The authors have no conflict of interest.
Intravenous Bisphosphonate Therapy Increases Radial Width in Adults With Osteogenesis Imperfecta†
Version of Record online: 14 MAR 2005
Copyright © 2005 ASBMR
Journal of Bone and Mineral Research
Volume 20, Issue 8, pages 1323–1326, August 2005
How to Cite
Gatti, D., Viapiana, O., Lippolis, I., Braga, V., Prizzi, R., Rossini, M. and Adami, S. (2005), Intravenous Bisphosphonate Therapy Increases Radial Width in Adults With Osteogenesis Imperfecta. J Bone Miner Res, 20: 1323–1326. doi: 10.1359/JBMR.050312
- Issue online: 4 DEC 2009
- Version of Record online: 14 MAR 2005
- Manuscript Accepted: 10 MAR 2005
- Manuscript Revised: 17 FEB 2005
- Manuscript Received: 13 DEC 2004
- osteogenesis imperfecta;
- bending breaking resistance;
- bone geometry
Neridronate therapy in adult patients with OI significantly increases the cross-sectional area of the proximal radius. This observation may provide an additional explanation for the antifracture efficacy of bisphosphonates.
Introduction: Bisphosphonate therapy decreases by 70-90% the fracture risk in patients with osteogenesis imperfecta (OI). This decrease is somewhat greater than that expected from the BMD changes, supporting the hypothesis that bisphosphonate therapy is associated with structural changes, not detectable by BMD measurements.
Materials and Methods: To explore this hypothesis, pQCT measurements at the nondominant radius were obtained in a group of adult OI patients participating in a randomized clinical trial with neridronate.
Results: The total volumetric BMD of the ultradistal radius rose significantly in patients treated with neridronate and calcium + vitamin D (neridronate group) compared with patients treated with calcium + vitamin D alone (control group). No significant differences were observed in trabecular BMD and in volumetric cortical density in either group. In the neridronate group, the cross-sectional area rose significantly versus both baseline values and the control group. These latter changes were associated with ∼20% increases in bending breaking resistance index (BBRI).
Conclusion: Our observation, if extended to postmenopausal osteoporosis, may provide a new explanation for the fracture risk reduction observed in osteoporotic patients treated with bisphosphonates.
OSTEOGENESIS IMPERFECTA (OI) is a heritable disease of connective tissue caused by heterologous mutations in the genes encoding for type I collagen and characterized by increased bone fragility.(1)
Recently, we have shown by pQCT measurements of the radius(2) that, in adults with OI, the most obvious defect is the inability to acquire an adequate thickness of the cortices of long bone and to achieve or maintain normal trabecular density. Bisphosphonates are nowadays considered the most promising therapy for OI.(1)
In 1995, in collaboration with the Istituto Superiore di Sanità (part of the Ministry of Health) and the Italian Association of patients with OI (AsItOI), we initiated a large randomized, controlled therapeutical trial of intravenous neridronate in patients of any age with OI. We have recently reported the results obtained in adults and in prepubertal children.(3, 4) In both studies, the observed decrease in fracture risk was somewhat greater than that expected from the BMD changes, a finding observed also during treatment in postmenopausal osteoporosis with antiresorbers.(5) This might suggest that structural skeletal changes not detected by traditional DXA may occur during treatment with bisphosphonates. To study this possibility, we report here the changes in pQCT parameters observed after intravenous neridronate therapy in adults with OI.
MATERIALS AND METHODS
The inclusion and exclusion criteria together with the details on recruitment and treatment of the clinical trial on the effect of intravenous neridronate in patients with OI are reported in detail elsewhere.(4) Patient recruitment initiated in 1996 and ended in March 2003, but we decided to include pQCT evaluation only in 2002. The study population of this report includes 26 of the adult patients described before(3) and 26 new patients who completed more recently the first year of follow-up. All patients were seen at 3-month intervals, but a pQCT evaluation was obtained at baseline and at the end of the first year of observation.
All patients had their dietary calcium intake regularly evaluated and maintained above 1000 mg daily through diet or supplementation. Vitamin D2 supplements (50,000 U monthly) were given if the serum 25(OH)vitamin D levels fell below 20 ng/ml3. The patients were randomized to treatment with either 100 mg neridronate diluted in 250 ml of saline infused intravenously every 3 months and calcium + vitamin D supplements (neridronate group) or calcium + vitamin D alone (control group) with OI.(3)
pQCT measurement was performed at the nondominant radius using a Stratec, XCT 960 machine (Unitrem, Roma, Italy). The Stratec XCT 960 determines by default only the trabecular BMD of the ultradistal radius. At this level, the CV was 2% for both dimension and density measurements.
The cross-sectional area of the proximal radius and its volumetric density were measured by nonstandard software at a constant 8 cm distance from the stylohyoid process, using a given attenuation coefficient. The system recognizes the tissue as cortical bone any voxel with a density of >80% of total absorbance. The cross-sectional area of cortical bone was estimated by the difference between total cross-sectional area and medullary area (any tissue with a density threshold <80). The partial volume effect is an inevitable source of error that is particularly important at the endosteal surface, where the separation between cortical tissue and bone marrow is less defined than at the periosteal surface. The cortical density (mg/cm3) is defined as the average density of all voxels predefined as cortical bone. The method, which has been described in detail elsewhere,(2, 6) has a CV of 1.8% for the overall cross-sectional area. The CV for the derived medullary area is much higher (3.9%).
The within-subject changes and the between-group differences in pQCT parameters were tested by ANOVA with Bonferroni's adjustment for multiple comparisons and by two-sided paired Student t-test (SPSS version 11.0).
Before treatment, all subjects had normal serum calcium, phosphate, 25(OH)vitamin D levels (>12 ng/ml), and PTH (Allegro intact PTH, normal values < 60 pg/ml), and there was no evidence of other relevant diseases. The two groups of patients (neridronate and control group) were comparable for age, weight, height, and female-male ratio, and nonstatistically different for pQCT parameters at baseline (Tables 1 and 2).
The mean changes in total volumetric BMDs of the ultradistal radius rose significantly (p < 0.02) in treated patients, and the changes were significantly different from those observed in control patients. Trabecular BMD significantly rose in the treated patients (+2.7 ± 5.3%), and these changes were significantly different from those observed in controls.
Bone volumetric cortical density at the proximal radius did change in the placebo or in the treatment group. A significant increase in the cross-sectional area of the proximal radius was observed in the treated group only (p < 0,04 versus control group; Table 2). The corresponding increase in cross-sectional diameter, estimated by assuming the sectional area as circular, was 0.56 mm. The cross-sectional area of cortical bone increased by 7.2% in treated patients (Table 2), but this change was not statistically significant for the large variability of the results. Figure 1 shows a typical example of the changes occurring in a patient after 1-year treatment with neridronate.
We have previously shown that bisphosphonate therapy significantly increases BMD, as measured by DXA technology, in adults with OI at both the hip and the spine.(3) Here, we have shown by pQCT that these changes in areal BMD are associated with significant increases in trabecular volumetric density but no changes in volumetric cortical density. However, the most striking new and unexpected finding is the increase in cross-sectional area of the proximal radius observed after 1 year of neridronate therapy, significantly different from the changes observed in the control group. In treated patients, the outer diameter rose by 0.56 mm. This might be seen as a small change, but it is associated with an estimated 20% increase in bending breaking resistance, which is proportional to the third power of the outer diameter.(6) If these changes occurred at any long bones, they may explain the strong decrease in fracture risk of the appendicular skeleton observed in adult OI patients treated with neridronate (−86%),(3) despite relatively modest DXA-BMD changes at both the spine and the hip.
An intriguing point raised by our observation is whether this effect of intravenous bisphosphonate therapy is peculiar to OI patients or may occur also in patients with postmenopausal osteoporosis. As far as we are aware, the changes in cross-sectional area after osteoporosis therapies were evaluated only in small groups of osteoporotic patients after therapy with PTH,(7) whereas no data are available after therapy with antiresorbing agents.
It has been frequently observed that the antifracture efficacy of antiresorbers such as alendronate, risedronate, and raloxifene occurs well before clinically meaningful BMD changes and are greater than expected from the BMD changes observed after 1–3 years.(5) Our observation in OI patients treated with a bisphosphonate may provide a clue for understanding these incongruences. In longitudinal observations, if bone mass remain constant (e.g., unchanged BMC as measured by DXA), even small increases in the cross-sectional area may actually be associated with decreases in areal BMD,(8) because the same BMC value is divided by a larger diameter. However, these decreases in areal BMD may be associated with remarkable increases in bone strength.
The mechanism by which bisphosphonate therapy increased the cross-sectional area of the proximal radius remains conjectural. The two most intuitive hypotheses are that either bisphosphonates promote periosteal apposition or, more likely, that inhibition of bone turnover at the periosteal surface is associated with increased mineralization above the density threshold used for detecting cortical bone surface by pQCT.
In conclusion, we have shown for the first time that bisphosphonate therapy may increase the cross-sectional area of the skeleton, and this may provide a new possible explanation to the mechanism by which antiresorbers decrease fracture rate in osteoporotic patients.