We evaluated the pattern of osteoporosis after spinal cord injury, determined the time-frame of the changes, and elucidated the relationship among parathyroid hormone levels, biochemical markers of bone formation, and the pattern of bone mass loss. We included 176 subjects with spinal cord injury and 62 subjects without spinal cord injury as controls in the study. Bone mineral density of the spine and the proximal femur was measured. The participants' age, level of injury, and length of time since injury were compared with the nonspinal cord-injured controls and with each other. Serum levels of calcium, calcitonin, biochemical markers of bone formation, and parathyroid hormone were determined. Our results revealed that bone mineral density of the proximal femur declined and reached fracture threshold at one to five years after injury. The decline was detected at 12 months after injury in all age groups. Spinal bone mineral density neither declined significantly nor reached fracture threshold. Parathyroid hormone levels declined before the end of the first year postinjury and increased at one to nine years postinjury in the 20- to 39-year age group. The increase correlated with the initial decline of bone mineral density of the proximal femur. Our studies in spinal cord-injured subjects revealed a pattern of osteoporosis similar to age and parathyroid dysfunction-related osteoporosis. No other correlation was detected between indexes of bone metabolism and bone mineral density measurements.

The purpose of the study was to use dual energy X-ray absorptiometry to measure bone mineral density (BMD) in the lumbar spine, the femoral neck, Ward's triangle, and the greater trochanter in 204 men (69 able-bodies controls and 135 spinal cord injured patients) strati®ed according to age (20 ± 39, 40 ± 59, and 60+ years old) in order to determine whether changes in BMD were age related, and to determine when these changes began to appear. The BMDs of the lumbar spine of both the 40 ± 59 year old and the 60+ year old patients were signi®cantly higher (P40.012) than the 40 ± 59 year old and 60+ year old controls, respectively. The femoral region BMDs of the 20 ± 39 year old and the 40 ± 59 year old patients were all signi®cantly lower (P40.027) than the 20 ± 39 year old and 40 ± 59 year old controls, respectively. When patients were grouped according to the time since their injury (0 ± 1, 1 ± 5, 6 ± 9, 10 ± 19, 20 ± 29, 30 ± 39, 40 ± 49, and 50 ± 59 years post injury) within the various age categories di erent results were obtained. In all the age categories, BMD loss occurred starting one year after spinal cord injury in the hip region. This bone loss took place gradually, reaching a signi®cant plateau (P40.017) at 19 years post injury and then started improving. The spine BMD in our patient population never signi®cantly decreased, and started improving as the age of the injury increased. Findings presented for the femoral regions were similar to other investigators' ®ndings; however, the steady bone mass maintained in the lumbar area, which increased with age regardless of the age of the injury, with the bone mass loss in the hip area, were the most notable new ®ndings.Keywords: spinal cord injury; spinal cord injury associated bone mass loss; bone mineral density (BMD); dual energy X-ray absorptometry (DEXA)

Disuse osteoporosis occurs in the lower extremities of patients with spinal cord injury (SCI). However, spinal osteoporosis is not usually observed in these patients. We investigated lumbar spine bone mineral density (BMD) in SCI patients using single energy quantitative computed tomography (QCT) and dual-energy X-ray absorptiometry (DXA). Our study population consisted of 64 patients with long-standing SCI. Spine BMD (g/cm3) was assessed by QCT at four vertebrae ranging from T11 to L4 with single mid-vertebral CT slices 1 cm thick parallel to the vertebral end-plates. Confounding variables affecting normal trabecular bone pattern, such as compression fractures, surgical hardware or fat replacement, were excluded. For a subset of 29 patients, DXA values of the spine and femoral neck were also measured, and QCT and DXA Z-scores were compared On the average, the 64 SCI patients had Z-scores 2.0 +/- 1.2 below those of age-matched controls. In the subset of 29 patients with both QCT and DXA measurements, the QCT and DXA Z-scores were 2.4 +/- 1.1 below and 1.3 +/- 2.3 above the mean, respectively (p < 0.0001). Our results indicate that QCT reveals osteoporosis of the spine after SCI, in contrast to DXA. We postulate that QCT is more valuable for evaluating spinal osteoporosis following SCI than DXA and thus recommend QCT for spinal BMD studies in SCI.

The aim of the present study was to compare bone mineral density (BMD in g/cm 2 ) in the lumbar spine and three hip regions of male spinal cord injured subjects at various times post injury to age-matched able-bodied controls and to correlate their BMDs to their age and level of their spinal cord lesion. Patients and controls were strati®ed into three 20 year age groups (20 ± 39, 40 ± 59, and 60 + years of age). BMD measurements were obtained using dual energy X-ray absorptiometry (DEXA, Lunar Model DPX). BMD levels taken within the ®rst year of injury were not signi®cantly lower than the age-matched able-bodies controls. The 20 ± 39 year old patients injured longer than 1 year had signi®cantly lower (P40.01) BMDs in their femoral region than both their age matched controls and the 20 ± 39 year old acutely injured (injured for less than 1 year) patients. Although femoral BMDs of both paraplegic and quadriplegic patients 40 ± 59 and 60 + years of age decreased over time, none showed signi®cant bone loss in this region until 10 years after their injury. These results indicate that spinal cord injury associated bone loss occurs most dramatically in the femoral region of young men. These results also indicate that initial bone mass loss does not occur prior to 1 year post-injury to the extent that it is detectable by densitometry, or at least it did not occur in our patients.