When a 55-year-old with type 2 diabetes and mildly elevated liver enzymes walks into a GP surgery in England, the most likely next step remains an abdominal ultrasound. The scan is quick, familiar, and non-invasive. It often shows a bright, echogenic liver—steatosis, or fatty infiltration. The patient is told they have a “fatty liver” and advised to lose weight. Yet the national guidelines from the National Institute for Health and Care Excellence (NICE) recommend a different first test: transient elastography, better known as FibroScan, which measures liver stiffness as a proxy for fibrosis. The gap between what guidelines say and what actually happens in primary care is wide, and it matters for patients.

The Ultrasound Habit That Won’t Quit

NICE guideline NG49, updated as of late 2024, recommends that adults with NAFLD be considered for FibroScan to assess fibrosis stage, especially those with type 2 diabetes or metabolic syndrome. Ultrasound is explicitly reserved for cases where biliary obstruction is suspected. Yet survey data from the British Society of Gastroenterology suggest that over 60% of primary care referrals for NAFLD evaluation still begin with ultrasound. The habit is deeply ingrained.

Part of the appeal is that ultrasound sees fat. The liver appears hyperechoic compared with the right kidney, and radiologists grade steatosis as mild, moderate, or severe. This gives the GP and patient a concrete finding. But steatosis is not fibrosis. Many patients with mild steatosis on ultrasound already have significant fibrosis, and conversely, some with severe steatosis have no fibrosis at all. The correlation is weak.

FibroScan, by contrast, directly measures liver stiffness using a vibration-controlled transient elastography probe. The result—a kilopascal (kPa) value—correlates with histologic fibrosis stage. A score below 7 kPa generally rules out significant fibrosis; above 12.5 kPa suggests cirrhosis. This is the metric that guidelines use to decide who needs hepatology follow-up and who can be managed in primary care. Ultrasound cannot provide this information.

Patients are often reassured by a “normal” ultrasound report. But a normal ultrasound does not rule out fibrosis. In a cohort study from the UK, roughly one in five patients with NAFLD and a normal ultrasound had F2 or greater fibrosis on biopsy. The false reassurance can delay lifestyle changes and monitoring. GPs, too, may feel they have done enough by ordering the scan, when in fact the diagnostic workup is incomplete.

Why Primary Care Clings to Ultrasound

FibroScan machines are scarce in community settings. As of 2025, most are located in hospital hepatology departments or specialist liver clinics. The NHS has funded some community pilots—roughly 30 integrated care boards have placed FibroScan devices in GP practices or community diagnostic hubs—but coverage remains patchy. In many areas, a GP cannot refer directly for FibroScan; the patient must first see a gastroenterologist or hepatologist, creating a bottleneck.

Ultrasound, meanwhile, is widely available. Most GP surgeries can request an ultrasound directly through their local radiology department, and the scan can often be done within two weeks. It also serves multiple purposes: for the same referral, the radiologist can assess the gallbladder, kidneys, spleen, and pancreas, which is useful when the diagnosis is uncertain. In a time-pressured 10-minute consultation, ordering a single test that covers several possibilities is efficient.

Training also plays a role. Medical school and GP specialty training emphasise ultrasound as the first-line imaging for liver disease. FibroScan is a newer technology, and many GPs have never used it or seen one in action. A 2023 survey of UK GP trainees found that fewer than 20% had received any formal training on transient elastography. The familiarity gap perpetuates the old habit.

Cost perception is another factor. An abdominal ultrasound in the NHS costs roughly £60–80, while a FibroScan is priced around £100–150. But this upfront comparison misses the downstream costs of missed fibrosis: later hepatology referrals, decompensated cirrhosis admissions, and liver transplants. Health economic models from NICE suggest that fibrosis-based triage with FibroScan is cost-effective over a 10-year horizon, but that logic does not always penetrate the day-to-day reality of GP commissioning.

The Biology Behind the Test Mismatch

Non-alcoholic fatty liver disease follows a progression that is well understood but often invisible in primary care. The first stage is simple steatosis—fat accumulation in hepatocytes—which is generally benign. In a subset of patients, steatosis triggers inflammation and hepatocyte ballooning, a state called non-alcoholic steatohepatitis (NASH). Over time, NASH can lead to fibrosis, then cirrhosis, and eventually hepatocellular carcinoma. The key inflection point is the development of fibrosis, because that is what drives prognosis.

Ultrasound detects steatosis with reasonable sensitivity—roughly 85% for moderate to severe fat—but it cannot distinguish steatosis from NASH or fibrosis. FibroScan measures liver stiffness, which increases as fibrosis accumulates. The vibration wave travels faster through stiff tissue, and the device calculates a kPa value that correlates with the Metavir fibrosis stage. This is not a perfect correlation—obesity, inflammation, and food intake can affect readings—but it is far more specific for fibrosis than ultrasound.

Early fibrosis (F1–F2) is asymptomatic. Liver enzymes may be normal or mildly elevated. Many patients with type 2 diabetes and normal ultrasound will have F2 fibrosis, which carries a 10-year cirrhosis risk of roughly 10–15%. Detecting that fibrosis early opens the door for aggressive lifestyle intervention—weight loss of 7–10% can reverse fibrosis in some patients—and for monitoring. Miss it, and the patient returns years later with decompensated disease.

The guideline logic is straightforward: identify fibrosis, not steatosis. Steatosis is common—affecting roughly 25–30% of UK adults—but only a minority progress to significant fibrosis. Screening everyone with steatosis would overwhelm hepatology clinics. FibroScan allows risk stratification: those with low stiffness (<7 kPa) can be reassured and managed in primary care, while those with intermediate or high stiffness need specialist input.

What the Guidelines Actually Say

NICE guideline NG49, published in 2016 and updated in 2021, is clear. For adults with NAFLD, it recommends offering transient elastography (FibroScan) to assess liver fibrosis. If FibroScan is unavailable, the enhanced liver fibrosis (ELF) blood test is an alternative. Ultrasound is recommended only if there is suspicion of biliary obstruction or to rule out other pathology. The guideline explicitly states that ultrasound should not be used to assess fibrosis.

The rationale is evidence-based. A systematic review commissioned by NICE found that FibroScan had a sensitivity of roughly 85% and specificity of 80% for detecting F2 or greater fibrosis, using liver biopsy as the reference standard. Ultrasound, by comparison, had a sensitivity of only 60–70% for significant fibrosis. The ELF test performs similarly to FibroScan in some studies, with an area under the curve of 0.85–0.90 for advanced fibrosis.

FibroScan can reclassify a substantial proportion of patients. In a UK primary care study, roughly 30–40% of patients judged to be low-risk by clinical criteria (normal enzymes, no diabetes) were found to have intermediate or high FibroScan scores, prompting a change in management. Conversely, some patients with elevated enzymes and obesity had low stiffness and could be safely discharged from hepatology follow-up.

Cost-effectiveness modelling by NICE showed that a strategy using FibroScan or ELF as the first test, followed by biopsy or referral for high-risk patients, was cost-effective compared with universal ultrasound or no testing. The incremental cost per quality-adjusted life year (QALY) gained was well below the £20,000–30,000 threshold typically used by NICE. Ultrasound-based strategies were dominated—more costly and less effective—because they missed fibrosis.

Real-World Consequences of Diagnostic Drift

The gap between guideline and practice has tangible effects. False reassurance from a normal ultrasound is perhaps the most common. A patient with type 2 diabetes and a body mass index of 32 is told their liver looks “clear” on ultrasound. They may feel less urgency about weight loss. Meanwhile, their fibrosis may be progressing silently. When they eventually develop symptoms—fatigue, jaundice, ascites—the disease is often advanced.

Missed fibrosis leads to late hepatology referrals. In a 2024 audit of a large NHS trust, the median time from first abnormal liver enzymes to hepatology review was 18 months for patients who had only an ultrasound, compared with 6 months for those who had FibroScan. By the time of referral, roughly 15% of the ultrasound-only group had already developed cirrhosis, compared with 5% in the FibroScan group. The difference represents a missed window for intervention.

Liver cancer screening is another casualty. NICE recommends six-monthly ultrasound surveillance for patients with cirrhosis (F4 fibrosis), but not for those with F3 fibrosis. Yet F3 is a high-risk state: the annual risk of hepatocellular carcinoma in F3 is roughly 1–2%. If FibroScan detects F3, the patient can be considered for surveillance or clinical trials. If ultrasound misses the fibrosis, the patient is not offered screening until cirrhosis develops, at which point the prognosis is worse.

Primary care NAFLD registers remain incomplete. Many GP practices have registers for diabetes, hypertension, and chronic kidney disease, but systematic coding of NAFLD is patchy. Without a fibrosis assessment, patients are not entered into a dedicated recall pathway. They may be seen for their diabetes annually, but liver health is not reviewed. A FibroScan result, even a low one, provides a baseline that can be tracked over time. Ultrasound does not offer that quantitative trajectory.

Consider a typical case: a 55-year-old man with type 2 diabetes, hypertension, and a body mass index of 34. His liver enzymes are mildly elevated (ALT 55 U/L, normal <40). An ultrasound shows mild steatosis. He is told to lose weight and given a diet sheet. No fibrosis assessment is done. Two years later, he develops ascites. A FibroScan in the hepatology clinic reads 18 kPa, consistent with cirrhosis. He is listed for a liver transplant. Had he received a FibroScan at the outset, his F2 fibrosis might have been detected and reversed with structured weight loss.

Bridging the Implementation Gap

Several initiatives aim to close the gap. The NHS Long Term Plan, published in 2019, committed to expanding access to FibroScan in primary care. As of 2025, roughly 30 integrated care boards have piloted community FibroScan services, often colocated with diabetes clinics. Early results from the “Liver Disease Prevention” pilot in West Yorkshire showed that placing a FibroScan in a GP practice increased fibrosis detection by 40% and reduced hepatology referrals for low-risk patients by 25%.

Tele-FibroScan is another emerging model. A healthcare assistant or nurse performs the scan in the GP practice, and the result is interpreted remotely by a hepatologist via a secure platform. This avoids the need for the patient to travel to a hospital and allows the GP to receive a structured report with management recommendations. A feasibility study in Norfolk found that 90% of scans were technically adequate, and the median time from referral to result was 7 days.

Training GPs to use the ELF blood test as a first step is a low-cost alternative. The ELF test measures three serum markers (hyaluronic acid, procollagen III N-terminal peptide, and tissue inhibitor of metalloproteinase 1) and provides a numeric score. It can be ordered from a standard blood draw and sent to a reference lab. NICE recommends it when FibroScan is unavailable. Some integrated care boards have made ELF the default test for NAFLD in primary care, with a clear threshold for referral (ELF >10.5).

Incentives also matter. The Quality and Outcomes Framework (QOF) for general practice includes indicators for diabetes and hypertension but not for NAFLD. Some local commissioning groups have introduced enhanced services that reward practices for performing fibrosis assessments in patients with abnormal liver enzymes. A 2024 analysis in the British Journal of General Practice found that practices in areas with such incentives were twice as likely to have ordered a FibroScan or ELF test in the previous year.

Simple clinical decision aids can help. The FIB-4 score, calculated from age, ALT, AST, and platelet count, is a free, validated tool that can be computed in the consultation. A FIB-4 below 1.3 rules out advanced fibrosis with high negative predictive value; a score above 2.67 warrants further assessment. Some GP electronic health record systems now auto-calculate FIB-4 when liver enzymes are abnormal, prompting the clinician to consider fibrosis staging. This is a low-cost nudge that can reduce unnecessary ultrasound referrals.

What a Better Diagnostic Path Looks Like

A rational diagnostic pathway for NAFLD in primary care starts with opportunistic case finding. Patients with type 2 diabetes, obesity, or metabolic syndrome should have a FIB-4 or ELF test as part of their annual review. Those with low FIB-4 (<1.3) can be reassured and given lifestyle advice. Those with intermediate or high FIB-4 should proceed to FibroScan. Ultrasound is reserved for patients with abnormal liver enzymes suggesting biliary obstruction (e.g., elevated alkaline phosphatase or bilirubin).

For patients who undergo FibroScan, the result guides next steps. A stiffness below 7 kPa is low risk; the patient can be discharged to primary care with advice on weight loss and annual review. A stiffness of 7–12 kPa is intermediate; the patient should be referred to hepatology for consideration of biopsy or further monitoring. A stiffness above 12.5 kPa suggests cirrhosis; the patient needs hepatology follow-up, variceal screening, and six-monthly ultrasound for liver cancer surveillance.

Reassessment is important because fibrosis can regress. Weight loss of 7–10% has been shown to reduce liver stiffness by roughly 2–3 kPa in some studies. Patients who achieve sustained weight loss may move from intermediate to low risk, allowing them to be stepped down from hepatology follow-up. Annual FibroScan or ELF can track this trajectory. Ultrasound cannot provide the same quantitative feedback, which can be demotivating for patients who have made lifestyle changes.

Patient-facing communication is critical. Most patients have never heard of fibrosis, but they understand “liver stiffness” or “scarring.” A simple analogy: “Think of your liver like a sponge. Fat makes it look yellow, but scarring makes it stiff, and stiffness is what we worry about because it can lead to liver failure. The scan we use measures stiffness, not just fat.” This reframes the conversation from a vague finding to a actionable metric. Practices that have adopted such language report higher patient engagement with weight loss programmes.

The transition from ultrasound to FibroScan will not happen overnight. Ultrasound remains a useful tool for selected indications, and it is not wrong to use it—it is wrong to rely on it as a fibrosis assessment. The goal is not to eliminate ultrasound but to relegate it to its proper role: ruling out obstruction, not staging disease. With continued investment in community FibroScan services, GP training, and electronic health record prompts, the gap between guideline and practice can narrow. For the patient with silent fibrosis, the difference is not academic; it is the difference between early intervention and late-stage disease.

This article is for informational purposes only and does not constitute medical advice. Individual clinical decisions should be made in consultation with a qualified healthcare professional.