1. Fracture risk factors

The main factors associated with the risk of bone fractures in patients presenting with osteoporosis include gender, age, bone mineral density (BMD), history of fragility fracture, history of hip fracture in a first-degree relative, and low body weight (i.e., body mass index [BMI] <20 kg/m²). Paradoxically, obesity can also be a risk factor for some peripheral fractures, including those of the humerus and distal third of the radius. Recognised risk factors also include various diseases including hypogonadism, early menopause, prolonged amenorrhea, anorexia nervosa, malabsorption, rheumatoid arthritis, diabetes (particularly type 1), immobilization, as well as their treatments, e.g., glucocorticoids, inhibitors of aromatase or gonado-tropin-releasing hormone agonists^2,3. Other disorders and medications that may be associated with the development of osteoporosis, (although probably less strongly) are hyperparathyroidism, hyperthyroidism, selective serotonin reuptake inhibitors, proton pump inhibitors, and anticonvulsants, as well as smoking and excessive alcohol consumption. Calcium deficiency and vitamin D deficiency have traditionally been considered risk factors for osteoporosis, although their precise role continues to be a subject of debate (Table 1).

Table 1. Diseases and treatments that constitute risk factors for osteoporosis 

Factors associated with an increased risk of falls, including postural instability, inability to get up from a chair, visual impairment, and some neurological problems are also associated with an increased risk of fractures.

After a first fracture, the greatest risk of sustaining a new fracture occurs within the first two years, particularly if the first fracture was vertebral⁴-⁶. This has led to the concept of an "imminent risk" of fracture. The main factors that have been associated with imminent risk are older age, female gender, white race, recent fracture, falls, and some comorbidities and treatments (e.g., very low bone mass, cardiovascular disease, obstructive pulmonary disease, chronic and depression, and anxiety, as well as the use of sedatives, hypnotics, glucocorticoids, and muscle relaxants).

In conclusion, recent evidence suggests that an assessment of clinical risk factors combined with the measurement of BMD is an effective method for assessing fracture risk (Recommendation A).

2. Bone densitometry and related techniques

Dual-energy X-ray absorptiometry (DXA) can be used to quantify BMD and is thus the procedure most commonly used to estimate fracture risk⁷. The results are expressed in terms of T-score, which is the number of standard deviations (SDs) by which the BMD value obtained differs from that of the normal young adult population (i.e., 20–29 years of age). The World Health Organization (WHO) guidelines state that osteoporosis can be diagnosed when the BMD is less than -2.5 T⁸. The organization has since clarified that this value must correspond to a measurement made on the neck of the femur using data from the National Health and Nutrition Examination Survey (NHA-NES III) study as a reference⁹. By contrast, the International Society for Clinical Densitometry (ISCD)¹⁰ states that this diagnosis can be established based on a -2.5 T value detected in the lumbar spine or total hip as well as the femoral neck. The WHO also defined normal bone density, osteopenia (i.e., low bone mass), as well as established or severe osteoporosis (Table 2).

Table 2. WHO diagnostic criteria for osteoporosis 

BMD: bone mineral density; T (T-score or T index): comparison with the BMD value reached in a young reference population.

BMD measured at the mid-third of the radius may also be used to diagnose osteoporosis when the hip and lumbar spine cannot be used or interpreted¹¹.

In, The ISCD recommends that instead of T-scores, Z-scores adjusted for ethnicity or race be used when diagnosing osteoporosis in premenopausal women, men younger than 50 years of age, and children. Z-scores ≤-2.0 are identified as "low bone mineral density for age chronological" or "below expected range for age". Z-scores >-2.0 are identified as"within expected range for age".

Evaluation of therapeutic efficacy is an indication for densitometry. This examination might be repeated after two to three years of treatment.

Other measurement techniques, including quantitative ultrasonometry and quantitative computed tomography, among others, also provide values that are related to fracture risk. However, they are not recommended as diagnostic procedures at this time.

Lateral projections of DXA studies can be used to identify vertebral fractures (i.e., VFA, or “vertebral fracture assessment”). However, the accuracy of this procedure is lower than that of conventional radiography, most notably for the diagnosis of fractures of the upper thoracic vertebrae.

The Trabecular Bone Score (TBS) is a parameter that describes bone texture based on data obtained from a DXA image of the lumbar spine. TBSs are typically reduced in patients who have sustained fragility fractures, and it is a useful value for assessing fracture risk in women and men over 50 years of age, independent of BMD findings. The combination of BMD and TBS is superior to BMD alone for the prediction of fracture risk. A TBS may be particularly useful in assessing fracture risk in patients diagnosed with diabetes or primary hyperparathyroidism as well as those treated with glucocorticoids. The TBS is also expressed in absolute terms and as a T- score.

A TBS value <1.230 (T <-3) is indicative of a degraded trabecular microstructure and a high risk of fracture. The TBS has been included in the Fracture Risk Assessment Tool (FRAX) which can be used to calculate the absolute risk of fracture in a given patient.

Despite the proven usefulness of DXA for assessing patients with an elevated risk of sustaining a fracture, the sensitivity and specificity of this modality remain limited. DXA does not identify all subjects at risk of fracture; more than 50% of peripheral fractures occur in patients with a T-score >-2.5^12,13. Current trends suggest that BMD measurements might be considered together with the clinical risk factors when calculating an absolute fracture risk^14,15.

There are no universally accepted criteria regarding when to perform densitometry. The general recommendation is that this procedure might be performed when risk factors that are strongly associated with osteoporosis or fractures emerge (Table 1), including:

In conclusion, DXA can be used to measure BMD in the proximal femur and lumbar spine to assess the risk of fracture (Recommendation A). A TBS can provide additional information on the risk of fracture in an individual patient (Recommendation B).

3. Markers of bone turnover

Bone turnover markers (BTMs) provide information on the dynamics of bone turnover. Among the markers of bone formation, significant research has focused on levels of osteocalcin, bone alkaline phosphatase, and the carboxy –and amino-terminal propeptides of type I procollagen (PICP and P1NP). Markers of bone resorption include the carboxy– and amino-terminal telopeptides of collagen I (CTX in blood, s-CTX and NTX in urine) and tartrate-resistant acid phosphatase 5b (FATR 5b). Various international organizations (for example, the International Federation of Clinical Chemistry) have recommended the use of P1NP and s-CTX as markers of bone formation and resorption, respectively, for ongoing and future clinical studies. It is important to control the variability of these measurements by obtaining biological samples consistently between 08:00 and 10:00 hrs after an overnight fast.

While BTMs are not useful for diagnosing osteoporosis, this information may be combined with other risk factors to identify, patients with a higher risk of sustaining a fracture. These values are particularly useful for the early assessment of responses to both antiresorptive and anabolic therapy (Evidence 2a)^16,17. For example, measurements of s-CTX and PINP are recommended as an effective means to monitor bone turnover after discontinuation of denosumab¹⁸.

In conclusion, BTMs can be useful for evaluating therapeutic responses (Recommendation B), but they must be measured under standardised conditions. They are not used routinely to diagnose osteoporosis.

4. Identification of vertebral fractures

Conventional radiography is not sufficiently sensitive or specific when used to assess changes in bone mass⁸. However, the use of this modality is essential when attempting to identify fractures.

A diagnosis of a vertebral fracture requires a decrease of at least 20–25% in height¹⁹. This is because slight wedging can be confused with deformities of another origin (e.g., sequelae of Scheuermann's disease, small wedging of a degenerative type)²⁰. Thus, VFA by DXA may be useful as a first step. Spinal radiography (or DXA) is recommended for patients over the age of 70 years with suspected osteoporosis who present with back pain, glucocorticoid treatment, or a significant decrease in height (>4 cm based on historical data or >2 cm in confirmed height)²¹.

In conclusion, reliable identification of vertebral fractures is important in decision-making because these lesions represent a risk for future fractures. Evaluation can be done by radiography or by VFA. However, radiography should not be used as a method of assessing bone mass to establish a diagnosis of osteoporosis (Recommendation A).

5. Study protocol

In addition to anamnesis and a physical examination, an evaluation of a patient with suspected osteoporosis should include a complete blood count and determination of basic biochemical parameters (kidney and liver function and serum levels of calcium, albumin, phosphorus, alkaline phosphatase, thyrotropin (TSH), and 25-hydroxyvitamin D, as well as a serum protein electrophoresis study). It is useful to quantify calciuria. These tests should be performed before starting treatment and then repeated if clinically indicated. The usefulness of parathyroid hormone (PTH) levels and BTMs remains controversial (see the previous section). Bone densitometry and an assessment of potential vertebral fractures by VFA or radiology will almost always be required. Pertinent studies should be performed to rule out secondary causes of osteoporosis (e.g., hypercortisolism, celiac disease, and systemic mastocytosis, among others) in younger patients (Recommendation C).

6. Risk prediction tools

Various scoring scales have been developed to assess either the risk of developing osteoporosis (i.e., low DXA), or sustaining osteoporotic fractures. Current scoring scales used to assess the risk of densitometric osteoporosis do not include BMD but are useful in deciding when densitometry evaluations should be performed.

The simplest method, known as the Osteoporosis Self-assessment Tool [OST])^22,23 includes only patient age and weight which are variables included in all assessment strategies.

To assess the risk of fractures, the addition of findings from DXA to the clinical data results in their improved predictive value. Several instruments have been developed for this purpose, including FRAX²⁴, the Garvan Medical Research Institute scale²⁵, and the QFracture Index²⁶. All three have similar discriminatory capacities albeit with only moderate performance^27,28. FRAX is the most widely used of these instruments on a worldwide basis. Unfortunately, its adaptation in Spain has been inadequate²⁹ and it underestimates the risk of fracture, most notably major osteoporotic fractures. Other tools, such as EPIC, which has been adjusted to the Spanish population, are currently undergoing validation.

In conclusion, although fracture risk prediction tools may be helpful in decision-making in some cases, their predictive value for our population is limited. Adaptations of FRAX may be used with caution pending the development and validation of newer and more precise instruments (Recommendation C).

1. Non-pharmacological interventions

A balanced diet should be maintained by all patients diagnosed with postmenopausal osteoporosis. This would include a protein intake of 1–1.5 g/kg/day. While sun exposure will promote essential vitamin D synthesis, additional supplementation may be needed (see below)³⁰. Furthermore, recent evidence suggests that physical exercise that loads the skeleton has a positive effect with respect to preventing falls and reducing the risk of fracture³¹. Routine exercise is recommended, for example, walking every day for at least 30 minutes.

Smoking and excessive alcohol consumption should be avoided, as both are factors associated with decreased bone mass and an increased risk of fractures^32,33.

Although the efficacy of fall prevention programs (beyond basic physical exercise) remains controversial, recent evidence suggests that they are useful in institutionalised elderly patients who undergo repeated falls^34,35. Hip protectors are slightly effective at reducing the risk of hip fracture. However, poor tolerance by some patients, poor adherence, and a slight increase in the risk of pelvic fractures limit its application³⁶.

2. Calcium and vitamin D

Patients treated with antiresorptive or anabolic drugs for osteoporosis should be certain to maintain an adequate intake of calcium and vitamin D^37,38. Serum levels of 25-hydroxyvitamin D (25(OH)D) should be maintained above 20–25 ng/ml, preferably above 30 ng/ml³⁹. The recommended daily dose of vitamin D is generally between 800–1200 IU/day, although some patients may need higher doses to maintain adequate serum levels of 25(OH)D. While bi-weekly or monthly equivalents can be considered, administration of large amounts of vitamin D in a single dose (e. g., 500,000 IU/year)⁴⁰ is not recommended. The standard dose of calcifediol (25(OH)D3) is 0.266 micrograms every 15–30 days. This form of vitamin D may be preferable in patients with advanced liver disease or problems with intestinal absorption. Occasionally, these patients may require parenteral administration.

Daily intake of calcium should be maintained at 1000-1200 mg/day³⁰. While it is preferable to obtain this amount from dietary sources, supplements can be added as necessary. The general population, particularly the elderly, should be advised to maintain adequate nutrient intake, including appropriate levels of calcium and vitamin D. However, the isolated effects of calcium and vitamin D on the progression of osteoporosis are not well-understood; if they exist at all, their impact seems to be limited⁴¹-⁴³.

In conclusion, patients at risk for developing osteoporosis and those undergoing treatment with antiresorptive or anabolic drugs should receive and be certain that they are taking in an adequate supply of calcium and vitamin D. However, these nutrients alone are insufficient treatments in patients who have developed osteoporosis (Recommendation A).

3. Calcitonin

Although treatment with calcitonin was associated with a slight reduction in the risk of vertebral fractures, it has no impact on the risk of peripheral fractures. Furthermore, long-term calcitonin use has been associated with an increased risk of tumors. Thus, calcitonin is not approved for the treatment of osteoporosis^44,45.

4. Thiazides

Although numerous observational studies suggested that treatment with thiazides resulted in increased bone mass and a concomitant reduction in the risk of fracture⁴⁶, we have no data that can be construed as recommending its use as a treatment for osteoporosis. Thiazide treatment (e. g., 12–50 mg/day of hydrochlorothiazide or chlortha-lidone) can be considered for patients presenting with hypercalciuria⁴⁷ (Recommendation D).

5. Estrogen therapy

The results of several clinical trials have revealed the efficacy of estrogens for the prevention of fractures. A recent network meta-analysis revealed that estrogen therapy (with or without progesterone) reduced the risk of vertebral fracture by 34% (hazard ratio [HR], 0.66; 95% confidence interval [CI], 0.49–0.89); hip fracture by 29% (HR, 0.71; 95% CI, 0.52–0.98); and non-vertebral fractures by 21% (HR, 0.79; 95% CI, 0.70-0.90)⁴⁸. However, the side effects of estrogen therapy revealed by the Women’s Health Initiative (WHI) study and other trials include an increase in cardiovascular events and breast cancer. Thus, estrogen is not recommended as a treatment for osteoporosis ex cept in women with early menopause or at a high risk of fracture in which there is no other therapeutic option available⁴⁹. Estrogens may be an effective treatment for osteoporosis in women already receiving these drugs as therapy for the climacteric syndrome.

In conclusion, although estrogen therapy is effective in preventing osteoporotic fractures, it is not recommended for routine use given the possibility of serious side effects (Recommendation A). Estrogens can be considered in patients exhibiting early menopause who have no other contraindications and/or in cases in which no other therapeutic options are available (Recommendation D).

6. Selective Estrogen Receptor Modulators (SERMs)

Results from several recent studies document that these drugs can increase BMD in the spine over follow-up periods as long as eight years^50,51. A recent meta-analysis revealed that raloxifene and bazedoxifene reduce the risk of vertebral fracture by 40%, although neither drug has any impact on non-vertebral fractures⁴⁸. The main complication associated with this class of drugs is an increased risk of venous thromboembolic disease.

In conclusion, SERMs may be indicated for the treatment of osteoporosis because they reduce vertebral fractures, but they do not reduce the risk of non-vertebral fractures (Recommendation A).

7. Tibolone

Although the use of this drug will reduce the risk of both vertebral and non-vertebral fractures in women under 60 years of age (or <10 years of menopause)^52,53, its cardiovascular side effects limit its use. At this time, tibolone may be prescribed for patients who are not at high risk for cardiovascular disease or breast cancer who cannot be treated with other drugs (Recommendation B). This drug has not been approved for the treatment of osteoporosis in Spain.

8. Phytoestrogens and isoflavones

Isoflavones may have a favorable effect on BMD⁵⁴. However, they are not currently recommended for the treatment of osteoporosis due to the lack of data focused on their efficacy in preventing fractures.

9. Bisphosphonates (BPs)

10. Denosumab

Denosumab is a monoclonal antibody with a powerful antiresorptive effect that translates into a reduction in the risk of fracture. In general, it has shown greater antiresorptive potency and results in a greater increase in BMD than achieved with BPs.

Denosumab therapy results in reductions in the risk of vertebral, non-vertebral, and hip fractures of ~70%, 20%, and 40% respectively⁹² (Evidence 1b). A post hoc analysis of these data suggests that its efficacy in reducing hip fracture may be greater in subjects older than 75 years of age⁹³ (Evidence 2b). Its beneficial impact on fracture risk appears to be maintained during treatment and persists for at least 10 years⁹⁴.

In the months following drug withdrawal, an increase in BTMs and a loss of the bone mass gained with subsequent stabilization at baseline values are observed. In some patients, these responses have been associated with multiple vertebral fractures⁹⁵. Therefore, any interruption of denosumab therapy should be followed by the administration of a BP for six months following the final dose. However, the ideal regimen has not yet been established (see below)¹⁸.

Denosumab is generally well tolerated. It is not associated with an increased risk of neoplasms, cardiovascular events, or infections and is safe to use in patients with diabetes⁹⁶. As with BPs, the risk of AFF and ONJ is very low. In a study performed with patients treated for a prolonged period (up to 10 years), the risk of AFF was determined to be ~1/10,000 patient-years. The risk of developing ONJ was 1/2,000 patient-years⁹⁴. Furthermore, denosumab can be used safely in patients with GFRs <30 ml/min and even in those on dialysis with no need for dose adjustment. However, hypocalcaemia may develop, especially in patients with advanced renal failure. Close follow-up will be necessary for these patients, together with an adequate supply of calcium and vitamin D.

In conclusion, denosumab therapy can reduce the incidence of vertebral, non-vertebral, and hip fractures. Thus, this agent has a definitive role in the treatment of osteoporosis (Recommendation A).

11. Strontium ranelate

Strontium ranelate reduces the incidence of vertebral and non-vertebral fractures by ~40% and 16%, respectively⁹⁷. However, the administration of this agent results in an increased incidence of cardiovascular events. It is not currently available for use in Spain or any other European country.

12. Parathyroid hormone (PTH) 1-34 (teriparatide)

Teriparatide is the amino-terminal (1-34) peptide fragment of human parathyroid hormone (PTH) that promotes bone formation. Administration of teriparatide reduces the risk of vertebral fracture by 65% and non-vertebral fracture by 50%⁹⁸ (Evidence 1a). While teriparatide has not yet been evaluated in trials designed to assess its specific impact on hip fractures, a review of observational studies suggested reductions of ~56%⁹⁹. A more recent meta-analysis found that teriparatide therapy resulted in no significant reductions in hip fractures⁴⁸, although another three reviews concluded that it reduced hip fractures between 56% and 65%^61,100,101. One study directly compared the effects of the BP, risedronate, and teriparatide in postmenopausal women with severe osteoporosis and vertebral fractures; the teriparatide-treated group experienced fewer vertebral and clinical fractures than the BP-treated group (5.4% versus 12.0% and 4.8% versus 9.8%, respectively)¹⁰². Teriparatide is administered as a daily subcutaneous injection for two years. The benefits with respect to BMD that are achieved with this drug decrease progressively after its withdrawal; thus, sequential treatment with an antiresorptive drug is recommended. Teriparatide is generally well tolerated. Several biological and biosimilar drugs have been approved for clinical use because they have met the standard bioequiva-lence requirements established for these drugs.

In conclusion, teriparatide reduces both vertebral and non-vertebral fractures and, although it is not approved for this indication, it probably also reduces the incidence of hip fractures (Recommendation A).

13. PTH (1-84)

This formulation is not currently licensed for osteoporosis treatment.

14. Abaloparatide

Abaloparatide is an analog of the 1-34 region of PTH and is a PTHrP (PTH-related peptide). The results of a clinical trial found that administration of abaloparatide reduced the risk of vertebral and non-vertebral fractures by 86% and 43%, respectively, compared to placebo¹⁰³.

A recent meta-analysis revealed that the use of this drug resulted in 87%, 50%, and 61% reductions in vertebral, non-vertebral, and wrist fractures respectively¹⁰⁴. Abaloparatide is approved for use in the US but not in Europe. It is not available in Spain.

15. Romosozumab

Romosozumab is a sclerostin-neutralizing monoclonal antibody. Sclerostin is a small protein pathway which is essential for osteoblastic activity. Various experimental and clinical studies have shown that romosozumab has a dual effect. Administration of romosozumab increases bone formation and also decreases the rate of bone resorption. The latter effect has been associated with the impact of this drug on levels of the osteoclast NKL. Consistent with its dual effect, romosozumab increases the levels of bone formation markers, such as PINP, and decreases the levels of resorption markers, such as CTX. Romosozumab induces notable increases in BMD in both the spine and the hip. The anabolic effects of this drug disappear after 6–12 months of treatment. Therefore, it is typically administered for periods of one year, after which an antiresorptive agent must be used to maintain or increase BMD.

The results of three pivotal trials and several meta-analyses^48,105-¹⁰⁸ reveal that treatment with romosozumab for 12 months reduces the incidence of vertebral fractures in postmenopausal women and men with osteoporosis (relative risk reduction [RRR], 66%–73%)¹⁰⁹. Likewise, the combined analysis revealed that romosozumab therapy decreases the risk of non-vertebral (RRR 33%) and hip fractures (RRR 56%). In postmenopausal women with severe osteoporosis and a history of previous fragility fractures, treatment with romosozumab for one year followed by alendronate significantly reduced the risk of new vertebral, hip, and clinical fractures, compared with treatment over the entire period with alendronate alone¹⁰⁷.

Romosozumab is generally well-tolerated, although the results of several studies suggested that it may increase the incidence of cardiovascular events¹¹⁰. While this difference was small in absolute terms (1.3% of events versus 0.9% in the control group), romosozumab is not indicated in patients with a history of myocardial infarction or stroke and should be carefully considered in patients presenting with multiple cardiovascular risk factors.

In conclusion, romosozumab has a defined role in the treatment of osteoporosis as it reduces the risk of both vertebral and peripheral fractures (Recommendation A). Potential cardiovascular risks and specific contraindications should be assessed in each patient.

16. Vertebroplasty and kyphoplasty

Although many uncontrolled studies have shown that these procedures are associated with a marked analgesic effect, randomised clinical trials have offered contradictory re-sults¹¹¹-¹¹⁴ and controversy regarding a potential increased risk of fracture in the adjacent vertebrae remains. Therefore, these procedures are not routinely recommended⁸⁷ for patients with asymptomatic vertebral fractures, mild pain, or those with symptoms that have persisted for more than one year. These procedures can be considered in patients who present with fractures that are less than six weeks old and severe pain despite appropriate medical treatment and in patients with fractures that have evolved over six weeks to one year ago with persistent pain that responds poorly to analgesics and evidence of edema on magnetic resonance imaging studies¹¹⁵. These procedures may also be useful in patients who present with contraindications or poor tolerance to analgesics. Vertebroplasty and kyphoplasty are similar in terms of effectiveness and safety¹¹⁶. There is insufficient evidence on the relative usefulness of procedures that include the insertion of expanding implants specifically when compared to vertebroplasty and balloon kyphoplasty (Recommendation B).

In conclusion, vertebroplasty, kyphoplasty, and related techniques are not routinely recommended for the treatment of vertebral fractures, although these procedures may help to control symptoms in carefully selected patients (Recommendation C). In any case, its use must be accompanied by medical treatment of osteoporosis to prevent new fractures.

1. Decision to commence treatment

There is no internationally agreed or apstandard on when to initiate treatment for osteoporosis. SEIOMM suggests that, in general, patients that present with the following attributes should be treated:

However, we recognise that some situations may require exceptions to these recommendations. All patients must undergo a careful, individualised assessment that considers the risk factors for fracture as well as other clinical characteristics. For example, it may be possible to delay the start of treatment in young women who present with only slightly low BMD, without fractures or other risk factors. By contrast, a patient who presents with several important risk factors may require early treatment. Scales that help estimate fracture risk (e. g., FRAX) may be helpful, although these instruments have not yet been fully validated for use in the Spanish population, as mentioned above.

2. Control of the therapeutic response

Adherence and therapeutic responses to treatment regimens can be assessed by changes in BTMs.

The beneficial effect of a given treatment regimen can be confirmed by increases in BMD and the absence of new fractures. However, it is critical to recognise that a single fracture while on a treatment regimen is not necessarily indicative of therapeutic failure. Elderly patients and those with dementia, poor quality of life, and/or multiple fractures are at greater risk for therapeutic failure. In cases where oral BPs have failed, parenteral drugs (zoledronate, denosumab, and [depending on patient characteristics] teriparatide or romosozumab) may represent good therapeutic alternatives.

Changes to treatment regimens due to a potential inadequate response may be considered in the following circumstances¹¹⁷:

Before proceeding with a therapeutic change, the following factors should be considered as possible causes of an inadequate response: a) vitamin D deficiency; b) secondary forms of osteoporosis; c) inadequate compliance; d) tendency to fall; e) defective techniques used to measure BMD and/or BTMs; f) serious bone deterioration, leading to the likelihood of new fractures despite active drug treatment.

If the reasons for the changes observed include an apparent lack of an appropriate response, the following options are recommended^117,118 (Recommendation D):

3. Duration of treatment

Interruption of treatment is justified when the risk/benefit ratio becomes unfavourable. These situations can include: a) therapeutic objectives have been achieved;

b) loss of effectiveness; or c) increased risk of developing secondary effects.

Based on these facts, the following recommendations are proposed. These recommendations represent expert consensus, albeit without published studies to provide definitive support¹²²-¹²⁶ (Recommendation D):

If none of these circumstances arise, BP treatment can be withdrawn, at least temporarily.

If treatment is maintained, the possibility of its withdrawal should be periodically reassessed at various intervals thereafter. There is currently no guidance as to how often each patient should be reassessed, nor if there is a defined maximum duration of treatment. A limit of 10 years is often set, as there are no studies that have evaluated the impact of these drugs over the longer term. However, if the patient remains at risk, anti-osteoporotic treatment should not be withdrawn. If anti-resorptive treatment is withdrawn, and the patient remains at risk for fracture, a drug from another class should be administered, for example, an anabolic.

When BP treatment is withdrawn, the suspension must be temporary (i.e., a “drug holiday”). It is not known how long the treatment regimen can be safely suspended or fully discontinued. Typically, the drug can be suspended for a period of 1 to 3 years, depending on the BP used (e. g., perhaps one year for risedronate, two years for alendronate, and three years for zoledronate). Some experts have suggested that BTMs and BMD measurements can help with this decision, although we are not in a position to confirm this. In theory, if the BMD remains above a “target” value (e. g., T >-2 or -2.5), drug withdrawal can be considered.

“Drug holidays” should not be scheduled for patients treated with denosumab, because after its withdrawal, not only is there no residual effect, but bone turnover increases to levels above baseline values (i. e., a “rebound effect”). This increased bone turnover has been associated with a rapid loss of bone mass and an increased risk of developing multiple vertebral fractures. Therefore, continuing denosumab therapy indefinitely is recommended. In cases in which denosumab must be discontinued, it should be replaced with a potent BP (see below)¹⁸.

There are some data available that address the efficacy and safety of SERMs (raloxifene and bazedoxifene) for up to eight years. In these cases, the treatment regimen can be maintained through this time or until the risk of hip fracture or complications, such as thromboembolic disease, increases. It is not usually recommended in patients older than 65-70 years of age.

Treatment with teriparatide or romosozumab should be maintained for 24 and 12 months, respectively, followed in both cases by an antiresorptive drug.

4. Sequential and combined treatment

5. Therapeutic decision algorithms

The proposed algorithm is based on data from published trials and considerations that are summarised below.

5.1. Initial treatment (Choice of a drug; Figure 1)

SERM: selective estrogen receptor modulator; (*): especially if T ≤-2 and factors strongly associated with fracture risk, such as hypogonadism, early menopause, or treatment with glucocorticoids or sex hormone antagonists. These general criteria may need to be adapted based on other clinical determinants of fracture risk, the characteristics of individual patients, and their preferences.

Figure 1. Algorithm for selecting the initial treatment in postmenopausal osteoporosis 

The main criterion for choosing the initial drug is the risk of fracture. We distinguish three levels of risk, including “moderate”, “high”, and “very high”.

1) Moderate risk. This category corresponds to the risk profile of a woman under 65 years of age, with no history of fracture, a spinal T-score between -2.5 and -3.0, and a relatively preserved hip BMD (T-score >-2). In this situation, a SERM is recommended because one can then delay the use of prolonged treatment strategies that can elicit AFF or ONJ. However, ibandronate and antiresorptive agents that are typically recommended for high-risk situations are the second choice in this situation. These drugs represent acceptable alternatives if for some reason SERMs are to be avoided.

2) High risk. Most of the cases seen in the clinic will present this level of risk (see section above “Decision to start treatment”). These patients do not meet the criteria that define either moderate or very high-risk cohorts as described further below. Alendronate, risedronate, zoledronate, or denosumab are indicated for the treatment of patients in the high-risk cohort. Oral BPs are considered preferable for patients <75 years of age when there are no inconveniences with respect to oral administration (digestive problems, polypharmacy, adherence). Injectable antiresorptive drugs are considered preferable in all other cases. As most individuals who have sustained hip fractures are over 75 years of age and belong to the second group, injectable antiresorptive agents are generally preferred for this group. Given the rebound effect after discontinuation of denosumab, zoledronate may be the preferred agent if there are doubts regarding compliance.

3) Very high risk. We consider women to be at very high risk in any of the following situations: a) two or more vertebral fractures, or an equivalent risk (i.e., T-score <-3.5); or b) vertebral or hip fracture with a T-score <-3.0. There may be other clinical situations that suggest that a patient is at a very high risk of fracture; these will require individualised consideration. For this level of risk, bone-forming drugs such as teriparatide or romosozumab should be used. Romosozumab may have a better cost-benefit ratio (although its marketing price was not known at the time that these guidelines were written), albeit a less favorable risk-benefit ratio due to the potential increase in cardiovascular events. Romosozumab should be avoided in all patients with or at high risk of developing cardiovascular disease. However, these guidelines and recommendations should be understood as provisional at this time, pending marketing in Spain and further experience with this drug in our population.

Although some authors have suggested that all patients with a recent fracture, especially a vertebral fracture, might benefit from treatment with a bone-forming drug. However, there is currently no consensus on this point among our panel of experts. Regardless of the treatment that is ultimately selected, therapy should be initiated as soon as possible given that these patients are at very high risk for new fractures.

5.2. Long-term treatment (Figure 2)

BP: bisphosphonates; SERM: selective estrogen receptor modulators; BTM: bone turnover markers; (*): there are not enough data to establish a recommendation after that treatment time, so the possible options are listed before a decision that must be individualized.

Figure 2. Long-term treatment continuation algorithm 

Romosozumab should only be administered for one year; teriparatide therapy is limited to two years. Likewise, given that efficacy and safety data are available for up to eight years of treatment only, withdrawal of SERMs should be considered after that period, when the patient reaches 65-70 years of age or if the risk of fracture increases. After one or more of these milestones are reached, it will likely be necessary to administer another antiresorptive. The discussion on long-term treatment is thus restricted to a consideration of BPs and denosumab. One key differentiating factor at this time is the potential impact of a temporary interruption or “therapeutic vacation” or “drug holiday”. While this is discouraged for individuals undergoing treatment with denosumab, it is currently accepted for BP regimens.

1) Denosumab. This agent can be administered continuously for 5–10 years. No information is currently available regarding longer periods of use. Thus, the decision to continue or discontinue drug treatment should be made carefully. Once administration of denosumab has been interrupted, the patient should be treated with a BP, for example, alendronate or zoledronate. Zoledronate is preferred if denosumab treatment was prolonged for more than 2–3 years. (See section 4.1).

2) Bisphosphonates (BPs). Three periods of use have been described: