About the MEFIB Index (MRE + FIB-4)
The MEFIB index is a dichotomous rule that combines magnetic resonance elastography (MRE) liver stiffness with the blood-based FIB-4 to identify significant fibrosis (stage F2 or higher) in metabolic (MASLD/NAFLD) liver disease. It is positive when MRE is at least 3.3 kPa and FIB-4 is at least 1.6, and this combination carried a positive predictive value of 91–97% for ≥ F2 across the derivation and validation cohorts — high enough to select candidates for fibrosis-directed treatment or a trial without a biopsy. When both values fall below their thresholds the rule is negative, with a negative predictive value of 86–93%. A result where only one criterion is met is indeterminate.
Formula
MEFIB-positive if (MRE ≥ 3.3 kPa) AND (FIB-4 ≥ 1.6). MEFIB-negative if both are below threshold. Otherwise indeterminate. [FIB-4 = (age × AST) ÷ (platelets × √ALT)]- MRE
- MR elastography liver stiffness in kPa; threshold 3.3.
- FIB-4
- Computed from age (years), AST (U/L), ALT (U/L) and platelets (×10⁹/L); threshold 1.6.
- MEFIB is a rule, not an equation — there is no continuous MEFIB number, only the three-way classification.
- Both conditions must hold for a positive; a single abnormal marker is deliberately not enough, which is what gives the positive result its high specificity.
- The FIB-4 sub-calculation uses the same conventional units as the standalone FIB-4: AST and ALT in U/L, platelets in ×10⁹/L.
Interpreting the result
Treat MEFIB as a high-confidence rule-in when positive and a reasonable rule-out when doubly negative. A positive result — both criteria met — makes significant fibrosis very likely (PPV 91–97%) and, in the context the score was built for, is strong enough to enrol a patient for fibrosis-directed therapy or a trial without histological confirmation. A doubly negative result makes significant fibrosis unlikely (NPV 86–93%) and supports continued metabolic management with reassessment over time; longitudinal data also link a MEFIB-negative result to a low rate of subsequent liver-related events. The discordant case, where only one marker is abnormal, is genuinely indeterminate and should prompt a further test rather than being forced into a positive or negative call.
| Score | Band | What it means | Action |
|---|---|---|---|
| MRE ≥ 3.3 & FIB-4 ≥ 1.6 | MEFIB-positive | Significant fibrosis (≥ F2) very likely — PPV 91–97% across cohorts | Suitable to identify a candidate for fibrosis-directed treatment or trial; specialist referral |
| Discordant (one criterion met) | Indeterminate | MEFIB neither rules significant fibrosis in nor out | Further testing — MRI-PDFF, repeat assessment or biopsy where it changes management |
| MRE < 3.3 & FIB-4 < 1.6 | MEFIB-negative | Significant fibrosis unlikely — NPV 86–93%; low rate of liver-related events on follow-up | Continue metabolic management and reassess over time |
What the MEFIB Index needs (2 inputs)
- MRE liver stiffness (kPa)
- Liver stiffness measured by 2D magnetic resonance elastography at 60 Hz. The MEFIB threshold is 3.3 kPa. MRE is the most accurate non-invasive stiffness measurement but requires an MRI scanner and specific software.
- FIB-4 (from age, AST, ALT, platelets)
- The FIB-4 index is computed internally as (age × AST) ÷ (platelets × √ALT); the MEFIB threshold is 1.6. Because age sits in the FIB-4 numerator, the same age-related caveats that affect FIB-4 carry into MEFIB.
What it returns
- MEFIB result
- Positive (both criteria met), negative (both below threshold) or indeterminate (discordant). It is a rule-in/rule-out classification, not a fibrosis stage.
- Component values
- The MRE stiffness and the computed FIB-4, each shown against its threshold, so it is clear which criterion is or is not met.
How it is calculated
MEFIB pairs a blood marker and an imaging marker that fail in different ways, so that requiring both to be abnormal removes most false positives. FIB-4 is cheap and sensitive but not specific — it is raised by age and by many non-fibrotic causes. MRE is highly accurate but, applied alone at a low threshold, still labels some people without significant fibrosis. The MEFIB authors searched combinations of the two in a biopsied MASLD cohort and found that the pairing of MRE ≥ 3.3 kPa with FIB-4 ≥ 1.6 maximised the positive predictive value for stage ≥ F2. The logic is conjunctive rather than additive: neither marker is weighted or summed, they simply both have to clear their cut-off for the rule to fire.
Facts & figures
| Cohort | Metric | Value |
|---|---|---|
| UCSD-NAFLD (derivation) | PPV of a positive rule | 97.1% |
| UCSD-NAFLD (derivation) | AUROC | 0.90 (0.85–0.95) |
| Japan-NAFLD (validation) | PPV of a positive rule | 91.0% |
| Japan-NAFLD (validation) | AUROC | 0.84 (0.78–0.89) |
| Both cohorts | NPV of a doubly negative rule | 86–93% |
Target is significant fibrosis (stage F2 or higher), not advanced fibrosis alone.
| Marker | Threshold | Type |
|---|---|---|
| MRE liver stiffness | ≥ 3.3 kPa | Imaging (MRI-based) |
| FIB-4 | ≥ 1.6 | Blood (age, AST, ALT, platelets) |
Both must be met for a positive result; both must be below for a negative result.
Evidence
Derivation — UCSD-NAFLD cohort
2021A prospective, biopsy-proven MASLD cohort at UC San Diego with contemporaneous MRE and FIB-4. Combinations of MRE and FIB-4 thresholds were tested against biopsy fibrosis stage, and MRE ≥ 3.3 kPa with FIB-4 ≥ 1.6 was selected as the rule maximising positive predictive value for stage ≥ F2.
For ≥ F2, the positive rule had a positive predictive value of 97.1% and an AUROC of 0.90 (95% CI 0.85–0.95).
External validation — Japan-NAFLD cohort
2021An independent biopsy-proven MASLD cohort in Japan, applying the same MRE ≥ 3.3 kPa and FIB-4 ≥ 1.6 rule.
Positive predictive value 91.0% for ≥ F2, AUROC 0.84 (95% CI 0.78–0.89) — confirming the rule transports across populations.
Head-to-head against FAST
2022Tamaki and colleagues compared MEFIB with the FibroScan-AST (FAST) score for identifying candidates for pharmacological treatment of NASH-related fibrosis in paired cohorts.
MEFIB identified treatment candidates with a higher positive predictive value than FAST, at the cost of requiring MRE rather than the more widely available FibroScan.
How it compares
MEFIB Index vs FIB-4 alone
MEFIB is FIB-4 plus MRE — use FIB-4 alone as the cheap first-line rule-out, and add MRE to form MEFIB when you need a high-confidence rule-in for significant fibrosis without biopsy.
FIB-4 on its own is sensitive but not specific: many raised results are not significant fibrosis, especially in older patients. Adding the requirement of an abnormal MRE lifts the positive predictive value for ≥ F2 from FIB-4's modest figure to around 90% or more. The trade-off is access and cost — FIB-4 is free and universal, while MEFIB needs an MRI-based scan — so the natural pathway is FIB-4 for everyone and MEFIB for those in whom a confident rule-in would change management.
MEFIB Index vs FAST score
Both select candidates for NASH fibrosis therapy, but MEFIB uses MRE and reached a higher positive predictive value head-to-head, while FAST uses the cheaper, more available FibroScan — accuracy versus access.
FAST combines FibroScan liver stiffness and CAP with AST to flag at-risk NASH, whereas MEFIB combines MRE with FIB-4 to flag significant fibrosis. In a direct comparison MEFIB identified treatment candidates with a higher positive predictive value than FAST, reflecting MRE's superior accuracy; but FibroScan is far more widely available than MRE, so FAST is the more deployable of the two where MRI capacity is limited.
MEFIB Index vs MAST score
Both are MRI-based, but they target different things — MAST estimates at-risk NASH (activity plus fibrosis) while MEFIB rules significant fibrosis in or out — so the choice follows whether the question is disease activity or fibrosis stage.
MAST folds MRI-PDFF, MRE and AST into a continuous probability of at-risk NASH (NAS ≥ 4 with F ≥ 2). MEFIB uses MRE and FIB-4 as a dichotomous rule for significant fibrosis (≥ F2) regardless of activity. Both rely on MRE, so availability is similar; the distinction is conceptual — an activity-and-fibrosis composite versus a pure fibrosis-stage rule.
Pearls & pitfalls
- MEFIB is a conjunction: both MRE ≥ 3.3 kPa and FIB-4 ≥ 1.6 must hold. A single abnormal marker is indeterminate, not positive.
- Its strength is the positive result. A MEFIB-positive is a high-confidence rule-in for ≥ F2; treat the negative result as a reasonable but less absolute rule-out.
- FIB-4's age dependence carries over. In older patients a raised FIB-4 may reflect age rather than fibrosis, so the MRE half does more of the work.
- It needs MRE, which requires an MRI scanner with elastography software, adds cost, and is unavailable to patients with common MRI contraindications.
- The target is significant fibrosis (≥ F2), not advanced fibrosis (≥ F3) — do not read a positive result as cirrhosis.
- It is validated in MASLD/NAFLD. Do not apply the 3.3 kPa and 1.6 thresholds to viral or alcohol-related liver disease without separate evidence.
- The discordant group is real and common; resist the urge to round it up to positive or down to negative.
Critical actions
- Confirm both thresholds explicitly before acting — a positive requires MRE ≥ 3.3 kPa and FIB-4 ≥ 1.6 together.
- On a positive result, proceed to specialist assessment and consider fibrosis-directed treatment or trial eligibility; biopsy is often avoidable given the high PPV.
- On a doubly negative result, continue metabolic management and set an interval for reassessment rather than discharging.
- On a discordant result, arrange a further test (MRI-PDFF, repeat MRE/FIB-4, or biopsy) rather than committing to a treatment decision.
- Interpret FIB-4 with its usual caveats — check for non-hepatic causes of thrombocytopenia or a raised AST before relying on the blood half.
Why this score exists
The MEFIB authors were solving a specific operational problem: how to select patients with genuine stage ≥ F2 fibrosis for NASH drug trials without biopsying everyone. Their design choice — demand that both an accurate imaging marker and a cheap blood marker be abnormal — was aimed squarely at the positive predictive value, because in a trial-enrichment setting a false positive is expensive and a missed borderline case is tolerable. That framing explains the rule's shape and its limits: it is optimised to be right when it says 'yes', it accepts an indeterminate middle group rather than forcing a call, and it was never intended as a population screen, because its imaging half is not something you can run at scale.
About the creator
First author, 2021 derivation study
Derived the MEFIB index, pairing MR elastography with FIB-4 to rule in significant fibrosis.
Senior author
Led the San Diego MASLD imaging programme behind MEFIB and much of the comparative non-invasive-test literature.
Limitations
- It requires MRE, which is expensive, not widely available, and precluded by common MRI contraindications, so it cannot serve as a first-line or population test.
- It is a rule for significant fibrosis (≥ F2), not a fibrosis stager and not specific to advanced fibrosis or cirrhosis.
- The blood half inherits FIB-4's weaknesses, including its rise with age and its distortion by non-hepatic causes of thrombocytopenia or transaminitis.
- A substantial discordant group falls into the indeterminate zone, where the rule provides no answer.
- It was derived and validated in MASLD/NAFLD and should not be transferred to other liver diseases without dedicated evidence.
- The high positive predictive value reflects cohorts with a meaningful prevalence of significant fibrosis; in a very low-prevalence setting the predictive values will differ.
If you are the patient
MEFIB is a way of combining two different liver tests to get a more confident answer about scarring. One is a special MRI scan of the liver (called MR elastography, or MRE) that measures how stiff the liver is. The other is FIB-4, a score worked out from your age and three routine blood results. On their own, each test can suggest scarring when there isn't much. MEFIB only calls the result 'positive' when both tests agree that the liver is abnormal — and when they do agree, significant scarring is very likely (right about 9 times out of 10), which can spare you a liver biopsy. If both tests are normal, significant scarring is unlikely. If the two disagree, the result is 'in between' and your team will usually arrange another test. The main catch is that MEFIB needs the special MRI scan, which isn't available everywhere, so it is used as a careful second step rather than a first screening test.
Frequently asked questions
What is the MEFIB index?#
MEFIB is a combined test for significant liver fibrosis (stage F2 or higher) in metabolic fatty liver disease. It is positive when MR elastography liver stiffness is at least 3.3 kPa and the blood-based FIB-4 is at least 1.6. Requiring both to be abnormal gives a high positive predictive value — around 91–97% — so a positive result can identify a treatment candidate without a biopsy.
What do the MEFIB thresholds mean?#
MEFIB-positive means both MRE ≥ 3.3 kPa and FIB-4 ≥ 1.6, indicating significant fibrosis is very likely. MEFIB-negative means both are below those cut-offs, indicating it is unlikely. If only one criterion is met the result is indeterminate, and a further test is needed. The two thresholds were chosen to maximise the positive predictive value for stage ≥ F2.
How accurate is MEFIB?#
In the UC San Diego derivation cohort a positive MEFIB had a positive predictive value of 97.1% for significant fibrosis, with an AUROC of 0.90; in an independent Japanese validation cohort the positive predictive value was 91.0% with an AUROC of 0.84. A doubly negative result had a negative predictive value of 86–93%.
Is MEFIB better than the FAST score?#
In a head-to-head comparison, MEFIB identified candidates for NASH fibrosis treatment with a higher positive predictive value than FAST, reflecting the greater accuracy of MR elastography. However, FAST uses FibroScan, which is far more widely available and cheaper than the MRI-based MRE that MEFIB requires, so FAST is more practical where MRI capacity is limited.
Does MEFIB replace a liver biopsy?#
For confirming a treatment candidate it often can. The whole purpose of MEFIB was to identify people with stage ≥ F2 fibrosis non-invasively, and the ~90%+ positive predictive value of a positive result is high enough that many patients no longer need biopsy. Indeterminate and some negative results may still require further testing, and biopsy remains an option where it would change management.
Which fibrosis stage does MEFIB detect?#
Significant fibrosis, meaning stage F2 or higher on the standard 0–4 scale. That is different from advanced fibrosis (F3–F4), so a positive MEFIB should not be read as cirrhosis; it indicates fibrosis meaningful enough to warrant treatment consideration.
References
Original / primary reference
Component tests
- Loomba R, Wolfson T, Ang B, et al. Magnetic resonance elastography predicts advanced fibrosis in patients with nonalcoholic fatty liver disease: a prospective study. Hepatology. 2014;60(6):1920-1928.
- Sterling RK, Lissen E, Clumeck N, et al. Development of a simple noninvasive index to predict significant fibrosis in patients with HIV/HCV coinfection (FIB-4). Hepatology. 2006;43(6):1317-1325.
Clinical practice guidelines
- Berzigotti A, Tsochatzis E, Boursier J, et al. EASL Clinical Practice Guidelines on non-invasive tests for evaluation of liver disease severity and prognosis – 2021 update. J Hepatol. 2021;75(3):659-689.
- EASL-EASD-EASO Clinical Practice Guidelines on the management of metabolic dysfunction-associated steatotic liver disease (MASLD). J Hepatol. 2024;81(3):492-542.
- 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.