Bedside Leg Ultrasound Predicts 6-Month Death Risk in Liver Cirrhosis Patients

Sarcopenia — defined by simultaneous loss of muscle mass and muscle strength — affects roughly 37–40% of patients with liver cirrhosis and independently doubles or triples mortality risk. Despite strong guideline recommendations to screen for sarcopenia, the standard imaging methods (CT or MRI at the L3 vertebral level) are costly, require radiation or complex scheduling, and are often impractical for acutely ill inpatients. This prospective single-center study evaluated whether a simple bedside ultrasound of the rectus femoris muscle (RFM) could fill that gap.

Eighty-four consecutive adults admitted to the hepatology ward of Würzburg University Hospital (May 2022–February 2024) were enrolled within 48 hours of admission. Nearly three-quarters (73.8%) had decompensated cirrhosis, predominantly alcohol-related. Within the first two days, each patient underwent RFM ultrasound (muscle thickness MT_RFM and cross-sectional area CSA_RFM, plus echogenicity by gray-scale analysis), bioelectrical impedance analysis (BIA), hand grip strength, chair rise test (CRT), timed up-and-go (TUG), and short physical performance battery (SPPB). Six-month outcomes were captured via clinic visits or telephone follow-up.

Key results: both MT_RFM and CSA_RFM — when normalized for height² — were significantly lower in patients with more advanced Child-Pugh stage (p<0.01). Height-indexed RFM measures predicted BIA-defined low muscle mass (ASMI <7/5.7 kg/m² for men/women) and low phase angle (≤4.9°) with AUROCs of 0.727–0.770, indicating good discriminative performance. Impaired physical performance (prolonged CRT and TUG times) was independently associated with reduced RFM mass (p<0.05). Higher muscle echogenicity — a proxy for fat infiltration — correlated with worse TUG and SPPB scores, suggesting ultrasound captures not just size but also muscle quality. The inter-observer intraclass correlation coefficient for MT_RFM was 0.983, confirming excellent reproducibility.

For mortality prediction, the combination of prolonged CRT (>15 s) and low MT_RFM/height² defined a sarcopenic phenotype associated with a hazard ratio of 7.19 (95% CI 2.25–22.98) for 6-month death in Cox regression and competing-risk analyses. This effect remained significant after adjustment for disease severity, underscoring RFM ultrasound's independent prognostic value beyond standard clinical scores such as Child-Pugh or MELD.

The clinical implication is substantial: unlike CT or MRI, RFM ultrasound requires only a portable linear-probe device already present on most wards, takes minutes to perform, and delivers bedside results without radiation. Adding a five-repetition chair rise test costs nothing. Together, these two assessments constitute a pragmatic, guideline-consistent sarcopenia evaluation — identifying the patients most urgently needing nutritional intervention, physical rehabilitation, or expedited transplant listing. Caveats include single-center design, moderate sample size, and the use of BIA-derived (rather than CT-derived) muscle mass as the reference standard, which may introduce bias in fluid-overloaded patients despite post-paracentesis timing.