To determine the effects of exercise training programs on bone mass of pre- and post- menopausal women.

A search was conducted for studies published between 1966 and December 1996 in MEDLINE. Details of the search strategy were reported. Reference lists of identified studies were examined and a handsearch was made of the journals Calcified Tissue International and Journal of Bone and Mineral Research. Studies published as abstracts or unpublished data were excluded.

Study designs of evaluations included in the review

Randomised controlled trials (RCTs) and controlled (CT) trials that assessed bone loss were included if separate analysis was carried out for pre- and post menopausal women, and for men and women and if two or more interventions were compared with each other. Studies were classified as RCT or CT according to criteria defined by the Cochrane Collaboration and presented in the review. Studies that did not report sufficient information to allow computation of study treatment effect and variance were excluded.

Specific interventions included in the review

Endurance training and strength training programs of at least 16 weeks duration were studied. Training included high and low impact exercise, weight bearing activities, flexibility exercises, walking, stepping up and down on a block, treadmill running, resistance training, high-intensity strength exercise, back strengthening exercise, high impact aerobics, weight training, running, cycle ergometry, jogging, stair climbing, and no exercise. The length of training programs ranged from 6 to 24 months. Co-interventions included hormone replacement therapy and calcium and Vitamin D supplementation.

Participants included in the review

Pre- and post menopausal women were studied. Mean ages across studies ranged from 19.9 years to 79 years.

Outcomes assessed in the review

Bone mass estimated as BMD or BMC at lumbar spine (LS) or femoral neck (FN) was assessed, with measurements undertaken using dual- photon absorptiometry or dual energy X-ray.

How were decisions on the relevance of primary studies made?

Selected articles were reviewed on the inclusion criteria by one of the authors and the decision to include or exclude subsequently discussed.

The authors did not state that they assessed validity.

If a study contained more than one experimental group then each group mean was compared with the control and treated as a single data point. If both BMC and BMD were reported only the BMD results were included in the meta-analysis. If an article reported results for both LS and FN then study treatment effects for both locations were used. Methods used to extract data were not described. Information was extracted with regards to the following: number of subjects in the training and control group; subject characteristics (mean age, baseline bone mass, calcium intake (when reported), and/or calcium supplementation; in post- menopausal women the use of hormone replacement therapy; training program characteristics (type of exercise: strength or endurance; length of program; frequency, duration and intensity (when reported) of training; measurement characteristics (location: lumbar spine or femoral neck; method: DPA or DXA; outcome: BMC or BMD).

How were the studies combined?

Overall treatment effects (OT) and 95% confidence intervals were calculated using the inverse variance weighting.

How were differences between studies investigated?

Separate analysis were conducted for RCTs and CTs, and within this grouping studies on pre and post menopausal women were analysed separately. If possible (minimum of 3 studies), additional separate analysis were conducted for strength and endurance training programs. All analyses were performed separately for the locations LS and FN. Other potential sources of heterogeneity were discussed. Results were considered significant at P < 0.05.

Sixteen RCTs (34 study treatment effects) and 9 CT (19 study treatment effects) were included in the meta-analysis.

42 treatment effects were positive, 11 were negative. 19 of the 53 treatment effects (only one of which was significantly negative) reached statistical significance. The range of study treatment effects was large ranging from - 6.6% per year to 10.2% per year.

Only the results for the RCTs are presented here. Exercise programs (both types combined) resulted in less reduction or more gain in bone mass than did the control regimes in pre-and post menopausal women. At both lumbar spine and femoral neck sites, endurance training in post-menopausal women resulted in less reduction or more gain in bone mass than did the control regimes but strength training in post-menopausal women produced no benefit compared to control.

Overall treatment effect on lumbar spine (21 treatment effects): 0.84 (95% CI: 0.53, 1.16; P < 0.05); pre-menopausal women: all exercise (5 treatment effects): 0.91 (95% CI: 0.44, 1.37; P < 0.05); post-menopausal women: all exercise (16 treatment effects): 0.79 (95% CI: 0.35, 1.22; P < 0.05). Endurance training in post-menopausal women: (11 treatment effects): 0.96 (95% CI: 0.43, 1.49; P < 0.05); strength training in post- menopausal women: (5 treatment effects): 0.44 (95% CI: - 0.32, 1.21).

Overall treatment effect on femoral neck (13 treatment effects): 0.89 (95% CI: 0.50, 1.29; P < 0.05); pre-menopausal women: all exercise (3 treatment effects): 0.90 (95% CI: 0.29, 1.50; P < 0.05); post-menopausal women: all exercise (10 treatment effects): 0.89 (95% CI: 0.36, 1.42; P < 0.05). Endurance training in post-menopausal women: (5 treatment effects): 0.90 (95% CI: 0.29, 1.51; P < 0.05); strength training in post- menopausal women: (5 treatment effects): 0.86 (95% CI: - 0.18, 1.91).

The CTs showed significant treatment effect only in post menopausal women for both LS and FN but not for pre-menopausal women (where the number of studies appears to be limited to 3).

Exercise training programs have the capability to prevent or reverse bone loss in the lumbar spine and femoral neck of pre-and post menopausal women.

The aims and inclusion criteria were defined. Details of the search strategy were reported. Methods used to select studies for inclusion were described. Some relevant information on the primary studies was presented in tabular format. The following potential sources of heterogeneity were discussed: large differences in type, duration, frequency and intensity of training; use of hormone replacement therapy, calcium and Vitamin D supplementation; differences in bone mass or age of subjects at baseline; confounding effect of weight loss; low compliance; and small sample sizes. Comment was made on the use of percentage change per year only being correct if bone density change were a function of time.

The literature search did not attempt to locate any unpublished studies. No details were given of methods used to extract data. Validity was not assessed. Although potential sources of heterogeneity were discussed, no statistical assessment of heterogeneity was undertaken. Were significant heterogeneity to exist, meta-analysis may not have been appropriate.

In view of the lack of validity and heterogeneity among studies, caution is advised when considering the conclusions.

Practice: The authors suggest that exercise training programs are worth considering in the prevention or treatment of osteoporosis.

Research: The authors consider there is a need for studies with a duration of more than 5 years in post menopausal women to evaluate the feasibility and relevance of the bone preservation by exercise.

This additional published commentary may also be of interest. Bone mass and exercise in women. Bandolier 1999;68:1-2.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.