Osteoporosis in Frail Patients: A Consensus Paper of the Belgian Bone Club

AN is associated with marked endocrine changes that also impair healthy bone turnover. As mentioned before, amenorrhea is common. In adolescents with AN, menarche is often delayed, contributing to low BMD. However, the severity of bone loss in women with AN is greater than in those with normal-weight hypothalamic amenorrhea, indicating that, in addition to estradiol deficiency, there are other factors including nutritional deficiencies and hormonal abnormalities that contribute to bone loss. GH resistance, low levels of IGF-1, hypercortisolemia, and low levels of testosterone have all been implicated in AN-associated osteopenia and osteoporosis. Abnormalities in hormones regulating appetite (oxytocin, leptin, and peptide tyrosine tyrosine (PYY)) may play an additional role [1]. Therefore, the etiology of bone loss in patients with AN is multifactorial.

Klibanski et al. observed an uncoupling of bone turnover in adults with AN, with a decrease in markers of bone formation and an increase in markers of bone resorption. However, in adolescents with AN, there is a low bone turnover state with a decrease in the levels of bone formation and resorption, as opposed to an increased bone turnover during normal adolescence [3]. Veronese et al. conducted a meta-analysis of vitamin D status in AN patients [4]. They identified 15 studies (totalizing 927 participants; AN = 408 and healthy controls = 519) and observed that, although AN patients reported similar dietary vitamin D intake compared to healthy controls, AN patients had significantly lower levels of 25OHD and 1,25(OH)₂D when not using vitamin D supplementation. Conversely, supplementation with cholecalciferol fully normalized serum levels of vitamin D [4].

Both trabecular and cortical bone sites are affected in AN, but since there is a marked estrogen deficiency component in these women, trabecular osteoporosis (such as at the lumbar spine) is more frequent and prominent [5]. Based on HR-pQCT studies, it has been suggested that bone geometry, microarchitecture, marrow adiposity, as well as bone strength estimates are affected [5]. When AN affects women before reaching their peak bone mass, a reduced bone accrual will impeach them from attaining the optimal BMD, increasing their risk of fracture [6].

Not surprisingly, therefore, a two- to sevenfold increase in fractures has been reported in patients with AN [7, 8]. Rigotti et al. followed a series of 27 women with AN for a median of 25 months and concluded that anorectic women have an increased risk of fracture (RR 7.1; 95% CI 3.2–18.5) and that BMD reductions appear not to be rapidly reversed by recovery from AN [7]. Vestergaard et al., using a case-control Danish Nationwide register study, also observed an increased fracture risk in AN, which persisted more than 10 years after the diagnosis [8].

The clinical assessment involves an osteoporosis assessment consisting of BMD measurement and assessment of other risk factors for osteoporosis such as a personal history of low-trauma fractures, a family history of osteoporotic fractures, lifestyle factors, medication, and serum level of 25OHD. In addition, checking for reproductive dysfunction (amenorrhea, follicle-stimulating hormone (FSH), luteinizing hormone (LH), estradiol) and measuring prolactin, thyroid-stimulating hormone (TSH), and human chorionic gonadotropin (hCG) is needed to rule out causes of amenorrhea other than AN.

Table 1 gives an overview of the management of low bone density in patients with AN. AN patients need to be encouraged to gain weight. Recovering a near-normal or normal BMI (18–25 kg/m²) will generally help to recover a normal reproductive functioning, which can be assessed by a regular menstrual cycle (not using hormone therapy). Restoration of body weight will also improve BMD although a complete catch-up does not always occur [5]. In addition, AN patients should take adequate calcium (e.g., 1200 mg) and vitamin D (e.g., 800 IU) daily, from diet and supplements.

High-dose estrogen–progestin contraception is not an effective treatment option for AN-associated bone loss in adolescents or adults. Indeed, prospective trials have failed to show a benefit of combination estrogen–progestin oral contraceptives in treating AN-associated bone loss in adolescents or adults, although one may be tempted to restore menstruation in these patients using oral contraceptives. Unfortunately, several studies have failed to observe a protective effect of the “pill” containing 50 µg ethynilestradiol [11]. On the other hand, lower-dose physiologic estrogen replacement using MHT such as 100 µg transdermal 17-beta-estradiol with cyclic micronized progesterone in adolescents resulted in BMD gains at the spine and hip compared with placebo, though the therapy did not restore BMD to normal [10]. Although AN patients are also deficient in androgens, androgen replacement is currently not recommended.

Limited data suggest that bisphosphonates may be of benefit in women with AN-associated bone loss. For instance, risedronate for 1 year resulted in a modest gain in BMD compared with placebo [13]. Another study, however, failed to report an improvement using alendronate vs. placebo [14]. Furthermore, there are safety concerns with bisphosphonates especially in younger patients who may want to become pregnant. Indeed, animal studies have suggested placental transfer of bisphosphonates and fetal skeletal development involvement, and in a small series of 10 women treated with bisphosphonates during pregnancy, congenital malformations were reported in 20% [15]. Although most other data in literature suggest that preconceptional and first-trimester use of bisphosphonates do not pose substantial fetal risks, it is recommended that, when bisphosphonates are prescribed to females of reproductive age, one should ensure that the patient is and will not become pregnant when using bisphosphonates [15]. Bisphosphonates are generally not approved for this indication, but may be used in elderly with AN. Also denosumab and teriparatide should not be used as first-line therapies in young patients suffering from AN-induced osteoporosis.

Dialysis status is associated with an outstanding risk of fracture compared to the general population. In 2006, results from the international observational cohort DOPPS (Dialysis Outcomes and Practice Patterns Study) were published. Such as in the general population, the risk of hip fracture in the dialysis population was higher in women and increased progressively with age, but, in both genders and in all age categories, the incidence of hip fracture was much higher in dialysis patients. For example, the incidence of fracture in women between 75 and 84 years was 1190 and 3136 per 100,000 patients in the general and dialysis population, respectively [16]. Similar data from USRDS (United States Renal Data System) showed a fourfold higher risk of fracture in dialysis compared to the general population [17]. There is also evidence that these fractures occur earlier in life, and fractures in hemodialysis are associated with higher morbidity and even mortality. A mortality rate as high as 64% in the year following a hip fracture has been described in dialysis patients [18]. Of course, the risk of fracture is also tightly associated with the high risk of falls in this population, which is enhanced by factors such as older age, diabetes, depression, antidepressants, previous falls, malnutrition, and frailty.

In clinical practice, bone fragility is evaluated by DXA that measures both cortical and trabecular bone quantity. New CT-scan techniques (quantitative computed tomography (QCT) or high-resolution peripheral QCT (HR-pQCT)) are now available and currently used in clinical research with the main advantage of being able to separate cortical and trabecular areas [19‐21]. However, bone quantity is only one piece of the puzzle. Indeed, bone strength, and thus fracture risk, is dependent not only on bone quantity but also on bone quality which is determined by both bone turnover and mineralization, concepts being assessed by bone biopsy and histomorphometry [19, 22].

Bone disorders have been described for more than a century in chronic kidney disease (CKD) patients [22]. In 2006, the international KDIGO (Kidney Disease: Improving Global Outcomes) guidelines proposed a common terminology using the term “Chronic Kidney Disease-Mineral and Bone Disorder” (CKD-MBD) to encompass all possible abnormalities associated with bone disease and mineral disturbances [23]. CKD and still more dialysis status is associated with severe mineral abnormalities, such as hypocalcemia, hyperphosphatemia, and secondary hyper- or hypoparathyrodism. It is beyond the scope of this article to review the pathophysiology of such abnormalities [23, 24] but basically they will be associated with either low (“adynamic bone disease”) or high (“osteitis fibrosa”) bone turnover disease in dialysis patients [22]. The vast majority of basic and clinical research in nephrology is thus focused on turnover abnormalities, and by consequence more on bone quality (turnover) than on bone quantity (osteoporosis) [19]. Briefly, both low and high turnover seems associated with an increase of fracture risk [19‐21, 23, 25]. Sensu stricto, such a diagnosis requires a bone biopsy but this technique remains relatively invasive, costly, and difficult to interpret and repeat. An important part of clinical research is thus about the role of biomarkers to estimate bone turnover [24]. Currently, the KDIGO recommends the measurement (and monitoring) of both parathyroid hormone (PTH) and bone-specific alkaline phosphatase (b-ALP), even if, as discussed later in this chapter, the interpretation of the results remains frequently difficult and sometimes disappointing [24].

Data on the utility of DXA in dialysis patients, notably for predicting the risk of fracture, are scarce and controversial in comparison to the literature in the general population. For this reason, but also because specific therapies for osteoporosis are not routinely used in dialysis, the KDIGO did not recommend routine DXA in dialysis patients when first published in 2009 [23]. This recommendation could, however, change in the next KDIGO recommendations. Indeed, a meta-analysis, including 6 studies and 683 (357 women) dialysis patients, suggested that low BMD was associated with fractures and this was true for all sites, except the femoral neck [26]. However, data were heterogeneous and only cross-sectional. A longitudinal observational study in 485 hemodialysis patients showed that DXA, especially at the hip region, was useful to predict incident fractures but only in patients with low PTH [25]. Beyond fracture risk, some authors suggested that in dialysis patients, osteoporosis and osteopenia were associated with an independent higher risk of mortality [27]. These data could lead to a broader use of DXA in dialysis patients in the future, even if currently the level of evidence is still relatively low. Therefore, at this time, a T-score threshold at which intervention should be started, cannot be recommended. Moreover, there are some specific technologic issues in dialysis patients. Indeed, these patients frequently suffer from spinal osteophytes and high vascular calcifications in the aorta making the interpretation of DXA results at the lumbar region questionable, with a risk of spuriously good results [19, 27]. It is thus suggested to measure BMD in the radius region, especially when there is secondary hyperparathyroidism, and this is another reason why QCT and HR-pQCT are of potential interest in dialysis patients [20].

FRAX does not include any “renal” parameter. This score has not been validated in dialysis patients and is thus considered neither in the KDIGO nor in clinical practice. Therefore, 10-year FRAX risks at which intervention should be initiated cannot be recommended.

If the available literature on bone biomarkers to detect and/or monitor bone turnover abnormalities in dialysis patients is abundant [24], current data on these biomarkers to detect and/or monitor osteoporosis are scarce and controversial. In the general population, biomarkers (especially β C-terminal cross-linked telopeptide (β-CTX) and N-terminal propeptide of type I procollagen (PINP)) may be used to monitor the response to osteoporosis therapy [28]. This is, however, not relevant in dialysis patients as nearly all osteoporotic treatments are not recommended in these patients. The current knowledge on the ability of such biomarkers to detect osteoporosis and/or to monitor bone loss is thus limited and controversial [21, 23]. A recent observational longitudinal study in 81 US dialysis patients measured several biomarkers as well as BMD by DXA and QCT at baseline and after one-year of follow-up [29]. At baseline and after multivariate analysis with age, gender, BMI, and ethnicity as covariates, an association was observed between DXA at the hip and baseline PTH and b-ALP (negative association) and sclerostin (positive association). At the spine, the associations were significant with PTH and sclerostin but also with FGF-23. Considering change in BMD over a one-year period, there was no association between change in BMD at the hip and any biomarker. At the spine, change in BMD was, however, associated with baseline FGF-23 values (negative association) and with change in sclerostin (positive association). The associations observed between change in BMD measured by QCT and biomarkers are also significant but with other biomarkers than those associated with DXA (for example, tartrate-resistant acid phosphatase-5b) and the associations are also different according to the region (hip or spine) [29]. The association between BMD or change in BMD and biomarkers are thus very heterogeneous, making the interpretation of this study difficult. Moreover, each biomarker, including “simple” biomarkers like PTH, has its own analytical and clinical limitations (accumulation in CKD, high intra-individual variation, absence of normal reference values) [24]. Other studies on the same topic are thus urgently needed. For example, for PTH, second- and third-generation PTH assays are currently available, which measure 1–84 and 7–84 PTH (‘intact’ PTH assay) and 1–84 PTH only, respectively. KDIGO experts conclude that both the second- and the third-generation assays are similarly informative. However, the assays have high inter-method variability. Therefore, KDIGO proposes to use a target range for serum PTH based on multiples of the upper limit of the normal values rather than on absolute concentrations [24].

The majority of therapeutic studies in dialysis patients focused on bone turnover abnormalities with, for example, vitamin D or calcimimetics, although the majority of these studies used biomarkers, and not bone biopsies, as endpoint [19]. Clinical studies reporting fractures as hard clinical endpoint are even more scarce. One large RCT suggests that fractures could be reduced in dialysis patients with hyperparathyroidism when treated with calcimimetics [30]. This was, however, a post hoc analysis with statistical adjustments, while the primary study was negative for the fracture endpoint [31].

There are few studies about specific osteoporosis therapies in dialysis patients as dialysis status was an exclusion criterion in all large interventional trials on this topic [19‐21]. The role of bisphosphonates in dialysis patients is still a subject of debate. Data in literature are scarce in severe CKD non-dialysis patients and even more limited in dialysis patients [20]. Some publications, however, suggest positive results on BMD [32]. Because 50% of the administered dose of bisphosphonates gets excreted by the kidneys, the dosage used in these studies is lower than in the general population. Limitations of these studies are the design, sample size, and follow-up (6–12 months). In theory, bisphosphonates are contraindicated in patients with low bone turnover. For this reason, the KDIGO recommends to perform a bone biopsy before a potential treatment with bisphosphonates. Also raloxifene has been hardly studied in dialysis patients. The safety profile seems acceptable in postmenopausal women on hemodialysis, but long-term data are lacking. The beneficial impact on BMD has been suggested by some prospective and sometimes randomized studies but the sample sizes and follow-up times were always limited (at best 26 patients in the treated group and follow-up of 1 year) [33]. Likewise, there are few clinical data on the benefit of strontium ranelate in dialysis patients and some authors have underlined the risk of osteomalacia with this treatment in dialysis patients [34]. Recent osteoporosis therapies have theoretical interest in dialysis patients. Denosumab has been shown to be effective in dialysis patients but the numbers of included patients were low and all studies were open-label and non-randomized. Moreover, dialysis patients are at higher risk of developing severe hypocalcemia with denosumab [35]. According to some authors, this antiresorptive therapy should also be avoided in dialysis patients with low bone turnover [20]. Teriparatide and other PTH analogs are theoretically interesting, especially in dialysis patients with low bone turnover but, once again, available data are limited to case reports and pilot studies [36]. Lastly, also in the context of low bone turnover disease, new anti-sclerostin antibodies could be beneficial but until now, only animal studies are available [37].

Because of the high risk of fracture in dialysis patients, there is a clear interest to promote clinical research in the field of bone disease. Nephrologists should certainly move from an “only bone turnover” (bone quality) point of view to a broader vision including the concept of bone quantity. This change of concept is necessary because many new and potentially useful therapies are now available. However, there is still need to clarify the place of DXA, QCT, and biomarkers to predict the risk of fractures and/or to evaluate or monitor BMD.

Cardiovascular mortality is the first cause of death in dialysis patients. At least in part, vascular calcifications explain this cardiovascular over-mortality. The pathophysiology of these vascular calcifications has been suggested to share similar pathways with bone physiology. In the future, potential connections between bone (and osteoporosis) and vessels (and vascular calcifications) will be an interesting field of basic and clinical research. Some data suggest an inverse correlation between vascular calcifications and osteoporosis [20, 27]. Therefore, future studies with osteoporosis therapies should be designed with two endpoints; the effect on bone (BMD, biomarkers and ideally fracture) and the effect on vessels (vascular calcification and ideally cardiovascular mortality).

Aromatase inhibitors (AIs) increase overall survival in RCTs against tamoxifen and have become the first-line hormonotherapy in the adjuvant setting of breast cancer. Bone loss is their main side effect. AI therapy is associated with an average 2% loss of lumbar spine BMD per year, and the effects of AIs on cortical bone and bone strength appear to be largely underestimated by classical dual-energy X-ray absorptiometry (DXA) [39]. Risk of fracture is 2–4 times higher in women treated with adjuvant AIs than with tamoxifen or placebo [38, 39]. The increased risk is independent of the type of AI and, with the exception of ABCSG-18, the only trial where fracture incidence was the primary endpoint, the risk has been underestimated because fractures were only reported as adverse events in oncology trials. The absolute risk of fracture in women treated with AIs ranges from 1–18%. Data from ABCSG-18 show a fracture rate of 9.6% after 3 years and 26% after 7 years in the placebo group on AIs only [40].

Published guidelines recommend that women with breast cancer receiving an AI or ovarian suppression therapy have their bone health monitored for fracture risk, for example, with BMD measurement and the FRAX algorithm [38]. In FRAX, however, anticancer treatments are not included as a unique risk factor and enter in the “secondary osteoporosis” group, which underestimates the effect of these therapies on fracture risk [38]. As for other conditions characterized by an increased fracture rate, patients receiving AIs should be advised to consume a calcium-enriched diet and/or receive calcium supplements, exercise moderately (resistance and weight-bearing exercise), and take 1000–2000 international units (IU) of vitamin D everyday. Table 2 summarizes treatment guidelines and possible therapeutic schemes with antiresorptives for women on AI. Guidance from expert groups for premenopausal women with therapy-induced early menopause recommends the use of antiresorptives if the BMD Z-score is <2.0 [41]. In postmenopausal women, the consensus from expert panels recommends treatment with antiresorptives in patients receiving AI therapy with a T-score <2.0 or with two or more clinical risk factors for fracture [38].

Data from randomized clinical trials (RCTs) in >5000 patients show that bisphosphonates and denosumab administered at doses and schedules that are most often similar to those used for postmenopausal osteoporosis can prevent bone loss in women with breast cancer and even lead to an increase in BMD [38, 39]. The ABCSG-18 trial, which randomized postmenopausal women on AIs to denosumab 60 mg q6m or placebo, found that active treatment reduced the risk of first clinical fracture (the primary endpoint) relative to placebo by 50%. Five years following randomization, 15% of placebo patients but little over 5% of denosumab-treated patients had experienced a fracture. A significant protective effect was seen both in women with a baseline T-score < -1 and in those with a T-score ≥ −1 [40]. These new findings will have to be considered when updating guidelines for the prevention of AI-induced bone loss, especially given that denosumab was not associated with additional toxicity. In particular, there was no concern over osteonecrosis of the jaw or atypical femoral fractures [40]. Important additional evidence for the use of antiresorptives in the adjuvant setting is provided by the recent Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) meta-analysis of data from postmenopausal breast cancer patients showing that adjuvant zoledronic acid and clodronate could reduce recurrence rate and prolong survival. Compelling evidence from this meta-analysis of trial data of >18,000 patients supports clinically significant benefits of bisphosphonates on the development of bone metastases and breast cancer mortality in postmenopausal women or those receiving ovarian suppression therapy [47]. An international panel of experts has recently recommended that bisphosphonates, either iv zoledronic acid or oral clodronate, are considered as part of the adjuvant breast cancer treatment in this population. Data from the adjuvant use of denosumab in this setting are eagerly awaited [44, 48].

Androgen deprivation therapy (ADT) is the cornerstone of treatment in prostate cancer but has several adverse effects, especially on bone health. ADT can be achieved by castration, luteinizing hormone-releasing hormone (LHRH) agonists or antagonists. Men undergoing ADT should have their bone health monitored for fracture risk [49]. As for breast cancer, FRAX underestimates the effects of hormone therapy on fracture risk. ADT leads to accelerated bone loss and an increase in fracture rate, as evidenced by large retrospective epidemiological studies [50, 51]. Men treated with LHRH agonists lose 1–5% of BMD within the first year and fracture risk increases with treatment duration [50]. A matched-cohort study of almost 20,000 men found that the risk of fragility fracture (all sites) was 17.2% for those on ADT (mean duration 6.5 years) compared with 12.7% among men not on ADT (HR 1.65; 95% CI 1.39–1.54) [51]. Increasing age is an independent risk factor for fractures, and fracture in prostate cancer more than doubles mortality. The recent introduction of the androgen synthesis inhibitor abiraterone for castration-resistant prostate cancer (CRPC) could aggravate the problem because of its impact on endogenous cortisone production. Abiraterone must indeed be combined with 10 mg prednisone and ADT has to be continued anyway. Abiraterone and other new agents significantly extend overall survival in CRPC, so that prolonged exposure to steroids and ADT may be expected, potentially increasing the risk of osteoporotic fracture. Because of its favorable toxicity profile, abiraterone is especially suited to elderly who are not good candidates for chemotherapy [52].

Table 3 summarizes possible therapeutic schemes with antiresorptives for men on ADT. Alendronate, risedronate, zoledronic acid, and pamidronate have all been shown to prevent BMD loss in patients with locally advanced prostate cancer under ADT [38]. 6–12 monthly zoledronic acid and 6-monthly denosumab are most often used. Prostate cancer is essentially a disease of elderly men who are more likely than younger men to require dose adjustment for renal impairment, which makes denosumab more attractive than bisphosphonates. Moreover, only denosumab has a specific license for ADT-induced bone loss. In a placebo-controlled trial of denosumab in 1468 men receiving ADT for non-metastatic prostate cancer, 3 years of denosumab treatment at doses used for the treatment of osteoporosis led to a 62% relative reduction in new vertebral fractures [53]. Although antiresorptive therapies are especially important for elderly patients with cancer, they are typically underutilized in this population and guidelines for this indication have received less attention. They should probably be broadly similar to those in breast cancer and recent guidelines advocate that all men aged over 75 years under ADT should receive antiresorptive agents at doses used to prevent osteoporosis [46]. Underuse of antiresorptive therapies may be more detrimental in elderly compared with younger patients because of multiple fracture risk factors, including physiological decreases in BMD and increases in fracture rate with advancing age.

Hematopoietic stem cell transplantation (HSCT) is the treatment of choice for many patients with malignant and non-malignant hematological diseases. Osteoporotic fractures constitute one of the main long-term complications. The etiology of bone loss after HSCT is multifactorial, including the use of corticosteroids and other immunosuppressive drugs, release of inflammatory cytokines, and treatment-induced hypogonadism. Even if patients undergoing HSCT are most often less than 50 years, the incidence of osteopenia and osteoporosis in adults after HSCT approaches 50% after 5 years and 20% after 2 years, respectively. The degree of bone loss is most often severe, especially after allo-HSCT, and cortical bone appears to be affected more than trabecular bone [56]. The relative risk of fractures is increased seven- to ninefold compared with 45- to 64-year-old adults in population-based cohort studies [57].

All HSCT patients should receive dietary and lifestyle advice, including calcium and vitamin D supplementation. Menopause hormone therapy (MHT) should be considered in menopausal women, and testosterone potentially in hypogonadal men. Antiresorptives used in this population are mostly bisphosphonates but there are few randomized trials [58]. The pamidronate trials included the largest number of patients and appear to be more representative of the real effects of bisphosphonates in this population. A randomized study including 99 patients under HSCT with or without five 60-mg pamidronate infusions over 1 year showed that in the pamidronate group, BMD remained stable at lumbar spine but still decreased at the femoral neck and total hip [59]. In a smaller study, using higher doses of pamidronate, comparable to the doses used for the treatment of bone metastases, similar findings were reported [60]. There is no reported experience with denosumab in patients after HSCT.

Several studies support the in vivo association of low BMD and sarcopenia. In 679 men aged 40–79 years from the European Male Ageing Study, sarcopenia was associated with low BMD and osteoporosis [68]. Similarly, in 17,891 subjects from various ethnicities, each standard deviation increase in relative appendicular skeletal muscle mass resulted in a 37% reduction in the risk of osteopenia/osteoporosis and subjects with sarcopenia were 2 times more likely to have osteopenia/osteoporosis [69].

In a study in 591 inpatients, sarcopenia was present in 64% of women and 95% of men who recently experienced a hip fracture [70]. Moreover, sarcopenia was associated with lower ability to perform activities of daily living compared to presarcopenia. This may reflect an increased risk of post-hip fracture complications, additional health resources utilization, and higher incidence of contralateral hip fracture. In a population of young patients (20–69 years) with a femoral neck fracture, those with low-energy trauma have significantly lower femoral neck BMD and fat-free mass than those with other trauma mechanisms [71]. These results re-emphasize the association between low bone and muscle mass in patients with hip fractures, and the need for a comprehensive management of these patients.

Furthermore, in Chinese community-dwelling elderly aged 65 and older, sarcopenia was a predictor of fracture risk independent of BMD and other clinical risk factors, and the diagnosis of sarcopenia added incremental value to the FRAX algorithm in predicting incident fracture risk [72].

Finally, in a cross-sectional study in 680 men and women with a mean age of 79 years, sarco-osteoporotic individuals were more likely to have a lower MNA (mini-nutritional assessment) and BMI compared to normal elderly [73]. In Italian hip fracture patients, a low intake of calories, protein, and leucine was associated with reduced muscle mass [74]. This illustrates the importance of detecting nutritional deficits and optimizing nutritional status in patients with osteoporosis and sarcopenia in order to prevent poor clinical outcomes such as falls and fractures.

At this stage, no RCTs assessing the concomitant effects of a new chemical entity (NCE) on bone and muscle have been published. This may be, in part, related to the absence of guidelines supporting regulatory studies for NCE to manage sarcopenia [75]. However, a better understanding of the interconnection of bone and muscle may shift our treatment paradigm to “kill two birds with one stone,” i.e., to treat sarcopenic patients and prevent fractures, as suggested by Gingis et al. [64].

A potential strategy is to target pathways that centrally regulate both bone and muscle (e.g., growth hormone (GH) and GH secretagogues, androgens, selective androgen receptor modulators, and vitamin D) or to investigate newly emerging pathways that might facilitate the communication between the two tissues (e.g., activin signaling inhibitors, including myostatin-neutralizing antibodies/propeptides, recombinant follistatin, follistatin derivatives, and soluble activin receptors or myokines) [63, 64]. An overview of the results of human clinical studies with these agents has recently been published [64].

All authors acknowledge the critical importance of regular exercise and adequate nutrition to optimize peak bone mass and maintain bone and muscle health throughout live [63, 67]. In this perspective, two intervention trials are worth a comment. Bauer et al. reported, in a well-designed 13-week RCT, the beneficial effects of a vitamin D- and leucine-enriched whey protein oral supplement on muscle mass and lower-extremity function among sarcopenic elderly [76]. Knowing the importance of vitamin D and proteins for bone health, this nutritional supplement could also take place in the armamentarium against osteoporosis [77, 78]. Chahal et al. investigated the impact of various intensities and frequencies of loading doses of physical activity on knee extension torque and broadband ultrasound attenuation at the heel in middle-aged women. They concluded that physical activity, especially at high intensity level and high frequency range, may have beneficial effects on muscle strength and bone density in this population [79].

The incidence of osteoporotic fractures increases with age. Today, the cumulative incidence of hip fractures in women at the age of 80 is close to 30% [80]. Vertebral fractures are even more common, with a prevalence of more than 40% in women older than 80 [81]. Moreover, the burden of osteoporosis will only increase in the future because of the aging of the population. In Belgium, it is expected that the number of osteoporotic fractures will increase with 25% in the next 10 years. Osteoporosis in old age is a challenge because of its significant burden in terms of morbidity, mortality, and economic cost.

According to the World Report on Ageing and Health of the World Health Organization (WHO) [82], healthy aging is the process of developing and maintaining the functional ability that enables well-being in older age. Functional ability is made up of the intrinsic capacity of the individual (the composite of all the physical and mental capacities of the individual), relevant environmental characteristics, and the interactions between the individual and the environment. An individual may have reserves of functional ability that he or she is not drawing on. These reserves contribute to the resilience of the individual. Resilience is the ability to maintain or improve a level of functional ability in the face of adverse events. Frailty can be considered as the progressive age-related decline in physiological systems that results in decreased reserves of intrinsic capacity, which leads to extreme vulnerability to stressors and increases the risk of adverse health outcomes [83].

Elderly patients with osteoporotic fractures are not “average” elderly, but should be considered as frail persons, with a high prevalence of underlying comorbidities and at risk of functional deficits [84]. Indeed, in old age, osteoporosis and osteoporotic fractures tend to occur in a particularly frail subset of the population [85]. This frailty will be reflected in poor post-fracture outcomes, such as functional decline, loss of quality of life, and an increased mortality which continues to be observed more than 10 years after the fracture [86].

Despite the increasing evidence for the frequency and severity of osteoporosis in the elderly, osteoporosis continues to be under-diagnosed and under-treated, particularly in individuals over the age of 80. This may, at least partly, be explained by the fact that evidence of the anti-fracture efficacy of osteoporosis treatment comes mainly from RCTs in women with a mean age of 70 to 75 years. Thus, there is an urgent need for treatment options with documented efficacy in older individuals, not only against vertebral fractures but even more so against non-vertebral fractures, as these account for most of the morbidity and mortality associated with osteoporosis. Treatment options should also be proven to be safe in elderly who are frail, with comorbidities and at increased risk of adverse events.

In this chapter, the evidence about the efficacy and safety of the available osteoporosis therapies in the elderly, and especially the oldest old (≥80 year), is discussed. Non-pharmacological interventions such as fall prevention strategies play an essential role in the treatment of osteoporosis, also in elderly, but will not be discussed.