Liver tissue is not an ideal elastic body, but rather a viscoelastic medium that exhibits both energy-storage and energy-dissipation characteristics. Shear wave speed (SWS) is primarily influenced by the shear modulus. Based on the assumptions that the tissue is homogeneous, isotropic, approximately incompressible, and purely elastic, some devices convert SWS to Young’s modulus and express liver stiffness in kPa. Thickening of the fibrous septa, collagen deposition, and remodeling of the hepatic lobule structure typically accelerate shear wave propagation; therefore, SWS or liver stiffness is relatively sensitive to fibrosis [11].

Shear-wave dispersion reflects frequency-dependent wave speed from energy dissipation. Evidence outside AILD links stiffness more closely to fibrosis, while dispersion or viscosity may also respond to necroinflammation and fluid changes [12]-[14]. In AIH, inflammation, edema, cholestasis, and obstruction may affect these outputs; dispersion is therefore not inflammation-specific.

Fibrosis staging describes scarring; activity grading describes necroinflammatory injury. METAVIR uses F0-F4 and A0-A3 [15], while Ishak uses a 0 - 6 fibrosis scale and the modified histological activity index [16]. They are not numerically interchangeable and do not capture all AILD lesions. Tables 1-4 summarize both systems.

Disease-specific histological interpretation must also consider interface hepatitis, bile-duct injury, and cholestatic features [17]-[19].

Table 1. METAVIR liver fibrosis staging system.

Table 2. Ishak liver fibrosis staging system.

Table 3. METAVIR necroinflammatory activity grading system.

Table 4. Ishak modified histological activity index for necroinflammatory grading.

Note: METAVIR and Ishak report fibrosis and inflammatory activity separately. METAVIR activity (A0-A3) is derived from interface hepatitis and lobular necrosis; Ishak mHAI sums four necroinflammatory components (0 - 18). The systems are not numerically interchangeable [15][16].

Inflammation may improve rapidly after immunosuppression, whereas collagen remodels slowly; early stiffness decreases may reflect less inflammatory swelling, not fibrosis regression. Cholestasis or obstruction may sustain stiffness. Histology remains important for discordant findings or suspected overlap despite biopsy sampling variation [17][18].

Vibration-controlled transient elastography (VCTE) uses a surface probe to generate low-frequency mechanical vibrations and tracks shear-wave propagation using pulse-echo techniques. Two meta-analyses demonstrated good overall performance for staging significant fibrosis, advanced fibrosis, and cirrhosis in patients with AIH, PBC, or PSC, although disease composition, treatment status, histological systems, and diagnostic thresholds varied substantially among the included studies [20][21]. VCTE assesses fibrosis severity within these diseases but does not establish the underlying autoimmune liver disease diagnosis.

In AIH, Mahmud et al. found that VCTE correlated with histological fibrosis stage, although probe selection and body habitus affected technical success [22]. Interpretation is also strongly influenced by treatment phase. Hartl et al. reported that liver stiffness was closely associated with inflammatory activity during the first 3 months of treatment, whereas its relationship with fibrosis became more reliable after at least 6 months of immunosuppression [23].

Accordingly, baseline elastography in AIH should not delay indicated immunosuppression or be interpreted as an unconfounded fibrosis stage. ALT, AST, and IgG should be recorded at each examination, and biochemical flares warrant cautious interpretation. Although liver stiffness may decrease after sustained biochemical remission, an early reduction more likely reflects improvement in inflammatory activity and does not by itself demonstrate collagen regression [24].

In PBC, Corpechot et al. reported VCTE cutoffs of 8.8, 10.7, and 16.9 kPa for F ≥ 2, F ≥ 3, and F4 fibrosis, respectively. An annual increase in liver stiffness exceeding 2.1 kPa was also associated with adverse outcomes, supporting the value of longitudinal rather than isolated measurements [25]. A reliable baseline examination near diagnosis and UDCA initiation may therefore support risk assessment after biliary obstruction has been excluded. Follow-up may coincide with biochemical response assessment at 6 or 12 months [5]. However, these intervals have not been validated specifically for SWDS or viscosity, and changes during UDCA treatment should not automatically be interpreted as antifibrotic effects.

In PSC, VCTE can distinguish severe fibrosis and cirrhosis and may provide prognostic information [26][27]. However, strictures, cholangitis, cholestasis, and asymmetric biliary dilatation create substantial spatial heterogeneity. A technically valid right-lobe measurement may therefore not represent other affected segments. Serial examinations should use documented, comparable sites, and the results should be interpreted together with cholangiography, biochemical findings, and the timing of biliary interventions [6].

When biliary obstruction or acute cholangitis is present, definitive fibrosis interpretation should be deferred until clinical and biochemical stabilization. Liver stiffness may decline rapidly after biliary drainage, but the available evidence does not establish a universal waiting interval before repeat assessment [28].

Across AILD subtypes, VCTE provides rapid and reproducible measurements but lacks real-time two-dimensional grayscale localization. Ascites, obesity, narrow intercostal spaces, and focal biliary abnormalities may increase the failure rate or introduce sampling bias. Results should therefore be interpreted in conjunction with liver biochemistry, immunoglobulin G, cholestatic markers, biliary imaging, and treatment status.

Shear wave elastography (SWE) includes point shear wave elastography (point SWE, pSWE) and two-dimensional shear wave elastography (two-dimensional SWE, 2D-SWE). Both techniques allow for the selection of sampling regions under conventional two-dimensional ultrasound guidance; pSWE generates SWS within a single small region, while 2D-SWE produces a larger color elastography image, enabling observation of the distribution of measured values within the region and avoidance of blood vessels, bile ducts, and artifacts. Li et al. included 51 patients with AILD; the area under the curve (AUC) for 2D-SWE in identifying METAVIR F ≥ 2 was 0.781, with a sensitivity of 83.3% and a specificity of 72.7% when using a cutoff of 9.15 kPa. Liver stiffness was found to be associated with both fibrosis and necroinflammation, suggesting that a single measurement is still subject to confounding by inflammation [29].

In 103 AIH patients, Xing et al. reported AUCs of 0.84, 0.84, and 0.94 for significant fibrosis, severe fibrosis, and cirrhosis; performance exceeded APRI and FIB-4 [30]. Sound touch elastography also discriminated fibrosis in AILD, although cutoffs were platform-dependent [31]. In 49 AIH and 41 PBC patients, pSWE showed an AIH F ≥ 2 AUC of 0.70 and acceptable PBC discrimination [32].

In PSC, B-mode guidance can avoid visible ducts and vessels, but 2D-SWE still samples only part of the liver. Serial studies should use consistent segments, depths, and protocols; additional valid sites may be reported separately when heterogeneity is relevant. Persistent discordance with biochemistry or cholangiography requires specialist assessment [6][9].

ARFI generates tissue displacement and shear waves with focused ultrasound. In 85 autoimmune or cholestatic liver disease patients, velocity correlated with Ishak stage (r = 0.510); a 2.04 m/s cutoff for Ishak ≥ 5 yielded an AUC of 0.872, 90.0% sensitivity, and 74.7% specificity [33]. The mixed cohort and few advanced cases limit disease-specific thresholds. ARFI is an excitation mechanism, not a viscosity measure.

Shear-wave dispersion imaging reports SWDS or model-derived viscosity. Evidence outside AILD suggests that dispersion may provide complementary tissue information, but these findings cannot be directly extrapolated to AILD [34]. Direct evidence in AILD remains limited to small and heterogeneous cohorts, with platform variation, incomplete follow-up, and limited histological validation (Table 5). Therefore, validated disease-specific thresholds for SWDS or viscosity have not yet been established.

Table 5.Direct evidence for dispersion and viscosity measurements in autoimmune liver diseases.

SWDS is the shear-wave dispersion slope, usually expressed in (m/s)/kHz; mHAI is the modified histological activity index; ALP is alkaline phosphatase. Model-derived viscosity is expressed in Pa·s and is not interchangeable with SWDS.

An increased SWDS should not be interpreted as inflammation-specific because fibrosis, fluid redistribution, and tissue architecture may also affect shear-wave dispersion [10].

In 148 AIH-PBC overlap patients, including 82 followed for more than 1 year, stiffness correlated with Scheuer stage and declined with biochemical remission, illustrating combined hepatitic and cholestatic influences. Interpretation should record immunosuppression and UDCA timing. Reduced stiffness does not prove histological resolution or justify treatment withdrawal, and AIH-PBC evidence cannot establish AIH-PSC or universal overlap thresholds [38].

Figure 1 summarizes a three-step framework that confirms acquisition quality, identifies reversible confounders, and integrates elastography with laboratory and biliary imaging findings. Within this framework, stiffness remains the principal fibrosis-related measure, whereas dispersion and viscosity are considered complementary. Stiffness is better established for fibrosis and prognosis; dispersion and viscosity remain complementary.

Figure 1. Proposed three-step framework for multiparametric ultrasound assessment in AILD.

Future multicenter prospective studies should prespecify AILD subgroups and standardize fasting, acquisition, quality control, treatment timing, and blinded histopathology when available. Serial work should separate short-term inflammatory or cholestatic effects from long-term fibrosis and outcomes, test incremental value beyond stiffness and routine laboratories, and validate results across vendors [36][37]. Thresholds should remain disease- and device-specific.