Rule-out Cut-off of 2D-Shear Wave Elastography for Significant Fibrosis in MASLD: Systematic Review, Meta-analysis, and External Validation.

BACKGROUND & AIMS Two-dimensional shear-wave elastography (2D-SWE) is a practical method for assessing liver fibrosis. However, an optimal cut-off value to rule out significant fibrosis (≥F2) in metabolic dysfunction-associated steatotic liver disease (MASLD) remains unclear. This study aimed to determine the evidence-based rule-out cut-off through diagnostic meta-analysis.

METHODS A systematic review was of MEDLINE, EMBASE, and Cochrane CENTRAL was performed through December 2025 for studies evaluating 2D-SWE with liver biopsy as the reference standard. Pooled sensitivity, specificity, and summary area under the receiver operating characteristic curve (sAUC) were estimated. The rule-out cut-off was defined as the highest threshold achieving pooled sensitivity ≥80%. External validation was performed using an independent clinical cohort.

RESULTS Twenty-two studies involving 3,171 adults with biopsy-proven MASLD were analyzed. The sAUC was 0.84 (95% confidence interval (CI), 0.81-0.87), with pooled sensitivity 79.6% and specificity 76.0%. The rule-out cut-off of 6.6 kPa (95% CI, 5.4-8.2) achieved sensitivity 80.0% and specificity 70.7%. At this threshold, negative predictive value exceeded 95% at prevalences of 5% and 10%. Diagnostic performance was generally preserved across subgroups, but was modestly attenuated in populations with greater metabolic burden and in studies using Supersonic ultrasound platforms. External validation (n=163) demonstrated consistent sensitivity (0.86) and specificity (0.68).

CONCLUSIONS 2D-SWE demonstrated good diagnostic accuracy for significant fibrosis in MASLD. A liver stiffness of 6.6 kPa can reliably rule out significant fibrosis. As 2D-SWE can be integrated into conventional ultrasound systems, it may facilitate broader implementation of fibrosis screening beyond specialized centers. [Clinical trial number: PROSPERO (International Prospective Register of Systematic Reviews), CRD42024616042].