Meta-analytic reviews on the epidemiological and clinical features of lean NAFLD.
| Author, year, [ref] | Method | Findings | Conclusion |
|---|---|---|---|
| Li et al., 2026 [17] | The analysis included 54 studies involving 146,714 patients with lean or non-obese NAFLD. | Among patients with lean or non-obese NAFLD, pooled global diabetes prevalence was 15.6% (95% CI 10.8–22.7%), highest in South America at 41.3% (CI 39.1–43.5%). Geographic region and mean age (p < 0.05) explained 51.61% of heterogeneity. Prediabetes prevalence was 22.9% (95% CI 12.5–41.9%), highest in Europe at 34.4% (CI 23.0–51.4%); geographic region and country (p < 0.05) explained 73.65% of heterogeneity. | Overall, the pooled global prevalence of diabetes and prediabetes among lean or non-obese patients with NAFLD was 15.6% and 22.9%, respectively. These findings highlight the potential value of considering routine diabetes screening in this patient population. |
| Miao et al., 2025 [18] | Fourteen studies were included, including ten retrospective cohort studies. | Compared to high-BMI NAFLD, lean NAFLD was associated with higher all-cause mortality (HR 1.34, 95% CI: 1.23–1.47) and liver disease-related mortality (HR 2.14, 95% CI: 1.18–3.87). However, it was not associated with increased cardiovascular (HR 0.92, 95% CI: 0.65–1.30) or cancer-related mortality (HR 1.20, 95% CI: 0.90–1.60). | Lean NAFLD patients exhibit higher risk of all-cause, and liver disease-related mortality compared to patients with high BMI. There is a need for tailored interventions and further research to understand specific mortality risks in this subgroup of patients. |
| Fatima et al., 2024 [19] | Twenty-five studies were included in the analysis. | Compared to obese NAFLD, lean NAFLD was associated with lower risks of diabetes mellitus (RR 0.78, 95% CI 0.71–0.87, p < 0.00001), dyslipidemia (RR 0.87, 95% CI 0.79–0.95, p = 0.002), arterial hypertension (RR 0.80, 95% CI 0.74–0.87, p < 0.00001) and metabolic syndrome (RR 0.45, 95% CI 0.31–0.64, p < 0.00001). Lipid profile, blood pressure and other clinical parameters were also more favorable in the lean NAFLD group. | Lean NAFLD is associated with lower metabolic diseases, FBG, BP, and a more favorable lipid profile compared to obese NAFLD. |
| Souza et al., 2024 [20] | This metanalysis included 22 studies with over 1 million NAFLD patients (13.0% were lean). | Lean NAFLD was associated with higher LRM (RR 2.22, 95% CI 1.57–3.15; HR 2.26, 95% CI 1.14–4.51). In contrast, cardiovascular risk was lower only in unadjusted analyses (RR 0.82, 95% CI 0.70–0.95) and was comparable after adjustment (HR 0.89, 95% CI 0.77–1.02), with no clear difference in cardiovascular mortality (RR 1.09, 95% CI 0.71–1.66; HR 1.26, 95% CI 0.89–1.78) vs. non-lean NAFLD. | Compared to non-lean NAFLD patients, lean NAFLD subjects exhibit worse liver outcomes, but similar CV outcomes highlighting the importance of closely monitoring NAFLD individuals irrespective of their BMIs. |
| Wongtrakul et al., 2024 [21] | Fourteen studies were included, comprising 94,181 patients with NAFLD, of whom 11.3% were lean, and 7,443 deaths over a median follow-up of 8.4 years. | Lean NAFLD was associated with increased all-cause mortality compared to non-lean NAFLD (random-effects HR 1.61, 95% CI 1.37–1.89; I2=77%). This association was consistent across stratified and sensitivity analyses and remained independent of age, sex and cardiometabolic risk factors, with no evidence of significant publication bias. | Lean NAFLD is associated with an approximately 1.6-fold higher mortality risk, supporting further mechanistic studies to clarify the complex relationship between lean NAFLD and excess mortality. |
| Souza et al., 2024 [22] | Eight studies totaling 56,745 NAFLD individuals (11% with lean NAFLD) and 704 cases of incident GI cancers were included. | Compared to non-lean NAFLD, lean NAFLD was associated with higher risks of hepatic (RR 1.77, 95% CI 1.15–2.73), pancreatic (RR 1.97, 95% CI 1.01–3.86) and colorectal cancers (RR 1.53, 95% CI 1.12–2.09), whereas risks of esophageal, gastric, biliary and small intestinal cancers did not differ significantly. | Lean NAFLD is associated with increased risks of liver, pancreatic and colorectal cancers, supporting tailored prevention strategies and further mechanistic research. |
| Zhang et al., 2023 [23] | The final analysis included 25 studies, comprising 229,091 lean or non-obese Chinese adults, of whom 22,641 had NAFLD. The pooled prevalence of NAFLD was 8.98% (95% CI 5.55–13.13) among lean participants and 13.77% (95% CI 11.13–16.63) among non-obese participants. | Compared to overweight or obese NAFLD, lean or non-obese NAFLD showed a more favorable cardiometabolic profile, including lower blood pressure (128.86/80.48 vs. 136.09/84.98 mmHg), WC (80.63 vs. 92.73 cm), FBG (5.53 vs. 5.69 mmol/L) and triglycerides (1.63 vs. 1.94 mmol/L), with higher HDL cholesterol (1.33 vs. 1.26 mmol/L). | Lean or non-obese NAFLD is generally associated with a more favorable metabolic profile than overweight or obese NAFLD. |
| Huang et al., 2024 [24] | 12 studies (totaling n = 26,329 individuals), 7 of which (n = 7,924) were used to evaluate the risk of all-cause mortality between lean and non-lean NAFLD patients. | Lean NAFLD subjects had a higher risk of death compared to non-lean patients (RR = 1.39, 95% CI 1.08–1.82, heterogeneity: I2 = 43%). Among the lean NAFLD population, all-cause mortality was 13.3 (95% CI: 6.7–26.1) per 1,000 person-years, 3.6 (95% CI: 1.0–11.7) for LRM and 7.7 (95% CI: 6.4–9.2) for CV-related mortality. The incidence of new-onset diabetes was 13.7 (95% CI 8.2–22.7) per 1,000 person-years, new-onset HTN was 56.1 (95% CI: 40.2–77.9), cirrhosis was 2.3 (95% CI: 1.0–5.0), and cancer was 25.7 (95% CI: 20.3–32.4). | Lean NAFLD is associated with higher all-cause mortality than non-lean NAFLD, indicating that BMI alone should not determine follow-up or treatment decisions. |
| Gao et al., 2023 [25] | This metanalysis included a total of 16 observational studies with 304,975 adult individuals (7.7% with lean NAFLD) followed up over a median period of 5.05 years. | Patients with lean NAFLD had a greater risk of incident diabetes than those without NAFLD (random-effects HR 2.72, 95% CI 1.56–4.74; I2 = 93.8%). Compared to the lean without NAFLD group, the aHRs of incident diabetes for those in the overweight/obese without NAFLD and overweight/obese with NAFLD groups were 1.32, CI 0.99–1.77) and 2.98, CI 1.66–5.32. NAFLD patients with high NFS had even greater risks (random-effects HR 3.48, 95% CI 1.92–6.31). | Lean NAFLD is associated with an approximately twofold higher risk of incident diabetes in non-overweight individuals, with risk increasing alongside NAFLD severity. These findings suggest that hepatic steatosis may confer a greater diabetogenic risk than excess body weight alone. |
| Ha et al., 2023 [26] | 10 cohort studies involving 109,151 NAFLD patients were analyzed. | Compared to non-lean NAFLD counterparts, lean NAFLD subjects had comparable risks for all-cause mortality (RR, 1.09, 95% CI, 0.66–1.90), CV mortality (RR, 1.12; 95% CI, 0.66–1.90), and adverse liver events including decompensated cirrhosis and HCC (RR, 0.81; 95% CI, 0.50–1.30). However, the risk of LRM was higher in patients with lean than non-lean NAFLD (RR, 1.88, 95% CI 1.02–3.45). | Lean NAFLD is associated with excess LRM compared with non-lean NAFLD, underscoring the need to define the mechanisms, prognostic factors and genetic or ethnic modifiers that should inform individualized management strategies. |
| Tang et al., 2023 [27] | This meta-analysis included 85 articles. | NAFLD in lean accounted for 13.11%, 95% CI 10.26–16.62% of the global population and 14.55%, 95% CI 11.32–18.51% in Asia. Significantly less metabolic dysfunction was found in lean NAFLD subjects than in NAFLD in overweight counterparts. For lean NAFLD only 19.56% (95% CI, 15.28–24.69%) of the subjects were diabetic, vs. 45.70% (95% CI, 35.01–56.80%) of obese subjects with NAFLD had diabetes (p < 0.01). | These findings emphasize the weight-dependent gradient of metabolic dysfunction in NAFLD: lean individuals are metabolically healthier than overweight or obese counterparts yet remain within the disease spectrum. Definitions that rely too heavily on metabolic dysfunction may therefore overlook established lean NAFLD phenotypes. |
| Alam et al., 2021 [28] | This study processed 22 studies totaling lean (n = 6,768) and obese (n = 9,253) patients with NAFLD were analyzed. Lean (n = 43,398) and obese (n = 9,619) subjects without NAFLD served as controls. | Compared to lean non-NAFLD controls, lean NAFLD was associated with higher adiposity, BP, glycemic indices and insulin resistance, including FBG (MD 5.17 mg/dL, 95% CI 4.14–6.16), HbA1c (MD 0.29%, 95% CI 0.11–0.48) and HOMA-IR (MD 0.49 U, 95% CI 0.29–0.68). Lipid parameters and uric acid were also generally increased, whereas HDL cholesterol was not. Across BMI strata, the most consistent NAFLD-associated differences involved DBP (slope 0.19, p < 0.037), HOMA-IR (slope 0.58, p < 0.001) and uric acid (slope 0.36, p = 0.022). | Lean and non-lean NAFLD appear to share overlapping metabolic features and risk determinants, supporting the view that lean NAFLD lies within the broader spectrum of metabolic liver disease. |
| Ito et al., 2021 [29] | Seventy-three eligible studies were included (n = 258,531). Pooled estimates were calculated using random-effects models, whereas trends and forecasts were evaluated with Bayesian modelling. Individual patient data were obtained for 14,887 participants from eight studies, including 7,752 with NAFLD and 7,135 without NAFLD. | NAFLD prevalence was 25.5% and is projected to reach 39.3% by 2030 and 44.8% by 2040. Incidence rates for NAFLD, HCC and all-cause mortality were 23.5, 7.6 and 5.9 per 1,000 person-years, respectively. Lean NAFLD accounted for 20.7% of NAFLD cases and was associated with higher mortality than non-lean NAFLD (8.3 vs. 5.6 per 1,000 person-years; p = 0.02), although lean status was not an independent mortality predictor. | In Japan, NAFLD prevalence has risen substantially and is projected to approach 45% by 2040. Lean NAFLD accounts for approximately one-fifth of cases and is associated with older age and higher mortality. |
| Lu et al., 2020 [30] | Thirty-three observational studies were included, totaling 205,307 individuals from 14 countries. | The global prevalence of lean NAFLD was 4.1% (95% CI: 3.4–4.8%) and reached 9.7% (95% CI: 7.7–11.8%) among lean individuals. Prevalence increased between 1988 and 2017, was highest in Asia (4.8%, 95% CI: 4.0–5.6%) and in middle-aged adults (45–59 years; 4.4%, 95% CI: 3.2–5.5%), and metabolic complications increased progressively from lean non-NAFLD to lean NAFLD and overweight/obese NAFLD. | Lean NAFLD is a prevalent phenotype associated with metabolic complications, with the highest burden reported among middle-aged individuals in Asian populations. |
| Ye et al., 2020 [31] | Ninety-three studies from 24 countries or regions were included (n = 10,576,383): 84 studies contributed to prevalence estimates, five to incidence analyses and eight to outcome analyses. | Within the NAFLD population, 19.2% (95% CI 15.9–23.0) were lean and 40.8% (36.6–45.1) were non-obese. In the general population, non-obese and lean NAFLD affected 12.1% (9.3–15.6) and 5.1% (3.7–7.0), respectively. Among individuals with non-obese or lean NAFLD, 39.0% had NASH, 29.2% had significant fibrosis and 3.2% had cirrhosis. Key incident outcomes per 1,000 person-years included NAFLD in non-obese individuals (24.6), all-cause mortality (12.1), LRM (4.1), new-onset diabetes (12.6), cardiovascular disease (18.7) and hypertension (56.1). | Non-obese individuals account for approximately 40% of the global NAFLD population, and nearly one-fifth are lean. These phenotypes carry substantial long-term hepatic and extrahepatic morbidity, indicating that obesity alone is an insufficient basis for NAFLD screening and that clinical trials should include participants across the full BMI spectrum. |
| Shi et al., 2019 [32] | Analysis included 45 studies comprising 55,936 lean or non-obese individuals, of whom 7,351 had NAFLD. | Pooled NAFLD prevalence was 10.2% (95% CI: 7.6–13.6%) in lean populations and 15.7% (95% CI: 12.5–19.6%) in non-obese populations, with lower estimates in Eastern than Western studies and an upward trend over time. Compared with non-lean or obese NAFLD, lean or non-obese NAFLD was associated with lower hypertension, uric acid and FBG levels, and higher HDL cholesterol. | Lean and non-obese NAFLD variants are common and clinically relevant across Eastern and Western populations. Better characterization of epidemiology and clinical profile should support more targeted care pathways and intervention strategies. |
| Sookoian et al., 2018 [33] | Eight studies were assessed that compared histological outcomes in lean (n = 493) vs. overweight/obese (n = 2,209) subjects. | Compared to lean NAFLD, overweight or obese NAFLD was associated with higher fibrosis scores (MD 0.28 ± 0.13; p = 0.032), NAFLD activity (0.58 ± 0.16; p = 0.0004) and steatosis (0.23 ± 0.07; p = 0.002). NASH was less frequent in lean NAFLD than in overweight or obese NAFLD (OR 0.58, 95% CI 0.34–0.97; 322 vs. 1,357 cases; p = 0.04). | Lean NAFLD is generally associated with less severe histological features than overweight or obese NAFLD. Longitudinal data are needed to define the clinical implications, although the ~25% higher mean fibrosis score in overweight or obese NAFLD suggests a potentially worse long-term prognosis. |
| Sookoian et al., 2017 [34] | A meta-analysis of 15 studies was performed together with data from the present cohort (n = 336 patients). Data from lean (n = 1,966) and obese (n = 5,938) patients with NAFLD were analyzed; lean (n = 9,946) and obese (n = 6,027) individuals without NAFLD served as controls. | Compared to lean non-NAFLD controls, lean NAFLD was associated with older age (3.79 ± 0.72 years, p = 1.36 × 10–6), higher FBG (6.44 ± 1.12 mg/dL, p = 10–10), HOMA-IR (0.52 ± 0.094-unit increment), triglycerides (48.37 ± 3.6, p = 10–10), BP and WC (5.88 ± 0.4 cm, p = 10–10). Alterations were greater in obese individuals, and meta-regression suggested that NAFLD modifies blood lipid levels. | Lean NAFLD shares metabolic and CV abnormalities with obesity-associated NAFLD, despite normal body weight. Excess abdominal adiposity and other features of metabolic syndrome remain central to this phenotype. |
aHR: adjusted hazard ratio; BMI: body mass index; BP: blood pressure; CI: confidence interval; CV: cardiovascular; FBG: fasting blood glucose; GI: gastrointestinal; HTN: arterial hypertension; HbA1c: glycated hemoglobin; HCC: hepatocellular carcinoma; HDL: high density lipoprotein; HOMA-IR: homeostasis model assessment of insulin resistance; HR: hazard ratio; MD: mean difference; NAFLD: nonalcoholic fatty liver disease; NFS: NAFLD fibrosis score; LRM: liver-related mortality; WC: waist circumference.