Appendix Table D7Treatment Systematic Reviews

Study, YearAimsDatabases searched; Literature search dates; Other data sourcesEligibility criteriaPatients/trialsCharacteristics of identified articles: study designsCharacteristics of identified articles: populationsCharacteristics of identified articles: interventionsMain efficacy outcomeMain efficacy resultsHarms resultsConclusionQuality ScoreComments
Cranney et al, 2002176To review the effect of calcitonin on bone density and fractures in postmenopausal womenMEDLINE, EMBASE
1966–2000; conference abstracts, FDA proceedings
RCTs ≥1 year duration enrolling post-menopausal women, comparing calcitonin to placebo or calcium/vitamin D with fracture or BMD outcomes30 trials; total n=3,993
Chesnut 2000 (n=1,255); Flicker 1997 (n=62); Grigoriou 1997 (n=45); Gurlek 1997 (n=20); Kapetanos 1997 (n=46); Ellerington 1996 (n=117); Hizmetli 1996 (n=107); Melis 1996 (n=102); Perez-Jaraiz 1996 (n=52); Thamsborg 1996 (n=72); Perez 1995 (n=73); Reginster 1995 (n=251); Reginster 1995 (n=150); Rico 1995 (n=72); Campodarve 1994 (n=236); Kollerup 1994 (n=54); Overgaard 1994 (n=134); Reginster 1994 (n=287); Meschia 1993 (n=46); Fioretti 1992 (n=60); Gennari 1992 (n=21); Overgaard 1992 (n=84); Perrone 1992 (n=85); Stevenson 1992 (n=86); Thamsborg 1991 (n=40); Meunier 1990 (n=109); Tremollieres 1990 (n=1990); Overgaard 1989 (n=52); Overgaard 1989 (n=40); Gennari 1985 (n=82)
RCTs; 16 treatment trials, 13 prevention trials, 1 combination treatment/prevention; 15 blinded; 16 concealed treatment allocationMean age 50–70 years 27 trials, <50 years in 3 trials
Mean baseline T-score −0.6 to − 2.9 in 15 trials; not reported in 15 trials
Calcitonin 50–400 IU qd placebo
calcium/vitamin D
Fracture incidence (also change in BMD)Vertebral fracture (4 trials): RR 0.46 (CI 0.25–0.87; p=0.02)
Non-vertebral fracture (3 trials): RR 0.52 (CI 0.22 to 1.23; p-0.14)
Described as poorly reported across the trials; loss to follow-up was similar in calcitonin and control groupsCalcitonin reduces the incidence of vertebral fracture, but the magnitude of effect is unclear due to small sample sizes in the trials used to calculate relative risks and the use of random-effects modeling which may place undue weight on smaller studiesFair
Harris et al, 2008173To assess the ability of ibandronate to reduce fracture risk relative to placeboNot applicableNot applicable4 trials: total n=8,710
Chesnut 2005 - BONE trial (n=2,928 ); Recker 2004 - IV Fracture Prevention trial (n=2,860 ); Reginster 2006 and Miller 2005 - MOBILE trial (n=1,566); Eisman 2006 and Delmas 2006 - DIVA trial (n=1,356 )
Double-blind RCTs reporting fracture outcomesAge 66–69 years
Baseline lumbar spine T-score − 2.81 to − 3.28
Ibandronate, varying doses, dosing schemes and methods of administration (IV and oral) placeboNonvertebral fracture incidence (also clinical fracture incidence)Non-vertebral fractures
High-dose ibandronate: adjusted HR
0.70 (CI 0.50 to 0.99; p=0.41)
Mid-dose ibandronate: adjusted HR
1.04 (CI 0.83 to 1.30; p=0.72)
Any clinical fracture: High-dose ibandronate: adjusted HR
0.73 (CI 0.56 to 0.95; p=0.19)
Mid-dose ibandronate: adjusted HR
0.92 (CI 0.77 to 1.09; p=0.33)
NRHigh-dose ibandronate was associated with demonstrable reductions in risk of nonvertebral and clinical fractureNot quality assessedResults were stratified according to accumulated exposure; High-dose includes FDA-approved 150mg/month oral and 3 mg/3 months IV; Mid-dose includes FDA- approved 2.5mg qd
MacLean et al, 2008187To compare the benefits in fracture reduction and the harms from adverse events of various therapies for osteoporosisCCRCT, MEDLINE, ACP Journal Club 1966–2006Efficacy: systematic reviews, meta-analyses, RCTs of low bone density treatments vs. placebo reporting fracture outcomes Safety: systematic reviews, RCTs and case-control or cohort studies with >1000 patientsEfficacy: 24 meta-analyses, 76 RCTs
Safety: 417 RCTs, 25 controlled clinical trials, 42 observational studies, 9 case reports/case series on osteonecrosis; total number of patients not calculated
Efficacy: 24 meta-analyses, 76
RCTs
Safety: 417 RCTs, 25 controlled clinical trials, 42 observational studies, 9 case reports/case series on osteonecrosis
Men or women with primary or secondary osteoporosis or low bone densityAlendronate, etidronate, ibandronate, pamidronate risedronate, zoledronic acid calcitonin, estrogen, teriparatide, raloxifene, tamoxifen, testosterone, vitamin D, calciumFracture reduction----Data are insufficient to determine relative efficacy or safety of included therapeutic agentsFair
Vestergaard et al, 2007185To examine the effects of parathyroid hormone (PTH) either alone or in combination with antiresorptive therapy on bone mineral density and fracture riskCCRCT (1990–2005); MEDLINE (1951–2005); EMBASE (1974–2005); Science Citation Index (1945–2005); conference abstracts; reference listsRCTs of PTH ≥6 months duration with fracture occurrence and/or BMD outcomes13 trials; total n=5,455
Greenspan 2005 (n=2,531); Lane 1998 (n=51); Body 2002 (n=146); Cosman 2001 (n=126); Neer 2001 (n=1,326); Orwoll 2003 (n=437); Finkelstein 1998 (n=43); Finkelstein 2003 (n=73); Kurland 2000 (n=23); McClung 2005 (n=203); Black 2003 (n=238); Hodsman 2003 (n=206)
RCTs; no further details on design provided
Quality of included trials ranged from 2–4 pts (Jadad)
Men or women age ≥18 years with primary or secondary (i.e. corticosteroid-induced) osteoporosisParathyroid hormone I-34 or I-84 20–100ug qd, alone or in combination with hormone replacement therapy (2 studies), bisphosphonates (5 studies) or nafarelin (1 study)Fracture incidence (also change in BMD)PTH alone results (results for PTH in combination with other treatments were similar with overlapping CIs)
Vertebral fracture (4 studies): RR 0.37 (CI 0.28 to 0.48; p<0.01) Non-vertebral fracture (2 studies): RR 0.62 (CI 0.46–0.82; p<0.01)
Back pain (5 studies): OR 0.68 (CI 0.53 to 0.87; p=0.09)PTH - alone and in combination - reduced incidence of vertebral fracture and, to a lesser extent, non- vertebral fractureGoodResults not pooled due to study heterogeneity
Wells et al, 2008162 AlendronateTo assess the efficacy of alendronate in the primary and secondary prevention of osteoporotic fractures in postmenopausal womenCCRCT, MEDLINE, EMBASE 1966–2007RCTs at least 1 year in duration enrolling postmenopausal women comparing alendronate to placebo or calcium/vitamin D11 trials; total n=12,068
Ascott Evans 2003 (n=144); Cummings 1998 (n=4,432); Hosking 1998 (n=120)
Black 1996 (n=2027); Bone 1997 (n=359); Chesnut 1995 (n=188); Durson 2001 (n=101); Greenspan 1998 (n=120); Greenspan 2002 (n=327); Liberman 1995 (n=994); Pols 1999 (n=1908)
10/11 double-blind RCTs; 1/11
RCT, blinding unclear
Post-menopausal women; age 53–78 years; baseline T-score −1.0 to −4.3Alendronate 5–20mg qd calcium ≤500mg qd vitamin D 125–400 IU qd placeboFracture incidencePrimary prevention Vertebral fracture: RR 0.55 (CI 0.38 to 0.80; p=0.002) Non-vertebral fracture: RR 0.89 (CI 0.76 to 1.04; p=0.14)
Hip fracture: RR 0.79 (CI 0.44 to 1.44; p=0.4) 5-year fracture risk (based on FRACTURE Index scores)
Score 1–2: ARR 0.5%; NNT 200 Score 3–4: ARR 1.1%; NNT 91 Score 5: ARR 2.4%; NNT 42 Score 6–7: ARR 3.2%; NNT 31 Score 8–13: ARR 5/0%; NNT 20
Secondary prevention Vertebral fracture: RR 0.55 (CI 0.43 to 0.69; p<0.001) Non-vertebral fracture: RR 0.77 (CI 0.64 to 0.92; p=0.005)
Hip fracture: RR 0.47 (CI 0.26 to 0.85; p=0.01)
No difference in tolerability or withdrawals due to AEs between alendronate and placebo/control groups with the exception of increased incidence of GI events (RR 1.03; CI 0.98 to 1.08) and esophageal ulcer (RR 1.16; CI 0.39 to 3.45) in the alendronate group; no reports of osteonecrosisFor primary prevention, clinically important reduction in vertebral fractures but not other types of fractures; secondary prevention clinically and statistically significant reduction in vertebral, non- vertebral, hip and wrist fractureGood
Wells et al, 2008163 EtidronateTo assess the efficacy of etidronate in the primary and secondary prevention of osteoporotic fractures in postmenopausal womenCCRCT, MEDLINE, EMBASE 1966–2007RCTs at least 1 year in duration enrolling postmenopausal women comparing oral etidronate to placebo or calcium/vitamin D11 RCTs; total n=1,248
Primary prevention: Herd 1997 (n=152); Meunier 1997 (n=54); Pouilles 1997 (n=109)
Secondary prevention: Ishida 2004 (n=132); Lyritis 1997 (n=100); Montessori 1997 (n=80); Pacifici 1988 (n=57); Shiota 2001 (n=40); Storm 1990 (n=66); Watts 1990 (n=423); Wimalawansa 1998 (n=35)
5/11 double blindPostmenopausal women age 53–72 years; baseline T-score −0.8 to −4.3Etidronate 200–400 mg qd calcium (dose not consistently reported across included trials) placeboFracture incidencePrimary prevention Vertebral fracture: RR 3.03 (CI 0.32 to 28.44; p=0.3) Non-vertebral fracture: RR 0.56 (CI 0.20 to 1.61; p=0.3) Hip fracture: no evidence available
Secondary prevention Vertebral fracture: RR 0.53 (CI 0.32 to 0.87; p=0.01) Non-vertebral fracture: RR 1.07 (CI 0.72 to 1.60; p=0.7) Hip fracture: RR 1.20 (CI 0.37 to 3.88; p=0.8)
Withdrawals: RR 0.91 (CI 0.71 to 1.26)
Withdrawals due to AEs: RR 0.61 (CI 0.25 to 1.49)
No statistically significant difference in AEs
No clinically or statistically significant reduction in fracture incidence was found with etidronate use with the exception of reducing vertebral fracture in a secondary prevention populationGood
Wells et al, 2008161 RisedronateTo assess the efficacy of risedronate in the primary and secondary prevention of osteoporotic fractures in postmenopausal womenCCRCT, MEDLINE, EMBASE 1966–2007RCTs at least 1 year in duration enrolling postmenopausal women comparing risedronate to placebo or calcium/vitamin D7 RCTs; total n=14,049
Hooper 2005 (n=381); Mortensen 1998 (n=111); Clemmesen 1997 (n=132; trial excluded from analysis due to study design); Fogelman 2000 (n=541); Harris 1999 (n=2,458); McClung 2001 (n=9,331); Reginster 2000 (n=1,222)
All double-blind studiesPostmenopausal women, age 51–78 years; baseline T-score −0.4 to 3.7Risedronate 2.5; 5 mg qd
cyclical risedronate 2.5; 5 mg qd
calcium 1000 mg qd
vitamin D 500 IU qd
placebo
Fracture incidencePrimary prevention Vertebral fracture: RR 0.97 (CI 0.42 to 2.25; p=0.94) Non-vertebral fracture: RR 0.81 (CI 0.25 to 2.58; p=0.72)
Hip fracture: inadequate evidence
Secondary prevention Vertebral fracture: RR 0.61 (CI 0.5 to 0.76; p<0.001) Non-vertebral fracture: RR 0.80 (CI 0.72 to 0.90; p=0.0002)
Hip fracture: RR 0.75 (CI 0.59 to 0.94; p=0.01)
Withdrawals (5 trials): RR 0.96 (CI 0.91 to 1.00)
Withdrawals due to AEs (5 trials): RR 0.96 (CI 0.88 to 1.05)
Adverse events - any upper GI event: RR 1.01 (CI 0.94 to 1.09)
Other specific AEs not pooled, reported as generally no difference between risedronate and placebo
Primary prevention Vertebral fracture: RR 0.97 (CI 0.42 to 2.25; p=0.94)
Non-vertebral fracture: RR 0.81 (CI 0.25 to 2.58; p=0.72)
Hip fracture: inadequate evidence
Secondary prevention Vertebral fracture: RR 0.61 (CI 0.5 to 0.76; p<0.001)
Non-vertebral fracture: RR 0.80 (CI 0.72 to 0.90; p=0.0002)
Hip fracture: RR 0.75 (CI 0.59 to 0.94; p=0.01)
Withdrawals (5 trials): RR 0.96 (CI 0.91 to 1.00)
Withdrawals due to AEs (5 trials): RR 0.96 (CI 0.88 to 1.05)
Adverse events - any upper GI event: RR 1.01 (CI 0.94 to 1.09)
Other specific AEs not pooled, reported as generally no difference between risedronate and placebo
Primary prevention Vertebral fracture: RR 0.97 (CI 0.42 to 2.25; p=0.94)
Non-vertebral fracture: RR 0.81 (CI 0.25 to 2.58; p=0.72)
Hip fracture: inadequate evidence
Secondary prevention Vertebral fracture: RR 0.61 (CI 0.5 to 0.76; p<0.001)
Non-vertebral fracture: RR 0.80 (CI 0.72 to 0.90; p=0.0002)
Hip fracture: RR 0.75 (CI 0.59 to 0.94; p=0.01)
Men
Sawka et al, 2005164To systematically review the anti-fracture efficacy of alendronate in men with low bone mass or with a history of prevalent fracture and incorporate prior knowledge of alendronate efficacy in women in the analysisCCRCT (through 2004), MEDLINE (1966–2004), EMBASE (1996–2004)RCTs of alendronate with men comprising at least half of the study population with ≤1 year follow-up reporting fracture outcomes2 trials; total n=375
Orwoll 2000 (n=241); Ringe 2004 (n=134)
RCTs; one double-blind (Orwoll), one open-label (Ringe)Mean age 63 years
Baseline T-score −1.0 to −2.0
Alendronate 10 mg qd calcium/vitamin D alfacalcidiolFracture incidenceVertebral fracture: OR 0.36 (CI 0.17 to 0.77)
Non-vertebral fracture: OR 0.73 (CI 0.32 to 1.67)
Bayesian random effects model (incorporating data from women)
Vertebral fracture: OR 0.44 (CRI 0.23 to 0.83)
Nonvertebral fracture: OR 0.60 (CRI 0.29 to 1.44)
NRVertebral fracture: OR 0.36 (CI 0.17 to 0.77)
Non-vertebral fracture: OR 0.73 (CI 0.32 to 1.67)
Bayesian random effects model (incorporating data from women)
Vertebral fracture: OR 0.44 (CRI 0.23 to 0.83)
Nonvertebral fracture: OR 0.60 (CRI 0.29 to 1.44)
NRVertebral fracture: OR 0.36 (CI 0.17 to 0.77)
Non-vertebral fracture: OR 0.73 (CI 0.32 to 1.67)
Bayesian random effects model (incorporating data from women)
Vertebral fracture: OR 0.44 (CRI 0.23 to 0.83)
Nonvertebral fracture: OR 0.60 (CRI 0.29 to 1.44)
Tracz et al, 2006186To estimate the effect of testosterone use on bone health outcomesCCRCT (through 2005), MEDLINE (1966–2005), EMBASE (1988–2005), reference lists, content expert filesRCTs of testosterone versus placebo reporting fractures as or BMD as an outcome8 trials; total n=388
Amory 2004 (n=48); Crawford 2003 (n=34); Fairfield 2001 (n=50); Hall 1996 (n=30); Kenny 2001 (n=67); Reid 1996 (n=16); Snyder 1999 (n=108)
RCTs; 7/8 studies blinded (know or presumed); 1 crossover studyMean age 60–75 years in 6 trials; <60 years in 2 trialsTestosterone 200–250mg qd or 2.5mg patch placeboFracture incidence (and change in BMD)No studies reported on fracture outcomesNRNo studies reported on fracture outcomesNRNo studies reported on fracture outcomes

Abbreviations: AE = adverse effects; ARR = absolute risk reduction; BMD = bone mineral density; CI = confidence interval; CRI = corresponding credibility interval; GI = gastro-intestinal; HR = heart rate; NNT = number needed to treat; NR = not reported; OR = odds ratio; PTH = parathyroid hormone; RR = relative risk; RCT = randomized controlled trial.

From: Appendix D, Appendix Tables

Cover of Screening for Osteoporosis
Screening for Osteoporosis: Systematic Review to Update the 2002 U.S. Preventive Services Task Force Recommendation [Internet].
Evidence Syntheses, No. 77.
Nelson HD, Haney EM, Chou R, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.