Diagnostic value of CEUS in detecting small (≤2 cm) HCC lesions in patients with liver cirrhosis: a scoping review considering obesity (BMI ≥30 kg/m²) and grade II–III ascites

Total
0
Shares
Share 0
Tweet 0
Pin it 0
Total
0
Shares
0
0
0
0

R Korczyk 12.09.2026

Background and objective: Small focal lesions (≤20 mm) in cirrhotic livers represent a major diagnostic challenge. Although contrast-enhanced ultrasound (CEUS) enables real-time perfusion assessment, its performance varies depending on the diagnostic algorithms and reference standards applied. This scoping review aimed to map the literature on the diagnostic performance of CEUS for nodules ≤20 mm, assess the role of CEUS LI-RADS, and identify evidence gaps in populations with a suboptimal acoustic window, including patients with high BMI or ascites.

Methods: A multi-database search of PubMed, Scopus, ScienceDirect, and Google Scholar was conducted through 2026. Eligible studies evaluated CEUS as the index test for lesions ≤20 mm in patients at risk of HCC. The analysis included the type of contrast agent (SonoVue vs Sonazoid), interpretation criteria, reference standard, and reporting of technical limitations.

Results: Standardized classification systems, particularly CEUS LI-RADS category LR-5, demonstrate excellent specificity and positive predictive value (≥90–100%) for nodules ≤20 mm, supporting their use as reliable rule-in tools. Sensitivity remains moderate (approximately 50–66%) and is influenced by lesion size (<10 mm vs 10–20 mm), lesion depth, washout timing, and differences between purely intravascular contrast agents and agents with a Kupffer phase. Heterogeneous reference standards (histopathology vs composite standards) introduce a risk of bias. A major evidence gap was identified: none of the studies reported diagnostic metrics stratified by BMI or by the presence or grade of ascites; these factors were addressed only as descriptive limitations.

Conclusions: CEUS is a highly useful problem-solving tool in the diagnostic pathway for lesions ≤20 mm, potentially shortening the diagnostic process without a risk of nephrotoxicity. Its performance nevertheless depends on the quality of the acoustic window. Future prospective studies should standardize reporting of visualization quality and evaluate CEUS performance in technically challenging clinical subgroups, particularly patients with obesity (BMI ≥30 kg/m²) and grade II–III ascites.

Keywords: contrast-enhanced ultrasound; CEUS LI-RADS; hepatocellular carcinoma; liver cirrhosis; focal liver lesions; body mass index; ascites.

  1. Introduction

1.1. Clinical significance of HCC in cirrhosis and the need for early detection

Hepatocellular carcinoma (HCC) is the most common primary liver malignancy in patients with cirrhosis, irrespective of the etiology of the underlying liver disease, and is associated with high mortality in this population. Consequently, patients with established cirrhosis undergo cancer surveillance aimed at detecting focal liver lesions as early as possible and thereby increasing the proportion of patients eligible for potentially curative treatment (46,47).

1.2. Small focal liver lesions as a particular diagnostic challenge

In this review, the term focal liver lesions (FLLs) is used as an umbrella term for focal abnormalities described in the literature. Small FLLs (≤2 cm) pose a particular diagnostic challenge because they may exhibit atypical or less pronounced vascular features. Moreover, technical limitations of ultrasonography affect both their detectability and reliable characterization. In clinical practice, this may delay the diagnosis of HCC or necessitate complex diagnostic pathways involving several imaging modalities (46,47).

1.3. Limitations of conventional B-mode ultrasonography in surveillance: the role of obesity and ascites

Conventional B-mode ultrasonography remains the principal surveillance tool in cirrhosis; however, its performance is substantially reduced by patient-related factors, particularly obesity and ascites. Ascites, which frequently accompanies portal hypertension, may affect imaging conditions in two directions: it can sometimes facilitate acoustic coupling, yet it may also impair image quality by displacing the liver and increasing the distance between the organ and the transducer. Obesity and the associated hepatic steatosis increase attenuation and scattering of the ultrasound beam, increase the imaging depth, and more frequently result in examinations of suboptimal quality. Consequently, the sensitivity for detecting small FLLs suspicious for HCC is substantially lower than in patients without these factors (40,41,46,47).

1.4. CEUS as a problem-solving method for evaluating FLLs in patients with cirrhosis

Contrast-enhanced ultrasound (CEUS) permits dynamic assessment of focal-lesion enhancement following intravenous administration of a microbubble contrast agent. It is an important adjunct to conventional ultrasonography for the diagnostic evaluation of detected FLLs—that is, as a problem-solving technique—in patients with cirrhosis (13–16,17,33,47,48). Gas-filled microbubble contrast agents remain within the intravascular compartment, are not nephrotoxic, and are eliminated through the respiratory system. These characteristics distinguish them from iodinated and gadolinium-based contrast agents used in contrast-enhanced computed tomography (CECT) and magnetic resonance imaging (MRI), respectively (33,48).

1.5. CEUS criteria for diagnosing HCC: APHE and contrast washout

The diagnosis of HCC relies on analysis of the time-dependent enhancement pattern, particularly the presence of arterial phase hyperenhancement (APHE) and the timing and degree of contrast washout in later phases (12–16,23–25,47,48). International ultrasound-society guidelines emphasize the growing role of CEUS in evaluating FLLs while noting that CEUS shares some of the acoustic-window limitations of conventional ultrasonography (17,33,48).

1.6. Current evidence and evidence gaps: small FLLs/HCC and diagnostically challenging populations

Numerous studies have evaluated the performance of CEUS for detecting and characterizing FLLs suspicious for HCC in cirrhotic livers, and some suggest that CEUS may outperform conventional B-mode ultrasonography in the assessment of small FLLs ≤2 cm. However, the evidence remains dispersed across heterogeneous patient populations and diverse examination protocols (1–3,5,8,10,12–16,19–22,28). Importantly, patients with obesity (BMI ≥30 kg/m²) and at least moderate ascites are poorly characterized in many studies, analyzed only as small subgroups, or excluded altogether. This limits the applicability of findings to routine clinical practice (48). Previous reviews have focused primarily on the general role of CEUS in diagnosing FLLs and have not systematically mapped the evidence regarding small FLLs/HCC in populations at increased risk of suboptimal examination quality.

1.7. Rationale for selecting a scoping-review design

Given the heterogeneity of the available studies, differences in definitions of high-risk populations, varied reference standards (CECT, MRI, histopathology, and follow-up), and diverse imaging protocols, a scoping-review approach is appropriate (14–16,19,28). This methodology makes it possible to organize the existing evidence, describe its scope and characteristics, and identify areas in which high-quality evidence is lacking—particularly in patients with obesity and advanced ascites.

1.8. Study rationale and objectives

Rationale. Small (≤2 cm) HCC lesions in cirrhotic livers are difficult to detect and evaluate with conventional ultrasonography. Obesity and ascites further impair the acoustic window and increase the risk of missed lesions during surveillance. CEUS may improve the vascular characterization of small FLLs and support diagnostic decisions within a problem-solving pathway; however, evidence regarding its performance in diagnostically challenging populations is heterogeneous and has not been fully synthesized (47,48).

Primary objective. To map and organize the literature on the diagnostic value of CEUS for detecting and characterizing small (≤2 cm) FLLs in patients with liver cirrhosis, with particular emphasis on HCC diagnosis.

Secondary objectives.

The secondary objectives were to compare CEUS with B-mode ultrasonography and other imaging modalities (CECT and MRI) for the detection and characterization of small lesions; identify factors that reduce the sensitivity of conventional ultrasonography in cirrhosis—particularly obesity and ascites—and assess the extent to which CEUS may compensate for these limitations; determine how and to what extent studies included patients with BMI ≥30 kg/m² and grade II–III ascites; and identify research gaps and areas requiring further investigation, including standardization of protocols and endpoint definitions.

2.1. Study design and methodological framework

The study was designed as a scoping review intended to map and organize the available evidence regarding the diagnostic value of contrast-enhanced ultrasound (CEUS) for detecting and characterizing small focal liver lesions (≤20 mm / <2 cm) in adults with cirrhosis or otherwise at risk of HCC. The scoping-review format was selected because of the heterogeneity of the available studies, including differences in patient populations, CEUS protocols, contrast agents, reference standards, and reporting of BMI and ascites subgroups, which would hinder a conventional meta-analysis. The review was planned and reported in accordance with PRISMA-ScR recommendations.

2.2. Research question (PIRD)

The research question was formulated using the PIRD framework (Population–Index test–Reference standard–Diagnosis):

  • P (Population): adults with liver cirrhosis, or a clearly defined population at risk of HCC in which cirrhosis constituted the predominant risk context, with a small focal liver lesion (≤20 mm / <2 cm).
  • I (Index test): liver CEUS, including interpretation according to CEUS LI-RADS, WFUMB–EFSUMB, or clearly described enhancement-pattern criteria.
  • R (Reference standard): CECT and/or MRI and/or histopathology and/or imaging follow-up, as defined by the study authors.
  • D (Diagnosis): HCC among small lesions measuring ≤20 mm / <2 cm.

2.3. Primary objective

The primary objective was to describe and synthetically map the available evidence on the diagnostic performance of CEUS for diagnosing and characterizing small (≤20 mm / <2 cm) focal lesions suspicious for HCC in patients with cirrhosis.

2.4. Secondary objectives

The secondary objectives were:

  • to assess how frequently and in what manner studies reported data on obesity (BMI ≥30 kg/m²) and grade II–III ascites, and whether results were stratified for these subgroups;
  • to compare the role of CEUS with:
    • B-mode ultrasonography in terms of image quality and lesion detectability;
    • CECT/MRI in differential diagnosis, as reference standards, and in problem-solving pathways;
  • to identify sources of heterogeneity and potential bias, including differences in reference standards, diagnostic criteria, contrast agents, and duration of washout observation;
  • to identify research gaps, particularly regarding the effect of BMI and ascites on CEUS diagnostic metrics in small HCCs.

2.5. Data sources and literature-search strategy

The literature search covered PubMed, Scopus, and ScienceDirect, with supplementary searches of Google Scholar. The time frame comprised publications from 2010 through 2026, with priority given to recent sources and additional searches restricted to 2020–2026 where appropriate.

The search logic combined four components:

  1. cirrhosis or an at-risk population;
  2. CEUS;
  3. HCC or focal liver lesions;
  4. small lesion size (≤20 mm, 2 cm, “small,” or “subcentimeter”).

An example of the conceptual search logic was:

  • (cirrhosis OR cirrhotic) AND (“contrast-enhanced ultrasound” OR CEUS OR “contrast enhanced ultrasonography”) AND (“hepatocellular carcinoma” OR HCC OR “focal liver lesion” OR nodule) AND (“20 mm” OR “2 cm” OR small OR subcentimeter).

Supplementary searches targeted factors that may impede imaging:

  • obesity/BMI: (obes* OR BMI OR “body mass index” OR overweight)
  • ascites: (ascites OR ascitic)

2.6. Supplementary AI-assisted search and archiving

To reduce the risk of omitting relevant studies—particularly those concerning CEUS LI-RADS in nodules ≤20 mm and comparisons between PFB-CEUS and MRI—a supplementary AI-assisted search was performed using OpenEvidence on 12 April 2026. The results were treated as supplementary evidence discovery rather than as a primary database, and records underwent the same eligibility assessment as all other studies. An archival search of the Web of Science Core Collection was also performed through OpenEvidence using a TS/Topic-type query without the “≤” symbol. Search dates, working queries, and the screening log were archived in a working document.

2.7. Eligibility criteria: inclusion

Studies were eligible when all the following criteria were met:

  • Publication type: original prospective or retrospective studies.
  • Population: adults with liver cirrhosis or a clearly defined population at risk of HCC, preferably with cirrhosis predominating.
  • Lesions: small FLLs ≤20 mm / <2 cm. Studies including a broader size range were accepted only when data for lesions ≤20 mm/≤2 cm could be extracted separately.
  • Index test: CEUS used to diagnose or characterize lesions, including in a problem-solving role.
  • Reference standard: CECT and/or MRI and/or histopathology and/or imaging follow-up sufficient to establish the final diagnosis.
  • Outcomes: quantitative diagnostic metrics, including sensitivity, specificity, PPV, NPV, accuracy, or AUC, or category-specific metrics such as those for CEUS LI-RADS LR-5, preferably reported for lesions ≤20 mm/≤2 cm.

Information concerning obesity (BMI ≥30) and/or grade II–III ascites was considered particularly desirable for extraction but was not required for inclusion because underreporting was anticipated.

2.8. Exclusion criteria

The following were excluded:

  • reviews, commentaries, letters, expert statements, and case reports without quantitative data; these could be retained as background for the discussion (25,26);
  • studies in which CEUS was not the index test, such as CT-only or MRI-only studies (6);
  • studies from which data for lesions ≤20 mm/≤2 cm could not be extracted;
  • studies concerning tumors other than HCC when HCC-specific data could not be separated (33);
  • studies in which CEUS was used exclusively to guide biopsy or ablation and did not provide CEUS-based diagnostic characterization. Such studies were considered, at most, for background information when contextually useful (9).

2.9. Study selection and handling of records classified as UNCLEAR

Study selection proceeded in two stages:

  1. title and abstract screening;
  2. full-text assessment of records classified as INCLUDE or UNCLEAR.

Screening was performed by one reviewer. Records classified as UNCLEAR during title and abstract screening were automatically advanced to full-text assessment. Duplicates were identified and flagged to prevent double counting in the PRISMA flow diagram.

2.10. Data charting and extracted variables

Data were extracted into a summary table or master screening table prepared for the review. Extracted information included:

  • bibliographic data, author, year, country, and study centers;
  • study design, including prospective or retrospective and single-center or multicenter status;
  • population characteristics, definition of the at-risk population, proportion with cirrhosis, and clinical setting—surveillance versus evaluation of a previously detected lesion (5,13,24,38);
  • lesion characteristics, including the ≤20 mm/≤2 cm size criterion, numbers of patients and lesions, and lesion location where available;
  • CEUS protocol, including contrast-agent type—purely intravascular versus Kupffer-phase agents—imaging phases, observation time, and interpretation criteria such as CEUS LI-RADS, WFUMB–EFSUMB, or other criteria (29,30,34,35,36,42,48);
  • reference standard, including CT, MRI, histopathology, or follow-up, and the definition of the final diagnosis (3,8,14);
  • diagnostic metrics, including sensitivity, specificity, PPV, NPV, accuracy, and AUC, as available in the abstract or full text;
  • information on BMI, obesity, steatosis, ascites, and other image-quality limitations when reported (23,40,41).

2.11. Data synthesis

Because of heterogeneity in study design, reference standards, and reporting methods, a narrative synthesis was planned. Data were organized into thematic groups, including:

  • studies strictly limited to lesions ≤20 mm/≤2 cm (1,2,4,7,10,15,17,22) versus studies requiring subgroup extraction (11,12,16,18,19,31,32,37,39);
  • CEUS LI-RADS-based studies (1,3,4,16,17,18,19,22,32,33,37) versus studies using enhancement-pattern criteria such as APHE and washout without a formal classification system (7,10,14,15,29,31);
  • comparisons of CEUS with CECT/MRI (1,12,14) and the problem-solving role of CEUS after inconclusive imaging (2,8,39), interpreted in the context of current international guidelines (20,21,46,47);
  • studies that did or did not report data on BMI, obesity, and/or ascites (23).

Results were presented descriptively, indicating the range of reported metrics and the conditions under which CEUS achieved its highest clinical utility, such as high specificity in lesions with typical enhancement features. Missing evidence in BMI and ascites subgroups was highlighted.

2.12. Assessment of study quality and limitations

A formal risk-of-bias assessment such as QUADAS-2 is not mandatory for study inclusion in a scoping review and was therefore not used to exclude studies. Potential sources of bias and other limitations—including composite reference standards, retrospective design, selective histopathological verification, variable duration of washout observation, and incomplete reporting of BMI and ascites—were documented descriptively during synthesis and considered in relation to previous systematic reviews (28,43,44,45).

3. Results

3.1. Record identification and study selection

The multi-database search of Scopus, PubMed, ScienceDirect, and Google Scholar, together with the supplementary AI-assisted OpenEvidence search, identified a mixed pool of publications comprising:

  1. original studies evaluating CEUS in small nodules ≤20 mm/≤2 cm (1,2,7,10,15,17,22);
  2. studies validating CEUS LI-RADS, often without a size restriction but potentially including relevant subgroup analyses (3,16,18,19,32,33,37);
  3. studies comparing CEUS with CECT/MRI (12,14,30,34,35,36); and
  4. guidelines and systematic reviews used as background evidence (20,21,28,43,44,45,46,47).

During title and abstract screening, records were excluded because CEUS was not the index test—for example, strictly CT- or MRI-based studies (6)—because no quantitative data were available (25,26), because the lesion-size range extended beyond ≤2 cm without an extractable subgroup, or because CEUS was used solely to guide biopsy or ablation (9).

The screening table, comprising batches 1–3, identified a core set of potentially eligible records for the primary table of lesions ≤2 cm/≤20 mm. Records marked INCLUDE were retained, whereas UNCLEAR records (11,12,16,18,19,23,29,30,31,32,34,35,36,37,39) were reserved for full-text verification to determine whether: (a) a ≤2 cm/≤20 mm subgroup was available; (b) diagnostic metrics were reported for that subgroup; and (c) the reference standard was histopathology, CECT/MRI, or follow-up.

3.2. Characteristics of the evidence: study types, populations, and reference standards

The provisionally included studies comprised:

  • diagnostic studies of 1–2 cm or ≤20 mm nodules in at-risk populations, frequently including patients with cirrhosis (2,7,15,17,24), with final diagnoses based on histopathology, CECT/MRI, follow-up, or a combination thereof;
  • studies comparing CEUS with MRI, including protocols using hepatocyte-specific contrast agents (1,14), or with CECT (12,14);
  • studies based on CEUS LI-RADS categories such as LR-5 and LR-M, reporting sensitivity, specificity, PPV, accuracy, or AUC (3,17,18,33,44,45).

A dominant methodological feature was heterogeneity of the reference standard. Some studies relied on histopathology (3), whereas others used combinations of CT, MRI, and follow-up (8,39). This heterogeneity must be considered when interpreting sensitivity, specificity, and the risk of bias.

3.3. Principal findings: CEUS performance in small nodules (≤2 cm / ≤20 mm)

Based on data available from abstracts and preliminary descriptions, subject to confirmation in the full texts, the evidence indicates that:

  • CEUS may achieve high specificity for diagnosing HCC in small nodules, particularly when typical features such as APHE and late, mild washout are present (2,15,29) and standardized criteria such as CEUS LI-RADS are applied (1,3,17,18,37).
  • Sensitivity is often moderate and depends on lesion size—particularly ≤10 mm versus 10–20 mm (10)—deep or subdiaphragmatic location, acoustic-window quality (42,48), and the interpretation criteria used, ranging from historical arterial-phase-only approaches (15) to modern LI-RADS assessment incorporating washout (17,37).
  • Some studies report very good diagnostic metrics when CEUS is used to evaluate an already detected FLL as a problem-solving tool (2,8,39), but these results cannot be directly extrapolated to primary screening or surveillance performance (5,13,38).

3.4. CEUS versus CECT/MRI and histopathology: comparisons and clinical context

In comparative studies, CEUS was evaluated as:

  1. a parallel modality, with each patient undergoing CEUS and CECT/MRI (1,12,14,35,36);
  2. a supplementary problem-solving modality after inconclusive CECT/MRI (2,8,39); or
  3. a component of an expedited diagnostic pathway after detection of a nodule on B-mode ultrasonography (24).

From a practical perspective, CEUS offers advantages in availability and safety because it is not nephrotoxic, whereas CECT and MRI generally provide more stable assessment conditions independent of the acoustic window. In studies using histopathology as the reference standard, CEUS yielded categories concordant with the final diagnosis in a substantial proportion of lesions (3). Nevertheless, the degree of agreement and the sensitivity–specificity trade-off depended on the diagnostic threshold, such as a restrictive LR-5 category versus broader criteria (3,17,22).

3.5. Did the literature clarify the diagnostic performance of CEUS in obesity and ascites?

At the screening stage, the available records only very rarely reported CEUS diagnostic metrics stratified by:

  • BMI or obesity, including BMI ≥30; and/or
  • the presence or grade of ascites.

Most sources referred to these factors indirectly as determinants of general ultrasound quality, including the acoustic window, penetration, imaging depth, and respiratory motion (23,40,41,42,48), rather than analyzing them quantitatively. Thus, the current evidence gap is not merely whether CEUS works in these groups, but how its sensitivity and specificity change in patients with obesity or grade II–III ascites who have small nodules ≤2 cm.

DISCUSSION

4.1. Role of CEUS in diagnosing small HCC lesions (<2 cm): comparison with reference diagnostics and position in the algorithm

4.1.1. What is actually measured in CEUS studies of lesions <2 cm?

CEUS uses intravenously administered microbubble contrast agents, typically 2–5 µm in diameter (42,48). These agents remain exclusively within the vascular compartment as pure blood-pool agents (25,26), enabling continuous real-time assessment of microvascular perfusion. Cirrhosis and multistep hepatocarcinogenesis involve progressive vascular remodeling, with gradual loss of portal venous supply and pathological arterial neoangiogenesis within the evolving nodule. This process favors intense arterial phase hyperenhancement (APHE) (10,15,31). The timing and degree of washout in later phases are key differentiating features (14), although their interpretation depends on the criteria applied—such as the standardized CEUS LI-RADS system (17,37,44)—and on the contrast agent used.

A clear distinction exists between pure blood-pool agents such as SonoVue (15,26,29) and newer agents with a Kupffer phase, such as perfluorobutane-based Sonazoid (1,13,35,36). In the literature, CEUS is most often evaluated as a problem-solving technique for characterizing previously detected FLLs or resolving inconclusive CECT/MRI findings (2,8,39), rather than as a primary screening method, for which conventional B-mode ultrasonography remains the foundation (20,46). For lesions <2 cm, it is methodologically essential to distinguish screening detection (5,38), CEUS characterization (10,14,15), and establishment of the final diagnosis of HCC versus non-HCC (3,33). CEUS also inherits the physical limitations of B-mode ultrasonography related to acoustic-window quality, including high BMI, obesity, and advanced ascites (23,40,41,42,48). These limitations may affect both the detection of a small lesion and the reliability of its dynamic-phase interpretation.

4.1.2. CEUS and MRI/CECT as benchmarks: complementarity rather than competition

CEUS, particularly when performed at the bedside, offers high availability, repeatability, and a favorable safety profile (26). Nevertheless, CECT and MRI remain the principal benchmarks for diagnostic accuracy in small lesions suspicious for HCC measuring <2 cm (1,12,14). CEUS should not be portrayed as a direct competitor to MRI or CECT in primary diagnosis, because most leading professional societies position it primarily as a problem-solving modality, particularly following an inconclusive CECT or MRI examination (20,21,46,47). Because CEUS shares the acoustic-window dependence of conventional ultrasound and cannot comprehensively evaluate the entire liver at one time (42,48), it does not replace cross-sectional imaging for mapping multifocal disease, global assessment, or complete staging. This is particularly relevant in patients with obesity and/or ascites, in whom imaging depth increases and acoustic-window stability deteriorates. These factors may limit the reliability of CEUS-based staging, while dedicated evidence for these specific clinical circumstances remains scarce (23).

Accordingly, CEUS occupies a complementary position in diagnostic algorithms by resolving uncertainty regarding small FLLs (2,8,39). MRI is relatively independent of acoustic-window quality and enables better assessment of subdiaphragmatic and deeply located lesions, as well as comprehensive evaluation of multifocality and ancillary features such as intralesional fat or a tumor capsule (1,14). CEUS may provide highly useful real-time clarification when MRI or CECT findings are inconclusive (2). In selected clinical settings, Sonazoid-enhanced CEUS and assessment of a Kupffer-phase defect may provide diagnostically valuable information approaching the performance of MRI for small nodules, provided that the examination is performed in an experienced center (1,32,34).

CECT remains a practical benchmark because it is widely available and permits rapid assessment of the entire liver (6,12). However, it requires iodinated contrast, exposes patients to ionizing radiation, and is highly sensitive to the precise timing of dynamic phases when small lesions are evaluated. In some multicenter studies, CECT forms part of a composite reference standard together with MRI and/or histopathology, which complicates interpretation of its isolated value as an independent benchmark (18,37,39). By contrast, CEUS eliminates the risk of contrast-induced nephrotoxicity (26,42).

Studies specifically addressing small nodules demonstrate the typical diagnostic-threshold trade-off. Huang et al. (2020) (17), evaluating CEUS LI-RADS for nodules ≤20 mm, showed that the restrictive LR-5 category maximizes specificity as a rule-in criterion at the cost of sensitivity, particularly in lesions close to the size threshold. Yang et al. (2024) (3), using histopathology as the reference standard for nodules <20 mm, confirmed that even with definitive tissue verification, the sensitivity of strict LR-5 criteria may be limited to approximately 50%, despite a high PPV. Giangregorio et al. (2023) (7) suggested that technical modifications such as high-frame-rate CEUS (HiFR-CEUS) may substantially improve characterization of microvascular flow and vascularity in lesions <2 cm compared with conventional protocols. Reports by Jang et al. (2009) (15) and Quaia et al. (2013) (14) place these findings within the evolution of interpretation criteria—from earlier models focused largely on the arterial phase alone (15) to contemporary, rigorous algorithms incorporating the timing and degree of washout (17,37) and the relevance of the Kupffer phase (1,32,34).

4.1.3. Reference standards in CEUS studies of small lesions (<2 cm): why they complicate interpretation

Histopathology is intuitively regarded as the strongest reference standard for diagnosing HCC because it provides direct tissue confirmation (3). In clinical practice, however, its role is substantially limited for small lesions (<2 cm). First, biopsy may be technically infeasible or associated with a considerable risk of complications (9), particularly in deep or subdiaphragmatic lesions, lesions adjacent to large vessels, or lesions in segments that cannot be accessed safely (42,48). Consequently, some patients are not referred for biopsy, whereas other studies employ histopathological verification selectively (18,37).

Second, even when tissue is obtained, small lesions are susceptible to sampling error if the needle does not target a representative area (9). This is especially important during early hepatocarcinogenesis, when regions with different degrees of dysplasia and early malignant transformation may coexist within a single lesion (31). Distinguishing a high-grade dysplastic nodule from early or small HCC may therefore be difficult not only with imaging algorithms but also histologically, particularly when the biopsy sample is scant (16,27).

For these reasons, many key studies comparing CEUS with MRI and/or CECT do not rely exclusively on histopathology. Instead, they use a composite reference standard in which the final diagnosis is established through a multistep algorithm (1,12,14). In such pathways, CECT/MRI findings form the basis of diagnosis and are supplemented, in selected cases, by histopathology and/or long-term follow-up (2,8,39). This approach better reflects the real clinical pathway and permits analysis of substantially larger cohorts (18,37). Its disadvantage is that the reference truth carries different degrees of certainty depending on whether the final diagnosis rests on biopsy, stringent imaging criteria, or morphological stability over time (11,12).

This gives rise to classic sources of bias that may distort reported CEUS performance independently of the technology’s true capabilities (28,43):

  1. Selection or spectrum bias: lesions selected for histopathological verification do not represent a random sample. They more often comprise technically accessible or diagnostically ambiguous cases (2,39). A pathologically proven cohort may therefore overrepresent difficult cases, limiting extrapolation to the typical surveillance population (3,17).
  2. Partial verification bias: different lesions within the same study are verified using different methods—histopathology, CT/MRI, or follow-up—and the chosen reference pathway often depends on the initial suspicion of malignancy. Pooled sensitivity and specificity are consequently calculated against references of nonuniform methodological quality (11).
  3. Incorporation bias: when the final diagnosis within a composite standard depends heavily on CT/MRI, CEUS is effectively evaluated against an HCC definition derived from another imaging modality with different physics, vascular dynamics, and contrast-distribution mechanisms (12,14). This impedes a fully symmetrical comparison of CEUS with CT/MRI (35,36). In PFB-CEUS, assessment of a Kupffer-phase defect adds a unique feature without a direct equivalent in conventional CECT or standard non-hepatocyte-specific MRI protocols (1,30,34).

CEUS diagnostic metrics in small lesions are therefore strongly dependent on the reference standard. A histopathology-only model increases the internal certainty of diagnosis (3) but usually reduces representativeness and may overrepresent difficult cases (17). A composite standard more closely reflects real-life practice but introduces heterogeneity and bias that must be clearly communicated (11,12). Accordingly, when CEUS LI-RADS LR-5 is used as a high-specificity rule-in category at the cost of sensitivity (17,18,37,44), reported sensitivity and specificity must always be interpreted in relation to the method used to establish the final diagnosis.

In summary, CEUS performs well in assessing small lesions <2 cm in patients with cirrhosis, but its apparent sensitivity and specificity depend heavily on how the clinical truth is defined. Histopathology-only models provide high diagnostic certainty but tend to include difficult cases (3), whereas composite standards based on CT, MRI, and follow-up better reflect clinical practice but introduce greater variability and a higher risk of bias (1,14,39). Therefore, particularly when interpreting CEUS LI-RADS LR-5 as a high-specificity rule-in category with reduced sensitivity (17,37,45), diagnostic metrics must always be considered in the context of the reference standard, which substantially determines how well CEUS appears to perform.

4.1.4. Clinical relevance: when CEUS changes decisions in small nodules

The principal clinical value of CEUS lies in real-time assessment of vascular perfusion. Compared with contrast-enhanced CECT and MRI, it is non-nephrotoxic, usually more readily available, and may reduce diagnostic costs in patients with suspected HCC (26). This is especially relevant within a problem-solving pathway after lesion detection on conventional ultrasonography (2,8,39). Diagnostic utility nonetheless requires technical feasibility and an adequate acoustic window. Common obstacles include limited patient cooperation, access to contrast agents, marked obesity, and advanced ascites (23,40,41). In this respect, CEUS shares some of the physical limitations of conventional ultrasonography (42,48).

CEUS may substantially shorten the pathway after an FLL is detected on B-mode ultrasonography by allowing immediate assessment of APHE and washout and, in some patients, earlier and more definitive lesion categorization (24). It may also reduce indeterminate results by enabling a categorical upgrade in equivocal cases (2), including after difficult-to-interpret MRI or CT findings. This helps determine the optimal next step: urgent MRI/CT, targeted biopsy, or long-term imaging follow-up (20,21,46,47). Its role as a precise, real-time problem-solving technique (2,8,39) may therefore enable immediate clarification of a small lesion and optimize care in high-risk patients.

4.1.5. Position of CEUS within the HCC diagnostic algorithm

CEUS is used primarily to characterize a detected FLL rather than as the principal screening method for HCC (2,8,39). It most often serves as a second-line examination used to resolve inconclusive CECT/MRI results (20,46) or support clinical decision-making within CEUS LI-RADS and subsequent diagnostic and therapeutic pathways (1,17,37).

Lesions measuring <20 mm require more cautious interpretation because they less frequently exhibit the full phenotype of typical enhancement features and may display subtler abnormalities, leading to lower sensitivity in this subgroup (10,15). As lesion size decreases, technical assessment becomes increasingly important, including acoustic-window quality, stability of arterial-phase imaging (42,48), equipment capability, operator experience, and the expertise of the center (21,47).

4.2. Quality of evidence and sources of heterogeneity

Evidence on CEUS for detecting and characterizing small nodules in cirrhosis (≤2 cm/≤20 mm) is dispersed because the available studies differ simultaneously across several domains that directly affect observed sensitivity and specificity (28,43).

First, the at-risk populations are heterogeneous (44,45). Some studies examine patients in conventional surveillance programs (5,13), others evaluate a direct diagnostic pathway after a nodule is identified on B-mode ultrasound (15,24), and still others investigate CEUS solely as a problem-solving technique after inconclusive CT/MRI (2,8,39). These settings involve different lesion spectra, HCC prevalence, etiologies, and stages of cirrhosis, generating spectrum bias and markedly altering PPV and NPV (11,28).

Second, definitions of a small FLL are inconsistent. Some authors use thresholds of ≤2 cm or ≤3 cm (12,14), whereas others apply <20 mm or ≤20 mm (1,3,17,22). Studies do not consistently separate subcentimeter lesions from those measuring 10–20 mm (10), despite the importance of this distinction for subtle or atypical vascular-enhancement features (10,31).

Third, CEUS protocols differ substantially, particularly in observation duration and interpretation of washout timing (29). The contrast agent is crucial: conventional pure blood-pool sulfur hexafluoride (SF6) agents differ from perfluorobutane (PFB) agents with a Kupffer phase. Thus, “CEUS” is not a single, fully standardized and comparable diagnostic intervention across centers worldwide (30,34,35,36).

Fourth, reference standards vary. Strict histopathological models (3) contrast with composite standards in which the reference truth is defined by combinations of CT/MRI and follow-up (12,14,18,37). This creates considerable heterogeneity and risks of partial verification and incorporation bias (11,12), especially for nodules <2 cm, for which biopsy is selective and prone to sampling error (9).

Finally, diagnostic thresholds differ. A rigorous CEUS LI-RADS approach using LR-5 as a rule-in criterion (17,18,37,44) addresses a different clinical question from less formalized criteria or historical models based on isolated arterial-phase assessment (15). Differences in reported metrics may therefore reflect both the method’s intrinsic performance and the chosen definition of a positive test (17,22).

Consequently, direct comparison of sensitivity and specificity across studies—and especially categorical claims that CEUS is superior or inferior to CT/MRI—requires careful specification of the population, lesion size, CEUS protocol, reference standard, and interpretive threshold (20,46,47).

4.3. Evidence gaps: obesity (BMI ≥30) and grade II–III ascites

Obesity as a physical barrier to CEUS is rarely investigated as a separate issue in studies dedicated to small FLLs, particularly using diagnostic metrics stratified by BMI (23). Obesity does not directly alter the physiological delivery of microbubble contrast into the hepatic circulation or the biological tissue-enhancement pattern (25,26). It may, however, create a major technical barrier to scanning by increasing imaging depth and ultrasound-beam attenuation and backscatter, thereby markedly reducing the signal-to-noise ratio (SNR) (42,48). The greatest detection and interpretation difficulties occur in subdiaphragmatic lesions in superior liver segments and small, deeply located nodules far from the transducer (23,41).

The increased thickness of adipose tissue frequently necessitates continuous adjustment of transducer position and scanning approach. This impairs stable maintenance of the region of interest (ROI) during the contrast bolus and subsequent dynamic phases (23,40). A particular problem in CEUS of small nodules is failure to capture the brief arterial phase, which may last only several seconds but is essential for demonstrating APHE (missed APHE) (15,29). Difficulty maintaining the ROI and increased motion artifacts may therefore impair image quality in a manner particularly detrimental to CEUS-based algorithms (41).

In everyday practice, patients with high BMI can consequently be expected to have:

  1. a higher proportion of examinations that are suboptimal or completely nondiagnostic for microvascular assessment (40);
  2. more intermediate or equivocal results corresponding to CEUS LI-RADS LR-3 and LR-4, at the expense of definitive LR-5 rule-in results (17,37);
  3. a greater need to redirect the diagnostic pathway toward MRI or CECT, which are less affected by abdominal-wall thickness (1,14).

Mild grade I ascites, reflecting portal hypertension in advanced cirrhosis, does not necessarily impair image quality. EFSUMB guidelines note that a small volume of intraperitoneal fluid may improve scanning conditions by providing a natural acoustic window and displacing bowel loops that generate gas-related reverberation artifacts (42,48). By contrast, grade II–III ascites can substantially displace the liver, increase the distance between the ROI and transducer, and impair imaging stability. Coexisting dyspnea at rest and tachypnea may increase motion artifacts, making it difficult to maintain a constant section through the nodule during both the arterial and late phases (23,42).

The literature shows a major evidence gap: ascites has not been analyzed quantitatively in relation to formal visualization-quality scores (41), and no studies have directly evaluated the effect of intraperitoneal fluid on CEUS sensitivity and specificity for lesions ≤20 mm. Available evidence therefore does not permit assessment of effect modification by BMI and/or ascites, and pooled overall findings may not apply fully to patients with poor visualization (28,43). Future prospective studies should report at least BMI, the presence and grade of ascites, a formal visualization-quality score, and CEUS diagnostic metrics stratified for lesions ≤10 mm and 10–20 mm (10,41).

4.4. Integrated diagnostic algorithm for small nodules (≤20 mm): synergy among CEUS, MRI/CECT, and clinical data

Integrated assessment of a small FLL (≤20 mm) may improve practical diagnostic effectiveness by achieving a more appropriate balance between sensitivity and specificity while permitting assessment of multifocal disease and stage through cross-sectional imaging (1,14,22). Within this framework, CEUS can be regarded as a problem-solving examination immediately after lesion detection on conventional B-mode ultrasound. It may provide rapid clarification at the point of care, even when nodules are approximately 2 cm or smaller, thereby minimizing the diagnostic interval (2,24,39).

MRI and CECT serve as selective second-line modalities, particularly when multifocality is suspected, the CEUS result is indeterminate or difficult to interpret (e.g., LR-3 or LR-4), precise vascular assessment or staging is required, the acoustic window is inadequate because of advanced ascites or marked obesity, or the lesion is located in a difficult subdiaphragmatic region (8,20,21,46,47).

The literature also supports a complementary model in which CEUS resolves some FLLs, whereas MRI/CECT verifies the remainder and provides a global reference assessment of the liver (12,14). CEUS may have unique value by enabling categorical upgrading after previously inconclusive CT or MRI, when dynamic real-time assessment of capillary perfusion is decisive (2,8,39). Conversely, MRI and CECT can complete the evaluation when CEUS remains inconclusive (1,12). They provide information that CEUS cannot reliably deliver in a comprehensive manner, including whole-liver assessment, detection of satellite lesions, evidence of major vascular invasion, and complete staging (42,48). Rather than positioning CEUS and MRI/CECT as competing modalities, it is more clinically appropriate to consider them complementary components of a sequential and integrated algorithm for nodules ≤20 mm (20,22,46).

4.5. Predictive value of CEUS in small HCC lesions: limited prognostic evidence but substantial clinical utility

Most available CEUS evidence in HCC concerns diagnosis and lesion characterization rather than prognosis, long-term disease course, or treatment response (28,43). In theory, CEUS perfusion features may reflect tumor biology through associations between a specific vascular phenotype and histopathological characteristics such as tumor grade or fibrosis stage (16). However, evidence linking these features to tumor aggressiveness is limited, heterogeneous, and rarely reported in a manner that permits firm conclusions for small lesions <2 cm in cirrhosis. Furthermore, obesity, a limited acoustic window, ascites, and deep or subdiaphragmatic lesion location (23,41,42,48) may reduce confidence in enhancement-pattern assessment and further limit prognostic generalization.

CEUS nevertheless has an established role in the separate setting of monitoring local treatment effectiveness, such as identifying residual perfusion after thermal ablation or transarterial therapy (9). These studies involve different populations and endpoints from the primary diagnosis of small nodules. Predictive applications should therefore be regarded chiefly as clinical context and a potential direction for further research (22), whereas the principal conclusions of this review concern the diagnostic value of CEUS for detecting and classifying small FLLs in patients with cirrhosis (17,37).

CEUS should not be presented as a predictive tool sensu stricto, because most evidence addresses diagnosis rather than prognosis (44,45). It does, however, have clear and documented clinical value as a precise problem-solving and complementary diagnostic method (2,8,39). Real-time assessment of vascular enhancement permits more reliable characterization of detected FLLs (14,15), may reduce the number of indeterminate diagnoses, and supports decisions regarding urgent CECT/MRI, targeted biopsy, or imaging follow-up (20,21,46,47). CEUS therefore optimizes the pathway not through prognostication but through better diagnostic stratification and sequencing of tests, particularly when the aim is rapid and safe characterization of small lesions in patients with cirrhosis (2,24).

4.6. Clinical application of CEUS in the diagnosis of small HCC lesions below 20 mm in routine practice

CEUS can be performed at the bedside as a natural extension of conventional ultrasonography in patients with cirrhosis (26). It simultaneously shares the fundamental physical limitations of B-mode imaging, particularly dependence on acoustic-window quality (42,48). Although the quality and completeness of ultrasound surveillance vary in routine practice (40), clinicians must continue to actively search for FLLs potentially representing malignancy and initiate appropriate management after diagnosis (20,46). In small FLLs, particularly those <20 mm or ≤20 mm, CEUS evaluation using either pure blood-pool SF6 contrast or PFB contrast with a Kupffer phase (35,36) may be technically demanding because of the brief arterial-phase diagnostic window (15,29) and ultrasound-quality limitations in clinically complex patients (23,41).

In cirrhosis, CEUS functions primarily as a precise problem-solving tool (2,8,39). It is useful when an FLL suspicious for HCC is clearly visible on conventional ultrasound, enabling stable maintenance of the ROI and reliable evaluation of enhancement dynamics (24). Typical HCC features—APHE followed by late, mild washout—are particularly important (14,29). In such cases, CEUS may rapidly confirm the lesion’s nature and reduce the need for further escalation of imaging in at-risk patients (17,37). Studies specifically concerning nodules ≤20 mm/≤2 cm show that standardized criteria, including CEUS LI-RADS-based approaches, can provide high diagnostic specificity and potentially reduce additional testing when a typical vascular pattern is present (1,3,7,10,14,15,17,22).

CEUS provides real-time perfusion information that may immediately establish whether an enhancement pattern is typical or atypical and whether escalation to MRI/CECT or targeted biopsy is required (2,20,46). In everyday practice, the cost and organizational burden of diagnosing small lesions often arise from repeated short-interval MRI/CECT follow-up of indeterminate lesions and the need for biopsy (9,12,14,39). CEUS may support other modalities and reduce the number of diagnostic steps by increasing certainty in typical cases, but it should not be considered a universal replacement for MRI or CT (21,47).

CEUS is also useful when CT/MRI contrast is contraindicated, when MRI is restricted by selected implants, or when transporting a critically ill patient for cross-sectional imaging would impose a substantial clinical burden (26,42). SF6- and PFB-based CEUS agents are not typically nephrotoxic and are safe in patients with renal impairment (26,48).

Limitations include those shared with conventional ultrasound, particularly obesity, steatosis, and grade II–III ascites (23,40). The reviewed literature rarely reported results stratified by BMI or the presence and grade of ascites, which represents a major evidence gap (28,43). CEUS can strongly support the diagnosis and characterization of an individual lesion (2) but does not replace complete MRI/CECT staging, which is essential for assessing multifocality and disease extent (1,12,14). Small lesions may also more often display atypical patterns, including less pronounced APHE, subtle washout, or a narrow observation window (10,31), increasing the proportion requiring additional evaluation (17,37).

A practical approach requires first determining whether image quality permits stable maintenance of the ROI and reliable evaluation of enhancement (23,41). If not, immediate escalation to cross-sectional imaging remains necessary (20,46). Interpretation of diagnostic metrics is additionally limited by heterogeneous reference standards—histopathology, CT/MRI, follow-up, or mixed standards—which affect the transferability of conclusions across populations and centers (1,3,11,12,14,28).

4.7. CEUS for small HCC lesions (<2 cm) in cirrhosis: evidence gaps and directions for further research

The literature on CEUS in small HCC lesions provides little precise information about patient-related factors and comorbidities such as marked obesity (BMI ≥30) and advanced ascites as impediments to CEUS, or about their direct impact on diagnosing small HCCs (23). Reference standards are mixed: some studies use histopathology (3), while others rely on CECT and MRI (12,14). This may introduce incorporation bias because the imaging test under evaluation is compared with another imaging modality that forms part of the reference standard (11,12). Many studies also fail to describe the reference hierarchy clearly—whether a diagnosis was based on histopathology, typical CECT/MRI criteria, or follow-up—or the proportion of lesions verified by each method (1,37). This prevents objective risk-of-bias assessment and reliable comparison across studies (28,43).

Patient populations are highly heterogeneous in cirrhosis etiology and baseline HCC risk (44,45). Diagnostic thresholds are inconsistent: some studies use historical criteria based predominantly on the arterial phase alone (15), while others do not clearly identify a standardized interpretation system or specific CEUS LI-RADS version (17,37). Washout assessment is also insufficiently standardized. Studies do not always specify the minimum late-phase observation time, which should extend for at least 4–6 minutes (29), report the distribution of time to washout using measures such as median and IQR, or characterize washout intensity as mild versus marked (17,32). This may produce substantial differences in lesion categorization, particularly in small lesions in which washout may be delayed (29).

Thus, although CEUS generally demonstrates substantial diagnostic value (44,45), it remains uncertain which patients benefit most, under what conditions sensitivity declines, and how the pathway should be optimized when conventional ultrasound and CEUS are likely to fail because of unfavorable technical conditions (20,46). Many studies do not report outcomes in patients with BMI >30, 35, or 40, or assess severe steatosis as a determinant of attenuation and penetration depth (23,40). Objective image-quality measures such as the US LI-RADS visualization score are also often unavailable (41). Reports frequently omit the transducer type, frequency, intercostal versus subcostal approach, and exact lesion depth (42,48). It is therefore impossible to determine whether diagnostic performance reflects the method’s intrinsic effectiveness or the technical limitations of a particular protocol and examination environment (23).

Ascites may exert a bidirectional effect. A small volume may improve imaging by providing an additional acoustic window (42,48), whereas grade II–III ascites can severely impair the examination by increasing imaging distance, reducing stability, and altering liver position (23). Yet the literature does not systematically distinguish ascites grades or analyze their impact on lesion visualization. Nor does it report whether ascites changes the proportion of suboptimal and nondiagnostic examinations, despite the practical importance of this information for planning diagnostic pathways (41).

Studies of small lesions frequently fail to distinguish nodules <10 mm from those measuring 10–20 mm, even though subcentimeter lesions may exhibit later washout and less pronounced APHE (10). Combining all lesions below 20 mm in a single group dilutes the findings and reduces clinical usefulness (1,17). An optimal research model should distinguish these two lesion populations (<10 mm and 10–20 mm) (10), report CEUS sensitivity and specificity, the proportions of LR-5, LR-M, and LR-4 categories (17,37), the timing and characteristics of APHE and late washout (29), and the proportion of unevaluable or nondiagnostic examinations in each subgroup (41).

Finally, there is a lack of pathway studies comparing real-world diagnostic strategies such as CEUS→MRI versus MRI→CEUS (1,2,12). These studies should account for B-mode visualization quality and technical factors (41). The clinically important question is not only whether CEUS works (28), but precisely where it should be positioned in the algorithm for a patient with high BMI, ascites, and a small 10–20 mm or subcentimeter lesion (22). Such evidence could reduce additional testing, shorten the time to final diagnosis, decrease indeterminate results, and rationalize biopsy while maintaining safety and accessibility (20,21,46,47).

4.8. Limitations of this scoping review and practical implications

Within this assessment of CEUS for HCC lesions up to 20 mm, the method represents an additional tool for detailed evaluation of FLLs and is frequently used as a problem-solving examination (2,8,39). Important limitations nevertheless concern the studied populations, including the absence of systematic differentiation by cirrhosis etiology (44,45) and the failure to distinguish clearly whether CEUS was performed as problem-solving after lesion detection or prior inconclusive imaging (2), or as part of routine HCC surveillance in patients with cirrhosis (5,24).

Another major methodological limitation is the use of mixed reference standards comprising definitive histopathology (3), cross-sectional imaging such as CECT and MRI (12,14), and frequently long-term follow-up (18,37,39). CEUS technology is itself heterogeneous in the microbubble agents and protocols used. Some studies administered sulfur hexafluoride (SonoVue) as a blood-pool agent (15,29), whereas many newer studies used perfluorobutane (Sonazoid) and its Kupffer phase (1,34,35,36). Older studies included in this review sometimes treated the arterial phase alone as sufficient evidence for HCC (15,27), complicating direct comparison with contemporary full dynamic assessment and standardized criteria (17,37).

Because of heterogeneous reference standards, pooled sensitivity and specificity cannot be directly compared across reports, and CEUS accuracy depends on how each study defined the diagnostic truth (11,12). Many studies present promising results (44,45), but frequently do not rigorously separate metrics for lesions ≤10 mm from those measuring 10–20 mm (10), or clearly specify the final diagnostic criteria when several reference methods are combined (1,37). These limitations preclude a reliable quantitative synthesis and justify the scoping-review rather than meta-analytic design (28,43).

Despite a meticulous multi-database search, some literature may have been missed, including studies published outside major indexed databases, non-English publications, and reports using atypical terminology such as indeterminate nodules or focal liver lesions without an explicit reference to HCC in the title or abstract.

Aggregated evidence indicates that CEUS can effectively fill a diagnostic gap in patients with incompletely characterized lesions and inconclusive MRI/CECT (2,8,39), as well as where access to cross-sectional imaging is limited (26), provided that adequate technical imaging conditions are achieved (23,41). Obesity and intraperitoneal fluid may critically restrict these conditions (40,42,48), yet the literature does not provide a reliable quantitative estimate of the reduction in sensitivity and specificity for lesions <2 cm because metrics stratified by BMI and ascites are exceptionally rare (23).

Future prospective studies should therefore include detailed BMI categories, robust assessment of the presence and grade of ascites, standardized acoustic-window quality measures such as a visualization score (41), precise lesion location and depth (23,42), and mandatory stratification into ≤10 mm and 10–20 mm groups (10). Clinically, when baseline imaging conditions are suboptimal or multifocal disease is present, MRI and CECT should generally be favored because they are far less dependent on acoustic-window quality (1,12,14,20,46).

CONCLUSIONS

CEUS is a valuable diagnostic tool for assessing small focal lesions (<2 cm) in cirrhotic livers for HCC, particularly when typical APHE and a characteristic washout pattern are observed (14,15,29). In clinical practice, it functions mainly as a problem-solving modality when diagnostic uncertainty persists after contrast-enhanced MRI or CT, or when cross-sectional findings require clarification through real-time perfusion assessment (2,8,39).

The available studies emphasize the high specificity of CEUS, particularly when standardized criteria such as CEUS LI-RADS are applied (17,18,37,44). Sensitivity in small lesions remains moderate and variable. Recurrent sources of this variability include lesion size—especially the distinction between <10 mm and 10–20 mm (10)—deep or subdiaphragmatic location (42,48), difficulty maintaining the ROI during the brief arterial phase (15,29), and the selected diagnostic threshold, such as the rule-in LR-5 category (3,17,22). The principal advantage of CEUS is real-time dynamic perfusion assessment (26), but the method remains dependent on the acoustic window, as does conventional ultrasonography (42,48).

Obesity reduces the detectability and reliability of small HCC-suspicious lesions through poorer penetration, greater organ depth, coexisting steatosis, and reduced image quality (23,40), the same mechanisms that impair unenhanced ultrasonography (41). This increases the risk of missing small lesions and incompletely assessing APHE and washout (15,29). Optimization of transducer selection, settings, and scanning strategy may improve CEUS interpretability in patients with high BMI (23), but the literature rarely reports robust diagnostic metrics for lesions <20 mm stratified by BMI (28,43). This remains one of the most important evidence gaps.

Ascites probably has a bidirectional effect. A small volume may improve acoustic conditions (42,48), whereas advanced ascites may hinder detection of small HCCs by increasing the distance from the transducer, altering liver position, and reducing examination stability (23). Most importantly, no studies directly quantify the effect of ascites grade on CEUS sensitivity and specificity for HCC <20 mm. This gap has practical implications for assessing CEUS effectiveness in diagnostically challenging populations (28,43).

CT and MRI are less dependent on the acoustic window and therefore remain essential complementary modalities when ultrasound quality is limited (1,12,14,20,46). CEUS nevertheless offers organizational and clinical advantages, including the absence of nephrotoxicity (26), bedside real-time assessment, and support for an expedited diagnostic pathway after detection of an FLL (24).

In summary, the literature clearly supports the usefulness of CEUS for small FLLs in cirrhosis, especially as a problem-solving method (2,8,39,44,45), but provides almost no evidence permitting reliable stratification by BMI or ascites grade (23). This review therefore supports standardized reporting of visualization quality using measures such as a visualization or quality score (41), together with prospective analyses in challenging subgroups, including patients with BMI >30 and grade II–III ascites. Research should also clearly separate the stages of lesion detection (5,38), contrast-enhanced characterization (10,14), and final diagnosis (3,33), because these stages are related but not equivalent.

References

  1. Li Y, Lu S, Mao S, Li Q, Zou X, Li S, et al. Contrast-Enhanced US Using Perfluorobutane for Diagnosing Small Hepatocellular Carcinoma in High-Risk Patients: Comparison with MRI LI-RADS Version 2018. Radiology. 2026;309(2):e252271. doi:10.1148/radiol.252271.
  2. Saborido B, Vilana R, Olmedo N, García-Criado A, Pérez-López J, Ayuso C. Contrast-enhanced ultrasound as a problem-solving tool after inconclusive gadoxetic acid-enhanced MRI for focal liver lesions ≤20 mm in cirrhosis. Eur Radiol. 2025;35(4):2840-2849. doi:10.1007/s00330-024-11002-w.
  3. Yang R, Lu Q, Huang J, Lu S, Jin J, Lyshchik A. Diagnostic performance of CEUS LI-RADS version 2017 for nodules ≤20 mm in patients at risk for hepatocellular carcinoma: a histopathological validation study. Ultrasound Med Biol. 2024;50(3):412-419. doi:10.1016/j.ultrasmedbio.2023.11.005.
  4. Huang J, Yang R, Huang X, Zeng K, Liu Y, Luo J, Lyshchik A, Lu Q. Feasibility of large language models for CEUS LI-RADS categorization of small liver nodules in patients at risk for hepatocellular carcinoma. Front Oncol. 2024;14:1513608. doi:10.3389/fonc.2024.1513608.
  5. Chan MV, Ridley L, Lall S, Descallar J. Perfluorobutane-enhanced ultrasound (Sonazoid) in surveillance for hepatocellular carcinoma compared with B-mode ultrasound and gadoxetic acid-enhanced MRI. J Med Imaging Radiat Oncol. 2024;68(2):167-175. doi:10.1111/1754-9485.13604.
  6. Yamaguchi T, Takahashi N, Sobajima A, Okuaki T, Tajima T. Virtual monoenergetic imaging versus low-tube-voltage CT for the detection of hypervascular hepatocellular carcinoma ≤2 cm. Abdom Radiol (NY). 2023;48(8):2580-2589. doi:10.1007/s00261-023-03912-1.
  7. Giangregorio F, de Sio I, d’Onofrio M, Iadevaia MD, del Prete A. High frame-rate contrast enhanced ultrasound (HIFR-CEUS) in the characterization of small hepatic lesions in cirrhotic patients. J Ultrasound. 2023;26(2):415-422. doi:10.1007/s40477-022-00729-w.
  8. Zeng Q, Xie S, He X, Guo Y, Wu Y, He N, Zhang L. FI-CEUS: a solution to improve the diagnostic accuracy in MRI LI-RADS-indeterminate (LR-3/4) focal liver lesions at risk for HCC. Front Oncol. 2024;13:1225116. doi:10.3389/fonc.2023.1225116.
  9. Wu M, Zhou J, Jiang T, Li S, Zou X, Li Q, et al. A Multicenter Randomized Controlled Study of Contrast-enhanced US versus US-guided Biopsy of Focal Liver Lesions. Radiology. 2022;305(2):464-473. doi:10.1148/radiol.212317.
  10. Mei Q, Yu M, Ji J. Clinical value of contrast-enhanced ultrasound in early diagnosis of small hepatocellular carcinoma (≤2 cm). World J Clin Cases. 2022;10(24):8525-8533. doi:10.12998/wjcc.v10.i24.8525.
  11. Jiang Z, Shen H, Feng Y, Chen J. Diagnostic role of contrast-enhanced ultrasonography versus conventional B-mode ultrasonography in cirrhotic patients with early hepatocellular carcinoma: a retrospective study. J Gastrointest Oncol. 2021;12(5):2245-2253. doi:10.21037/jgo-21-611.
  12. Hsiao CY, Chen TH, Chiu NC, Yeh YC, Huang PC. Contrast-enhanced ultrasound versus computed tomography and magnetic resonance imaging for diagnosis of small (≤3 cm) hepatocellular carcinoma. J Med Ultrason (2001). 2019;46(4):453-462. doi:10.1007/s10396-019-00958-z.
  13. Kudo M, Ueshima K, Osaki Y, Hirooka M, Imai Y, Aso K, et al. B-Mode Ultrasonography versus Contrast-Enhanced Ultrasonography for Surveillance of Hepatocellular Carcinoma: A Prospective Multicenter Randomized Controlled Trial. Liver Cancer. 2019;8(4):271-280. doi:10.1159/000501082.
  14. Quaia E, de Paoli L, Degrassi F, Cova MA. Enhancement pattern of hepatocellular carcinoma ≤2 cm: comparison of contrast-enhanced ultrasound, multiphasic computed tomography and gadobenate dimeglumine-enhanced magnetic resonance imaging. Eur Radiol. 2013;23(11):3087-3097. doi:10.1007/s00330-013-2917-y.
  15. Jang HJ, Kim TK, Wilson SR. Small focal liver lesions (1–2 cm) in liver cirrhosis: characterization with contrast-enhanced ultrasound. Eur J Radiol. 2009;72(3):418-424. doi:10.1016/j.ejrad.2009.03.042.
  16. Cheng W, Jin J, Yang R, Lu Q. Pathologic correlates of CEUS LI-RADS version 2017 categories: tumor grade and liver fibrosis stage in patients at risk for hepatocellular carcinoma. AJR Am J Roentgenol. 2021;217(3):653-662. doi:10.2214/AJR.20.24521.
  17. Huang JY, Li JW, Lu Q, Luo Y, Liu JB, Jiang TA, et al. Diagnostic Accuracy of CEUS LI-RADS for the Characterization of Liver Nodules 20 mm or Smaller in Patients at Risk for Hepatocellular Carcinoma. Radiology. 2020;294(2):329-339. doi:10.1148/radiol.2019191138.
  18. Terzi E, Iavarone M, Peri V, Buda A, Marisi C, d’Onofrio M, et al. Contrast ultrasound LI-RADS LR-5 identifies hepatocellular carcinoma in cirrhosis in a multicenter retrospective study of 1,006 nodules. J Hepatol. 2018;68(3):485-492. doi:10.1016/j.jhep.2017.11.014.
  19. Strobel D, Jung EM, Ziesch M, Hackel C, Schönowsky R, Sultanova A, et al. Real-life assessment of standardized contrast-enhanced ultrasound (CEUS) and CEUS algorithms (CEUS LI-RADS/ESCULAP) in hepatic nodules in cirrhotic patients: a prospective multicenter study. Eur Radiol. 2021;31(9):6504-6513. doi:10.1007/s00330-021-07872-3.
  20. Singal AG, Llovet JM, Yarchoan M, Mehta N, Heimbach JK, Dawson LA, et al. AASLD practice guidance on prevention, diagnosis, and treatment of hepatocellular carcinoma. Hepatology. 2023;78(6):1922-1965. doi:10.1092/hep.32655.
  21. National Comprehensive Cancer Network (NCCN). Hepatocellular Carcinoma. NCCN Clinical Practice Guidelines in Oncology. Version 1.2026. Plymouth Meeting: NCCN; 2026.
  22. Hu XY, Sun YK, Miao Y, Chen XL, Lu D, Zhou BY, et al. Preoperative identification of hepatocellular carcinoma from focal liver lesions ≤20 mm in high-risk patients using clinical and contrast-enhanced ultrasound features. Eur J Radiol. 2025;187:112076. doi:10.1016/j.ejrad.2025.112076.
  23. Merrill CD, Samuel A, Gupta S, et al. A novel technology for resolution of CEUS imaging problems in patients with high BMI and fatty liver. J Ultrasound Med. 2023;42(11):2603-2614. doi:10.1002/jum.16296.
  24. Giorgio A, Montesarchio L, Gatti P, et al. Contrast-enhanced ultrasound: a simple and effective tool in defining a rapid diagnostic work-up for small nodules detected in cirrhotic patients during surveillance. J Gastrointestin Liver Dis. 2016;25(2):205-211. doi:10.15403/jgld.2014.1121.252.chu.
  25. Numata K, Luo W, Morimoto M, Kondo M, Kunishi Y, Sasaki T, et al. Contrast enhanced ultrasound of hepatocellular carcinoma. World J Radiol. 2010;2(2):68-82. doi:10.4329/wjr.v2.i2.68.
  26. Lencioni R, Piscaglia F, Bolondi L. Contrast-enhanced ultrasound in the diagnosis of hepatocellular carcinoma. J Hepatol. 2008;48(5):848-857. doi:10.1016/j.jhep.2008.02.005.
  27. Fracanzani AL, Burdick L, Borzio M, Roncalli M, Bonelli N, Fanfulla R, et al. Contrast-enhanced Doppler ultrasonography in the diagnosis of hepatocellular carcinoma and premalignant lesions in patients with cirrhosis. Hepatology. 2001;34(5):1109-1114. doi:10.1053/jhep.2001.29373.
  28. Fraquelli M, Nadarevic T, Colli A, Manzotti C, Giljaca V, Miletic D, et al. Contrast-enhanced ultrasound for the diagnosis of hepatocellular carcinoma in adults with chronic liver disease. Cochrane Database Syst Rev. 2022;9(9):CD013483. doi:10.1002/CD013483.pub2.
  29. Schellhaas B, Bernatik T, Dirks K, Blank W, Görg C, Schuler A, et al. Contrast-enhanced ultrasound patterns for the non-invasive diagnosis of hepatocellular carcinoma: a prospective multicenter study in histologically proven liver lesions in a real-life setting demonstrating the benefit of extended late phase observation. Ultrasound Med Biol. 2021;47(11):3170-3180. doi:10.1016/j.ultrasmedbio.2021.07.003.
  30. Kang HJ, Lee JM, Yoon JH, Lee K, Han JK, Choi BI. Contrast-enhanced US with sulfur hexafluoride versus perfluorobutane for characterization of focal liver lesions in patients at risk for hepatocellular carcinoma: a prospective, intraindividual comparison study. Radiology. 2020;297(3):573-582. doi:10.1148/radiol.2020201382.
  31. Duan Y, Xie XY, Li Q, Mercaldo ND, Lin MX, Wang W, et al. Differentiation of regenerative nodule, dysplastic nodule, and small hepatocellular carcinoma in cirrhotic patients: a contrast-enhanced ultrasound-based multivariable model analysis. Eur Radiol. 2020;30(10):5401-5410. doi:10.1007/s00330-020-06915-x.
  32. Li W, Wang J, Zhang L, et al. Modified LI-RADS for perfluorobutane contrast-enhanced ultrasound incorporating Kupffer phase defect: diagnostic performance for hepatocellular carcinoma in patients at risk. Abdom Radiol (NY). 2022;47(12):4012-4022. doi:10.1007/s00261-022-03681-7.
  33. Vidili G, Zocco MA, Biolato M, et al. Contrast-enhanced ultrasound Liver Imaging Reporting and Data System: Lights and shadows in hepatocellular carcinoma and cholangiocellular carcinoma diagnosis. World J Gastroenterol. 2022;28(27):3488-3499. doi:10.3748/wjg.v28.i27.3488.
  34. Li S, Zou X, Li Q, et al. Comparison of contrast-enhanced ultrasound using sulfur hexafluoride versus perfluorobutane with a modified algorithm for diagnosing hepatocellular carcinoma. Ultraschall Med. 2023;44(05):e211-e220. doi:10.1055/a-1956-4221.
  35. Huang J, Yang R, Jin J, Lu Q. Head-to-head prospective comparison of Sonazoid and SonoVue contrast-enhanced ultrasound for the diagnosis of primary hepatic malignancies in patients at risk. Clin Hemorheol Microcirc. 2023;84(2):165-176. doi:10.3233/CH-231711.
  36. Guo Y, Wu Y, Zhang L, et al. Contrast-enhanced ultrasound for hepatocellular carcinoma diagnosis: an intra-individual prospective comparison of SonoVue and Sonazoid. J Ultrasound Med. 2024;43(4):689-701. doi:10.1002/jum.16410.
  37. Lyshchik A, Wessner CE, Bradigan K, et al; CEUS LI-RADS Trial Group. Contrast-enhanced ultrasound liver imaging reporting and data system: clinical validation in a prospective multinational study in North America and Europe. Hepatology. 2024;79(2):380-391. doi:10.1097/HEP.0000000000000571.
  38. Park JH, Park MS, Lee SJ, Jeong WK, Lee JY, Park MJ, et al. Contrast-enhanced US with perfluorobutane for hepatocellular carcinoma surveillance: a multicenter diagnostic trial (SCAN). Radiology. 2019;292(3):638-646. doi:10.1148/radiol.2019190183.
  39. Loiacono R, Boccatonda A, Ercolani G, Serra C. The role of CEUS in the characterization of indeterminate focal liver lesions at second-level imaging methods: INFOLIL STUDY. Intern Emerg Med. 2026;21(2):345-354. doi:10.1007/s11739-025-04210-z.
  40. Loomba R, Lim JK, Patton H, El-Serag HB. AGA Clinical Practice Update on Screening and Surveillance for Hepatocellular Carcinoma in Patients With Nonalcoholic Fatty Liver Disease: expert review. Gastroenterology. 2020;158(6):1822-1830. doi:10.1053/j.gastro.2020.01.056.
  41. King MJ, Lee KM, Rosberger S. Multicenter study of ACR ultrasound LI-RADS visualization scores on serial examinations: implications for surveillance strategies. Abdom Radiol (NY). 2023;48(1):263-270. doi:10.1007/s00261-022-03714-3.
  42. Claudon M, Dietrich CF, Choi BI, Cosgrove DO, Kudo M, Nolsøe CP, Piscaglia F, et al. Guidelines and good clinical practice recommendations for contrast enhanced ultrasound (CEUS) in the liver—update 2012: a WFUMB-EFSUMB initiative in cooperation with representatives of AFSUMB, AIUM, ASUM, FLAUS and ICUS. Ultraschall Med. 2013;34(1):11-29. doi:10.1055/s-0032-1325499.
  43. Fraquelli M, Nadarevic T, Colli A, Manzotti C, Giljaca V, Miletic D, et al. Contrast-enhanced ultrasound for the diagnosis of hepatocellular carcinoma in adults with chronic liver disease. Cochrane Database Syst Rev. 2022;9(9):CD013483. doi:10.1002/CD013483.pub2.
  44. Qin Z, Zhou Y, Ding J, Wen J, Chen Y, Zhou H, et al. Risk stratification for hepatocellular carcinoma of contrast-enhanced ultrasound Liver Imaging Reporting and Data System (LI-RADS) and the diagnostic performance of LR-5 and LR-M: a systematic review and meta-analysis. Clin Radiol. 2022;77(4):e286-e294. doi:10.1016/j.crad.2022.01.041.
  45. Peng W, Wang J, Fan W. The Value of Contrast-Enhanced Ultrasound Liver Imaging Reporting and Data System (CEUS LI-RADS) for Diagnosing Hepatocellular Carcinoma: A Systematic Review and Meta-Analysis. J Ultrasound Med. 2022;41(6):1385-1394. doi:10.1002/jum.15837.
  46. Singal AG, Llovet JM, Yarchoan M, Mehta N, Heimbach JK, Dawson LA, et al. AASLD Practice Guidance on prevention, diagnosis, and treatment of hepatocellular carcinoma. Hepatology. 2023;78(6):1922-1965. doi:10.1092/hep.32655.
  47. National Comprehensive Cancer Network (NCCN). Hepatobiliary Cancers: Hepatocellular Carcinoma. NCCN Clinical Practice Guidelines in Oncology. Version 1.2026. Plymouth Meeting: NCCN; 2026.
  48. Claudon M, Dietrich CF, Choi BI, Cosgrove DO, Kudo M, Nolsøe CP, Piscaglia F, et al. Guidelines and good clinical practice recommendations for contrast enhanced ultrasound (CEUS) in the liver—update 2012: a WFUMB-EFSUMB initiative in cooperation with representatives of AFSUMB, AIUM, ASUM, FLAUS and ICUS. Ultraschall Med. 2013;34(1):11-29. doi:10.1055/s-0032-1325499.
Total
0
Shares
Share 0
Tweet 0
Pin it 0
Share 0