Regular Strength and Sprint Training Counteracts Bone Aging: A 10‐Year Follow‐Up in Male Masters Athletes

TL;DR

In this 10-year follow-up of 69 male masters sprinters, continued regular strength and sprint training was associated with maintained or improved tibial bone properties, whereas less-trained athletes showed declines. The authors conclude that regular intensive exercise counteracts bone aging, though the observational design limits causal inference.

Key findings

• Over 10 years, group × time interactions (raw p < 0.05) were found for distal tibia total BMC, trabecular vBMD, and compressive strength index, and for mid-tibia cortical CSA, medullary CSA, total BMC, and anterior/posterior BMC; these reflected maintained or improved bone properties in well-trained athletes and decreased properties in less-trained athletes.
• Depending on the bone variable, the difference in change favoring the well-trained group ranged from 2% to 5%. The greatest differences were distal tibia trabecular vBMD (2.8% difference; maintained in well-trained vs −3.2% in less-trained; adjusted p = 0.048) and mid-tibia posterior BMC (5.1% difference; adjusted p = 0.008).
• Other distal tibia changes: total BMC and compressive strength index were maintained in well-trained athletes but decreased by 3.5% and 5.9% in less-trained athletes, with differences of 3.1% and 5.2% favoring the well-trained group. Mid-tibia changes: cortical CSA increased +1.1% in well-trained vs −1.4% in less-trained, total BMC +0.8% vs −0.9%, and medullary CSA was maintained vs +4.9% in less-trained, with differences of 2.5%, 1.8%, and 4.2%, respectively; these were not significant after multiple-testing adjustment.
• In exploratory age-group analyses, 40–64-year-olds showed differences favoring well-trained athletes of 3.2% for distal total BMC, 3.5% for trabecular vBMD, and 5.9% for compressive strength index; 65–85-year-olds showed differences of 4.1% for mid-tibia cortical CSA, 5.3% for medullary CSA, and 6.2% for posterior BMC.

Methods

• Design: 10-year longitudinal follow-up of male masters sprinters aged 40–85 years with a long-term training background. Baseline n = 83; follow-up n = 69 (83%). At follow-up, athletes were categorized by self-reported training/competition status into well-trained (actively competing, sprint training including strength training ≥2 times/week; n = 36) and less-trained (<2 times/week, no strength training, switched to endurance training; n = 33).
• Measurements: peripheral quantitative computed tomography (pQCT) of the distal tibia (5% site) and tibial midshaft (50% site) at baseline and follow-up, including structural, strength, densitometric, and polar mass distribution parameters; muscle CSA, anthropometry, health/training questionnaires, and 60-m sprint time were also assessed. Mean follow-up was 9.8 ± 0.2 years.
• Analysis: linear mixed models with group × time interaction adjusted for age; multiple-testing correction used an effective number of 16 comparisons from 19 outcomes, giving a Meff-Šidák-adjusted alpha of 0.00317. Sensitivity analyses examined age groups 40–64 and 65–85 years and compared longitudinal changes with cross-sectional predictions.

Limitations

• Stated: the study is observational, so only associations rather than causal relationships can be reported; possible influences of genetics or diseases cannot be fully excluded; group allocation was based on self-reported physical activity, though recall bias was considered less likely in this athlete population; there were no sedentary controls, so direct comparisons with non-exercisers cannot be made; age-group analyses were exploratory/hypothesis-generating, and the sample size was not sufficient for more fine-grained age-group analyses; many bone variables were analyzed, so results are presented both raw and multiple-testing corrected.
• Stated: 15 participants could not attend the main follow-up visit; their pQCT data were later obtained during a bone examination in the same laboratory at the World Masters Indoor Championships.
• Evident from the text: follow-up grouping depended on training status at follow-up rather than at baseline, so reverse causation or selection effects cannot be excluded; for some participants, follow-up pQCT scans occurred at a different time/event than the main follow-up visit, which may introduce timing differences.