# Liver Disease in NAFLD: A Patient's Guide to Noninvasive Testing and What It Means for Your Care Nonalcoholic fatty liver disease (NAFLD) affects roughly 1 billion people worldwide — about one in four adults — yet most don't know they have it until it becomes serious. This expert medical review, published in *Gastroenterology*, examines how doctors can now evaluate liver health in NAFLD patients using blood tests and advanced imaging instead of invasive liver biopsies. The key finding is that advanced liver scarring (fibrosis) is the strongest predictor of poor long-term outcomes, and that a combination of serum biomarkers, ultrasound-based elastography, and MRI-based fat measurement can identify high-risk patients with good accuracy. While no noninvasive test yet perfectly distinguishes simple fatty liver from the dangerous inflammatory form called NASH, these tools are transforming clinical practice for the millions of people at risk. # Liver Disease in NAFLD: A Patient's Guide to Noninvasive Testing and What It Means for Your Care ## Table of Contents - Key Points - Understanding NAFLD and NASH - Why Noninvasive Testing Is Needed - How Noninvasive Tests Work: Blood Markers and Imaging - Diagnosing Fatty Liver (Steatosis) - Diagnosing NASH: The Aggressive Form - What This Means for Patients - What These Tests Cannot Tell Us - Recommendations for Patients and Doctors - Frequently Asked Questions - Source Information ## Key Points - Advanced liver fibrosis, not fat amount, is the strongest predictor of poor outcomes in NAFLD. - Noninvasive tests include blood biomarkers, ultrasound-based elastography, CAP for fat, and MRI-PDFF for precise fat measurement. - Liver biopsy samples only 1/50,000th of the liver, so it can miss patchy fibrosis and carries rare bleeding risks. - In a study of 4,282 patients, liver stiffness predicted liver-related events, but fatty liver severity did not. - MRI-PDFF outperformed CAP in detecting all grades of steatosis in several small studies, with AUROC 0.99 vs 0.85 in one. ## Understanding NAFLD and NASH Nonalcoholic fatty liver disease (NAFLD) is exactly what it sounds like: fat accumulating in the liver in people who drink little or no alcohol. It is remarkably common. The review authors estimate that NAFLD affects around one-fourth of the general population worldwide. In the United States, it is expected to become the leading reason for liver transplantation within the next decade. NAFLD rarely travels alone. The review cites that it is frequently accompanied by metabolic conditions: - **Obesity:** present in 51% of NAFLD patients (95% confidence interval, 41%–61%) - **Type 2 diabetes:** present in 22% (95% CI, 18%–28%) - **Hyperlipidemia (high cholesterol or fats):** present in 69% (95% CI, 50%–83%) - **Hypertension (high blood pressure):** present in 39% (95% CI, 33%–46%) - **Metabolic syndrome:** present in 42% (95% CI, 30%–56%) Most people with NAFLD will never develop serious liver problems. But a smaller subset — roughly 1.5% to 6.5% of the general population — has the progressive form called **nonalcoholic steatohepatitis (NASH)**. In NASH, the liver shows not just fat but also inflammation and damaged liver cells (called "ballooning"), which together drive faster scarring. The statistics on mortality are sobering. For patients with NAFLD, the liver-specific death rate is 0.77 per 1,000 person-years, and overall death rate is 11.77 per 1,000 person-years. For patients with NASH, these numbers jump dramatically: liver-specific mortality rises to 15.44 per 1,000 person-years, and overall mortality to 25.56 per 1,000 person-years. Person-years is a way researchers measure time; one person followed for 10 years equals 10 person-years. So, for example, 25.56 deaths per 1,000 person-years means that if you followed 1,000 patients with NASH for a year, roughly 26 of them would die from any cause. The leading cause of death in NAFLD patients is actually cardiovascular disease, not liver disease. But NASH can progress to advanced fibrosis (extensive scarring), cirrhosis (severe scarring that distorts liver function), hepatocellular carcinoma (liver cancer), and liver failure requiring transplantation. ## Why Noninvasive Testing Is Needed Two critical questions drive clinical care in NAFLD: First, does this patient have NASH or just simple fat? Second, how much liver scarring (fibrosis) has already occurred? The answer to the second question matters most because **advanced fibrosis has been shown to be the major driver of long-term outcomes and death** in these patients. Until recently, liver biopsy was the gold standard for answering both questions. A biopsy involves inserting a needle through the skin to remove a small piece of liver tissue for examination under a microscope. But as the authors point out, biopsy has well-known limitations: - **Invasiveness:** It is a needle procedure with real risks, including rare but potentially life-threatening complications such as bleeding or bile leakage. - **Poor acceptability:** Many patients find the procedure painful and worrying. - **Sampling variability:** A needle sample represents only about 1/50,000th of the liver, and fibrosis is often patchy, so the sample may miss the worst — or best — areas. - **Cost:** It is expensive, requiring skilled personnel and sometimes hospital admission. There is also a practical arithmetic problem: with an estimated 1 billion people affected worldwide, there is no way to biopsy everyone who needs assessment. The review authors emphasize that there is a "substantial unmet need" for efficient, cost-effective, and noninvasive ways to identify which NAFLD patients are at risk for progressing to advanced fibrosis, cirrhosis, and liver-related complications. This need has driven a decade of intensive research into blood tests and imaging techniques. ## How Noninvasive Tests Work: Blood Markers and Imaging Noninvasive methods fall into two broad approaches: **The "biological" approach** measures biomarkers (biological molecules) in a blood sample. These biomarkers reflect liver inflammation, cell death, or the activity of scar-forming cells. Examples include the NAFLD Fibrosis Score, FIB-4, APRI, and the Enhanced Liver Fibrosis (ELF) score. Their practical advantages are high applicability (tests can be obtained in more than 95% of patients), good reproducibility between laboratories, and low cost. Their weakness: none is perfectly liver-specific, and results can be influenced by other medical conditions. **The "physical" approach** measures the mechanical stiffness of the liver using a technique called elastography. Think of it as feeling how firm the liver is: a healthy, soft liver is like a freshly baked sponge, while a scarred (fibrotic) liver is more like a tough, aged one. Elastography sends vibrations through the liver and measures the speed of the resulting shear waves. Faster waves mean a stiffer liver. The liver stiffness measurement (LSM) is reported in kilopascals (kPa) or meters per second. There are two main types of elastography: - **Ultrasound-based elastography:** Includes vibration-controlled transient elastography (TE, best known by the brand name FibroScan), point shear wave elastography (pSWE), which includes acoustic radiation force impulse imaging (ARFI), and two-dimensional shear wave elastography (2D-SWE). The first is the most widely used worldwide. - **Magnetic resonance elastography (MRE):** Uses an MRI scanner to measure waves, offering a larger sample of liver tissue. Some devices measure fat at the same time. TE can measure fat using the **controlled attenuation parameter (CAP)**, while MRI can calculate the **proton-density fat fraction (PDFF)**. ## Diagnosing Fatty Liver (Steatosis) Steatosis means fat buildup in liver cells. Knowing how much fat is present is useful but, as we'll see, less critical than knowing how much fibrosis exists. ### Blood-Based Steatosis Scores Several scoring systems have been developed to detect steatosis from blood tests alone, including the SteatoTest, Fatty Liver Index (FLI), Hepatic Steatosis Index (HSI), lipid accumulation product, Index of NASH, and NAFLD Liver Fat Score. In one head-to-head comparison of 324 patients with suspected NAFLD who underwent liver biopsy, three of these scores performed almost identically: the Fatty Liver Index had an area under the receiver operating characteristic (AUROC) of 0.83, the NAFLD Liver Fat Score 0.80, and the Hepatic Steatosis Index 0.81. To understand AUROC: it is a measure of a test's discrimination ability. An AUROC of 0.5 means the test is no better than a coin flip. A score of 1.0 means a perfect test. An AUROC of 0.83 is generally considered "good" accuracy. Despite reasonable accuracy, these scores have not become popular. The authors note they add little to information already gathered from routine clinic visits, labs, and imaging in patients suspected of having NAFLD. ### Ultrasound: First-Line Imaging Conventional ultrasound is the most commonly used imaging method for diagnosing fatty liver because it is widely available, inexpensive, and well tolerated. Typical ultrasound findings include a "bright" liver (hyperechogenicity compared with the right kidney), weakening of the ultrasound beam as it passes deeper into the liver (distal attenuation), and patchy areas of normal tissue called "focal sparing." How accurate is it? A large meta-analysis of 34 studies including 2,815 patients found that ultrasound distinguishes moderate-to-severe fatty liver from no fat with a pooled sensitivity of 85% (80%–89%) and specificity of 93% (87%–97%). In plain terms: if the disease is there, ultrasound spots it 85% of the time, and if it isn't, ultrasound correctly says so 93% of the time. However, there is an important catch. Ultrasound can only reliably detect steatosis when liver fat content is greater than roughly 2.5% to 20%. Since NAFLD is typically defined by fat content starting at 5%, a "normal" ultrasound can still miss relevant steatosis in a substantial number of patients. Accuracy also drops in people with obesity and in those with coexisting kidney disease. That said, European guidelines for managing NAFLD still recommend ultrasound as the first-choice imaging test for adults at risk. ### Controlled Attenuation Parameter (CAP) CAP is a measurement built into the transient elastography (FibroScan) device. It estimates liver fat based on how much the ultrasound beam weakens (attenuates) as it passes through fat-laden tissue. CAP offers a simple, immediate, bedside fat reading. The first study of CAP's performance involved 115 patients with chronic liver disease (only 15% with NAFLD). It detected steatosis of 11% or more, 33% or more, and 66% or more with AUROCs of 0.91, 0.95, and 0.89, respectively — quite strong accuracy. A more recent individual-patient meta-analysis pooled 19 studies using the M-probe (a certain probe size), covering 2,735 patients, including 537 with NAFLD (19.6%). Results were good but slightly more modest: - For steatosis ≥11%: AUROC 0.82, sensitivity 0.69, specificity 0.82, optimal cutoff **248 dB/m** (95% CI, 237–261) - For steatosis ≥33%: AUROC 0.86, sensitivity 0.77, specificity 0.81, optimal cutoff **268 dB/m** (95% CI, 257–284) - For steatosis ≥66%: AUROC 0.88, sensitivity 0.88, specificity 0.78, optimal cutoff **280 dB/m** (95% CI, 268–294) Notice that CAP values were influenced by the presence of NAFLD itself, diabetes, and body mass index (BMI). Other researchers, using MRI-PDFF as the reference test, have suggested 288 dB/m as the best cutoff for detecting at least 5% fat. A recent U.S. multicenter study using the XL-probe (designed for larger patients) in 393 NAFLD patients found CAP had an AUROC of 0.76 for detecting steatosis greater than 5%, with a 96% positive predictive value at a cutoff of 263 dB/m. In plain terms, when CAP briefly indicated significant fat at that threshold, it was right 96% of the time. However, CAP's accuracy for separating moderate (≥33%) from severe (≥66%) steatosis was only suboptimal. When compared head-to-head with ultrasound, CAP generally performed better at detecting and grading steatosis — but it also overestimated steatosis much more often: 30.5% versus 12.4% (*P* < .05), meaning the difference was statistically significant (less than a 5% chance it occurred randomly). Against MRI-PDFF, CAP came up short. In 78 American NAFLD patients, MRI-PDFF outperformed CAP for diagnosing all grades of steatosis (AUROC 0.99 vs 0.85; *P* = .0091). Similar results were seen in 127 Japanese and 55 Dutch patients. One large longitudinal study followed 4,282 patients who had both reliable liver stiffness measurements and at least 10 successful CAP readings. The result: neither the presence nor the severity of fatty liver predicted liver-related events, cancer, or cardiovascular events in the near term, but liver stiffness measurements (reflecting fibrosis) and the cause of liver disease independently predicted liver-related events. Subgroup analyses of viral hepatitis patients (37.0% hepatitis B, 2.9% hepatitis C) and NAFLD patients (40.7% of the whole cohort) showed similar patterns. In short: **fat content matters less than stiffness** when predicting who will get into trouble. ### MRI-PDFF: The Most Precise Fat Measurement Magnetic resonance imaging proton-density fat fraction (MRI-PDFF) is an advanced MRI technique that precisely quantifies liver fat. It has been widely used in epidemiological studies and is now FDA-approved and commercially available on scanners from GE Healthcare, Siemens, and Philips. In single-center studies, MRI-PDFF was more sensitive than liver biopsy itself at measuring changes in liver fat over a 24-week period. Multicenter studies in both adults and children confirmed these findings. When MRI-PDFF measurements were carefully aligned (co-localized) with magnetic resonance spectroscopy (MRS) — a very precise chemical analysis method — the correlation between changes over time was excellent, ranging from 0.96 to 0.99. MRI-PDFF has also become an important tool in clinical trials. The MOZART trial, which tested the drug ezetimibe versus placebo in NASH patients, demonstrated that the liver fat is unevenly distributed — which is why imaging that captures a larger region of the liver is valuable. Trial analyses showed that a **relative 30% reduction in MRI-PDFF** was associated with significantly higher odds of a 2-point improvement in the NAFLD Activity Score (a biopsy-based scoring system used to gauge NASH severity). Newer research also suggests that patients with higher liver fat at the start — even before fibrosis develops — have significantly higher odds of fibrosis progression later. One major limitation: MRI-PDFF cannot assess liver inflammation, ballooning, or whether NASH has resolved or fibrosis has improved. It measures fat, not the disease process itself. ## Diagnosing NASH: The Aggressive Form Separating NASH from simple steatosis is one of the hardest challenges in liver disease, because the distinction historically required a biopsy to see inflammation and cell damage. ### Blood Tests for NASH Many serum biomarkers have been investigated, but **cytokeratin-18 (CK-18)** fragments have been studied the most. When liver cells die through the process of apoptosis (programmed cell death), the enzyme caspase 3 chops up a structural protein called keratin 18. These fragments circulate in the blood and can be measured with immunoassays. The M30 test specifically detects apoptosis fragments — a hallmark of steatohepatitis — while the M65 test measures total cell death. In the original study by Feldstein and colleagues, a CK-18 level of about 250 U/L predicted NASH with an AUROC of 0.83, a sensitivity of 0.75 (detecting 75% of true NASH cases), and a specificity of 0.81 (correctly identifying 81% of patients without NASH). Many subsequent studies confirmed this, but in relatively small patient groups. Two meta-analyses pooled the data: CK-18 achieved a pooled AUROC of **0.82** (95% CI, 0.76–0.88) in predicting NASH, with a median sensitivity of 66%–78% and specificity of 82%–87%. Those numbers place CK-18 in the "moderately useful" range. However, real-world use of CK-18 has stalled due to several issues: - There is **no commercially available clinical test** for it (it remains a research tool). - Sensitivity at the individual patient level is limited — meaning a normal CK-18 doesn't fully rule out NASH. - Studies have proposed wildly varying cutoff values, making it hard to know which threshold to use in a given patient. Some researchers have combined CK-18 with other markers to boost sensitivity, including sFas levels, uric acid, the hormones adiponectin and resistin (in a panel dubbed "NASH diagnostics"), and the "Nice Model," which combines ALT with the presence of metabolic syndrome. Several other predictive models have been developed: - **HAIR score:** hypertension, elevated ALT, and insulin resistance - **Palekar score:** age, sex, AST, BMI, AST/ALT ratio, and hyaluronic acid - **Gholam score:** AST and diabetes mellitus - **oxNASH:** a ratio of oxidized fat compounds, plus age, BMI, and AST - **NAFIC score:** ferritin, insulin, and type IV collagen 7s - **NashTest:** a proprietary formula including 12 variables (age, sex, height, weight, triglycerides, cholesterol, α2-macroglobulin, apolipoprotein A1, haptoglobin, γ-glutamyltransferase, ALT, AST, and total bilirubin) NashTest had a weighted AUROC of 0.84 in a meta-analysis of 494 obese patients with a NASH prevalence of 17.2%. But many of these models were developed in small, highly selected populations — often morbidly obese patients — and have not been externally validated in broader groups. Newer approaches explore genetic biomarkers, including variations (single nucleotide polymorphisms) in the PNPLA3 gene. The NASH Score combines PNPLA3 genotype, AST, and fasting insulin; the NASH ClinLipMet Score adds specific metabolites (glutamate, isoleucine, glycine, lysophosphatidylcholine 16:0, phosphoethanolamine 40:6). Other researchers have studied microRNA expression, particularly miR-122. So far, these genetic tools have shown only "moderate clinical utility." The bottom line from the authors: **no currently available serum marker can differentiate NASH from simple steatosis with high sensitivity and specificity**, though combining different approaches does improve diagnostic accuracy. ### Imaging Techniques for NASH The original review goes on to discuss imaging-based approaches for diagnosing NASH after this point, but the available excerpt concludes here. The authors do note that imaging tools like elastography and MRI-PDFF can provide crucial complementary information about fibrosis and fat, and that research is ongoing to find imaging signatures that reflect the inflammatory component of NASH. The key message for patients: diagnosing NASH is still challenging, and an accurate diagnosis may require an experienced doctor to weigh multiple test results together. ## What This Means for Patients The biggest practical takeaway: **knowing your fibrosis stage matters more than knowing your liver fat percentage.** Advanced fibrosis is the major driver of mortality in NAFLD, and it is now possible to screen for it noninvasively with a combination of blood tests (like FIB-4 or NAFLD Fibrosis Score) followed by elastography when needed. For the doctor's office, the review sketches a practical approach: 1. Start with serum biomarkers to categorize the patient's risk of advanced fibrosis. 1. Use ultrasound as the first-line imaging test for diagnosing steatosis, per European guidelines. 1. If more precise liver fat quantification is needed — for instance, in a clinical trial — use MRI-PDFF, which outperforms CAP. 1. Use liver stiffness measurement (by transient elastography, ARFI, 2D-SWE, or MRE) to assess fibrosis severity at the point of care. This staged strategy means many patients can avoid biopsy altogether. It also makes population-level screening feasible for the first time — a crucial step given that NAFLD is expected to become the leading indication for liver transplantation in the United States within the next decade. ## What These Tests Cannot Tell Us The review is candid about the limits of current noninvasive tools: - **CAP cannot reliably differentiate adjacent grades of steatosis** — results overlap between mild, moderate, and severe fat categories. - **Ultrasound misses low-level steatosis** (below the 2.5%–20% fat threshold) and performs poorly in obese patients. - < ## Frequently Asked Questions ### What is the difference between NAFLD and NASH? NAFLD means fat builds up in the liver even if you drink little or no alcohol. NASH is the more aggressive form: the liver also shows inflammation and damaged cells, called ballooning. NASH leads to faster scarring, higher death rates, and a greater need for liver transplant than simple fatty liver. ### Why can’t doctors just use a liver biopsy to check my liver? Liver biopsy is invasive, can cause bleeding or bile leakage, and samples only about 1/50,000th of your liver, so it can miss patchy scarring. It is also expensive and painful. Worldwide, about 1 billion people have NAFLD, so biopsy for everyone is not practical or safe. ### What are the noninvasive tests for liver fat and scarring? Doctors use blood tests like FIB-4, NAFLD Fibrosis Score, and Enhanced Liver Fibrosis score to gauge scarring. Ultrasound and a special measurement called CAP detect liver fat. More precise fat measurement uses MRI-PDFF. Shear wave elastography measures liver stiffness, which reflects fibrosis, without needing a needle. ### What does a liver stiffness measurement tell me? Liver stiffness measurement, done with transient elastography or MRI elastography, estimates how scarred your liver is. A stiffer liver suggests more fibrosis. In a large study, liver stiffness independently predicted liver-related events, cancer, and cardiovascular events, while liver fat amount did not predict these outcomes in the near term. ### Is there a blood test that can diagnose NASH? Currently, no blood test can reliably distinguish NASH from simple fatty liver. The most studied marker, CK-18, had a pooled AUROC of 0.82, meaning moderate accuracy. It is not clinically available, and other models were developed in small, selected groups. Doctors often combine multiple tests to weigh the likelihood of NASH. ### When should a patient with NAFLD or NASH seek a second opinion about noninvasive testing and fibrosis assessment? Patients with NAFLD should consider a second opinion when noninvasive test results conflict or when the fibrosis stage remains unclear, because advanced fibrosis is the strongest predictor of long-term outcomes and may determine whether a biopsy is needed. Since no single blood test or imaging method perfectly distinguishes NASH from simple fatty liver, an experienced evaluation combining serum biomarkers, elastography, and MRI-based fat measurement can clarify risk. A second opinion can also help interpret borderline stiffness measurements or decide if additional testing is warranted. Diagnostic Detectives Network provides independent expert second opinions. --- Publisher: Diagnostic Detectives Network (https://diagnosticdetectives.com) — independent multi-expert medical second opinions, worldwide, private-pay. Author byline: Anton Titov, MD, PhD. Contact: https://diagnosticdetectives.com/pages/contact Canonical page: https://diagnosticdetectives.com/products/liver-disease-in-nafld-a-patients-guide-to-noninvasive-testing-and-what-it-means-for-your-care