Osteoporosis is a chronic disease that seriously affects human health and quality of life. This study is aimed at determining whether swimming had an effect on the bone mineral density (BMD) of the spine and femoral neck in postmenopausal and premenopausal osteoporosis patients. We retrieved relevant literature and analyzed data from randomized controlled trials to assess the effect of swimming on BMD in postmenopausal and premenopausal women. Relevant studies, with no language restrictions, from inception to September 2019, were retrieved from the PubMed, Cochrane, EMBASE, and EBSCO databases independently by two investigators. The keywords used for the literature search were “osteoporosis” and “swimming.” The main results included BMD and
Osteoporosis is a musculoskeletal disease characterized by decreased bone mass and destruction of bone microstructure. Osteoporosis has many causes, including age, genetic factors, hormone therapy, and long-term bed rest. From an epidemiological point of view, osteoporosis mainly occurs in postmenopausal and premenopausal women and older men aged >50 years. Osteoporotic fracture is extremely harmful to older people; a hip fracture in older people is called the “last fracture” in life. In addition, 30% of women and 20% of men aged >50 years experience fractures [
Meanwhile, studies reported that exercise intervention in osteoporosis may be a better treatment option [
However, for osteoporosis patients, the responsiveness of all organs, including bone, to external stimuli is lower than that of young people [
At present, sports can be roughly divided into weight-bearing and non-weight-bearing sports. The National Osteoporosis Foundation of the United States recommended that high- and low-intensity weight-bearing training should be carried out at the same time for skeletal load, at least 30 min a day for 5–7 days a week. Moreover, attention should be paid to the muscle target of the exercise. Strong muscles can intensify the auxiliary role of the bones. It can improve posture, reduce falls, and promote bone metabolism [
Swimming, as a sport suitable for all ages, is rapidly becoming accepted by the general population. Studies have shown that swimming can improve cardiopulmonary function, reduce blood lipid levels, and improve body’s antioxidant capacity, as well as delay aging. Previous studies have shown that swimming is a non-weight-bearing exercise and has no effect on bone mass. However, Orwoll et al. suggested that long-term adherence to swimming is beneficial to increase bone mass in older people [
While some hypothesize that swimming has an effect on the bone density of patients with osteoporosis, others think it has no effect. Thus, a meta-analysis is needed to summarize past clinical studies on swimming and osteoporosis. This study is aimed at determining whether swimming has an effect on the BMD of the spine and femoral neck in postmenopausal and premenopausal osteoporosis patients.
This meta-analysis was based on the PRISMA statement. Relevant studies, with no language restrictions, from inception to September 2019, were retrieved from the PubMed, Cochrane, EMBASE, and EBSCO databases independently by two investigators. The keywords used for the literature search were “osteoporosis” and “swimming.” The inclusion criteria were as follows: (1) clinical trials involving comparison of swimmers with inactive subjects and (2) BMD data being provided. The exclusion criteria were as follows: (1) clinical trials without control group, (2) osteoporosis was due to causes other than postmenopausal and premenopausal osteoporosis, and (3) animal studies of osteoporosis.
In this study, the quality of randomized controlled trials was assessed by two independent researchers using the Cochrane risk-of-bias tool. Quality indicators are divided into low risk, high risk, and unclear risk. The features of interest of the Cochrane manual include sequence generation, allocation sequence concealment, blinding, incomplete outcome data, selective outcome reporting, and other sources of bias. Articles are classified as high quality, medium quality, and low quality according to the following criteria: (1) If randomized sequence generation and allocation concealment are identified with high risk of bias, studies are graded as low quality. (2) When randomization and allocation concealment are considered to have low risk of bias, articles are rated as high quality. Evaluation of other features was excluded. (3) If these two criteria are not met, the literature will be rated as ambiguous.
Two investigators (Y.S. and Z.C.) extracted data from the identified article, including the study title, journal, country, design, mean age, sample size, and relevant outcomes. If the selected articles contained two or one more groups of data, only relevant data were extracted for analysis. If there are differences between the two investigators, such differences were settled through consensus.
The primary indicators were bone density in the lumbar spine, upper extremity, lower extremity, and femoral neck.
Extracted data were analyzed using RevMan 5.3.5 (Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014). The mean difference (MD) and 95% confidence interval (CI) were used for continuous variables. Cochrane’s
Our literature search retrieved a total of 423 articles. After removing duplicates, the remaining 351articles were examined. Article title, abstract, and full text were read; finally, five articles met the inclusion criteria, and the total number of participants was 263 (Figure
Flow diagram of the study selection process.
We also summarized the basic information of the five articles and presented them in Table
Characteristics of the included studies.
Study year | Journal | Country | Design | Mean age | Simple size | Relevant outcomes | ||
---|---|---|---|---|---|---|---|---|
Swimming | Placebo | Swimming | Placebo | |||||
Andreoli 2012 | U.K. | Retrospective study | 58.4 | 60.8 | 12 | 24 | Total body and regional BMD, BMC, fat mass and lean body mass | |
Czeczuk 2012 | Poland | Case control study | 50.7 | 52.1 | 18 | 18 | Body fat, BMC, BMD | |
Mohr 2015 | Germany | Randomized controlled trials | 46 | 45 | 21 | 20 | BMD, BMC, P1NP, CTx | |
Greenway 2012 | Germany | Case control study | 40.4 | 43.8 | 43 | 44 | BMD, BMC | |
Nagata 2002 | Japan | Retrospective study | 59.7 | 60.9 | 41 | 22 | BMD, height, body weight, and BMI |
DB: double-blind; P1NP: procollagen type 1 amino-terminal propeptide; CTx: carboxy-terminal crosslinking telopeptide of type I collagen; BMD: bone mineral density; BMC: bone mineral content.
The included articles were published between 2002 and 2015. All included participants were women aged >40 years. The risk assessments of all five articles are shown in Figures
Assessment of risk of bias in all included randomized controlled trials.
Assessment of risk of bias in all included randomized controlled trials.
Of the five studies, the three measured the BMD of the lumbar spine. After analyzing data of the three studies, although the overall lumbar spine density of the experimental group was significantly higher than that of the control group, we found that the data of the three studies showed medium heterogeneity [heterogeneity:
Forest plot of meta-analysis showing the effect of swimming on the bone mineral density of the lumbar spine.
Forest plot of meta-analysis showing the effect of swimming on the bone mineral density of the subgroup of the lumbar spine.
Our results suggest that swimming may have an effect on the BMD of postmenopausal swimmers if the swimming time is between 3 and 6 h, but not in premenopausal swimmers with swimming time less than 3 h. This may prove wrong the notion that swimming does not increase BMD in osteoporosis.
At present, many studies report on the effect of swimming on osteoporosis; most of which support that swimming does not improve BMD. However, some experiments have confirmed that it affects not only BMD but also the level of bone turnover markers, such as CTX (decreased bone resorption marker) [
Our results also suggest that swimming, as a fitness program, may have an effect on BMD. Although only one trial has reported biomarkers and no data can be compared, we believe that the effect of swimming on bone turnover markers cannot be underestimated. Thus, more clinical trials on the effects of swimming are needed.
We believe that the effect of swimming on osteoporosis is mainly reflected in the following aspects. First, swimming stimulates osteoblasts by inducing muscle movement and water pressure on the bone, which ultimately delay bone mass decline. Second, swimming may affect the balance of bone mass regulation by increasing the content of estrogen in the body. Studies have shown that the levels of testosterone and estradiol in the blood of swimming trainers are significantly higher than those of the control group [
Because there are differences in the experimental design among the three articles which report the data of lumbar BMD, we designed a subgroup analysis design in the experimental design, which is based on age (or menopause, i.e., premenopausal and postmenopausal groups) and exercise time (3–6 h in subgroup 1, <3 h in subgroup 2). It is not clinically reasonable to group participants according to age to explain the moderate heterogeneity, i.e., the increase of BMD decreases with age, because as we aged, the sensitivity of bones to forces decreases. It seems acceptable to explain the heterogeneity from the perspective of exercise time. Previous studies have also confirmed that bone growth and development are directly related to exercise time. Exercise can increase muscle contraction. In a proper range, as you increase the exercise time, muscle contraction will also be strengthened, so the effect of muscle on bone will also be enhanced. Exercise can also accelerate blood circulation, increase metabolic efficiency, and reduce the negative effects of obesity on bone. In proper time, increasing exercise time can enhance BMD [
Bone tissue is a hard connective tissue composed of bone cells, fibers, and matrix. A large amount of calcium salt is deposited in the mechanism, which can play a supporting role. Osteocytes responded significantly to external loading or mechanical loading. The protruding processes of the osteocytes are embedded in the bone matrix to support the load of the whole matrix [
The regulation of sex hormones on bone may be achieved by regulating the estrogen receptor of osteoblasts. In animal experiments, the osteoclast activity of castrated mice was increased [
Swimming may affect osteoporosis by means of low-intensity vibration. Under the no-load condition, low-intensity mechanical signals can promote bone formation [
Human experiments have confirmed that low-intensity vibration can promote bone mass in disabled children. It also promotes the synthesis of skeletal muscle and skeleton in osteoporotic women aged 15–20 years and can regulate bone balance in women with anorexia nervosa [
Swimming may ultimately reduce inflammatory bone loss by reducing the inflammatory state associated with obesity. Experiments have shown that obesity increases the number of osteoblasts and lymphocytes and decreases the number of myeloid cells and B cells in the bone marrow immune system [
Obesity can also lead to insulin resistance and poor glucose tolerance, leading to type 2 diabetes mellitus and ultimately to bone lesions, including decreased cortical density and destruction of bone trabecular structure. The increase in inflammation level is an important factor leading to bone loss [
When infectious or noninfectious stimuli enter the body, the body is able to resist the stimuli through soluble factors secreted by immune cells, thereby enhancing the body’s defensive response. However, these inflammatory factors, including interferon, IL, and chemokines, can affect the growth and differentiation of osteocytes. These inflammatory factors are also known as inflammatory osteoporosis mediators. The effects of inflammatory factors on osteoblasts and osteoclasts should not be underestimated. First, inflammatory factors can amplify the role of inflammation in progressive transmission and induce other cytokines, noncytokine inflammatory mediators, and proteases. These factors can stimulate osteoblasts and osteoclasts, enhance osteoclast function, and inhibit osteoblast function [
Some scholars say that inflammation may be the main cause of bone loss and can cause disability and mortality [
Swimming may increase bone density by strengthening muscles. Muscle aging plays an important role in the pathogenesis of osteoporosis. At present, the lack of muscle capacity will lead to osteoporotic fractures. Age, disease, cell aging, decreased physical activity, and decreased sex hormone synthesis are important factors affecting muscle loss [
Swimming is the best way to prevent and treat osteoporosis when taking into account side effects, but other methods are still needed to manage and treat osteoporosis. While inhibiting osteoclasts, promoting the differentiation and maturation of osteoblasts is also an effective measure to treat osteoporosis. Currently, effective measures include intermittent use of monoclonal antibodies, parathyroid hormone (PTH), and Dickkopf WNT Signaling Pathway Inhibitor 1 (DKK1). Some researchers used PTH in the experimental model of fracture arthritis, which eventually reversed bone loss and repaired local bone destruction [
The strength of this article is its finding that swimming may improve the BMD of participants. BMD plays a decisive role in osteoporosis, leading to an increased risk of osteoporosis. Unlike previous studies, the researchers used swimming as a placebo for other exercises. The findings of this paper may be used as a reference for the effect of swimming on osteoporosis. Of course, more high-quality studies will be needed in the future to further confirm this conclusion.
Although more clinical randomized controlled trials are needed to study the effect of swimming on BMD in other parts of the human body, we have preliminary evidence to show that swimming may have an effect on the lumbar vertebra density of premenopausal swimmers and that swimming may improve the BMD or the radius in these participants. This may also be a good program for the clinical prevention and treatment of osteoporosis.
Zhe Chen and Yanlin Su are co-first authors.
All authors declared that they had no conflicts of interest.
We would like to thank Editage (