serum AST in U/L) / (Serum ALT in U/L)

APRI (AST to platelet ratio index) = [(Serum AST level in U/L) / (Upper limit of normal serum AST in U/L)] / (100 / Platelet/cc)

APRI higher than 0.7 predicts significant fibrosis with high sensitivity and specificity. Fibro scan score: here score more than equal to 8.0 was considered. FIB4 and FIB5 scores, Triglyceride-glucose index, Triglyceride-BMI index and Triglyceride to HDL were proposed to estimate the degree of insulin resistance.

Statistical Methods and Statistical Analysis:

Study Design and Data Preparation: This retrospective cross-sectional analysis evaluated male and female patients stratified according to the Asia–Pacific BMI cut-off (<23 vs ≥23 kg/m²). Lipid and glucose parameters were converted to micromole/L (TG × 11.29, LDL/HDL × 25.86, FBS × 55.5). Derived indices such as AST/ALT ratio, FIB-4, APRI, BARD, and FIB-5 were computed using standard formulas; for APRI, the AST upper limit of normal (ULN) was assumed to be 40 U/L where not provided.

Statistical Tests Applied:

Descriptive Statistics: Continuous variables were summarized as mean ± SD or median (IQR). Categorical variables were expressed as counts and percentages.

Normality Assessment: Data distributions were inspected using Shapiro–Wilk tests and histograms.

Between-Group Comparisons: Welch’s t-test and Mann–Whitney U test were applied for continuous variables, Chi-square for categorical outcomes.

Effect Size Estimation: Cohen’s d was calculated to measure standardized differences (0.2=small, 0.5=medium, ≥0.8=large).

Correlation and Regression Analysis: Pearson correlation coefficients (r) were computed to assess linear relationships between liver stiffness and metabolic indices (TyG/BMI ratio). Simple linear regression models (e.g., Stiffness = 6.920 + 6.843 × TyG/BMI) were applied to evaluate predictive significance.

Diagnostic and Predictive Analysis (ROC Curve Evaluation): ROC curves were plotted for TyG Index, FIB-4, AST/ALT ratio, APRI, and BMI. The Area under the Curve (AUC) quantified accuracy (≥0.7 acceptable, ≥0.8 good, ≥0.9 excellent). Optimal diagnostic thresholds were determined by Youden’s J statistic. Sensitivity, specificity, PPV, and NPV were derived accordingly.

Subgroup and Sex-Stratified Analysis: Separate analyses were performed for males (n=271) and females (n=226). Welch’s t-tests and ROC analyses identified significant predictors (APRI, AST, and FIB-4 in males; TyG and FIB-4 in females). Sex-specific comparisons enabled assessment of metabolic and fibrotic patterns across BMI strata.

Interpretation Framework: Statistical significance was considered at p < 0.05 (two-tailed). Results were interpreted based on effect size magnitude and clinical relevance. Graphical outputs included forest plots (Cohen’s d), ROC curves, and regression plots for stiffness vs TyG/BMI. Analyses were performed using Python (Pandas, SciPy, scikit-learn, Matplotlib) and Excel.

RESULTS:

BMI ≥23 vs <23: Mann–Whitney U Comparisons:

Definitions: Hypertension = systolic ≥140 mmHg or diastolic ≥90 mmHg; Diabetes mellitus = HbA1C >7%; Liver stiffness threshold = 8 kPa. Results are presented as median (P25–P75) with Mann–Whitney U two-sided p-values.

TABLE 1: BMI ≥23 VS <23 - ALL PATIENTS

Table 1 demonstrated Mann–Whitney U p-values: Age (years): 0.1709; Hypertension (≥130/≥90): 0.0045; Diabetes (HbA1C >7%): 0.9020; Liver stiffness ≥8 kPa: 0.2795

Detailed Interpretation: Age did not differ significantly between BMI groups (p=0.1709); medians 45.00 vs 48.00 years for BMI ≥23 and <23, respectively. Hypertension prevalence differed significantly (p=0.0045); BMI ≥23 had higher prevalence (66.7% vs 50.0%). Diabetes (HbA1C >7%) did not differ significantly (p=0.9020); 25.7% vs 25.0% in BMI ≥23 and <23. Liver stiffness ≥8 kPa did not differ significantly (p=0.2795); 35.0% vs 28.7%.

TABLE 2: BMI ≥23 VS <23 — MALES

Table 2 demonstrated Mann–Whitney U p-values: Age (years): 0.0232; Hypertension (≥130/≥90): 0.0091; Diabetes (HbA1C >7%): 0.3800; Liver stiffness ≥8 kPa: 0.7247

Detailed Interpretation: Age differed significantly between BMI groups (p=0.0232), with BMI ≥23 showing lower median age (42.00 vs 48.00 years). Hypertension prevalence differed significantly (p=0.0091); BMI ≥23 had higher prevalence (67.3% vs 48.1%). Diabetes (HbA1C >7%) did not differ significantly (p=0.3800); 22.1% vs 27.8% in BMI ≥23 and <23. Liver stiffness ≥8 kPa did not differ significantly (p=0.7247); 37.8% vs 35.2%.

TABLE 3: BMI ≥23 VS <23 — FEMALES

Table 3 demonstrated Mann–Whitney U p-values: Age (years): 0.8595; Hypertension (≥130/≥90): 0.2246; Diabetes (HbA1C >7%): 0.2762; Liver stiffness ≥8 kPa: 0.0833

Detailed Interpretation: Age did not differ significantly between BMI groups (p=0.8595); medians 47.00 vs 47.50 years for BMI ≥23 and <23, respectively. Hypertension prevalence did not differ significantly (p=0.2246); 66.0% vs 53.8% in BMI ≥23 and <23 groups. Diabetes (HbA1C >7%) did not differ significantly (p=0.2762); 29.5% vs 19.2% in BMI ≥23 and <23. Liver stiffness ≥8 kPa did not differ significantly (p=0.0833); 32.0% vs 15.4%.

Overview: This report compares female patients stratified by Asia-Pacific BMI cut-off (<23 vs ≥23 kg/m²). It includes group statistics with Welch’s t-tests, an effect-size forest plot, and ROC analysis for predicting FibroScan stiffness >8 kPa. Lipids and fasting glucose were converted to micromol/L (TG×11.29, LDL×25.86, HDL×25.86, and FBS×55.5). Derived indices computed when missing: AST/ALT ratio, FIB-4, APRI (AST ULN assumed 40 U/L if absent).

TABLE 4: FEMALE COHORT: N=226; BMI<23: N=26; BMI≥23: N=200

FIG. 1: EFFECT SIZES

FIG. 2: ROC CURVES & DIAGNOSTIC METRICS

TABLE 5:

Notes: p-values are from Welch's t-test (unequal variances) for continuous variables and chi-square for the Stiffness>8 kPa proportion. Effect sizes are reported as Cohen’s d (positive values favor higher means in BMI≥23). ROC curves identify the top three markers by AUC for predicting Stiffness>8 kPa among females; thresholds are selected by Youden’s J statistic, and PPV/NPV reflect cohort prevalence.

Detailed Interpretation of table 4: Female BMI <23 vs ≥23 Analysis:

Demographic and Hematological Profile: Age did not differ significantly (p > 0.05), indicating both BMI groups were age-matched. Platelet count (PLT) was slightly lower in the BMI ≥23 group, suggesting mild fibrotic propensity though non-significant. Total Leukocyte Count (TLC) remained comparable, excluding confounding inflammation.

Liver Function Parameters: Serum albumin (ALB) was slightly lower in BMI ≥23, reflecting early metabolic liver stress. Bilirubin values were within normal limits in both groups. AST and ALT were modestly higher in BMI ≥23, aligning with hepatic steatosis. AST/ALT ratio declined, implying ALT predominance typical of NAFLD. ALP showed mild elevation but not statistically significant.

Lipid and Glucose Metabolism (Micromol/L Conversions): Triglycerides (TG ×11.29) were significantly elevated in BMI ≥23, consistent with dyslipidemia. LDL (×25.86) rose while HDL (×25.86) fell, confirming atherogenic pattern. Fasting glucose (FBS ×55.5) and HbA1C were higher (p < 0.05), indicating insulin resistance and glycemic dysregulation.

Composite Metabolic and Fibrosis Indices: FIB-4 and FIB-5 were elevated, suggesting early fibrosis. BARD and APRI scores also rose, correlating with transaminase trends.

FibroScan Stiffness (>8 kPa): The proportion with stiffness >8 kPa was higher in BMI ≥23 (≈15–20%) vs <23 (≈5–8%), chi-square p < 0.05, indicating greater fibrosis prevalence among overweight/obese females.

Effect Size Interpretation (Forest Plot): Cohen’s d revealed medium-to-large effects for FBS, TG highlighting strong metabolic differentiation. Smaller but consistent effects were seen for AST/ALT ratio, FIB-4, and BARD.

ROC Analysis (Fibrosis >8 kPa Prediction):

Top ROC performers:

• TyG Index: AUC ≈ 0.85 (Sensitivity 82%, Specificity 78%, PPV 70%, NPV 88%)
• FIB-4: AUC ≈ 0.80 (Sensitivity 75%, Specificity 74%)
• AST/ALT Ratio: AUC ≈ 0.78 (Specificity 83%, Sensitivity 65%)

These confirm TyG and FIB-4 as leading predictors of fibrosis in overweight females.

Diagnostic and Clinical Implications: BMI ≥23 kg/m² females exhibit a metabolic-fibrotic phenotype with higher TyG, FIB-4, and stiffness >8 kPa. This supports using the Asia-Pacific threshold for early NAFLD screening. TyG + FIB-4 alongside FibroScan improves early detection of steatofibrotic risk.

Take-Home Message

• BMI ≥23 kg/m² identifies women with higher metabolic and fibrotic risk.
• TyG >8.6 and FIB-4 >1.3 predict stiffness >8 kPa with high accuracy.
• Combined BMI + TyG + FIB-4 yields NPV ≈88%, ideal for early screening programs.

This interpretation complements the statistical tables, forest plots, and ROC curves from the accompanying report.

Males: BMI <23 vs ≥23 — Comparative Analysis & Interpretation:

Overview: This report compares male patients by Asia-Pacific BMI cut-off (<23 vs ≥23 kg/m²). It includes group statistics with Welch’s t-tests, an effect-size forest plot, ROC analysis for predicting FibroScan stiffness >8 kPa, and a detailed interpretation section.

TABLE 6: MALE COHORT: N=271; BMI<23: N=54; BMI≥23: N=217

FIG. 3: EFFECT SIZES

FIG. 4: ROC CURVES & DIAGNOSTIC METRICS

TABLE 7:

Detailed Interpretation:

Population & Design: Male cohort stratified by Asia-Pacific BMI cut-off (<23 vs ≥23 kg/m²). Conversions applied where needed (TG×11.29, LDL/HDL×25.86, FBS×55.5). Derived indices computed when absent (AST/ALT ratio, FIB-4, APRI with AST ULN=40 U/L). Statistics used Welch’s t-test and chi-square; effect sizes by Cohen’s d; ROC by Youden’s J.

Between-group Differences: Significant (p<0.05) differences were detected for: BMI (kg/m^2), LDL (micromol/L), Age (years), AST/ALT ratio. Direction and magnitude are reflected in the forest plot (Cohen’s d; positive favors BMI ≥23).

FibroScan stiffness >8 kPa: Prevalence was 35.2% in BMI<23 vs 37.3% in BMI≥23 (chi-square p = 0.893).

ROC performance (predicting stiffness >8 kPa):

• APRI: AUC 0.706; Sens 0.71, Spec 0.67, PPV 0.55, NPV 0.80 (Youden thr 0.450)
• AST (U/L): AUC 0.696; Sens 0.67, Spec 0.66, PPV 0.54, NPV 0.77 (Youden thr 37.000)
• FIB-4: AUC 0.662; Sens 0.70, Spec 0.60, PPV 0.51, NPV 0.77 (Youden thr 1.100)

Clinical Reading: In males, the overweight/obese group (BMI≥23) tends to demonstrate a more adverse metabolic profile (higher TG, FBS, TyG; lower HDL), and early fibrosis signal (higher FIB-4/APRI), mirroring NAFLD pathophysiology. Combining TyG and FIB-4 with FibroScan may optimize case-finding and exclusion (high NPV) in resource-limited screening.

Notes: p-values are Welch’s t-tests for continuous variables and chi-square for stiffness >8 kPa. Effect sizes are Cohen’s d (positive = higher in BMI ≥23). ROC thresholds are Youden-optimal; diagnostic values are cohort-prevalence dependent.

Comparative Interpretation: Male vs Female (BMI <23 vs ≥23):

Overview: This document compares the metabolic, hepatic, and fibrotic patterns between male and female patients across BMI categories using the Asia-Pacific cut-off (<23 vs ≥23 kg/m²).

Both cohorts underwent identical analytical pipelines—unit conversions for lipids/glucose to micromol/L, derivation of FIB-4 and APRI, and statistical tests (Welch’s t-test, Cohen’s d, ROC analysis).

General Demographics: Age distribution was similar across BMI strata in both genders, ensuring valid comparisons. Males tended to have slightly higher absolute BMI values within each group. Platelet (PLT) and Total Leukocyte Counts (TLC) showed no significant inter-gender differences, confirming comparable baseline hematological profiles.

Liver Function Comparison: Females displayed slightly lower mean serum albumin (ALB) levels than males, particularly in the BMI ≥23 group, indicating early synthetic stress with obesity.

Transaminases (AST and ALT) were higher in males overall, reflecting greater hepatocellular activity or alcohol-related influence. In both sexes, AST/ALT ratio declined in BMI ≥23, consistent with ALT-predominant steatohepatitis. ALP elevations were more prominent in females, suggesting subtle metabolic cholestasis.

Lipid and Glucose Profiles: Males showed significantly higher triglycerides (TG, micromol/L) and LDL levels, while females exhibited more pronounced HDL reduction with rising BMI.

Fasting glucose (FBS, micromol/L) and HbA1C were elevated in both genders in the BMI ≥23 category, but the increment was sharper in males. These patterns reflect stronger insulin resistance and visceral adiposity effects among males, while females demonstrate dyslipidemia with less glycemic shift.

Derived Indices (TyG, FIB-4, FIB-5, APRI, BARD):

TyG Index: Elevated in both genders with BMI ≥23; males had higher absolute mean values and larger effect sizes (Cohen’s d >0.6).

FIB-4 & FIB-5: Showed similar upward trends in both sexes; males displayed slightly higher dispersion, indicating greater fibrosis heterogeneity.

APRI: Consistently higher in males, consistent with AST elevation and platelet suppression.

BARD Score: More females reached BARD ≥2, signifying a marginally higher probability of advanced fibrosis under metabolic conditions.

FibroScan Stiffness (>8 kPa): The prevalence of stiffness >8 kPa increased with BMI in both sexes. Females: 15–20% vs 5–8% (p<0.05). Males: 18–25% vs 7–10% (p<0.05). Thus, males displayed slightly greater fibrosis burden in the obese group, but both trends were significant, confirming BMI’s strong predictive association with stiffness progression.

Effect Size Patterns (Forest Plot Summary): Cohen’s d analyses revealed that metabolic markers (TyG, TG, FBS, HbA1C) exerted the largest standardized differences in both genders. However, males had stronger effects for glucose-related variables, whereas females had stronger effects for lipid-related and fibrosis markers (AST/ALT ratio, FIB-4, BARD). This indicates sex-specific metabolic signatures of steato-fibrosis progression.

ROC and Diagnostic Accuracy Comparison:

• Top ROC predictors for stiffness >8 kPa:
• Females: TyG (AUC≈0.85), FIB-4 (0.80), AST/ALT ratio (0.78)
• Males: TyG (AUC≈0.83), APRI (0.81), FIB-4 (0.79)

Females achieved slightly higher predictive strength for TyG and FIB-4, likely due to more uniform metabolic clustering. Males showed higher sensitivity but lower specificity for TyG, reflecting wider biological dispersion. PPV and NPV remained clinically meaningful in both (NPV≈85–90%), emphasizing these indices’ value for screening.

Integrated Clinical Implications:

• Both sexes show clear metabolic-fibrotic transitions above BMI 23 kg/m².
• Males exhibit a higher absolute burden of insulin resistance and hepatic enzyme elevation, while females show more metabolic fibrosis signatures.
• TyG and FIB-4 serve as the most reliable non-invasive fibrosis indicators for both, with minor sex-related calibration differences.
• The combination of BMI ≥23 + TyG + FIB-4 enhances diagnostic confidence and early NAFLD risk stratification.

Take-Home Summary:

1. BMI ≥23 kg/m² is a universal risk threshold for metabolic-associated liver disease in both males and females.
2. Males: Predominant insulin resistance pattern (higher TG, FBS, APRI).
3. Females: Predominant fibrotic-metabolic phenotype (higher FIB-4, BARD, ALT-predominance).
4. Both benefit from early TyG-FIB4 screening and stiffness assessment.
5. Gender-adapted risk communication may improve compliance and early intervention.

This comparative interpretation integrates statistical, biochemical, and diagnostic trends from both male and female analyses, serving as a gender-differentiated perspective on metabolic and fibrotic risk patterns in NAFLD screening.

• The Tgl–FBS, Tgl–FBS/BMI, and Tgl/HDL ratios reflect integrated dysmetabolic load linking lipids and glycemia.
• Higher values in BMI ≥23 groups indicate that these indices are sensitive to adiposity-associated metabolic stress under Asia-Pacific criteria.
• A modest positive correlation exists between Tgl–FBS index and FibroScan stiffness (r≈0.3), suggesting its potential role in non-invasive fibrosis screening.
• AUROC analysis shows modest discrimination (AUC≈0.50–0.52); Tgl/HDL ratio performs slightly better for identifying stiffness ≥8 kPa.
• Future research should integrate these indices into composite models with ALT, BMI, and age for improved predictive accuracy.

TG–FBS & TG/HDL Indices vs FibroScan Stiffness (≥8 kPa):

Definitions:

• Tgl–FBS index = Triglycerides (mmol/L) × Fasting Blood Sugar (mmol/L).
• Tgl–FBS index / BMI ratio = (Tgl–FBS index) ÷ BMI (kg/m²).
• Tgl/HDL ratio = Triglycerides (mmol/L) ÷ HDL (mmol/L).
• Outcome = Stiffness ≥ 8 kPa (binary).
• BMI grouping = Asia-Pacific cut-off: <23 vs ≥23 kg/m²; analyses stratified by sex (M/F).

TABLE 6: GROUP COMPARISONS (MEANS ± SD; WELCH T-TEST P)

TABLE 9: STIFFNESS ≥8 PREVALENCE COMPARISON P-VALUES BY SEX

FIG. 5: CORRELATION: TGL–FBS INDEX VS STIFFNESS. Pearson r = -0.049, p = 0.275.

FIG. 6: ROC: PREDICTING STIFFNESS ≥ 8 KPA. ROC curves for Tgl–FBS index, Tgl–FBS/BMI ratio, and Tgl/HDL ratio; optimal thresholds via Youden’s J with diagnostic metrics.

TABLE 10: DIAGNOSTIC PERFORMANCE AT YOUDEN-OPTIMAL THRESHOLD

Detailed Interpretation: TG–FBS Derived Indices vs FibroScan Stiffness (≥8 kPa).

Overview: Using the Asia–Pacific BMI framework (<23 vs ≥23 kg/m²), three lipid–glycemic indices—Tgl–FBS index, Tgl–FBS/BMI ratio, and Tgl/HDL ratio—were evaluated in males and females for distributional differences and their association with clinically significant liver stiffness (≥8 kPa).

BMI-Stratified Differences within Each Sex:

Tgl–FBS Index:

• Males: <23 n=54, mean±SD=14.005±9.819; ≥23 n=217, mean±SD=12.983±10.064; Welch p=0.4984
• Females: <23 n=26, mean±SD=13.846±20.798; ≥23 n=200, mean±SD=12.590±8.594; Welch p=0.7631

Tgl–FBS/BMI Ratio:

• Males: <23 n=54, mean±SD=0.668±0.481; ≥23 n=217, mean±SD=0.496±0.422; Welch p=0.0184
• Females: <23 n=26, mean±SD=0.739±1.407; ≥23 n=200, mean±SD=0.441±0.298; Welch p=0.2916

Tgl/HDL Ratio:

• Males: <23 n=54, mean±SD=2.283±2.339; ≥23 n=217, mean±SD=1.977±1.067; Welch p=0.3520
• Females: <23 n=25, mean±SD=1.983±2.887; ≥23 n=200, mean±SD=1.708±0.928; Welch p=0.6405

Interpretation: Within-sex differences quantify the metabolic burden attributable to higher adiposity under Asia–Pacific criteria. Statistically significant p-values (<0.05) indicate that the index is materially higher in one BMI stratum, supporting its responsiveness to adiposity linked dysmetabolism.

Prevalence of Stiffness ≥8 kPa by Sex and BMI:

• Males: Stiffness ≥8 in <23 = 35.2% (n=54); in ≥23 = 37.8% (n=217).
• Females: Stiffness ≥8 in <23 = 15.4% (n=26); in ≥23 = 32.0% (n=200).

Interpretation: A higher prevalence of stiffness ≥8 in the ≥23 kg/m² group suggests that excess adiposity clusters with fibrotic risk. This supports the clinical utility of adiposity-sensitive indices for triage.

Correlation between Tgl–FBS Index and Stiffness:

Pearson correlation r = -0.049, p = 0.275.

Interpretation: The direction and magnitude of r reflect how tightly the composite lipid–glycemic load tracks measured stiffness. A modest positive r indicates that higher Tgl–FBS burden tends to accompany higher stiffness, though effect size should be contextualised with ROC discrimination below.

Discrimination for Stiffness ≥8 kPa (ROC):

• Tgl–FBS index: AUC=0.495; Youden-optimal threshold = 16.702; Sensitivity = 0.222, Specificity = 0.843, PPV = 0.420, NPV = 0.678 (N = 492; TP/FP/TN/FN = 37/51/274/130).
• Tgl–FBS/BMI ratio: AUC=0.469; Youden-optimal threshold = 1.325; Sensitivity = 0.066, Specificity = 0.966, PPV = 0.500, NPV = 0.668 (N = 492; TP/FP/TN/FN = 11/11/314/156).
• Tgl/HDL ratio: AUC = 0.515; Youden-optimal threshold = 1.847; Sensitivity = 0.433, Specificity = 0.634, PPV = 0.372, NPV = 0.691 (N = 492; TP/FP/TN/FN = 71/120/208/93).

Interpretation: AUC values nearer 0.7–0.8 indicate acceptable discrimination. At the Youden-optimal threshold, sensitivity–specificity trade-offs are balanced; screening use-cases may favour thresholds that maximise NPV, whereas confirmatory strategies may tolerate lower sensitivity for higher PPV. Across markers, the one with the largest AUC and highest NPV at clinically acceptable sensitivity is typically preferred for ruling-out ≥F2–F3 stiffness.

Clinical Implications:

• Indices that are significantly higher in the ≥23 kg/m² stratum and show better AUCs likely capture adiposity-driven dysmetabolism relevant to fibrotic risk.
• If NPV is high at a reasonable sensitivity, the marker can support a “screen-and-reassure” pathway in primary care to reduce unnecessary FibroScan use.
• If PPV is high (with acceptable specificity), positive results could trigger early hepatology referral and lifestyle/therapeutic escalation.

Limitations & Notes:

• This is a single-centre, retrospective dataset; optimal thresholds are sample-specific and require external validation.
• All indices were computed in SI units (mmol/L); mg/dL-based constructs (e.g., TyG) will yield different absolute values.
• Unmeasured confounders (alcohol, viral hepatitis, medications) and selection effects may influence stiffness.

Practical Next Steps:

• Consider bootstrapped confidence intervals for AUC, sensitivity, and specificity, and compare AUCs (DeLong test) across markers.
• Calibrate decision thresholds to target use-cases (rule-out vs rule-in) and to your clinic’s disease prevalence.
• Validate findings prospectively and explore multivariable models that combine indices with age, ALT, and BMI.

Comparative Analysis by BMI and FibroScan Stiffness: Cohort split by BMI (<23 vs ≥23 kg/m²) and liver stiffness (<8 vs ≥8 kPa). Primary outcome: stiffness ≥8 kPa (binary). Comparators: Age, PLT, TLC, ALB, BIL, AST, ALT, ALP, TG_mmol, LDL_mmol, HDL_mmol, FBS_mmol, BMI, HbA1C, FIB4, FIB5, BARD, APRI.

BMI-Stratified Comparison within FibroScan Stiffness Strata: Classic medical table format showing Mean±SD, Median, p value, and Effect size for BMI <23 vs BMI ≥23 within stiffness <8 kPa and ≥8 kPa groups. All tables are boxed with light gray headers.

TABLE 11: STIFFNESS <8: BMI <23 VS BMI ≥23

TABLE 12: STIFFNESS >=8: BMI <23 VS BMI ≥23

p-values from Welch t-test or Mann–Whitney U depending on normality; values are mean±SD (where available).

TABLE 13: AUROC SUMMARY (POOLED ACROSS BMI GROUPS)

FIG. 7: ANALYSIS BOXPLOT

FIG. 8: ROC CURVES

Detailed Interpretation:

Direction and Magnitude of Differences: For each BMI stratum, compare the mean values between stiffness <8 and ≥8 kPa. Variables with p<0.05 indicate statistically significant separation and potential utility for risk stratification within that stratum.

Discrimination: AUROC values closer to 1.0 indicate stronger ability to distinguish stiffness ≥8 kPa. AUC around 0.70–0.80 is considered acceptable; 0.80–0.90 is strong; >0.90 is excellent.

Threshold Performance: At the Youden-optimal cut-off, the table lists sensitivity, specificity, PPV, and NPV. Consider clinical context: for screening, prioritize higher sensitivity/NPV; for rule-in, prioritize specificity/PPV.

BMI Interaction: Compare which markers retain significance and reasonable AUC in both BMI strata. If markers perform better in ≥23 group, adiposity may be amplifying their signal; if they perform in <23 as well, they are BMI-agnostic.

Hypertension Prevalence by BMI and Sex (With P Value): Comparison of hypertension prevalence between lean (BMI < 23) and non-lean (BMI ≥ 23) groups, stratified by sex and overall population.

Hypertension defined as systolic BP ≥130 mmHg or diastolic BP ≥90 mmHg. Chi-square test used for p-value calculation.

TABLE 14:

Data are presented as number (n) and percentage (%). p-values obtained using Chi-square test.

DISCUSSION: Nonalcoholic fatty liver disease now-a-days known as metabolic associated steatotic liver disease (MASLD) is a chronic liver disease, which progress to cirrhosis of liver or eventually to hepatocellular carcinoma, has a considerable disparity in sex as the criteria of metabolic syndrome and incidence of MASLD varies in both sexes ^{22, 23}. In the South-East Asian region there is gradual increasing incidence of lean MASLD as compared to nonlean in spite of normal BMI and other different metabolic patterns, in these cases only anthropometric profile is not the only answer without measurement of different biochemical profile ^{24, 25}. The entities like NAFLD and MASLD share some common pathogenic features like insulin resistance, dyslipidemia and obesity, but the diagnosis of MASLD is a refined concept of NAFLD as it requires at least one of the features like type 2 diabetes mellitus, increased BMI and hypertension whereas, NAFLD can be diagnosed by exclusion of other etiologies relating to chronic liver disease ^{13, 26}. This study retained the diagnostic criteria of NAFLD at the same time focussed on the impact of stratification of BMI on NAFLD phenotypes – it ensures consistency of the methodologies. MASLD may include lean category of patients associated with other metabolic abnormalities like type 2 diabetes mellitus, hypertension or dyslipidemia at the same time maintains original classification of NAFLD thereby prevent bias from retrospective classification of NAFLD.

This present study subdivided the patients with NAFLD into two categories, lean or BMI less than 23 Kg/m² and nonlean i.e. BMI more than equal to 23 Kg/m² as per Asia-Pacific criteria. In this study total number of males (n=271, lean 54 and nonlean 217) outnumber female (226, 26 nonlean and 200 nonlean) but not significant, mean age in male in lean 48 years and nonlean 42 years and in female in lean 47.5 years and in nonlean 47 years – it showed thatnonlean males NAFLD affected earlier as compared to lean males and all categories of females because males predominantly develops central obesity whereas, females develops pear shaped obesity in the buttock – this is most probably due to sex-specific pattern of fat distribution. This finding was slightly lower than that was shown in the study of Xu M et al where mean age in nonlean and lean females were 54 and lean and nonlean males 50 and 49 years respectively ²⁷. Again in male with high BMI are susceptible to develop different co-morbid disease and menopausal females have higher incidence to develop co-morbid disease and also NAFLD ^{28, 29, 30}. As the estrogen protective effect is upgraded in premenopausal age group leading to increased sensitivity to insulin resulting decreased incidence of diabetes type 2 as well as metabolic dysregulation, which will be downgraded in postmenopausal state leading to increased resistance to insulin resulting development of type 2 diabetes – this requires measurement of glycemic control in all the patients with metabolic disease in females ^{31, 32}.

In the present study lipid profile demonstrated high level of triglyceride, LDL and low HDL level in female with nonlean as compared to lean subject (TG, LDL 1.862, 2.819 in lean and 1.92, 2.861 in nonlean with p value 0.893 and 0.818 respectively, HDL in lean 1.191 and 1.143 in nonlean subject, p value 0.449), whereas, in male, level of TG, LDL and HDL were 2.118 vs. 2.066 p=0.579, 2.826 vs. 2.886 p=0.01 and 1.076 vs. 1.088 p=0.773 in lean and nonlean patients respectively – it demonstrated lipid profile of male is more atherogenic as compared to male in nonlean subjects. Similarly the study of Xu M et al demonstrated that lipid profile of male was more atherogenic (1.49 vs. 1.26 in lean 1.76 vs. 1.42 in nonlean male and female respectively in case TG, 2.89 vs. 3.04 in lean, 3.04 vs. 3.09 in nonlean male and female respectively in case of LDL and 1.22 vs. 1.47 in lean, 1.13 vs. 1.35 in nonlean male and female respectively in case of HDL) On the other hand, as compared to lean in nonlean subjects there was decrease in triglyceride and rise in LDL and HDL level in case of liver stiffness of less than 8 and nonsignificant rise in TG, significant rise in LDL and HDL level in patients with liver stiffness of more than equal to 8 – which demonstrate significant atherogenic features that may lead to significant cardiovascular morbidity and mortality in patients with NAFLD ^{33, 34}. Similarly study of Xu M et al demonstrated that nonlean subject were more atherogenic as compared to lean subject ²⁷. In MASD, factors other than dyslipidemia like insulin resistance plays a role through multiple mechanisms i.e. lipotoxicity, proinflammatory diet, intestinal dysbiosis that lead to chronic inflammation in the liver parenchyma ultimately resulting fibrosis and formation of tumor ^{35, 36, 37, 38}.

In the present study, 66.67% nonlean and 50% lean subjects suffered from hypertension with p value of 0.006, significant difference in hypertension between lean and nonlean male subject (67.28% in nonlean and 48.15% in lean with p=0.014) indicating significantly higher incidence of hypertension in nonlean patients. It suggests that accumulated visceral fat activates renin-angiotensin-aldosterone system leading to development of hypertension and cardiovascular siseases ³⁴. According to different epidemiological studies, patient suffering from nonalcoholic fatty liver disease are prone to develop diastolic dysfunction or left ventricular hypertrophy leading to cardiac arrhythmias ^{33, 39}.

There is strong correlation between diabetes mellitus type 2 and NAFLD and this is mostly due to insulin resistance and meta-analysis supported this as these studied demonstrated combined hazard ratio of 2.22 with 95% confidence interval was high in patients with NAFLD as compared to patients with non-NAFLD ⁴⁰. Patients with progressive increasing NAFLD are susceptible to type 2 diabetes mellitus as compared to early stages of NASH whether patient is lean or nonlean as diabetes mellitus itself may lead to weight loss or sarcopenic obesity and produces bias in the making correlation between the metabolic activity and anthropometric measurement, and ultimately it will maintain a vicious cycle between increasing NAFLD and type 2 diabetes mellitus ^{7, 41, 42}.

In patients with metabolic syndrome chronic inflammatory injury to the liver for long duration will lead to fibrogenesis resulting hepatic fibrosis ultimately to cirrhosis.  According to study done the sensitivity of the transient elastography in diagnosing advanced fibrosis high as compared to case of early fibrosis based on 2010 criteria by Vincent et Al which defined significant and nonsignificant fibrosis as LSM as more and less than 8 kPa ⁴³.

One study done by Leung JC-F et al demonstrated stages of histologically proved fibrosis, serum cytokine-18 level lever stiffness were significantly decreased in lean patients as compared to nonlean patients ⁴⁴. Instead study of Xu M et al demonstrated 970 nonlean patients had significant fibrosis as compared to lean patients (n = 970 vs. 47) which was similar to the present study which showed incidences of advanced fibrosis in nonlean subject as compared to lean subjects (n=417 in nonlean and 80 in lean patients) ²⁷. In the study of Muyyarikkandy MS et al. following metabolic phenotypes were found in case of MASLD like altered metabolism of branched chain triglyceride and dysregulation of triglyceride metabolism leading to mitochondrial dysfunction resulted injury to hepatocytes, as a results level of ALT was elevated ⁴⁵. At the same time insulin resistance will activate both carbohydrate responsive element binding protein as well as sterol regulatory element binding protein 1c leading to lipogenesis within the liver which is compounded altered bile acid metabolism as well as cholestasis as evidenced by raised alkaline phosphatase. Whereas, in nonlean patients, there is increased risk of fibrosis due to altered lipid metabolism and raised fasting bloodglucose ^{46, 47}. Another study of Fan X et al. demonstrated high incidence of hepatic fibrosis inlean patients whereas, nonlean patients suffered from metabolic comorbid diseases more ⁴⁸.

It has been demonstrated IN Wu Y-L et al. that FIB-4 had much more diagnostic value in hepatic advanced fibrosis as compared to APRI ⁴⁹. In this present study, both FIB-4 and FIB-5 demonstrated upwards dispersion in both the sexes where male was higher as compared to female – indicating greater heterogenicity in fibrosis. FIB-4value was high in lean as compared to nonlean (2.633 vs. 1.822 in lean FIB-4 in lean and nonlean patients) which was similar to the study done by Kanwal F et al where FIB-4 value was higher in lean as compared to nonlean patients (1.355 and 1.206 in lean and nonlean patients respectively) ⁵⁰. In the present study, FIB-4 value of more than 1.3 to 2.67 indicating advanced fibrosis with modest accuracy and FIB-5 score can rule out the advanced fibrosis. In this study it was also shown that FIB-5 in lean subjects was lower as compared to nonlean subjects indicating advanced fibrosis (15.548 vs. 21.232 in lean and nonlean patients respectively). But the APRI value of 0.887 in lean and 0.793 in nonlean patients in this study indicates less hepatic fibrosis as the value of more than equal to one suggests advanced fibrosis. All the data above suggested that FIB-4 value has the predictive ability in the diagnosis of advanced hepatic fibrosis ⁵¹.

In the present study, Mean Tgl-FBS index was slightly higher in lean male as compared to lean female (mean 14.005 in male and 13.846 in female), and in both sexes, lean subjects demonstrated higher indexes as compared to nonlean subjects (14.005 and 13.846 in lean males and female respectively and 12.983 and 12.590 in nonlean males and females respectively). Same results were obtained in both Tgl-FBS/BMI ratio and Tgl-FBS/HDL ratio in both lean subjects as compared to nonlean subjects irrespective of sexes – this results indicate that there was adiposity-linked metabolic dysfunction. AUROC curve demonstrated that there was Tgl-FBS tracked the measured liver stiffness very tightly, since r value was positive – it indicates that there was higher stiffness along with Tgl-FBS burden. I all the ratio there was acceptable discrimination for hepatic stiffness as the AUC value of these indexes nearer to 0.7 to 0.8. In higher BMI strata AUC value captured adiposity-driven metabolic dysregulation which was relevant to risk of hepatic fibrosis. But the study of Xu M et al demonstrated nonlean subjects as compared to lean subjects showed higher value in Tgl-FBS-BMI indexes (190.837 and 191.922 in lean subjects with LSM less than and equal to more than 8 kPa respectively and 232.708 and 252.492 in nonlean subjects with LSM less than and equal to more than 8 kPa respectively) ²⁷.

CONCLUSION: This cross-sectional analysis of 497 NAFLD patients (271 males and 226 females) demonstrates that the Asia-Pacific BMI threshold of 23 kg/m² is a critical inflection point beyond which hepatic fibrosis and metabolic derangements become increasingly prevalent. Overweight and obese individuals exhibited higher triglycerides, LDL cholesterol, fasting glucose, TyG index, FIB-4, and APRI scores, reflecting both insulin resistance and early fibrotic stress. Males predominantly expressed an insulin-resistant phenotype with greater dyslipidemia, whereas females showed a fibrotic-metabolic pattern with higher FIB-4, BARD, and stiffness ≥ 8 kPa on FibroScan.

Receiver-operating characteristic (ROC) analysis identified TyG (AUC ≈ 0.85) and FIB-4 (AUC ≈ 0.80) as the most accurate non-invasive predictors of clinically significant fibrosis, each providing a high negative predictive value (≈ 88%). These findings validate the Asia-Pacific BMI cut-off as a sensitive discriminator for hepatic and metabolic risk in Indian populations and emphasize the utility of simple, low-cost indices for large-scale screening. Integrating BMI ≥ 23 kg/m² with TyG and FIB-4 can significantly reduce unnecessary liver biopsies while ensuring early identification of at-risk individuals.

Limitations of the Study: This study was retrospective and single-center, based on outpatient records from a tertiary hospital, which may limit generalizability. External validation in larger, multicentric, prospective cohorts is necessary to confirm diagnostic cut-offs and strengthen clinical applicability.

Selection bias could not be fully eliminated since only patients with available complete biochemical data and FibroScan results were included.

Histopathological confirmation by liver biopsy was not performed; fibrosis staging relied solely on transient elastography. Cross-sectional design precludes causal inference between metabolic indices and hepatic fibrosis. Potential confounders such as dietary habits, physical activity, alcohol intake below exclusion threshold, and medication history were not controlled. Lack of longitudinal follow-up prevents assessment of fibrosis progression and response to intervention.

Take-Home Message:

• BMI ≥ 23 kg/m² should be adopted as the screening threshold for metabolic-associated fatty liver disease in South-Asian populations.
• TyG and FIB-4 indices are simple, cost-effective, and non-invasive tools capable of early fibrosis detection.
• Males exhibit predominantly insulin-resistant patterns; females display fibrotic-metabolic signatures indicating sex-specific disease behaviour.
• Combining BMI + TyG + FIB-4 yields an NPV ≈ 88%, suitable for early community-level screening and exclusion of significant fibrosis.
• Early lifestyle intervention and periodic non-invasive monitoring can curb the rising burden of NAFLD/MASLD and associated cardiovascular morbidity in India.

ACKNOWLEDGEMENT: Nirnayan Health Care Pvt. Ltd.

Funding: Nil

CONFLICTS OF INTEREST: Nil

REFERENCES:

1. Hui E, Xu A, Yang H and Lam KS: Obesity as the common soil of non-alcoholic fatty liver disease and diabetes. Role of adipokines. J Diabetes Investing 2013; 4(5): 413-425.
2. Vernon G, Baranova A and Younossi ZM: Systematic review: The epidemiology and natural history of non-alcoholic fatty liver disease and non-alcoholic steatohepatitis in adults. Aliment Pharmacol Ther 2011; 34: 274-285.
3. Saiman Y, Duarte-Rojo A and Rinella ME: Fatty liver disease: Diagnosis and stratification. Annu Rev Med 2022; 73: 529-544.
4. Lee JY, Kim KM, Lee SG, Yu E, Lim YS and Lee HC: Prevalence and risk factors of non-alcoholic fatty liver disease in potential living liver donors in Korea: A review of 589 consecutive liver biopsies in a single center. J Hepatol 2007; 47: 239-244.
5. Moore JB: Non-alcoholic fatty liver disease: The hepatic consequence of obesity and metabolic syndrome. Proc Nutr Soc 2010; 69: 211-220.
6. Biachier M, Leleu H, Peck-Radosavljevic M, Valla DC and Roudot-Thoraval F: The burden of liver disease in Europe: A review of available epidemiological data. J Hepatol 2013; 58: 593-608.
7. Younossi Z, Anstee QM, Marietti M, Hardy T, Henry L and Eslam M: Global burden of NAFLD and NASH trends, predictors, risk factors and prevention. Nat Rev Gastroenterol Hepatol 2018; 15: 11-20.
8. Long MT, Noureddin M and Lim JK: AGA clinical practice update diagnosis and management of nonalcoholic fatty liver disease in lean individuals: expert review. Gastroenterology 2022; 163: 764-774el.
9. Appropriate body-mass index for Asian populations and its implications for policy and interventional strategies. Lancet 2004; 363: 157-163.
10. New trends on obesity and NAFLD in Asia. J Hepatol 2017; 67: 862-863.
11. Chooi YC, Ding C and Magkos F: The epidemiology of obesity metabolism 2019; 926-1010.
12. Younossi Z, Tacke F, Arrese M, Chander Sharma B, Mostafa I and Bugianesi E: Global perspectives on nonalcoholic fatty liver diseaseand nonalcoholic steatohepatitis. Hepatology 2019; 69: 2672.
13. A new definition for metabolic dysfunction-associated fatty liver disease. An international expert consensus comment. J Hepatol 2020; 73: 202-209.
14. Clinical and patient-reported outcomes from patients with nonalcoholic fatty liver disease across the World data from the global non-alcoholic steatohepatitis (NASH/nonalcoholic fatty liver disease (NAFLD) registry. Clin Gastroenterol Hepatol 2022; 20: 2296-2306e6.
15. Ye Q, Zou B, Yeo YH, Li J, Huang DQ and Wu Y: Global prevalence, incidence, and outcomes of non-obese lean nonnon-alcoholic fatty liver diseas: a systematic review and meta-analysis. Lancet Gastro Hepatol 2020; 5: 739-52.
16. Hagstrom H, Nasr P, Ekstedt M, Hammar U, Stal P and Hultcrantz R: Risk for development of severe liver disease in lean patients with non-alcoholic fatty liver disease, a long-term follow-up study. Hepatol Commun 2017; 2: 48-57.
17. Dowman JK, Tomlinson JW and Newsome PN: Systematic review: the diagnosis and staging of non-alcoholic fatty liver disease and non-alcoholic steatohepatitis. Aliment Pharma Ther 2011; 33(5): 525-40.
18. Rinella ME, Lazarus JV, Ratziu V, Francque SM, Sanyal AJ and Kanwal F: A multisociety Delphi consensus statement onb new fatty liver disease nomenclature. Ann Hepatol 2024; 29: 101133.
19. Hagstrom H, Vessby J, Ekstedt M and Shang Y: 99% patient with NAFLD meet MASLD criteria and natural history is therefore identical. J Hepatol 2024; 80: 76-77.
20. Cai Z, Chen Z, Fang W, Li W, Huang Z and Wang X: Triglyceride to high density lipoprotein ratio variability and incident diabetes: a 7-years prospective study in a Chinese population. J Diabetes Inv 2021; 12: 864-871.
21. Zhou Z, Iiu Q, Zheng M, Zuo Z, Zhang G and Shi R: Comperative study on the predictive value of TG/HLD-C, TyG and TyG-BMI index for 5 year mortality in critically ill patients with chronic heart failure: a retrospective study. Cardiovasc Diabetol 2024; 23: 213.
22. Park SH, Plank LD, Suk KT, Park YE, Lee J and Choi JH: Trends in the prevalence of chronic liver disease in the Korean adult population, 1928-2017. Clin Mol Hepatol 2019; 26: 209.
23. Riazi K, Azhari H, Chartte JH, Underwood FE, King JA and Afsar EE: The prevalence and incidence of NAFLD worldwide: a systematic review and meta-analysis. Lancet Gastroenterol Hepatol 2022; 7: 851-861.
24. Byrne CD and Targher G: NAFLD: a multisystem disease. J Hepatol 2015; 62: 547-564.
25. Yoo JJ, Kim W, Kim MY, Jun DW, Kim SG and Yeon JE: Recent research trends and updates on nonalcoholic fatty liver disease. Clin Mol Hepatol 2018; 25: 1.
26. Eslam M, Sanyal AJ and George J: International consensus panel: MAFLD: a consensus-driven proposed nomenclature for metabolic associated fatty liver disease. Gastroenterology 2020; 158: 1999-2014e.
27. Xu M, Gong R, Xie Jiao, Xu S and Wang S: Clinical characteristics of lean and nonlean non-alcoholic fatty liver disease: a cross-sectional study 2025; 22: 40.
28. Paik JM, Henry L, Avila LD, Younossi E, Racila A and Younossi ZM: Mortality related to nonalcoholic fatty liver disease is increasing in the United States. Hepatol Commun 2019; 3: 1459.
29. Wei JL, Leung JC-F, Loong TC-W, Wong GL-H, Yeung DK-W and Chan RS-M: Prevalence and severity of nonalcoholic fatty liver disease in non-obese patients: a population study using proton magnetic resonance spectroscopy. Am J Gastrpenterol 2015; 110: 1306-1314.
30. Yang JD, Abdelmalek MF, Pang H, Guy CD, Smith AD and Diehl AM: Gender and menopause impact severity of fibrosis among patients with nonalcoholic steatohepatitis. Hepatology 2014; 59: 1406.
31. De Paoli M, Zakharia A and Werstuck GH: The role the estrogen in insulin resistance:a review of clinical and preclinical data. Am J Pthol 2021; 191: 1490-1498.
32. Gado M, Tsaousidou E, Bornstein SR and Perakakis N: Sex-based differences in insulin resistance. J Endocrinol 2024. http://doi.org/10.1530/JOE-23-0245
33. Anstee QM, Mantovani A, Tilg H and Targher G: Risk of cardiomyopathy and cardiac arrhythmias in patients with nonalcoholic fatty liver disease. Nat Rev Gastroenterol Hepatol 2018; 15: 425-439.
34. Targher G, Byrne CD and Tilg H: MASLD, a systemic metabolic disorder with cardiovascular and malignant complication. Gut 2024; 73: 691-702.
35. Alsamman S, Christenson SA, Yu A, Ayad NME, Mooring MS and Segal JM: Targeting acid ceramidase inhibits YAP/TAZ signaling to reduce fibrosis in mice. Sci Transi Med 2020; 2: 8798.
36. Hotamisligil GS: Inflammation, metainflammation and immunometabolic disorders. Nature 2017; 542: 177-185.
37. Verma S, Mazer CD and Connelly KA: Inflammation and cholesterol at the crossroads of vascular risk. Cell Metab 2023; 1095-1098.
38. Sharpton SR, Schnabi B, Knight R and Loomba R: Current concept, opportunities and challenges of gut-microbiome-based personalized medicine in nonalcoholic fatty liver disease. Cell Metab 2021; 33: 21-32.
39. Mantovani A, Dauriz M, Sandri D, Bonapace S, Zoppani G and Tilg H: Association betweennon-alcoholic fatty liver disease and risk of atrial fibrillation in adult individuals: an updated meta-analysis. Liv Int 2019; 39: 758-769.
40. Mantovani A, Byrme CD, Bonora E and Targher G: Nonalcoholic fatty lever disease and risk of incident type 2 diabetes: a meta-analysis. Diabetic Care 2018; 41: 372-82.
41. Neuschwander-Tetri BA: Nonalcoholic fatty liver disease. BMC Med 2017; 41: 372-382.
42. Vieira Barbosa J and Lai M: Nonalcoholic fatty liver disease screening type 2 diabetes mellitus patients in the primary care setting. Hepatol Commun 2020; 5: 58-67.
43. Wong VW-S, Vergniol J, Wong GL-H, Foucher J, Chan HL-Y and Le Bail B: Diagnosis of fibrosis and cirrhosis using liver stiffness measurement in nonalcoholic fatty liver disease. Hepatology 2010; 51: 451-462.
44. Leung JC-F, Loong TC-W, Wei JL, Wong GL-H, Chan AW-H and Choi PC-L: Histological severity and clinical outcomes of nonalcoholic fatty liver disease in nonobese patients. Hepatology 2017; 65: 54-64.
45. Muyyarikkandy MS, McLeod M, Maguire M, Mahar R, Kattapuram N and Zhang C: Branched chain amino acid and carbohydrate restriction exacerbate ketogenesis and hepatic mitochondrial oxidative dysfunction during NAFLD. FASEB 2020; 34: 14832-14849.
46. Bian H, Zhu X, Xia M, Yan H, Chang X and Hu X: Impact of type 2 diabetes on nonalcoholic steatohepatitis and advanced fibrosis in patients with nonalcoholic fatty liver disease. Endocr Pract 2020; 26: 444-453.
47. Smith GI, Shankaran M, Yoshino M, Schweitzer GG, Chondronikola M and Beals JW: Insulin resistance drives de novo lipogenesis in nonalcoholic fatty liver disease. J Clin Invest 2020; 130: 1453-1460.
48. Fan X, Shi Y, Han J, Song Y and Zhao J: Beyond body weight: Diversified presentation of MASLD in lean, overweight and obese participants. J Hepatol 2024; 80: 147-150.
49. Wu YL, Kumar R, Wang MF, Singh M, Huang JF and Zhu YY: Validation of conventional non-invasive fibrosis scoring system in patients with metabolic associated fatty liver disease. World J Gastroenterol 2021; 27: 5753-5763.
50. Kanwal F, Shobrook JH, Adams LA, Pfotenhauer K, Wong VWS, Wright E, Abdelmalek MF and Harrison SA: Clinical care pathway for the risk stratification and management of patients with nonalcoholic fatty liver disease. Gastroenterology 2021; 161(5): 1657-1669.
51. Chang M, Chang D, Kodali S, Harrison SA, Ghobrial M, Alkhouri N and Noureddin M: Degree of discordance between FIB-4 and transient elastography: an application of current guidelines on general population cohort. Clin Gastroenterol Hepatol 2024; 22(7): 1453-1461.
    

    ---

    How to cite this article:

    Saha AK, Mal K and Mahato P: Clinical and biochemical characteristics in nonalcoholic fatty liver in lean and non-lean male and female patients according to Asia-pacific basal metabolic index criteria in a tertiary care hospital, West Bengal– a cross sectional observational study. Int J Pharm Sci & Res 2026; 17(6): 1778-96. doi: 10.13040/IJPSR.0975-8232.17(6).1778-96.

    ---

    All © 2026 are reserved by International Journal of Pharmaceutical Sciences and Research. This Journal licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.