Determining the sensitivity and specificity of the calculated fatty liver index in comparison with ultrasound Authors: Arash Saberian, Azizallah Dehghan, Reza Homayounfar, Saeid Kaffashan, Fariba Zarei, Sepideh Niknejad, Mojtaba Farjam Published: 2024 ABSTRACT BackgroundNon-alcoholic fatty liver disease (NAFLD) is a common chronic liver disease in human history and it is expected to surpass other causes of liver disease mortality by 2030. Therefore, finding an alternative way to diagnose steatosis in the early stage when imaging modalities are not available is crucial. This study decided to validate the optimal cut-off points and the sensitivity and specificity of the Fatty Liver Index (FLI) based on the Iranian population compared to ultrasonography.MethodsThe data of 367 individuals, 108 males and 259 females over 35, were analyzed. Hepatic steatosis was identified by ultrasound. FLI was determined from waist circumference, gamma-glutamyl transferase, triglyceride, and body mass index data. The receiver operating characteristic curve (ROC) was used to determine the best FLI index cut point for diagnosing nonalcoholic fatty liver. The sensitivity and specificity indices were calculated for the determined cut point.ResultsThe AUC of the FLI index in diagnosing NAFLD in the total population was 0.733 (95% CI: 0.68–0.77, specificity = 0.6705, sensitivity = 0.7320) with the optimal COP of 40.6. There was a statistically significant association between non-alcoholic liver disease and FLI-based ultrasound (p < 0.0001). Furthermore, the sex-specific optimal COPs of FLI was 33.4, specificity = 0.6071, sensitivity = 0.8462 in men vs. 27.8, sensitivity = 0.8233, specificity = 0.7655 in women.ConclusionFLI is a reliable tool for identifying individuals with NAFLD. It has the potential to aid in detecting and managing this condition in large-scale populations while other methods are not available. We also determine an optimal COP of 40.6 with sensitivity and specificity of 73.20% and 67.05% in the general population, respectively. Background Non-alcoholic fatty liver disease (NAFLD) is a common chronic liver disease in human history, characterized by the excessive accumulation of lipid granules in liver hepatocytes (steatosis) in the absence of other etiologies, such as consumption Excessive alcohol intake or using hepatotoxic drugs [1–5]. (More than a daily intake of 30 g for men and 20 g for women) Although NAFLD is often asymptomatic, it includes a spectrum ranging from non-alcoholic steatohepatitis (NASH), steatohepatitis, liver tissue scarring (fibrosis), and hepatocellular carcinoma [6–9]. With the significantly increasing prevalence of metabolic syndrome-related disorders, like type ΙΙ diabetes, which are NAFLD’s leading causes, it is expected to surpass other causes of liver disease mortality by 2030 [10–12]. About 33% of people worldwide have NAFLD, with rates increasing from 25 to 38% in the last thirty years [13]. Yet, it is poorly understood by global health communities and the general population [14]. While liver biopsy is considered the gold standard technique for diagnosing hepatic steatosis, it is invasive and may result in clinical complications [15, 16]. Some non-invasive imaging techniques have been proposed as alternative methods to liver biopsy, which include transient elastography, magnetic resonance imaging (MRI), and computed tomography. However, these can be expensive and not readily available [6]. Conversely, ultrasound is a convenient and safe imaging method that can be performed easily, even on conscious patients.; it is broadly available and relatively cheap. Therefore, ultrasound is the primary diagnostic method for detecting NAFLD in most cases [17–19]. Early diagnosis of NAFLD is necessary to prevent its progression, and finding an alternative way to diagnose steatosis when imaging modalities are unavailable is crucial [20]. by using simple clinical and laboratory parameters such as waist circumference, triglycerides, gamma-glutamyl transferase, and body mass index to calculate Fatty Liver Index (FLI), it is a convenient index for screening and identifying high-risk patients for NAFLD [21, 22]. It has become a popular screening tool for NAFLD due to its simplicity and cost-effectiveness [20, 23]. However, optimal cut-off points for FLI may vary across different populations, including the Iranian population [24]. Therefore, due to the variation among population lifestyles and the differing cut-off points (COPs) for FLI parameters, this study aimed to identify the optimal cut-off points, as well as the specificity and sensitivity of the Fatty Liver Index based on the Iranian population compared to ultrasonography. Methods Participants This cross-sectional study gathered data from the Fasa Cohort Study as a branch of the PERSIAN cohort, focusing on non-communicable diseases [25]. Informed consent was obtained from all subjects. Legal guardians were involved to ensure that consent was appropriately managed for any individuals who may have cognitive impairments or other conditions that could affect their ability to provide informed consent independently. Age over 35, no alcohol consumption, no history of congenital hepatic diseases, B, C, or autoimmune hepatitis were considered as inclusion criteria. The individuals who underwent abdominal surgery within the past six months and those with a history of drug use that could result in liver steatosis, like consumption of corticosteroids and valproate sodium, were excluded. Anthropometric and biochemical assessment Demographic data and the participants’ status of alcohol consumption, physical activity, smoking, and medical history, including blood hypertension, cardiovascular disease, diabetes, gastrointestinal disease, and stroke, were recorded in the questionnaire. Weight and height were measured and recorded by a trained healthcare worker (Behvarz). Blood pressure was taken two times, with a 15-minute gap between each measurement. After twelve hours of fasting, venous blood samples were drawn to analyze serum lipid profiles, aspartate aminotransferase, alanine transaminase, Gamma-glutamyl transpeptidase, and fasting blood sugar. Samples were analyzed at the Noncommunicable Diseases Research Center (NCDRC) laboratory. Physical activity was evaluated by the answers given in questionnaires. More information is explained in the cohort protocol [25, 26]. NAFLD diagnosis Ultrasonography Ultrasonography (US) was performed with a Samsung WS80A ultrasound machine by a trained radiologist at the ultrasound center of Valiasr Hospital in Fasa. To ensure accuracy and reliability, the US films of the patients were recorded and subsequently double-checked by another experienced radiologist. In cases where discrepancies between the radiologists’ interpretations of ultrasound results arise, the opinion of a senior radiologist would typically be accepted to resolve any differences and establish a conclusion. Based on the echogenicity of the liver parenchyma and the comparison with the echogenicity of the renal cortex (for patients with parenchymal renal disease, the liver parenchyma was compared with the spleen), participants were separated into two groups: without NAFLD and with NAFLD [27]. Fatty liver index FLI was determined based on the following calculation: FLI = [ 0.139 × BMI + e0.953 × ln (TG) + 0.718 × ln (GGT) + 0.053 × WC − 15.745 / (1 + e0.953 × ln (TG) + 0.718 × ln (GGT) + 0.139 × BMI + 0.053 × WC − 15.745)] × 100. The cut-off points (COPs) of the FLI used in this study were based on previous literature, specifically at values of 30 and 60, as established in earlier research [20, 22]. Statistical analysis All statistical analyses were performed using MedCalc20.0.26. Results were described as mean ± standard deviations for quantitative data and number and percentage for qualitative data. T-tests were used to compare means between groups, while chi-square tests were applied to evaluate associations between categorical variables. Receiver operating characteristic curve (ROC) analysis was employed to determine the optimal FLI cut-off point for diagnosing nonalcoholic fatty liver disease. The sensitivity and specificity indices were calculated for the determined cut point. Discrimination was assessed using the C-statistic (Area Under the Curve, AUC). Calibration was also evaluated through calibration plots, although specific metrics were not reported because the sample size was insufficient to provide reliable estimates. Confidence intervals (CIs) for sensitivity and specificity were calculated and reported. A p-value < 0.05 was considered significant in the analysis. Result This cross-sectional study evaluated 367 individuals, 108 males, and 259 females, with a mean age of 49.3 ± 8.94 years. The mean and standard deviation of BMI and physical activity index were 26.95 ± 4.60 kg/m2 and 40.66 ± 9.30 weekly MET (Metabolic Equivalent of Task)-minutes, respectively. Demographic and clinical variables of nonalcoholic fatty liver disease and non-NAFLD diagnosed in the US are reported in Table 1. A total of 194 people (52.9%) had NAFLD. T-test analysis revealed that age (P = 0.007) and physical activity index (P = 0.034) are associated with NAFLD. Moreover, our study showed a significant statistical association between diabetes mellitus (DM) and nonalcoholic fatty liver disease based on chi-square (p = 0.018) (Table 1). Table 1General Characteristics of Ultrasound Diagnosed NAFLD and Non-NAFLDVariablesNAFLD, N (%)Non-NAFLD, N (%)P ValueSex0.253Male52 (26.8)56 (32.4)Female142 (73.2)117 (67.6)DM0.018Yes42 (21.6)21 (12.1)No152 (78.4)152 (87.9)HTN0.251Yes46 (23.7)32 (18.5)No148 (76.3)141 (81.5)Smoking0.377Yes25 (12.9)28 (16.2)No169 (87.1)145 (83.8) Mean (SD) Mean (SD) Age48.68 (8.16)50 (9.72)0.007Physical activity index39.58 (10.04)41.88 (8.48)0.034FLI56.13 (25.62)33.71(25.37)< 0.001BMI28.75 (4.37)24.94 (3.99)< 0.001TG152.47(74.39)129.01(89.61)< 0.001GGT28.76(28.14)21.42(19.83)< 0.001WC100.79(9.86)91.84(10.68)< 0.001Abbreviations: BMI, body mass index; DM, diabetes mellitus; FLI: Fatty Liver Index ⁏ GGT: gamma-glutamyl transferase⁏ HTN, hypertension; N, number; NAFLD, non-alcoholic fatty liver disease; SD, standard deviation; TG: triglycerides⁏ US, ultrasonography⁏ WC: waist circumference. Statistical tests: Chi-square test, independent t-test Table 1General Characteristics of Ultrasound Diagnosed NAFLD and Non-NAFLDVariablesNAFLD, N (%)Non-NAFLD, N (%)P ValueSex0.253Male52 (26.8)56 (32.4)Female142 (73.2)117 (67.6)DM0.018Yes42 (21.6)21 (12.1)No152 (78.4)152 (87.9)HTN0.251Yes46 (23.7)32 (18.5)No148 (76.3)141 (81.5)Smoking0.377Yes25 (12.9)28 (16.2)No169 (87.1)145 (83.8) Mean (SD) Mean (SD) Age48.68 (8.16)50 (9.72)0.007Physical activity index39.58 (10.04)41.88 (8.48)0.034FLI56.13 (25.62)33.71(25.37)< 0.001BMI28.75 (4.37)24.94 (3.99)< 0.001TG152.47(74.39)129.01(89.61)< 0.001GGT28.76(28.14)21.42(19.83)< 0.001WC100.79(9.86)91.84(10.68)< 0.001Abbreviations: BMI, body mass index; DM, diabetes mellitus; FLI: Fatty Liver Index ⁏ GGT: gamma-glutamyl transferase⁏ HTN, hypertension; N, number; NAFLD, non-alcoholic fatty liver disease; SD, standard deviation; TG: triglycerides⁏ US, ultrasonography⁏ WC: waist circumference. Statistical tests: Chi-square test, independent t-test General Characteristics of Ultrasound Diagnosed NAFLD and Non-NAFLD Abbreviations: BMI, body mass index; DM, diabetes mellitus; FLI: Fatty Liver Index ⁏ GGT: gamma-glutamyl transferase⁏ HTN, hypertension; N, number; NAFLD, non-alcoholic fatty liver disease; SD, standard deviation; TG: triglycerides⁏ US, ultrasonography⁏ WC: waist circumference. Statistical tests: Chi-square test, independent t-test Table 2 shows the frequency of NAFLD based on FLI’s different cut-off points and its relations between the study’s variables. Physical activity index and HTN are related to NAFLD in cut-off points 30 and 60 (FLI ≥ 30: p < 0.01, p = 0.017. FLI ≥ 60: p = 0.003, p = 0.014). DM is related to the NAFLD in the FLI cut-off point ≥ 60 (p = 0.014) (Table 2). Table 2Association offattyyliverrindexx and variables with and without NAFLDVariableFLI ≥ 30FLI ≥ 60NAFLD (Yes / No)Mean (SD)tP ValueNAFLD (Yes / No)Mean (SD)tP ValueAge (years)-1.140.255-1.5660.434Yes49.70(8.55)Yes50.30 (8.87)No48.59(9.58)No48.77 (8.95)BMI (kg/m2)-17.5470.000-16.3470.078Yes29.30(3.83)Yes31.04 (4.03)No22.82(2.44)No24.76 (3.18)Physical Activity Index (weekly MET-minutes)3.5950.0002.8760.003Yes39.37(8.15)Yes38.77 (7.27)No42.94(10.71)No41.68 (10.10)NAFLD, N (%)Non-NAFLD, N (%)P ValueNAFLD, N (%)Non-NAFLD, N (%)P ValueSex0.8131.000Male70 (29.9)38 (28.6)38 (29.7)70 (29.3)Female164 (70.1)95 (71.4)90 (70.3)169 (70.7)DM0.1130.014Yes46 (19.7)17 (12.8)24 (18.8)200 (83.7)No188. (80.3)116 (87.2)39 (16.3)104 (81.3)HTN0.0170.014Yes59 (25.2)114 (85.7)36 (28.1)42 (17.6)No19 (14.3)175 (74.8)92 (71.9)197 (82.4)Smoking1.0000.372Yes200 (85.5)114 (85.7)108 (84.4)206 (86.2)No34 (14.5)19 (14.3)20 (15.6)33 (13.8)Abbreviations: BMI, body mass index; DM, diabetes mellitus; FLI, Fatty Liver Index; HTN, hypertension; MET: Metabolic Equivalent of Task; N, number; NAFLD, non-alcoholic fatty liver disease; SD, standard deviation Table 2Association offattyyliverrindexx and variables with and without NAFLDVariableFLI ≥ 30FLI ≥ 60NAFLD (Yes / No)Mean (SD)tP ValueNAFLD (Yes / No)Mean (SD)tP ValueAge (years)-1.140.255-1.5660.434Yes49.70(8.55)Yes50.30 (8.87)No48.59(9.58)No48.77 (8.95)BMI (kg/m2)-17.5470.000-16.3470.078Yes29.30(3.83)Yes31.04 (4.03)No22.82(2.44)No24.76 (3.18)Physical Activity Index (weekly MET-minutes)3.5950.0002.8760.003Yes39.37(8.15)Yes38.77 (7.27)No42.94(10.71)No41.68 (10.10)NAFLD, N (%)Non-NAFLD, N (%)P ValueNAFLD, N (%)Non-NAFLD, N (%)P ValueSex0.8131.000Male70 (29.9)38 (28.6)38 (29.7)70 (29.3)Female164 (70.1)95 (71.4)90 (70.3)169 (70.7)DM0.1130.014Yes46 (19.7)17 (12.8)24 (18.8)200 (83.7)No188. (80.3)116 (87.2)39 (16.3)104 (81.3)HTN0.0170.014Yes59 (25.2)114 (85.7)36 (28.1)42 (17.6)No19 (14.3)175 (74.8)92 (71.9)197 (82.4)Smoking1.0000.372Yes200 (85.5)114 (85.7)108 (84.4)206 (86.2)No34 (14.5)19 (14.3)20 (15.6)33 (13.8)Abbreviations: BMI, body mass index; DM, diabetes mellitus; FLI, Fatty Liver Index; HTN, hypertension; MET: Metabolic Equivalent of Task; N, number; NAFLD, non-alcoholic fatty liver disease; SD, standard deviation Association offattyyliverrindexx and variables with and without NAFLD Abbreviations: BMI, body mass index; DM, diabetes mellitus; FLI, Fatty Liver Index; HTN, hypertension; MET: Metabolic Equivalent of Task; N, number; NAFLD, non-alcoholic fatty liver disease; SD, standard deviation The AUC of the FLI index in the diagnosis of nonalcoholic fatty liver disease in the total population was 0.733 (95% CI: 0.68–0.77, specificity = 0.6705, sensitivity = 0.7320, 95% CI for sensitivity = [0.67–0.76], 95% CI for specificity = [0.61–0.71]) and the optimal cut-off is 40.6. Analysis showed that there was a statistically significant association between the non-alcoholic fatty liver disease and FLI based on ultrasound (p < 0.0001) (Table 3) (Fig. 1). Furthermore, the sex-specific optimal COP of FLI was 33.4 in men vs. 27.8 in women. The AUC of FLI in males was 0.75 (95% CI: 0.66–0.83, specificity = 0.6071, sensitivity = 0.8462, 95% CI for sensitivity = [0.75–0.89], 95% CI for specificity = [0.50–0.69]) (p < 0.0001) while in female was 0.72 (95% CI: 0.66–0.78, sensitivity = 0.8233, specificity = 0.7655, 95% CI for sensitivity = [0.79–0.88], 95% CI for specificity = [0.47–0.60]) (p < 0.0001) (Table 3) (Fig. 1). Table 3The cut-off points of fatty liver index based on the results of liver USVariableNAFLD, N (%)AUCSensitivity (95% CI)Specificity (95% CI)PVTotal n = (367)0.7330.73 (0.67–0.76)0.67(0.61–0.71)< 0.0001Yes194 (52.86)No173 (47.14)Male n = (108)0.7550.84(0.75–0.89)0.60(0.50–0.69)< 0.0001Yes52 (48.15)No56 (51.58)Female n = (259)0.7270.84(0.79–0.88)0.54(0.47–0.60)< 0.0001Yes142 (54.83)No117 (45.17)Abbreviation: AUC: area under the curve, CI: confidence interval, NAFLD: non-alcoholic fatty liver disease Table 3The cut-off points of fatty liver index based on the results of liver USVariableNAFLD, N (%)AUCSensitivity (95% CI)Specificity (95% CI)PVTotal n = (367)0.7330.73 (0.67–0.76)0.67(0.61–0.71)< 0.0001Yes194 (52.86)No173 (47.14)Male n = (108)0.7550.84(0.75–0.89)0.60(0.50–0.69)< 0.0001Yes52 (48.15)No56 (51.58)Female n = (259)0.7270.84(0.79–0.88)0.54(0.47–0.60)< 0.0001Yes142 (54.83)No117 (45.17)Abbreviation: AUC: area under the curve, CI: confidence interval, NAFLD: non-alcoholic fatty liver disease The cut-off points of fatty liver index based on the results of liver US Abbreviation: AUC: area under the curve, CI: confidence interval, NAFLD: non-alcoholic fatty liver disease Fig. 1Receiver Operating Characteristic Curves (ROC) of FLI with a cut-off point of A: 40.6 in total population. B: 27.8 in females. C: 33.4 in males Receiver Operating Characteristic Curves (ROC) of FLI with a cut-off point of A: 40.6 in total population. B: 27.8 in females. C: 33.4 in males The area under the curve was calculated for the predictive model. Sensitivity and specificity were higher in males than in females. The AUC values indicated consistent performance across sexes, illustrating the model’s robustness in various populations. Discussion Nonalcoholic fatty liver disease (NAFLD) serves as an important indicator of metabolic syndrome and is associated with various metabolic-related conditions, including cardiovascular disease (CVD), diabetes mellitus, renal disease, thyroid dysfunction, polycystic ovarian syndrome (PCOS), and colorectal cancer [15, 28]. Given the increasing prevalence of NAFLD among individuals with these conditions, establishing an accessible, non-invasive, and cost-effective screening method is essential [29, 30]. Liver biopsy is an invasive procedure that may result in clinical complications and MRI, while accurate for detecting steatosis, is costly and not widely available [31]. Transient elastography is a straightforward, quick, and non-invasive method that can predict NAFLD in lean patients, but its accuracy in diagnosing NAFLD in obese patients may be limited [32]. Ultrasonography is widely available and relatively cheap, and it has become a standard method used for diagnosing hepatic steatosis based on increasing echogenicity of the liver by fat accumulation in hepatocytes. However, it is operator-dependent [33, 34]. A study by Irene Cantero et al. found that in the absence of MRI and biopsy, ultrasound (with ROC-AUC: 0.746) demonstrated the highest correlation with these methods [35]. The Fatty Liver Index is also a helpful method that can be easily calculated in a medical setting. Research has shown that FLI is strongly correlated with NAFLD diagnosed by ultrasonography [24]. FLI can be utilized for screening individuals with fatty liver disease and for identifying those at high risk for metabolic and cardiovascular disorders [36]. However, it is crucial to consider potential variations in waist and BMI cut-offs due to factors such as ethnicity, diet, and the environment [37]. Therefore, validation of FLI is necessary when implementing it in diverse populations. Previous studies based on Western populations showed acceptable accuracy of cut-off points for FLI, with an AUC of 0.81–0.84. These studies suggested that an FLI below 30 effectively rules out hepatic steatosis with a sensitivity of 87%, while an FLI above 60 predicts the condition with a specificity of 86% [20, 22]. Juan Wu also in the American population compared both transient elastography and US with FLI and proposed a value of 45.60 and 59.54 for the optimal COP of FLI, AUC of 0.833 and 0.681, specificity of 70.50%, and 75.15%, sensitivity of 80.85% and 55.53%, respectively [38]. In an Asian population, similar values yielded an acceptable AUC of 0.87 [39]. In the Iranian population, Dehnavi et al. validated 26.2 as the optimal cut-off point based on a controlled attenuation parameter technique with an AUC of 0.85, sensitivity of 0.83, and specificity of 0.7 [40]. In the present study, we compared the use of abdominal ultrasonography, finding the optimal COP of FLI to be 40.6, with sensitivity and specificity of 73.20% and 67.05%. The variations in optimal cut-off points, sensitivity, and specificity of FLI across different studies may stem from differences in diagnostic methods, study populations, sample sizes, and ethnic backgrounds. Therefore, further research is needed to establish standardized guidelines for FLI interpretation, taking into account the diverse characteristics of populations and settings. The findings regarding Body Mass Index and diabetes mellitus indicate that a lower cut-off point of 30 is more sensitive for detecting early stages of liver fat accumulation, as it shows a significant difference in BMI between NAFLD and non-NAFLD. In contrast, no such difference is observed at a cut-off of 60. Additionally, it is important to consider the effective sample size of individuals with liver steatosis at FLI values of 30 and 60, as this can significantly impact the robustness and reliability of the results. A larger sample size at lower thresholds may provide more reliable insights into the association with NAFLD. This suggests that BMI may be a more effective marker for identifying at-risk individuals when using lower thresholds. Similarly, the relationship between DM and NAFLD becomes less pronounced at higher FLI cut-off points, indicating that metabolic implications are clearer at lower thresholds. Notably, using a cut-off point of 30 yielded a prevalence of NAFLD of 63.8% (234 individuals), compared to 34.9% (128 individuals) at a cut-off of 60, emphasizing the impact of cut-off selection on diagnostic results and highlighting the potential for overdiagnosis at lower thresholds, which could affect clinical practice and patient management. The studies of Bi-Ling Yang et al. [41] and Dehnavi et al. [40] showed that the optimal COPs of the FLI are higher in males than females. In line with these findings, in the current study, we calculated the optimal COPs of Fatty Liver Index 33.4 and 27.8 for males and females, respectively. The sensitivity and specificity of FLI were higher in males than in females, with sensitivity exceeding 82% and specificity exceeding 60% for both genders. Inconsistent with our findings, in a study conducted in 2016 on 5052 subjects, the cut-off value in females was suggested higher than in males [23]. Resolving these conflicting results is challenging; however, the differences in optimal cut-off points may partly be explained by the observation that males generally have a higher body mass index and more severe metabolic disturbances, contributing to the development of NAFLD. As males age, they experience an increase in visceral fat, leading to fatty liver disease and insulin resistance due to the release of adipocytokines and free fatty acids [41]. In women Estrogen plays a role in suppressing the accumulation of visceral fat and triglycerides though they have an increased risk of being obese and metabolic syndrome after menopause, suggesting a potential protective effect of estrogen in preventing the onset of fatty liver disease [42, 43]. Based on our findings, if FLI is considered more than 30, there was no significant relationship between NAFLD and the independent variables of gender, diabetes, and smoking. Body mass index, physical activity index, and hypertension (HTN) were significantly related to NAFLD. Moreover, with a cut-off point of more than 60 for FLI, no statistically significant relationship was observed between NAFLD and gender, DM, and smoking; however, a significant correlation was found between physical activity index and increased blood pressure with NAFLD. A study by Bi-Ling Yang involving 23,797 participants indicated positive relations between age, gender, BMI, fasting blood sugar, blood pressure, and FLI [44]. Variations in sample size, population characteristics, and different cut-off points for FLI may account for discrepancies in results. Acknowledging that relying on abdominal ultrasonography as a comparison technique for diagnosing NAFLD may introduce potential limitations is essential. Ultrasound results can be influenced by the experience of the radiologist and may be subject to interpretation variability. Future studies should consider incorporating additional diagnostic techniques to enhance the accuracy and reliability of NAFLD diagnosis. In the present study, we focused on the area under the curve for evaluating the effectiveness of the FLI. This approach provides a more comprehensive understanding of diagnostic performance across different populations. Differences in AUC values across studies can be attributed to various factors, including sample size, population characteristics, and diagnostic methods. Despite limitations related to sample variation and diagnostic methods, findings from comparative studies support the efficiency of FLI in predicting NAFLD. Therefore, we recommend the Fatty Liver Index as a reliable method for detecting NAFLD and suggest determining optimal cut-off points based on population characteristics. Conclusion The Fatty Liver Index is a reliable tool for identifying individuals with NAFLD. It has the potential to aid in the detection and management of this condition in large-scale populations where other methods may not be available. We determined an optimal cut-off point of 40.6, with sensitivity and specificity of 73.20% and 67.05% in the general population, respectively.