Osteoporosis (OP) is a skeletal disorder characterized by a decrease in bone strength predisposing an increased risk of fracture.1 The risk of fracture resulting from bone fragility is the most important clinical aspect of the disease.2 In the United States, approximately 10 million individuals suffer from OP or have a low bone mineral density (BMD), which can lead to fractures.3 In Spain, the disease affects approximately 15% of the female population, with a prevalence of 26% in women aged 50.4

Although fractures can occur in different locations, vertebral and hip fractures are of particular clinical importance because of their high prevalence, their impact on the quality of life of patients and their associated morbidity. The presence of one or more vertebral fractures multiplied by 5 the risk of new vertebral fractures in a period of one year5 and also increases the likelihood of suffering a hip fracture.6,7 However, vertebral fractures are not always recognized in time, which reduces the opportunities to identify and properly treat patients.8 Moreover, hip fractures are characterized by slow and often incomplete recovery, so a high proportion of patients require long-term care and ongoing support for walking.2,9 Hip fracture is associated with a high mortality,10 most deaths occurring in the first 3–6 months after the fracture.2

Osteoporotic fractures also have a major impact on healthcare budgets. It is estimated that by 2025 the annual cost of these fractures may rise to 25000 million dollars in United States11 and in 2050 the cost in Europe will be 76700 million euros.12 In Spain, the cost of hospital admissions resulting from hip fracture amounts to 220 million euros a year.13 Therefore, appropriate fracture risk stratification may help identify those patients most at risk and thus facilitate early treatment, with a consequent reduction in the likelihood of future fractures. Unfortunately, there is no universally accepted methodology on how to identify patients at high risk for fracture2,14,15 and there are no well-known criteria used in everyday practice of specialists for the management of these patients. For this reason, we set out to gather information about the Spanish rheumatologist assessment of the risk of fracture and how the diagnosis is made, the disease classified and patients when at high risk of osteoporotic fracture during their practice.

As part of the OSTEOPAR project, a group of experts in OP designed a survey that included 27 questions to obtain information on the profile of the surveyed rheumatologists, characteristics of clinical activity, assessment of risk factors (RF), patient assessment, choice of diagnostic tests and circumstances for which they are requested, and the treatment for which they qualify based on the patient's clinical characteristics (see Appendix A).

All responses were obtained by selecting one or more predetermined options in the questionnaire, except the last question, which asked respondents to define patients at high risk of fracture, using up to 200 characters of free text.

Different sections of the questionnaire specifically collected data on:

• (1)

  Characteristics of the participating physician. Workplace, years of professional experience in their specialty and Autonomous Community where the respondent was located at.

• (2)

  Characteristics of clinical activity. Total number of patients and patients with OP seen per week, percentage of patients with primary OP, origin, percentage with fractures and fracture type.

• (3)

  Assessment of the RF and the patient. Questions were made about the importance of several RF and high risk of osteoporotic fracture according to their degree of influence over diagnostic and/or therapeutic decision-making. Respondents made the assessment on a numerical scale from 1 to 5, where: 1=not important, 2=of little importance, 3=moderately important, 4=very important and 5=highest importance. Also, the surveyed were requested to order, by degree of importance, the 3 RF for fracture considered as more relevant. Regarding the assessment of the patient, we asked which option or options they preferred when faced with a patient with OP presenting back pain of recent onset or changes in their usual features, if exploring the spine for the presence of kyphosis or other deformities, and if they commonly used scales for assessing fracture risk and frequency of the use of the FRAX tool.

• (4)

  Diagnostic tests. Circumstances in which the respondent requested a lateral spine X-ray, value given to the different characteristics of morphometric vertebral fracture (on the above mentioned scale of 1–5) and whether or not an analysis of bone metabolism was performed.

• (5)

  Assessment of fracture risk and treatment options. How respondents rate the risk of fracture in 4 pre-defined patient-profiles (using the terms “high”, “medium” or “low”) and what treatment would be chosen as first, second and third options for each profile. Predefined profiles in the questionnaire were as follows: Profile 1: 65-year-old woman presented vertebral fracture after 5 years of treatment with bisphosphonates. Profile 2: 59-year-old woman in chronic treatment with bisphosphonates and a T-score of −2.5. Profile 3: 70-year-old woman without treatment with a T-score of −2.8 in the femoral neck and presenting 2 vertebral fractures. Profile 4: A 55-year-old man with a T-score of −1.5 in chronic treatment with glucocorticoids. Additionally, respondents were asked to determine whether supplemental calcium and vitamin D should be used for each profile, as well as if treatment for OP should be employed and what would the physician choose among the following: estrogen receptor modulators, oral or intravenous bisphosphonates, strontium ranelate, PTH (1–84) and teriparatide.

Finally, respondents were asked to evaluate whether, from their point of view, there was a group of patients at high risk of fracture and whether their behavior would be different toward these patients.

Descriptive statistics based on the calculation of percentages, means and medians was used and data analysis was carried out using SPSS software version 18.0. When appropriate, the analysis also included the standard deviation, mode, variance and percentiles, as well as maximum and minimum values.

Because the clinical significance of osteoporosis lies primarily in the risk of fracture and the morbidity and mortality associated with it, treatment guidelines suggest that it may be more important to identify patients at high risk of fracture than the OP patient properly.2 Therefore, fracture risk stratification is necessary to distinguish between patients with a high probability of presenting them from those in which this probability is lower. Risk assessment helps to identify individuals at higher risk and therefore guides therapeutic decisions as well, and this estimation probably produces a greater awareness on the part of the clinicians who care for these patients and results in a more complete intervention.

Although guidelines recommend14 drug treatment be assigned based on the risks and benefits in individual patients, there are no specific indication about the risks and benefits that should be assessed by the specialist and therefore, in many cases are difficult to apply in the clinical practice.

Through the OSTEOPAR project, using an online survey, we tried to identify the views of rheumatologists in Spain on various RF and especially what they mean by patients at “high risk of fracture” and how the management is carried out in clinical practice.

Respondents felt that previous fracture and age were the two most important standard RF, while a single value of BMD was considered of lesser impact. A previous fracture as the best predictor of fracture is consistent with the evaluation of other authors16 and the guidelines published by the SEIOMM,17 although the order of priority established by the respondents to prior fracture and age varies slightly from previous18 guidelines. Other studies have shown that in both men and women, age is an important fracture RF, independent of BMD.2,19 The secondary role assigned to the BMD by the respondents is consistent with the general view of the guidelines and the previous observation that many patients can present fractures despite having a relatively well preserved BMD.2,18,20 Although BMD has traditionally been considered a predictor of fracture, there is no accurate way to identify individuals who will suffer it.21 Indeed, the opinion of respondents agrees with previous analysis, in which it must be considered in conjunction with biochemical indices of remodeling and other laboratory values16,17,22 and that treatment should be considered rather in terms of risk of fracture given by the clinical RF rather than by the isolated value of BMD.2,20 In contrast to other published23 risk assessments, our findings tended to evaluate BMI as a moderate RF, even lower than height loss or a tendency to fall. It is possible to take into account that FRAX computes BMI as a proxy for BMD24 and therefore assigned a lower relevance to this parameter. The majority used FRAX-including scores used as a preventive strategy, a finding that is supported by the guidelines.25

Among other factors, it has been suggested that the presence of vertebral deformities and timing of fractures may be a first sign of weakness that identifies high-risk patients.5,26 In this sense, almost 90% of rheumatologists surveyed said an X-ray of the thoracolumbar spine was useful to assess the presence of vertebral fractures. While the vast majority of reviewers explore the spine in search of kyphosis or other deformities, they felt that both factors are secondary in importance to the number of fractures. This exploratory strategy is consistent with recommendations previously set, according to which physical examination and radiography are indicated for the detection of vertebral alterations.17,27

According to survey results, treatment usually begins in patients with previous osteoporotic fractures, which is consistent with the American (ACP) 14 and the European (NICE) guidelines.9

Teriparatide occupied a prominent position as a treatment option in the profiles of those considered at high risk for fracture (profiles 1 and 3), while oral bisphosphonates were the treatments of choice for the majority of respondents in medium-risk patients (profiles 2 and 4). The choice of teriparatide as a treatment in patients at high risk of fracture is supported by the pivotal trial (Fracture Prevention Trial), in which the drug significantly reduced the risk of new fractures at 18 months in patients with at least one fracture, mild vertebral fractures or two mild vertebral fractures.28 The high cost of treatment could explain a decision to reserve its use for these patients. In contrast, the use of raloxifene and ranelate in the high risk profiles was negligible, although its use is also recommended in these patients.17,27

In conclusion, the definition of OP patients with high risk of fracture in daily practice is not uniform, although the majority of rheumatologists considered the most important factor is a history of previous fragility fractures, followed by treatment with glucocorticoids, advanced age and a low BMD. Early identification of high-risk patients may facilitate the selection of more effective treatments that reduce the risk of future fractures. Among the Spanish rheumatologists, anabolic treatments are the therapeutic option of first choice in patients at high risk of fracture.

Enrique Casado, Manuel Caamaño, Jorge Malouf, Enrique Salas, Juan Sánchez-Bursón, Gabriel Herrero-Beaumont have no disclosures to make.

María Luz Rentero belongs to the medical department of Eli Lilly & Co., the company that manufactures teriparatide.