Impact of osteoporosis and osteoporosis medications on fracture healing: a narrative review.

Antiresorptive medications do not negatively affect fracture healing in humans. Teriparatide may decrease time to fracture healing. Romosozumab has not shown a beneficial effect on human fracture healing. BACKGROUND: Fracture healing is a complex process. Uncertainty exists over the influence of osteoporosis and the medications used to treat it on fracture healing. METHODS: Narrative review authored by the members of the Fracture Working Group of the Committee of Scientific Advisors of the International Osteoporosis Foundation (IOF), on behalf of the IOF and the Société Internationale de Chirurgie Orthopédique et de Traumatologie (SICOT). RESULTS: Fracture healing is a multistep process. Most fractures heal through a combination of intramembranous and endochondral ossification. Radiographic imaging is important for evaluating fracture healing and for detecting delayed or non-union. The presence of callus formation, bridging trabeculae, and a decrease in the size of the fracture line over time are indicative of healing. Imaging must be combined with clinical parameters and patient-reported outcomes. Animal data support a negative effect of osteoporosis on fracture healing; however, clinical data do not appear to corroborate with this. Evidence does not support a delay in the initiation of antiresorptive therapy following acute fragility fractures. There is no reason for suspension of osteoporosis medication at the time of fracture if the person is already on treatment. Teriparatide treatment may shorten fracture healing time at certain sites such as distal radius; however, it does not prevent non-union or influence union rate. The positive effect on fracture healing that romosozumab has demonstrated in animals has not been observed in humans. CONCLUSION: Overall, there appears to be no deleterious effect of osteoporosis medications on fracture healing. The benefit of treating osteoporosis and the urgent necessity to mitigate imminent refracture risk after a fracture should be given prime consideration. It is imperative that new radiological and biological markers of fracture healing be identified. It is also important to synthesize clinical and basic science methodologies to assess fracture healing, so that a convergence of the two frameworks can be achieved.

Denosumab for the treatment of primary pediatric osteoporosis.

Primary osteoporosis is rare in children and adolescents and its optimal pharmacological management is uncertain. Bisphosphonates are commonly used while denosumab has only been administered to a few children with osteogenesis imperfecta. We studied a treatment-naïve 13.5-year-old boy with severe osteoporosis and multiple vertebral deformities who presented with back pain and difficulty in walking. Causes of secondary osteoporosis were excluded and there were no abnormalities in genes known to cause bone fragility. He was treated with denosumab 60 mg subcutaneously every 3 months for 30 months, and he was pain-free within 6 weeks after the first injection. Lumbar spine BMD and femoral neck BMD increased with treatment by 65.6% and 25.3%, respectively, and deformed vertebrae regained their normal shape; linear growth was not impaired. During the second year of treatment, transient hypercalcemia (maximum 3.09 mmol/l) before the denosumab injection was observed. In conclusion, denosumab was highly effective in this case of primary pediatric osteoporosis, with remarkable clinical and radiological response. Transient hypercalcemia was probably due to amplification of the effect of growth spurt and puberty on bone remodeling by the transient, short-term discontinuation of the drug. Furthermore, our data suggest that mobilization of calcium from treatment-induced sclerotic transverse lines in bone metaphyses may contribute to the development of hypercalcemia.

Cost-effective osteoporosis treatment thresholds for people living with HIV infection in Greece.

OBJECTIVES: We aimed to specifically define the FRAX-based cost-effective treatment thresholds for osteoporosis among people living with HIV (PLWHIV) in Greece and to compare them with those of the general population. METHODS: A previously described state transition Markov cohort model was used in order to estimate the cost-effective intervention thresholds for osteoporotic therapy among Greek PLWHIV employing the FRAX® tool. The model-derived relative risk at which an incremental cost-effectiveness ratio of 30,000€/QALY gained was observed for treatment versus no intervention was multiplied by the average Greek FRAX-based 10-year probabilities for both major osteoporotic and hip fractures. RESULTS: There exists no significant difference in the cost-effective FRAX® based thresholds between PLWHIV and general population. The absolute 10-year probabilities of 2.5 and 10% for hip and major osteoporotic fractures, respectively, could be used for the initiation of treatment for PLWHIV of both genders under the age of 75; for older subjects the proposed intervention threshold is raised to 5 and 15% 10-year probability for hip and major osteoporotic fracture, respectively. CONCLUSIONS: Our study confirms the general recommendation for the use of country specific FRAX® thresholds when managing bone fragility within PLWHIV. In any case, clinical judgment and appropriate screening are mandatory and irreplaceable.

Factors associated with high 24-month persistence with denosumab: results of a real-world, non-interventional study of women with postmenopausal osteoporosis in Germany, Austria, Greece, and Belgium.

Persistence with osteoporosis therapy is vital for fracture prevention. This non-interventional study of postmenopausal women receiving denosumab in Germany, Austria, Greece, and Belgium found that persistence with denosumab remains consistently high after 24 months in patients at high risk of fracture. PURPOSE: Continued persistence with osteoporosis therapy is vital for fracture prevention. This non-interventional study of clinical practice evaluated medication-taking behavior of postmenopausal women receiving denosumab in Germany, Austria, Greece, and Belgium and factors influencing persistence. METHODS: Subcutaneous denosumab (60 mg every 6 months) was assigned according to prescribing information and local guidelines before and independently of enrollment; outcomes were recorded during routine practice for up to 24 months. Persistence was defined as receiving the subsequent injection within 6 months + 8 weeks of the previous injection and adherence as administration of subsequent injections within 6 months ± 4 weeks of the previous injection. Medication coverage ratio (MCR) was calculated as the proportion of time a patient was covered by denosumab. Associations between pre-specified baseline covariates and 24-month persistence were assessed using multivariable logistic regression. RESULTS: The 24-month analyses included 1479 women (mean age 66.3-72.5 years) from 140 sites; persistence with denosumab was 75.1-86.0%, adherence 62.9-70.1%, and mean MCR 87.4-92.4%. No covariate had a significant effect on persistence across all four countries. For three countries, a recent fall decreased persistence; patients were generally older with chronic medical conditions. In some countries, other covariates (e.g., older age, comorbidity, immobility, and prescribing reasons) decreased persistence. Adverse drug reactions were reported in 2.3-6.9% patients. CONCLUSIONS: Twenty-four-month persistence with denosumab is consistently high among postmenopausal women in Europe and may be influenced by patient characteristics. Further studies are needed to identify determinants of low persistence.

Persistence, adherence, and medication-taking behavior in women with postmenopausal osteoporosis receiving denosumab in routine practice in Germany, Austria, Greece, and Belgium: 12-month results from a European non-interventional study.

UNLABELLED: Persistence with and adherence to osteoporosis therapy are critical for fracture reduction. This non-interventional study is evaluating medication-taking behavior of women with postmenopausal osteoporosis (PMO) receiving denosumab in Germany, Austria, Greece, and Belgium. Patients were representative of the PMO population and highly persistent with and adherent to denosumab at 12 months. INTRODUCTION: Persistence with and adherence to osteoporosis therapy are important for optimal treatment efficacy, namely fracture reduction. This ongoing, non-interventional study will evaluate medication-taking behavior of women with postmenopausal osteoporosis (PMO) receiving denosumab in routine practice in four European countries. METHODS: The study enrolled women who had been prescribed subcutaneous denosumab (60 mg every 6 months) in accordance with prescribing information and local guidelines. Persistence was defined as receiving the subsequent injection within 6 months + 8 weeks of the previous injection. Adherence was defined as receiving two consecutive injections within 6 months ± 4 weeks of each other. Medication coverage ratio (MCR) was calculated using the time a patient was covered with denosumab, as assessed from prescription records. Treatment was assigned prior to and independently of enrollment; outcomes are recorded during routine practice. RESULTS: These planned 12-month interim analyses included data from 1500 patients from 141 sites. Mean age was 66.4-72.4 years, mean baseline total hip T-scores ranged from -2.0 to -2.1 and femoral neck T-scores from -2.2 to -2.6, and 30.7-62.1% of patients had prior osteoporotic fracture. Persistence was 87.0-95.3%, adherence 82.7-89.3%, and MCR 91.3-95.4%. In a univariate analysis, increased age, decreased mobility, and increased distance to the clinic were associated with significantly decreased persistence; parental history of hip fracture was associated with significantly increased persistence. CONCLUSIONS: These data extend the real-world evidence regarding persistence with and adherence to denosumab, both of which are critical for favorable clinical outcomes, including fracture risk reduction.

Denosumab versus zoledronic acid in patients previously treated with zoledronic acid.

UNLABELLED: Denosumab and zoledronic acid are potent antiresorptives. In this study in patients pre-treated with zoledronic acid, denosumab achieved similar increases with zoledronic acid in lumbar spine BMD despite the more prominent reduction of bone turnover markers. Denosumab reversibly reduced endogenous RANKL. INTRODUCTION: We aimed to compare yearly changes in lumbar spine (LS) bone mineral density (BMD), bone turnover markers, free soluble receptor activator of nuclear factor kappaB ligand (sRANKL) and sclerostin levels between denosumab and zoledronic acid. METHODS: Postmenopausal women with low bone mass previously treated with zoledronic acid for 1 year were assigned to denosumab injection (n = 32) or zoledronic acid infusion (n = 26). Procollagen type 1 N-terminal propeptide (P1NP), C-terminal cross-linking telopeptide of type 1 collagen (CTx), sRANKL, and sclerostin levels were measured in serum samples obtained at baseline and 3, 6, and 12 months after denosumab injection or zoledronic acid infusion. LS BMD was measured at baseline and 12 months. RESULTS: The mean LS increase was 4.5 and 4.4% with denosumab and zoledronic acid, respectively (p = 0.560). Denosumab caused a larger decrease in CTx at 3 months (p < 0.001) and P1NP at 3 (p < 0.001), 6 (p = 0.021), and 12 months (p = 0.042). Denosumab significantly decreased sRANKL by 87.4% at 3 months (p < 0.001). Sclerostin levels were not changed with either intervention (p = 0.162 and p = 0.214, respectively). CONCLUSIONS: In patients previously treated with zoledronic acid, denosumab reduces bone turnover more than zoledronic acid, but the increases in LS BMD are comparable. Furthermore, denosumab administration results in reversible inhibition of the metabolically significant endogenous free soluble RANKL levels. Serum sclerostin is not affected by either agent.

Cost-effective osteoporosis treatment thresholds in Greece.

UNLABELLED: A Greek-specific cost-effectiveness analysis determined the FRAX-based intervention thresholds. Assuming a willingness to pay of 30,000 , osteoporosis treatment is cost-effective in subjects under the age of 75 with 10-year probabilities for hip and major osteoporotic fractures of 2.5 and 10 %, respectively, while for older patients, the same thresholds are raised to 5 and 15 %. INTRODUCTION: The purpose of this study was to determine the FRAX calculated fracture probabilities at which therapeutic intervention can be considered as cost-effective in the Greek setting. METHODS: A Markov cohort model was populated with Greek data, and quality-adjusted life years (QALYs) were used to calculate the cost-effective thresholds for an annual medication cost of 733.7 by gender and age. Average FRAX-based 10-year probabilities for both major osteoporotic and hip fractures were multiplied by the model-derived relative risk at which a cost of 30,000 for each QALY gained was observed for treatment versus to no intervention. RESULTS: A biphasic intervention threshold model is supported by our findings. Osteoporosis treatment becomes cost-effective when absolute 10-year probabilities for hip and major osteoporotic fractures reach 2.5 and 10 %, respectively, among both men and women under the age of 75. For older subjects, the proposed intervention thresholds are raised to 5 and 15 % 10-year probability for hip and major osteoporotic fractures, respectively. CONCLUSIONS: Cost-effective osteoporosis treatment may be facilitated in Greece if FRAX algorithm is used to identify subjects with 10-year probabilities for hip and major osteoporotic fractures of 2.5 and 10 %, under the age of 75, while for older patients, the relevant thresholds are 5 and 15 %, respectively.

Circulating irisin is associated with osteoporotic fractures in postmenopausal women with low bone mass but is not affected by either teriparatide or denosumab treatment for 3 months.

UNLABELLED: In vitro data suggest that myokine irisin may affect bone metabolism by promoting osteoblast differentiation while inhibiting osteoclast differentiation. In this study, circulating irisin levels were associated with previous osteoporotic fractures but not with bone mass and were not affected by denosumab or teriparatide treatment for 3 months. INTRODUCTION: This study aimed to evaluate predictors of circulating irisin in postmenopausal women with low bone mass and to assess a potential effect of denosumab or teriparatide treatment for 3 months. METHODS: Serum samples for irisin measurement were obtained from (a) postmenopausal women with low bone mass (lumbar spinal [LS] or femoral neck [FN] bone mineral density [BMD] T-score ≤-2.0) and their age-matched controls at baseline and 3 months after denosumab (Dmab) injection (Dmab group, n = 50; Dmab control group, n = 25) and (b) women with more severe disease (LS or FN BMD T-score ≤-2.8) and their age-matched controls at the above-mentioned time points after teriparatide (TPTD) initiation (TPTD group, n = 25; TPTD control group, n = 25). RESULTS: At baseline, irisin levels were inversely correlated with age (partial coefficient (r p ) = -0.24; p = 0.009), parathyroid hormone (PTH) (r p = -0.30; p = 0.001), and creatinine (r p = -0.23; p = 0.016) in univariate analysis, and were lower in women with (n = 26; 41.6 ± 2.7 ng/dL) than without previous osteoporotic fracture(s) (n = 99; 51.0 ± 1.6 ng/dL; p = 0.007). In multiple linear regression, previous osteoporotic fracture(s) and PTH were independently negatively associated with irisin [p = 0.04, CI -16.1 to -0.4 and p = 0.002, CI -0.3 to -0.07, respectively], but only the association with PTH remained after controlling for creatinine levels. Serum irisin levels were not different between women with or without low bone mass and were not affected by either Dmab or TPTD treatment for 3 months. CONCLUSIONS: Circulating irisin levels were associated with previous osteoporotic fracture(s); whether this association is independent or is due to confounding by lower muscle mass, potentially reflected by lower creatinine levels, remains to be fully clarified.

Circulating periostin levels do not differ between postmenopausal women with normal and low bone mass and are not affected by zoledronic acid treatment.

Periostin is a secreted extracellular matrix protein preferentially expressed in bone by osteocytes and periosteal osteoblasts. Reduced periostin expression may affect osteoblast differentiation and collagen type I synthesis and predispose to osteoporosis and increased fracture risk. We aimed to evaluate circulating periostin levels in postmenopausal women with low bone mass, their possible correlations with clinical and laboratory parameters, as well as the 3-month effect of zoledronic acid. Serum samples for periostin, 25-hydroxyvitamin D, parathyroid hormone (PTH), C-terminal telopeptide of type I collagen (CTx), and total alkaline phosphatase (tALP) were obtained from 46 postmenopausal women with low bone mass at baseline and 3 months after zoledronic acid infusion and from 30 age-matched, postmenopausal controls with normal bone mass at baseline. There was no difference in periostin levels between women with normal and low bone mass (250±15 vs. 272±14 ng/ml, respectively; p=0.279). Periostin remained essentially unchanged after zoledronic acid infusion (262±18 ng/ml; p=0.130). Serum periostin levels at baseline were not affected by previous bisphosphonate treatment, and were correlated only to tALP (rs=0.351; p=0.018). In multiple linear regression analysis, tALP (B=3.17; 95% CI=0.59-5.79; p=0.018) was associated with serum periostin levels at baseline, independently from previous anti-osteoporotic treatment, age, body mass index, and 25-hydroxyvitamin D. In conclusion, serum periostin levels do not differ between postmenopausal women with normal and low bone mass and are not affected by zoledronic acid treatment. Women with higher tALP have independently higher periostin levels.

Incidence of hip fractures in Greece during a 30-year period: 1977-2007.

UNLABELLED: The incidence of hip fractures doubled in Greece from 1977 to 2007 among people aged 50 and over. A mild decrease was observed after 2002, although the future trend cannot be safely anticipated at the moment. Half of all hip fractures in 2007 were derived from the age group of 80 and over. INTRODUCTION: The purpose of this study was to determine the incidence of hip fractures during a 30-year period in Greece among people aged 50 and over and to document possible alterations in secular trends. METHODS: We studied hip fractures during 2007 and compared them with those of previous years starting from 1977 with an in-between 5-year interval (1977, 1982, 1987, 1992, 1997, 2002). Age- and sex-specific incidence was calculated, and secular trends were recorded. The relative risk of hip fracture in every age group was estimated according to the corresponding incidence of 1977. RESULTS: The adjusted incidence of hip fractures increased approximately 100 % throughout the study; it progressively increased from 1977 to 2002 and exhibited a mild significant decrease thereafter. The relative risk of hip fractures among subjects aged 60-69 in 2007 has declined compared with 1977 [0.85, 95 % confidence intervals (CI) 0.79-0.92, p < 0.0005]. Among people aged 70-79, an increased relative fracture risk (1.53, 95 % CI 1.45-1.61, p < 0.0005) was estimated in 2007 compared with 1977. People ≥80 years old were responsible for half of the hip fractures in 2007 but only for the 22.5 % of fractures in 1977. The relative fracture risk in people aged ≥80 was 2.81 times higher (95 % CI 2.64-2.98, p < 0.0005) in 2007 than in 1977. CONCLUSIONS: The incidence of hip fractures doubled during the last 30 years among people aged ≥50 years, although a mild decrease was observed in almost all age groups after 2002. The most affected group is 80 and over.

Circulating activin-A is elevated in postmenopausal women with low bone mass: the three-month effect of zoledronic acid treatment.

UNLABELLED: Activin-A is expressed in bone and seems to regulate osteoclastogenesis. In this study, serum activin-A was increased in postmenopausal women with low bone mass and was positively correlated to age and negatively to lumbar spinal bone mineral density (BMD). Serum activin-A levels did not change 3 months after zoledronic acid infusion. INTRODUCTION: The aims of the study were to evaluate prospectively the circulating activin-A levels in postmenopausal women with low bone mass and explore possible correlations with clinical and laboratory data, as well as the 3-month effect of zoledronic acid infusion. METHODS: Postmenopausal women with low bone mass assigned to receive zoledronic acid infusion (Patients, n = 47) and age-matched, postmenopausal women with normal bone mass (Controls, n = 27) were recruited on an outpatient basis. Main outcome measurement was serum activin-A levels. RESULTS: Serum activin-A was higher in patients at baseline compared to controls (p < 0.001) and activin-A in the serum of patients and controls was positively correlated with age (Spearman's coefficient of correlation [rs] = 0.325; p = 0.005) and negatively with lumbar spinal (LS) BMD (rs = -0.425; p < 0.001). In multiple linear regression analysis, only age (B = 8.93; 95 % CI = 4.39-13.46; p < 0.001) was associated with serum activin-A levels at baseline, independent from group (patients or controls), previous anti-osteoporotic treatment, LS BMD and follicle-stimulating hormone. Circulating activin-A levels were not affected 3 months after zoledronic acid infusion. CONCLUSIONS: Serum activin-A is increased in postmenopausal women with low bone mass compared with postmenopausal women with normal bone mass and is positively correlated to age and negatively to LS BMD.

Performance of the mini nutritional assessment score in the detection of vitamin D status in an elderly Greek population.

The primary aim of the study was to explore the potential relationship between serum 25-hydroxyvitamin D [25(OH)D] levels and Mini Nutritional Assessment (MNA) score, a surrogate for protein energy undernutrition, in elderly (≥65 years old) subjects with and without a hip fracture. A secondary aim of the study was to provide estimates of the MNA discriminatory performance in the detection of subjects with low levels of 25(OH)D (<20 ng/ml). The study population consisted of 101 patients with a hip fracture, recruited from a single urban Hospital in Athens, Greece, and 85 community dwelling subjects with no history of hip fracture. Serum 25(OH)D was measured, nutritional status was determined by the MNA questionnaire in all subjects, and linear correlation between variables was investigated. Receiver operator characteristic (ROC) curve analysis was performed and discriminatory performance was further assessed by calculating positive and negative likelihood ratios (LR). MNA scores were significantly correlated with 25(OH)D levels (rho=0.685, p<0.001) and this finding was robust in both groups and unaffected by gender. ROC curve analysis demonstrated an area under the curve (AUC) of 0.860 [standard error (SE): 0.026, 95% confidence interval (CI): 0.810-0.910], which provided a significantly better estimation of 25(OH)D status than simple guess (p<0.001). The lowest cutoff value in MNA score, providing a sensitivity over 90% was 25.25, which was associated with a sensitivity of 90.9% and a specificity of 53.6%. The same analysis revealed acceptable results only within hip fracture patients. MNA score might be a satisfactory surrogate marker for 25(OH)D levels with which it is linearly correlated. However, it appears that its discriminatory performance, as a diagnostic tool for 25(OH)D insufficiency, is rather suboptimal.

Paget's disease of bone and calcium homeostasis: focus on bisphosphonate treatment.

Paget's disease of bone (PDB) is the second most common metabolic bone disease. Bisphosphonates (BPs) are currently the drugs of choice for PDB. PDB and osteomalacia are both common in the elderly. The concept of relative vitamin D deficiency in patients with PDB was suggested long ago, but it has not yet elucidated. Both diseases predispose to fractures, but their combined action to fragility has not been studied yet. The older BPs, mainly etidronate, further inhibit bone mineralization. Mineralization defects have also been described in patients with PDB treated with pamidronate. Moreover, hypocalcemia and secondary hyperparathyroidism after treatment with BPs have been described in PDB. Hypocalcemia seems to be more severe after treatment with the more potent, intravenous zoledronic acid, which is currently the treatment of choice for PDB. The counteracting hyperparathyroidism pathophysiologically intends to increase renal reabsorption of calcium and 1.25-dihydroxy vitamin D production and to stimulate osteoclasts in order to prevent long-term hypocalcemia. However, the effect of PTH on osteoclasts is, at least partly, restricted in patients taking BPs. Secondary hyperparathyroidism is a potentially detrimental condition, especially in patients already suffering from another bone disease. Serum calcium and vitamin D deficiency should be restored before BP treatment and calcium and vitamin D administration should be possibly continued for longer after achieving normocalcemia, which may shorten the duration of secondary hyperparathyroidism. QUICK SUMMARY: Mineralization defects and hypocalcemia with secondary hyperparathyroidism have been described in patients with Paget's disease of bone treated with bisphosphonates. Secondary hyperparathyroidism may be a potentially detrimental condition for patients with Paget's disease of bone.

Erythrocyte fatty acid composition and insulin sensitivity in daughters of Type 2 diabetic patients and women with no family history of diabetes.

BACKGROUND: There is evidence that the impact of environmental factors on insulin sensitivity is modified by the presence of family history of diabetes. AIM: To compare the association between the erythrocyte phospholipid fatty acid composition (a biomarker of dietary fatty acids) and insulin sensitivity in daughters of Type 2 diabetic patients with the corresponding association in women without family history of diabetes. MATERIAL/SUBJECTS AND METHODS: Eighteen offspring of Type 2 diabetic patients [age 30+/-6.5 yr; body mass index (BMI) 22.2+/-2.5 kg/m2; body fat 31.8+/-5.1%] and 18 matched women (age 30.1+/-6.8 yr; BMI 22.2+/-1.8 kg/m2; body fat 32.2+/-6.0%) participated in the study. RESULTS: Insulin Sensitivity Index (ISI)-Matsuda tended to be lower (p=0.06) in the Offspring than the control group. Weight proportions of erythrocyte phospholipid saturated (SFA), polyunsaturated (PUFA), and monounsaturated fatty acids (MUFA) were similar between the two groups. In the offspring, erythrocyte total SFA were negatively correlated with ISI-Matsuda [r=-0.47, p<0.05), ISI(gly)-Belfiore (r=-0.52, p<0.05) and ISI(ffa)-Belfiore (r=-0.53, p<0.05)], whereas total PUFA were positively correlated with insulin sensitivity [ISI-Matsuda, r=0.46, p<0.05; ISI(gly)-Belfiore, r=0.53, p<0.05; ISI(ffa)-Belfiore, r=0.54, p<0.05]. No significant correlations were observed in the control group. CONCLUSIONS: The associations between erythrocyte fatty acid composition and insulin sensitivity are distinct between daughters of Type 2 diabetic patients and women without family history of diabetes.

Bisphosphonate dose and incidence of fractures in postmenopausal osteoporosis.

INTRODUCTION: The specific pharmacological properties of bisphosphonates have raised concerns about their long-term effects on skeletal fragility that may be related to the total dose of bisphosphonate given. However, the effect of different doses on the incidence of osteoporotic fractures has not been adequately studied. METHODS: In this retrospective analysis, we investigated the effect of different doses of intravenous pamidronate given at 3-monthly intervals on the incidence of fractures in 92 women with severe postmenopausal osteoporosis. RESULTS: The risk of sustaining a new vertebral fracture on treatment was significantly increased by 32% for every prevalent vertebral fracture (OR: 1.32, CI: 1.05, 1.66; p=0.02). Patients with nonvertebral fractures received a significantly lower dose of pamidronate and their risk for these fractures increased by 25% for every prevalent vertebral fracture at baseline (OR: 1.25, CI: 1.01, 1.53; p=0.03). Patients who had received oral bisphosphonate before intravenous pamidronate had a significantly higher incidence of nonvertebral fractures which, however, did not hold true after adjustment for baseline BMD and prevalent fractures. CONCLUSIONS: In patients with established osteoporosis bone fragility during treatment with intravenous pamidronate is mainly determined by the severity of the disease, assessed by the presence and numbers of prevalent fractures, rather than the dose of the bisphosphonate or the rate of bone turnover.