Swimming Can Contribute To Rebuilding Bone Strength

New health strategies possible from the first study to report the relative effects of weight-bearing and non-weight-bearing exercises on bone mechanical properties

July 13, 2003 — Bethesda, MD — Skeletal bone is living tissue and requires constant turnover, accomplished through a process that repairs damage and alters bone mass in response to needs. For most of the time, the skeleton responds to mechanical loading generated though exercise by increasing bone mass and strength.

In human studies, bone development and bone mechanical properties are indicated by bone mineral density (BMD). Physical activity with weight-bearing has been shown to improve BMD. Conversely, previous studies have reported that non-weight-bearing exercises such as swimming does not have an equivalent beneficial effect on BMD.

The BMD in people who have participated in a non-weight-bearing swimming exercise program has been shown to not be significantly different from the BMD in respective control groups. Because of ethical issues, measurements of bone development in human studies are typically limited to noninvasive radiography or serum bone marker assays. Thus, it is difficult to investigate the effects of exercise mode on bone by using these techniques.

There is difficulty in equalizing the energy expenditure or total work of different modes of exercise, which has lead to a lack of studies in this area. There have been studies that assessed the effects of either weight-bearing or non-weight-bearing exercise on bone separately, which determined that non-weight-bearing exercises, generally swimming, have a positive influence on bone strength in ovariectomized rats, osteotomized rats, and growing rats.

The published findings suggest that water sports could enhance the mechanical properties of bone under specific conditions, and direct comparison of its effects with those of weight-bearing exercise would be useful.

A New Study

Accordingly, a team from Taiwan controlled the relative training intensities of two different exercise modes in an attempt to equalize the energy expenditure between the two training programs. They hypothesized that weight-bearing exercise would lead to significantly greater levels of BMD, compared with non-weight-bearing exercise; and that non-weight-bearing exercise may have some beneficial influence on the mechanical properties of bone.

The authors of the study, “Effects of Different Exercise Modes on Mineralization, Structure, and Biomechanical Properties of Growing Bone,” are T. H. Huang, S. C. Lin, F. L. Chang, S. S. Hsieh, S. H. Liu, and R. S. Yang, all from the National Taiwan University, Taipei, Taiwan. Their findings appear in the July 2003 edition of the Journal of Applied Physiology.


To investigate the effects of exercise mode on growing bone, 29 male Wistar rats (seven weeks old) were randomly assigned to one of three groups: a running exercise group (Run, eight rats); a swimming exercise group (Swim, ten rats), or a non-exercise control group (Con, ten rats). An eight-week training session was instituted that consisted of 20–60 minutes a day of exercise, five days a week. The Run group rats were trained at progressively increasing running speeds, and weights attached to the tail of the Swim rats were progressively increased from zero to two percent of their body weight.

According to a previous study, the weight gain of growing rats varied with the intensity of the running exercise. Therefore, body weight was also used as an additional indicator to help equalize the energy expenditure and training intensities of the Run and Swim groups. All animals were weighed weekly to make sure that there was no significant difference between the two exercise groups.

At the completion of the training period, the BMD of the tibiae and femora was measured with an X-ray absorptiometer. Each entire left tibia and femur was scanned, and the images were divided into three segments: proximal quarter, distal quarter, and diaphyseal portion.

Each area was analyzed separately to capture the variation in cortical and cancellous bone density. Mechanical properties of bone tissues were measured in three-point bending in a materials testing system. The span of the two support points was 20 mm, and the deformation rate was 1 mm/min. Load deformation curves were analyzed.


Key findings of the study included:

• Body weight: During the training period, the increase in body weight of both exercise groups was less than the increase in body weight of the control group. After the third week of training, the mean body weights of the rats in the Run and Swim groups were significantly less than the weights of the control (Con) animals at each time point.

• BMD measurements. The mean BMD of the proximal tibiae in the Run group rats was significantly greater than the respective mean BMD in the Swim group. In addition, the rats in the Run group also exhibited higher mean total tibia BMD and distal femur BMD, compared with the Swim and Con groups, although these differences were not statistically significant.

• Bone weight, geometry, and biomechanical three point-bending measurements. The two exercise groups had a higher mean femoral wet weight compared with the controls. The mean water content ratio of the tibiae of the Swim animals was significantly greater than the mean water content ratio of the tibiae of the Run animals. The mean water content ratio of the femora in the Swim and Run rats was significantly greater than the water content ratio of the control animal femora.


This experiment investigated the effects of exercise with or without weight bearing on the mineral and mechanical properties of growing bone. Only the running rats exhibited a higher site-specific BMD compared with the swimming and control groups. However, both running and swimming resulted in increased long bone biomechanical properties compared with controls.

In the BMD analysis, the highest levels of BMD were found in the proximal tibia and distal femur of the Run rats. In addition, the Run animals exhibited higher levels of total tibia BMD than the animals in the Swim and Con groups, although these comparisons were not significant.

Similar to studies in the human, the higher BMD in the knee would be due to the site-specific effects of the weight-bearing exercise. Yet, the swimming exercise, because of its non-weight-bearing mode, showed the lowest BMD value in the proximal tibia and distal femur but not significantly lower than animals in the control group.


This appears to be the first study to report the relative effects of weight-bearing and non-weight-bearing exercises on bone mechanical properties. The increase in some bone mechanical properties with swimming does not mean that non-weight-bearing exercise would improve all bone mechanical properties more than weight-bearing exercise.

However, contrary to traditional attitudes, this study demonstrated that there were positive effects of non-weight-bearing exercise on some bone mechanical properties. Further studies are needed to investigate the effects of muscle contraction and systemic activation (e.g., growth hormone-insulin like growth factors pathway) on bone strength. Moreover, exercise might benefit bone mechanical properties by changing the composition of the bone (e.g., water content ratio, collagen formation) in addition to the bone mineral density.

Source: July 2003 edition of the Journal of Applied Physiology

The American Physiological Society (APS) was founded in 1887 to foster basic and applied science, much of it relating to human health. The Bethesda, MD-based Society has more than 10,000 members and publishes 3,800 articles in its 14 peer-reviewed journals every year.

1 Star2 Stars3 Stars4 Stars5 Stars (138 votes, average: 3.20 out of 5)

Leave a Reply