Objectives: To assess the paraspinal and abdominal muscle activities during different therapeutic exercises and to study how load increment produced by varying limb movements and trunk positions could affect these muscle activities.

Design: A cross-sectional study comparing muscle activities between men and women.

Setting: Rehabilitation clinic in university hospital.

Participants: Twenty-four healthy volunteers (14 women, 10 men) aged 21 to 39 years.

Interventions: Subjects performed 16 different therapeutic exercises commonly used to treat low back pain.

Main outcome measures: Surface electromyography was recorded from the paraspinal (T9, L5) and abdominal (rectus abdominis, obliquus externus) muscles during these exercises. Average electromyographic amplitudes obtained during the exercises were normalized to the amplitude in maximal voluntary contraction (% MVC) to produce interindividually comparable muscle activity assessments.

Results: Mean average normalized electromyographic amplitudes (% MVC) of the exercises were below 50% MVC. At L5 level, the multifidus muscle activities were significantly higher (p <.05) in women than in men, whereas no significant difference was found at T9 level. Similarly, rectus abdominis and obliquus externus activities were significantly higher (p <.001, p <.05) in women than in men. Load increment in hands or unbalanced trunk and limb movements produced higher paraspinal and abdominal muscle activities (p <.05).

Conclusions: Simple therapeutic exercises are effective in activating both abdominal and paraspinal muscles. By changing limb and trunk positions or unbalancing trunk movements, it is possible to increase trunk muscle activities. Women were better able to activate their stabilizing trunk muscles than men; but it is also possible that men, having a much higher degree of strength on maximal contraction, only need to activate a smaller amount of that maximum to perform a similar activity.