USE OF THE FIBROSIS-4 INDEX FOR SCREENING LIVER FIBROSIS IN PEOPLE LIVING WITH HUMAN IMMUNODEFICIENCY VIRUS AND GLUCOSE METABOLISM DISORDERSABSTRATC
Background and Objective: To evaluate the performance of the Fibrosis-4 (FIB-4) index for liver fibrosis screening in people living with HIV (PLWH) with glucose metabolism disorders (diabetes and prediabetes).
Methods: Cross-sectional study including 100 PLWH on continuous antiretroviral therapy at a reference center in Brazil (March-November 2025). Clinical, laboratory, and ultrasonographic data were collected for FIB-4 calculation. Participants with diabetes were compared to non-diabetic individuals.
Results: Altered FIB-4 values were observed in 32 participants, with no sex differences. Mild steatosis (S≥1) was present in 47.0%, and clinically significant steatosis (S≥2) in 11.0%. S≥1 was associated with higher BMI, waist circumference, visceral fat, total cholesterol, LDL, triglycerides, and ALT levels. FIB-4 showed limited ability to classify S≥2.
Conclusion: FIB-4 is useful for fibrosis screening, particularly to exclude advanced disease, but does not adequately identify clinically significant steatosis (S≥2). Its use should be combined with metabolic assessment and elastography when values are ≥1.3.
Keywords:
HIV; non-alcoholic fatty liver disease; diabetes mellitus
HIGHLIGHTS
FIB-4 is effective as a first-line noninvasive tool to screen liver fibrosis, particularly to rule out advanced disease due to its high negative predictive value.
FIB-4 shows poor performance for detecting clinically significant steatosis (S≥2), with low sensitivity at standard cutoffs.
32% had elevated FIB-4; steatosis prevalence was 47% (S≥1) and 11% (S≥2), with no sex differences.
FIB-4 ≥1.3 should prompt further evaluation with elastography and metabolic assessment, as it is insufficient as a standalone tool for steatosis detection.
RESUMO
Contexto e Objetivo: Avaliar o risco de fibrose hepática pelo índice Fibrosis-4 (FIB-4) em pessoas vivendo com HIV (PVHIV) com distúrbios do metabolismo glicídico (pré-diabetes e diabetes).
Métodos: Estudo transversal com 100 PVHIV em terapia antirretroviral contínua, atendidas em centro de referência em Belém, Brasil (março-novembro de 2025). Foram coletados dados clínicos, laboratoriais e ultrassonográficos para cálculo do FIB-4. Participantes com diabetes foram comparados aos não diabéticos.
Resultados: Valores alterados de FIB-4 foram observados em 32 participantes, sem diferença entre os sexos. Esteatose leve (S≥1) ocorreu em 47,0% e esteatose clinicamente significativa (S≥2) em 11,0%. S≥1 associou-se a maiores valores de IMC, circunferência abdominal, gordura visceral, colesterol total, LDL, triglicerídeos e ALT. O FIB-4 apresentou capacidade limitada para classificar S≥2.
Conclusão: O FIB-4 é útil para triagem de fibrose hepática, especialmente para excluir doença avançada, mas não identifica adequadamente esteatose clinicamente significativa (S≥2). Deve ser associado à avaliação metabólica e à elastografia quando ≥1,3.
Palavras-chave:
HIV; doença hepática gordurosa não alcoólica; diabetes mellitus
INTRODUCTION
With the increased life expectancy of people living with HIV (PLWH), it has become evident that chronic HIV infection predisposes individuals to a persistent inflammatory state associated with disturbances in glucose metabolism and insulin resistance1. In this context, the liver plays a central role. Metabolic dysfunction-associated steatotic liver disease has become particularly prevalent, ranging from simple steatosis to steatohepatitis with fibrosis, and is driven by obesity, type 2 diabetes, hypertension, and dyslipidemia2,3.
Biopsy remains the gold standard for assessing inflammation and fibrosis, but routine clinical practice requires feasible noninvasive methods. Elastography and clinical-laboratory scores have gained prominence, such as the Fibrosis-4 (FIB-4) index4. The FIB-4 combines age, aspartate aminotransferase (AST), alanine aminotransferase (ALT), and platelet count, showing acceptable performance for screening significant fibrosis5. Recent guidelines recommend its use as a first-line approach, particularly in adults with prediabetes or diabetes; values ≥1.3 typically prompt transient elastography, a strategy that correlates with histological findings and helps reduce unnecessary biopsies6. In individuals with diabetes, higher FIB-4 scores have been associated with overall mortality, cardiovascular events, carotid intima-media thickness, and chronic kidney disease7,8. Current international guidelines, including those from the European Association for the Study of the Liver (EASL) and the American Association for the Study of Liver Diseases (AASLD), recommend the use of noninvasive scores such as the FIB-4 index as a first-line screening tool for liver fibrosis, followed by additional assessment in indeterminate or high-risk cases9,10.
In this context, the present study aims to assess the risk of liver fibrosis in PLWH exhibiting altered glucose metabolism disorders, using the FIB-4 index. Specifically, it seeks to estimate the prevalence of fibrosis risk in this population, characterize the frequency of prediabetes and diabetes, and identify factors associated with elevated FIB-4 scores, focusing on anthropometric, metabolic, and ultrasonographic indicators of steatosis.
METHODS
This analytical cross-sectional observational study was conducted at the Tropical Medicine Center in Belém, Brazil, between March and November 2025. The target population comprised PLWH receiving regular follow-up at the endocrinology outpatient clinic. A convenience sample of 100 participants was included. The study was initiated following informed consent and approval by the Research Ethics Committee of the Institute of Health Sciences, Núcleo de Medicina Tropical-NMT/Universidade Federal do Pará - UFPA (approval number 7.423.664; CAAE: 85431324.2.0000.5172).
Eligibility criteria were age ≥18 years, continuous use of antiretroviral therapy (ART) for at least 12 months, and willingness to participate in the study. Individuals using chronic corticosteroids or with active opportunistic infections were excluded. The minimum sample size was estimated based on a hepatic fibrosis prevalence of 8.3% with a 95% confidence interval and a maximum error of 5%, resulting in a minimum of 55 participants11. The study followed the STROBE guidelines for observational research12.
Data were obtained from medical records and routine laboratory and imaging exams. AST, ALT, platelet count, and abdominal ultrasonography with visceral fat measurement (distance between the inner rectus abdominis and anterior aortic wall) performed within six months prior to inclusion were recorded; older or missing exams were repeated. FIB-4 was calculated as (Age×AST) / [Platelets×√(ALT)], with platelets in 109/L. Additional variables included sex, age, ART type and duration, fasting glucose, HbA1c, total cholesterol, LDL, HDL, triglycerides, Body mass index (BMI), and waist circumference. Participants were grouped by type 2 diabetes mellitus status. Data were stored in Microsoft® Excel 2010 and analyzed using BioEstat® 5.3 and Epi Info™ 7.0. Continuous variables were expressed as mean ± SD or median ± IQR; categorical variables as absolute and relative frequencies. 95% CIs were calculated. Normality was assessed by D’Agostino-Pearson test; comparisons used chi-square, Fisher’s exact, t-test, or Mann-Whitney U test, with P<0.05.
To evaluate the accuracy of FIB-4 for clinically significant steatosis (S≥2; grade ≥2) on ultrasonography, 2×2 contingency tables were constructed using prespecified cutoffs. For participants younger than 65 years, a FIB-4 cutoff of 1.3 was applied. For those aged 65 years or older, a higher cutoff of 2.0 was used, in accordance with established age-adjusted criteria. These thresholds were selected to improve diagnostic accuracy and minimize misclassification of fibrosis severity. Confusion matrices were reported, including true positives (TP), false positives (FP), true negatives (TN), and false negatives (FN), and sensitivity and specificity with 95%CI were derived using the Wilson method. Analyses were conducted for the overall sample and stratified by glycemic status-normoglycemia versus elevated glycemia (prediabetes + diabetes)-according to the American Diabetes Association (ADA) criteria: diabetes (fasting glucose ≥126 mg/dL and/or HbA1c ≥6.5%) and prediabetes (100-125 mg/dL and/or HbA1c 5.7-6.4%)13.
RESULTS
A total of 100 people living with HIV receiving care at the Tropical Medicine Center were included, with an equal distribution between females (n=50) and males (n=50), as shown in Table 1.
The overall median FIB-4 value was 1.25 (IQR:0.65). Stratified by sex, the median was 1.19 (IQR:0.53) in women and 1.26 (IQR:0.85) in men, with no statistically significant difference between groups (Mann-Whitney U test, P=0.072).
Using an age cutoff of 65 years, the prevalence of altered FIB-4 was 29.9% (23/77) in participants <65 years and 39.1% (9/23) in those ≥65 years (chi-square test, p=0.403). Regarding sex, 38 women (76.0%) had normal FIB-4 and 12 (24.0%) had altered FIB-4, whereas 30 men (60.0%) had normal FIB-4 and 20 (40.0%) had altered FIB-4, with no statistically significant association (chi-square test, P=0.086). Table 2 presents the distribution of FIB-4 across different glycemic profiles.
A significant negative correlation was found between FIB-4 and estimated glomerular filtration rate (eGFR) (r= −0.41; P<0.001; pearson), indicating that higher FIB-4 values were associated with reduced renal function. Participants with elevated FIB-4 were older (61.0±10.1 vs 56.3±11.1 years; P=0.043) and showed lower eGFR (66.4±25.2 vs 76.6±21.1 mL/min/1.73m²; P=0.038), higher AST (34.0±18.7 vs 23.0±8.8 U/L; P<0.001), higher ALT (31.2±34.7 vs 23.1±17.2 U/L; P=0.012), and lower platelet counts (248.5±201.6 vs 332.0±250.8 ×10³/µL; P=0.008). BMI (26.1±6.8 vs 26.9±5.9 kg/m²; P=0.790), waist circumference (93.0±18.8 vs 91.5±15.0 cm; P=0.625), visceral fat (3.8±3.3 vs 4.1±1.6 cm; P=0.877), fasting glucose, HbA1c, total cholesterol, LDL, HDL, and triglycerides did not differ between groups. Consistently, no significant correlations were observed between FIB-4 and fasting glucose (r= −0.08; P=0.449; spearman), HbA1c (r= −0.02; P=0.823; spearman), total cholesterol (r= −0.13; P=0.208; spearman), LDL (r= −0.12; P=0.233; spearman), HDL (r=0.06; P=0.563; pearson), or triglycerides (r= −0.13; P=0.189; spearman).
Among the 100 patients evaluated by ultrasonography, the prevalence of mild hepatic steatosis (S≥1) was 47.0% (47; 95%CI 37.5-56.7). Stratified by FIB-4, S≥1 was observed in 45.6% of participants with normal FIB-4 (31/68) and 50.0% of those with altered FIB-4 (16/32), with no significant association (chi-square test, P=0.843; OR=0.84; 95%CI 0.36-1.94). Clinically significant steatosis (S≥2) had an overall prevalence of 11.0% (11; 95%CI 6.3-18.6), occurring in 13.2% of participants with normal FIB-4 (9/68) and 6.3% of those with altered FIB-4 (2/32), also without significant association (Fisher’s exact test, P=0.495; OR=2.29; 95%CI 0.46-11.27).
The performance of FIB-4 was assessed at prespecified cutoffs of 1.3 (screening) and 2.67 (high risk) for the outcome of clinically significant steatosis (S≥2) on ultrasonography. Table 3 presents the confusion matrices (TP/FP/TN/FN) and the corresponding sensitivity and specificity with 95% confidence intervals for the overall sample (n=100) and for participants with elevated glycemia (prediabetes + diabetes, ADA criteria; n=75).
DISCUSSION
In this study population, 32% of participants presented with elevated FIB-4, with an equal sex distribution (50% women; 50% men) and no sex differences in continuous FIB-4 values (overall median 1.25; IQR:0.65; women 1.19 [IQR:0.53] vs men 1.26 [IQR:0.85]; Mann-Whitney U, P=0.072). Population-based studies have reported lower proportions of elevated FIB-4, such as 9%14. In a U.S. cohort, 52.3% maintained normal values over one year15, whereas a Japanese cohort reported 78.5%16. Among people living with HIV (PLWH) with hepatic steatosis, 72.8% were above the upper limit17.
Another finding was the proximity of the median FIB-4 in our cohort to previously published metabolic series. The median FIB-4 in our sample was 1.25 (IQR:0.65), comparable to reported means in type 2 diabetes: 1.15±0.52 and 1.4±0.7218,19. In HIV-positive women, a median of 0.86 was reported20. In men with type 2 diabetes, a mean of 1.64±1.19 was observed21. Differences across studies likely reflect the use of different metrics (median vs mean), as well as sample composition, age, and metabolic profile.
Analysis of FIB-4 in relation to anthropometric parameters revealed no significant differences between participants with normal and elevated FIB-4: median BMI was 26.85 kg/m² in the normal group versus 26.06 kg/m² in the elevated group (P=0.790); waist circumference 91.5 cm vs 93.0 cm (P=0.625); and visceral fat on ultrasonography 4.10 cm vs 3.80 cm (P=0.877). These findings are consistent with literature reporting modest or inconsistent associations between adiposity and FIB-4 in metabolic contexts. A negative correlation between BMI and FIB-4 was described in individuals with insulin resistance without diabetes22. Similar FIB-4 means across BMI strata (<25 and ≥25 kg/m²) in diabetes were also observed23.
Regarding glycemic profiles, no statistically significant association was observed between elevated FIB-4 and prediabetes (OR 1.19; 95%CI 0.51-2.77; P=0.680), diabetes (OR 1.57; 95%CI 0.63-3.92; P=0.330), or elevated glycemia (prediabetes + diabetes; OR 2.25; 95%CI 0.76-6.68; P=0.137), nor with fasting glucose (r= −0.08; P=0.449) or HbA1c (r= −0.02; P=0.823). These results align with metabolic series where central FIB-4 values in type 2 diabetes were modest and similar to ours: 1.15±0.5 and 1.4±0.7 vs 1.25 in the present sample18,19.
Ultrasonography revealed a prevalence of 47.0% for mild hepatic steatosis (S≥1) and 11.0% for clinically significant steatosis (S≥2), with no sex differences. Participants with S≥1 had higher BMI (28.4±5.9 vs 25.5±3.8), waist circumference (95.4±12.6 vs 88.7±11.2), visceral fat (4.7±1.8 vs 3.9±1.4), total cholesterol (170.0 vs 148.0 mg/dL), LDL (95.0 vs 77.0 mg/dL), triglycerides (150.0 vs 121.0 mg/dL), and ALT (31.9±29.6 vs 24.0±11.5), all P<0.05. HDL, eGFR, fasting glucose (106.0±45.5 vs 96.0±19.0), and HbA1c (6.0±1.0 vs 5.9±0.7) did not differ (P>0.05). Continuous FIB-4 values were similar across steatosis strata (1.3±0.7 vs 1.2±0.6; P=0.820), reinforcing its utility primarily for fibrosis screening rather than for diagnosing steatosis22. In PLWH, lower imaging-based prevalences have been reported, highlighting clinical heterogeneity17. In the present study, the prevalence of S≥1 was 47.0%.
At prespecified cutoffs (Table 3), FIB-4 showed limited performance in classifying S≥2. In the overall sample, at 1.3, sensitivity was 27.3% (95%CI 9.7-56.6) and specificity 53.9% (43.6-63.9), with a negative predictive value of 85.7% (74.3-92.6) in a low-prevalence setting (S≥2: 11.0%). At 2.67, sensitivity decreased to 0.0% (0.0-25.9) while specificity increased to 93.3% (86.1-96.9). A similar pattern was observed among participants with elevated glycemia (1.3: 30.0% and 49.2%; 2.67: 0.0% and 90.8%, for sensitivity and specificity, respectively). Likelihood ratios close to 1 indicate limited reclassification ability, consistent with evidence that FIB-4 is more effective for excluding advanced fibrosis than for detecting clinically significant steatosis24.
CONCLUSION
Elevated FIB-4 values were observed in 32% of participants, with no sex differences. Steatosis prevalence was 47.0% for S≥1 and 11.0% for S≥2; S≥1 was associated with higher BMI, waist circumference, visceral fat, total cholesterol, LDL, triglycerides, and ALT, while HDL, eGFR, fasting glucose, and HbA1c did not differ. FIB-4 remains useful for fibrosis screening, particularly to exclude advanced disease, but does not replace methods for quantifying hepatic fat when identifying S≥2; in practice, its use should be integrated with metabolic assessment and elastography when FIB-4 ≥1.3. Study limitations include its cross-sectional design, convenience sampling, few S≥2 events (n=11), ultrasound-based outcomes without systematic elastography or histology, and single-center setting, which limits generalizability and underscores the need for larger longitudinal studies.
REFERENCES
-
1 Bourgi K, Wanjalla C, Koethe JR. Inflammation and Metabolic Complications in HIV. Curr HIV/AIDS Rep. 2018;15:371-81. Doi: 10.1007/s11904-018-0411-2.
» https://doi.org/10.1007/s11904-018-0411-2 -
2 Milic J, Menozzi V, Schepis F, Malagoli A, Besutti G, Franconi I, et al. Liver steatosis and nonalcoholic fatty liver disease with fibrosis are predictors of frailty in people living with HIV. AIDS. 2020;34:1915-21. Doi: 10.1097/QAD.0000000000002650.
» https://doi.org/10.1097/QAD.0000000000002650 -
3 Godoy-Matos AF, Valério CM, Silva Júnior WS, de Araujo-Neto JM, Bertoluci MC. 2024 UPDATE: the Brazilian Diabetes Society position on the management of metabolic dysfunction-associated steatotic liver disease (MASLD) in people with prediabetes or type 2 diabetes. Diabetol Metab Syndr. 2024;16:23. Doi: 10.1186/s13098-024-01259-2.
» https://doi.org/10.1186/s13098-024-01259-2 -
4 Lai JC, Liang LY, Wong GL. Noninvasive tests for liver fibrosis in 2024: are there different scales for different diseases? Gastroenterol Rep (Oxf). 2024;12:goae024. Doi: 10.1093/gastro/goae024.
» https://doi.org/10.1093/gastro/goae024 -
5 Sterling RK, Lissen E, Clumeck N, Sola R, Correa MC, Montaner J, et al. Development of a simple noninvasive index to predict significant fibrosis in patients with HIV/HCV coinfection. Hepatology. 2006;43:1317-25. Doi: 10.1002/hep.21178.
» https://doi.org/10.1002/hep.21178 -
6 Yilmaz Y, Kaya E. The role of fibroscan in the era of metabolic (dysfunction)-associated fatty liver disease. Hepatol Forum. 2023;4:1-2. Doi: 10.14744/hf.2022.2022.0044.
» https://doi.org/10.14744/hf.2022.2022.0044 -
7 Guan L, Li L, Zou Y, Zhong J, Qiu L. Association between FIB-4, all-cause mortality, cardiovascular mortality, and cardiovascular disease risk among diabetic individuals: NHANES 1999-2008. Front Cardiovasc Med. 2023;10:1172178. Doi: 10.3389/fcvm.2023.1172178.
» https://doi.org/10.3389/fcvm.2023.1172178 -
8 Watanabe K, Takakubo N, Saigusa T, Nagasawa A, Yamana M, Ojima M, et al. Fibrosis-4 Index Is Closely Associated with Arterial Damage and Future Risk of Coronary Heart Disease in Type 2 Diabetes. Int J Hypertens. 2022;2022:2760027. Doi: 10.1155/2022/2760027.
» https://doi.org/10.1155/2022/2760027 -
9 Rinella ME, Neuschwander-Tetri BA, Siddiqui MS, Abdelmalek MF, Caldwell S, Barb D, et al. AASLD Practice Guidance on the clinical assessment and management of nonalcoholic fatty liver disease. Hepatology. 2023;77:1797-1835. Doi: 10.1097/HEP.0000000000000323.
» https://doi.org/10.1097/HEP.0000000000000323 -
10 European Association for the Study of the Liver. EASL Clinical Practice Guidelines on non-invasive tests for evaluation of liver disease severity and prognosis - 2021 update. J Hepatol. 2021;75:659-89. Doi: 10.1016/j.jhep.2021.05.025.
» https://doi.org/10.1016/j.jhep.2021.05.025 -
11 DallaPiazza M, Amorosa VK, Localio R, Kostman JR, Lo Re V 3rd. Prevalence and risk factors for significant liver fibrosis among HIV-monoinfected patients. BMC Infect Dis. 2010;10:116. Doi: 10.1186/1471-2334-10-116.
» https://doi.org/10.1186/1471-2334-10-116 -
12 Ghaferi AA, Schwartz TA, Pawlik TM. STROBE reporting guidelines for observational studies. JAMA Surg. 2021;156:577-8. Doi: 10.1001/jamasurg.2021.0528.
» https://doi.org/10.1001/jamasurg.2021.0528 -
13 American Diabetes Association Professional Practice Committee. 2. Diagnosis and Classification of Diabetes: Standards of Care in Diabetes-2024. Diabetes Care. 2024;47:S20-42. Doi: 10.2337/dc24-S002.
» https://doi.org/10.2337/dc24-S002 -
14 Iacob DG, Luminos M, Benea OE, Tudor AM, Olariu CM, Iacob SA, et al. Liver fibrosis progression in a cohort of young HIV and HIV/ HBV co-infected patients: A longitudinal study using non-invasive APRI and Fib-4 scores. Front Med (Lausanne). 2022;9:888050. Doi: 10.3389/fmed.2022.888050.
» https://doi.org/10.3389/fmed.2022.888050 -
15 Vieira Barbosa J, Milligan S, Frick A, Broestl J, Younossi Z, Afdhal NH, et al. Fibrosis-4 Index as an Independent Predictor of Mortality and Liver-Related Outcomes in NAFLD. Hepatol Commun. 2022;6:765-79. Doi: 10.1002/hep4.1841.
» https://doi.org/10.1002/hep4.1841 -
16 Wada T, Zeniya M. Background of the FIB-4 index in Japanese non-alcoholic fatty liver disease. Intern Med. 2015;54:127-32. Doi: 10.2169/internalmedicine.54.2685.
» https://doi.org/10.2169/internalmedicine.54.2685 -
17 Yendewa GA, Khazan A, Jacobson JM. Risk Stratification of Advanced Fibrosis in Patients With Human Immunodeficiency Virus and Hepatic Steatosis Using the Fibrosis-4, Nonalcoholic Fatty Liver Disease Fibrosis, and BARD Scores. Open Forum Infect Dis. 2024;11:ofae014. Doi: 10.1093/ofid/ofae014.
» https://doi.org/10.1093/ofid/ofae014 -
18 Karagiannakis DS, Stefanaki K, Petrea F, Zacharaki P, Giannou A, Michalopoulou O, et al. Elevated FIB-4 Is Associated with Higher Rates of Cardiovascular Disease and Extrahepatic Cancer History in Patients with Type 2 Diabetes Mellitus. Biomedicines. 2024;12:823. Doi: 10.3390/biomedicines12040823.
» https://doi.org/10.3390/biomedicines12040823 -
19 Barana L, Nelva A, Scivetti P, Costanzo M, Fanelli CG. Diagnosis of hepatic steatosis and steatohepatitis in people with new-onset type 2 diabetes: a multidisciplinary approach. Endocrine. 2024;84:412-19. Doi: 10.1007/s12020-024-03715-5.
» https://doi.org/10.1007/s12020-024-03715-5 -
20 Blackard JT, Welge JA, Taylor LE, Mayer KH, Klein RS, Celentano DD, et al. HIV mono-infection is associated with FIB-4 - A noninvasive index of liver fibrosis - in women. Clin Infect Dis. 2011;52:674-80 .Doi: 10.1093/cid/ciq199.
» https://doi.org/10.1093/cid/ciq199 -
21 Miyauchi S, Miyake T, Miyazaki M, Eguchi T, Niiya T, Yamamoto S, et al. Free testosterone concentration is inversely associated with markers of liver fibrosis in men with type 2 diabetes mellitus. Endocr J. 2017;64:1137-42. Doi: 10.1507/endocrj.EJ17-0225.
» https://doi.org/10.1507/endocrj.EJ17-0225 -
22 Lopes A, Perozo A, Fontes R, Serfaty F, Castelo M, Cunha M, et al. A-073 Is there a relationship of FIB 4 with weight and insulin resistance in subjects without type 2 diabetes and with normal transaminases? Clinical Chemistry. 2024;70. Doi: 10.1093/clinchem/hvae106.072.
» https://doi.org/10.1093/clinchem/hvae106.072 -
23 Sung MJ, Lim TS, Jeon MY, Lee HW, Kim BK, Kim DY, et al. Sarcopenia Is Independently Associated with the Degree of Liver Fibrosis in Patients with Type 2 Diabetes Mellitus. Gut Liver. 2020;14:626-35. Doi: 10.5009/gnl19126.
» https://doi.org/10.5009/gnl19126 -
24 Castellana M, Donghia R, Guerra V, Procino F, Castellana F, Zupo R, et al. Fibrosis-4 Index vs Nonalcoholic Fatty Liver Disease Fibrosis Score in Identifying Advanced Fibrosis in Subjects With Nonalcoholic Fatty Liver Disease: A Meta-Analysis. Am J Gastroenterol. 2021;116:1833-41. Doi: 10.14309/ajg.0000000000001337.
» https://doi.org/10.14309/ajg.0000000000001337
