Postoperative Gait Performance Following Pertrochanteric Fractures Is Influenced by the Preoperative Condition of the Gluteal Muscles

Abstract

Introduction Amid an increasing number of patients with pertrochanteric fractures, early prediction of postoperative gait potential could reduce unnecessary rehabilitation and hospitalization. The relationship between preoperative gluteal muscle condition and postoperative gait outcomes remains unclear. The gluteal muscles are crucial for mobility, and their cross-sectional area (CSA) and fatty infiltration are indicators of physical function. Preoperative computed tomography (CT) provides quantitative data on muscle CSA and density, measured in Hounsfield Units (HU). This study aimed to identify which preoperative muscle index, CSA, BMI-adjusted CSA, or muscle density, best predicts postoperative gait ability. We hypothesized that a higher adjusted CSA and muscle density in the gluteus muscles would be associated with superior gait performance. Materials and methods Preoperative assessments included radiographs and CT scans. Patients under 75, with less than four weeks of follow-up, prior contralateral hip surgery, prefracture immobility, male patients, high-energy trauma, or conditions impairing physical performance were excluded. Gait performance was evaluated four weeks postoperatively, classifying patients into two groups: the 'parallel bar group,' requiring parallel bars, and the 'walker group,' walking independently. Patients underwent CM nailing. Preoperative CT scans measured the CSA and muscle density of the gluteus maximus and medius. Measurements were taken from the non-injured side to avoid interference from the fracture. Muscle borders were manually traced, and the CSA and muscle density in HU were calculated. Results Out of 81 patients, 49 met the inclusion criteria (mean age: 87). The patients were divided into the 'parallel bar group' (n=19) and the 'walker group' (n=30) based on postoperative gait performance. No significant differences in age, weight, height, or fracture laterality were observed between groups. The mean (and standard deviation (SD)) of CSA for the gluteus maximus in the 'parallel bar group'/in the 'walker group' was 2211.8 ± 469.8 mm²/2440.0 ± 586.2 mm², respectively (p=0.15), and for the gluteus medius, it was 1751.7 ± 415.2mm²/1869.1 ± 448.3mm², respectively (p=0.36). The mean (and SD) muscle density for the gluteus maximus in the 'parallel bar group'/in the 'walker group' was 13.6 ± 12.7 HU / 20.6 ± 13.0 HU (p=0.07), and for the gluteus medius, it was 25.2 ± 8.4 HU/31.8 ± 10.1 HU, respectively (p=0.02). The ROC curve identified a 30.9 HU cut-off for gluteus medius density, with sensitivity and specificity of 60.7% and 78.9%. The mean (and SD) of BMI-adjusted CSA for the gluteus maximus in the 'parallel bar group'/ in the 'walker group' was 116.4 ± 26.8 m²/10⁶ kg/124.3 ± 29.2 m²/10⁶ kg, respectively (p=0.35), and for the gluteus medius, it was 93.3 ± 27.2 m²/10⁶ kg/95.4 ± 21.3m²/10⁶kg, respectively (p=0.78). Conclusion This study analyzed preoperative CT images of women aged 75 and older with pertrochanteric fractures, comparing gluteal muscle CSA and density with postoperative walking ability. The gluteus medius density was significantly higher in the superior gait group, with a cut-off value of 30.9 HU. However, no significant differences were found in the gluteus maximus density, CSA, or BMI-adjusted CSA. These findings partially support the hypothesis, emphasizing the importance of muscle evaluation in predicting postoperative gait performance.

Keywords: body mass index: bmi; computed tomography; cross-sectional area; density; gait; gluteal muscles; gluteus medius; pertrochanteric fractures; postoperative.

Figure 1. CT images of the gluteus maximus and medius muscles

Panel (a) shows an axial image at the level of the greater trochanter's tip, highlighting the gluteus maximus. Panel (b) displays an axial image at the level of the third sacral vertebra, representing the gluteus medius.

Figure 2. Cross-sectional area (CSA) of the gluteus medius muscle

The boundaries of the muscle were manually traced by an observer, and the area was calculated based on the outlined contours.

Figure 3. Flowchart of patient selection

This flowchart illustrates the process of selecting patients for inclusion in the study, detailing the criteria applied and the number of patients at each stage.

Figure 4. Receiver operating characteristic (ROC) curve illustrating the relationship between sensitivity and specificity

The ROC curve demonstrates the balance between sensitivity and specificity for gluteus medius density in distinguishing between the 'parallel bar group' and the 'walker group.' The area under the curve (AUC: 0.719, 95% CI: 0.568-0.869) highlights the accuracy of gluteus medius density as a predictive factor. The optimal cut-off value, as determined by Youden's index, is 30.9 HU.

Figure 5. Case presentation of a 93-year-old female patient

Panel (a) displays an axial image of the gluteus maximus, while Panel (b) shows the gluteus medius, with black arrows indicating intramuscular fat. This patient, who could walk with a cane before the injury, exhibited reduced muscle density of 9.8 HU in the gluteus maximus and 19.8 HU in the gluteus medius. Postoperatively, her walking ability was limited to the parallel bars, highlighting the significant impact of diminished muscle quality on gait performance.