# Association between CT-derived skeletal muscle and fat indices and fracture healing following operative treatment for intertrochanteric fractures: a multicenter retrospective study

## TL;DR

This study finds that lower skeletal muscle and higher fat levels in elderly patients are linked to slower healing after hip fracture surgery.

## Contribution

The study demonstrates a novel association between CT-derived muscle and fat indices and fracture healing rates in intertrochanteric fracture patients.

## Key findings

- Lower skeletal muscle index (SMI) is associated with a higher risk of delayed fracture healing.
- Higher visceral fat index (VFI) increases the risk of slower healing after hip fracture surgery.
- SMI showed better predictive power for healing rates than VFI in ROC analyses.

## Abstract

Intertrochanteric fractures account for nearly 50% of hip fractures in elderly patients and are primarily treated with internal fixation. Considerable variability in postoperative healing persists, with delayed union or nonunion prolonging immobilization and increasing complications and healthcare costs. Accurate, objective assessment of healing is essential, and the Radiographic Union Score for Hip (RUSH) offers a reliable quantitative tool. Sarcopenia and visceral adiposity are linked to poor surgical outcomes, yet their roles in fracture healing remain unclear. This study hypothesizes that low skeletal muscle index (SMI) and high fat indices are associated with delayed healing.

A total of 619 participants from two institutions were enrolled to assess the skeletal muscle index (SMI), subcutaneous fat index (SFI), visceral fat index (VFI), and the visceral-to-subcutaneous fat area ratio (VSR) at the T12 level, and subsequently categorized into control and experimental groups based on one-month postoperative RUSH scores (≥18 vs. <18). Fracture healing was quantified using RUSH scores assessed by six blinded orthopedists (ICC = 0.824 at 1 month). At the same follow-up, a higher RUSH score was interpreted as faster fracture healing, whereas a lower score indicated delayed healing. Slower healing was defined as a RUSH score of less than 18 at 1 month.

Multicenter analysis demonstrated that each unit increase in SMI was associated with a 10–29.4% reduction in the risk of slower healing (Institution 1: OR = 0.900, 95% CI: 0.859–0.942; Institution 2: OR = 0.706, 95% CI: 0.610–0.818), whereas each unit increase in VFI was associated with a 1.7–6.4% increase in risk (OR = 1.021–1.064). In both institutions, the change in RUSH score from 1 day to 1 month was positively correlated with SMI and negatively correlated with VFI, with stronger associations observed in Institution 1 (r = 0.591, p < 0.001; r = −0.438, p < 0.001). ROC analyses confirmed that SMI had better discrimination (AUC = 0.682–0.862) compared with VFI (AUC = 0.614–0.691).

Preoperative T12-level SMI and VFI were associated with the rate of fracture healing after adjustment, with lower SMI and higher VFI linked to slower healing.