Background Previous studies suggest that handgrip strength may correlate with muscle mass in cirrhosis, but validation across diverse transplant cohorts remains limited. We sought to validate whether previously reported correlations between handgrip strength and computed tomography (CT)-derived psoas-area estimates hold in a single-center cohort of liver transplant candidates and to explore whether a bedside prediction model could be derived. Methods In this retrospective cohort study, 96 patients with end-stage liver cirrhosis undergoing transplant evaluation at University Hospital Essen (January 2020–December 2024) underwent standardized handgrip strength, 4-meter gait time, and diameter-product-derived psoas area estimates. Pearson correlation, multiple linear regression, and Cox proportional hazards analyses were performed with Benjamini–Hochberg correction for multiple comparisons. Results Handgrip strength significantly correlated with total psoas area (<InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(r = 0.35\)</EquationSource> </InlineEquation>, 95% CI 0.16–0.51, <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(p_\textrm{adj} = 0.002\)</EquationSource> </InlineEquation>). Multiple linear regression yielded a prediction model (<InlineEquation ID="IEq3"> <EquationSource Format="TEX">\(R^2 = 0.27\)</EquationSource> </InlineEquation>) achieving 72% sensitivity and 68% specificity for sarcopenia detection (AUC = 0.74). In multivariate Cox regression, each 100 mm2/m2 increase in psoas index reduced waiting list mortality by 15% (HR = 0.85, 95% CI 0.73–0.98, <InlineEquation ID="IEq4"> <EquationSource Format="TEX">\(p = 0.032\)</EquationSource> </InlineEquation>) and post-transplant mortality by 18% (HR = 0.82, 95% CI 0.71–0.95, <InlineEquation ID="IEq5"> <EquationSource Format="TEX">\(p = 0.002\)</EquationSource> </InlineEquation>). Conclusion Handgrip strength shows a modest, statistically significant association with diameter-product-derived psoas area estimates in liver transplant candidates. The derived prediction model has limited accuracy (<InlineEquation ID="IEq6"> <EquationSource Format="TEX">\(R^2 = 0.27\)</EquationSource> </InlineEquation>, AUC = 0.74) and cannot defer clinically indicated CT imaging; it may assist in identifying candidates who warrant confirmatory imaging or closer nutritional follow-up.

Handgrip strength and diameter–product–derived psoas area estimate in end–stage liver cirrhosis: an exploratory single–center analysis

期刊论文
出版模式：

开放获取
发表日期：

2026年7月29日

查看全文

Eugen Malamutmann,
Christina Neuhaus,
Lale Umutlu,
Sophia T. Schmitz,
Miroslav Jalc,
Christian G. Klein,
Jan Bednarsch,
Ulf Neumann,
Arzu Oezcelik

摘要

Background

Previous studies suggest that handgrip strength may correlate with muscle mass in cirrhosis, but validation across diverse transplant cohorts remains limited. We sought to validate whether previously reported correlations between handgrip strength and computed tomography (CT)-derived psoas-area estimates hold in a single-center cohort of liver transplant candidates and to explore whether a bedside prediction model could be derived.

Methods

In this retrospective cohort study, 96 patients with end-stage liver cirrhosis undergoing transplant evaluation at University Hospital Essen (January 2020–December 2024) underwent standardized handgrip strength, 4-meter gait time, and diameter-product-derived psoas area estimates. Pearson correlation, multiple linear regression, and Cox proportional hazards analyses were performed with Benjamini–Hochberg correction for multiple comparisons.

Results

Handgrip strength significantly correlated with total psoas area ( \(r = 0.35\) , 95% CI 0.16–0.51, \(p_\textrm{adj} = 0.002\) ). Multiple linear regression yielded a prediction model ( \(R^2 = 0.27\) ) achieving 72% sensitivity and 68% specificity for sarcopenia detection (AUC = 0.74). In multivariate Cox regression, each 100 mm²/m² increase in psoas index reduced waiting list mortality by 15% (HR = 0.85, 95% CI 0.73–0.98, \(p = 0.032\) ) and post-transplant mortality by 18% (HR = 0.82, 95% CI 0.71–0.95, \(p = 0.002\) ).

Conclusion

Handgrip strength shows a modest, statistically significant association with diameter-product-derived psoas area estimates in liver transplant candidates. The derived prediction model has limited accuracy ( \(R^2 = 0.27\) , AUC = 0.74) and cannot defer clinically indicated CT imaging; it may assist in identifying candidates who warrant confirmatory imaging or closer nutritional follow-up.