Qualitative synthesis of scientific evidence and consensus techniques have been used in this project which reflect the agreement by experts based on their clinical experience and scientific evidence.

A series of steps were taken during the development of the recommendations document, which were:

• 1.

  Creation of the work group. The document began with the formation of a panel of experts formed by 9 SER member rheumatologists. They were selected through an open call to all SER members. The Clinical practice Guideline and Recommendations Committee (Spanish acronym: GPC) of the SER assessed the curriculum vitae of all applicants, in accordance with objective criteria of contribution to knowledge on OP, taking into consideration publications in significant journals over the last 5 years. Coordination of clinical and methodological aspects was made, respectively, by one of these rheumatologists as the main researcher (MR) and a methodology specialist executive from the Research Unit of the SER.

• 2.

  Identification of the key areas for updating of the previous agreement. All workgroup members were involved in giving shape to the document, and in establishing its content and key aspects. First they identified the clinical questions of research which could have the greatest impact in offering information on OP management: diagnosis, assessment, prevention, treatment and special circumstances. After this, they decided which of these required a response through the formulation of a PICO question (patient, intervention, comparison, and outcome). The methodology to be followed in the recommendation creation process was also defined.

• 3.

  Literature review search. To respond to the non PICO clinical questions an updated systematic review (SR) search and GPC was made in MEDLINE and specialised sources in guidelines. The other clinical questions were reformatted in four questions with the PICO format. To respond to the PICO questions a search strategy was designed and a review of scientific evidence was made of studies published up until May 2017. The following databases were used: PubMed (MEDLINE), EMBASE and Cochrane Library (Wiley Online). The process was completed with a reference manual, posters and congress abstracts search which the reviewers and experts considered to be of interest.

• 4.

  Analysis and synthesis of scientific evidence. The SR and GPC identified for the non PICO questions were assessed by the methodology coordinator. It was agreed that only those of high quality were considered apt for their incorporation as a source of evidence. Two rheumatologists from the workgroup of SER evidence reviewers were responsible for systematically reviewing the scientific evidence available for the PICO questions. After a critical reading of the complete test of the selected studies for each review, they drew up a summary using a homogenised formula including tables and texts to describe the methodology, results and quality of each study. They made in-depth notes on the reasons for exclusion of articles not included in the selection. The overall level of the scientific evidence was assessed using the Scottish Intercollegiate Guidelines Network (SIGN) (see Appendix 1) evidence levels.

• 5.

  Formulation of recommendations. When the critical reading was completed, the MR and the panel of experts proceeded with the formulation of specific recommendations based on the available scientific evidence. This formulation was based on “formal assessment” or “reasoned judgement,” previously summarising the evidence for each of the clinical questions. They also took into account quality, quantity and consistency of the scientific evidence, the generality of the results, its applicability and its clinical impact. For the formulation of the recommendations two rounds of consensus were used. First, with the “reasoned judgement” system, where all of the experts drafted and discussed the recommendations. Then, in the presence of the methodologist, using the modified technique of nominal consensus, the level of agreement of the experts with the drafting of each recommendation was agreed upon. It was established that there was a high level of consensus when the percentage of experts who agreed with the recommendation draft was 75% or above. The agreement was evaluated using a dichotomous reply (yes: agreed with the wording; no: disagreed with the wording). Grading of the recommendations was made with the SIGN system (see Appendix 1). The recommendations were divided into four main areas: diagnosis and evaluation; treatment; special circumstances; new treatments.

• 6.

  Public exposure. The draft of this document of SER recommendations was submitted to a process of public exposure by the members of SER and different interest groups (pharmaceutical industry, other scientific societies and patient associations) aimed at receiving scientific evaluation and discussion of its methodology and recommendations. The complete information of this process is found in the appendix to the SER web site (www.ser.es), in the section under SER Research and Recommendations.

The document includes all the formulated recommendations subdivided into the different above-mentioned areas. Management algorithms were formulated from the recommendations and these offer a summarised approach to the treatment of OP.

OP is a diffuse skeletal disease characterised by a reduction in bone resistance which leads to a higher risk of fragility fractures. The concept of “bone resistance” encompasses factors relating both to bone mineral density (BMD) and the quality of the bone tissue.6

A fragility fracture is that brought on by a low impact trauma. A fall, standing up or being seated would be included in this concept. The most frequent and relevant fractures are those of the proximal femur, the spine and the distal forearm.7

OP may be defined in any of the following circumstances:

• a)

  T-score in spine columnar, femoral neck or total hip ≤−2.5 SD.

• b)

  Femoral fracture due to fragility, regardless of the BMD value, in postmenopausal women and in men >50 years of age.

• c)

  Fracture due to fragility of the spine, proximal humerus or pelvis in postmenopausal women and in men >50 years of age, if there is a low BMD (T-score <−1.0SD).

The conceptual definition of OP by the World Health Organisation (WHO) is solely based on the results of the BMD. An individual has OP if the T-score in lumbar spine, femoral neck or total hip is equal to or under −2.5 standard deviation (SD) compared with the population peak bone mass.8,9

In normal clinical practice this definition fails to be operative due to many fragility fractures occurring in patients with a T-score >−2.5 SD. For this reason the National Bone Health Alliance has proposed the consideration diagnosing OP in postmenopausal women and in men >50 years of age with femoral fracture caused by fragility, regardless of the BMD value, and also when there is a low impact spinal fracture, of the proximal humerus or the pelvis, if the T-score <−1.0 SD.10 Doubts exist regarding forearm fracture, which in some cases could be included depending on age or the mechanism of the fracture. The authors also included OP diagnosis for the American population a quantified risk with the Fracture Risk Assessment (FRAX®) for the primary fracture ≥20% (proximal humerus, forearm, hip or clinical spine fracture) or a risk of femoral fracture ≥3%.10

In view of the recommendations of these groups of experts and in accordance with clinical experience, the panel of experts considered it appropriate to include postmenopausal women and men >50 years of age with fragility fracture of the femur (regardless of BMD) in the OP diagnosis, as well as spinal fracture, fracture of the humerus or the pelvis, if it was accompanied by low bone mass (AL: 89%).

The main aim in OP management is fracture prevention and it is therefore essential to identify those individuals with the highest risk of presenting with them.11

Table 2 describes high fracture risk factors (RF) (relative risk ≥2) and moderate RF (relative risk 1–2).5

Several risk factors, such as age, fracture history and the use of glucocorticoids, among others contribute to the risk of fracture regardless of the BMD. Patients with a recent fracture are at a particularly high risk of suffering from a new fracture.11

The term “imminent fracture risk” has recently been proposed to refer to patents with a short term highs risk, such as those with a recent fracture, elderly fragile people with frequent falls or people treated with glucocorticoids at high level doses.12,13

Recommendation:It is recommended that evaluation of fracture risk due to fragility should not be exclusively based on BMD figures but also bear in mind clinical risk factors (RR: B; AL: 100%).

The combination of the BMD with clinical RF provides the best estimation of fracture risk.14,15 This has led to the development of fracture risk calculation tools which are able to include many factors. Among the most popular and widely used is the FRAX®.9

Recommendation:It is recommended that the tool FRAX®be used, with or without BMD, to assess the fracture risk (RR: √; AL: 78%).

The FRAX® calculates the probability of presenting a primary fracture (this include hip, clinical spine, humerus and forearm) or a hip fracture in the following 10 years, with the inclusion or non inclusion of the BMD score. When the BMD is included in the FRAX® tool the prediction of fracture risk is more precise and reliable.16,17

FRAX® limitations are mostly determined by insufficiencies in the data from which it was calculated. Thus the FRAX® does not consider the dose-response factor for several RF, nor lumbar BMD, nor the greater risk of fracture in patients with a recent fracture, and it does not encompass falls.9,16–18

The Spanish version of the FRAX® underestimates the risk of the primary fracture.19 However, the application of a decision algorithm with bone densimetry threshold indication (DXA) and treatment based on FRAX® has been shown to be cost-effective with regard to the exclusive use of T-score −2.5SD.20 Despite its limitations, the current version of the Spanish FRAX® may help to classify patients better with regard to fracture risk.

Recommendation:It is recommended that patients be classified as high fracture risk when the risk quantified by FRAX®for hip fracture is ≥3% (RR: √; AL: 100%).

Recommendation:To classify patients as high risk using the FRAX®for the primary fracture it is suggested that the threshold ≥10% without BMD be used or ≥7.5% with BMD (RR: √; AL: 78%).

The validity of the Spanish FRAX® was analysed in two prospective cohorts studies and in both the primary fracture risk was underestimated. The relationship between observed fractures and expected fractures was 2.2 in the FRIDEX20 study and 1.5 in the ECOSAP21 Study. However, the estimation of the Spanish FRAX® for hip fracture appears to be appropriate. A workgroup estimated the thresholds of the FRAX® in the Spanish population20,22,23 defining high primary fracture risk as a FRAX®, with or without BMD, ≥7.5% and intermediate risk between 5% and 7.5%. This was a single study which defined risk thresholds in Spain and had methodological limitations such as not being validated in regions other than Catalonia, including only 8.9% of subjects with intermediate risk and 8.5% with high fracture risk, as well as overlapping in the confidence interval of the high risk threshold. In this sense, the panel considers that a women aged around 70 years with a prior fracture (is approximately equivalent to as Spanish FRAX® for the principal fracture without BMD of 10%) is an example of high fracture risk.

From the scarce evidence and low level of agreement in classifying the patient as high fracture risk based on the FRAX®, the panel of experts agreed to consider a high risk fracture as FRAX® ≥3% for a hip fracture. Since no further studies on the FRAX® thresholds and/or national official positioning were available, the panel of experts also considered high fracture risk a FRAX® for primary fracture ≥10% without BMD or ≥7.5% with BMD.

The panel of experts also considered as high fracture risk, but without reaching an acceptable level of agreement (AL: 55%), the combination of two or more high RF. Should the FRAX® tool not be used, the presence of two high risk factors may serve as the starting point for assessing the BMD.

Recommendation:It is recommend that a basic analytical study be performed to rule out secondary causes of osteoporosis (RR: √; AL: 100%).

Basic laboratory tests to identify secondary OPs and make the differential diagnosis with other bone disease include30,42,43: haemogram, serum calcium and phosphate tests, alkaline phosphatase, test, proteingram, albumin, creatinine, liver function tests, calciuria and 25-hydroxy vitamin D tests.

Different scientific societies30,42,43 have produced documents with the following recommendations:

• •

  Additional tests will be carried out if there is suspicion of an associated condition:

  • 1.

    PTH (primary or secondary hyperparathyroidism).

  • 2.

    TSH (hyperthyroidism).

  • 3.

    Immunoelectrophoresis (myeloma).

  • 4.

    Antitransglutaminase antibodies (celiac disease).

  • 5.

    Urinary cortisol (Cushing's).

  • 6.

    Serum tryptase (mastocytosis).

  • 7.

    Sexual hormone study (young people).

  • 8.

    Genetic study (imperfect osteogenesis, hypophosphatase and others).

Recommendation:The routine use of bone turnover markers has not been established. However, they could be considered in initial assessment and in follow-up of patients with osteoporosis (RR: B; AL: 89%).

The reductions of the greatest magnitude in markers of bone turnover (MBT) are significantly associated to more marked reductions in the non vertebral fracture risk,44 and the initial changes in a formation marker predict the efficacy of the bisphosphonates (BP) in the reduction of vertebral fracture risk.45 However, there is not sufficient evidence to conclude that MBT are able to identify a reduction in risk fracture unequivocally.46

The majority of GPC recommend the consideration of the use of MBT in the initial evaluation and in follow-up, as an additional test. In the initial evaluation, the high levels may predict a faster loss of bone mass and a higher risk of fracture. But their main indication is in follow-up, since they can contribute to assessing treatment adherence and efficacy, and also contribute to monitoring the duration of therapeutic holidays.30,42,43,47 The groups of international experts conclude that further research studies are required prior to making a recommendation based on the evidence.43 To conclude, although the use of bone turnover markers has not been established as a routine, the panel of experts suggest that they could provide additional information in the initial evaluation and in the patient follow-up with OP.

The two main serum references of MBT which are recommended to use by the IOF and the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) are the precollagen type I N-terminal propeptide (PINP), bone turnover marker and the Carboxyl-terminal telopeptide of type I collagen (CTX), a reabsorption marker.48

Recommendation:It is recommended that the presence of vertebral fractures be assessed in the patient with a high fracture risk or with osteoporosis, when kyphosis or significant loss of height has presented and when there is back pain or low back pain of recent onset (RR: D; AL: 100%).

Chest and lumbar spine X-rays (focused on T7 and L3, respectively) are the method of choice for diagnosis of a vertebral fracture. A side projection is usually sufficient, thus minimising the radiation the patient receives.49 These are useful in the initial evaluation of all patients diagnosed with OP when there is a significant loss of height or a change in relevant statics (especially thoracic kyphosis) and dorsolumbar pain of recent onset with or without a history of trauma50 in a patient with a risk of OP or fracture. Height loss which is significantly associated with the presence of vertebral fractures is not defined, varying between 2cm compared with measurement in previous years to 4cm of historic loss since youth.30,51 The presence of a vertebral fracture, the number of them and their degree of severity are highly relevant indicators of the risk of new vertebral fractures, even those of the hip.52

The panel of experts has thus considered that the presence of vertebral fractures should be assessed in the evaluation of the patient with OP or with a high risk of fracture when kyphosis or significant loss of height is present and there are clinical features of vertebral fracture. Some DXA teams also use the Vertebral Fracture Assessment (VFA) to assess fractures.

Recommendation:Included in the prevention of primary and secondary osteoporosis and fractures is the recommendation for a healthy lifestyle, consisting of a balanced diet and regular physical exercise, not smoking, limiting alcohol consumption and using preventative measures against falls (RR: √; AL: 100%).

Several studies have concluded that the first step to the prevention of OP and the avoidance of fractures is to maintain healthy life habits.53–57 Evidence suggests that exercise has a modest effect on reducing fractures and preventing falls.57Table 3 contains a summary of the most important measures to be taken into consideration.

Studies show that the adoption of fall prevention measures, in which exercise programmes and actions on environmental safety and minimisation of the use of drugs such as hypnotics, diuretics, antihypertensives and others which impair balance, also contribute both to primary and secondary fragility fracture prevention.57,58

In contrast, exercises which require great effort or are abrupt and those which involve repeated flexion or twisting of the trunk and abdominals are ill-advised.

Regarding specific measures for the patient who has kyphosis, studies have proven the effectiveness of exercise, and training in muscular strength and balance.59

The effectiveness of hip protectors in the prevention of fractures is inconclusive, and adherence to their use is very low.60 Finally, vibratory platforms have not demonstrated any consistent efficacy.61

Recommendation:It is recommended that a daily intake of calcium between 1000 and 1200mg be taken and that it should mainly originate from a regular diet (RR: D; AL: 100%).

Recommendation:In the following cases a daily intake of 800UI of vitamin D is recommended (RR: D; GA: 100%):

• 1.

  Patients with osteoporosis.

• 2.

  People over 65 years of age with fracture risk.

• 3.

  People with a vitamin D deficit.

• 4.

  People with limited exposure to the sun.

• 5.

  People with an insufficient calcium intake (under 700–800mg daily).

Studies coincide in the need to ensure an adequate provision of calcium as an essential feature in any OP treatment programme.47 Thus patients in pharmacological treatment for OP must use calcium and vitamin D supplements because practically all clinical trials which have demonstrated the efficacy of the antiosteoporotic drugs routinely include calcium supplements and cholecalciferol (vitamin D3). When the calcium provided in the diet is insufficient, it is recommended that calcium and vitamin D supplements be taken. In postmenopausal women with OP the international expert groups advise an intake of 800–1.200mg/of calcium per day and 800UI/of vitamin D per day.62,63

In primary prevention a daily intake of calcium of 1000–1200mg for women >50 years of age and men >70 years of age is advised, preferably in the diet, and 1000mg/of calcium per day in women <50 years of age and men aged between 51 and 70.

Some foods are rich in calcium, such as milk products (milk, cheese, and yoghurt) tinned oily fish, beans and almonds. A list of foods and an online calculator are available in the IOF website.64,65 if supplements are used, the experts advise bearing in mind the optimum dose for calcium absorption which is situated in 500mg of calcium, and if larger doses are administered they should be spread over time.66

With regard to the possible side effects of calcium supplements, constipation and dyspeptic upsets are the most notable. An increased risk of nephrolithiasis was demonstrated (RR 1.17) in the Women's Health Initiative67 study, attributed to a diet rich in calcium of the women studied. However, a systematic review concluded that the calcium supplements in OP treatment, alone or in combination with other types of treatment, did not significantly increase the risk of nephrolithiasis or renal colic.68 Controversy exists regarding the possible increased cardiovascular risk associated with calcium supplements, particularly if the recommended maximum dose is exceeded. The available evidence suggests that a total daily intake of calcium (obtained by adding up dietary intake plus supplements) under 2000mg did not increase cardiovascular risk.69,70

Regarding vitamin D, in OP patients, the conclusion reached by the experts is to maintain minimum serum concentrations of 25-hydroxy cholecalciferol (calcidiol) of 30ng/ml. They therefore recommend measuring the calcidiol levels in patients with a risk of vitamin D deficiency and in patients with OP. Given the essential role of sunlight as a source of vitamin D, they advise an exposure to the sun of 10–15min per day.62,63 When necessary; they advise vitamin D supplements with a dose between 800 and 2000UI/day, depending on baseline levels. In patients with hepatopathies, malabsorption syndromes, treatment with anticonvulsives or other situations where the vitamin 25D hydroxylation may be compromised, administration of the metabolite calcidiol is recommended.

Current scientific evidence enables us to state that neither the increase of dietary calcium nor the isolated intake of calcium supplements is able to protect against the presence of fractures.71,72 Vitamin D administered as a monotherapy is not effective in reducing fragility fractures either, in non institutionalised elderly people.72,73 Calcium and vitamin D supplements have been show to be effective in the population >65 years of age who reside in care homes, reducing the risk of non vertebral fracture and marginally of hip fracture.72,74 This effect depends to a large extent on the vitamin D dose (≥800UI daily) and is more obvious at a greater age and with lower vitamin D levels.72 Calcium and vitamin D supplements do not reduce the risk of vertebral fracture.72

The efficacy of calcium and vitamin D supplements in the non institutionalised population >65 years of age (residents in the community) is questionable, as evidenced by the high number required to treat (NNT) to prevent fracture in this low risk population group74 and the results of a recent meta-analysis.75

There are contradictory data on the efficacy of vitamin D in the prevention of falls. Some results indicate that vitamin supplements, with and without calcium, would be effective in the prevention of falls,76 and especially in the elderly with vitamin D deficit.77 Other studies do not show this beneficial effect of vitamin D,78 and even the administration of annual high doses could increase the risk of falls.79

Recommendation:It is recommended that pharmacological treatment be initiated to reduce the Risk of osteoporotic fracture in the following situations (RR: √):

• 1.

  Fragility fracture of vertebra or hip in patients >50 years of age (AL: 100%).

• 2.

  Other fragility fractures in patients >50 years of age and with low bone mass (T-score <−1 S) (AL: 89%).

• 3.

  OP defined by T-score ≤−2.5 SD in spine, femoral neck or total hip, always evaluating age, BMD figures and other risk factors (AL: 100%).

• 4.

  Patients with a high hip fracture risk according to FRAX®with BMD ≥3% (AL: 100%).

• 5.

  Males in androgen deprivation treatment and T-score ≤−2.5 SD (AL: 100%).

• 6.

  Women in treatment with aromatase inhibitors and T-score ≤−2 or with a T-score <−1.5 SD and an additional risk factor, or with ≥2 risk factors without BMD (AL: 100%).

• 7.

  Patients in treatment with glucocorticoids and: (a) initial dose ≥30mg/day of prednisone; (b) postmenopausal women and men >50 years of age with a dose of ≥5 mg/day and previous fragility fracture or T-score ≤−1.5 SD or high risk of fracture: FRAX®for hip ≥3% or for primary fracture ≥10% without BMD or ≥7.5% with BMD (AL: 78%).

The drugs used for the treatment of OP have been shown to be more effective in patients with a high risk of fracture and in those with an existing fracture or with a T-score ≤−2.5 SD.80

The NOF recommendations and those of other expert groups support the use of FRAX® in selecting candidates for treatment, including cases without OP using DXA.30,80

The panel thus considers that treatment is indicated in patients with fragility fracture, OP from DXA T-score ≤−2.5 SD, high fracture risk (FRAX® for hip fracture with BMD ≥3%)23,34 and in sub-groups of patients who take glucocorticoids, aromatase inhibitors35 or antiandrogens.37 The specific circumstances where osteoporotic antifracture treatment should be assessed is commented upon in detail in other sections.

Regarding patients with fragility fractures, the NOF recommends treating all hip and vertebra fractures, whilst the other fractures would be treated depending on the results of the DXA and the risk of new fractures calculated using the FRAX®.30

Regarding the evaluation of treatment based on the DXA outcome with a T-score ≤−2.5 SD, the panel wishes to highlight that in these cases clinical risk factors and the FRAX® should also be considered when decision are to be taken.

Prophylaxis and treatment of OP induced by glucocorticoids are dealt with in a separate section.

Several international guidelines, including those on oncology, advise that treatment be indicated for OP in women who receive aromatase inhibitors and have a T-score below or equal to −2 SD or with a T-score <−1.5 SD with an additional risk factor, or with ≥2 risk factors without BMD.35 For men in androgen deprivation therapy, the IOF fixes the threshold for treating OP in a T-score of −2.5 SD or a raised fracture risk by FRAX®37 (Fig. 1).

Fig. 1.

Recommendations for initiating treatment for the prevention of fractures in patients with osteoporosis or risk of osteoporosis.

(0.22MB).

Recommendation:When selecting treatment, among other factors, it is recommended to take the following into consideration (RR: √; AL: 100%):

• 1.

  The efficacy and safety of the drugs.

• 2.

  Cost-effectiveness ratio.

• 3.

  The value of the BMD.

• 4.

  The presence of fractures, particularly vertebral or multiple fractures due to fragility.

• 5.

  Treatments prior to this and adherence to them.

• 6.

  Age.

• 7.

  Comorbidities and polymedication.

• 8.

  Limitations for oral administration.

The final aim of OP treatment is the prevention of fractures. Table 3 contains the efficacy in primary and secondary prevention of vertebral and non vertebral fractures of the drugs used in Spain.81–94

In the main studies conducted with drugs in patients who only had osteopenia without prevalent fracture (primary prevention) no fracture reduction occurred in primary outcome.

In primary fracture prevention in patients with OP, the drugs which demonstrated a reduction of vertebral fracture were: alendronic acid, zoledronic acid, raloxifene, bazedoxifene and denosumab, whilst denosumab also reduced non vertebral fracture and hip fracture in this group.

In patients with OP and vertebral fracture, alendronic acid, risedronic acid and zoledronic acid reduced new vertebral, non vertebral and hip fractures, whilst teriparatide reduced vertebral and non vertebral fractures and denosumab reduced vertebral fractures. For their part, ibandronic acid, raloxifene and bazedoxifene reduced vertebral fractures in patients with previous fracutre.81–94 With regard to forearm fractures, alendronic acid demonstrated a significant reduction of new wrist fractures in meta-analysis in patients with previous vertebral fracture.81

Tables 4–7 contain details of the risk reduction for different fractures in the published studies.83–94

In patients with at least two moderate vertebral fractures or at least one severe fracture teriparatide was significantly more effective than risedronic acid for reducing new morphometric and clinical vertebral fractures after 24 months of treatment. There were no significant differences in the rate of new non vertebral fractures.95

Strontium renalate has not been included in these recommendations because its production has been suspended, nor calcitonin, because of the limitation its technical record has in treatment indication and duration. However, therapy with oestrogens reduces vertebral and hip fractures, but is currently not used due to its poor risk/benefit profile and because safer treatments exist.

The following aspects have to be taken into consideration when integrating trial outcomes into clinical practice: (1) the patient profile included is a postmenopausal woman with a high baseline risk due to age and the presence of previous fractures; (2) the populations of the trials are different to one another, and therefore may not be directly compared; (3) the studies were conducted with calcium and/or vitamin D supplements and (4) real treatment adherence is lower than that reported in the trials. Moreover, treatment should be individualised according to the circumstances of each patient.

Another aspect to take into account is the cost/effectiveness ratio of treatment. The studies undertaken in other countries fix the threshold of FRAX® for primary fracture, so that the cost-effectiveness of the treatment is 8.8% (Portugal),96 13.8% (Switzerland)97 and 10%–15% (Greece).98 For estimation the cost of alendronic acid was considered (Portugal and Switzerland) and the average cost of treatment for OP (Greece). When drugs other than alendronic acid were considered, the FRAX® threshold for cost effectiveness was higher.96–98 Despite no cost-effectiveness studies existing for Spain based on fracture risk, the panel of experts advised taking into account BP as first line option in OP treatment.

The panel of experts therefore consider that when choosing treatment to prevent fractures the efficacy and safety of the drugs, cost-effectiveness ratio, BMD figures, and presence of previous features (location, number and time passed since fracture) and oral administration route limitations should be contemplated. Other factors which also were to be taken into account were previous treatments and adherence to them, co morbidities and patient preferences. Thus, in patients with several fractures, oral intolerance, dementia, malabsorption and poor adherence, parenteral therapy could be considered.

The risk of long-term complications in OP treatments99–101 and the increase in the risk of fracture after their suspension102,103 often obliges physicians to put forward a sequential treatment strategy.

Although there are currently no outcomes on fracture reduction with sequential treatment,104,105 the bone forming sequence (teriparatide) followed by antiresorptive sequence (denosumab or BP)104–106 is the most effective in terms of gaining BMD. Teriparatide followed by denosumab appears to be more effective than teriparatide followed by BP.107

The administration of teriparatide after an antiresorptive drug is also effective, although its anabolic effect may slow down (less change in bone markers and lower increase of BMD),108 and if administered after denosumab it has been described as a loss of initial BMD in the hip.106 Sequential treatment with romosozumab followed by an antiresorptive drug is dealt with in the section relating to new treatments.

Little information exists on the use of one antiresorptive drug followed by another, especially if the first was administered for years. A review of 11 prospective studies of patients previously treated with alendronic acid or risedronic acid showed that the change to ranelate from strontium or denosumab achieved an additional increase in BMD.109 However, no data were available on fracture reduction either.109 In one study denosumab showed increases in BMD in patients who had been treated with zoledronic acid.110

There are studies which show greater efficacy in combined therapy (bone formation and antiresorptive) compared with monotherapy. The combination of zoledronic acid and teriparatide has been associated with a higher increase in femoral BMD one year after treatment compared with teriparatide in monotherapy.104,111 The combination of denosumab and teriparatide has been associated with a greater increase in lumbar spine and proximal femur BMD, compared with both drugs in monotherapy.112 Notwithstanding, there is no evidence of reduction of fractures with these combinations. After a review of the literature, the panel concluded that combined (bone formation and antiresorptive) treatment did not appear to be an option to recommend in general, although its use could be justified in highly selected serious cases of OP.

OP is a chronic disease, and treatment must therefore be maintained indefinitely or at least for several years. However, there is no recommended maximum treatment for each drug used.42,100

The antiresorptive drugs (selective oestrogen receptor modulators, SERM, BP and denosumab) generally reduce fracture risk from 12 to 18months88,91,113 and have no maximum approved duration.

In the case of BP, its efficacy, in terms of BMD and fracture risk reduction is maintained for at least 3 years of treatment.114,115 Some studies show that from the 4th or 5th year BMD slightly rises or remains stable and the risk of some complications increases so that after 5 years of treatment with oral BP or 3 years with zoledronic acid it is advisable to evaluate the risk/benefit of continuing with treatment.116 The Groups of experts agrees in recommending maintaining treatment (up to 10 years for oral BP or up to 6 years for zoledronic acid) in patients with a high fracture risk (fractures prior or during treatment, age >75 years or T-score ≤−2.5 SD in femoral neck or total hip).42,100,117,118

Denosumab produces a progressive increase of BMD for at least 10 years of treatment, maintaining its ant fracture efficacy over time.119 Although there is no consensus regarding the duration of this treatment, some authors indicate that it could be established according to a predefined objective (treat to target strategy), i.e. after reaching a certain BMD value.120 however, they should not suspend treatment with denosumab without establishing other treatment for OP (see section “Suspension or discontinuation of treatment”).

Although data exist on efficacy and safety of SERM (raloxifene and bazedoxifene) up to 8 years, maximum duration of the treatment is conditioned by the risk of complications such as thromboembolism, which is more frequent in patients over 70.121

Regarding patients who receive glucocorticoids, aromatase inhibitors or androgen deprivation therapy, the panel of experts consider that antiresorptive treatment should be maintained at least whilst the patient receives the inductor drug of OP (see “Osteoporosis by glucocorticoids”).

Teriparatide has been shown to reduce the risk of fracture after 6–12months of treatment initiation. However, to achieve maximum efficacy, both in trabecular and cortical bones, different groups of experts agree that it is recommendable to maintain it for 2 years, which is the maximum period approved.42,100,108

OP treatment monitorisation includes the analytical determinations and the DXA, as well as asking the patient about new fractures. Several studies conclude that the bone remodelling markers could be useful for early monitoring of compliance and response to treatment,139 but the panel of experts consider that they cannot recommend their systematic determination in patient follow-up. The panel recommends monitoring the response to treatment through central DXA, bearing in mind patient characteristics.

There are no controlled, good quality studies on the frequency with which measurement of the BMD should be repeated during treatment.140 Possible variation in BMD with current treatments is slow to occur and usually of little magnitude. Given the risk of DXA measurement error, international experts concluded that measurement should be made when the expected change is equal to or higher than the minimum significant difference.32 For this reason, in general, it is not advised to repeat measurement of the BMD before 2–3 years.5,141 Other guidelines advise lengthening this time period to 3 years for zoledronic acidy and up to 5 years for oral BP.126,142,143 In situations of high fracture risk,5 in treatment with high doses of glucocorticoids or when there is a suspicion of failure in therapy, such as when a new fracture occurs, the period for repeating DXA may be shortened.144 The aim is to detect the patients who, despite treatment, suffer from a significant drop in BMD. It is also advised to carry out a DXA when planning temporary suspension of the drug for “therapeutic holidays”.124 Although the greatest changes in BMD are detected in the lumbar spine, it is useful to also monitor the hip, as it is less dependent on artefacts impacted by degenerative changes.

Several authors propose using a treat to target strategy in OP where the objectives would be to reach a certain value of the T-score (for example, T-score: −2.5 SD) or BMD. If this occurred, monitorisation of the BMD would enable the knowledge of whether the therapeutic aim had been reached. However, not all drugs lead to an improvement of BMD proportional to the reduction of fracture risk, and the real value of this strategy in clinical practice is also yet to be defined.117

Determination of TBS provides information on the changes which treatment may produce on bone microarchitecture,145 but insufficient data is available to recommend its use in clinical follow-up.

The definition of therapeutic failure in the treatment of OP is complex and there are no established criteria. Experts agree that it is advisable to ensure that treatment adherence is correct, confirm an adequate supply of calcium and vitamin D and rule out secondary OP causes. With regard to treatment failure, after one year of treatment with any antiosteoporotic drug, a group of international experts suggest the following criteria of assessment for treatment changes144:

• 1.

  Two or more fragility fractures during treatment.

• 2.

  One fragility fracture together with one of the following factors:

  • -

    Significant loss of BMD (>5% in lumbar spine or >4% in proximal femur).

  • -

    Absence of significant changes in the MBT (reduction of PINP or CTX >25% with antiresorptive treatment, or increase >25% with bone turnover treatment after 6 months).

• 3.

  Absence of significant changes in the MBT together with a significant loss of BMD.

In general the drugs approved for OP treatment are safe and present with good tolerability.146 Oral BP may lead to a series of adverse digestive effects (pyrosis, dyspepsia, esophagitis, dysphagia or abdominal pain) and oral effects (conjuntivitis, anterior uveitis), of minor intensity in most cases, which disappear when the drug is withdrawn. After the first infusion of zoledronic acids symptoms of unclarified origin have been described characterised by fever, myalgias and bone pains and this occurs in between 20% and 30% of patients. In subsequent infusions the symptoms tend to disappear. This effect may occur rarely with oral BP.

Treatment with BP does not increase cardiovascular risk. No association has been demonstrated between the development of auricular fibrillation and treatment with oral BP. In one study an increased risk of auricular fibrillation was observed with the use of intravenous zoledronic acid.147 Isolated cases of cancer of the oesophagus have been published in patients on oral BP, but the most recent studies have not confirmed this relationship.148 Osteomuscular pain, kidney damage and hepatotoxicity from BP are exceptional and will rarely lead to withdrawal of the drug.149 Use of BP is not advised in patients with glomerular filtration rate <30ml/min, since there are no available studies that have demonstrated a reduction in fractures in these patients and on the contrary, they increase the risk of impairment of a reduced remodelling osteodystrophy.

Denosumab, in the guideline approved for the treatment of OP, is a drug which is generally well tolerated. A higher rate of infections has been observed, and in particular cutaneous88 and urinary,150 but their overall rate is very low.151 However, due to their structure, they are not eliminated by the kidney, which is an advantage for patients with kidney diseases where a higher risk of hypocalcaemia must be monitored. As with BP, denosumab should not be used when there is clinical suspicion or histomorphometric evidence of reduced remodelling osteodystrophy.

Adverse reactions to teriparatide are generally not serious. Muscle pain, cramps and dizziness are frequent. Treatment with teriparatide may raise calcaemia and calciuria, and it is therefore advisable to determine the levels of calcium in the blood and urine prior to initiating treatment.

In recent years complications have appeared which are associated with prolonged treatment with antiresorptives (BP and denosumab) and which have generated uncertainty about safety. The relative importance of these rare adverse effects (osteonecrosis of the jawbones and atypical femur fracture) in the treatment of OP is a highly controversial issue.

The fear of patients in suffering these complications may be an impediment for adherence and therapeutic compliance, although the benefit–risk balance of maintaining treatment is highly favourable for high fracture risk patients. A brief analysis will now be made of both complications, based on the systematic reviews carried out to complete this document.

Recommendation:If any invasive dental procedures has been indicated to the OP patient (dental extraction or implant) it is recommended that antiresorptive treatment initiation be postponed until the surgical wound has completely healed (RR: D; AL: 100%).

Recommendation:In patients treated with antiresorptives who are going to carry out a dental procedure it is not recommended that they discontinue treatment with BP or denosumab. If other additional osteonecrosis of the jawbone risks prevail or the surgical procedure is to be extensive, temporary suspension of the treatment with BP may be considered (RR: D; AL: 78%).

Osteonecrosis of the jawbones has been associated with the prolonged use of BP or denosumab, although its rate in patients with OP is very low (between 1/10,000 and 1/100,000).152–154

According to the outcomes of a previous review, published in 2013, available evidence on BP is scarce and of low quality.154 There is insufficient data to confirm that oral BP or intravenous BP used for OP treatment confers any significant risk of osteonecrosis of the jawbones. There have not been enough studies to assess the role of denosumab in the development of osteonecrosis of the jawbones (level of evidence 3).152,153 Several systemic factors such as previous treatment with BP or dental extractions are associated with the development of osteonecrosis of the jawbones in patients treated with denosumab (level of evidence 3).153

Several medical associations have published recommendations to reduce the risk of this complication. In general, guidelines include correct dental hygiene in their proposals and dental check-up if poor dental health is observed. If any invasive dental procedures has been indicated (dental extraction or implant) it is recommended that treatment initiation be postponed until the surgical wound has completely healed.155 The use of CTX marker thresholds have been suggested to assess the baseline risk of osteonecrosis of the jawbones. However, available evidence does not endorse its use.156

There is some controversy regarding the which approach to take in patients who are already taking BP or denosumab. The majority of guidelines advise not to suspend antiresorptive treatment in non oncological patients. In 2011, the American Dental Association guide recognised that the risk of osteonecrosis of the jawbones was very low in patients with OP, and it was therefore not necessary to suspend an oral BP prior to a dental procedure.157 However, and according to the recommendations made by the international panel of experts, in patients who have undergone extensive oral surgery and with associated risk factors (diabetes, periodontal disease, immunodeficiency, tobacco habit), clinical judgement may suggest temporal suspension of treatment.155 In this situation, and given that withdrawal of denosumab leads to a bone remodelling and the risk of appearance of multiple vertebral fractures, the panel considers that treatment with denosumab should not be interrupted.

During OP treatment, invasive dental procedures may be made only after careful consideration of their necessity. In the case of denosumab, and whenever possible, surgery must be avoided in periods close to drug administation.158 In this situation a prolonged unnecessary suspension should be avoided, especially in patients treated with denosumab. Should a relevant adverse event occur such as osteonecrosis of the jawbones, and although no scientific evidence demonstrates withdrawal of the drug will help the process evolve, the same document considers it prudent to suspend treatment and assess drug indication with a different mechanism of action.155

Recommendation:In patients with high fracture risk and prolonged treatment with bisphosphonates (>5 years for orals or >3 years for intravenous) it is recommended not to suspend treatment, since the risk of atypical femur fracture is very low and the benefits of fracture reduction greatly outweigh the risk of atypical fracture (RR: D; AL: 100%).

Atypical fracture of the femur is a potential complication from prolonged treatment with BP or denosumab, although real incidence is very low.

According to the definition proposed by the American Society for Bone and Mineral Research (ASBMR),159 the atypical fracture of the femur is located between the lesser trochanter and the suprachondylar crest and should present with a minimum of 4 out of the following 5 criteria: (1) minimum or absent trauma; (2) origin of the fracture in the external cortical and directed cross-sectionally or obliquely; (3) without comminution or, if this exists, it must be minimal; (4) periostic or endostic thickening of the external, localised cortical, and (5) involvement of the external side cortical (incomplete fracture) or both corticals (complete fracture). Minor criteria are not essential for diagnosis, but they increase it: (a) cortical thickness increased diffusely in femoral diaphysis; (b) pain in thighs or groin area, prior to fracture; (c) bilateral fracture of similar characteristics, and (d) delay in consolidation.

Apart from prolonged treatment with antiresorptives, the atypical fracture of the femur has also been associated with several comorbidities or with taking medication such as proton pump inhibitors or glucocorticoids.

According to the results of a previous 2011 review, the rate of atypical fracture of the femur was very low and it could not be confirmed that there was a significant increase in the risk of atypical fracture of the femur in patients with Op treated with BP. Since then several poor quality studies have been published which show that the rate of atypical fracture of the femur as very low. Two of the 3 published meta-analyses,160,161 based on randomised clinical trials (RCT), cohort, case and control studies, with a high degree of overlapping, demonstrated a significant increase in the risk of subtrochanteric or diaphysary fracture with the use of BP ranged between 1.70 and 1.99, although with a high level of heterogeneity. In a third meta-analysis which only explored alendronic acid,162 the relative risk of atypical fracture of the femur with this drug was 3.23 (95% CI: .88–11.84), but the differences were not significant. When a stratified analysis was made with the studies to assess the effect of alendronic acid over 5 years, heterogeneity was insignificant. Using a model of fixed effects, the relative risk was 2.55 (95% CI: 2.26–2.88), concluding that the alendronic acid did not increase the risk of atypical fracture of the femur short-term, but it did in the long term (>5 years).162

A recent systematic review162 concluded that the rate of atypical fracture of the femur associated with BP was very low, but that the relative risk (RR) increased with prolonged use (especially if it was over 3 years). According to the authors, the benefits of BP in reduction of OP fractures greatly outweigh the risk of atypical femur fracture. A limiting feature highlighted was that almost all studies were retrospective and of variable duration (4–9 years), limiting the interpretation of long term exposure.

Regarding the RCT,163,164 isolated cases have been reported on atypical femur fractures, but these studies present with relevant limitations, such as the short duration of the same, the non-inclusion of this outcome as a main variable or the losses in follow-up, which impede the obtainment of reliable outcomes. If we analyse observational studies,165,166 the incidence and absolute risk of atypical fracture of the femur in patients treated with BP have very low rates (between 10.8 and 19.1 per 100,000 patients per year).

The rate of atypical femur fracture represents 0.3% of the total of femur fractures or 5% of the total of subtrochanteric/diaphyseal fractures. Women represent 80% of all diaphyseal fractures and 95% of the atypical fractures of the femur, with a relative risk which is higher in men. The outcome of recent review studies and meta-analyses suggest a favourable benefit/risk ratio to maintain treatment up until 10 years with antiresorptives.

Regarding denosumab, the systematic review results on the risk of atypical femur fracture are very limited due to the scarcity of high quality studies. The studies included were one RCT,147 two follow-up studies167,168 and one cohort study.169

According to observational studies, long-term treatment with denosumab (between 7 and 10 years) is associated with a nil168 or very low rate of atypical fracture of the femur (.8 per 10,000 patients per year).167 A clinical trial of postmenopausal women with OP who had switched from BP to denosumab was unable to demonstrate that the rare cases of atypical fracture of the femur which appeared early on could be attributable to the use of denosumab, in contrast they appeared to be derived from prior use of BP.160

Although cases of atypical femur fracture have been described in patients treated with denosumab, the panel of experts decided not to make a specific recommendation on the risk of atypical femur fracture with denosumab due to low evidence and the need for further studies with a higher number of cases and longer time period.

To reduce the clinical impact of atypical femur fracture, in patients under prolonged treatment with BP or denosumab attention must be paid to the presence of pain in the thighs. In these cases it is recommended that X-rays be performed to detect stress fractures or cortical thickening and for early determination of atypical femur fracture. It may also be useful to use magnetic resonance or a bone scan to confirm diagnosis.

Treatment of an atypical femur fracture is surgical. When this complication arises, studies conclude that antiresorptive treatment should be withdrawn, the pain treated, an adequate provision of calcium and vitamin D ensured and a drug with a different mechanism of action be indicated.170 No control studies exist On the efficacy of teriparatide in patients who have suffered an atypical femur fracture. In a short series of cases, treatment with teriparatide increased the bone mass and the bone remodelling markers, but with no consistent effect on fracture consolidation.170

Recommendation:It is recommended that all patients’ ≥50 years of age with a recent fragility fracture be assessed systematically to prevent further fractures (RR: D; AL: 100%).

Patients with fragility fracture have a significant increase in the risk of further fractures. However, it is normal that fewer than 25% of them start treatment for OP after a fracture.4 The term “imminent fracture risk” has recently been introduced to refer to patients with recent fractures, fragile elderly people with frequent falls and fracture patients who have started treatment with high doses of glucocorticoids.12

In keeping with the EULAR171 recommendations the panel of experts suggested that all patients with recent osteoporotic fracture aged ≥50 years of age should be systematically assessed to prevent new fractures.

The FLS model, or Fracture Unit appears to be the most effective172 in terms of evaluation and treatment initiation. It is also cost-effective173 and positive experiences have been had with it in Spain.136 The central element of a Fracture Unit is the coordinator (in Spain this is usually a medical bone metabolism expert), who works in close collaboration with a specialised nurse. The keys to success in the Fracture Unit are: (a) identification of all cases, usually from a list of emergencies and of hospitalised patients; (b) study of the patients, including determination of risk factors, DXA and analysis; (c) the advice of the nurse on non pharmacological measures and adherence to medication; (d) early initiation of antifracture treatment and (e) the drawing up of a report with precise indications for the primary care physician.

In patients with fragility fracture the panel of experts concludes that it would be useful to assess fracture risk factors and falls and identify secondary OP. They advise the carrying out of DXA and assessment of the presence of vertebral fractures.

Recommendation:The treatment of choice for an acute vertebral fracture is rest, rapidly scaled analgesics to control pain, and if necessary the use of a brace (RR: √; AL: 100%).

Acute vertebral fracture may lead to serious pain and disability, lasting for several months.174 General measures that have been shown to be useful and shorten the recovery period are bedrest (for 4 days or less) and rapidly scaled analgesics (including the use of powerful opiates when necessary).175

Ortheses (rigid or semi-rigid brace) are useful for acute episode in combination with analgesic treatment, as they help to immobilise the fracture, reduce load and improve spinal alignment in patients who can tolerate them. The type of orthesis will depend on fracture location. Prolonged use is not advisable. It has not been demonstrated that the rate of new fractures will decrease.176

In patients with persistent and highly intense pain (visual analogue scale [VAS] ≥7), after 4–6 weeks of treatment as described above, some studies concluded that facet infiltrations could be considered or radiofrequency ablation of the lumbar medial branch,177 before prescribing a vertebroplasty or kyphoplasty. According to other studies, in the sub-acute and chronic phase it would be beneficial to have physical therapy using heat applied techniques, ultrasound or hydrotherapy and a progressive initiation of an exercise programme to strengthen the paravertebral and thoracoabdominal muscles.178 Weight-bearing exercises and balance training prevent falls and fractures and improve the patient's quality of life.59,176

Recommendation:In patients who receive or who are going to receive treatment with glucocorticoids fracture risk must be assessed and the initiation of treatment for osteoporosis as soon as possible (RR: D; AL: 100%) (Fig. 2).

Fig. 2.

Recommendations to initiate prophylaxis or Op treatment in patients who will receive glucocorticoids in the following 3 months.

(0.18MB).

Recommendation:It is recommended that pharmacological treatment be initiated to prevent OP in patients who are going to receive glucocorticoids for over 3 months in the following cases (RR: D):

• 1.

  Patients with an initial dose ≥30 mg/day of prednisone. In these cases it is recommended that treatment be imitated for the OP immediately (AL: 100%).

• 2.

  Postmenopausal women and men >50 years of age with doses of prednisone ≥5mg/day who also present with some of the following conditions:

  • a)

    Previous fragility fracture (AL: 100%).

  • b)

    Low BMD (T-score ≤−1.5 SD in spine or hip) (AL: 89%).

  • c)

    High fracture risk: FRAX®for hip ≥3% or for primary fracture ≥10% without BMD or ≥7.5% with BMD (AL: 78%).

• 3.

  Premenopausal women and men <50 years of age and dose ≥7.5mg/day of prednisone who also present with some of the following conditions:

  • a)

    Previous fragility fracture (AL: 100%).

  • b)

    Low BMD (Z-score ≤−3 SD) (AL: 78%).

OP induced by glucocorticoids is the most frequent cause of secondary OP, and is associated with the presence of fractures in up to 30%–50% of cases.182 The risk of fracture depends on factors such as age, prior BMD, daily accumulated dose of glucocorticoids and baseline disease. The loss of BMD is fast, especially in the first 6–12 months, even with low doses of glucocorticoids, with trabecular bone being the most affected.182 According to studies, the fractures associated with treatment with glucocorticoids appear with BMD which is not too low, due to the fact that the glucocorticoid mainly affects the whole bone micro architecture, and the TBS and FRAX® adjusted with TBS could have a certain additional usefulness in the assessment of the risk of these patients.183 However, the application of the FRAX® in OP through glucocorticoids has certain limitations and underestimates the real risk of fracture, since it does not adjust from accumulated dose from daily dose or treatment duration.

ACR recommendations consider that in patients who receive or are going to receive treatment with glucocorticoids the fracture risk must be assessed (preferably using FRAX®) and treatment initiation as soon as possible should be assessed. Furthermore, general preventative OP measures would have to be used in all patients, such as healthy life habits for bones and diet (see section on non pharmacological treatment, calcium and vitamin D). Furthermore, the lowest possible dose of glucocorticoid should be used and for as short a time as possible, and using immunosuppressants if necessary.36

In patients with an intermittent corticoid regime and who have an accumulated dose of prednisone >5g/year or those with annual losses of BMD above 10% the panel of experts consider that treatment for OP would also be indicated.

Premenopausal women or men <50 years of age are at a lower fracture risk. There are also fewer data on efficacy and safety or OP treatments for this population group and treatment indications appear to be more restrictive.

The pharmacological treatment of choice is oral BP (risedronic acid and alendronic acid) or intravenous zoledronic acid since they stabilise or increase the BMD and reduce the risk of vertebral fractures in comparison with control groups.184 According to the technical specifications, the drugs which currently have an glucocorticoid OP indication in Spain are teriparatide, risedronic acid and zoledronic acid.

In women with established OP, teriparatide has been demonstrated to be more effective than risedronic acid in reducing vertebral fractures and clinical fractures, regardless of having received previous treatment with BP.185 In fact, teriparatide has demonstrated superiority against alendronic acid in both gain in BMD and reduction in morphometric vertebral fractures, in premenopausal, postmenopausal women and men treated with glucocorticoids after 18 months of follow-up.186 As a result, the panel of experts consider that high risk patients or those refractory or who have contraindications to antiresorptive treatment could evaluate the treatment with teriparatide.47

In patients who start or have already received glucocorticoids, the administration of denosumab for 12 months has shown an increase in BMD which is significantly higher than with risedronic acid,187 and the European Medicines Agency (EMA) has recommended that denosumab be added into the technical specification as a new indication: for treatment of the loss of BMD associated with the prolonged use of glucocorticoids in adult patients at high risk of fracture.

According to the recommendations of the International Groups, the treatment of choice should be maintained for as long as treatment with glucocorticoids lasts and for at least 6 months after suspending the glucocorticoid, unless fractures exist, in which case it would be continued.36

Recommendation:It is advisable to assess the risk of fracture and consider performing DXA in patients with rheumatoid arthritis, systemic lupus erythematosus and ankylosing spondylitis especially those over 50 years of age and who are being treated with glucocorticoids or with a severe or evolved disease (RR: √; AL: 100%).

Patients with rheumatoid arthritis, systemic lupus erythematosus and ankylosing spondylitis are at greater risk of OP and/or fractures.188–194 One study reported an increased fracture risk in psoriasis arthritis and in severe psoriasis.195 The conclusions of some studies in patients with rheumatoid arthritis advise screening for fragility fracture and performing rheumatoid DXA, especially in patients >50 years of age in those with severe disease or evolved disease or those treated with glucocorticoids.188,189 Moreover, several documents of consensus on the management of comorbidities in rheumatoid arthritis, psoriasis arthritis and systemic lupus erythematosus suggested making an accurate prophylaxis of OP.196–198 In relation to antifracture treatments, all the drugs approved for OP were used. In the case of denosumab, safety data was limited in patients treated with biologic agents for their baseline disease, although according to the outcome of some studies there does not appear to be a different safety profile from the postmenopausal OP one.199

Recommendation:The study of the causes of secondary osteoporosis in all men with low bone mass or fragility fracture is recommended (RR: D; AL: 100%).

It has been observed in different studies that OP in the male is underdiagnosed.200 in fact it has been estimated that a third of hip fractures worldwide occur in males, especially after the age of 70 years. Furthermore, mortality after a hip fracture (over 37% in the first year) is higher than that in women.201 Also, in the male secondary OP is more frequent and studies therefore conclude that appropriate, accurate clinical assessment is needed for systematic exclusion. The most frequent OP causes and risk factors in the man are hypogonadism, alcohol consumption, tobacco and glucocorticoids. Fracture risk assessment in the male would include clinical risk factors, the existence of previous fractures and the measurement of BMD by DXA (if indicated) to better stratify those patients more susceptible for treatment.202 With regard to DXA, the same reference values are used in men as in women.203

With regard to non pharmacological prevention measures, the same may be considered as for OP in women. For hormonal treatment, there are no data on fracture reduction efficacy with testosterone treatment, and the experts therefore only indicate treatment in males with symptomatic hypogonadism and also advising it should be combined with another antiosteoporotic treatment.204 The drugs which have shown improvement of BMD and reduction of vertebral risk fracture both in eugonadal and hypogonadal males are oral BP (alendronic acid and risedronic acid),205,206 zoledronic acid (the only one which was established as the primary aim)207 and teriparatide.208,209 Denosumab was also approved in males with a high fracture risk and to reduce the loss of BMD in the patient with prostate cancer in androgen deprivation treatment.210 All these drugs, except the alendronic acid, have an indication in Spain for OP in the male.

Recommendation:It is recommended that a study of secondary osteoporosis causes be made in all premenopausal women with a low bone mass measured by DXA or with a fragility fracture (RR: √; AL: 100%).

According to the different scientific societies and expert groups, in the premenopausal woman a DXA should be carried out when there are fragility fractures and/or risk factors associated with the loss of bone mass.42,211 In this patient group the Z-score would have to be used (which compares with individuals of the same age and sex) instead of the T-score.42,211,212 Thus, “low bone mass” is considered when the Z-score is lower than −2 SD.211 The presence of fragility fractures and particularly when associated with low bone mass, means that OP diagnosis may be established in the premenopausal woman.212

Considering that over 50% of premenopausal women with OP present with a secondary OP212,213 the experts recommend making an exhaustive study to rule out an underlying cause,212,213 with the most frequent being the following: treatment with glucocorticoids, oestrogen deficit, malabsorption diseases and OP associated with pregnancy or breast feeding.212 Idiopathic forms are frequent in premenopausal women and may be associated with hypercalciuria and with a family history of OP.213

The therapeutic approach in the premenopausal woman includes optimum calcium and vitamin D intake, regular physical exercise, suspending tobacco consumption, limiting alcohol consumption and treating the underlying cause.212

For premenopausal OP experts, the specific antiosteoporotic drugs (BP, teriparatide, denosumab) will be considered in patients with fragility fracture or for certain secondary causes such as treatment with glucocorticoids.213 The drugs which they have studied in premenopausal OP are BP (alendronic acid and risedronic acid) and teriparatide.212 According to experts, in young women certain caution should be used in the administration of BP given its long retention in the bone tissue, particularly in women who wish to get pregnant in the future. They also conclude that none of these drugs should be used in pregnancy (BP and teriparatide are in the C category according to the FDA).214,215 With regard to the SERM, it seems reasonable to avoid its use in premenopausal women, whilst with denosumab there is only an indication in postmenopausal OP (according to the FDA: category in pregnancy).212

Some experts suggests that in fertile women anti conceptive measures should be prescribed if any antiosteoporotic treatment is initiated.213