Osteoporosis: Complementary and Integrative Approaches

Introduction

Osteoporosis is a common condition with both significant public health and individual impacts. According to United States (US) Department of Health and Human Services resources, approximately 10 million Americans age 50+ are living with osteoporosis (with about 80% of those affected being women and 20% men), and an additional 43 million people have low bone mass, putting them at risk of developing osteoporosis in the future. ¹ Worldwide, about 200 million women are living with osteoporosis, with that number only expected to increase as the world population ages and longevity in developing countries continues to improve. ² A 2021 analysis based on 86 studies with over 100 million participants ages 15–105 estimated the worldwide prevalence of osteoporosis in women to be 23.1%, whereas in men it was 11.7%. ³

Osteoporosis is characterized by reduced bone strength and therefore an increased risk of bone fracture. A staggering 50% of women will experience at least one bone fracture after menopause. ⁴ The most common types of osteoporotic fractures are vertebral fractures (whether clinical or subclinical), and a prior vertebral fracture is associated with a five times increased risk for additional vertebral fractures, plus a two- to threefold increase in the risk of future fractures at other areas. ⁵ Hip fractures cause the most morbidity and mortality, with only 30%–40% of people who suffer a hip fracture recovering their previous level of functionality, and with the cumulative mortality ranging from 20% to 40% 1 year post hip fracture. ⁶ Importantly, osteoporosis does not cause symptoms (until the occurrence of bone fracture itself). This means an individual with osteoporosis may be completely unaware that they are living with the condition. This makes early diagnosis especially important, so that adequate treatment can be used upfront to slow or even reverse the progression of bone loss.

The diagnosis of osteoporosis is made via dual energy X-ray (DEXA) scan, which generates a T-score. DEXA assesses bone density, which is conventionally used as a proxy for bone strength (with the concept that fracture risk increases as bone mineral density [BMD] decreases being well established in multiple epidemiological studies). The T-score is the number of standard deviations by which the person’s results surpass or fall under the mean for young adults. A T-score that falls over this mean would be positive, and one falling under this is expressed as a negative value. ² Hip (femoral neck) and lumbar BMD measurements are considered most accurate. The diagnostic T-score criteria used to define osteoporosis are seen in the sidebar. Osteoporosis might also be assessed using a Z-score. Rather than using the BMD of healthy young adults as the T-score does, the Z-score compares the individual’s BMD to that of age-matched adults.

In clinical practice, the Fracture Risk Assessment Tool (FRAX), a free online web-based calculator validated in independent cohorts, might also be used to better understand an individual’s risk. FRAX estimates a person’s 10-year probability of experiencing a hip or other major osteoporotic fracture using a number of clinical risk factors. These include age, sex, body mass index, past history of fragility fracture, family history of hip fracture in a parent, smoking, glucocorticoid therapy, excess alcohol intake, and rheumatoid arthritis. FRAX can be used to calculate fracture probability either with or without femoral neck BMD, so may be a useful tool in situations where DEXA is unavailable. As such, it is included in many international guidelines for the diagnosis and management of osteoporosis. ⁷ Although BMD is the standard indicator used for osteoporosis diagnosis, keep in mind that BMD alone has limitations for assessing both risk and adequacy of treatment. The concept of bone strength as a combination of bone density and bone quality may be a more useful concept. Properties impacting bone quality and strength are seen below. ²

Normal bone remodeling and turnover entails balanced activity of osteoclasts (which resorb bone via acidification and proteolytic digestion) and osteoblasts (which secrete bone matrix into the area of resorption). ² This normal, dynamic remodeling process is important for bone repair (such as healing microfractures) and allows bone to be modified in response to stress or biomechanical force. Generally, formation of new bone is closely linked with the process of bone resorption. If these processes become unlinked, bone disease can then be the result. ⁸ In women who are menopausal, the activity of osteoclasts surpasses that of osteoblasts, leading to net bone loss. This alteration is thought to be related to a decrease in osteoblast lifespan with a concurrent increase in osteoclast lifespan. ²

This process of bone remodeling may be reflected in bone turnover markers. These markers, which can be measured in urine or serum, have been used to better elucidate the mechanisms of action for therapeutic agents used in the treatment of osteoporosis. In spite of that, how these markers might best be used in clinical practice to help individual patients remains unknown. Clinical utility is hampered by both laboratory and individual/biological variability seen with measurement of these markers, so these markers alone are not considered sufficient for diagnostic purposes. Reference ranges have been established for P1NP and CTX, and these two markers have been identified as possibly having the greatest clinical utility. ⁸ Research into how these markers might best be used in clinical practice is ongoing, with areas of interest including use of these markers to stratify fracture risk and to assess treatment response. ⁹

Risk factors for osteoporosis include smoking, older age, a history of previous fracture or falls, glucocorticoid therapy, and a family history of hip fracture. Additional risk factors associated with osteoporosis include human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS) and use of antiretroviral drugs, anorexia, history of gastric bypass surgery, diabetes, premature menopause, hyperparathyroidism, and use of aromatase inhibitors, gonadotropin-releasing hormone agonists, and some antiepileptic drugs. Nutritional risk factors include insufficient dietary calcium and protein intake. ¹⁰ Heavy alcohol consumption also increases risk, decreasing bone density and impacting the mechanical properties of bone tissue. ¹¹ Sedentary lifestyle and lack of physical activity also increase the risk of osteoporosis and lower BMD. ^12,13

In addition, low levels of vitamin D may increase risk. Vitamin D helps maintain the balance of calcium to phosphorus and promotes bone matrix mineralization. Absorption of dietary calcium may be hampered by vitamin D deficiency, leading to reductions in bone quality. ¹⁰ In adults age 45 plus, lumbar spine and hip BMD have been shown to be greatest among those with higher serum vitamin D levels. In postmenopausal women, the risk of osteoporosis at different sites (total femur, femoral neck, lumbar spine) has been shown to decrease as serum vitamin D concentration increases. ¹⁴

Treatment of osteoporosis may include lifestyle change and risk factor mitigation, ensuring optimum intake of calcium and vitamin D adequacy, and correcting or treating any underlying contributors such as hyperparathyroidism or hyperthyroidism. Lifestyle modification should include evaluation and counseling for fall prevention, as well as the addition of weight-bearing exercise. Regarding diet, good nutrition is key to bone health at all life stages, especially as it pertains to sufficient intake of dietary proteins and calcium and vitamin D adequacy. ^15,16 In their 2019 guidance publication, the European Society for Clinical and Economic Aspects of Osteoporosis (ESCEO) and the Committees of Scientific Advisors and National Societies of the International Osteoporosis Foundation (IOF) recommend a daily calcium intake between 800 and 1200 mg for women at increased risk of fracture. Supplementation of calcium can be used if the dietary intake falls under 800 mg per day. ESCEO and IOF also recommend 800–1000 IU cholecalciferol daily for postmenopausal women at increased risk of fracture, and note that vitamin D supplementation should also be considered in patients at risk for, or with documentation of, vitamin D insufficiency. ¹⁶ Vitamin D supplementation is discussed at more length below.

Regarding exercise, it should be noted that muscle is a critical factor in bone health. Skeletal muscle mass has been shown to positively correlate with bone mineral content and BMD, and changes in bone mass and muscular strength have been shown to track together over the lifespan. ¹⁷ In addition, many of the factors that contribute to low muscle mass or sarcopenia are also implicated in the development of osteopenia or osteoporosis. These include aging, sedentary lifestyle, environmental factors, and endocrine factors. The co-occurrence of these two processes simultaneously (i.e., osteosarcopenia) is also highly prevalent. ^18,19 Mechanical forces on bone tissue are essential for bone remodeling and increasing bone strength and mass. Both body weight itself and weight-bearing exercise can provide this mechanical force. Muscle contraction itself is the greatest source of voluntary load to bone tissue. ¹⁷ This highlights the importance of exercise throughout the lifespan as well as in people who are either at risk of or with a diagnosis of osteopenia or osteoporosis.

Weight-bearing exercise, individualized for the patient’s ability and needs, is a key part of osteoporosis treatment. ¹⁶ Exercise not only builds muscle strength and bone formation, but also improves balance and coordination. Although the exact “dose” of exercise that is optimal for bone health in people with osteoporosis or osteopenia is unknown, the American College of Obstetrics and Gynecology guidelines defer to the Centers for Disease Control physical activity guidelines: 150–300 minutes per week of moderate intensity aerobic exercise, or 75–150 minutes per week of vigorous intensity aerobic exercise, is recommended. Exercise prescriptions from additional international guiding bodies are seen in the sidebars. Physical therapy consultation to help individualize an appropriate exercise regimen may be necessary, depending on the patient’s functional and clinical status. ¹⁷

Bisphosphonate medications are considered the first-line pharmacologic therapy for osteoporosis, with receptor activator of nuclear factor kappa beta (RANK) ligand inhibitor medication (denosumab) a secondary option for people who have contraindications or adverse effects with bisphosphonates. ²⁰ Bisphosphonates work by inhibiting the activity of osteoclasts, reducing bone resorption. First-line bisphosphonates are generally used for a period of 3–5 years. Although increasing the duration of bisphosphonate treatment beyond 3–5 years reduces the risk of new vertebral fractures, it does not appear to reduce the risk of other fractures, and also carries an increased risk for long-term adverse effects. Therefore, bisphosphonate therapy would generally not exceed 5 years unless there is a significant indication for ongoing treatment. ²⁰ Bisphosphonates are taken up into bone tissue and continue to have an antiresorption effect in the bone even after the medication is discontinued (for at least 10 years). ²¹ The most common reason for oral bisphosphate discontinuation is gastrointestinal side effects. Oral bisphosphonates should be avoided in people at risk of or experiencing gastritis, esophagitis, or gastroesophageal reflux, or in those unable to sit upright for 30 minutes after dosing the medication. Additional adverse effects of bisphosphonates are seen in the sidebar. ²² The risk of osteonecrosis of the jaw is increased with periodontal and dental disease, invasive dental procedures, and dental extraction or trauma. Patients initiating bisphosphonate therapy should have a baseline dental examination to assess for the need for surgical dental procedures (if personal risk factors permit) before initiating treatment. ²³

This article reviews complementary and integrative interventions shown to have evidence in people with osteoporosis. These include strontium, choline-stabilized orthosilicic acid (OSA), isoflavones, vitamin D, and vitamin K.

Strontium

Strontium is an alkaline metal that naturally occurs as a mixture of four isotopic forms. Strontium readily oxidizes to form strontium oxide, and as such is not naturally found in its free form. Strontium is considered a trace element in the human body and is obtained via the diet (leafy vegetables, dairy foods, and grains being the main sources) and drinking water. The estimated daily intake of strontium from such sources in adults is ∼4 mg. Most body strontium is located in bone, but compared with calcium, it is a minor component of bone tissue (strontium levels are about 0.035% that of calcium levels in bone tissue). ²⁴

Strontium and calcium share some physical and chemical properties. ²⁴ Strontium ranelate, a synthetic strontium salt of ranelic acid, has been shown to have an anabolic effect and stimulates bone formation while reducing bone resorption. ^25,26 Aside from the ranelate form, strontium is also available in a citrate form (more on this below), and the citrate, gluconate, and carbonate forms of strontium are naturally occurring rather than synthetic. ²⁷ Although the complete mechanisms by which strontium performs its actions in bone are still incompletely understood, the following has been demonstrated or proposed:

Strontium increases osteoblast proliferation and differentiation.

Strontium (particularly strontium ranelate) has been the subject of numerous clinical studies in people with osteoporosis. Some of the initial clinical trials demonstrating the efficacy of strontium ranelate in people with osteoporosis were the Spinal Osteoporosis Therapeutic Intervention (SOTI) study and the Treatment of Peripheral Osteoporosis (TROPOS) study. In SOTI, 1649 postmenopausal women with vertebral osteoporosis experienced a reduction in the relative risk (RR) of new vertebral fractures of 41% at 3 years, with strontium ranelate at 2 g daily. ²⁹ In the multinational TROPOS study, over 10,000 subjects were included, with results showing the strontium ranelate at this same dose (2 g daily) resulted in increased BMD and a reduced risk of both vertebral and nonvertebral fractures over 3 years. ³⁰ Additional trials then followed SOTI and TROPOS, confirming strontium ranelate’s efficacy.

In a 2011 meta-analysis, Kanis et al. pooled results from two phase III studies on the effects of strontium ranelate in postmenopausal women with osteoporosis. Both of the included trials were multicenter, double-blind, randomized, placebo-controlled studies. Included trials were conducted across 11 European countries and Australia (N = 6740). Strontium ranelate was dosed at 2 g daily for 3 years. All subjects, whether randomized to strontium or placebo, were also give calcium and vitamin D. Participants were followed for up to 4.6 years. Strontium ranelate treatment was found to lead to a 31% decrease in osteoporotic fractures (95% confidence interval [CI]: 20%–39%) and a 40% decrease in vertebral fractures (95% CI: 31%–48%) compared with placebo. The effects of strontium were found to be independent of baseline fracture probability (as assessed by FRAX), suggesting that the efficacy of strontium ranelate is comparable over a range of FRAX values. ²⁸

In a 2019 network meta-analysis comparing various osteoporosis treatments, strontium ranelate did not have a significant effect in reducing hip or nonvertebral fractures, but did reduce the risk of vertebral fractures. Compared with placebo, strontium demonstrated an RR of 0.60 in vertebral fracture reduction. This was compared to the following osteoporosis therapies: abaloparatide (RR: 0.14), teriparatide (RR: 0.27), parathyroid hormone (RR: 0.41), romosozumab (RR: 0.33), denosumab (RR: 0.32), zoledronate (RR: 0.38), risedronate (RR: 0.61), alendronate (RR: 0.57), ibandronate (RR: 0.67), raloxifene (RR: 0.59), bazedoxifene (RR: 0.61), lasofoxifene (RR: 0.67), estrogen with progesterone (RR: 0.65), tibolone (RR: 0.56), and calcitonin (RR: 0.65). ³¹

In a 2010 analysis of placebo-controlled, double-blind, phase III randomized controlled trials (RCTs) of antiosteoporotic agents in postmenopausal osteoporosis patients with one fracture at baseline, the number needed to treat (NNT) was calculated for various forms of treatment. In this analysis, vertebral and hip fracture over a 3-year period were utilized as outcome since these are the most common sites for osteoporotic fracture. The NNT to prevent one vertebral fracture over 3 years was 9 for strontium ranelate at a dose of 2 g per day. The NNT to prevent a hip fracture over 3 years was 48 for strontium ranelate at a dose of 2 g per day. This is compared to an NNT of 21 for ibandronate and 91 for bisphosphonates, for vertebral and hip fracture, respectively. ³²

In a 2006 Cochrane systematic review, four RCTs in postmenopausal women, comparing strontium ranelate to placebo, were included. All of the included trials were at least 1 year in duration. Three of these examined the effects of supplementation in a treatment population, with doses ranging from 0.5 to 2 g daily. The remaining RCT examined the effects of strontium supplementation preventively at doses of 0.125, 0.5, and 1 g daily. ³³

In treatment populations, 3 years’ supplementation of strontium ranelate at 2 g daily resulted in a 37% decrease in vertebral fractures and a 14% decrease in nonvertebral fractures (RR 0.63, 95% CI: 0.56–0.71, and RR 0.86, 95% CI: 0.75-0.98, respectively). Increased BMD was seen at all sites after 2–3 years of strontium supplementation in both treatment and prevention populations. The highest dose of strontium demonstrated the greatest BMD increases and vertebral fracture reductions. Although there were no serious adverse effects, there was an increased risk of diarrhea with the 2 g per day dose of strontium ranelate. The authors also noted that the risk of vascular and nervous system side effects is increased with taking 2 g strontium ranelate daily for a 3- to 4-year period. The overall rate of vascular events (such as venous thromboembolism [VTE] or pulmonary embolism) was 26.3% for the strontium group subjects (880/3352 subjects) compared with 24.4% in controls (809/3317), and the overall rate of nervous system disorders (such as headache or memory changes) was 20.9% (699/3352 subjects) for strontium and 18.9% for controls (627/3317 subjects). The cause of this increased risk for neurological or vascular side effects was unknown. ³³

With the above studies indicating its efficacy, strontium ranelate was previously available in Europe in prescription form (it was never approved by the FDA and hence was not available in prescription form in the US). In 2014, however, the Pharmacovigilance Risk Assessment Committee of the European Medicines Agency (EMA) recommended that strontium ranelate no longer be used for osteoporosis. This decision was based on annual periodic safety update reporting (the mechanism by which manufacturers regularly submit safety data to the EMA) indicating possible cardiac events with strontium ranelate, which prompted further analysis of the benefits and risks of the medication. ^34,35 Subsequent studies did not definitively demonstrate an increased risk of myocardial infarction with strontium, compared with bisphosphonate use, or with current use of strontium ranelate versus prior use, but did demonstrate an increased risk of VTE (both with current strontium ranelate, compared with current bisphosphonate use, and for current versus past strontium ranelate use). ³⁴

The questions of whether or not other doses, as well as other forms of strontium (such as strontium citrate or strontium chlorate), work differently or have a different safety profile than the ranelate form have not been extensively studied. Two small, shorter-term (12-month) trials of lower dose strontium citrate in combination with other bone nutrients have been performed. In the 2012 COMB study (Combination of Micronutrients for Bone), a combination of vitamin D3, vitamin K2, strontium citrate at a dose of 680 mg daily, magnesium, and docosahexaenoic acid was given for 12 months (N = 77, 5 men and 72 women). Of note, 29 subjects (38%) reported lack of results with previous use of bisphosphonates, and 48 subjects (62%) had declined a recommendation for standard drug therapy such as bisphosphonate. After 12 months of the nutrient combination, there was a significant improvement in bone density (as evidenced by Z-scores) at the femoral neck and spine (mean pretreatment Z-score at the femoral neck −0.51 ± 0.74, post-treatment −0.24 ± 0.81, P = 0.03; mean pretreatment Z-score at lumbar spine −0.85 ± 0.98, post-treatment −0.39 ± 1.07, P = 0.006). Improvement was also seen in the mean total hip score, but this was not statistically significant (mean pretreatment Z-score −0.27 ± 0.82, post-treatment − 0.06 ± 0.84, P = 0.12). It was also noted that BMD change at the hip was more pronounced in subjects who had reported lack of success with past bisphosphonate use. The authors made no mention of adverse effects in this study, but noted that to their knowledge, there are no thrombotic effects associated with strontium citrate use. ²⁷

In the 2017 MOTS study (Melatonin-micronutrients Osteopenia Treatment Study), a nutrient combination with strontium citrate was also used. Postmenopausal women with osteoporosis and with moderate fracture risk per FRAX were randomized to either placebo or to the supplement group (taking 5 mg melatonin, 450 mg strontium citrate, 2000 IU vitamin D3, and 60 µg vitamin K2 as menaquinone-7) for 1 year. This was a small study, with a total N of 22 (age 58.6 ± 1.12 [age range 49–75], 11 subjects each in the placebo and intervention group). At the completion of the trial, participants who received the nutrient supplement had an average BMD change at the femoral neck of +0.015 (2.2%), whereas placebo group participants saw an average femoral neck BMD change of −0.023 (−3.6%). For the lumbar spine, women taking the nutrient combination saw a BMD change of +0.035 (4.3%), compared with −0.029 (−3.2%) for the placebo group (P < 0.001). There were no adverse effects reported for either the active intervention or the placebo group in this small trial. ³⁶

Other sources have raised the concern that replacement or substitution of bone calcium with strontium might lead to defective bone mineralization, similar to what is seen in osteomalacia. Strontium has a higher atomic mass than calcium, so it is thought that some portion of the increased BMD seen on DEXA with strontium supplementation may be artefactual. ³⁴ In relation to calcium, strontium and calcium appear to share similar routes of absorption, but calcium is absorbed preferentially over strontium in the gut ³⁷ (suggesting it might be best to separate dosing of calcium and strontium supplements). Regarding dose, it is thought that while lower doses of strontium stimulate bone formation, higher doses may have a negative effect on bone mineralization via decreased calcium absorption and potential changes to bone mineral content. This may be especially pertinent if calcium intake is low. ³⁸ This effect has been demonstrated in animal models of osteoporosis, but again, the question of how this might translate into bone effects in humans has not been extensively explored. Osteonecrosis of the jaw has not been demonstrated with strontium use. ³⁸

Silicon and Choline-Stabilized OSA

The trace element silicon (Si) has been proposed to play a potential role in bone formation as well as bone strength. Silicon participates in the formation of cross-linkages between collagen and proteoglycans, and with the bone matrix predominantly being composed of collagen, would therefore participate in supporting the mechanical properties of bone. ³⁹ Silicon deficiency in animals has been shown to result in the production of more porous bone tissue, and greater dietary intake of silicon is positively associated with some measures of BMD in humans. ^39,40 Silicon is rarely seen in nature in its free/elemental form, instead occurring mostly as silicon dioxide. Both Si and silicon dioxide are poorly water soluble and also considered poorly bioavailable. OSA, on the other hand, is a water-soluble form of Si and is readily absorbed in the digestive tract. Excess OSA is cleared from the body by renal excretion, so is thought to be unlikely to accumulate or become toxic provided kidney function is healthy. ⁴¹

A single clinical trial has been conducted on a choline-stabilized OSA complex in postmenopausal women with osteopenia. For 12 months, 136 women (mean age 60.7 ± 10.4 years; baseline T-score < −1.5 at the lumbar spine as evidenced by DEXA scan) took 1000 mg calcium, 20 µg of cholecalciferol, and either choline-stabilized OSA or a placebo. For OSA, participants were randomized to one of three doses, providing 3, 6, or 12 mg Si daily. At the conclusion of the trial, P1NP levels were significantly higher with supplementation of choline-stabilized OSA at 6 and 12 mg daily, compared with placebo (P < 0.05). Femoral and lumbar spine BMD were not significantly different with choline-stabilized OSA compared with placebo. In subgroup analysis, femoral neck BMD was significantly improved at the 6 mg per day dose of choline-stabilized OSA compared with placebo (P < 0.05). There were no adverse events related to the use of choline-stabilized OSA during the trial. The authors speculated that the lack of observed effect of OSA supplementation on lumbar spine BMD may have been related to a reduced power of this study to detect small changes in BMD. ⁴²

Isoflavone Supplementation

Phytoestrogens are naturally occurring nonsteroidal plant-derived compounds with structures similar to 17-β-estradiol. Although there are different classes of phytoestrogens, isoflavones are one of the most common types and are generally found in their conjugated forms in legumes as part of the diet. After ingestion, isoflavones are hydrolyzed into aglycones by gut flora. This enhances bioavailability of the parent compound as well as its estrogenic activity. ^43,44 Both ipriflavone, a synthetic flavonoid compound, and soy isoflavone may have therapeutic effects in women with osteoporosis. Ipriflavone is not thought to possess direct estrogenic activity, but rather might inhibit osteoclast function by reducing the recruitment or differentiation of pre-osteoclasts. ⁴⁵ Likewise, soy isoflavones are thought to inhibit the differentiation and activation of osteoclasts. ⁴⁶

Studies on ipriflavone have ranged from trials in a preventive setting, to studies in women with existing osteoporotic fractures. In one preventive clinical trial, the effects of ipriflavone alone, or ipriflavone with low-dose hormone therapy (HT), were compared. Women in early menopause (N = 105, last menstrual period 6–36 months prior) were randomized to one of four groups for 12 months:

Control group: 500 mg calcium daily

After 12 months, women in the control group saw a significant decrease in spine BMD by −3.41% (P < 0.05). Women in the HT group who received the lower dose of E2 (25 µg/day) also experienced a decrease in BMD, −0.55%. Those who received the higher dose E2 (50 µg/day) experienced a 1.84% increase in BMD. Women in the ipriflavone-only group had a 0.11% increase in BMD after 12 months. Ipriflavone and HT were not found to potentiate each other: in the combination group, participants saw a −0.22% change in BMD after 12 months (the authors offered no explanation/speculation for why this was the case, although note that the combination group received the lower dose of E2). ⁴⁷

In women with existing osteoporosis or osteopenia, ipriflavone supplementation has compared favorably to calcium alone for maintaining BMD. In the study by Ohta et al., 60 postmenopausal women with osteopenia or osteoporosis were given either ipriflavone at 600 mg daily or calcium at 800 mg daily, for 12 months. At the conclusion of the trial, spine BMD was similar before and after treatment in the ipriflavone group (0.78 and 0.77 g/cm², respectively). For women receiving calcium alone, however, spine BMD significantly decreased from 0.81 to 0.79 g/cm² (P < 0.0001). Median deoxypyridinoline levels also decreased significantly with ipriflavone supplementation, from 10.2 mmol/mmol creatinine (Cr) to 5.8 mmol/mmol Cr, whereas in the calcium group, they were unchanged. This suggests that ipriflavone works by suppressing bone resorption. ⁴⁸

Two trials in osteoporotic women with existing vertebral fractures have also indicated a benefit to ipriflavone supplementation. In one of these, 149 women (ages 65 to 79) were randomized to either ipriflavone 600 mg daily or a placebo, for 2 years. All participants took calcium 1000 mg daily as well. In total, 111 women completed the study. Compared with the placebo, ipriflavone resulted in a significant increase in BMD measured at the forearm (+4%, P < 0.05), a significant reduction in urinary hydroxyproline (suggesting, as with the study above, a reduction in bone turnover), and a reduction in the incidence of vertebral fractures. Some subjects also reported reduced bone pain and analgesic use. The occurrence of adverse reactions, mostly gastrointestinal (GI) symptoms, was similar for ipriflavone (14.5%) and placebo (16.1%). ⁴⁹

In a second trial in women with existing vertebral fractures, 100 women (age 65+) were again randomized to either ipriflavone 600 mg daily or a placebo, for 2 years (all participants also received calcium at 1000 mg daily). In total, 84 women completed the trial (41 in the ipriflavone group and 43 in the placebo group). Subjects receiving ipriflavone experienced a significant increase in BMD at the completion of the trial (P < 0.05) and again saw decreases in pain and analgesic use, as well as urinary hydroxyproline. Two new fractures occurred in women taking ipriflavone during the trial (11 fractures occurred in the placebo group). By contrast, women in the placebo group saw a decrease in BMD (P < 0.05) and reported increased pain and use of analgesic medications. ⁵⁰

In a 2020 systematic review and meta-analysis, 11 RCTs of ipriflavone with a total of 1605 subjects with osteopenia or osteoporosis were assessed. Ipriflavone was dosed at 600 mg daily, and most participants also received 1000 mg calcium daily. Trials ranged from 6 months to 4 years in duration. Overall, the increase in lumbar spine BMD for ipriflavone was greater than that seen with placebo (standardized mean difference = 0.36; 95% CI: 0.09–0.62). GI symptoms were the most frequently reported adverse reactions with ipriflavone, but overall withdrawals related to side effects were no different for ipriflavone and placebo groups. The authors noted the occurrence of subclinical lymphocytopenia (total lymphocyte concentration <500/µL) in two of the included trials (although in one of these trials the decrease in lymphocyte count was not statistically significant), which was reversible with discontinuation of ipriflavone. ⁵¹

Regarding the safety of ipriflavone supplementation, in Alexandersen et al.’s 2001 clinical trial (this was a 4-year trial of 600 mg ipriflavone daily, compared with placebo, which found no significant improvement of BMD with supplementation), 31 subjects (13.2%) in the ipriflavone group developed subclinical lymphocytopenia. By 1 year, 52% of these cases of lymphocyte decrease had resolved spontaneously, and by 2 years, 81% had resolved. ⁵² In the study by Agnusdei and Bufalino (1997), <3% of patients experienced lymphocyte values out of the normal range with long-term ipriflavone treatment. ⁴⁹ Until additional data are available, it would seem reasonable to check the complete blood count in people initiating ipriflavone treatment, which could then be repeated at 6 and 12 months, and yearly thereafter, for observation.

Human pharmacokinetic studies indicate that ipriflavone and its metabolites are eliminated within about 120 hours. Ipriflavone does not accumulate in body tissues (including bone). Agnusdei and Bufalino suggest adjusted dosing in people with renal impairment, based on the finding that plasma ipriflavone levels are slightly increased in people with renal impairment. They recommend a dose of 400 mg daily in people with mild to moderate renal impairment (i.e., at a Cr clearance of 40–80 mL per minute). ⁴⁹

Aside from ipriflavone, soy isoflavones have also been shown to improve BMD. Both a 2008 meta-analysis and a 2010 meta-analysis demonstrated a positive effect of soy isoflavones. In the first of these, 10 studies with a total of 608 subjects were included. Spine BMD among women who supplemented with soy isoflavones significantly increased from baseline (by 20.6 mg/cm², 95% CI: 4.5–36.6 mg/cm²), and soy isoflavones increased spine BMD, compared with placebo as well (but this finding was considered borderline significant). When the analysis was restricted to studies using doses >90 mg daily, the increase in BMD was even greater. Increases in spine BMD with soy isoflavones at doses >90 mg/day, and for periods of at least 6 months, were 28.5 mg/cm² (95% CI: 8.4–48.6 mg/cm²) and 27 mg/cm² (95% CI: 8.3–45.8 mg/cm²). Equol-producer status of the subjects was not examined. ⁵³

In the second of these meta-analyses on soy isoflavones, 11 RCTs with a total of 1240 menopausal women were included. Supplementation with an average of 82 mg soy isoflavones daily for 6 to 12 months (doses for included trials ranged from 47 to 150 mg daily) led to significant increases in spine BMD by 22.25 mg/cm² (95% CI: 7.62–32.89 mg/cm²; P = 0.002), an increase of 2.38% compared with controls (95% CI: 0.93–3.83 mg/cm²; P = 0.001). ⁵⁴ Soy protein supplementation at doses of 25 g daily (providing 60 mg isoflavones) for 1 year does not cause lymphocytopenia in postmenopausal women. ⁵⁵

Vitamin D

Vitamin D, the hormone synthesized with exposure of the skin to sunlight, is crucial for bone health. It helps control the homeostasis of calcium and phosphorus via its activities in the kidneys, gut, and bone tissue. This occurs via a number of mechanisms:

Promotes the absorption of dietary calcium in the intestine.

Cutaneous vitamin D production can be affected by a number of factors. These include latitude, aging, sunscreen and clothing use, seasonal changes, skin pigmentation, and time of day. ^57,58 Vitamin D deficiency is considered to be a global public health problem and is thought to affect up to billion people worldwide. Deficiency can even be seen in significant portions of the population in locations with favorable/sunny climates. ⁵⁹ For example, an estimated 80% of the urban population in India is vitamin D deficient (and hip fractures in India occur roughly a decade earlier than they do in Western countries). ³ Although the benefits of vitamin D sufficiency would seem obvious, clinical trials on the impact of vitamin D supplementation on BMD have provided conflicting results. A summary of various meta-analyses on vitamin D supplementation and BMD or fracture outcomes is seen in Table 1.

As seen in the table, a variety of dosing regimens, durations, and outcomes were seen in these analyses. ^{60 –63} In Reid et al.’s 2014 meta-analysis, specifically examining effects on BMD, there was a small benefit to BMD at the femoral neck (weighted mean difference [WMD] 0.8%, 95% CI: 0.2–1.4) but not at the hip with vitamin D supplementation. ⁶¹ The remainder of these meta-analyses specifically looked at fracture risk, with two pointing to a reduction of fracture risk with calcium plus vitamin D, and one finding no association. ^60,62,63

Some trials included in these analyses assessed for baseline and follow-up 25OHD3 concentrations, while others did not, so it is unknown if the doses used in these studies were sufficient to correct underlying deficiency in all subjects, or even how many participants may have been vitamin D replete versus deficient to start. It seems reasonable that the individual’s baseline vitamin D status and response to supplementation could impact the results seen. It has also been suggested that some trials may use doses that are too low to demonstrate a benefit, that the duration of some studies may be too short to detect a desired outcome, or that the selection of the study population may be responsible for inconsistent results seen in these vitamin D trials. ⁶⁴ Systematic reviews suggest that a reduction in fall risk with vitamin D supplementation is more likely to be seen in vitamin D-deficient individuals, and many trials looking at fall risk have used doses <1000 IU daily, with only small numbers of participants who are vitamin D deficient (again, probably impacting the results achieved). ⁶⁵

Based on the available evidence, the ESCEO advises that 1000 IU daily vitamin D3 should be recommended in patients at increased risk of vitamin D deficiency. This at-risk group is likely to include large numbers of older patients with a variety of risk factors or comorbidities (see sidebar). ⁶⁵ For the achievement of bone health in the general population, the Institute of Medicine and the Endocrine Society recommend cutoff levels for vitamin D sufficiency of 50 and 75 nmol/L, respectively (20 ng/mL and 30 ng/mL). ¹⁴

Vitamin K

The vitamin K family of nutrients are fat-soluble compounds that play a role in bone health via their participation in osteocalcin function. Osteoblasts synthesize and release osteocalcin, which, once carboxylated, binds to hydroxyapatite to facilitate calcium deposition in the bone. Undercarboxylated osteocalcin is not biologically active and does not have this binding activity. ⁶⁶ Vitamin K is a coenzyme for vitamin K-dependent carboxylase, which participates in the carboxylation of osteocalcin. ^66,67 Therefore, vitamin K is required for the conversion of osteocalcin to the active, carboxylated form, and for appropriate hydroxyapatite binding.

The vitamin K compounds include phylloquinone (also known as vitamin K1) and the menaquinones (i.e., MK forms of vitamin K2, of which MK-4, MK-7, and MK-9 are the most studied). Vitamin K1 is primarily obtained from leafy greens in the diet, whereas menaquinones are primarily sourced from bacteria. Menaquinones are found in small amounts in fermented foods and animal foods and are also produced by human gut flora. ⁶⁷ Dietary sources of vitamin K compounds are seen in Table 2. ^68,69

Although vitamin K is necessary for appropriate osteocalcin function, studies have yielded varying results in terms of the impact of supplemental vitamin K on BMD or fracture risk. Older studies give a sense of these mixed results. These included the following:

A 2006 meta-analysis of RCTs, most of which were conducted in Japan and in postmenopausal women. Seven RCTs included fracture data and all of these used MK-4 at either 15 mg or 45 mg daily, finding a significant reduction of hip, vertebral, and nonvertebral fractures. ⁷⁰

Two newer (both published in 2022) and larger meta-analyses have indicated positive results with vitamin K supplementation. In Zhou et al., nine RCTs with 6853 postmenopausal women with osteoporosis were included. Vitamin K2 (form not further specified) was dosed at 45 mg daily in all trials, and follows ups ranged from 12 to 48 months. Use of vitamin K2 was associated with a significantly increased percent change in lumbar and forearm BMD (WMD 2.17, 95% CI: 1.59–2.76; and WMD 1.57, 95% CI: 1.15–1.99, respectively). Five of the included trials used calcium supplementation alone as a comparator, but in subgroup analysis, vitamin K2 plus calcium did not perform significantly differently in terms of fracture risk (RR = 0.95, 95% CI: 0.81–1.12, P = 0.55; although, also note that doses of calcium varied widely in those 5 trials, from 150 mg to 2000 mg per day). In terms of side effects, data on adverse reactions were available from 6 studies for a total of 6179 participants (3077 for vitamin K2 and 3102 controls). Although pooled analysis indicated that vitamin K2 groups seemed to have higher rates of side effects (RR = 1.33, 95% CI: 1.11–1.59, P = 0.002), there were no serious adverse events, and the side effects reported largely consisted of minor GI symptoms. ⁷⁵

In Ma et al., 16 RCTs of vitamin K2 with 6425 postmenopausal women were analyzed. All but one of the included trials identified the form of vitamin K2 used (MK-7 in 4 trials, in doses ranging from 180 to 375 µg daily, and MK-4 in 11 trials, almost all dosed at 45 mg daily), and follow ups ranged from 6 to 48 months. For the 10 studies that reported on lumbar spine BMD, there was a significant improvement with the use of vitamin K2 (mean difference [MD] = 1.02, 95% CI: 0.30–1.75, P = 0.006). In subgroup analysis of those trials that utilized vitamin K2 as part of a combination therapy (with bisphosphonate, calcium, and/or vitamin D), lumbar spine BMD was also significantly improved (MD = 1.97, 95% CI: 0.20–3.74, P = 0.03). A pooled analysis showed no significant difference between vitamin K2 and control groups for adverse reactions (RR = 1.03, 95% CI: 0.87–1.21, P = 0.76). ⁶⁶

Vitamin K would of course be expected to work hand-in-hand with other bone nutrients, such as calcium and vitamin D. It seems plausible that varying doses of calcium and vitamin D given in vitamin K trials (or failure to co-administer these nutrients at all) might be responsible for conflicting results seen in some earlier trials. ⁶⁷ In addition, MK-7 is thought be more active than MK-4 at the same concentration, ⁷⁵ making it challenging to compare trials using different forms of vitamin K2 that demonstrate different results.

Discussion

Although a diagnosis of osteoporosis may engender fear, patients can be reminded that bone is a highly dynamic tissue. Bone is a living structure, continually adapting over the course of the lifespan. An estimated 10% of the skeleton is replaced annually. ⁷⁶ People with existing osteoporosis or osteopenia, or who are interested in prevention, have the opportunity to make changes that capitalize on this normal bone remodeling process.

A diagnosis of osteoporosis needs to be met with a multimodal intervention plan that may incorporate pharmacotherapy, smoking cessation, avoidance of excess alcohol consumption, calcium intake, weight-bearing exercise, and prevention of falls. ⁵ This review provides the evidence for use of additional integrative therapies that may be considered to help support bone density in people with osteoporosis. This includes strontium, choline-stabilized OSA, isoflavones (ipriflavone and soy isoflavones), vitamin D, and vitamin K. Although the studies detailed above have generally examined these substances as isolated interventions, many clinicians would be using these substances in combination, and as a complement to the multimodal intervention plans and lifestyle modifications mentioned above. New clinical trials with high-quality methodology that examine the effects of combination (rather than isolated) interventions would be highly clinically relevant and are much needed.