Cirrhosis screening in dementia

Screening for Cirrhosis in Dementia: The Encephalopathy in the Chart

Screening for cirrhosis in dementia means computing a FIB-4 index from values already in the chart. In two national cohorts, 5% to 13% of patients with dementia had scores pointing to undiagnosed advanced fibrosis or cirrhosis, and hepatic encephalopathy, unlike most dementia, responds to treatment.

A dementia diagnosis ends most cognitive workups. In a meaningful share of patients, routine laboratory values point to a liver that nobody has examined.

Screening for cirrhosis in dementia is not routine, and two national cohorts show it has to be. Hepatic encephalopathy produces confusion, slowed thinking, sleep reversal and personality change, the same features that earn an older adult a dementia label, and unlike most dementia it responds to treatment. When cirrhosis has never been diagnosed, nobody asks, and a scored cognitive assessment gets read as proof of decline rather than a baseline that could improve. A FIB-4 index computed from values already in the chart asks the question, and a positive answer reopens a workup the diagnosis had quietly closed.

What the veteran and TriNetX cohorts found

Bajaj and colleagues examined Veterans Health Administration records from 2009 to 2019 for 177,422 US veterans with a dementia diagnosis at two or more visits, no prior cirrhosis diagnosis and enough laboratory data to calculate FIB-4. The cohort was 97.1% men with a mean age of 78.35 years, and FIB-4 was capped at age 65 in the calculation.

5.3% (9,373) had FIB-4 above 3.25, the threshold suggestive of cirrhosis.
10.3% (18,390) had FIB-4 above 2.67, suggestive of advanced fibrosis.
A local validation cohort of patients with dementia showed high FIB-4 in 4.4% to 11.2%.
FIB-4 above 3.25 was associated with viral hepatitis (odds ratio 1.79) and congestive heart failure (1.48), and inversely associated with diabetes (0.78).

Veterans are overwhelmingly male, so the group repeated the analysis in the multi-center TriNetX network: 68,807 patients with dementia and no cirrhosis diagnosis, 44.7% male, mean age 72.73 years. FIB-4 exceeded 3.25 in 7.6% (5,815) and 2.67 in 12.8% (8,683). Congestive heart failure (odds ratio 1.73) again rose with high scores and diabetes (0.82) again ran inverse. The authors concluded that FIB-4 could be used to screen for potential undiagnosed cirrhosis in patients with dementia. Across both cohorts, somewhere between one in twenty and one in eight patients carrying a dementia diagnosis had a laboratory signature demanding a liver workup that nobody had ordered. The case for liver screening despite normal enzymes is in MASLD screening in primary care.

Reading FIB-4 in the oldest patients

Age is a multiplier in the formula, so the index behaves differently in a memory clinic than in the populations that validated it. In a biopsy cohort from European specialist hepatology clinics, specificity of FIB-4 for advanced fibrosis fell with age and was 35% in those aged 65 and older (n=76). A cut-off of 2.0 in that age group restored specificity to 70% with a sensitivity of 77%. That is why the veteran analysis capped age at 65 years even for those above this cutoff. An uncapped calculator run across a dementia panel overcalls, and overcalling is how a useful screen gets abandoned.

Three confounders belong in the note. Congestive hepatopathy raises FIB-4 through the liver, and both cohorts found heart failure among its strongest associations, so a high value in heart failure is a reason to look at both organs, not to dismiss either. Platelet counts fall for reasons other than portal hypertension, including marrow disease and medication, so read them against the full laboratory panel. And the inverse association with diabetes runs against the metabolic expectation; a retrospective design cannot explain it, and it is no reason to skip patients with diabetes, whose fatty liver is the commonest road to cirrhosis in a metabolic practice. A plain-language version for patients holding their own score is what a fatty liver fibrosis score means.

Why covert encephalopathy goes unrecognized

Covert hepatic encephalopathy has no asterixis and no overt confusion. It lives in attention, processing speed and executive function, the exact domains that dementia screening reads as decline. At 16 medical centers across China, 528 patients with cirrhosis underwent psychometric and Stroop testing, and the prevalence of covert encephalopathy was 50.4%. Among cirrhotic patients referred to a dedicated clinic for cognitive complaints and given a full neuropsychological, EEG and MRI workup, about two thirds had covert encephalopathy; 73% of those also had another cause of impairment, most often cerebrovascular or psychiatric, and clinical improvement was observed in 77% of the patients re-evaluated after specific management. A second diagnosis does not cancel the treatable one.

Liver disease short of cirrhosis already shows up on cognitive testing. In 69 outpatients with fatty liver disease, mean age 50.4 years, 32% scored below a Montreal Cognitive Assessment of 26 and 12% had encephalopathy detected on the EncephalApp test. Genetic evidence points the same way: in a two-sample Mendelian randomization analysis, genetically predicted liver fibrosis and cirrhosis carried an odds ratio of 1.849 for vascular dementia.

The cascade runs through three drivers. First, gut-derived ammonia and bacterial products reach the brain once a scarred liver and portosystemic shunting stop filtering them; the first brain, the gut, becomes the source of the second brain’s toxin load. Second, skeletal muscle is the backup clearance route for ammonia, so sarcopenic wasting in an older patient removes the reserve at the moment the liver fails, which is why measured muscle mass belongs in this workup. Third, and most preventable, is the system: once a chart says dementia, every later cognitive change is pre-explained, alcohol use and isolation in older adults go unasked, and the liver panel is filed instead of calculated. The diagnosis becomes the reason nobody looks.

What a positive screen changes

A high FIB-4 in a patient with dementia is a referral trigger, not a diagnosis. Confirming fibrosis with elastography, staging cirrhosis and diagnosing encephalopathy belong to hepatology and imaging services. What changes in primary care is the frame, because hepatic encephalopathy is treatable. At 11 teaching hospitals in China, 98 cirrhotic patients with minimal encephalopathy were randomized to lactulose or no therapy, and at day 60 the reversal rate was 64.18% with lactulose against 22.58% without, a number needed to treat of 2.4. The 2014 joint American and European guideline builds management around correcting precipitants: sedatives, opioids, dehydration, infection and constipation.

A positive screen also demands cognitive testing that is repeated rather than recorded once. If the score improves after encephalopathy treatment, the dementia label was at least partly wrong; if it does not, the patient has lost nothing but an unexamined assumption. Nutrition becomes treatment with a mechanism: adequate protein and muscle loading rebuild the ammonia-clearing capacity the liver has lost. The broader metabolic reading of a slipping cognitive score is in cognitive screening in primary care: add the metabolic picture. Hepatic encephalopathy belongs on the same differential as the changeable dementia risks sorted in turning dementia risk factors into orders.

Where the measurements fit

Measura [Cardiometabolic and Autonomic Health Analysis] is a testing service. It does not diagnose cirrhosis or hepatic encephalopathy, and it does not treat. It supplies the measurements:

Cognitive assessment establishes a scored baseline that can be repeated after hepatology review or treatment, turning a diagnostic assumption into a measured change.
Laboratory panels include hepatic panel groups selected to the clinical question; the ordering physician can confirm that AST, ALT and a platelet count are drawn together so FIB-4 can be calculated in the chart.
Bioimpedance body composition quantifies muscle mass, the second ammonia-clearance organ and a fall risk in its own right.

Building the screen into standing orders

The screen survives only when it is automatic. A rule in standing orders can require FIB-4 calculation, age capped or age-adjusted, whenever a dementia or mild cognitive impairment diagnosis is added and liver chemistry and a platelet count are on file. The annual wellness visit already requires cognitive assessment; recording the FIB-4 result and any referral beside it keeps the finding visible (see annual wellness visit integration), and pairing it with fall screening follows the structure in cognitive assessment and fall prevention. Unmeasured is unmanaged, and a reversible encephalopathy filed as dementia is the most expensive thing never measured.

Frequently asked questions

Which FIB-4 threshold should be used for patients over 65?

Standard cut-offs lose specificity with age. In a biopsy cohort, a threshold of 2.0 in patients aged 65 and older improved specificity to 70% while keeping sensitivity at 77%. The dementia cohorts instead capped age at 65 in the formula and used 2.67 and 3.25. Either approach is defensible if it is written down and applied consistently. Why a documented baseline matters for later comparison is covered in cognitive screening: what a baseline is for.

Does a high FIB-4 mean the dementia diagnosis is wrong?

It means the diagnosis is incomplete until the liver is evaluated. Covert encephalopathy frequently coexists with vascular and neurodegenerative disease, so improvement after treatment may be partial. The aim is to find every reversible contributor, the same logic that applies to vitamin B12 and homocysteine in cognitive decline, discussed in B vitamins in mild cognitive impairment.

How does hepatic encephalopathy differ from delirium in an older patient?

Overt encephalopathy can look exactly like delirium and is precipitated by the same triggers: infection, dehydration, constipation and sedating drugs. Covert encephalopathy is subtler and persistent, affecting attention and processing speed without an acute change in consciousness. Untreated pain is another reversible driver of confusion that gets sedated rather than examined, as covered in delirium in elderly patients: rule out pain before sedation.

Should patients with fatty liver and insulin resistance get cognitive testing?

Cognitive deficits were common even in younger outpatients with fatty liver disease. A baseline cognitive score in a patient with MASLD, diabetes or central adiposity takes one visit and gives every later result a reference point. The metabolic side of that screening decision is laid out in insulin resistance and dementia: screening past the glucose.

Which medications deserve review when cirrhosis is suspected in a patient with dementia?

Sedative-hypnotics, opioids, anticholinergic drugs and anything that causes constipation or dehydration can precipitate or deepen hepatic encephalopathy, and hepatic clearance of many drugs falls in cirrhosis. A structured medication review belongs in the same encounter as the referral, anchored to measured function rather than a list, as argued in deprescribing starts with a baseline, not a drug list.

Make the liver question routine

Learn how the Measura protocol pairs repeatable cognitive assessment with laboratory and body composition data, and how it fits your standing orders.

Send a message

4477 Woodson Rd, Suite 201, St. Louis, MO 63134. Monday to Friday, 9:00 a.m. to 5:00 p.m. Please do not send symptoms, diagnoses or images through a web form — a website form is not a secure medical channel. Send your name and number and we will call you back.

References

Bajaj, J. S., Silvey, S. G., Rogal, S., O’Leary, J. G., Patton, H., Morgan, T. R., Kanagalingam, G., Gentili, A., et al. (2024). Undiagnosed cirrhosis and hepatic encephalopathy in a national cohort of veterans with dementia. JAMA Network Open, 7(1), e2353965. https://doi.org/10.1001/jamanetworkopen.2023.53965
Silvey, S., Sterling, R. K., French, E., Godschalk, M., Gentili, A., Patel, N., & Bajaj, J. S. (2024). A possible reversible cause of cognitive impairment: undiagnosed cirrhosis and potential hepatic encephalopathy in patients with dementia. The American Journal of Medicine, 137(11), 1082-1087.e1. https://doi.org/10.1016/j.amjmed.2024.06.014
McPherson, S., Hardy, T., Dufour, J. F., Petta, S., Romero-Gomez, M., Allison, M., Oliveira, C. P., Francque, S., et al. (2017). Age as a confounding factor for the accurate non-invasive diagnosis of advanced NAFLD fibrosis. The American Journal of Gastroenterology, 112(5), 740-751. https://doi.org/10.1038/ajg.2016.453
Zeng, X., Yin, C., Sun, C. Y., Lu, C. H., Zhao, S. S., Gao, X. H., Chen, D. F., Wen, L. Z., et al. (2023). Prevalence and risk factors of covert hepatic encephalopathy in cirrhotic patients: a multicenter study in China. Journal of Digestive Diseases, 24(2), 122-132. https://doi.org/10.1111/1751-2980.13171
Sultanik, P., Kheloufi, L., Leproux, A., Bouzbib, C., Mouri, S., Santiago, A., Galanaud, D., Navarro, V., et al. (2024). Other causes of neurocognitive impairment than covert hepatic encephalopathy (CHE) are very frequent, either alone or associated with CHE, in cirrhotic patients with cognitive complaints. Alimentary Pharmacology & Therapeutics, 60(6), 749-764. https://doi.org/10.1111/apt.18148
Parikh, N. S., Wahbeh, F., Tapia, C., Ianelli, M., Liao, V., Jaywant, A., Kamel, H., Kumar, S., et al. (2024). Cognitive impairment and liver fibrosis in non-alcoholic fatty liver disease. BMJ Neurology Open, 6(1), e000543. https://doi.org/10.1136/bmjno-2023-000543
Li, Y. S., Xia, Y. G., Liu, Y. L., Jiang, W. R., Qiu, H. N., Wu, F., Li, J. B., & Lin, J. N. (2024). Metabolic-dysfunction associated steatotic liver disease-related diseases, cognition and dementia: a two-sample mendelian randomization study. PLoS One, 19(2), e0297883. https://doi.org/10.1371/journal.pone.0297883
Wang, J. Y., Bajaj, J. S., Wang, J. B., Shang, J., Zhou, X. M., Guo, X. L., Zhu, X., Meng, L. N., et al. (2019). Lactulose improves cognition, quality of life, and gut microbiota in minimal hepatic encephalopathy: a multicenter, randomized controlled trial. Journal of Digestive Diseases, 20(10), 547-556. https://doi.org/10.1111/1751-2980.12816
Vilstrup, H., Amodio, P., Bajaj, J., Cordoba, J., Ferenci, P., Mullen, K. D., Weissenborn, K., & Wong, P. (2014). Hepatic encephalopathy in chronic liver disease: 2014 practice guideline by the American Association for the Study of Liver Diseases and the European Association for the Study of the Liver. Hepatology, 60(2), 715-735. https://doi.org/10.1002/hep.27210
Bloom, P. P. (2025). The misdiagnosis and underdiagnosis of hepatic encephalopathy. Clinical and Translational Gastroenterology, 16(2), e00784. https://doi.org/10.14309/ctg.0000000000000784