A 40-year-old schoolteacher with type 2 diabetes comes to the medicine outpatient clinic with an ultrasound that says “fatty liver, grade 1”. She does not drink. Her AST is 45 U/L, her ALT 40 U/L and her platelet count 180 ×10⁹/L. She wants to know whether her liver is damaged. This is an illustrative case, not a real patient.
The FIB-4 and Elastography tool gives you her score, the threshold each guideline applies, and what to do next. It no longer tells you where those thresholds came from. That history, the study figures and the worked examples used to sit in its Thresholds and Evidence tabs, and they have moved here. By the end of this article you will know what her FIB-4 can tell you, what it cannot, and why the same blood test reads differently at 40 and at 70.
If she asks you what fatty liver is, the MEDiscuss article written for patients, Why ‘Fatty Liver’ When I Don’t Even Drink?, explains it in plain words.
Use this at the bedside
Open the tool: FIB-4 and Elastography
Get the CDSS app: install the MEDiscuss CDSS on your phone
1. Where FIB-4 came from
The thresholds were derived in two specific populations. Accuracy declines as a patient’s characteristics diverge from those populations.
Sterling and colleagues derived it in 2006, in 832 patients co-infected with HIV and hepatitis C, to separate Ishak stage 4 to 6 from stage 0 to 3 (Sterling 2006). The alternative then was a liver biopsy. Every other use of it since has been a borrowing. Shah 2009 was the first use in fatty liver disease: best of eight scores in 541 patients, in a paper that also called for better markers (Shah 2009).
“So FIB-4 was never built for my patient?” Not for her, no. It was borrowed, and then tested again in people like her. That is why each guideline’s cut-off is printed with the population it was tested in.
APRI has a similar story. Wai derived it in 2003 in 270 patients with chronic hepatitis C (Wai 2003).
The change of name
The 2023 multisociety Delphi replaced non-alcoholic fatty liver disease with metabolic dysfunction-associated steatotic liver disease, MASLD: at least one of five cardiometabolic criteria, and MetALD for the intermediate drinker. INASL adopted it in 2025 and kept the same work-up, risk stratification and management (INASL). Over 98 per cent of the 7,721 patients in the Indian consortium cohort met the new criteria, and FIB-4 and stiffness performed the same under both definitions. Not one threshold in the tool moved.
2. Age is in the numerator
Age enters the numerator of the score, so identical laboratory values return different results at different ages. Look at our schoolteacher’s values: AST 45 U/L, ALT 40 U/L, platelets 180 ×10⁹/L.
- At 40 years old, FIB-4 is 1.58: intermediate, and the guideline sends this patient for a scan.
- At 70 years old, FIB-4 is 2.77: above 2.67, and on the standard reading this patient goes to a hepatologist.
The laboratory values are identical; only the age term differs. McPherson raised the rule-out threshold to 2.0 above 65 for this reason: at the standard cut-off, specificity in that age group fell to 35 per cent, so two of every three positive results were false (McPherson 2017).
3. What the predictive values depend on
The two-step design assumes a population in which advanced fibrosis is uncommon, and both steps are calibrated to that prevalence. Applied in a hepatology clinic, where a third of patients have advanced disease, the predictive values change although no threshold does. The resulting asymmetry is reported in the derivation papers:
- FIB-4 below 1.30 in fatty liver disease: negative predictive value 90 per cent, which is a reliable rule-out.
- FIB-4 at or above 2.67: positive predictive value 80 per cent in Shah’s derivation, and 60 to 80 per cent across the studies the AGA pathway reviewed (AGA 2021).
- Stiffness above 12 kPa in MASLD: AASLD 2023 records the positive predictive value as ranging from 0.34 to 0.71 (AASLD 2023).
In other words, a low score is good at ruling out. A high score is a reason to look further, not a diagnosis. Our schoolteacher, at 1.58, is in the middle band: she needs the scan.
4. APRI and FIB-4 in Indian patients
Kolhe and colleagues biopsied 100 consecutive patients with fatty liver disease from an urban slum population in Mumbai in 2019, 27 of whom had significant fibrosis (Kolhe 2019). For ruling out significant fibrosis, the area under the curve was 0.95 for APRI and 0.78 for FIB-4. That single Indian study is why the tool prints APRI beside FIB-4 on every result.
5. Lean fatty liver and the waist
“She is not even overweight. Can she still have fibrosis?” She can. Roughly 17 per cent of Indian patients with fatty liver disease are lean, pooled across 17 Indian studies, and lean fatty liver has a community prevalence of about 6.5 per cent (Shalimar 2022). In 170 lean Indian patients, 56.5 per cent nonetheless had central obesity by waist circumference. The correlation between body mass index and waist was poor, at r = 0.24.
Within that lean group, those with central obesity had advanced fibrosis on FIB-4 in 19.8 per cent against 8.1 per cent, and on stiffness in 9.5 per cent against none at all. The adjusted odds ratio was 3.11. In a 1,040-patient cohort, FIB-4, stiffness, CAP and the FAST score were all statistically indistinguishable between lean and non-lean patients, and the biopsies showed no difference in steatohepatitis or fibrosis either. So measure the waist, and use the same thresholds.
6. What the machine is doing
A probe on the skin over the right lobe delivers a low-frequency mechanical pulse, a 50 Hz thump, into the liver. That thump travels as a shear wave, and the same probe tracks it by ultrasound. Stiff tissue carries a shear wave faster. The machine converts the velocity to a Young modulus in kilopascals. It samples a cylinder roughly 1 cm across and 4 cm long, about a hundred times the volume of a biopsy core, which is the technique’s real advantage over histology and the reason it is not defeated by sampling error the way a needle is.
Think of tapping a melon to judge it. A firm one answers quickly. But a melon can be firm for more than one reason, and so can a liver. Stiffness is not fibrosis: oedema, inflammation, cholestasis and congestion stiffen a liver too, and the tool lists each of them.
The probe matters
With the M probe in patients above a body mass index of 28, measurement failed in 16 per cent of cases against 1.1 per cent with the XL probe, and a quarter of XL readings were still unreliable against half of M readings. That is why the tool checks the probe against the body mass index.
The spleen
The spleen enlarges and stiffens as portal pressure rises, so its stiffness tracks portal hypertension more directly than the liver’s does (Baveno VII). Its thresholds, and their restrictions, stay in the tool.
7. The threshold tables, whole
The tool now keeps each of these tables in two halves so that they fit a phone. Here they are whole, as the tool carried them on 3 November 2026. Read each row with its source and its population: a threshold taken from one document should not be applied to a disease that another document addresses.
FIB-4
| Source | Population | Rules out | Middle | Rules in |
|---|---|---|---|---|
| Sterling 2006, the original | HIV and hepatitis C co-infection | <1.45, NPV 90 per cent | 1.45 to 3.25 | >3.25, PPV 65 per cent, specificity 97 per cent |
| Shah 2009 | 541 adults with fatty liver disease | ≤1.30, NPV 90 per cent | 1.30 to 2.67 | ≥2.67, PPV 80 per cent |
| McPherson 2017, age 65 and over | 76 patients aged 65 and over, of 634 | <2.0, sensitivity 77 per cent | 2.0 to 2.67 | >2.67 |
| AASLD 2023 and AGA 2021 | MASLD, adopting Shah and McPherson | <1.3, or <2.0 above age 65 | 1.3 to 2.67 | >2.67, refer |
| EASL, EASD and EASO 2024 | MASLD | <1.3, reassess in 1 to 3 years | 1.3 to 2.67 | >2.67, hepatology referral |
| India, NAFLD operational guideline v2.0, September 2024 | The national NCD programme | <1.30, manage at the primary health centre | 1.30 to 2.67, refer to the district hospital | >2.67, refer to the district hospital |
| INASL 2022 | Indian fatty liver disease | <1.3 | 1.3 to 2.67 | >2.67 |
| INASL 2022, the Indian-data caveat in its text | Indian patients, for significant fibrosis | FIB-4 <1, with APRI <0.45 | not stated | not stated |
| India, NVHCP hepatitis B technical guideline 2019 | Chronic hepatitis B | not stated | >1.45 indicates significant fibrosis | ≥3.25 indicates cirrhosis |
| Baveno VII 2022, as a corroborating test | Any aetiology, alongside a stiffness of 10 kPa or more | not used this way | not used this way | ≥2.67 supports the stiffness reading |
APRI
| Source | Question | Rules out | Rules in |
|---|---|---|---|
| Wai 2003, the original | Significant fibrosis, Ishak 3 or more, in hepatitis C | ≤0.5 | >1.5 |
| Wai 2003 | Cirrhosis in hepatitis C | ≤1.0 | >2.0 |
| INASL 2022 | Fibrosis in Indian fatty liver disease | <0.5 | >2.0, grey zone 0.5 to 2.0 |
| India, NVHCP viral hepatitis guideline 2018 | Significant fibrosis, then cirrhosis | not stated | >1.5 significant fibrosis, >2.0 cirrhosis |
| India, NVHCP hepatitis B guideline 2019 | Significant fibrosis, then cirrhosis | not stated | >1.5 significant fibrosis, ≥2.0 cirrhosis |
| Indian diabetologists' consensus 2025 | Fibrosis in MASLD | <0.5 | >1.5, grey zone 0.5 to 1.5 |
Liver stiffness, by cause
| Cause and source | Rules out | Middle | Rules in |
|---|---|---|---|
| MASLD: AASLD 2023, AGA 2021, EASL 2021 and 2024 | <8 kPa, advanced fibrosis unlikely | 8 to 12 kPa | >12 kPa, advanced fibrosis likely |
| MASLD: India, NAFLD operational guideline v2.0, 2024 | 7.9 kPa or below | no middle band is given | >7.9 kPa is significant fibrosis, F2 to F4, refer to a hepatologist |
| MASLD: INASL 2022 | <6.0 kPa, significant fibrosis unlikely | 6.0 to 8.2 kPa | ≥8.2 kPa is F2 or more, ≥9.7 kPa is F3 or more, ≥13.6 kPa is cirrhosis |
| Alcohol-related: EASL 2021 | <8 kPa, advanced fibrosis ruled out | 8 to 12 kPa | ≥12 to 15 kPa, after excluding false positives |
| Hepatitis B with a normal ALT: EASL 2015 and 2017 | <6 kPa | 6 to 9 kPa, the grey area | >9 kPa, severe fibrosis or cirrhosis |
| Hepatitis B with ALT raised but below 5 times normal | <6 kPa | 6 to 12 kPa, the grey area | >12 kPa, severe fibrosis or cirrhosis |
| Hepatitis B: India, NVHCP 2019 | not stated | not stated | ≥8.0 kPa significant fibrosis, ≥12.5 kPa cirrhosis |
| Hepatitis C, untreated: EASL 2018 | not stated as a rule-out | not stated | 10 kPa for F3, 13 kPa for cirrhosis |
| Viral hepatitis: India, NVHCP 2018 | not stated | not stated | >8 kPa significant fibrosis, >11 kPa cirrhosis |
| Hepatitis C after cure: EASL 2021 | EASL gives no cut-off and issues the opposite instruction: the cut-offs used before treatment must not be used to stage fibrosis after a sustained virological response. Baveno VII gives one usable rule instead, below. |
The controlled attenuation parameter
| Source and probe | S1, any steatosis | S2, a third or more | S3, two thirds or more |
|---|---|---|---|
| Karlas 2017, individual patient data, M probe | 248 dB/m | 268 dB/m | 280 dB/m |
| India, NAFLD operational guideline v2.0, 2024 | 248 to 268 dB/m | 268 to 280 dB/m | above 280 dB/m |
| EASL 2021, the only endorsed figure | above 275 dB/m, over 90 per cent sensitivity | no figure given | no figure given |
| Shalimar 2020, Indian, body mass index under 25 | 275 dB/m | 319 dB/m | 337 dB/m |
| Shalimar 2020, Indian, obese | 285 dB/m | 340 dB/m | 355 dB/m |
| Kuchay 2021, Indian, against MRI-PDFF | 262 dB/m for 5 per cent fat | 295 dB/m for 10 per cent fat | not derived |
| EASL 2021, XL probe | 263 dB/m at high sensitivity, 294 dB/m at the best trade-off | not derived | not derived |
Two of the Indian studies in the last table, Shalimar 2020 and Kuchay 2021, are named as the tool names them.
Back to the schoolteacher
Her FIB-4 is 1.58. At 40 years old that is the middle band, so the next step is a stiffness scan, not reassurance and not a hepatologist. Her APRI is printed beside it, because the one Indian biopsy study favoured it for ruling out. Her waist is measured, because a normal body mass index does not protect her. If the scan is low, she is reassessed in 1 to 3 years rather than discharged. Had she been 70, the same blood test would have read 2.77, and the rule-out threshold of 2.0 above 65 would have changed the conversation.
Try this
A 58-year-old man with type 2 diabetes has a stiffness of 22 kPa and a platelet count of 130 ×10⁹/L. His body mass index is 33. Does he have clinically significant portal hypertension? Illustrative case.
Answer. Baveno VII does not permit an answer. Both the 25 kPa rule-in and the ANTICIPATE model, which would otherwise put 20 to 25 kPa with platelets under 150 at a risk of 60 per cent or more, are restricted to viral disease, alcohol-related disease, and steatohepatitis below a body mass index of 30. At 33 he is outside all three. Three statements remain valid: above 20 kPa with platelets below 150 he needs a screening endoscopy if he cannot take a beta-blocker; above the 15 kPa mark compensated advanced chronic liver disease is highly likely; and the next step is a hepatology opinion, not a number.
Take-home points
- FIB-4 was built for HIV and hepatitis C co-infection and borrowed for fatty liver. Read each threshold with the population it was tested in.
- Age is in the numerator. Above 65, the rule-out threshold is 2.0, not 1.3.
- A low FIB-4 rules out well. A high one is a reason to look further, not a diagnosis.
- Measure the waist even when the body mass index is normal.
- Stiffness is not fibrosis. Check the probe and the confounders before you believe the number.
Use this at the bedside
Open the tool: FIB-4 and Elastography
Get the CDSS app: install the MEDiscuss CDSS on your phone
Declaration: the author built and maintains the tool. It is free to use. The figures in this article are those the tool carried in its Thresholds and Evidence tabs on 3 November 2026, each from the source named beside it.
References
- Sterling RK, Lissen E, Clumeck N, et al. Development of a simple noninvasive index to predict significant fibrosis in patients with HIV/HCV coinfection. Hepatology 2006;43(6):1317-25.
- Shah AG, Lydecker A, Murray K, et al. Comparison of noninvasive markers of fibrosis in patients with nonalcoholic fatty liver disease. Clin Gastroenterol Hepatol 2009;7(10):1104-12.
- McPherson S, Hardy T, Dufour JF, et al. Age as a confounding factor for the accurate non-invasive diagnosis of advanced NAFLD fibrosis. Am J Gastroenterol 2017;112(5):740-51.
- Wai CT, Greenson JK, Fontana RJ, et al. A simple noninvasive index can predict both significant fibrosis and cirrhosis in patients with chronic hepatitis C. Hepatology 2003;38(2):518-26.
- Rinella ME, Neuschwander-Tetri BA, Siddiqui MS, et al. AASLD Practice Guidance on the clinical assessment and management of nonalcoholic fatty liver disease. Hepatology 2023;77(5):1797-1835.
- Kanwal F, Shubrook JH, Adams LA, et al. Clinical Care Pathway for the risk stratification and management of patients with nonalcoholic fatty liver disease. Gastroenterology 2021;161(5):1657-69.
- European Association for the Study of the Liver, European Association for the Study of Diabetes, European Association for the Study of Obesity. EASL-EASD-EASO Clinical Practice Guidelines on the management of metabolic dysfunction-associated steatotic liver disease. J Hepatol 2024;81(3):492-542.
- 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(3):659-89.
- European Association for the Study of the Liver. EASL 2017 Clinical Practice Guidelines on the management of hepatitis B virus infection. J Hepatol 2017;67(2):370-98, read with the EASL-ALEH 2015 guidelines on non-invasive tests, J Hepatol 2015;63(1):237-64.
- European Association for the Study of the Liver. EASL Recommendations on Treatment of Hepatitis C 2018. J Hepatol 2018;69(2):461-511.
- de Franchis R, Bosch J, Garcia-Tsao G, Reiberger T, Ripoll C, Baveno VII Faculty. Baveno VII: renewing consensus in portal hypertension. J Hepatol 2022;76(4):959-74.
- Karlas T, Petroff D, Sasso M, et al. Individual patient data meta-analysis of controlled attenuation parameter (CAP) technology for assessing steatosis. J Hepatol 2017;66(5):1022-30.
- Duseja A, Singh SP, De A, et al. Indian National Association for Study of the Liver (INASL) guidance paper on nomenclature, diagnosis and treatment of nonalcoholic fatty liver disease. J Clin Exp Hepatol 2023;13(2):273-302, read with the 2025 INASL statement adopting the MASLD nomenclature, J Clin Exp Hepatol 2025;15(5):102590.
- Directorate General of Health Services, Ministry of Health and Family Welfare, Government of India. Operational Guidelines for Non-alcoholic Fatty Liver Disease under the National Programme for Prevention and Control of Non-Communicable Diseases, Version 2.0, September 2024.
- Kolhe KM, Amarapurkar A, Parikh P, et al. Aspartate transaminase to platelet ratio index (APRI) but not FIB-5 or FIB-4 is accurate in ruling out significant fibrosis in patients with non-alcoholic fatty liver disease (NAFLD) in an urban slum-dwelling population. BMJ Open Gastroenterol 2019;6(1):e000288.
- Shalimar, Elhence A, Bansal B, et al. Prevalence of non-alcoholic fatty liver disease in India: a systematic review and meta-analysis. J Clin Exp Hepatol 2022;12(3):818-29, read with De A, Bhagat N, Mehta M, et al. Central obesity is an independent determinant of advanced fibrosis in lean patients with non-alcoholic fatty liver disease. J Clin Exp Hepatol 2025;15(1):102400.
Leave a Reply