Osteoporosis/Fracture Prevention Clinical Practice Guidelines

Transcription

1 NATIONAL CLINICAL PRACTICE GUIDELINES Osteoporosis/Fracture Prevention Clinical Practice Guidelines Reviewed/Approved by the National Guideline Directors September 2017 Next Review/Approval: September 2019 Developed by the National Osteoporosis/Fracture Prevention Guideline Development Team Disclaimer These guidelines are informational only. They are not intended or designed as a substitute for the reasonable exercise of independent clinical judgment by practitioners, considering each patient s needs on an individual basis. Guideline recommendations apply to populations of patients. Clinical judgment is necessary to design treatment plans for individual patients.

2 Table of Contents National Osteoporosis/Fracture Prevention Clinical Practice Guidelines... 3 All adults: Non-pharmacologic primary prevention of osteoporosis... 4 Low Bone Mass (T-score -1.1 to -2.4): Pharmacologic Treatment... 5 Osteoporosis (T-score -2.5 or prior fragility fracture): Pharmacologic Treatment... 6 Discontinuation of Bisphosphonate Treatment... 8 Screening... 8 Appendix A: National Osteoporosis/Fracture Prevention GDT Appendix B1. Summary of Recommendation Decisions Appendix B2. Crosswalk of Recommendation Ratings Appendix C. Rationales Low bone mass (T-score -1.1 to -2.4): Pharmacologic treatment Osteoporosis (T-score -2.5 or prior fragility fracture): Pharmacologic treatment Discontinuation of Bisphosphonate Treatment Appendix D. Evidence Review All adults Non-pharmacologic primary prevention of osteoporosis Low Bone Mass (T-score -1.1 to -2.4) Pharmacologic treatment Validity of FRAX in postmenopausal women with low bone mass Fall risk in postmenopausal women with low bone mass Osteoporosis (T-score -2.5 or prior fragility fracture) Pharmacologic treatment (excluding calcitonin) Calcitonin Treatment in men with osteoporosis Men and women taking corticosteroids Discontinuation of Bisphosphonate Treatment Bisphosphonate discontinuation Long-term Safety Screening Screening for osteoporosis Screening for Vit D deficiency AMSTAR assessments References

3 National Osteoporosis/Fracture Prevention Clinical Practice Purpose Guidelines This guideline was developed by the KP National Osteoporosis/Fracture Prevention Guideline Development Team (GDT) (Appendix A) to assist Primary Care physicians in the primary prevention, screening, and treatment of low bone mass and osteoporosis. Background This guideline is an update of the 2015 KP National Osteoporosis/Fracture Prevention Guideline. Major changes include updates to the section Discontinuation of Bisphosphonates. Minor changes include an update to the language and strength of the recommendations to align with current National Guideline Program methodology. Methods KP National Guideline Program follows a methodology 1 that incorporates well-established scientific frameworks to critically appraise evidence and evaluate external guidelines. In developing these recommendations, the GDT adopted the 2015 KP Osteoporosis/Fracture Prevention Guideline recommendations, and updated the recommendations for discontinuation of bisphosphonates. The updated recommendations are based on data from landmark efficacy trials (FLEX, HORIZON-PFT) and multiple published systematic reviews for harms. The GDT s decisions and justification are outlined in Appendix B1. The KP National Guideline methodology, updated in its entirety in 2012, categorizes recommendation strength as strong, conditional, or no recommendation for or against. In December 2017, the methodology was revised to change the strength of recommendation label from weak to conditional. The intent of the recommendation remains the same. Clinicians should consider the strength of the recommendation when determining the appropriate course of action based on patient s individual situation. The current KP guideline recommendation language and strength evolved from previous KP rating systems. Appendix B2 provides a crosswalk to compare language between current and previous methodology. GDTs develop guideline recommendations using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) criteria, which considers the balance between desirable and undesirable effects, quality of evidence, patient values and preferences, and resource use. When the GDT changes an existing KP guideline, adapts an external guideline with modifications, or reviews new evidence to inform its updated recommendation, the GRADE framework provides transparency into those decisions. Appendix C summarizes GDT deliberations from the GRADE process. KP recommendations that were unchanged and external guidelines that were adopted verbatim are not subject to GRADE assessment. Additional evidence review work that informed the development of these recommendations is documented in Appendix D. This includes assessment of the methodological rigor of systematic reviews using the AMSTAR criteria; descriptions of original systematic reviews conducted by KP; and summaries and excerpts from individual studies or systematic reviews used to inform the recommendations. 3

4 Recommendations Prevention and Treatment of Osteoporosis All adults: Non-pharmacologic primary prevention of osteoporosis Lifestyle Choices: For all adults, consider recommending regular weight-bearing and muscle-building exercise for prevention of osteoporosis and falls. (Conditional recommendation) For all adults who are current smokers, consider recommending smoking cessation for prevention of osteoporosis and fragility fractures. (Conditional recommendation) Fall Prevention: For adults at increased risk of falling, consider safety proofing their residence. (Conditional recommendation) o Home safety proofing includes removing rugs, adding grab bars, establishing adequate lighting (eg, nightlights), and securing electrical cord placement. For post-menopausal women and men aged 50 years, the routine use of hip protectors to reduce the risk of hip fractures is not recommended. (Strong recommendation) Supplementation: For all adults, advise patients to consume the recommended total daily intake* of calcium and vitamin D. (Good Practice Statement) Consider prescribing supplemental calcium and vitamin D if recommended daily intake is not achieved through diet alone. (Conditional recommendation) In adults aged 50 years without osteoporosis, do not screen for vitamin D deficiency. (Strong recommendation) *USDA recommended daily allowance (RDA): Calcium o Women: o Aged years: 1,000 mg Aged 51 years: 1,200 mg Men: Aged years: 1,000 mg Aged > 70 years: 1,200 mg Vitamin D o All adults: Aged years: 600 IU Aged > 70 years: 800 IU Clinical Considerations o If supplementation is necessary, vitamin D3 (ergocalciferol) is preferred o Calcium carbonate contains the most elemental calcium per dose. It should be taken with food to enhance absorption. o Calcium citrate contains less elemental calcium than the carbonate salt, but it is better absorbed and may be preferred in patients with reduced gastric acid production or high gastric ph requiring long term H2 antagonist or proton pump inhibitor therapy and in patients who have undergone bariatric surgery. It is more expensive and usually requires more tablets to be taken per day than calcium carbonate. 4

5 Low Bone Mass (T-score -1.1 to -2.4): Pharmacologic Treatment Consider treating women with low bone mass (and no history of prior fragility fracture) who have at least one of the following additional risk factors: o A FRAX 10-year risk of hip fracture 3% or a FRAX 10-year risk of major osteoporotic fracture 20% a o Increased risk of falling b (Conditional recommendation) Preferred therapy o Optimize nonpharmacologic therapies, including vitamin D supplementation, calcium intake, and fall prevention efforts (via physical activity or therapy, ambulatory assistive devices, and reduction in sedative/hypnotic medications). (Strong recommendation) o Consider prescribing alendronate (70 mg/week) as a first-line therapy, particularly in those with a lowest T-score in the range of -2.1 to -2.4 and at least one of the following additional risk factors: A FRAX 10-year risk of hip fracture 3% or a FRAX 10-year risk of major osteoporotic fracture 20% a Increased risk of falling (Conditional recommendation) Alternative therapy o Consider prescribing risedronate, ibandronate, zoledronic acid, or raloxifene (the latter for women at low cardiovascular risk c ) for women who cannot tolerate or should not take alendronate. (Conditional recommendation) o There is no recommendation for or against the use of teriparatide (PTH) or denosumab in postmenopausal women with low bone mass. Evidence is insufficient to determine benefits and harms of therapy. Therapies to avoid o Do not prescribe hormone replacement therapy (estrogen or estrogen/progesterone) solely for the treatment of low bone mass. (Strong recommendation) o Do not prescribe nasal calcitonin for the treatment of low bone mass. (Strong recommendation) Clinical Considerations Calcium supplementation o Use calcium citrate in patients taking proton pump inhibitors (PPIs) or H2 blockers. Fall risk o Hip fracture risk increases with falling. The most common risk factors for falls include: A history of falls a For women, whose high FRAX score (> 3% for hip fracture and > 20% for major osteoporotic fracture) is due in part to long-term, daily use of corticosteroids (> 5 mg of prednisone or equivalent therapy), please refer to the recommendation for Treatment for Men and Women Taking Corticosteroid Therapy. b The most common risk factors for falls include: A history of falls Psychoactive medications (sedatives, antipsychotics, and antidepressants) anticonvulsants, or antihypertensive medications A high number of medications consumed (> 4), independent of medication indication Strength, gait and balance impairments Visual impairment Age > 80 years old c Those at increased risk include postmenopausal women with known coronary artery disease, peripheral vascular disease, or cerebrovascular disease; or a combination of diabetes with 1 additional risk factor (age > 65, current smoking, hypertension, hyperlipidemia), or a combination of all 4 of the listed risk factors together (Mosca 2001). 5

6 Psychoactive medications (sedatives, antipsychotics, and antidepressants) anticonvulsants, or antihypertensive medications o A high number of medications consumed (> 4), independent of medication indication. o Strength, gait, and balance impairments a o Visual impairment o Age > 80 years old Chronic kidney disease o Use bisphosphonates with caution in patients with chronic kidney disease and reduced glomerular filtration rate. Current drug monographs state that an estimated GFR < 35 ml/min is a contraindication to bisphosphonate use. Dental hygiene o Educate patients starting a bisphosphonate about the importance of regular dental cleanings and good dental hygiene. For those patients who have a planned tooth extraction or dental implant surgery, consider delaying the start of bisphosphonate therapy until 3 months after completion of the dental procedure, or until maxillofacial bone healing is complete. Both considerations are based on moderate evidence for association of osteonecrosis of the jaw (ONJ) with use (0.001% to 0.069% per year increased incidence over no bisphosphonate use). b Choice of alternative therapies o Consider significant side effects when choosing alternative therapies for patients who do not tolerate alendronate: Zoledronic acid: There is strong evidence for an acute phase reaction within 3 days of zoledronic acid administration (up to 25% increased risk over placebo of any of the following symptoms: pyrexia, myalgia, headache, arthralgia, chills). A 650-mg dose of acetaminophen initiated 45 minutes before zoledronic acid infusion and continuing every 6 hours for 3 days has been shown to reduce severity of symptoms. It is also common practice to ensure the patient is well hydrated prior to infusion. Raloxifene: There is strong evidence for increased hot flashes (2%-7% increase over placebo) and venous thromboembolic events (0.2%-0.7% increased risk) and death due to stroke (0.07% increased risk over placebo) as side effects of raloxifene. Osteoporosis (T-score -2.5 or prior fragility fracture): Pharmacologic Treatment Women: Preferred therapy o Optimize non-pharmacologic therapies, including vitamin D supplementation, calcium intake, and fall prevention efforts (via physical activity or therapy, ambulatory assistive devices, and reduction in sedative/hypnotic medications). (Strong recommendation) o Prescribe alendronate (70 mg/week) as a first-line therapy for postmenopausal women with osteoporosis (T-score -2.5 or a prior fragility fracture). (Strong recommendation) Alternative therapy o Prescribe alternative bisphosphonates, including ibandronate, risedronate, and zoledronic acid, for women who cannot tolerate or should not take alendronate. (Strong recommendation) Additional therapeutic options o Consult with specialty care about prescription of denosumab or teriparatide (PTH) for women who cannot tolerate or should not take bisphosphonates. (Strong recommendation) a This can be assessed with the timed Get-Up-and-Go test. The test is performed by observing the time it takes a person to rise from an armchair, walk 3 meters (10 feet), turn, walk back, and sit down again. The average healthy adult aged > 60 years can perform this task in less than 10 seconds. b Recommendations on interruption of bisphosphonate therapy can be found in the discontinuation section of this guideline. 6

7 o Consider prescribing raloxifene for women at low cardiovascular risk who cannot tolerate or should not take bisphosphonates. (Conditional recommendation) Therapies to avoid o Do not prescribe hormone replacement therapy (estrogen or estrogen/progesterone) solely for the treatment of osteoporosis. (Strong recommendation) o Do not prescribe nasal calcitonin for the treatment of osteoporosis. (Strong recommendation) Clinical Considerations Calcium supplementation o Use calcium citrate in patients taking proton pump inhibitors (PPIs) or H2 blockers. Chronic kidney disease o Use bisphosphonates with caution in patients with chronic kidney disease and reduced glomerular filtration rate. Current drug monographs state that an estimated GFR < 35 ml/min is a contraindication to bisphosphonate use. Dental hygiene o Educate patients starting a bisphosphonate about the importance of regular dental cleanings and good dental hygiene. For those patients who have a planned tooth extraction or dental implant surgery, consider delaying the start of bisphosphonate therapy until 3 months after completion of the dental procedure, or until maxillofacial bone healing is complete. Both considerations are based upon the moderate evidence for association of osteonecrosis of the jaw (ONJ) with bisphosphonate use (0.001% to 0.069% per year increased incidence over no bisphosphonate use). Choice of alternative therapies o Consider significant side effects when choosing alternative therapies for patients who do not tolerate alendronate: Zoledronic acid: There is strong evidence for an acute phase reaction within 3 days of zoledronic acid administration (up to 25% increased risk over placebo of any of the following symptoms: pyrexia, myalgia, headache, arthralgia, chills). A 650-mg dose of acetaminophen initiated 45 minutes before zoledronic acid infusion and continuing every 6 hours for 3 days has been shown to reduce severity of symptoms. It is also common practice to ensure the patient is well hydrated prior to infusion. Raloxifene: There is strong evidence for hot flashes (2%-7% increase over placebo), venous thromboembolic events (0.2%-0.7% increased risk), and death due to stroke (0.07% increased risk over placebo) as side effects of raloxifene. Men: Consider prescribing alendronate (70 mg/week) as a first-line therapy for men with osteoporosis (T-score -2.5 or a prior fragility fracture). (Conditional recommendation) Men and Women Taking Corticosteroid Therapy: Consider prescribing alendronate (70 mg/week) or risedronate (35 mg/week) as a first-line therapy for men and women who are taking oral corticosteroid medication at a dose of 5 mg/day prednisone or equivalent for 3 months and have a FRAX 10-year risk of hip fracture 3%. (Conditional recommendation) Consult with specialty care about prescription of teriparatide (PTH) for glucocorticoid-treated patients who cannot tolerate or should not take bisphosphonates. (Strong recommendation) 7

8 Discontinuation of Bisphosphonate Treatment In adults taking bisphosphonates for the treatment of osteoporosis or low bone mass, consider discontinuing therapy after 5 years of oral (alendronate, risedronate, ibandronate) or 3 years of intravenous (zoledronic acid) therapy. (Conditional recommendation) o Refer to Clinician Guide, Figures 1, 2, and 4 In adults taking bisphosphonate therapy for the treatment of osteoporosis or low bone mass, bisphosphonate therapy is generally not recommended after 10 years of continuous use of oral (alendronate, risedronate, ibandronate) or 6 years of continuous use of intravenous (zoledronic acid) therapy. (Conditional recommendation) Screening o Refer to Clinician Guide, Figures 3 and 4 Women: Postmenopausal Women o o For postmenopausal women aged 65 years who are not taking prescription antifracture medication, consider offering a bone mineral density (BMD) test by DXA. (Conditional recommendation) For postmenopausal women aged < 65 years with a 10-year fracture risk of 9.3%, consider offering a BMD test by DXA. (Conditional recommendation) Premenopausal Women o For premenopausal women, routine screening for osteoporosis is generally not recommended. (Conditional recommendation) Men: For men aged 70 years and/or with previous fragility fracture, consider offering screening for osteoporosis with a bone mineral density (BMD) test by DXA. (Conditional recommendation) Bone Mineral Density Testing with Dual Energy X-ray Absorptiometry (DXA): Measurement Sites o Total proximal femur (total hip), femoral neck, and lumbar spine Alternative Measurement Sites o Forearm (distal one-third of the radius) is acceptable for patients in whom hip and spine BMD cannot be measured or interpreted. Interpretation o The lowest T-score from the measurements of the total hip, femoral neck, and lumbar spine (L1 to L4, composite score) is recommended to establish a diagnosis of osteoporosis (T-score -2.5) or low BMD (T-score -1.1 to -2.4). DXA Screening Intervals: For women aged 65 years who are not taking prescription antifracture medication and who have had a baseline BMD test, future rescreening for low BMD with DXA is an option. (Conditional recommendation) o If DXA testing is obtained, suggested rescreening intervals based on initial T- score (lowest T- score from total hip, femoral neck, or lumbar spine). Evidence is insufficient to recommend a rescreening interval for women aged < 65 years or men. Clinical Considerations Clinicians should assess the patient's willingness to initiate treatment before deciding to rescreen. 8

9 Clinicians should consider calculating a current FRAX score using the patient's most recent T- score to make rescreening and treatment decisions. If DXA testing is obtained, suggested rescreening intervals based on initial T-score (lowest T-score from total hip, femoral neck, or lumbar spine) are as follows: Initial T-Score Suggested Minimum Interval years -1.5 to years -2.0 to years 9

10 Appendix A: National Osteoporosis/Fracture Prevention GDT Lead Team Shireen Fatemi, MD, Clinical Lead, Endocrinology, KP Southern California Elizabeth Liles, MD, Methodologist, Internal Medicine, KP Northwest Deborah Regidor, PhD, Principal Consultant, CMI Evidence Services, KP Program Office Guideline Development Team Colorado Katherine Weber, MD, Endocrinology, KP Colorado Rachel Heilmann, PharmD, Pharmacy, KP Colorado Georgia Elham Syed, MD, Internal Medicine, KP GA Negah Rassouli, MD, Endocrinology, KP GA Hawaii Jennifer Loh, MD, Endocrinology, KP Hawaii Northern California/Mid-Atlantic States Jean Kayser, MD, OB/GYN, KP Northern California Kendal Hamann, MD, Endocrinology, KP Northern California Ma Clara Padero, MD, Endocrinology, KP Northern California Northwest Alistair Bahar, MD, Endocrinology, KP Northwest Preston Peterson, MD, Geriatrics, KP Northwest Southern California Marguerite Koster, Practice Leader, EBM, KP Southern California Nora Strick, MD, Internal Medicine, KP Southern California Jaime Natoli, Senior Consultant, EBM, KP Southern California Washington Judy Lee, MD, Endocrinology, KP Washington Drug Information Services Natalie Aboubechara, PharmD, Pharmacy Conflicts of Interest The NGP manages conflicts of interest in accordance with the Guidelines International Network Principles for Disclosure of Interests and Management of Conflicts in Guidelines (Schünemann HJ, Al-Ansary LA, Forland F, Kersten S, Komulainen J, Kopp IB, et al. Guidelines International Network: Principles for Disclosure of Interests and Management of Conflicts in Guidelines. Ann Intern Med. 2015; 163: ). GDT members have the following disclosures: None. 10

11 Appendix B1. Summary of Recommendation Decisions 2017 KP Recommendation 2015 KP Recommendation a Prevention and Treatment of Osteoporosis All adults: Non-pharmacologic primary prevention of osteoporosis Lifestyle choices KP GDT Decision Justification For all adults, consider recommending regular weightbearing and muscle-building exercise for prevention of osteoporosis and falls. (Conditional recommendation) Lifestyle changes are recommended for all adults: Exercise regular weight-bearing and musclebuilding exercise (Consensus-based) Smoking cessation (Consensus-based) Update recommendation language to current NGP terminology Approved by GDT with no additional changes needed. For all adults who are current smokers, consider recommending smoking cessation for prevention of osteoporosis and fragility fractures. (Conditional recommendation) Fall prevention For adults at increased risk of falling, consider safety proofing their residence. (Conditional recommendation) Home safety proofing includes removing rugs, adding grab bars, establishing adequate lighting (eg, nightlights), and securing electrical cord placement. Safety Proofing is recommended for postmenopausal women and men at risk of falling. (Consensus-based) NOTE: Home safety proofing includes removing rugs, adding grab bars, establishing adequate lighting (eg, nightlights), and securing electrical cord placement. Update recommendation language to current NGP terminology Approved by GDT with no additional changes needed. a Various recommendation rating and evidence grading systems are used; see Appendix B2 for a crosswalk with KP recommendation ratings. [if applicable] 11

12 2017 KP Recommendation 2015 KP Recommendation a KP GDT Decision Justification For postmenopausal women and men aged 50 years, the routine use of hip protectors to reduce the risk of hip fractures is not recommended. (Strong recommendation) The routine use of hip protectors is not recommended as an intervention for reducing the risk of hip fractures in postmenopausal women and men aged 50 years or older (Evidence-based: D) Update recommendation language to current NGP terminology Approved by GDT with no additional changes needed. Supplementation For all adults, advise patients to consume the recommended total daily intake* of calcium and vitamin D. (Good practice statement) Consider prescribing supplemental calcium and vitamin D if recommended daily intake is not achieved through diet alone. (Conditional recommendation) *USDA recommended daily allowance (RDA) a : Calcium Women: Aged years: 1,000 mg Aged 51 years: 1,200 mg Men: Aged years: 1,000 mg Aged > 70 years: 1,200 mg Total daily intake of calcium is recommended for all pre- or postmenopausal women and older men (1,000 mg/day for premenopausal women; 1,200 mg/day for postmenopausal women and men aged 50 years or older). Many individuals require supplemental calcium therapy. (Evidence-based: B) Update recommendation language to current NGP terminology Update requirements to current USDA recommended RDA Approved by GDT with no additional changes needed. Vitamin D All adults: Aged years: 600 IU Aged > 70 years: 800 IU Clinical considerations If supplementation is necessary, vitamin D3 (ergocalciferol) is preferred NOTE: Calcium carbonate contains the most elemental calcium per dose. It should be taken with food to enhance absorption. Calcium citrate contains less elemental calcium than the carbonate salt, but it is better absorbed and may be preferred a 12

13 2017 KP Recommendation 2015 KP Recommendation a KP GDT Decision Justification Calcium carbonate contains the most elemental calcium per dose. It should be taken with food to enhance absorption. Calcium citrate contains less elemental calcium than the carbonate salt, but it is better absorbed and may be preferred in patients with reduced gastric acid production or high gastric ph requiring long term H2 antagonist or proton pump inhibitor therapy and in patients who have undergone bariatric surgery. It is more expensive and usually requires more tablets to be taken per day than calcium carbonate. in patients with reduced gastric acid production or high gastric ph requiring long term H2 antagonist or proton pump inhibitor therapy and in patients who have undergone bariatric surgery. It is more expensive and usually requires more tablets to be taken per day than calcium carbonate. Remove Total daily intake of vitamin D (at least 1,000 IU/day), preferably vitamin D3, is recommended for all pre- or postmenopausal women and men aged 50 or older. (Consensus-based) Combine Calcium and Vitamin D into one recommendation In adults aged 50 years without osteoporosis, do not screen for Vitamin D deficiency (Strong recommendation) Screening for vitamin D deficiency is not recommended for identifying vitamin D deficiency in adults aged 50 years or older without osteoporosis. (Consensus-based) Update recommendation language to current NGP terminology Approved by GDT with no additional changes needed. Low Bone Mass (T-score -1.1 to -2.4): Pharmacologic Treatment Consider treating women with low bone mass (and no history of prior fragility fracture) who have at least one of the following additional risk factors: A FRAX 10-year risk of hip fracture 3% or a FRAX 10-year risk of major osteoporotic fracture 20% a Consider treating women with low bone mass (and no history of prior fragility fracture) who have at least one of the following additional risk factors: No changes from 2015 KP guideline See Appendix C a For women whose high FRAX score ( 3% for hip fracture and 20% for major osteoporotic fracture) is due in part to long-term, daily use of corticosteroids ( 5 mg of prednisone or equivalent therapy), please refer to the recommendation for Treatment for Men and Women Taking Corticosteroid Therapy. 13

14 2017 KP Recommendation 2015 KP Recommendation a KP GDT Decision Justification Increased risk of falling a (Conditional recommendation) Preferred therapy Optimize nonpharmacologic therapies, including vitamin D supplementation, calcium intake, and fall prevention efforts (via physical activity or therapy, ambulatory assistive devices, and reduction in sedative/hypnotic medications). (Strong recommendation) Consider prescribing alendronate (70 mg/week) as a firstline therapy, particularly in those with a lowest T-score in the range of -2.1 to -2.4 and at least one of the following additional risk factors: A FRAX 10-year risk of hip fracture 3% or a FRAX 10-year risk of major osteoporotic fracture 20% b Increased risk of falling (Conditional recommendation) Alternative therapy Consider prescribing risedronate, ibandronate, zoledronic acid, or raloxifene (the latter for women at low cardiovascular risk c ) for women who cannot tolerate or should not take alendronate. A FRAX 10-year risk of hip fracture 3% or a FRAX 10-year risk of major osteoporotic fracture 20% Increased risk of falling (Weak recommendation) Preferred therapy Optimize nonpharmacologic therapies, including vitamin D supplementation, calcium intake, and fall prevention efforts (via physical activity or therapy, ambulatory assistive devices, and reduction in sedative/hypnotic medications). (Strong recommendation) Consider prescribing alendronate (70 mg/week) as a first-line therapy, particularly in those with a lowest T-score in the range of -2.1 to -2.4 and at least one of the following additional risk factors: o A FRAX 10-year risk of hip fracture 3% or a FRAX 10-year risk of major osteoporotic fracture 20% o Increased risk of falling (Weak recommendation) Alternative therapy Consider prescribing risedronate, ibandronate, zoledronic acid, or raloxifene (the latter for women a The most common risk factors for falls include: A history of falls Psychoactive medications (sedatives, antipsychotics, and antidepressants) anticonvulsants, or antihypertensive medications A high number of medications consumed (> 4), independent of medication indication. Strength, gait and balance impairments Visual impairment Age > 80 years b For women whose high FRAX score ( 3% for hip fracture and 20% for major osteoporotic fracture) is due in part to long-term, daily use of corticosteroids ( 5 mg of prednisone or equivalent therapy), please refer to the recommendation for Treatment for Men and Women Taking Corticosteroid Therapy. c Those at increased risk include postmenopausal women with known coronary artery disease, peripheral vascular disease, or cerebrovascular disease; or a combination of diabetes with 1 additional risk factor (age > 65, current smoking, hypertension, hyperlipidemia), or a combination of all 4 of the listed risk factors together (Mosca 2001). 14

15 2017 KP Recommendation 2015 KP Recommendation a KP GDT Decision Justification (Conditional recommendation) There is no recommendation for or against the use of teriparatide (PTH) or denosumab in postmenopausal women with low bone mass. Evidence is insufficient to determine benefits and harms of therapy. Therapies to avoid Do not prescribe hormone replacement therapy (estrogen or estrogen/progesterone) solely for the treatment of low bone mass. (Strong recommendation) Do not prescribe nasal calcitonin for the treatment of low bone mass. (Strong recommendation) Clinical considerations Calcium supplementation Use calcium citrate in patients taking proton pump inhibitors (PPIs) or H2 blockers. Fall risk Hip fracture risk increases with falling. The most common risk factors for falls include: o A history of falls o Psychoactive medications (sedatives, antipsychotics and antidepressants) anticonvulsants, or antihypertensive medications o A high number of medications consumed (> 4), independent of medication indication. o Strength, gait and balance impairments a o Visual impairment o Age > 80 years with a low cardiovascular risk) for women who cannot tolerate or should not take alendronate. (Weak recommendation) Therapies to avoid Do not prescribe hormone replacement therapy (estrogen or estrogen/progesterone) solely for the treatment of low bone mass. (Strong recommendation) Do not prescribe nasal calcitonin for the treatment of low bone mass. (Strong recommendation) Clinical Considerations Calcium supplementation Use calcium citrate in patients taking proton pump inhibitors (PPIs) or H2 blockers. Fall risk Hip fracture risk increases with falling. Therefore, in this update we have added language to increase awareness of overall fall risk in the context of fracture prevention. The most common risk factors for falls include: o o o A history of falls Psychoactive medications (sedatives, antipsychotics and antidepressants) anticonvulsants, or antihypertensive medications A high number of medications consumed (> 4), independent of medication indication. a This can be assessed with the timed Get-Up-and-Go test. The test is performed by observing the time it takes a person to rise from an armchair, walk 3 meters (10 feet), turn, walk back, and sit down again. The average healthy adult aged > 60 years can perform this task in less than 10 seconds. 15

16 2017 KP Recommendation 2015 KP Recommendation a KP GDT Decision Justification o o o Strength, gait and balance impairments b Visual impairment Age > 80 years old Chronic kidney disease Use bisphosphonates with caution in patients with chronic kidney disease and reduced glomerular filtration rate. Current drug monographs state that an estimated GFR < 35 ml/min is a contraindication to bisphosphonate use. Dental hygiene Educate patients starting a bisphosphonate about the importance of regular dental cleanings and good dental hygiene. For those patients who have a planned tooth extraction or dental implant surgery, consider delaying the start of bisphosphonate therapy until 3 months after completion of the dental procedure, or until maxillofacial bone healing is complete. Both considerations are based on moderate evidence for association of osteonecrosis of the jaw (ONJ) with bisphosphonate use (0.001% to 0.069% per year increased incidence over no bisphosphonate use). a Choice of alternative therapies Consider significant side effects when choosing alternative therapies for patients who do not tolerate alendronate: o Zoledronic acid: There is strong evidence for an acute phase reaction within 3 days of zoledronic acid administration (up to 25% increased risk over placebo of any of the following symptoms: pyrexia, myalgia, Chronic kidney disease Use bisphosphonates with caution in patients with chronic kidney disease and reduced glomerular filtration rate. Current drug monographs state that an estimated GFR < 35 ml/min is a contraindication to bisphosphonate use. Dental hygiene Educate patients starting a bisphosphonate about the importance of regular dental cleanings and good dental hygiene. For those patients who have a planned tooth extraction or dental implant surgery, consider delaying the start of bisphosphonate therapy until 3 months after completion of the dental procedure, or until maxillofacial bone healing is complete. Both considerations are based on moderate evidence for association of osteonecrosis of the jaw (ONJ) with bisphosphonate use (0.001% to 0.069% per year increased incidence over no bisphosphonate use). Choice of alternative therapies Consider significant side effects when choosing alternative therapies for patients who do not tolerate alendronate: o Zoledronic acid: There is strong evidence for an acute phase reaction within 3 days of zoledronic acid administration (up to 25% increased risk over placebo of any of the following symptoms: pyrexia, myalgia, a Recommendations on interruption of bisphosphonate therapy can be found in the discontinuation section of this guideline. b This can be assessed with the timed Get-Up-and-Go test. The test is performed by observing the time it takes a person to rise from an armchair, walk 3 meters (10 feet), turn, walk back, and sit down again. The average healthy adult aged > 60 years can perform this task in less than 10 seconds. 16

17 2017 KP Recommendation 2015 KP Recommendation a KP GDT Decision Justification o headache, arthralgia, chills). A 650 mg dose of acetaminophen initiated 45 minutes before zoledronic acid infusion and continuing every 6 hours for 3 days has been shown to reduce severity of symptoms. 1 It is also common practice to ensure the patient is well hydrated prior to infusion. Raloxifene: There is strong evidence for increased hot flashes (2-7% increase over placebo) and venous thromboembolic events ( % increased risk) and death due to stroke (0.07% increased risk over placebo) as side effects of raloxifene. o headache, arthralgia, chills). A 650 mg dose of acetaminophen initiated 45 minutes before zoledronic acid infusion and continuing every 6 hours for 3 days has been shown to reduce severity of symptoms. It is also common practice to ensure the patient is well hydrated prior to infusion. Raloxifene: There is strong evidence for hot flashes (2%-7% increase over placebo) and venous thromboembolic events (0.2%- 0.7% increased risk) and death due to stroke (0.07% increased risk over placebo) as side effects of raloxifene. Treatment of Osteoporosis (T score -2.5 or Prior Fragility Fracture) Women Preferred therapy Optimize nonpharmacologic therapies, including vitamin D supplementation, calcium intake, and fall prevention efforts (via physical activity or therapy, ambulatory assistive devices, and reduction in sedative/hypnotic medications). (Strong recommendation) Prescribe alendronate (70 mg/week) as a first-line therapy for post-menopausal women with osteoporosis (T-score < or a prior fragility fracture) (Strong recommendation) Preferred therapy Optimize no-pharmacologic therapies, including vitamin D supplementation, calcium intake, and fall prevention efforts (via physical activity or therapy, ambulatory assistive devices, and reduction in sedative/hypnotic medications). (Strong recommendation) Prescribe alendronate (70 mg/week) as a first-line therapy for post-menopausal women with osteoporosis (T-score -2.5 or a prior fragility fracture). (Strong recommendation) No changes from 2015 KP guideline See Appendix C Alternative therapy Prescribe alternative bisphosphonates, including ibandronate, risedronate, and zoledronic acid, for women who cannot tolerate or should not take alendronate. (Strong recommendation) Alternative therapy Prescribe alternative bisphosphonates, including ibandronate, risedronate, and zoledronic acid, for 17

18 2017 KP Recommendation 2015 KP Recommendation a KP GDT Decision Justification Additional therapeutic options Consult with specialty care about prescription of denosumab, or teriparatide (PTH) for women who cannot tolerate or should not take bisphosphonates. (Strong recommendation) Consider prescribing raloxifene for women with a low cardiovascular risk who cannot tolerate or should not take bisphosphonates. (Conditional recommendation) women who cannot tolerate or should not take alendronate. (Strong recommendation) Additional therapeutic options Consult with specialty care about prescription of denosumab or teriparatide (PTH) for women who cannot tolerate or should not take bisphosphonates. (Strong recommendation) Consider prescribing raloxifene for women with a low cardiovascular risk who cannot tolerate or should not take bisphosphonates. (Weak recommendation) Therapies to avoid Do not prescribe hormone replacement therapy (estrogen or estrogen/progesterone) solely for the treatment of osteoporosis (Strong recommendation) Do not prescribe nasal calcitonin for the treatment of osteoporosis (Strong recommendation) Therapies to avoid Do not prescribe hormone replacement therapy (estrogen or estrogen/progesterone) solely for the treatment of osteoporosis. (Strong recommendation) Do not prescribe nasal calcitonin for the treatment of osteoporosis (Strong recommendation) Clinical Considerations Calcium supplementation Use calcium citrate in patients taking proton pump inhibitors (PPIs) or H2 blockers. Chronic kidney disease Use bisphosphonates with caution in patients with chronic kidney disease and reduced glomerular filtration rate. Current drug monographs state that an estimated GFR < 35 ml/min is a contraindication to bisphosphonate use. Dental hygiene Clinical Considerations Calcium supplementation Use calcium citrate in patients taking proton pump inhibitors (PPIs) or H2 blockers. Chronic kidney disease Use bisphosphonates with caution in patients with chronic kidney disease and reduced glomerular filtration rate. Current drug monographs state that 18

19 2017 KP Recommendation 2015 KP Recommendation a KP GDT Decision Justification Educate patients starting a bisphosphonate about the importance of regular dental cleanings and good dental hygiene. For those patients who have a planned tooth extraction or dental implant surgery, consider delaying the start of bisphosphonate therapy until 3 months after completion of the dental procedure, or until maxillofacial bone healing is complete. Both considerations are based on moderate evidence for association of osteonecrosis of the jaw (ONJ) with bisphosphonate use (0.001% to 0.069% per year increased incidence over no bisphosphonate use). a Choice of alternative therapies Consider significant side effects when choosing alternative therapies for patients who do not tolerate alendronate: o Zoledronic acid: There is strong evidence for an acute phase reaction within 3 days of zoledronic acid administration (up to 25% increased risk over placebo of any of the following symptoms: pyrexia, myalgia, headache, arthralgia, chills). A 650 mg dose of acetaminophen initiated 45 minutes before zoledronic acid infusion and continuing every 6 hours for 3 days has been shown to reduce severity of symptoms. 1 It is also common practice to ensure the patient is well hydrated prior to infusion. o Raloxifene: There is strong evidence for hot flashes (2-7% increase over placebo) and venous thromboembolic events ( % increased risk) and death due to stroke (0.07% increased risk over placebo) as side effects of raloxifene. an estimated GFR < 35 ml/min is a contraindication to bisphosphonate use. Dental hygiene Educate patients starting a bisphosphonate about the importance of regular dental cleanings and good dental hygiene. For those patients who have a planned tooth extraction or dental implant surgery, consider delaying the start of bisphosphonate therapy until 3 months after completion of the dental procedure, or until maxillofacial bone healing is complete. Both considerations are based on moderate evidence for association of osteonecrosis of the jaw (ONJ) with bisphosphonate use (0.001% to 0.069% per year increased incidence over no bisphosphonate use). Choice of alternative therapies Consider significant side effects when choosing alternative therapies for patients who do not tolerate alendronate: o Zoledronic acid: There is strong evidence for an acute phase reaction within 3 days of zoledronic acid administration (up to 25% increased risk over placebo of any of the following symptoms: pyrexia, myalgia, headache, arthralgia, chills). A 650 mg dose of acetaminophen initiated 45 minutes before zoledronic acid infusion and continuing every 6 hours for 3 days has been shown to reduce severity of symptoms. It is also common practice to ensure the patient is well hydrated prior to infusion. o Raloxifene: There is strong evidence for hot flashes (2%-7% increase over placebo) and venous thromboembolic events (0.2%- a Recommendations on interruption of bisphosphonate therapy can be found in the discontinuation section of this guideline. 19

20 2017 KP Recommendation 2015 KP Recommendation a 0.7% increased risk) and death due to stroke (0.07% increased risk over placebo) as side effects of raloxifene. KP GDT Decision Justification Men Consider prescribing alendronate (70 mg/week) as a firstline therapy for men with osteoporosis (T-score -2.5 or a prior fragility fracture) (Conditional recommendation) Alendronate (70 mg/week) is recommended as a first-line therapy for men aged 70 or older diagnosed with osteoporosis or with a FRAX 10- year risk of hip fracture 3%. (Consensus-based) Update recommendation language to current NGP terminology Approved by GDT with no additional changes needed. Align language with recommendation for PMO women Remove from guideline Pharmacologic treatment for osteoporosis is optional in men under the age of 70 who are diagnosed with osteoporosis (T-score -2.5) but without a FRAX 10-year risk of hip fracture 3%. (Consensus-based) Not necessary to stratify recommendation by age Men and Women Taking Corticosteroid Therapy Consider prescribing alendronate (70 mg/week) or risedronate (35 mg/week) as a first-line therapy for men and women who are taking oral corticosteroid medication at a dose of 5 mg/day prednisone or equivalent for a duration of 3 months and have a FRAX 10-year risk of hip fracture 3%. (Conditional recommendation) Bisphosphonates Alendronate (70 mg/week) or risedronate (35 mg/week) is recommended as first-line therapy for men and women who are taking oral corticosteroid medication at a dose of 5 mg/day prednisone or equivalent for a duration of 3 months or more and have a FRAX 10-year risk of hip fracture 3%. (Consensus-based) Update recommendation language to current NGP terminology Approved by GDT with no additional changes needed. 20

21 2017 KP Recommendation 2015 KP Recommendation a KP GDT Decision Justification Consult with specialty care about prescription of teriparatide (PTH) for glucocorticoid-treated patients who cannot tolerate or should not take bisphosphonates. (Strong recommendation) Teriparatide Teriparatide (recombinant PTH) by daily injection is an anabolic agent that is an option for treating osteoporosis in glucocorticoid-treated patients not tolerant of or responsive to other agents. It should be used only after specialist evaluation. (Evidencebased: B) Update recommendation language to current NGP terminology Approved by GDT with no additional changes needed. Monitoring Treatment Bone Mineral Density Testing with DXA Removed from guideline Routine BMD testing by DXA is not recommended for monitoring the rate of bone loss after initiation of treatment in women or men. (Consensus-based) NOTE: A major determinant of fracture risk reduction with bisphosphonate therapy is continuing to take the therapy. Monitoring during treatment is integrated into Clinician Guide algorithms Bone Turnover Testing Removed from guideline There is no recommendation for or against routine bone turnover testing with biochemical markers for monitoring women and men taking antiresorptive therapy for osteoporosis. (Evidence-based: I) Monitoring during treatment is integrated into Clinician Guide algorithms Discontinuation of Bisphosphonate Treatment 21

22 2017 KP Recommendation 2015 KP Recommendation a KP GDT Decision Justification In adults taking bisphosphonates for the treatment of osteoporosis or low bone mass, consider discontinuing therapy after 5 years of oral (alendronate, risedronate, ibandronate) or 3 years of intravenous (zoledronic acid) therapy. (Conditional recommendation) o Refer to Clinician Guide, Figures 1, 2 and 4 In adults taking bisphosphonate therapy for the treatment of osteoporosis or low bone mass, bisphosphonate therapy is generally not recommended after 10 years of continuous use of oral (alendronate, risedronate, ibandronate) or 6 years of continuous use of intravenous (zoledronic acid) therapy. (Conditional recommendation) o Refer to Clinician Guide, Figures 3 and 4 Bisphosphonate therapy is generally not recommended after 10 years of continuous use. (Weak recommendation) There is insufficient evidence to recommend for or against discontinuation of bisphosphonates after 5 to < 10 years of therapy. Clinical Considerations For patients who have taken bisphosphonates for 5 years, the decision to continue or discontinue treatment may take the following issues into account: For patients at lower risk of fragility fracture, the risk of rare but serious atypical femur fractures may outweigh the relatively small additional reduction in fragility fracture risk conferred by continuing treatment. The primary factor that increases the risk for atypical femur fracture is duration of bisphosphonate use. o The risk of atypical femur fracture ranges from ~1 per 50,000 for < 2 years of treatment to ~1 per 1,000 for 8-10 years of treatment. For patients at higher risk of fragility fracture, the benefit of reduced fragility fracture risk may outweigh the potential harms. Factors that may increase the risk of fragility fracture include the following: o o o o Prior history of fragility fracture Risk for falling Age (increased risk with increased age, especially > 75 years) Low pre- and post-treatment bone density Major changes to recommendation Separate recommendation for drug holiday and discontinuation at maximum cumulative treatment duration Removal of clinical consideration and transferred content into algorithms published in Clinician Guide See Appendix C 22

23 2017 KP Recommendation 2015 KP Recommendation a Screening for Osteoporosis o High-risk pre-treatment FRAX score Based on one small study (the FLEX trial), continuous use of alendronate for 10 years reduced the risk for symptomatic vertebral fracture from ~1 in 20 to ~1 in 40, compared to stopping after 5 years of treatment. If bisphosphonate treatment is discontinued, there is little evidence to guide the length of discontinuation, including if and when treatment should resume. o o o The estimated half-life of bisphosphonates is measured in years and is approximately 10 years for alendronate. In the FLEX trial, the relative risk for clinical vertebral fracture began to increase about 2-3 years after alendronate was discontinued. Patients with secondary causes of osteoporosis (e.g., rheumatoid arthritis, long-term corticosteroid use) may require additional discussion with their respective specialists. KP GDT Decision Justification Women 23

24 2017 KP Recommendation 2015 KP Recommendation a KP GDT Decision Justification For postmenopausal women aged 65 years who are not taking prescription antifracture medication, consider offering a bone mineral density (BMD) test by DXA (Conditional recommendation) Post-menopausal Women Age 65 years A bone mineral density (BMD) test by DXA is recommended for postmenopausal women aged 65 or older who are not on drug treatment for osteoporosis. (Evidence-based: B) Update recommendation language to current NGP terminology Approved by GDT with no additional changes needed. For postmenopausal women aged < 65 years with a 10- year fracture risk of > 9.3%, consider offering a BMD test by DXA. (Conditional recommendation) Age < 65 years For postmenopausal women under age 65, a BMD test by DXA is an option when selected risk factors* are present. (Consensus-based) * In addition to advancing age and female sex, risk factors in the FRAX* model include low body mass index (BMI), personal history of fragility fracture after age 50, parental history of hip fracture, rheumatoid arthritis, long-term exposure to systemic corticosteroids (3 months or more at doses 5 mg), high alcohol intake (about 3 ounces per day), cigarette smoking, and other causes of secondary osteoporosis (e.g., type 1 diabetes, osteogenesis imperfecta in adults, untreated long-standing hyperthyroidism, hypogonadism or premature menopause (< 45 years), chronic malnutrition, or malabsorption and chronic liver disease). It is useful to review the FRAX model and risk factors with patients. Update recommendation language to current NGP terminology. Premenopausal women For premenopausal women, routine screening for osteoporosis is generally not recommended (Conditional recommendation) Pre-menopausal Women Routine screening for osteoporosis with a BMD test by DXA is not recommended for premenopausal women. (Consensus-based) Update recommendation language to Approved by GDT with no additional changes needed. 24

25 2017 KP Recommendation 2015 KP Recommendation a KP GDT Decision current NGP terminology Justification Men For men aged 70 years and/or with previous fragility fracture, consider offering screening for osteoporosis with a bone mineral density (BMD) test by DXA. (Conditional recommendation) Screening with DXA is an option for men aged 70 or older with risk factors. (Consensus-based) Update recommendation language to current NGP terminology Approved by GDT with no additional changes needed. Bone Mineral Density Testing with Dual Energy X-ray Absorptiometry (DXA) Measurement sites: Total proximal femur (total hip), femoral neck, and lumbar spine Alternative measurement sites: Forearm (distal one-third of the radius) is acceptable for patients in whom hip and spine BMD cannot be measured or interpreted. Interpretation: The lowest T-score from the measurements of the total hip, femoral neck, and lumbar spine (L1 to L4, composite score) is recommended to establish a diagnosis of osteoporosis (T-score -2.5) or low BMD (T-score -1.1 to -2.4). Screening Intervals Measurement sites When BMD testing is indicated, the total proximal femur (total hip), femoral neck, and lumbar spine are recommended measurement sites for DXA to predict risk of osteoporotic fracture in women and men. (Evidence-based: B) Alternative Measurement Sites DXA of the forearm (distal one-third of the radius) is an option for patients in whom hip and spine BMD cannot be measured or interpreted. (Evidencebased: B) T-score Interpretation The lowest T-score from the measurements of the total hip, femoral neck, and lumbar spine (L1 to L4, composite score) is recommended to establish a diagnosis of osteoporosis (T-score -2.5) or low BMD (T-score -1.1 to -2.4). (Consensus-based) Not a recommendation: Ungraded explanatory information re: DXA testing 25

26 2017 KP Recommendation 2015 KP Recommendation a KP GDT Decision Justification For women aged 65 years *who are not taking prescription antifracture medication and who have had a baseline BMD test, future rescreening for low BMD with DXA is an option. (Conditional recommendation) Evidence is insufficient to recommend a rescreening interval for women aged < 65 years or men. Post-menopausal Women For women aged 65 years or older who are not taking prescription antifracture medication (and who have had a baseline BMD test), future rescreening for low BMD with DXA is an option. (Weak recommendation) Clinical considerations: Clinicians should assess the patient's willingness to initiate treatment before deciding to rescreen. Clinicians should consider calculating a current FRAX score using the patient's most recent T-score to make rescreening and treatment decisions. If DXA testing is obtained, suggested rescreening intervals based on initial T-score (lowest T-score from total hip, femoral neck, or lumbar spine) are as follows: Clinical Considerations Clinicians should assess the patient's willingness to initiate treatment before deciding to rescreen. Clinicians should consider calculating a current FRAX score using the patient's most recent T- score to make rescreening and treatment decisions. If DXA testing is obtained, suggested rescreening intervals based on initial T-score (lowest T-score from total hip, femoral neck, or lumbar spine) are as follows: Minor changes from 2015 added missing footnote concerning women < 65 and men Approved by GDT with no additional changes needed. 26

27 Appendix B2. Crosswalk of Recommendation Ratings Grade Kaiser Permanente Current Grading System Intended action Strong affirmative Provide the intervention. Most individuals should receive the intervention; only a small proportion will not want the intervention. Recommendation language: start, initiate, prescribe, treat, provide, offer, evaluate Strong Strong negative Do not provide the intervention. Most individuals should not receive the intervention; only a small proportion will want the intervention. Recommendation language: do not start, initiate, prescribe treat, provide, offer, evaluate Conditional Conditional affirmative Assist each patient in making a management decision consistent with personal values and preferences. The majority of individuals in this situation will want the intervention, but many will not. Different choices will be appropriate for different patients. Recommendation language: consider starting, initiating, prescribing, treating, providing, offering, evaluating Conditional negative Assist each patient in making a management decision consistent with personal values and preferences. The majority of individuals in this situation will not want the intervention, but many will. Different choices will be appropriate for different patients. Recommendation language: consider stopping, consider not starting, initiating, prescribing, treating, providing, offering, evaluating No recommendation for or against Given that the balance between desirable and undesirable effects, the evidence quality, the values & preferences, and the resource allocation implications of an intervention do not drive a recommendation in one particular direction, recommendations will be made at the discretion of the individual clinician. Recommendation language: No recommendation for or against 27

28 Kaiser Permanente Old Grading System: Common Methodology Translation options Grade Description Evidence-based A Evidence-based B Evidence-based C Evidence-based D Evidence-based I Consensus-based The intervention improves important health outcomes, based on good evidence Benefits substantially outweigh harms The intervention improves important health outcomes, based on either: 1) Good evidence that benefits outweigh harms and costs 2) Fair evidence that benefits substantially outweigh harms and costs Evidence is sufficient to determine the benefits, harms, and costs of an intervention, and there is at least fair evidence that the intervention improves important health outcomes. Balance of benefits and harms too close to justify a recommendation At least fair evidence that the intervention is ineffective or harms outweigh benefits Evidence is lacking, of poor quality or conflicting and the balance of benefits, harms and costs cannot be determined Recommendation based on consensus, typically in the setting of insufficient evidence Strong affirmative Strong affirmative Conditional affirmative Conditional affirmative No rec for or against Conditional negative Strong negative No rec for or against Conditional affirmative/negative No rec for or against Conditional affirmative/negative 28

29 Appendix C. Rationales Recommendations using the GRADE framework are supported by a rationale adapted from the GRADE Evidence to Decision framework. The justification for recommendations created prior to the use of GRADE are documented in Appendix D. Low bone mass (T-score -1.1 to -2.4): Pharmacologic treatment Preferred therapy Optimize non-pharmacologic therapies, including vitamin D supplementation (1,000 IU daily), calcium intake (total of 1,200 mg daily) and fall prevention efforts (via physical activity or therapy, ambulatory assistive devices, and reduction in sedative/hypnotic medications). (Strong recommendation) Consider prescribing alendronate (70 mg/week) as a first-line therapy, particularly in those with a lowest T score in the range of -2.1 to -2.4 and additional risk factors as outlined above. (Conditional recommendation) Alternative therapy Consider prescribing risedronate, ibandronate, zoledronic acid or raloxifene (the latter for women with a low cardiovascular risk a ) for women who cannot tolerate or should not take alendronate. (Conditional recommendation) Recommendation There is no recommendation for or against the use of teriparatide (PTH) or denosumab in postmenopausal women with low bone mass. Evidence is insufficient to determine benefits and harms of therapy. Therapies to Avoid Do not prescribe hormone replacement therapy (estrogen or estrogen/progesterone) solely for the treatment of low bone mass. (Strong recommendation) Do not prescribe nasal calcitonin for the treatment of low bone mass. (Strong recommendation) Clinical considerations Fall risk Hip fracture risk increases with falling. Therefore, in this update we have added language to increase awareness of overall fall risk in the context of fracture prevention. The most common risk factors for falls include: a Those at increased risk include postmenopausal women with known coronary artery disease, peripheral vascular disease, or cerebrovascular disease; or a combination of diabetes with 1 additional risk factor (age >65, current smoking, hypertension, hyperlipidemia), or a combination of all 4 of the listed risk factors together (Mosca 2001). 29

30 o A history of falls o Psychoactive medications (sedatives, antipsychotics and antidepressants) anticonvulsants, or antihypertensive medications o A high number of medications consumed (>4), independent of medication indication. o Strength, gait and balance impairments a o Visual impairment o Age >80 years old Chronic kidney disease Use bisphosphonates with caution in patients with chronic kidney disease and reduced glomerular filtration rate. Current drug monographs state that an estimated GFR <35 ml/min is a contraindication to bisphosphonate use. Dental hygiene Educate patients starting a bisphosphonate about the importance of regular dental cleanings and good dental hygiene. For those patients who have a planned tooth extraction or dental implant surgery, consider delaying the start of bisphosphonate therapy until 3 months after completion of the dental procedure, or until maxillofacial bone healing is complete. Both of these considerations are based upon the moderate evidence for association of osteonecrosis of the jaw (ONJ) with bisphosphonate use (0.001% to 0.069% per year increase over nonbisphosphonate users). b Choice of alternative therapies Consider significant side effects when choosing alternative therapies for patients who do not tolerate alendronate: o Zoledronic acid: There is strong evidence for an acute phase reaction within 3 days of zoledronic acid administration (up to 25% increased risk over placebo of any of the following symptoms: pyrexia, myalgia, headache, arthralgia, chills). A 650-mg dose of acetaminophen initiated 45 minutes before zoledronic acid infusion and continuing every 6 hours for 3 days has been shown to reduce severity of symptoms. 1 It is common practice also to ensure the patient is well hydrated prior to infusion. o Raloxifene: There is strong evidence for hot flashes (2-7% increase over placebo) and venous thromboembolic events ( % increased risk) and death due to stroke (0.07% increased risk over placebo) as side effects of raloxifene. a This can be assessed with the timed Get-Up-and-Go test. The test is performed by observing the time it takes a person to rise from an armchair, walk 3 meters (10 feet), turn, walk back, and sit down again. The average healthy adult older than 60 years can perform this task in less than 10 seconds. b Recommendations on interruption of bisphosphonate therapy can be found in the discontinuation section of this guideline. 30

31 Primary reduction of fracture through treatment of women with low bone mass (osteopenia) plus other risk factors Nearly half of the fragility fractures in the population arise in women with low bone mass (defined as a T score of -1.1 to -2.4), due to its higher prevalence compared with osteoporosis. Low bone mass is not adequately predictive of fragility fracture when considered alone, and therapies to reduce fracture have a small risk of serious side effects. Therefore, we recommend only offering treatment to women with low bone mass and an elevated FRAX score or increased risk of falling, especially if the factors conferring additional risk cannot be adequately reduced with non-pharmacologic therapies (e.g., physical activity or therapy, use of a walker or cane, reduction in psychoactive medications). The rationale for making alendronate the primarily recommended therapy is based upon its pharmacodynamics (a 10-year halflife), likely effectiveness in treating low bone mass, clear guidelines around discontinuation, and lower cost. We recommend the option of offering risedronate, ibandronate or zoledronic acid as alternatives to alendronate, based on the single study of risedronate and on assumed class effectiveness of ibandronate and zoledronic acid, i.e. similar mechanism of action and pharmacologic effects, ease of monthly administration, favorable side effect profile, and low cost. Basis of recommendation The bisphosphonates also have well-known side effect profiles. Serious adverse events from bisphosphonates (osteonecrosis of the jaw, atypical femur fracture) are rare; because most patients affected by ONJ develop this complication after dentoalveolar surgery, it is appropriate to delay initiation of bisphosphonate therapy for fracture prevention until 3 months after a planned dental extraction of implant. 2 Less serious effects of bisphosphonates (i.e., upper gastrointestinal symptoms, myalgias, cramps, limb pain) are common. We recommend the alternative option of raloxifene in women at low risk for thrombotic complications based on moderate quality evidence for effectiveness and low risk of serious complications. Because of the increased risk of death from stroke (0.07% absolute risk increase over placebo) in women with increased cardiovascular risk, raloxifene is considered an alternative option to the bisphosphonates. The rationale for recommending against hormone replacement or calcitonin for the treatment of low bone mass is consistent with that in the osteoporosis guideline. The evidence for fracture reduction effectiveness of bisphosphonates in women with low bone mass is considered low quality, as most studies in women with low bone mass were not designed to detect fracture outcomes; this has led to imprecise effect estimates. We do not recommend initiating or continuing hormone therapy (HT) (estrogen or estrogen/progesterone) solely for the prevention of osteoporosis. Although a single large trial supports 31

32 the effectiveness of estrogen and progesterone in reducing hip fractures and all fractures, multiple reports from the Women s Health Initiative have identified association of HT with increased risk for stroke, breast cancer, and venous thromboembolic events. We recommend against the use of calcitonin for the treatment of osteoporosis, given the availability of more effective treatments and the potential for harm with long-term use. Low quality evidence supports the effectiveness of calcitonin in reducing the incidence of vertebral fracture only. In 2013 the FDA reviewed a meta-analysis of 21 studies suggesting an increased risk of cancer in calcitonin-treated vs. placebo-treated patients and concluded that the benefits for the individual patient should be carefully evaluated against all possible risks. FRAX as a risk predictor for fracture FRAX is considered a standard assessment tool by the World Health Organization, the National Osteoporosis Foundation and the United States Preventive Services Task Force. Its area under the ROC curve (AUC) as a predictor of fragility fracture is in an acceptable range for usage ( for hip fractures and 0.54 to 0.81 for other osteoporotic fractures). 2 Of note, 3 studies compared FRAX with simple models, such as age and BMD or age and fracture history, and found the simple models performed as well as FRAX in predicting hip and other clinical fractures 3, 4 and vertebral fractures 5. Because FRAX has been incorporated as a standardized tool in many KP regions, we recommend continued use as a risk stratification tool in adults with low bone mass. The questions needed for FRAX calculation may be incorporated in a DXA screen, and FRAX calculation is incorporated into the DXA software. Alternatively, clinicians may utilize a FRAX calculator, which can be accessed at: Fall risk Hip fracture risk increases with falling. Therefore, in this update we have added language to suggest assessment of overall fall risk in the context of fracture prevention. A systematic review of 33 studies examining risk factors for falls in community living adults ages 65 years 6 listed the strongest risk factors for falling as a history of previous falls; strength, gait, and balance impairments; and use of specific medications. GRADE criteria GRADE assessment Balance of desirable and undesirable effects In post-menopausal women with low bone mass, the potential benefits of bisphosphonate treatment, absolute risk reduction (ARR) of 5% in non-vertebral fractures (risedronate), hazard ratio (HR) = 0.09, 95%CI= may outweigh the low risk of potential serious harms in some cases (Bisphosphonates - Atypical subtrochanteric femur fracture, incidence = 0.002% [with up to 2 years of treatment] to 0.1%[ with greater than 8 years of 32

33 treatment]; osteonecrosis of the jaw, incidence = 0.03% 4.30%). Clinicians should consider non-pharmacologic means of reducing risk and individualize treatment decisions in shared decision-making conversations. The primary evidence for use of alendronate in postmenopausal women with low bone mass comes from two post hoc analyses of women enrolled in the FIT trial. 7, 8 These trials reported a nonstatistically significant trend towards reduction in vertebral fractures. A post hoc analysis from the FIT trial 8 of 2797 postmenopausal women with low bone mass (femoral neck T- score between 1.6 and 2.5 SD and no prevalent fracture) with 4.5 years follow-up found a non-statistically significant trend towards reduction in vertebral fractures. For clinical vertebral fractures, HR = 0.46, 95% CI= ; ARR = 12 per 10,000 person-years. For radiographic vertebral fractures, HR=0.64, 95%CI= ; ARR = 0.2 per 10,000 person-years. There is weak evidence that risedronate is effective for reducing non-vertebral fractures in post-menopausal women with low bone mass. Effectiveness of other bisphosphonates in preventing fracture in women with low bone mass is assumed based on class effect, i.e. similar mechanism of action and pharmacologic effects, ease of monthly administration, favorable side effect profile, and low cost. Only one post hoc analysis of 4 trials of risedronate shows a significant decrease in nonvertebral fractures. This analysis of 620 postmenopausal women with low bone mass (femoral neck T-score between 1 and 2.5 SD and no prevalent fracture) with years follow-up, found that risedronate compared with placebo reduced the risk of nonvertebral fracture: HR= 0.09, 95%CI= No statistically significant reduction in vertebral fractures was found. There are no studies of ibandronate or zoledronic acid in postmenopausal women with low bone mass, measuring fracture as an endpoint. The RUTH trial examined the efficacy of raloxifene among a group of 10,101 post-menopausal women with or at risk for coronary heart disease and unselected by bone mineral density 10, 11, 12, 13 Raloxifene compared with placebo reduced the risk of clinical vertebral fractures, Hazard Ratio (HR) = 0.65, 95%CI = , absolute risk reduction (ARR) = 1.3 per 1000 persons. Raloxifene also reduced the risk of all vertebral fractures. The authors found no reduction in hip or other non-vertebral fractures. In post-menopausal women with low bone mass, the potential benefits of treatment with raloxifene: relative risk reduction of 35% in clinical vertebral fractures, HR = 0.65, 95%CI = , may outweigh the risk of potential serious harms: venous thromboembolic events ( % increased risk over placebo) and death due to stroke (0.07% increased risk over placebo). Clinicians should only consider raloxifene in women of low cardiovascular risk, as the RUTH trial specifically enrolled 33

34 women of increased cardiovascular risk. They should also consider non-pharmacologic means of reducing risk and individualize treatment decisions in shared decision-making conversations. One small RCT of denosumab in postmenopausal women with low bone mass (n=332) did not find a statistically significant reduction in vertebral or non-vertebral fractures; however, this study was not powered to detect fracture outcomes. 14 Evidence for the efficacy of nasal calcitonin-salmon is weak. The largest fracture prevention trial of calcitonin in post-menopausal women with osteoporosis (PROOF) shows borderline efficacy for reduction of vertebral fractures. Results are limited by serious risk of bias due to very high loss to follow-up (59%) and other methodological limitations. Therefore, due to the weak evidence of efficacy, concerns around a potential increased risk of cancer, and the availability of therapies with a stronger and higher quality evidence base, calcitonin is no longer suggested as an alternate therapy for treatment of osteoporosis. There is strong evidence, driven by results of the Women s Health Initiative (WHI), that menopausal estrogen or estrogenprogesterone therapy is effective for reducing the risk of vertebral and non-vertebral fractures in post-menopausal women. However, the risk of serious adverse events including stroke and thromboembolic events may outweigh the benefits of hormone therapy as a treatment for osteoporosis. Risk of vertebral fractures o Estrogen: HR (95%CI) = 0.62( ) o Results for estrogen+progesterone are similar Risk of hip fractures o Estrogen: HR=0.61( ) o Results for estrogen+progesterone are similar Risk of stroke (pooled estimate) o Estrogen: OR(95%CI) = 1.34( ), 3 studies o Estrogen+progesterone: OR=1.28( ), 2 studies Risk of thromboembolic events (pooled estimate) o Estrogen: OR=1.36(1.01-1,86), 4 studies o Estrogen+progesterone: OR=2.27( ), 3 studies Quality of evidence Benefits of therapy Therapy Outcome Strength of evidence-ahrq (key) BISPHOSPHONATES Alendronate Vertebral Low fractures Risedronate Non-vertebral Low Alendronate, ibandronate and zoledronic acid Ibandronate, risedronate and zoledronic acid fractures Non-vertebral/ hip fractures Vertebral fractures Insufficient evidence Insufficient evidence 34

35 Denosumab Vertebral, nonvertebral, hip Insufficient evidence Teriparatide (PTH) Vertebral, nonvertebral, hip Insufficient evidence Raloxifene Vertebral Moderate Harms of therapy Therapy Outcome Strength of Raloxifene Pulmonary Embolism Thromboembolic events myalgias, cramps, and limb pain. Hot flashes Mortality evidence (AHRQ) High/strong Bisphosphonates Atrial fibrillation Esophageal cancer Insufficient evidence Osteonecrosis - Low jaw Atypical subtrochanteric fracture of the femur Mild upper GI High/strong symptoms Alendronate Hypocalcemia Moderate Zoledronic acid Hypocalcemia High/strong Influenza-like symptoms Denosumab mild upper GI symptoms rash High/strong Moderate infection Teriparatide (PTH) Hypercalcemia Headaches Mild upper GI symptoms High/strong Values and preferences This recommendation places a high value on avoidance of serious adverse events; however, risk is low, and can be mediated. Values and preferences were derived by polling the GDT. Uncertainty regarding values and preferences is estimated to be high as direct patient input was not obtained. Variability of values and preferences with respect to initiation of therapy is estimated to be low. Variability in the choice of specific therapeutic agent is estimated to be moderate. Resource implications There exist lesser resource implications for some of the oral agents (alendronate, ibandronate, raloxifene), given their ease of administration and cost. Alendronate is the least expensive of 35

36 the oral bisphosphonates with ibandronate (the second least expensive) exceeding its cost by more than $100 per year s supply. Denosumab is given subcutaneously once every six months, and is comparable to Risedronate in estimated cost. Zoledronic acid is convenient as it is given once yearly, but the necessity of IV administration complicates the ease of delivery, and the need for it to be administered in a health care setting adds to its cost. Teriparatide (PTH) is given subcutaneously daily and is the most expensive of all the treatment options by a factor of at least 5. Osteoporosis (T-score -2.5 or prior fragility fracture): Pharmacologic treatment Preferred therapy Optimize non-pharmacologic therapies, including vitamin D supplementation (1,000 IU daily), calcium intake (total of 1,200 mg daily) and fall prevention efforts (via physical activity or therapy, ambulatory assistive devices, and reduction in sedative/hypnotic medications). (Strong recommendation) Prescribe alendronate (70 mg/week) as a first-line therapy for post-menopausal women with osteoporosis (T-score < -2.5 or a prior fragility fracture) (Strong recommendation) Recommendation Alternative therapy Prescribe alternative bisphosphonates, including ibandronate, risedronate and zoledronic acid, for women who cannot tolerate or should not take alendronate. (Strong recommendation) Additional therapeutic options Consult with specialty care about prescription of denosumab, or teriparatide (PTH) for women who cannot tolerate or should not take bisphosphonates. (Strong recommendation) Consider prescribing raloxifene for women with a low cardiovascular risk who cannot tolerate or should not take bisphosphonates. (Conditional recommendation) Therapies to Avoid Do not prescribe hormone replacement therapy (estrogen or estrogen/progesterone) solely for the treatment of osteoporosis. (Strong recommendation) 36

37 Do not prescribe nasal calcitonin for the treatment of osteoporosis. (Strong recommendation) Clinical considerations Chronic kidney disease Use bisphosphonates with caution in patients with chronic kidney disease and reduced glomerular filtration rate. Current drug monographs state that an estimated GFR <35 ml/min is a contraindication to bisphosphonate use. Dental hygiene Educate patients starting a bisphosphonate about the importance of regular dental cleanings and good dental hygiene. For those patients who have a planned tooth extraction or dental implant surgery, consider delaying the start of bisphosphonate therapy until 3 months after completion of the dental procedure, or until maxillofacial bone healing is complete. Both of these considerations are based upon the moderate evidence for association of osteonecrosis of the jaw (ONJ) with bisphosphonate use (0.001% to 0.069% per year increase over non-bisphosphonate users). a Choice of alternative therapies Consider significant side effects when choosing alternative therapies for patients who do not tolerate alendronate: o Zoledronic acid: There is strong evidence for an acute phase reaction within 3 days of zoledronic acid administration (up to 25% increased risk over placebo of any of the following symptoms: pyrexia, myalgia, headache, arthralgia, chills). A 650-mg dose of acetaminophen initiated 45 minutes before zoledronic acid infusion and continuing every 6 hours for 3 days has been shown to reduce severity of symptoms. 1 It is common practice also to ensure the patient is well hydrated prior to infusion. o Raloxifene: There is strong evidence for hot flashes (2-7% increase over placebo) and venous thromboembolic events ( % increased risk) and death due to stroke (0.07% increased risk over placebo) as side effects of raloxifene. Basis of recommendation In post-menopausal women with osteoporosis (defined by T- score and by prior fragility fracture), strong evidence (i.e., at least one large prospective randomized trial comparing drug to placebo with at least 12 months of follow-up, with fracture as a Recommendations on interruption of bisphosphonate therapy can be found in the discontinuation section of this guideline. 37

38 the primary endpoint) supports the effectiveness of all bisphosphonates, denosumab, teriparatide (PTH) and raloxifene in reducing vertebral fractures. Strong evidence supports the effectiveness of alendronate, risedronate, zoledronic acid, denosumab and teriparatide in reducing nonvertebral and hip fractures in postmenopausal women. We recommend alendronate as the preferred therapy for osteoporosis because of (1) strong and abundant evidence for effectiveness in reducing fracture risk, (2) low-level evidence to support continued effectiveness after discontinuation, and (3) low cost. Other bisphosphonates, including ibandronate, risedronate and zolendronic acid are also effective, though they do not have as extensive of an evidence base as alendronate. In addition, costs of these drugs are currently substantially greater than the generically available alendronate. Therefore, these drugs may be used when alendronate is not well-tolerated or is contraindicated. The effect of ibandronate on hip fracture risk reduction is unclear due to lack of evidence. Nonetheless, in absence of direct evidence of hip fracture reduction, it remains recommended as alternative therapy due to class effect,i.e.. similar mechanism of action and pharmacologic effects, ease of monthly administration, favorable side effect profile, and low cost. 2 Serious adverse events from bisphosphonates (osteonecrosis of the jaw, atypical femur fracture) are rare; because most patients affected by ONJ develop this complication after dentoalveolar surgery, it is appropriate to delay initiation of bisphosphonate therapy for fracture prevention until 3 months after a planned dental extraction of implant. 3 Less serious effects of bisphosphonates (i.e., upper gastrointestinal symptoms, myalgias, cramps, limb pain) are common. Raloxifene is considered as a therapeutic option for women who cannot take or tolerate bisphosphonates. Strong evidence supports its effectiveness in reducing vertebral (but not nonvertebral) fracture, but there are risks of serious complications. It is important that patients be informed of side effects such as hot flashes (2.1% to 7.0% increased risk over placebo), and the complications of thromboembolism (0.2% to 0.7% increased risk over placebo) and death from stroke (0.07% increased risk over placebo), while considering raloxifene and alternative therapies. Denosumab and teriparatide both have strong evidence to support reduction in both vertebral and non-vertebral fractures. We recommend referral to specialty care when primary and alternative therapies are not tolerated or indicated in postmenopausal women who have had a fragility fracture or osteoporosis diagnosis. The high cost and unique side effect profiles of these medicines (strong evidence reports headache and hypercalcemia with teriparatide and moderate evidence reports major infection with denosumab) preclude prescription in a primary care setting. 38

39 We do not recommend initiating or continuing hormone therapy (HT) (estrogen or estrogen/progesterone) solely for the prevention of osteoporosis. Although a single large trial supports the effectiveness of estrogen and progesterone in reducing hip fractures and all fractures, multiple reports from the Women s Health Initiative have identified association of HT with increased risk for stroke, breast cancer, and venous thromboembolic events. We recommend against the use of calcitonin for the treatment of osteoporosis, given the availability of more effective treatments and the potential for harm with long-term use. Low quality evidence supports the effectiveness of calcitonin in reducing the incidence of vertebral fracture only. In 2013 the FDA reviewed a meta-analysis of 21 studies suggesting an increased risk of cancer in calcitonin-treated vs. placebotreated patients and concluded that the benefits for the individual patient should be carefully evaluated against all possible risks. GRADE criteria GRADE assessment In post-menopausal women with osteoporosis, the potential benefits of osteoporosis treatment (Reduction of vertebral fractures: NNT = to prevent 1 vertebral fracture over 1 to 3 years of treatment; reduction of non-vertebral fractures: NNT= to prevent 1 non-vertebral fracture over 1 to 3 years of treatment) significantly outweigh the low risk of potential serious harms: (Bisphosphonates - Atypical subtrochanteric femur fracture, incidence = 0.002% [up to 2 years of treatment] to 0.1% [greater than 8 years of treatment]; osteonecrosis of the jaw, incidence = 0.03% 4.30%) Balance of desirable and undesirable effects There is strong evidence that all bisphosphonates, denosumab, teriparatide (PTH), and raloxifene are effective in reducing the risk for vertebral fractures in post-menopausal women, and with the exception of ibandronate and raloxifene, effective in reducing non-vertebral fractures. Denusomab is an effective therapy for fracture reduction in hip, nonvertebral, vertebral and new clinical vertebral fractures with hazard ratios of 0.31 to There exist adverse effects of mild upper gastrointestinal symptoms (OR= , risk= events per 1000 persons) and infection (RR, 1.3; number needed to harm = 118). Because it is a relatively new therapy (FDA approved in 2010 for this indication), with incomplete post-marketing harms data, we recommend specialty referral when considering its use. Raloxifene is an effective therapy for fracture reduction and has chemopreventive effects against certain breast cancers, and therefore may be an alternative osteoporosis therapy in women at high risk for breast cancer. However, there is strong evidence that raloxifene is associated with an increased risk of thromboembolic events and hot flashes (OR=1.6, risk =

40 events per 1000 persons). There is also moderate quality evidence that raloxifene is associated with a slightly increased risk of death from stroke (0.07% over placebo). Teriparatide (PTH) is an effective therapy for fracture reduction in vertebral and non-vertebral fractures. Adverse events noted are mild upper gastrointestinal symptoms (OR= ), headache (OR=1.5), and hypercalcemia (OR=13). Evidence for the efficacy of nasal calcitonin-salmon is weak. The largest fracture prevention trial of calcitonin in postmenopausal women with osteoporosis (PROOF) shows borderline efficacy for reduction of vertebral fractures. Results are limited by serious risk of bias due to very high loss to follow-up (59%) and other methodological limitations. Therefore, due to the weak evidence of efficacy, concerns around a potential increased risk of cancer, and the availability of therapies with a stronger and higher quality evidence base, calcitonin is no longer suggested as an alternate therapy for treatment of osteoporosis. There is strong evidence, driven by results of the Women s Health Initiative (WHI), that menopausal estrogen or estrogenprogesterone therapy is effective for reducing the risk of vertebral and non-vertebral fractures in post-menopausal women. However, the risk of serious adverse events including stroke and thromboembolic events may outweigh the benefits of hormone therapy as a treatment for osteoporosis. Risk of vertebral fractures o Estrogen: HR (95%CI) = 0.62( ) o Results for estrogen + progesterone are similar Risk of hip fractures o Estrogen: HR=0.61( ) o Results for estrogen + progesterone are similar Risk of stroke (pooled estimate) o Estrogen: OR(95%CI) = 1.34( ), 3 studies o Estrogen + progesterone: OR=1.28( ), 2 studies Risk of thromboembolic events (pooled estimate) o Estrogen: OR=1.36(1.01-1,86), 4 studies o Estrogen + progesterone: OR=2.27( ), 3 studies Quality of evidence Benefits of therapy Therapy Outcome Strength of evidence- AHRQ ( key) Bisphosphonates Alendronate, risedronate, ibandronate, and zoledronic acid Vertebral fractures High/strong 40

41 Alendronate, risedronate, and zoledronic acid Non-vertebral/ hip fractures High/strong Ibandronate Hip fractures Insufficient evidence Alendronate, Wrist fractures Low risedronate Denosumab Vertebral, nonvertebral, hip High/strong Menopausal hormone therapy PMO women Vertebral, hip High/strong PMO women with osteoporosis Fracture reduction, not otherwise specified(nos) Moderate teriparatide (PTH) Vertebral High/strong Non-vertebral Moderate Raloxifene Vertebral High/strong Calcitonin Vertebral Low Non-vertebral Moderate Head to head superiority trials Bisphosphonatesintraclass Superiority for fracture reduction Insufficient evidence Bisphos vs. calcium, teriparatide, or raloxifene Superiority for fracture reduction Insufficient evidence Bisphos vs. menopausal hormone therapy. Superiority for fracture reduction Moderate [no superiority] Harms of therapy Therapy Outcome Strength of evidence (AHRQ) Raloxifene Pulmonary Embolism Thromboembolic events myalgias, cramps, and limb pain. Hot flashes Mortality High/strong Menopausal hormone therapy Bisphosphonates CVA, thromboembolic events Atrial fibrillation Esophageal cancer High/strong Insufficient evidence 41

42 Osteonecrosis - Low jaw Atypical subtrochanteric fracture of the femur Mild upper GI High/strong symptoms Alendronate Hypocalcemia Moderate Zoledronic acid Hypocalcemia High/strong Influenza-like symptoms Denosumab mild upper GI symptoms rash High/strong Moderate Teriparatide (PTH) infection Hypercalcemia Headaches Mild upper GI symptoms High/strong Calcitonin Malignancy Unpublished FDA analysis major limitations noted; unable to assess causality This recommendation places a high value on avoidance of serious adverse effects of medications, while focusing on optimal prevention of fragility fractures. Values and preferences Given superior evidence of efficacy and incidentally low cost, it is presumed that patients would prefer alendronate over other bisphosphonates. Choice of specific therapeutic agent may be influenced for preferences for specific dosing schedules (e.g., ibandronate and risedronate can be given once per month, zoledronate once per year) and routes of administration. Raloxifene is a lower priority option given the (albeit small) increased risk of death from stroke. Values and preferences were derived by polling the GDT. Uncertainty regarding values and preferences is estimated to be high as direct patient input was not obtained. Variability of values and preferences with respect to initiation of therapy is estimated to be low. Variability in the choice of specific therapeutic agent is estimated to be moderate. Resource implications There exist lesser resource implications for some of the oral agents (alendronate, ibandronate, raloxifene) given their ease of administration and cost. Alendronate is the least expensive of the oral bisphosphonates with ibandronate (the second least expensive) exceeding its cost by more than $100 per year s 42

43 supply. Denosumab is given subcutaneously once every six months, and is comparable to risedronate in estimated cost. Zoledronic acid is convenient as it is given once yearly, but the necessity of IV administration complicates the ease of delivery, and the need for it to be administered in a health care setting adds to its cost. Teriparatide (PTH) is given subcutaneously daily and is the most expensive of all the treatment options by a factor of at least 5. Discontinuation of Bisphosphonate Treatment Recommendation In adults taking bisphosphonates for the treatment of osteoporosis or low bone mass, consider discontinuing therapy after 5 years of oral (alendronate, risedronate, ibandronate) or 3 years of intravenous (zoledronic acid) therapy. (Conditional recommendation) o Refer to Clinician Guide, Figures 1, 2 and 4 In adults taking bisphosphonate therapy for the treatment of osteoporosis or low bone mass, bisphosphonate therapy is generally not recommended after 10 years of continuous use of oral (alendronate, risedronate, ibandronate) or 6 years of continuous use of intravenous (zoledronic acid) therapy. (Conditional recommendation) o Refer to Clinician Guide, Figures 3 and 4 The recommendations for bisphosphonate holiday and discontinuation are based primarily on limited evidence of benefit of prolonged use. Though the potential harms of prolonged use are rare, they are serious. Basis of recommendation The balance of benefits and harms for continuing oral bisphosphonates beyond 5 years and intravenous bisphosphonates beyond 3 years of therapy is likely to vary among patients based on their fragility fracture risk and their preferences. However, for patients with risk factors for fragility fracture, the likelihood of benefit is significantly greater than that of harm. After 3 years of bisphosphonate therapy, for each 100 fragility fractures prevented, bisphosphonates are estimated to cause only 0.02 to 0.16 atypical femur fractures. Drug Holiday The extension trials of alendronate (FLEX) and zoledronic acid (HORIZON-PFT) indicate that, with discontinuation of alendronate after 5 years of use or of zoledronic acid after 3 years of use, risk of fragility fracture remains stable for 2 to 3 years. Due to the long half-life of bisphosphonates, there may be a residual effect after discontinuation, and limited evidence suggests the risk of fragility fracture remains stable for 2 to 3 43

44 years after discontinuation. Because of an increased risk of atypical femur fracture (AFF) and perhaps osteonecrosis of the jaw (ONJ) with duration of therapy, coupled with a residual effect of bisphosphonates, it is reasonable to consider discontinuation for patients who, after several years of therapy, are no longer considered at high-risk of fragility fracture. Therefore, we make a conditional recommendation to consider discontinuation of bisphosphonates after 5 years (oral) or 3 years (IV) years of treatment. Drug Discontinuation There are no studies evaluating the use of oral bisphosphonates beyond 10 years and there is insufficient evidence regarding the use of intravenous bisphosphonates for reduction of the risk of fragility fracture beyond 6 years. Therefore, given the uncertain benefit continuing bisphosphonate therapy beyond 10 years (oral) or 6 years (intravenous) of therapy is generally not recommended. GRADE criteria GRADE assessment Fracture reduction Low quality evidence indicates that alendronate use for 5 to <10 years reduces the risk for symptomatic vertebral fracture although no benefit was observed for non-vertebral fracture, including hip fracture. Low quality evidence indicates that zoledronic acid use for 3 to <6 years reduces the risk for morphometric a vertebral fracture although no benefit was seen on hip and total non-vertebral fractures. There are no trials evaluating the efficacy of oral bisphosphonates beyond 10 years and no trials beyond 6 years for intravenous bisphosphonates that has sufficient power to form conclusions specific to reducing the risk of fragility fractures. Balance of desirable and undesirable effects Risk of atypical femur fracture Low quality evidence supports an association between increased duration of bisphosphonate use and increased risk for atypical femur fracture, though at a much lower incidence than that of fragility fracture. Using Kaiser Permanente Southern California data, risk of atypical fracture increased from 1.36 per 10,000 per year for exposure of 2 to 3.9 years to 11.3 per 10,000 per year for exposure of 8 to 9.9 years. Combining data from FIT, FLEX and HORIZON, the relative risk of atypical femur fracture was 1.7, 95%CI = b After 3 years of bisphosphonate therapy, for each 100 fragility fractures prevented, bisphosphonates might cause 0.02 to 0.16 atypical femur fractures, assuming hypothetical risk ratios for atypical femur fracture of 1.2 to 2.4 (NNH= 6,200 to 43,300). With an estimated half-life of alendronate of 10 years, and evidence to suggest risk of fragility fracture remains stable for up to 2 to 3 years after discontinuation, combined with the low, a Sub-clinical vertebral fracture identified by radiographic imaging b Black 2016 NEJM 374;3 44

45 but increasing risk of atypical femur fracture with duration of use, we make a conditional recommendation for considering discontinuation after 5 years of oral bisphosphonate and 3 years of intravenous bisphosphonate use. Certainty of net benefit We have a high level of uncertainty regarding the balance between desirable and undesirable effects. For patients at lower risk of fragility fracture, the risk of rare but serious atypical femur fractures may outweigh the relatively small additional reduction in fragility fracture risk conferred by continuing treatment. However, for patients at higher risk of fragility fracture, the benefit of reduced fragility fracture risk may outweigh the potential harms. Quality of evidence Values and preferences Resource implications The overall quality of evidence for 5 to <10 years of oral bisphosphonate therapy and 3 to <9 years of intravenous bisphosphonate therapy was low (efficacy) to very low (safety). There are no studies examining the efficacy of oral bisphosphonates beyond 10 years of use and insufficient evidence evaluating the use of intravenous bisphosphonates beyond 6 years to determine whether there is a clinical benefit; therefore, the efficacy beyond this timeframe is unknown. This recommendation places a high value on avoidance of possible iatrogenic harm associated with bisphosphonates and an equally high value on prevention of fragility fracture. The uncertainty about the values and preferences of individual patients is high. The variability of values and preferences among providers also is likely to be high (see prior discussion on differences between patients at low vs. high risk of fracture). Continuing bisphosphonate therapy may decrease resource use through reduced incidence of fragility fractures, but may lead to increased resource use and medical-legal risk through increased atypical fractures. Discontinuing bisphosphonates would reduce prescription-related costs but may increase resource use through an increased incidence of fragility fractures. There is likely to be high uncertainty and high variability regarding the resource implications of bisphosphonate continuation versus discontinuation. A variety of clinical considerations affect the decision to continue or discontinue treatment. The time that is required to consider these issues and discuss preferences with individual patients will increase resource utilization. 45

46 Appendix D. Evidence Review This guideline is an update of the KP 2010 Osteoporosis/Fracture Prevention Guideline. 3 The evidence to support the recommendations included in this guideline was derived from multiple externally published and KP conducted. systematic reviews. All adults Non-pharmacologic primary prevention of osteoporosis Evidence Review The 2017 recommendations for non-pharmacologic recommendations were updated to align with current KP methodology. Weight-bearing exercise This recommendation was translated to a conditional recommendation to align with current NGP terminology. Previously, this recommendation was designated consensus-based, and was based on very weak and inconsistent indirect evidence. The recommendation did not fit the criteria for a good practice statement as there is no belief in the medical community nor evidence to suggest a large net positive effect on fracture reduction. Current evidence landscape To get a snapshot of current evidence on this topic, we reviewed Dynamed websites Osteoporosis, prevention and screening and Falls in the elderly, exercise for fall prevention. a Several publications suggest exercise may increase bone mineral density in pre- and post-menopausal women and men; however, results are not consistent across studies. Justification Although studies do not find an effect on fracture reduction, evidence suggest exercise may increase bone mineral density. As well, there is direct evidence that exercise or physical activity reduces the risk of falls in older adults. The justification to translate this recommendation to a conditional recommendation includes: Indirect evidence that exercise may increase BMD and therefore may impact fracture reduction and possibly reduce the risk of falls Direct evidence that exercise or physical therapy reduces the risk of falls in older adults Exercise is recommended for all people as a general preventive healthcare measure. There is minimal cost to the healthcare system, and can be implemented at low cost to the patient. The harms of simple weight-bearing exercise, when performed appropriately, are minimal. Smoking cessation Background This recommendation was translated to a conditional recommendation to align with current NGP terminology. Previously, this recommendation was designated consensus-based, and was based on very weak and inconsistent indirect evidence. The recommendation did not fit the criteria for a good practice statement as there is no belief in the medical community nor evidence to suggest a large net positive effect on fracture reduction. Current evidence landscape Evidence from cohort studies report a lower risk of fragility fracture in former vs. current smokers, and indirectly suggests that smoking cessation may lower fracture risk. Kanis et al., in a meta-analysis of 10 cohort studies (n=59,232) reported A smoking history was associated with a significantly increased risk of a DynaMed [Internet]. Ipswich (MA): EBSCO Information Services Cited 2017 Aug 18. Available from Registration and login required. Copyright (c) , DynaMed LLC. All rights reserved. Accessed by license through KP Clinical Library reference shelf, evidence-based references. 46

47 fracture compared with individuals with no smoking history, but the risk ratios were lower than for current smoking. 4 Justification The justification to translate this recommendation to a conditional recommendation includes: Smoking is a recognized predictor of fracture risk and is included in the FRAX calculation. Observational evidence suggests previous smokers are at a lower risk of hip fracture than current smokers, indirectly implying that smoking cessation may reduce the risk of hip fractures. Smoking cessation is recommended for all people to reduce the risk of multiple morbidities. Smoking cessation programs are already integrated into the healthcare delivery system, and therefore only minimal added cost. Home Safety Proofing Background This recommendation was translated to a conditional recommendation to align with current NGP terminology. Previously, this recommendation was designated consensus-based, based on the assumption that home safety proofing would lower the risk of falls. The recommendation did not fit the criteria for a good practice statement as there is no belief in the medical community nor evidence to suggest a large net positive effect on fracture reduction. Current evidence landscape To get a snapshot of current evidence on this topic, we reviewed the Dynamed website Falls in the Elderly, home assessment and modification. a Included is a large Cochrane systematic review that concluded home safety assessments and interventions appear to reduce falls in high-risk community-dwelling adults. 5 Justification The justification to translate this recommendation to a conditional recommendation includes: RCT level evidence of decreased risk and rate of falling with home safety interventions Hip protectors Background This recommendation was translated to a strong recommendation against hip protectors to align with current NGP terminology. Previously, this recommendation was designated Evidence-based: D. The decision was based on evidence from pooled analyses of RCTs show no (community dwelling) to marginal (long term care setting) association between use of hip protectors and fracture incidence. Current evidence landscape To get a snapshot of current evidence on this topic, we reviewed the Dynamed website, Hip Fractures, external hip protectors. b Evidence suggests that hip protectors may reduce the risk of hip fractures in institutionalized, but not community-dwelling older adults. Compliance was variable across studies. A 2014 Cochrane review 6 concluded Hip protectors probably reduce the risk of hip fractures if made available to older people in nursing care or residential care settings, without increasing the frequency of falls. However, hip protectors may slightly increase the small risk of pelvic fractures. Poor acceptance and adherence by older people offered hip protectors is a barrier to their use. Better understanding is needed of the personal and design factors that may influence acceptance and adherence. Justification The justification to translate this recommendation to a strong recommendation against the use of hip protectors includes: a DynaMed [Internet]. Ipswich (MA): EBSCO Information Services Cited 2017 Aug 18. Available from Registration and login required. Copyright (c) , DynaMed LLC. All rights reserved b DynaMed [Internet]. Ipswich (MA): EBSCO Information Services Cited 2017 Aug 18. Available from Registration and login required. Copyright (c) , DynaMed LLC. All rights reserved 47

48 Lack of benefit in community-dwelling adults and marginal benefit in institutionalized setting Although evidence is inconclusive, the possibility that hip protectors may slightly increase the risk of pelvic fractures cannot be ruled-out Adherence is variable and ranges from 24%-80% from multiple studies Calcium and vitamin D supplementation Background Previously there were separate recommendations for calcium and vitamin D supplementation, graded as Evidence-based B for calcium and consensus-based for Vitamin D. These recommendations were based on fair quality evidence that calcium supplementation, with or without vitamin D, reduces subsequent fractures in postmenopausal women. Total daily intake thresholds of calcium were derived from USDA guidance; published studies informed the thresholds for vitamin D intake. In this update, we make a good practice statement to ensure adequate total daily intake of calcium and vitamin D, and include a conditional recommendation for supplementation if recommended daily allowance (RDA) is not achieved through diet alone. The thresholds were updated to align with current USDA RDA guidance. 7 A conditional recommendation to add calcium and vitamin D supplements if RDA is not achieved through diet is warranted because: Effectiveness of calcium and vitamin D supplementation for reducing the risk of fracture is unclear Calcium with or without vitamin D supplementation may improve bone density Vitamin D supplementation alone has moderate benefit in preventing falls among older adults The likelihood of harm from calcium and vitamin D supplementation among those who cannot obtain these through diet is unclear. a o The WHI trial reported an increased risk for nephrolithiasis (hazard ratio, 1.17 [95% CI, 1.02 to 1.34]) among those taking calcium and vitamin D supplements. The absolute risk was 2.5% in the intervention group and 2.1% in the placebo group, with a number needed to harm of 273. It is uncertain if this adverse effect occurs in vitamin D deficient populations. 8 o A meta-analysis of calcium supplementation suggests an association between calcium use and increased risk for cardiovascular disease, but the link has not been consistently demonstrated. Low Bone Mass (T-score -1.1 to -2.4) Pharmacologic treatment Executive Summary Nearly half of the fragility fractures in the population arise in women with low bone mass (osteopenia) (defined as a T score of -1.1 to -2.4), due to its higher prevalence. 9, 10, 11 Preventing fractures reduces mortality and morbidity. However, low bone mass is not adequately predictive of fragility fracture when considered alone. Therefore, we recommend only offering treatment to post-menopausal women with low bone mass with an elevated FRAX score or increased risk of falling. The strength of evidence supporting these fracture prediction tools is low; the prediction models are constructed primarily from observational studies of association of variables with fracture. Because pharmacologic therapies have a small risk of serious harm, we recommend optimizing non-pharmacologic therapies (e.g., vitamin D/calcium supplementation, physical activity or therapy, use of a walker or cane, reduction in psychoactive medications) prior to consideration of one of the medications. The evidence for effectiveness of bisphosphonates in women with low bone mass is indirect and imprecise. Nonetheless, it does indicate a trend toward reducing risk of fracture. Because of an acceptable side effect a Draft Recommendation Statement: Vitamin D, Calcium, or Combined Supplementation for the Primary Prevention of Fractures in Adults: Preventive Medication. U.S. Preventive Services Task Force. September

49 profile and because of the availability of evidence regarding discontinuation, bisphosphonates remain the first-line recommended prescribed therapy. Raloxifene has direct evidence for effectiveness in reducing risk of vertebral fracture in a large randomized trial of women with increased cardiovascular risk, however there is also an increased risk of death due to stroke (0.07% over placebo) and an increased risk of venous thromboembolic events ( % increased risk over placebo) in this population of women. Because of this increased risk of serious adverse events, raloxifene was determined to be recommended only as a 2nd line alternative prescribed therapy option. Evidence Review Key Questions This evidence update uses the literature obtained in AHRQ s Comparative Effectiveness Review #53 (AHRQ Publication No. 12-EHC023-EF March 2012) Treatment to Prevent Fractures in Men and Women with Low Bone Density or Osteoporosis: Update of a 2007 Report. 12 In 2014, AHRQ updated the same review of the same list of therapies (see below), and also added review of benefits and risks of denosumab (Prolia), a newer therapy. 13 Key Question 1: What are the comparative benefits in fracture risk reduction among the following therapeutic modalities for low bone density? Bisphosphonate medications, specifically: o Alendronate (Fosamax, oral, once weekly o Risedronate (Actonel ; oral, once-a-week) o Ibandronate (Boniva ); once monthly o Zoledronic acid (Reclast IV), once annually Parathyroid hormone (PTH) 1-34 (teriparatide) (Forteo ) Selective estrogen receptor modulators (SERMs), specifically Raloxifene (Evista ) Added to 2014 update 13 : Denosumab (Prolia ); 60 mg subq once every 6 months Key Question 2: What are the short- and long-term harms (adverse effects) of the above therapies (when used specifically to treat or prevent low bone density/osteoporotic fracture)? 49

50 Population Post-menopausal women with low bone mass, defined by t-score at either lumbar spine, femoral neck or hip of -1.1 to -2.4 with no prevalent fragility fractures Health Intervention Bisphosphonates: alendronate, risedronate, ibandronate, zolendronate SERMS: raloxifene PTH: Teriparatide Denosumab (Prolia) Comparison Most important outcomes Head-to-head, drug vs. drug Comparison to placebo Comparison to usual care (no intervention) Efficacy: Osteoporotic fractures: vertebral, hip or total fractures Serious harms Acute coronary syndrome, including AMI Atrial fibrillation Pulmonary embolism (PE) Venous thromboembolic events (VTE) Esophageal cancer Other types of cancer Osteonecrosis of the jaw Atypical femur fractures Mortality Non-serious, important harms Hot flashes (raloxifene) Acute phase reaction, including any of the following symptoms: pyrexia, myalgia, influenza-like symptoms, headache, arthralgia, chills (zoledronic acid) Literature search Because the AHRQ EPC included women with low bone mass, we elected to use their search results to examine this question. Additionally, to ensure no articles were missed, we performed a related article search to key articles that focused on women with low bone mass. The literature search conducted by AHRQ s Evidence Practice Center (EPC) and published in 2012 examined studies published through March Subsequently, an updated review was published in 2014, examining studies published through March 4, Because of the existing strength of the evidence, we elected to not bridge the literature search to include any 2015 publications. 50

51 We conducted a Related Articles search in Pub Med to ensure that no additional publications with these types of analyses existed. Out of 256 citations, we evaluated 17 articles individually that seemed relevant; no new articles emerged that met inclusion criteria. We searched these sources to obtain information about results pertaining to postmenopausal women with low bone mass only. Utilizing the AHRQ 2007, 2012 and 2014 reports as source documents, we also searched for any subgroup or post hoc analyses of randomized trials comparing therapy to placebo, specifically in women with low bone mass. Methodology The search strategy and other methodological details from AHRQ s 2012 evidence synthesis may be found here: Methods. Details of search strategy found here: Appendix A Search Methodology. Study design was limited to RCTs, systematic reviews and meta-analyses. Outcomes were limited to fracture incidence (vertebral, hip, non- vertebral, total fractures). Included in this review were studies of post-menopausal women with low bone mass, or studies that reported results on women with low bone mass separately from those with osteoporosis. Results Benefits of therapy Of the evidence identified in the 2007, 2012 and 2014 AHRQ reports, 12 RCTs reported results for women with low bone mass: 10 RCTs of bisphosphonates , one RCT of denosumab 24, and one study (2 publications) of raloxifene Details of the included studies are found in the following links: Evidence Table C-1. Randomized Controlled Trials Evidence Table C-2. Post-hoc and Subgroup Analyses and Follow-up Studies Evidence Table C-3. Large Randomized Controlled Trials from Original Report Harms of therapy Please refer to the Harms of Treatment section in the systematic review for therapy in post-menopausal women with osteoporosis. We consider the harms of these medicines to likely be the same whether used in osteoporotic or women with low bone mass. 51

52 Evidence Synthesis Benefits of treatment in postmenopausal women with low bone mass Bisphosphonates We consider the evidence for bisphosphonates as a class, with the assumption there exists a class effect across individual medications (alendronate, risedronate, ibandronate and zoledronic acid), i.e. similar mechanism of action and pharmacologic effects, ease of monthly administration, favorable side effect profile, and low cost. We identified studies of alendronate and risedronate, however, there are no studies of ibandronate or zoledronic acid in postmenopausal women with low bone mass, measuring fracture as an endpoint. Vertebral fractures Eight studies reported vertebral fracture results in women with low bone mass, six of alendronate and two of risedronate. Evidence for alendronate includes two post-hoc analyses of the FIT trial 15 16, two post hoc analyses of the FLEX trial 22 23, one meta-analysis of 2 RCTs 20 and one additional RCT 14. A post hoc analysis from the FIT trial of 2797 postmenopausal women with low bone mass (femoral neck T-score between 1.6 and 2.5 SD and no prevalent fracture) with 4.5 years follow-up found a non-statistically significant trend towards reduction in vertebral fractures 16. For clinical vertebral fractures, hazard ratio (HR) = 0.46, 95% CI= , absolute risk reduction (ARR) = 12 per 10,000 person-years. For radiographic vertebral fractures HR=0.64, 95%CI= , ARR = 0.2 per 10,000 person-years. A post hoc analyses of the FLEX trial of 536 women with femoral neck Tscore >- 2.5 comparing 10 years to 5 years of alendronate therapy found a statistically significant reduction in risk difference (RD) for clinical vertebral fractures. 27 Relative risk was not reported in this study; however, results are not likely to be statistically significant as absolute number of fractures were low (7 fractures in the entire group of women with low bone mass). o For women with baseline T-score between -2 and -2.5: RD=1.6, 95%CI= , number needed to treat to prevent one fracture (NNT) = 63 over 5 years of treatment. o For women with baseline T-score >-2.0: RD=1.0, 95%CI= , NNT = 102 over 5 years of treatment. A second post hoc analyses of the FLEX trial (10 years vs. 5 years alendronate) found no reduction in relative risk of morphometric vertebral fractures

53 o For women with baseline T-score between -2 and -2.5: RR= 0.69, 95%CI= o For women with baseline T-score >-2.0: RR=0.84, 95%CI= Results from a meta-analysis of 2 RCTs including 1355 postmenopausal women with mean T- score of -1.8 and 2 to 3 years follow-up were imprecise, with confidence limits spanning a strong benefit to strong harm: RR=0.45, 95%CI= One additional small RCT of 144 postmenopausal women who had recently discontinued hormone replace therapy reported no vertebral fracture events in either arm. 14 This study was not powered for fracture outcomes. Evidence for risedronate includes one post hoc analysis of 4 RCTs 17, and one meta-analysis of 2 RCTs 23 : A post hoc analysis of 620 postmenopausal women with low bone mass (femoral neck T-score between 1 and 2.5 SD and no prevalent fracture) from 4 trials who received 5 mg risedronate (n=311) or placebo (n=309) daily for years did not find a statistically significant reduction in vertebral fractures: HR=0.44, 95%CI= A meta-analysis of 2 RCTs with 327 women with a T-score >- 2.5 using 5 mg dose of risedronate and 2-year follow-up did not find a reduction in vertebral fractures: RR= 0.97, 95%CI=0.42, Non-vertebral fractures Three studies reported non-vertebral fracture results in women with low bone mass, one of alendronate and 2 of risedronate. Evidence for alendronate includes one post hoc analysis of the FLEX trial 21 : A post hoc analyses of the FLEX trial of 536 women with femoral neck Tscore >- 2.5 comparing 10 years to 5 years of alendronate therapy did not find a reduction in nonvertebral fractures: In women with T-score>-2.5 to -2, RR=0.79, 95%CI= In women with T-score > -2, RR=1.41, 95%CI= Evidence for risedronate includes one post hoc analysis 17 and one meta-analysis of 2 RCTs 23. A post hoc analysis of 4 trials of risedronate shows a significant decrease in non- vertebral fractures 17. This analysis of 620 postmenopausal women with low bone mass (femoral neck T- score between 1 and 2.5 SD and no prevalent fracture) with years follow-up, found that risedronate compared with placebo reduced the risk of non-vertebral fracture: HR=0.09, 95%CI= No statistically significant reduction in vertebral fractures was found. 53

54 A meta-analysis of 2 RCTs with 327 women with a T-score >- 2.5 using 5 mg dose of risedronate and 2-year follow-up did not find a reduction in non-vertebral fractures: RR=0.8, 95%CI= Quality Assessment Overall, the quality of the evidence is very low to low. None of the studies were designed to measure fracture outcomes in postmenopausal women with low bone mass. Studies included in this assessment were either small, underpowered RCTs or post-hoc subgroup analyses. There is a high risk of bias inherent in post hoc analyses. As participants were not randomized by subgroup, the potential for imbalances in confounding characteristics exists. In addition, analytics should account for multiple analyses of the same data. Raloxifene The RUTH trial examined the efficacy of raloxifene among a group of 10,101 post-menopausal women with or at risk for coronary heart disease and unselected by bone mineral density Raloxifene compared with placebo reduced the risk of clinical vertebral fractures, HR=0.65, 95%CI= , ARR=1.3 per 1000, and all vertebral fractures. The authors found no reduction in hip or other non-vertebral fractures. There were no significant differences in all-cause mortality or total stroke, but raloxifene was associated with an increased risk of fatal stroke (59 vs. 39 events; HR=1.49, 95%CI= , absolute risk increase (ARI) = 0.7 per 1000 woman-years and venous thromboembolism (103 vs. 71 events; HR=1.44, 95%CI= ; ARI, 1.2 per 1000 woman-years). In addition, there was a statistically significant increase in risk of hot flashes (8.0% vs. 4.8%), leg cramps (9.7% vs 6.7%), peripheral edema (14.4% vs 12.1%) and gallbladder disease (5.6% vs 4.5%), for raloxifene and placebo, respectively. Quality Assessment Overall quality is considered moderate, downgraded for indirectness with respect to study population, as participants were not selected by bone mineral density (Table 1). 54

55 Table 1: Quality assessment of evidence for raloxifene in post-menopausal women 55