Questioning skeletal muscle index for muscle mass reduction assessment in computed tomography: Why square the height?

Background & aims

Computes Tomography (CT) scans are commonly used to assess muscle mass reduction by measuring skeletal muscle area (SMA) at the third lumbar vertebra (L3) level. The skeletal muscle index (SMI), which normalizes SMA by height squared, is the current reference standard for evaluating muscle mass reduction. However, recent studies suggest that SMI may not be the most reliable indicator due to its height dependency. This study aimed to assess the clinical relevance of different SMA-derived metrics to determine the most accurate measure of muscle mass reduction.

Methods

This retrospective study included inpatients from a tertiary hospital who underwent abdominal CT scans. Skeletal muscle area (SMA) was measured using an AI-based software. The study population was divided into four groups based on age (≤40 years and >40 years) and gender. Four SMA-derived metrics were evaluated: SMI, skeletal muscle height (SMH, calculated as SMA divided by height), and Z-scores for both SMI and SMH (z(SMI) and z(SMH)). Pearson correlations between SMA-derived metrics and age, BMI, and height were calculated, alongside regression models to determine the optimal height scaling power.

Results

The study included 674 participants (362 males, 312 females), with a median age of 61 years and body mass index (BMI) of 24.7 kg/m². BMI positively correlated with SMA, SMH, and SMI, but not with z(SMH) or z(SMI). SMH showed no significant correlation with height, except in males over 40. Linear regression for participants under 40 revealed height scaling coefficients of 1.547 (95 % CI: [0.736, 2.359]) for females and 0.982 (95 % CI: [0.047, 1.917]) for males.

Conclusion

SMH could be a more suitable metric than SMI for assessing muscle mass reduction in CT scans, particularly due to its better independence from height. We emphasize the importance of clear terminology and recommend adopting “SMH” for SMA divided by height, to differentiate it from SMI in future studies.