Bone mineral density (BMD) and bone mineral content (BMC) were measured using DXA at lumbar spine and tibial diaphyses at the beginning and at the end of a 15-week training period in 151 military recruits of the Swiss army belonging to 5 different troop categories (infantry grenadiers, tank drivers, tank gunners, signalmen, and privates) who each were exposed to physical training of various intensity. At baseline, height, body mass index, and degree of physical fitness independently correlated with vertebral and tibial BMD. Over the 15 weeks of physical training BMD at tibial diaphyses increased by 2.2 +/- 0.3% at the left Leg (p = 0.0001) and by 1.1% at the right leg (p = 0.002) with differences between troop categories. At lumbar spine, BMD decreased significantly in tank drivers (-1.2 +/- 0.4%, p = 0.001) and particularly in infantry grenadiers (-2.1 +/- 0.4%) who had the most strenuous weight-bearing training, but not in other troop categories. This decrease was twice as large at the center of the vertebra than for the whole vertebra. These BMD changes were associated with increments in serum levels of osteocalcin and alkaline phosphatase activity. From the initial cohort, 48 subjects volunteered for a third investigation carried out 2 years after the end of the military training period. At this time, lumbar BMD and BMC had risen back to baseline, whereas at tibial diaphyses bone width and BMC but not BMD increased by 5.8 +/- 1.1% and 6.2 +/- 0.9%, respectively, vs. baseline (p = 0.0001 for both). We conclude that (1) bone mass appears to be determined by both anthropometric parameters and the degree of physical fitness; (2) an acute four-month physical exercise period leads to increased BMD tibial diaphyses and to increased bone turnover, resulting in a transitory bone loss at lumbar spine with complete recovery within two years; and (3) diameter of tibial diaphyses increases after age 20, leading to a rise in absolute tibial bone mass.
