Weak psoas and spine extensors potentially predispose to hip fracture

We performed a computed tomography analysis of muscle composition characteristics in hip fracture patients and non-hip fracture controls.

In total, 43 patients (9 men, 34 women) were included in the hip fracture group, matched 1 to 1 with non-hip fracture controls. Muscle cross-sectional areas were measured in axial CT scan at the body level of the 4th lumbar vertebra (L4), intervertebral disc level between the 5th lumbar vertebra and the 1st sacral vertebra (L5-S1) and just below level of the lesser trochanter (LT). Attenuation was also evaluated through the mean Hounsfield unit (HU) in these areas.

The cross-sectional area per weight (CSA/Wt, mm²/kg) of psoas muscle and extensor muscles of the spine showed a significant difference between the 2 groups at both L4 (9.7 vs. 12.4, p < 0.001 and 26.3 vs. 29.2, p = 0.025) and L5-S1 (9.6 vs. 11.5, p = 0.001 and 8.8 vs. 10.3, p = 0.041) levels. In addition, the HU of these muscles differed significantly between the 2 groups at both L4 (33.3 vs. 47.6, p < 0.001 and 13.7 vs. 30.2, p < 0.001) and L5-S1 (39.7 vs. 52.6, p < 0.001 and 3.8 vs. 15.1, p = 0.012) levels. There was no difference in abdominal wall, gluteal, or thigh compartment musculature between the groups.

Poorer quantity and quality of psoas muscle and extensor muscles of the spine rather than whole body muscles may contribute to falls and were characteristic features of the hip fracture patients in this series. These findings should be considered when recommending a preventive exercise and rehabilitation protocol.