We compared indices of three-dimensional microstructure of iliac trabecular bone between 26 patients with vertebral compression fractures due to postmenopausal osteoporosis and 24 control subjects without vertebral fracture, who were matched for age, sex, race, menopausal status, and several densitometric and histologic indices of both cortical and trabecular bone mass. The patients with fracture had a significantly lower mean value (1.03 +/- 0.15 vs. 1.26 +/- 0.26; P less than 0.005) for indirectly calculated mean trabecular plate density, an index of the number and connectivity of structural elements, and as a necessary corollary, a significantly higher mean value for the mean thickness of structural elements. Plate density was more than one standard deviation below the age-adjusted mean value for normal postmenopausal white females in 19 (73%) of the fracture cases and in only 5 (21%) of the nonfracture cases (P less than 0.001). We conclude that the biomechanical competence of trabecular bone is dependent not only on the absolute amount of bone present but also on the trabecular microstructure.

No abstract

No abstract

We measured iliac bone formation rates on all surfaces after double tetracycline labeling, serum levels of type 1 procollagen carboxy-terminal extension peptide (pColl-I-C), and serum levels of total alkaline phosphatase activity (TAP) in four normal subjects and in 44 patients with various forms of metabolic bone disease. In three patients with enzymatic evidence of liver disease both biochemical serum markers were disproportionately raised. In a patient with idiopathic axial osteosclerosis serum pColl-I-C was selectively increased by more than ten-fold. In the remaining 44 subjects pColl-I-C and TAP levels correlated significantly with each other (r = 0.70) and both showed the same directional changes and broadly similar correlations with iliac bone formation rate expressed in different ways. In general, pColl-I-C levels correlated better with cancellous bone formation rates and TAP levels cortical bone formation rates. There was a modest improvement in prediction of bone formation rate with multiple regression using both markers. In 15 patients with typical uncomplicated postmenopausal osteoporosis, neither biochemical marker, singly or jointly, correlated significantly with any expression of bone formation rate. Disadvantages to the use of pColl-I-C as a marker include a significant contribution to the serum level from type 1 collagen biosynthesis in tissues other than bone, and (probably) variable metabolic clearance. For both biochemical markers the most consistently high correlations (r = 0.77-0.79) were found with total bone formation rate for the entire biopsy core volume, which is the best estimate available from a biopsy of formation rate at the bone organ level of organization in vivo. The core volume as a referent also allows the amount of bone formed on cortical, endocortical, and cancellous surfaces to be compared. Measurement of serum pColl-I-C levels merits further study as a noninvasive index of bone metabolism. Differences between normal and abnormal subjects in the relationships between a variety of biochemical markers and a variety of histologic indices have the potential for providing insight into the pathogenesis of osteoporosis.

No abstract