Somewhere in the gap between what people worry about and what actually kills them, the liver sits largely ignored. Ask most adults what they know about liver disease and the conversation turns quickly to alcohol, to cirrhosis, to the chronic afflictions of people who drink too much. What they don’t picture is themselves — or the roughly 38 percent of the world’s adult population who, by current estimates, carry some degree of fatty liver disease whether they know it or not (Cusi et al., Diabetes Care, 2025).
The organ responsible for more than 300 distinct physiological functions — glucose regulation, fat packaging, protein synthesis, immune surveillance, the neutralization of virtually every toxin that enters the bloodstream — has become the site of a quiet epidemic. Not the dramatic kind that generates headlines, but the slow metabolic kind that spends years accumulating silently before announcing itself at a stage where the options have narrowed.
That lag is the central problem. The liver is both the organ most sensitive to metabolic dysfunction and the one most likely to look normal on standard bloodwork until something has gone seriously wrong. Understanding why that is, what the progression actually looks like, and what the evidence says about reversing it, turns out to be one of the more consequential things a person can learn about their own health.
What the liver actually does
Most people’s mental model of liver function begins and ends with alcohol metabolism. That’s not wrong, exactly — the liver does break down ethanol — but it captures perhaps one percent of the organ’s actual workload.
The liver processes every macronutrient the body absorbs. After a meal, glucose rises, the pancreas releases insulin, and insulin signals the liver to pull that glucose from circulation and store it as glycogen. When fasting begins and blood sugar starts to fall, the liver reverses the process, breaking glycogen back down and releasing glucose into the bloodstream. If fasting extends long enough, it synthesizes glucose from scratch through gluconeogenesis. If a meal delivers more sugar than glycogen stores can accommodate, the liver converts the surplus into triglycerides, packages them into apolipoprotein B-containing particles, and ships them to adipose tissue for long-term storage. All of this happens continuously, with a precision that is easy to take for granted until it begins to fail.
The scale of the blood sugar regulation is worth pausing on. A typical fasting blood glucose of 90 mg/dL represents roughly 4.5 grams of glucose in the entire bloodstream — about a teaspoon. A single moderate meal may deliver 90 grams of glucose, twenty times that amount. In a healthy person, blood sugar rarely climbs more than a teaspoon above baseline despite that flood of incoming carbohydrate. That homeostatic feat depends almost entirely on the liver’s ability to absorb, store, and release glucose in precisely calibrated increments. The reserve capacity built into that system is also why early dysfunction is so easy to miss — the system compensates until the compensation itself becomes the problem.
Beyond glucose, the liver synthesizes the majority of proteins circulating in blood: albumin, clotting factors, apoB, IGF-1. It manufactures and exports cholesterol. It is the first destination for all blood draining from the gut, which makes it the body’s primary interface between the digestive system and systemic circulation — and the first line of defense against anything pathogenic that crosses the gut wall.
The reason liver disease matters beyond the liver itself is embedded in this anatomy. An organ that sits at the center of glucose metabolism, lipid metabolism, and protein synthesis cannot malfunction without the malfunction spreading. The leading cause of death in people with metabolic liver disease is not liver failure; it’s cardiovascular disease — because a liver under metabolic stress overproduces apoB-containing particles and amplifies the insulin resistance that accelerates atherosclerosis throughout the body.
A four-stage map
The progression from a metabolically healthy liver to one that is permanently scarred follows a sequence with four distinct stages, and the distinction between them is not merely academic. Which stage a person occupies determines whether the disease is reversible, how urgently intervention matters, and what that intervention needs to accomplish.
Stage one is metabolic stress — the liver beginning to receive more substrate than it can efficiently process. No visible pathology yet, but the biochemical environment is shifting.
Stage two is steatosis: the liver has begun storing excess energy as fat. This is what imaging refers to when it reports a “fatty liver.” Steatosis is a warning sign, not damage in the strict sense, but it is not benign. Fat inside the liver creates the conditions for everything that follows.
Stage three is steatohepatitis. The fat burden has tipped the liver into inflammation, and hepatocytes — the liver’s primary functional cells — begin to die. Each dying cell recruits immune cells that release inflammatory signals into neighboring tissue, killing those cells in turn. It becomes a self-propagating loop. This transition from stage two to stage three is the move from what was once called NAFLD to what was once called NASH, now renamed under a nomenclature designed to better reflect the metabolic origin of the disease: MASLD (metabolic dysfunction-associated steatotic liver disease) and MASH (metabolic dysfunction-associated steatohepatitis).
Stage four is fibrosis. The liver responds to the death of its cells the way any tissue does — by laying down scar tissue. Fibrosis is the body’s attempt at repair. The problem is that scar tissue cannot perform the functions of the cells it replaces. As fibrosis accumulates, the liver moves toward cirrhosis, the end-stage scarring in which so much functional tissue has been replaced that the organ can no longer regulate blood sugar, synthesize proteins, manage clotting, or clear toxins.
Stages one through three are largely reversible. Fibrosis occupies a different category. Early fibrosis can regress, particularly with aggressive intervention. But once scarring disrupts the liver’s architecture at scale, that damage is permanent. The transition to fibrosis is therefore not just one more step along a spectrum — it represents a qualitative change in the nature of the disease and in the urgency of doing something about it.
How a liver accumulates fat
The chain of events from chronic caloric surplus to fatty liver runs through insulin resistance, and the mechanism is specific enough to be worth understanding rather than just accepting as given.
When calorie intake consistently exceeds expenditure, the body stores the surplus as fat. The liver’s role in that process involves converting excess energy into triglycerides through de novo lipogenesis and packaging them into apoB-containing particles — mainly VLDLs — for transport to adipose tissue. This is normal physiology, not pathology. It becomes problematic when adipose tissue becomes overfilled.
As fat cells accumulate more lipid than they can properly manage, a molecule called diacylglycerol (DAG) builds up inside them. DAG interrupts insulin signaling. The fat cells, no longer responding normally to insulin, begin releasing fatty acids back into the bloodstream — the opposite of what the situation calls for. The liver now has to deal not only with the caloric excess arriving from the diet, but also the surplus fat being released by dysfunctional adipose tissue.
The same process eventually takes hold in the liver itself. Lipid intermediates accumulate, insulin signaling becomes impaired, and the liver becomes insulin-resistant. What follows is a phenomenon called selective hepatic insulin resistance, and it has a particular cruelty to it: insulin normally tells the liver to stop releasing glucose into the bloodstream and stop manufacturing new fat. As resistance develops, the first signal fails before the second. The liver continues releasing glucose even when blood sugar is already elevated, while the pancreas responds by producing more insulin — which continues driving fat production. The result is simultaneous hyperglycemia and accelerating fat accumulation in the liver. Production of liver triglycerides eventually outpaces the liver’s capacity to export them, and fat begins building up inside the organ. That is steatosis.
Visceral fat and the portal vein
Not all body fat poses the same risk to the liver, and the reason has everything to do with anatomy.
Fat stored subcutaneously — beneath the skin — releases fatty acids into systemic circulation. Those fatty acids dilute across the entire bloodstream before reaching the liver. Visceral fat, stored around the abdominal organs, drains directly into the portal vein, which delivers blood straight to the liver. The liver receives a concentrated, continuous stream of fatty acids from visceral deposits in a way it never does from subcutaneous fat.
The data that bear this out are not subtle. In a cohort study using CT-measured visceral fat area, people with visceral fat above 200 cm² had a 7.5-fold greater likelihood of hepatic steatosis compared to those below 100 cm², independent of BMI and liver enzymes (Lee et al., PLoS One, 2017). Among people already diagnosed with MASLD in the NHANES database, all-cause mortality in the top quartile of visceral adiposity was nearly 3.5 times that of the lowest quartile (Xie et al., BMC Gastroenterology, 2025).
The practical implication is that body weight and BMI are poor proxies for liver risk. Two people at the same BMI can carry dramatically different visceral fat burdens, and the person with more visceral fat faces a categorically different metabolic threat to their liver. This is part of why metabolic liver disease appears in people who would not conventionally be considered overweight — and why people of Asian ancestry, who tend to accumulate visceral fat at lower body weights, are underserved by weight-based screening criteria.
What muscle has to do with it
Skeletal muscle is the body’s largest glucose sink. Roughly three-quarters of total glucose storage capacity resides in muscle, with the remaining quarter in the liver. More muscle means more capacity to buffer glucose spikes after meals; less muscle means more of that burden lands on the liver.
A 7-year longitudinal Korean cohort study tracked the relationship between changes in skeletal muscle mass and the trajectory of NAFLD. People who gained the most muscle over the study period resolved their NAFLD at more than four times the rate of those who gained the least (adjusted hazard ratio 4.17, 95% CI 1.90–6.17) (Kim et al., Hepatology, 2018). The finding held for both prevention of new cases and resolution of existing disease.
The clinical category of sarcopenic obesity — low muscle mass with excess fat — explains why people who appear metabolically “normal” by weight can develop severe liver disease. The skinny-fat phenotype, as it’s sometimes called colloquially, removes the buffering capacity of muscle while preserving all the metabolic stressors of excess fat.
This is why resistance training occupies an outsized role in any serious approach to liver health. Not because it burns enough calories to solve a caloric surplus problem on its own, but because building and maintaining skeletal muscle fundamentally changes the liver’s metabolic environment.
The fructose question
Dietary sugar generates more popular confusion than almost any other nutrition topic, and the liver is where the fructose-versus-glucose debate is most relevant.
A randomized controlled trial enrolled 94 healthy men and had them consume fructose- or glucose-sweetened beverages for seven weeks while maintaining body weight stability (Geidl-Flueck et al., Journal of Hepatology, 2021). Fructose and sucrose roughly doubled the liver’s de novo lipogenesis activity — its fat-manufacturing machinery — while glucose did not produce this effect. The fructose-specific signal on hepatic fat production is real.
The complication is that on the harder outcome — actual steatosis — controlled feeding studies find that the dominant driver is caloric excess, not fructose specifically. When fructose is substituted isocalorically for other carbohydrates, liver fat barely moves. The reason fructose has earned its reputation in the context of liver disease is not its molecular properties in isolation; it’s the form in which it typically arrives. Liquid sugar in sodas and high-fructose corn syrup beverages is calorie-dense, does not generate satiety signals, and is easy to consume in large quantities. Cohort data link sugar-sweetened beverage consumption to higher MASLD risk (Tseng et al., Nutrients, 2023), but the mechanistic story is almost certainly downstream of calories rather than a fructose-specific toxicity.
The practical upshot: eliminating sugary beverages is one of the highest-yield dietary changes for someone with insulin resistance or early liver disease, primarily because it reduces caloric intake with essentially no compensatory hunger response. The fructose itself matters less than the vehicle in which it arrives.
Alcohol is not a separate conversation
Alcohol-associated liver disease is frequently treated as a distinct clinical entity from metabolic liver disease, which obscures the degree to which they overlap and compound each other.
Alcohol causes fatty liver through a mechanism different from caloric excess — the primary driver is acetaldehyde, the toxic metabolite produced during ethanol oxidation, which accumulates faster when intake exceeds approximately one drink per hour. But the downstream stages of the disease are the same: steatosis, inflammation, fibrosis, cirrhosis. The route is different; the destination is not.
The danger of combining metabolic dysfunction with alcohol consumption is not merely additive. An NHANES cohort study with 26.7 years of follow-up examined adults with established cardiometabolic risk factors who also had hepatic steatosis (Kwak et al., Hepatology, 2025). Steatosis alone, in people with cardiometabolic risk, was not associated with a statistically significant increase in all-cause mortality. But steatosis combined with moderate alcohol consumption — self-reported at roughly 30–60 grams of ethanol per day for men, 20–50 grams for women — produced a hazard ratio of 1.4 for all-cause mortality, 2.35 for cancer mortality, and approximately 15 for liver-specific mortality, compared to people with no steatotic liver disease. A 40 percent increase in death from any cause. A 135 percent increase in cancer death. A 1,400 percent increase in liver-specific death.
This combination has recently been formally recognized under the designation MetALD — metabolic and alcohol-associated liver disease — acknowledging that the two processes are not just co-occurring but synergistically harmful. The alcohol transplant data are equally clarifying: alcohol-associated conditions accounted for roughly 41 percent of liver transplants in 2024 (HRSA, Table LI 7), making it the single largest contributor in the transplant registry.
The genetics of predisposition
Individual variation in liver disease risk is substantial enough to be clinically meaningful, and much of it traces back to specific genetic variants.
The most consequential single-gene factor is a variant in PNPLA3. Carriers of two copies of the risk allele have approximately twice the likelihood of accumulating liver fat, with elevated risk of inflammation and fibrosis that persists even after controlling for standard metabolic risk factors. The variant is far more common in people of Hispanic ancestry, which accounts for much of the elevated population-level MASLD prevalence in that group. Conversely, a loss-of-function variant in HSD17B13 appears protective, associated with lower liver enzyme levels and reduced fibrosis risk, and may partially offset PNPLA3-related harm. Other variants — TM6SF2 E167K and MBOAT7 rs641738 — are associated with increased MASLD severity.
These genetic predispositions are real but not deterministic. Two people with identical metabolic profiles can have profoundly different liver outcomes depending on their genotype. That variation is part of what makes population-level risk tools like BMI unreliable at the individual level.
Hormonal factors are equally consequential. Premenopausal women carry meaningful protection against hepatic fat accumulation, with estrogen appearing to restrain both visceral and hepatic fat deposition. After menopause, that protection fades, and fatty liver becomes more common and can progress more aggressively. For women, menopause status belongs in the liver risk assessment alongside conventional metabolic markers.
Why normal enzymes don’t mean a normal liver
The standard bloodwork interpretation most people receive — “your liver enzymes are normal, your liver is fine” — is based on a misunderstanding of what those enzymes actually measure.
ALT (alanine aminotransferase) and AST (aspartate aminotransferase) are often described as liver function tests, but they are not. They are markers of active hepatocyte damage. They rise in response to acute injury — hepatitis, drug toxicity, alcohol-related inflammation. Early MASLD, and even early MASH, can develop almost entirely while those enzymes remain within the reference range. Fat constituting 15 to 20 percent of the liver’s volume can be present with AST and ALT appearing normal.
The reference range itself compounds the problem. The conventional upper limit of normal — around 40 U/L — was derived from population data that necessarily included large proportions of people with undetected metabolic dysfunction. If more than a third of adults have some degree of MASLD, using that population to define “normal” sets the baseline in the wrong place. More conservative thresholds — AST below 30, ALT below 25 — better reflect genuinely healthy liver metabolism.
The earliest warning signs of liver stress live not in the liver panel but in the metabolic panel: the triglyceride-to-HDL ratio, apoB, fasting insulin, and the oral glucose tolerance test. These markers reflect the insulin resistance and dyslipidemia that precede liver disease. When they trend in the wrong direction, the liver is already under stress, regardless of what the enzymes show.
GGT (gamma-glutamyltransferase) adds useful discrimination. AST can rise from muscle damage — intense exercise, for instance — and alkaline phosphatase can originate from bone as well as liver. An elevated GGT in that context points toward liver origin. A 2:1 AST-to-ALT ratio with elevated GGT suggests an alcohol-driven component rather than a purely metabolic one.
When metabolic markers indicate sustained dysfunction, the FIB-4 score provides a non-invasive proxy for fibrosis risk. The calculation combines age, platelet count, AST, and ALT. By incorporating platelet count — which falls as the liver’s synthetic function deteriorates — the score captures early fibrosis that the enzyme values alone would miss. A FIB-4 above 1.3 generally warrants imaging. Using the age-adjusted threshold that some guidelines endorse for older patients risks missing real pathology in people who have had years of unaddressed metabolic disease; a conservative cutoff applied uniformly is preferable given the low cost and low invasiveness of the follow-up.
What imaging finds
The primary imaging tool for characterizing metabolic liver disease is vibration-controlled transient elastography, sold under the brand name FibroScan. A 10-minute outpatient procedure, it produces two numbers: a score for hepatic fat content and a stiffness score that reflects fibrosis. Those two numbers correspond directly to stages two and four of the disease progression map — the two pathological hallmarks that determine prognosis.
FibroScan costs a few hundred dollars out of pocket or a modest copay with insurance coverage. MRI provides greater resolution and is the reference standard for quantifying proton density fat fraction and assessing fibrosis through magnetic resonance elastography sequences, but it adds cost, patient burden, and is less accessible as a first-line tool. Liver biopsy remains the definitional standard — MASH is technically a histological diagnosis, confirmed by hepatocyte ballooning and immune cell infiltration on biopsy tissue — but clinical practice has moved substantially away from routine biopsy. Non-invasive staging through FibroScan, combined with metabolic markers and FIB-4, now guides most clinical decision-making outside of trial settings or cases with irresolvable diagnostic ambiguity.
What fibrosis actually costs
The four-stage map comes into focus when the mortality data are laid against it.
A systematic review and meta-analysis of biopsy-proven NAFLD studies reported the following hazard ratios for all-cause mortality relative to F0 (no fibrosis): 1.5 at F2, 2.0 at F3, and 3.7 at F4 (cirrhosis) (Ng et al., Clinical Gastroenterology and Hepatology, 2023). The 10-year mortality rate was 7.7 percent for people at stages F0 through F2 combined, and 41.5 percent at F4. The relationship between fibrosis stage and mortality is non-linear — each increment of scarring extracts a disproportionately larger toll.
The cancer data are particularly striking. Hepatocellular carcinoma — a liver cancer with poor detection rates and essentially no curative options once metastatic — occurs at an incidence of 0.03 cases per 100 person-years in MASLD patients below stage F4 (Orci et al., Clinical Gastroenterology and Hepatology, 2021). At F4 cirrhosis, that rate rises to 3.8 cases per 100 person-years. A 125-fold increase in a cancer that is difficult to catch early and harder still to treat.
A retrospective longitudinal case-control study examined patients stratified by FIB-4 score and found a 14-fold higher odds ratio for hepatocellular carcinoma at 10 years among those with high scores, along with a 2-to-3-fold increase for lung, colon, and breast cancers, a 5-fold increase for pancreatic cancer, and more than a 6-fold increase for any metastatic cancer diagnosis (Haimi et al., Digestive Diseases and Sciences, 2025). These are association-level findings from a single retrospective study and should be interpreted with appropriate caution about causality, but the consistency of the signal across cancer types reflects something systemic — fibrosis as a marker of profoundly disrupted metabolic regulation, with consequences that extend well beyond the liver.
Stages two and three, serious as they are, mostly track broader metabolic risk. The fibrosis stage carries independent risk beyond all the other metabolic factors. That asymmetry is what makes preventing the transition to F1 and above the operative goal.
What actually reverses it
The disease is reversible. That point deserves emphasis before any discussion of intervention, because the four-stage progression map can read as a one-way march toward cirrhosis when it is anything but. The liver is the most regeneratively capable organ in the body. The transition from MASLD to MASH typically takes a decade or more; each fibrosis stage can take another five to seven years to advance. That slowness is not passivity — it reflects the liver fighting back. The reversal data are correspondingly real.
Weight loss is the primary tool, and the dose-response relationship is well-characterized. A loss of 5 percent or more of body weight begins to reduce hepatic steatosis. Reaching 7 to 10 percent loss allows inflammation to resolve. Weight loss above 10 percent produces measurable reductions in early-stage fibrosis. These thresholds are averages and vary with starting weight, but the direction is consistent and the mechanism clear: reduced caloric load means less substrate for de novo lipogenesis, less hepatic fat, less inflammatory pressure.
No specific dietary pattern has shown superiority over others in controlled trials when calories are matched. Mediterranean, low-carbohydrate, high-protein — the evidence does not favor any one approach. What matters is a sustained caloric deficit. Liquid calorie elimination — sugary beverages first, then any high-calorie drink — is among the most effective single dietary moves because it reduces intake without generating a significant compensatory hunger response.
GLP-1 receptor agonists have changed what pharmacologic intervention can accomplish. Semaglutide is the only GLP-1 agonist with formal approval for MASH at stages F2 and F3, based on phase 3 trial data: resolution of steatohepatitis without worsening fibrosis occurred in 63 percent of patients, and fibrosis reduction was observed in 37 percent, compared to 34 percent and 22 percent respectively in the placebo arm (Sanyal et al., NEJM, 2025). Tirzepatide, which adds GIP receptor agonism to GLP-1 agonism, completed a successful phase 2 trial in MASH (Loomba et al., NEJM, 2024) and awaits phase 3 results and formal approval. Whether these agents produce liver benefits through weight loss alone or through additional mechanisms remains an open question.
Resmetirom is the other approved agent for MASH. A thyroid hormone receptor beta agonist, it acts preferentially on liver cells, directing hepatocyte mitochondria to burn accumulated fat rather than store it. The beta selectivity is clinically meaningful — broad thyroid activation would affect cardiac tissue, but resmetirom’s targeting profile largely avoids that. A phase 3 trial confirmed its efficacy in NASH with fibrosis (Harrison et al., NEJM, 2024), and it can work alongside caloric restriction rather than depending on it.
Pioglitazone, a PPAR-gamma agonist that fell out of favor in type 2 diabetes management, improves insulin sensitivity in adipose tissue and promotes redistribution of fat from visceral to subcutaneous depots — the direction that reduces hepatic fat load. Paired with an SGLT2 inhibitor or GLP-1 agonist, the edema that has historically limited its use is substantially mitigated. SGLT2 inhibitors independently benefit liver disease, particularly in people with coexisting hypertension, kidney disease, or elevated cardiovascular risk. Metformin, notably, has not demonstrated liver-specific histological benefit in MASLD or MASH trials and does not belong on this list despite its broader metabolic role.
Exercise contributes through two paths. The first is creating and sustaining a caloric deficit, though dietary restriction is typically more efficient at generating that deficit than exercise alone. The second is preserving or building skeletal muscle — the glucose sink whose loss shifts metabolic burden back onto the liver. For patients on caloric restriction, particularly those also using GLP-1 agonists (which can disproportionately reduce lean mass), resistance training is not optional. Any intervention aimed at reversing MASLD that does not include a muscle-preserving exercise component risks trading hepatic fat for sarcopenia, which re-creates the metabolic conditions it was trying to resolve.
What supplements won’t do
The liver detox industry starts from a philosophical error. The liver is a detoxification organ. Hepatocyte-driven enzymatic processes are what “detoxification” means in a vertebrate body — a sophisticated two-phase system for neutralizing and excreting harmful molecules. Marketing a product as something that detoxifies the liver is like offering to sweep a broom.
The irony is sharper than that. If a supplement contains anything biologically active, the liver is the organ that metabolizes it, and during phase I metabolism, hepatocytes are directly exposed to high concentrations of whatever they’re processing. Supplement-driven liver injury has risen substantially: dietary supplements now account for approximately 20 percent of acute liver injury cases reported to the Drug-Induced Liver Injury Network, up from 7 percent in the early 2000s (Navarro et al., Hepatology, 2016). The share of non-acetaminophen drug-induced liver injuries leading to transplant waitlisting that are attributable to supplements rose from under 3 percent in 1995 to approximately 25 percent by 2020 (Ghabril et al., Liver Transplantation, 2022).
Against that background, the list of supplements with real evidence in metabolic liver disease is short.
Vitamin E has the most rigorous trial data, with RCTs showing modest benefit specifically in confirmed MASH. The European MASLD clinical practice guidelines (Tacke et al., Journal of Hepatology, 2024) acknowledge this evidence while noting concerns about high-dose long-term use. It is not a prophylactic and the magnitude of effect is considerably smaller than weight loss or pharmacotherapy, but in confirmed MASH it is worth considering as an adjunct.
Coffee has a consistent protective signal in observational studies, particularly for fibrosis. A UK Biobank analysis found similar associations for decaffeinated coffee, which points toward non-caffeine constituents as the active elements (Kennedy et al., BMC Public Health, 2021). The observational data are robust enough that coffee is unlikely to cause liver harm, but not robust enough to recommend taking up coffee specifically for liver protection. The effect appears only at three or more cups per day, and caffeine sensitivity creates real tradeoffs with sleep quality for many people.
Omega-3 fatty acids show no liver-specific histological benefit in MASH trials despite being marketed extensively for liver health. Their cardiovascular and triglyceride-management benefits may justify use independently, but not for liver disease pathology.
NAC (N-acetylcysteine) has a legitimate application in acetaminophen overdose, where it replenishes glutathione and prevents acute hepatic failure. That mechanism has essentially no overlap with the pathophysiology of metabolic liver disease. Extrapolating from overdose treatment to general liver support is a category error.
Choline deficiency genuinely impairs the liver’s capacity to export fat, and correcting a deficiency can reduce hepatic fat in deficient individuals. Supplementation in people who are not deficient has not demonstrated benefit — the dose-response relationship does not extend into surplus territory.
An RCT of 99 adults with biopsy-proven MASH found that silymarin — the active compound in milk thistle — taken three times daily for one year did not meet its primary endpoint on histological activity score improvement (Kheong, Mustapha, Mahadeva, Clinical Gastroenterology and Hepatology, 2017). Brewed green tea appears safe and marginally studied, but trial data have not demonstrated benefit worth endorsing. Green tea extract in supplement form is a documented cause of clinically apparent liver injury, with the NCBI LiverTox repository listing more than 100 cases including acute liver failure leading to transplant or death. Green tea and green tea extract are not the same thing, and that distinction matters.
A framework for assessment
The clinical picture that emerges from this evidence resolves into a workflow rather than a checklist.
The metabolic panel — fasting insulin, triglyceride-to-HDL ratio, apoB, glucose — reflects the insulin resistance that precedes liver disease. Fasting insulin below 6.0 is reassuring; values consistently in the 6–9 range warrant closer tracking; above 10, the metabolic environment is already hostile regardless of what the A1c or fasting glucose show. A DEXA scan that quantifies visceral fat provides information no blood test can: visceral fat above approximately 800 grams in women or 1,000 grams in men is highly correlated with insulin resistance, and the trajectory over time matters as much as any single measurement.
When metabolic markers indicate sustained dysfunction, the FIB-4 score adds fibrosis risk stratification at no additional cost. A FIB-4 above 1.3, combined with any other concerning indicator, justifies proceeding to FibroScan. For someone already diagnosed with steatosis, annual FibroScan tracking allows the response to intervention to be measured rather than assumed.
The five-year trajectory of a liver that has been accumulating fat for a decade can be reversed. That reversal requires sustained caloric deficit, muscle preservation through resistance training, and in many cases pharmacologic support — not because willpower is insufficient but because the pharmacology has become genuinely effective and the stakes are high enough to use every available tool. What it does not require is a detox protocol, a liver cleanse, or any supplement marketed as “liver support.” The liver’s job is to handle what the body throws at it. The patient’s job is to stop throwing so much.
Sources
Cusi K et al. Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD) in People With Diabetes: The Need for Screening and Early Intervention. Diabetes Care, 2025.
Lee H et al. The relationship between visceral obesity and hepatic steatosis measured by controlled attenuation parameter. PLoS One, 2017.
Xie C et al. Metabolic score for visceral fat is correlated with all-cause and cardiovascular mortality among individuals with non-alcoholic fatty liver disease. BMC Gastroenterology, 2025.
Kim G et al. Relationship Between Relative Skeletal Muscle Mass and Nonalcoholic Fatty Liver Disease: A 7-Year Longitudinal Study. Hepatology, 2018.
Geidl-Flueck B et al. Fructose- and sucrose- but not glucose-sweetened beverages promote hepatic de novo lipogenesis: A randomized controlled trial. Journal of Hepatology, 2021.
Tseng T et al. Sugar-Sweetened Beverages and Artificially Sweetened Beverages Consumption and the Risk of Nonalcoholic Fatty Liver (NAFLD) and Nonalcoholic Steatohepatitis (NASH). Nutrients, 2023.
Kwak M et al. MASLD/MetALD and mortality in individuals with any cardio-metabolic risk factor: A population-based study with 26.7 years of follow-up. Hepatology, 2025.
Ng CH et al. Mortality Outcomes by Fibrosis Stage in Nonalcoholic Fatty Liver Disease: A Systematic Review and Meta-analysis. Clinical Gastroenterology and Hepatology, 2023.
Orci LA et al. Incidence of Hepatocellular Carcinoma in Patients With Nonalcoholic Fatty Liver Disease: A Systematic Review, Meta-analysis, and Meta-regression. Clinical Gastroenterology and Hepatology, 2021.
Haimi M et al. Association Between High FIB-4 Score and the Risk of Malignancy Development and Mortality: A Retrospective Longitudinal Case-Control Study. Digestive Diseases and Sciences, 2025.
Sanyal A et al. Phase 3 Trial of Semaglutide in Metabolic Dysfunction–Associated Steatohepatitis. NEJM, 2025.
Loomba R et al. Tirzepatide for Metabolic Dysfunction–Associated Steatohepatitis with Liver Fibrosis. NEJM, 2024.
Harrison S et al. A Phase 3, Randomized, Controlled Trial of Resmetirom in NASH with Liver Fibrosis. NEJM, 2024.
Navarro V et al. Liver Injury from Herbal and Dietary Supplements. Hepatology, 2016.
Ghabril M et al. Eight-Fold Increase in Dietary Supplement-Related Liver Failure Leading to Transplant Waitlisting Over the Last Quarter Century in the United States. Liver Transplantation, 2022.
Tacke F et al. EASL-EASD-EASO Clinical Practice Guidelines on the management of metabolic dysfunction-associated steatotic liver disease (MASLD). Journal of Hepatology, 2024.
Kennedy O et al. All coffee types decrease the risk of adverse clinical outcomes in chronic liver disease: a UK Biobank study. BMC Public Health, 2021.
Kheong CW, Mustapha NRN, Mahadeva S. A Randomized Trial of Silymarin for the Treatment of Nonalcoholic Steatohepatitis. Clinical Gastroenterology and Hepatology, 2017.
LiverTox: Clinical and Research Information on Drug-Induced Liver Injury — Green Tea. NCBI, 2020.
Afdhal N. Fibroscan (Transient Elastography) for the Measurement of Liver Fibrosis. Gastroenterology & Hepatology, 2012.
HRSA. Clinical Characteristics of Adult Liver Transplant Recipients, Table LI 7, 2024.
