PSAP-VII • Women’s and Men’s Health 203 Drug-Induced Osteoporosis Learning Objectives 1. Apply an understanding of drug mechanisms to explain drug-induced osteoporosis and its consequences. 2. Use the diagnostic process to develop treatment and monitoring plans for drug-induced osteoporosis. 3. Assess for potential risk factors associated with drug- induced osteoporosis. 4. Evaluate the epidemiologic literature to assess the absolute increase in risk of drug-induced osteoporo- sis associated with a specific class of agents. Introduction Osteoporosis is a disease characterized by low bone mass and deterioration of the bone architecture leading to increased fragility and fractures. A fragility fracture is defined as one that results from no identifiable trauma or minimal trauma, such as a fall from standing height or less. Although a fragility fracture can occur in any bone, fractures of the hip, spine, and wrist are particularly common in individuals with osteoporosis. Primary osteo- porosis refers to a reduction in bone mass related to aging and menopause, whereas secondary osteoporosis results from specific diseases or drugs. e relationship between osteoporosis and oral gluco- corticoids is well understood, but in recent years, many other agents have been reported to affect bone health and should therefore also be considered in the risk assessment for osteoporosis. is chapter reviews new evidence for drugs other than glucocorticoids in the development of secondary osteoporosis and provides a framework for puing risk into context for suitable discussion with patients. Epidemiology Osteoporotic fractures are a major public health prob- lem, with about 1.5 million occurring each year. In 2005, the costs associated with osteoporosis were estimated at $17 billion. Epidemiologic reports estimate that 10 million Americans have established osteoporosis, and another 34 million Americans are at risk because of osteo- penia. Currently, there are no reliable estimates of the prevalence of drug-induced osteoporosis or osteopenia. An estimated 80% of those affected by osteoporosis are women, and as discussed in another chapter in this book, men also develop the disease. Although those of Asian or white background are at the highest risk, osteo- porosis crosses all ethnic boundaries and can occur in anyone regardless of ethnicity. Bone health behaviors should be assessed regularly in all patients because every- one is at risk of osteoporosis and osteoporotic fractures. Seniors and adults with risk factors generally qualify for bone mineral density (BMD) testing with dual-energy x-ray absorptiometry (DEXA). Drug-Induced Osteoporosis By Susan K. Bowles, Pharm.D., MSc, FCCP Reviewed by Mary Beth O’Connell, Pharm.D., FCCP, FASHP, BCPS; and Michelle M. Richardson, Pharm.D., FCCP, BCPS Baseline Review Resource e goal of PSAP is to provide only the most recent (past 3–5 years) information or topics. Chapters do not provide an overall review. Suggested resources for background information on this topic include: • O’Connell MB, Vondracek SF. Osteoporosis and other metabolic bone diseases. In: DiPiro JT, Talbert RL, Yee GC, Matzke GR, Wells BG, Posey LM, eds. Pharmacotherapy: A Pathophysiologic Approach, 7th ed. New York: McGraw-Hill Medical, 2008[12]:1483–504. • National Osteoporosis Foundation. Clinicians Guide to Prevention and Treatment of Osteoporosis. 2010. Available at www.nof.org/professionals/clinicians_guide.htm. Accessed May 25, 2010. • Silverman SL, Lane NE. Glucocorticoid-induced osteoporosis. Curr Osteoporos Rep 2009;7:23–6.
Transcript
PSAP-VII • Women’s and Men’s Health 203 Drug-Induced Osteoporosis
Learning Objectives1. Apply an understanding of drug mechanisms
to explain drug-induced osteoporosis and itsconsequences.
4. Evaluate the epidemiologic literature to assess theabsoluteincreaseinriskofdrug-inducedosteoporo-sisassociatedwithaspecificclassofagents.
Introduction Osteoporosis is a disease characterized by low bonemassanddeteriorationofthebonearchitectureleadingtoincreasedfragilityandfractures.Afragilityfractureisdefined as one that results from no identifiable traumaor minimal trauma, such as a fall from standing heightor less. Although a fragility fracture can occur in anybone,fracturesofthehip,spine,andwristareparticularlycommoninindividualswithosteoporosis.Primary osteo-porosisreferstoareductioninbonemassrelatedtoagingand menopause, whereas secondary osteoporosis resultsfromspecificdiseasesordrugs. Therelationshipbetweenosteoporosisandoralgluco-corticoidsiswellunderstood,butinrecentyears,manyotheragentshavebeenreportedtoaffectbonehealthand
shouldthereforealsobeconsideredintheriskassessmentforosteoporosis.Thischapterreviewsnewevidencefordrugsotherthanglucocorticoidsinthedevelopmentofsecondary osteoporosis and provides a framework forputting risk into context for suitable discussion withpatients.
Epidemiology Osteoporoticfracturesareamajorpublichealthprob-lem,withabout1.5millionoccurringeachyear.In2005,the costs associated with osteoporosis were estimatedat $17 billion. Epidemiologic reports estimate that 10million Americans have established osteoporosis, andanother34millionAmericansareatriskbecauseofosteo-penia. Currently, there are no reliable estimates of theprevalenceofdrug-inducedosteoporosisorosteopenia. An estimated 80% of those affected by osteoporosisare women, and as discussed in another chapter in thisbook, men also develop the disease. Although those ofAsianorwhitebackgroundareatthehighestrisk,osteo-porosis crosses all ethnic boundaries and can occur inanyone regardless of ethnicity. Bone health behaviorsshouldbeassessedregularlyinallpatientsbecauseevery-oneisatriskofosteoporosisandosteoporoticfractures.Seniorsandadultswithriskfactorsgenerallyqualifyforbone mineral density (BMD) testing with dual-energyx-rayabsorptiometry(DEXA).
• National Osteoporosis Foundation. Clinicians Guide to Prevention and Treatment of Osteoporosis. 2010.Availableatwww.nof.org/professionals/clinicians_guide.htm. AccessedMay25,2010.
PSAP-VII • Women’s and Men’s Health204Drug-Induced Osteoporosis
Risk Factors Among the most important strategies in managingdrug-inducedosteoporosis isawarenessofother factorsthancanaffectbonehealthandfracturerisk(Box1-1).Themostcommonoftheseriskfactorsareageandfemalesex.Womenexperiencemorehipfracturesthanmen;thisis thought tobe related toa lowerpeakbonemassandgreatereffectofmenopause.However,one-thirdofallhipfracturesoccurringafterage65occurinmen.Therefore,bonehealthandfractureriskareimportanttoconsiderinallpatientstakingdrugswiththepotentialtocauseboneloss. Many osteoporotic fractures are related to falls, soevaluation of risk factors for falls (Box 1-2) is also animportantcomponentintheassessmentofbonehealth.Many falls are preventable, and reducing fall rates canpreventfracturesandthemorbidityandmortalityassoci-atedwiththem. AsshowninFigure1-1,severalfactorsareassociatedwith declining BMD and impaired bone quality. Theseinclude factors related to aging and hypogonadism, aswellastheclinicalriskfactorsoutlinedinBox1-1.
Diagnosis The diagnosis of osteoporosis is primarily based onthemeasurementofBMD,butaclinicaldiagnosiscanbe
Abbreviations in This ChapterAI AromataseinhibitorBMD BonemineraldensityDEXA Dual-energyx-ray
DrugsassociatedwithosteoporosisFallsaFactors included in the WHO fracture risk assessment tool(FRAX).bOnlyhipfractureofparentincludedasaFRAXriskfactor.cOnlythesesecondarycauses—type1diabetes,osteogenesisimperfectainadults,untreatedlong-standinghyperthyroidism,hypogonadismorprematuremenopause(<45years),chronicmalnutrition,ormalabsorptionandchronicliverdisease—areincludedasFRAXriskfactors.FRAXdoesnotincludeprescriptiondrug–inducedcausesofosteoporosisexceptforglucocorticoids.
PSAP-VII • Women’s and Men’s Health 205 Drug-Induced Osteoporosis
made insomeonewithseveral risk factorsexperiencingafragilityfracture.MethodsformeasuringBMDincludecentralDEXA,peripheralDEXA,quantitativecomputedtomography, and quantitative ultrasound densitometry.Althoughallofthesecanpredicttheriskoffragilityfrac-tures, evidence supports the use of DEXA as the mostaccuratemethod,withthebestpositivepredictivevalue,toestimatefractureriskinpostmenopausalwomenandmenolderthan50. Measurements are taken at common sites for frac-ture(e.g.,forearm,hip,spine)andexpressedasgramsof
mineralpercentimeterscanned.ThesemeasurementsarethenexpressedaseitheraT-scoreorZ-score.AT-scoreisessentiallyacomparisonofpeakbonemassbasedonthedifferencebetweenapatient’sBMDandtheaveragefound in a white, healthy, young adult of the same sex.TheZ-scorerepresentsacomparisonofapatient’sBMDwiththatexpectedforsomeoneofthesamesexandsim-ilar age. Both are expressed as the number of standarddeviations(SD)aboveorbelowthemean. The World Health Organization (WHO) has devel-opedaclassification forbonehealthbasedonBMD,asdeterminedbyDEXA, forpostmenopausalwomenandmen50yearsorolder.AsshowninTable1-1,osteopo-rosis is defined as a T-score of −2.5 or less. The WHOdiagnostic criteria should not be applied to children,premenopausal women, or men younger than 50. Inthese groups, the International Society for ClinicalDensitometry recommends that race-adjusted Z-scoresbeused.AZ-scoreof−2.0or lowerisdefinedasbelowthe expected range for age. A Z-score of greater than−2.0 is considered within the expected range for age.TheZ-scorecanbeusedtoguidetreatmentdecisionsfordrug-induced osteoporosis in individuals younger than50andprovideanestimateoffractureriskfordiscussionwithpatients.
PSAP-VII • Women’s and Men’s Health206Drug-Induced Osteoporosis
TheWHOhasdevelopedaFractureRiskAssessmentTool(FRAX)thatcanbeusedwithorwithoutfemoralneck BMD measurements (expressed either as gramsper square centimeter or a T-score). The machine typeshould be specified when using FRAX to estimate frac-ture risk. In general, FRAX is not used if the person isalready receiving treatment with osteoporosis prescrip-tiondrugs.TheonlineFRAXtool(availableatwww.shef.ac.uk/FRAX)canbeusedtoestimatethe10-yearprob-ability of fracture risk for overall major osteoporoticfractures(i.e.,spine,forearm,hip,andshoulder)andhipfracture for women and men aged 40–90. In general, a10-year overall fracture risk of less than 10% is consid-eredlowrisk;10%to20%ismoderaterisk;andgreaterthan 20% is high risk. The only drug-induced osteopo-rosisdrugsevaluatedwith this toolareglucocorticoids,nicotine, and alcohol. The effect of alcohol use is morerelatedtotheriskoffallsthantoanactualeffectonBMD. The peripheral DEXA and quantitative ultrasounddensitometrytestsareoftenusedincommunitypharma-ciesorambulatoryclinicstoidentifypatientswhoshouldbe referred for further evaluation by DEXA. However,these methods of measuring BMD should not be usedinindividualswithseveralriskfactors,fragilityfractures,or secondarycauses forosteoporosis.Theyarealsonotappropriate for children, premenopausal women, oryoungmenunlessthepatienthasatleastonemajorriskfactor.Therefore,theydonothavearoleinthescreeningofpatientsfordrug-inducedosteoporosis.
Osteoporosis Related to Specific DrugsHormonal Therapies Estrogenandtestosteroneareimportantregulatorsoftheboneremodelingprocess,soitisnotsurprisingthatosteoporosis is associated with a decline in hormonalconcentrations after menopause. Similarly, testosteronedeficiencyisthemostcommoncauseofosteoporosisinmen,althoughtheroleoftestosteroneisnotasstraight-forward as once thought. Drugs inhibiting secretionor altering the metabolism of sex hormones have thepotential to induce osteoporosis. These drugs includethearomataseinhibitors(AIs)andgonadotropin-releas-inghormone(GnRH)agonistsusedinthetreatmentofbreastandprostatecancers,aswellasthecontraceptivedepotmedroxyprogesteroneacetate(DMPA). Thyroidhormonesalsoaffectbonemetabolism,withincreasedboneresorptionobservedinhyperthyroidism.Theboneeffectsresultfrombothendogenousandexog-enouscausesofhyperthyroidism.
Aromatase Inhibitors The use of AIs as adjuvant treatment for breast can-cerhasbeenshowntoimprovedisease-freesurvivaland
decrease the occurrence of metastatic disease in post-menopausal women with estrogen receptor–positivedisease. However, the pharmacologic activity of theseagentsalsoaffectsBMDandfracturerisk. Aftermenopause,estrogenisproducedintheperiph-eral tissues by the conversion of adrenal androgens toestrogen.TheAIsinhibitthearomataseenzyme,respon-sibleforthisconversion,andresultindecreasedestrogenconcentrations. Because many postmenopausal womenhave several underlying risk factors for osteoporosis,furtherestrogen loss fromtreatmentwithAIsmightbeexpectedtocausebonelossandincreasedfracturerisk. The potential for AIs to affect bone health wasanticipated based on their pharmacologic activity andtherefore was evaluated during clinical trials studyingthe efficacy and tolerability of these agents in treat-ingbreastcancer.TherateofbonelossassociatedwithAIs varies depending on the patient population stud-ied.However,adeclineofabout4%to6%over5yearshas generally been observed for both anastrozole andletrozole. The relationship between this bone loss andan increased risk of fragility fractures was definitivelyestablishedforbothdrugs,withfractureratesof11%foranastrozoleand8.6%forletrozolereportedinrandom-izedcontrolledclinicaltrials.Comparedwithtamoxifen,whichisthoughttoofferboneprotection,theabsoluteriskincrease(ARI)forallfragilityfracturesis3.3%foranastrazole and 2.8% for letrozole. This translates intooneexcessfractureforevery30–35womentreatedwitheitherAIover5years. Earlystudiesinanimalssuggestthatexemestane,aste-roidalAI,doesnotsharethesameriskofbonelossandfragility fractures as anastrozole and letrozole. Becauseof itssteroidalstructure,exemestanealsoexhibitssomeandrogenic activity, potentially offsetting the bone lossproperties of AI-induced estrogen deficiency. However,thisbenefithasnotbeenshown inclinical trials,whichfoundtherateofbonelosscomparablewiththatofnon-steroidal AIs. Similarly, when exemestane is comparedwith tamoxifen, an ARI of 2% for fragility fractures isobserved;thistranslatestooneexcessfractureforevery50 patients treated with exemestane during a 5-yearperiod.
GnRH Agonists Gonadotropin-releasing hormone agonists, used inconjunction with chemotherapy, improve disease-freesurvivalinpremenopausalwomenwithhormonerecep-tor–positivebreastcancer.Althoughmorecontroversial,these agents are sometimes used in combination withAIs or tamoxifen. The GnRH agonists down-regulatethesecretionof luteinizinghormone(LH)and follicle-stimulatinghormone(FSH),resultinginsuppressionofovarianfunctionandacorrespondingdeclineinestrogenproduction. Suppression of ovarian function by GnRHagonists is a treatment strategy also employed in the
PSAP-VII • Women’s and Men’s Health 207 Drug-Induced Osteoporosis
managementofendometriosis.Regardlessoftheindica-tionforovariansuppression,bonemetabolismislikelytobeaffectedandresultinboneloss. Whetherusedaloneor incombinationwithanAIorchemotherapy, goserelin or leuprolide in premenopausalwomencausesarapiddeclineinBMDofabout5%to6%over 6–12 months. Limited data suggest some recoveryin BMD on discontinuation of GnRH agonists, but thedegreeandtimeframearenotwellunderstood.Asaresult,the relationship between GnRH-related bone loss andfutureriskoffragilityfracturehasnotbeenestablished. ThetherapeuticeffectofGnRHagonistsinthetreat-mentofprostatecancer is related to theirantiandrogeneffects, which result in decreased serum testosteroneconcentrations.Bone loss isawell-knownconsequenceofhypogonadism,soitfollowsthatdecreasedBMDisapotentialcomplicationofGnRHagonists.Longitudinalstudies documented significant bone loss within 1 yearof initiating therapy with a GnRH agonist (2% to 5%decreaseinBMD,dependingontheanatomicsitemea-sured).Severalsmallstudieshaveevaluatedfractureriskamong men receiving treatment with GnRH agonists,butthefindingswerelimitedbysmallsamplesizeandthelackofcontrolgroups. One large retrospective cohort study of more than50,000 men observed an increased risk of fracture inmen receiving androgen deprivation therapy for themanagement of prostate cancer. The relative risk (RR)during a 5-year period was significantly higher for menwhoreceivedGnRHagonistsinthefirstyearafterdiag-nosis than for men without antiandrogen therapy. TheARIwasestimatedat3.6%,resultinginacalculatednum-berneededtoharm(NNH)of28forallmenreceivinga GnRH. This means that one excess fracture occurredforevery28mentreatedover5years.Givenanestimatedannualincidenceofprostatecancerof220,000casesperyear, and current drug use patterns for GnRH agonists,thistranslatestoabout3000excessfracturesannually. Furthermore, dose-response and age effects wereobserved.TheNNHdeclinedsignificantlyformenreceiv-ing nine or more doses compared with those receivingfour or fewer doses. A decline in the NNH is indicativeofahigherARI.Forexample,thegroupofmenreceivingonetofourdoseshadanNNHof74,whereasthegroupreceivingnineormoredoseshadanNNHof18.Thissug-geststhattheriskoffractureincreaseswiththenumberofGnRH doses received. A similar trend toward decliningNNHwasobservedinalldosecategorieswithincreasingage,indicatingthattheriskoffractureincreaseswithage.Thisisnotsurprisingbecauseoldermenhavelowerbase-lineBMDandwouldbeexpectedtobeatincreasedriskoffracturewhenbonelossisacceleratedbydrugs.
including some 400,000 adolescents. This agent pre-vents pregnancy by inhibiting LH and FSH, causinganovulation and a corresponding decrease in estrogenproduction. The potential loss of bone owing to DMPA-relatedestrogendeprivationisofparticularconcernforteenagegirlsandwomenyoungerthan30,atimewhenBMDnor-mallyincreases.Prolongedusecouldpotentiallydecreasethepeakbonemassandincreasetheriskoffragilityfrac-turesin20–30years.Longitudinaldatashowanannualrateofbonelossof1%to3%inadolescentsaged12–18,withthegreatestreductioninBMDoccurringinthefirst2yearsofuse.Adultwomenshowasimilarpattern,withmostbonelossobservedinthefirst2yearsofuse,slow-ingthereaftertoanannualrateofabout0.5%. WhenDMPAisdiscontinued,adolescentbonemassundergoes full recovery at the spine and at least par-tiallyat thehipwithin24–36months.ProspectivedataindicatethatBMDinadultssimilarlyrecovers.Thelong-term effect of continued DMPA use on fracture risk isunknown;however,someresearchhasexaminedwhethertheuseofDMPAincreasestheriskoffracturesduringtheshortterm. One prospective study of female Army recruitsage 16–35 found that DMPA use was associated withincreasedriskofstressfractureofthecalcaneusinwhitewomen.Unlikefragility fractures,stress fracturesareanincomplete fracture of the bone; caused by unusual orrepeated stress, they are often associated with athleticactivitysuchasmightbeexpectedinmilitarylife.Studyparticipants had several other risk factors, including alower-than-averageBMIandhighratesofcigarettesmok-ingandalcoholuse. A second cross-sectional study examined the risk offractureassociatedwithDMPAuseinacohortofdevelop-mentally delayed premenopausal women. Fracture risk,evenwhencorrectedforage,race,andconcomitantanti-convulsant use, was significantly increased with DMPAuse.Theoveralleventratewassmall,however,with3.6%of DMPA users experiencing a fracture compared with1.6%ofnonusers.AlthoughtheARIof2%(NNHof50)is similar to that observed with AIs, these findings arefromaretrospectivecohortstudyratherthanarandom-izedcontrolledtrial,andnotallpotentiallyconfoundingvariables (e.g., seizure activity) could be accounted forusing this study design. Larger and longer-term longi-tudinal studies are needed to accurately determine theassociationbetweenDMPAuseandfractures. Available data have prompted the U.S. Food andDrugAdministration(FDA)to issueablackboxwarn-ingstatingthatprolongedusemightresultinasignificantandcumulativebone loss thatmightnotbecompletelyreversibleondiscontinuation.ThecautionconcernstheuseofDMPAbeyond2yearsinpremenopausalwomen;the manufacturer recommends BMD testing by DEXAafter2yearsofuse.
PSAP-VII • Women’s and Men’s Health208Drug-Induced Osteoporosis
Thyroid Replacement Therapy Hyperthyroidism and thyroid replacement therapyarebothassociatedwithbone loss.Thyroid-stimulatinghormone (TSH) receptors have been identified onosteoclastic and osteoblastic precursor cells, withacceleratedboneresorptionoccurringduringhyperthy-roid states when TSH concentrations are suppressed.Oversupplementation of thyroid replacement hormonecausesanexogenoushyperthyroidism,suppressingTSHconcentration, with direct effects on bone remodelingthatresultinboneloss. Epidemiologic studies have reported an associationbetweenbonelossandsuppressedTSHconcentrations.Afteradjustmentforfactorswithpotentialtoaffectbonedensity,totalhipBMDmeasurementsweredecreasedbyabout4%inmenandbyalmost8%inpostmenopausalwomen who had serum TSH concentrations below thereferencerange.Incontrast,TSHconcentrationswithinthereferencerangewereassociatedwithpreservationofbone mass. A large cross-sectional study of more than6000 women with self-reported thyroid disease foundthatwomenwiththelowestTSHconcentration(lessthan0.5milli-internationalunit/L)had lower forearmBMDand the highest prevalence of osteoporosis. There werenodifferencesinBMDbetweenwomenwithTSHcon-centrations of 0.5–1.49 milli-international units/L andthose with concentrations greater than 1.49 milli-inter-national units/L. Hyperthyroidism was also associatedwithanincreaseinfracturerisk,witha4.5-foldincreasein vertebral fractures and a 3.2-fold increase in nonver-tebralfractureswhentheserumTSHconcentrationfellbelow0.1milli-internationalunit/L. TherealsoappearstobearelationshipbetweenTSHconcentrations and BMD in euthyroid postmenopausalwomen. Recent data suggest a relationship betweenincreasing TSH concentrations within the therapeuticrange and BMD. The odds ratio (OR) for the medianTSH reference range (0.39–1.8 milli-internationalunits/L) versus the higher end of the reference range(1.8–4.5 milli-international units/L) was 3.4 for osteo-porosis and 2.2 for osteopenia.. Furthermore, BMD ofthetotalhipwasalmost5%lower inwhitewomenand9.7% lower in African American women in the lowestquintileofthereferencerangethaninwomenwithTSHconcentrationsinthehighestquintile. These data suggest that overt exogenous hyperthy-roidism should be avoided with thyroid replacementtherapy. Consideration should also be given to dosingadjustments when the TSH concentration falls belowthemedianofthereferencerange,particularlyinwomenwithosteopeniaorpreexistingosteoporosis.
Central Nervous System Agents Several classes of central nervous system agentshave been associated with an increased risk of fracture.
These include anticonvulsants, antidepressants, andantipsychotics.
Anticonvulsants There are several mechanisms by which anticonvul-sants might affect bone metabolism. Initially, it wasthoughtthattheanticonvulsantsthatarepotentinducersofcytochromeP450(CYP)(i.e.,carbamazepine,pheno-barbital,andphenytoin)mightincreasethemetabolismofvitaminD, leading toa reduction incalciumabsorp-tion,subsequentelevationinparathyroidhormone,andincreasedboneturnover.IthasalsobeensuggestedthatCYP induction leads to lower circulating concentra-tionsofestrogenandtestosterone,resultinginboneloss.However, many anticonvulsants that do not affect CYPmetabolism are associated with bone loss, indicatingthat other mechanisms are likely responsible. However,these other mechanisms are poorly understood. Earlydata in animals suggested that anticonvulsants directlyinhibit intestinal calcium absorption. More recently,in vitro studies suggested that anticonvulsants directlyinhibit osteoblasts, resulting in decreased bone forma-tion. Further animal and clinical studies are needed tofullyelucidatethemechanismsassociatedwithanticon-vulsant-relatedeffectsonbonemetabolism. AlthoughBMDhasbeenshowntodecreasewiththeuseofbothCYP-inducingand-noninducinganticonvul-sants,itisdifficulttoquantifytheamountandrateofbonelossbecausestudieshaveenrolledmanydifferentpatientsamples,fromchildrentoolderadults,andincludebothsingle and combination therapy. Reports indicate thatZ-scores decline in the range of−0.08 to−0.91 for thespineand−0.09to−0.90forthehip.Similarly,measuringthe relationship between fracture risk and anticonvul-santsisdifficultbecauseepilepsyitselfmaybeassociatedwithfractures. Pooleddatafrom11studiesshowedanoverallRRof2.18 for all fractures among patients with epilepsy; thisis much larger than expected considering the degree ofbone loss observed with anticonvulsants. One studyincluded in the analysis excluded seizure-related frac-turesandestimatedanRRof1.3,whichisintheexpectedrangeofobservedBMDmeasurements.Thesedatasug-gest that, although there is an increased fracture riskassociated with anticonvulsant use, it is not as high asgenerallythought,andmanyfracturesmayberelatedtothediseaseratherthanthedrug. This result is supported by observations that 34% to40% of fractures are related to trauma during seizureactivity.Furthermore,patientsexperiencinggeneralizedseizures have been reported to experience more frac-turesthanthosewithothertypesofepilepsy.Comorbidconditions and concomitant drug use are importantconsiderations when assessing fracture risk associatedwithepilepsyandanticonvulsantdrugs.Epilepsymightbe related to central nervous system neoplasms,
PSAP-VII • Women’s and Men’s Health 209 Drug-Induced Osteoporosis
developmental delay, or stroke. Such individuals mightbe more physically inactive, which can decrease bonedensity.Somediseasestatesalsoaffectbalanceandgait,rendering individuals more susceptible to falls. Thesepatients might be taking other psychotropic drugs thatfurther increase fall risk, or they might be taking drugssuch as corticosteroids that are known to have a detri-mentaleffectonbonehealth.
Antidepressants Theserotoninergicsystemappears toplayan impor-tant role in bone physiology, which has implicationsfor the effect of selective serotonin reuptake inhibitors(SSRIs) and serotoninergic tricyclic antidepressants(TCAs) on bone health. Specifically, serotonin appearsto modulate skeletal response to parathyroid hormone,possibly through receptors and transporters found onosteoblasts and osteocytes. Several studies have shownbonelossamongSSRIusers,suggestingaclinicaleffecton bone metabolism. Cross-sectional and longitudinaldata in older men and women indicate higher rates ofannual bone loss among SSRI users and about a 4% to5% lower BMD, depending on the anatomic site, com-pared with nonusers. These studies did not observe asimilardeclineinBMDamongusersofnonserotoniner-gic TCAs, suggesting that the serotoninergic activity ofSSRIshasaclinicaleffectonbonemetabolism. The association between antidepressant use andfractures is well established; however, recent evidencesuggeststhatdepressionitselfisassociatedwithdecreasedBMD and increased fracture risk. In addition to drugs,behavioralandbiologicfactorscaninteractinanindivid-ualtonegativelyaffectbonehealth.Poorhealthbehaviorssuchassmoking,alcoholuse,andphysicalinactivityarehighamongdepressedindividuals.Furthermore,changesto the hypothalamic-pituitary-adrenal axis have beenidentified in patients with depression, causing ele-vated cortisol concentrations and lower amounts ofsex hormones, both of which can result in lower BMD.Comorbidconditionsandconcomitantdrugsalsoplayarole,potentiallyincreasingtheriskoffalls.AsoutlinedinFigure1-1,whenallofthesefactorscometogether,theyresultinskeletalfragilityandexcessiveboneloadingthatultimatelyleadtofractures. Although both SSRIs and TCAs are associated withan increased risk of falls, some data suggest that SSRIscarryagreaterriskoffracturesthanTCAs.Thisfindingmight simply be caused by selection bias; clinicians arelesslikelytoprescribeaTCAforthepatientathighriskoffracturebecauseoftheadverseeffectprofile.Tricyclicantidepressants might also be used at lower doses, notfor their antidepressant activity but for management ofneuropathic pain syndromes and insomnia. Most stud-ies also do not differentiate TCAs on the basis of theirserotoninergicactivity.Nevertheless,assessmentofbone
health is important to theoverall well-being ofpatientswithdepression.
Antipsychotic Agents Similartoantidepressants,awell-establishedrelation-ship exists between antipsychotic agents and falls andfracture. The postulated biologic mechanism by whichantipsychoticagentsaffectbonephysiologyisrelatedtotheir effect on prolactin concentrations. Conventionalantipsychotics,inparticular,areknowntocauseariseinprolactinconcentration;thisinturnlowersestrogenandtestosteroneconcentrations,potentially leadingtoboneloss. As with depression, other mental illnesses mightalso represent an independent risk factor for osteopo-rosis.Schizophreniaandotherpsychoticdisorderswereassociated with higher rates of osteoporosis and fragil-ityfractures.Notably,thehighestriskofanyfracturewasreported among premenopausal women, with an RRof2.5).Theriskofhipfracturewasincreased5-foldforolderwomenand6-foldforoldermen. Data are limited on the effect of antipsychotic drugson bone density; however, available data suggest thata higher proportion of premenopausal women takingprolactin-raising agents have lower BMD than those tak-ing prolactin-sparing agents. Spinal BMD has also beenobserved tobeabout20% lower inyoungmenandpre-menopausal women taking conventional antipsychoticsthan in those taking atypical agents or in healthy youngcontrols.NoBMDdifferenceswereseenbetweenpatientstaking atypical agents and healthy controls. When con-sidered together, these data suggest that conventionalantipsychoticagentscausebonelossandthatatypicalanti-psychoticagentsdonot.However,theeffectofbehavioralfactorscouldnotbeaccountedforinthesestudies. Despite an incomplete understanding of the role ofbehavioral,biologic,anddrug-relatedfactorsinindivid-uals taking antipsychotic agents, the potential of theseagentstonegativelyaffectbonehealthshouldbeconsid-eredforallpatients.
Gastric Acid–Reducing AgentsProton Pump Inhibitors Interest in the association between proton pumpinhibitor (PPI) use and hip fracture arose from studiesthatshoweddecreasedcalciumabsorptioninpatientstak-ingPPIs.Lesspotentgastricacidagents,theH2-receptorantagonists(H2RAs),werenotobservedtohavethesameeffect. However, the studies varied in method and maynothaveusedcorrecttestingtodocumentthesepotentialdrug-drugordrug-foodinteractions.OtherdatasuggestthatPPIshaveadirecteffectonbonemetabolism.Protonpumpshavebeenidentifiedonosteoclastsandappeartobeusedduringtheexcretionofhydrogenionsforboneresorption.Inhibitionoftheseprotonpumpsmayinter-fere with the resorption process, resulting in decreasedbonedensitywithtime.
PSAP-VII • Women’s and Men’s Health210Drug-Induced Osteoporosis
Proton pump inhibitors appear to affect BMD inmen;asmallbutsignificantdifference inhipBMDwasobserved among male PPI users compared with non-users. However, similar observations were not found inwomen,suggestingthatmenareatsomewhatincreasedriskcomparedwithwomen. Observationsthatmenareatincreasedriskoffracturearesupportedbyoneepidemiologicstudy that foundastronger association between duration of PPI use andfracture in men than in women, although the odds offracture were increased in both. However, this findingwasnotsupportedbyfindingsfromotherepidemiologicstudies. Withrespecttotheoverallriskoffracture,twocase-control studies found that PPIs were used more oftenby patients with fracture. The results were consistentbetweenstudies,withtheadjustedORreportedas1.44and as 1.45 for PPI use in the previous year. Therapydurationappearedtobeanimportantfactorbecausethestrength of association increased with each year of use.AthirdstudyfoundnoassociationbetweenhipfractureandPPIusefor lessthan5years,althoughPPIusewasmorefrequentinthosewithanyfractureafter5ormoreyearsofuse(OR1.92)orhipfracture(OR1.62).After7 or more years of PPI use, the OR increased to 4.55,furthersupportingthenotionthatdurationofuseisanimportantfactorindeterminingrisk.
H2-Receptor Antagonists IncontrasttoPPIs,dataonH2RAusewereequivocal;onestudyfoundtheseagentstohaveaprotectiveeffectonBMD,whereasanothershowedasignificantassocia-tion between hip fracture and H2RA use, although thisassociationwasnotasstrongasthatobservedwithPPIs. Althoughepidemiologicdataaloneareinsufficienttoproveacausalrelationshipbetweengastricacid–reducingagents(particularlyPPIs)andanincreaseinosteoporoticfracture, gastric acid reducers may contribute to overallrisk when assessing bone health in patients using theseagents.
Thiazolidinediones Theriskoffractureappearstobeincreasedinindividu-alswithtype2diabetes,withsomesuggestionthatgoodglucosecontrolreducestheassociationbetweenthedis-ease and fracture risk. However, there is an apparentincreased risk of fracture associated with the thiazoli-dinedionesrosiglitazoneandpioglitazone; thiswasfirstidentified in randomized controlled trials examiningtheefficacyoftheseagentsinthemanagementoftype2diabetes. A meta-analysis that pooled data from 10 random-ized controlled trials showed a small but significantARI of 0.6% for all fractures in thiazolidinedione users.However, when stratified by sex, the ARI was signifi-cantly increased for women at 2.8% compared with no
differenceinriskformen.Infact,menusingthiazolidine-dioneexperiencedfewerfracturesthanthecontrolgroupofeithermetforminorsulfonylureausers.Mostfractureswere observed in the periphery rather than the hip orspine,butthismaysimplybeareflectionoftheyoungerpatientsampleintherandomizedcontrolledtrials(aver-ageage50–60years).Furthersupportingthisnotion isonecase-controlstudythatidentifiedasignificantasso-ciationbetweenhipfractureandthiazolidinedioneuseinolderwomenthatwasnotseeninnonusers. Thesex-baseddifferenceinriskobservedinthemeta-analysiswasnotconfirmedbysubsequentepidemiologicstudies, which found a significant association betweenthiazolidinedione use and fractures in both men andwomen.Oneofthesestudiesstratifieditsresultsbybothagentandsexandfoundasignificantassociationbetweenfractures and both rosiglitazone and pioglitazone inwomen,butonlypioglitazoneincreasedtheriskoffrac-tureformen.Thesedatasuggestedthatwhereasbothmenandwomentakingthiazolidinedionescouldbeatriskoffractures,pioglitazonepresentsthegreaterriskformen.However,thisfindingrequiresconfirmationthroughfur-therresearch. Both randomized controlled trials and observationalstudiesshowaconsistentdeclineinBMDovertime,sug-gestingthatthiazolidinedionesdohaveaneffectonBMD.Thiazolidinedioneusersshowmoreboneloss(about1%at the spine and 1.5% at the hip) than nonthiazolidin-edione users. These agents are thought to affect bonemetabolismthroughseveralmechanisms,whichincludeadecrease inosteoblastic function, increasingadiposityofthebonemarrow,andreducedaromataseactivity.Thislastmechanismmightpartiallyexplainwhyclinicaltrialshaveobservedwomentobeatagreaterriskthanmen.
Summary of Drug-Induced Osteoporosis There is a clear association between bone loss, fra-gility fractures, and the use of AIs in postmenopausalwomen and GnRH agonists in men. The use of GnRHagonists and DMPA by premenopausal women is asso-ciatedwithbone loss,butwhether this loss isassociatedwithincreasedriskoffractureremainstobedetermined.The issue of osteoporosis secondary to central nervoussystemactiveagentsisdifficulttoassessinquantifyingtheriskofbonelossandfracturethatcanbeattributedtodruguse.Morelikely,theoveralleffectofthesedrugsisrelatedto complex interactions between the underlying diseasestate,drugs,comorbidconditions,andlifestylebehaviors.Thesameholds true forPPIsandthiazolidinediones.Ofinterest,eachofthesetwodrugclasseshasshownsexdif-ferences, with the association between fractures and PPIusestronger inmenthanwomen.Incontrast,somedatawiththiazolidinedionessuggestagreaterriskoffractureforwomenthanmen.Furthermore,thereissomesuggestionthatpioglitazonepresentsagreaterriskthanrosiglitazone,althoughthismustbeconfirmedbyfurtherstudy.
PSAP-VII • Women’s and Men’s Health 211 Drug-Induced Osteoporosis
Managing Drug-Induced OsteoporosisNonpharmacologic Interventions Limiteddataexistregardingtheeffectofnonpharma-cologic interventionsforthetreatmentofdrug-inducedosteoporosis. Lifestyle interventions are important,particularly smoking cessation. Cigarette smoking candecrease sex hormone concentrations and interferewithcalciumabsorption,anditmayimpairosteoblasticfunction. Although not yet studied, there is a potentialadditivenegativeeffectonBMD,especiallyamongado-lescents and young women using DMPA, that couldminimize peak bone mass. High rates of smoking andalcoholuse inpatientswithmental illnessmaycontrib-utetothebonelossassociatedwithantidepressantsandantipsychotic agents, further increasing risk of fracture.Thesameholdstrueforthosetakinganticonvulsants. Physical activity among young women is importanttotheachievementofpeakbonemass.Forolderpeople,exerciseimprovesmusclestrength,balance,andcoordi-nationandassistsinmaintainingmobility.Allmedicallyfitpatientsshouldbeencouragedtoparticipateinmod-erate-intensity weight-bearing exercise such as walkingfor 30 minutes on most days of the week. Ideally, thisshould be supplemented with some resistance activitytwotimes/week. Fallsandfracturesareclosely linked,soreducingtherisk of falls can decrease the risk of fractures. Livingenvironmentsshouldbereviewedtoidentifyfactorsasso-ciatedwithfallssuchaspoorlighting,unsafebathroomsor stairs, and tripping hazards. Drug profiles should bereviewed,withdrugsknowntoincreasefallriskdiscon-tinuedwheneverpossible. Although some clinical trialshaveshownasignificantreduction in fallswithvitaminD supplementation, this has not been consistent in allstudies.
Pharmacologic Interventions Severalalgorithms(Figures1-1,1-2,and1-3)stresstheimportanceofensuringadequatecalciumandvitaminDintakeinpatientstakingdrugsassociatedwithincreasedriskoffracture.Adequateintakeisbestachievedthroughdietary intake; however, it is estimated that the dailyintake of the average American is only 600 mg of ele-mentalcalcium.Thislevelofconsumptioniswellbelowthe recommended daily allowance of 1300 mg for ages14–18,1000mgforages19–50,and1200mgforthoseolder than 50. Vitamin D deficiency is common, espe-cially among older individuals, those living in northernlatitudesorininstitutions,orthosewithpoornutrition. Vitamin D status is diagnosed by measuring25-hydroxyvitamin D concentrations, with generaldefinitions of deficiency (less than 20 ng/mL), insuf-ficiency (21–29 ng/mL), and sufficiency (greater than30 ng/mL). Although this test is expensive and subjectto variability between laboratories, it is appropriate in
individualstakingdrugsknowntoaffectvitaminDcatab-olismorthoseatriskofvitaminDdeficiency. There is considerable controversy regarding the cur-rently recommended daily allowance of vitamin D. TheNational Osteoporosis Foundation recommends 800–1000units/dayforindividuals50yearsorolder.However,many clinical experts feel that up to 2000–4000 units/dayarenecessarytomaintainastateofvitaminDsuffi-ciency. Patients taking CYP-inducing anticonvulsantsmayrequirethehigherdoseof4000units/daytomain-tain25-hydroxyvitaminDconcentrationsgreaterthan30mg/mL.Muchhigherdosagesaresometimesprescribedfor those with documented vitamin D deficiency, suchasergocalciferoloneortwo50,000-unitcapsulesweeklyfor1–3months.Becauseofthelonghalf-lifeofvitaminD, about 3 months of therapy are required before newsteady-state 25-hydroxyvitamin D concentrations areobserved. Severaldrugshavebeenapprovedforthetreatmentofosteoporosis,butonlythebisphosphonates,teriparatide,anddenosumabhavebeenstudiedinthemanagementofdrug-induced osteoporosis. The bisphosphonates havebeenstudiedinglucocorticoid-inducedosteoporosisandbone loss secondary to hormonal therapies used in thetreatmentofbreastandprostatecancer.Denosumabhasrecentlybeenapprovedforthetreatmentofosteoporosisrelatedtoandrogendeprivationtherapyofprostatecan-cer.Teriparatidehasbeenstudiedinthemanagementofglucocorticoid-inducedosteoporosis.
Specific GuidelinesOsteoporosis Induced by AIs and GnRH Agonists in Women An approach to the management of osteoporosisrelatedtotheuseofAIsisillustratedinFigures1-2and1-3.Whenwomenhaveadditionalriskfactorsorestab-lishedosteoporosis,thebisphosphonatesareconsideredfirst-linetherapy,eitherorallyorintravenously,althoughthis recommendation is based on limited information.Risedronate, administered as 35 mg once weekly, wasstudiedin87postmenopausalwomenwithbreastcancerduringa2-yearperiod.OnlyafewwomenwerereceivinganAI(i.e.,five intheplacebogroupandsix intherise-dronate group). At 24 months, BMD at the spine haddeclined by almost 5% in the placebo group comparedwitha2%lossintherisedronategroup. More recently, zoledronic acid has been studiedbecauseofdatasuggestingthatitexhibitsantimetastaticandantitumorproperties inaddition to itspreservativeeffectonbonedensity.Dataonthisagent’seffectonbonelossfromhormonaltherapiescomeprimarilyfromlargerandomized controlled trials that examined the effectof zoledronic acid on disease-free survival from breastcancerinpre-andpostmenopausalwomentreatedwithGnRH agonists and/or AIs. The current data are onlyshortterm,withpatientsmonitoredfornomorethan1
PSAP-VII • Women’s and Men’s Health212Drug-Induced Osteoporosis
year.Zoledronicacid4mgadministeredevery6monthsresultedinsignificantgainsinBMDof3%to5%amongpostmenopausalwomenreceivingadjuvanttherapywithAIsinearlybreastcancer.However,itisunknownwhetherthistranslatestoareductionofAI-relatedfracturesinthispopulation.Thedoseusedinthesestudieswasalsomuchhigher thantheannualdoseof5mgofzoledronicacidindicatedfornon–drug-inducedosteoporosis. A5-yearstudyofpremenopausalwomenwithbreastcancer undergoing ovarian suppression plus hormonaltherapy found that BMD stabilized after 36 months ofintravenous zoledronic acid 4 mg administered every 6months. At 5 years, those receiving goserelin plus hor-monal therapy with zoledronic acid had an increase ofabout4%inBMDatboththespineandhipcomparedwith those receiving placebo. However, this did nottranslateintoadecreaseinfracturesbecausethefracturerateswere1.1%forbothgroups.Similarly,nosignificantdifferences were observed when stratified by hormonaltherapy,withARIestimatedat0.7%forthegoserelinplusanastrozolegroupnottreatedwithzoledronicacidcom-paredwithtreatment.
Osteoporosis Induced by GnRH Agonists in Men The general approach to managing osteoporosisrelatedtoantiandrogentherapyinmenwithprostatecan-cerisbasedonBMDand/ortheoccurrenceoffragilityfractures. Men who have experienced fragility fracturesshould receive counseling regarding adequate calciumand vitamin D intake and be treated with bisphospho-nates or denosumab. Men who have not experiencedfracturesshouldhavetheirBMDmeasured.ThosewithosteoporosisshouldreceivecalciumandvitaminDsup-plements, as well as treatment with bisphosphonates.If the T-score indicates osteopenia, patients should betreated with calcium and vitamin D supplements, withBMD repeated in 6–12 months. Adequate calcium andvitamin D intake should also be ensured in men forwhom BMD indicates normal bone, with repeat BMDtestingin2years. SimilartoosteoporosisrelatedtoAIsandGnRHago-nistsinwomenwithbreastcancer,thebisphosphonatesplayanimportantroleinthemanagementofosteoporosisrelatedtoantiandrogentherapyinmen.Oralalendronatehasbeenstudiedinmenreceivingandrogendeprivation
PSAP-VII • Women’s and Men’s Health 213 Drug-Induced Osteoporosis
therapy.Menreceivingalendronate70mgonceweeklyshowedalmosta4%increase inBMDof thespineand1.6%atthehipat1year.Incomparison,meninthepla-cebogrouphadadecline inBMDofabout1%inboththespineandhip. Theeffectofzoledronicacidhasalsobeenstudiedinafewmenreceivingantiandrogentherapyforthetreat-mentofnonmetastaticprostatecancer.Zoledronicacidadministeredatadoseof4mgevery3monthsfor1yearresulted in a 5.6% increase in BMD compared with alossof2%for thosenotreceiving therapy.A12-monthstudy of 40 men randomized to either a single 4-mgdoseofzoledronicacidorplaceboobservedanincreasein spinal BMD of 4% in men assigned to active treat-mentcomparedwithareductionof3%inmenreceiving
placebo.Atpresent,nodataexistregardingtheeffectofzoledronicacidontheriskof fracture.Alargerandom-izedcontrolledtrialthatiscurrentlyexaminingtheeffectof zoledronic acid on disease outcomes in men withmetastatic prostate cancer should provide health careproviders with more information regarding the role ofthisagentonbonehealth. Denosumab,amonoclonalantibodyagainstRANKL,was recently studied for the treatment of osteoporo-sis secondary to androgen deprivation therapy in menwith prostate cancer. This randomized controlled trialcomparing denosumab 60 mg subcutaneously every 6months with placebo observed significant increases inBMDofthelumbarspine(7.9%),totalhip(5.7%),anddistal radius (6.9%) after 3 years of active treatment.
PSAP-VII • Women’s and Men’s Health214Drug-Induced Osteoporosis
Vertebralfractureswerealsosignificantlyreducedinthedenosumab treatment group, representing an absoluteriskreductionof2.4%andanumberneededtotreatof42.
Glucocorticoid-Induced Osteoporosis Although glucocorticoid-induced osteoporosis isnot the focus of this chapter, recent data with teripara-tidewarrantdiscussion.Teriparatide20mcgoncedailywascomparedwithalendronate10mgoncedailyfor18months in 428 women and men with glucocorticoid-inducedosteoporosis.AnincreaseinBMDofthelumbarspine was observed in both treatment groups, but theincreasewassignificantlygreaterintheteriparatidegroup(7.2%)thaninthealendronategroup(3.4%).Therewerealso fewer vertebral fractures in the teriparatide group(0.6%) than in the alendronate group (6.1%), with anabsoluteriskreductionof5.5%andanumberneededtotreatof18(i.e.,18patientswithglucocorticoid-inducedosteoporosisneedtobetreatedwithteriparatideduringan 18-month period to prevent one fracture). The inci-dence of nonvertebral fractures did not differ betweenthegroups. Current guidelines, which were developed beforethis study, recommend the use of oral bisphosphonatesas first-line therapy in the treatment of glucocorticoid-induced osteoporosis. These recommendations maychange when longer-term data comparing teriparatidewiththebisphosphonates,aswellasacost-effectivenessevaluationofeachapproach,becomeavailable.
Osteoporosis Related to Other Drugs Information is limited regarding the managementof osteoporosis related to drugs other than AIs, GnRHagonists, and glucocorticoids. Lifestyle counseling isimportant regardless of the suspect drug, particularlyfor patients with depression and psychotic disorders.Ensuring appropriate calcium and vitamin D intake,whether by diet or supplements, is equally important.Vitamin D intake might need to be higher in patientsreceiving anticonvulsant agents with CYP-inducingactivity,withtheseindividualspotentiallyrequiringdos-ages of 2000–4000 units/day to maintain vitamin Dsufficiency. Overt hyperthyroidism should be avoidedin patients receiving thyroid replacement therapy, andconsideration should be given to maintaining the TSHconcentrationwithinthemedianofthereferencerange. Althoughthedegreeofincreasedriskrelatedtolong-termuseofPPIs,H2RAs,andthiazolidinedioneshasnotbeendefined,theseagentsshouldbeusedcautiously inpatients at high risk of osteoporosis. Based on existingdata,H2RAsmaybeamoreappropriatefirst-linechoicewhenmorepotentgastricacidreductionisnotnecessary.If PPIs are needed, their use should be reassessed on aregularbasis.Considerationshouldbegiventotheuseofotheroralantihyperglycemicagentsinpatientswithtype
Role of the Pharmacist Pharmacists providing direct or managed patientcareshouldbeawareoftheriskfactorsforosteoporosisin patients taking agents associated with drug-inducedosteoporosis.Inaddition,pharmacistsneedtobefamiliarwithdrugssuspectedofcausingdrug-relatedosteoporo-sisandtheriskoffractureassociatedwithspecificagents.Allpatientstakingdrugsassociatedwithanincreasedriskoffractureshouldreceiveabaselineriskassessment,andthoseatthehighestriskshouldbereferredforBMDtest-ingbyDEXA. Based on estimates of the NNH for AIs and GnRH,drug-inducedosteoporosishas thepotential tocontrib-utesignificantlytothenumberof fracturesexperiencedbywomenandmenwithhormonalcancers.Pharmacistsworkinginoncologyclinicsshouldincludeassessmentofbonehealthaspartoftheirdirectpatientcareactivitiesandbeinvolvedintreatmentdecisionsforthepreventionandtreatmentofosteoporosis.Similarly,healthcarepro-viders working in other specialty areas (e.g., neurology,psychiatry, diabetes) should be familiar with drugs thatmightincreasetheriskofosteoporosisintheirpatients.Pharmacists also play an important role in educatingotherhealthcareprovidersaboutdrug-inducedosteopo-rosis,particularlythoseworkinginspecialtyareaswhereosteoporosis is not a major focus of care, and helpingtheseprovidersidentify,prevent,andtreatdrug-inducedosteoporosis.
Conclusion Pharmacists should play a prominent role in recog-nizing the potential for drug-related osteoporosis andpreventing associated fractures. Given the morbidityandmortalityofosteoporoticfractures,interventionstoreducethisriskmayhaveimportantpublichealthbene-fits.Pharmacistsshouldacquireageneralunderstandingofdrug-inducedosteoporosisandidentifypatientswhorequire treatment. Prevention of fractures is the overallgoal of therapy and requires effective communicationwithpatientsandotherhealthcareprovidersalike.
Annotated Bibliography1. Heneghan C, Badenoch D. Evidence-Based Medicine
Muchoftheevidencetodateregardingdrug-inducedosteoporosiscomesfromepidemiologicdataratherthanrandomized controlled clinical trials. This book is anessential reference for clinicians who need to criticallyappraisethisliteraturesothattheycancommunicaterisk
PSAP-VII • Women’s and Men’s Health 215 Drug-Induced Osteoporosis
to patients. The first chapter guides the reader throughtheprocessofformulatingtheappropriateclinicalques-tionusingthePICOapproach(i.e.,Patientorproblem,Intervention, Comparison, Outcome). The chapteraddressing appraisal of studies on harm and etiology isparticularly useful in the context of drug-induced dis-ease,explainingthedifferencesbetweencase-controlandcohort studies, the correct interpretation of the variousexpressionsofrisk,andthemostappropriateexpressionsforthetypeofstudy.Examplesofhowtocorrectlyinter-pretRRincrease,ARI,andNNHarealsoprovided.
2. FRAX. The World Health Organization fracture riskassessment tool. Available at www.shef.ac.uk/FRAX.AccessedMay26,2010.
ClinicianswillfindtheFRAXassessmenttoolhelpfulwhen counseling patients about their risk of osteopo-rosis-related fractures. The easy-to-use FRAX tool isappropriate for risk assessment in women and menolderthan50whoarenottakingosteoporosisprescrip-tiondrugssuchasbisphosphonates.TheFRAXtoolwasdevelopedtoevaluatethefractureriskofpatientsbasedonclinicalriskfactors,anditalsoallowstheuseoffemo-ralneckBMDmeasurementswhenavailable.TheFRAXmodels were developed from population-based cohortsin several countries, including North America. The toolprovides 10-year probability of hip fracture, as well as10-yearprobabilityofamajorosteoporoticfractureinthespine,forearm,shoulder,orhip.Theseresultsaredepictedinafigurethatcanbegiventopatients.FortheU.S.cal-culation, the correct race/ethnic group (white, AfricanAmerican, Asian American, and Hispanic American)must be selected. Clinical risk factors included in theprobabilitycalculationincludeBMI(basedonheightandweight), age, sex, previous fracture, parental history ofhipfracture,smokingstatus,currentalcoholuse(greaterthan 3 units/day), rheumatoid arthritis, current or pastuseoforalglucocorticoids,andselect secondarycausesofosteoporosis.TheFRAXtooldoesnot includedrugsother than glucocorticoids known to decrease BMD orincreasefracturesintheriskassessment.
3. Kanis JA, Borgstrom F, De Laet C, Johanasson H,Johnell O, Jonsson B, et al. Assessment of fracture risk.OsteoporosInt2005;16:581–9.
Thisreviewprovidesasummaryoftheimportanceofconsideringclinicalriskfactorsforosteoporoticfracturesin addition to results of BMD testing. The literature forthese factors that provides information on fracture riskabove and beyond that of BMD (e.g., age, prior fragil-ityfracture,prematuremenopause,familyhistoryofhipfracture) is concisely summarized, with a good discus-sionontheprosandconsofusingvariousexpressionsofrisk.Inparticular,areviewof theuseofRRiswellpre-sented. There is also a discussion about integrating thepresenceorabsenceofriskfactorswithBMDtoestimatethe probability of fracture over time, and what appro-priateinterventionthresholdsshouldbewithrespecttotreatment.
4. Ioannidis G, Papaioannou A, Hopman WM, Akhtar-Danesh N, Anastassiades T, Pickard L, et al. Relationbetween fractures and mortality: results from theCanadian Multicentre Osteoporosis Study. CMAJ2009;181:265–71.
This cohort study evaluated the 5-year relationshipbetween incident fractures and mortality in 7753 par-ticipants older than 50 years. The cohort was drawnfrom the Canadian Multicentre Osteoporosis Study, alarge national prospective study of ambulatory commu-nity-dwellingpatientswithosteoporosis.Fractureswereidentified by questionnaires mailed annually to all par-ticipants. Reported fractures were confirmed by reviewof medical records. The primary outcome of the studywastheassociationbetweenfracturesandmortality.Theabsolutemortalityratewas8.1%fornofractures,16%forvertebral fractures, and 23.5% for hip fractures, result-ing inanARIof7.9%forvertebral fracturesand15.4%for hip fractures. This translates into one excess deathforevery12vertebralfracturesoreverysixhipfractures.Comparedwithparticipantswhohadnofracture inthe5-year follow-up, those who experienced vertebral frac-ture during the second year were at increased risk ofdeath(hazardratio[HR]2.7,95%CI,1.1–6.6).Theasso-ciationwasstrongest inwomen,withtheARIfordeathinwomenwithvertebralfracturesbeing8.8%.Vertebralfracturewasnotassociatedwithmortalityinmen.Thosewith hip fracture in the first year were also at increasedriskofdeath(HR3.2;95%CI,1.4–7.4),againwiththestrongest association being in women, with an ARI of16.9%. As with all cohort studies, residual confoundersandbiascouldpotentiallyinfluencetheresults.However,these findings confirm the relationship between frac-ture and mortality in ambulatory community-dwellingwomenwithosteoporosis.Furtherresearchisneededtoconfirmtheassociationformenandtoexaminetherela-tionshipbetweenmultiplefracturesandmortality.
5. ATACTrialistsGroup.Resultsof theATAC(Arimidex,Tamoxifen, Alone or in Combination) trial after com-pletion of 5 years’ adjuvant treatment for breast cancer.Lancet2005;365:60–2.
Thisrandomizedcontrolledtrialpresentsthecompara-tiveefficacyandtolerabilityofanastrozoleandtamoxifenafteramedianfollow-upof68monthsin9366postmeno-pausal women with localized breast cancer. Treatmentwithanastrozoleresulted insignificantly longerdisease-freesurvivalandtimetorecurrence.Fractureratesweremonitored throughout the study to determine whethera decline in functional estrogen concentration wouldbe associated with increased fracture risk. Fracturesoccurredin11%ofwomentakinganastrozoleand7.7%inthetamoxifengroup,representinganARIof3.3%.ThisresulttranslatestoanNNHof30.Thatis,oneexcessfrac-tureoccursforevery30womentreatedwithanastrozolecomparedwithtamoxifen.Thesedatainformcliniciansoftheriskof fracturesassociatedwithanastrozoleuseandtheimportanceofassessingbonehealthforwomenusingthese drugs in the treatment of breast cancer. SimilarresultshavebeenobservedwiththeotherAIs, letrozoleandexemestane,indicatingaclasseffect.
PSAP-VII • Women’s and Men’s Health216Drug-Induced Osteoporosis
6. Reid DM, Doughty J, Eastell R, Heys SD, Howell A,McCloskey EV, et al. Guidance for the management ofbreast cancer treatment–induced bone loss: a consen-suspositionfromaUKExpertGroup.CancerTreatRev2008;34:S3–S18.
Thisconsensusstatementwasdevelopedbyagroupofcliniciansfromarangeofmedicaldisciplines(e.g.,oncol-ogy, endocrinology, rheumatology) for the purpose ofproviding guidelines for the management of bone lossinducedbyAIsandGnRHagonistsduringthetreatmentofbreastcancer.Asystematic literaturereviewwascon-ducted to identify the evidence around the impact ofAIsandovariansuppression therapyonbonehealth,aswellasthecurrentliteratureregardingtreatmentofboneloss secondary to these therapies. This article providesagoodoverallreviewoftheeffectofeachbreastcancertreatmentonbonehealth.Theliteratureregardingtreat-ment of bone loss was used to develop two algorithms:oneforpremenopausalwomenwithovariansuppression,withorwithoutconcomitantAItherapy;andasecondforpostmenopausalwomenreceivingAIs.Althoughthecon-sensusstatementdoesnotprovidethelevelofevidencefor each algorithm, it does incorporate the best avail-ableevidencetodateandprovideclinicianswithausefulguideinmakingtreatmentdecisions.However,theserec-ommendations might require revision, particularly thepostmenopausalAIalgorithm,asnewdatabecomeavail-able (e.g., the data discussed in annotated bibliography7). In the current treatment algorithm, treatment deci-sionsfortheuseofbisphosphonatesaremadebasedontheT-score;thismightchangeoncethelong-termeffectson fracture rates of early bisphosphonate therapy, espe-ciallyintravenousregimens,aredetermined.
7. Brufsky A, Bundred N, Coleman R, Lambert-Falls R,MenaR,HadjiP,etal.Integratedanalysisofzoledronicacidforthepreventionofaromataseinhibitor-associatedbone loss in post-menopausal women with early breastcancerreceivingletrozole.Oncologist2008;13:503–14.
Thisarticlepoolstheresultsoftheinterim12-monthanalysis of two large randomized controlled trials(Z-FAST and ZO-FAST) investigating the effect ofintravenouszoledronicacid4mgevery6monthsincom-bination with letrozole for the treatment of early breastcancer. In both studies, women were randomized in anopen-labelfashiontoreceivezoledronicacidearlyintheirletrozole treatment regimen or to delay treatment untiltheirT-scorebecamelessthan−2.0orafragilityfractureoccurred. The primary purpose of the two randomizedcontrolledtrialswas todeterminewhetherbisphospho-natetreatmentimprovesdisease-freesurvivalandreducesdistantmetastasesduringa5-yearperiod.TherewasalsointerestintheeffectofzoledronicacidonAI-associatedboneloss,withfractureastheprimaryoutcomeforthissubstudy. The results of this interim analysis evaluatethesurrogateendpointofbonelossinacombinedtotalof 1667 women by monitoring BMD using DEXA at 6months and 12 months. Six-month results showed thatthosereceivingearlyzoledronicacidexperiencedasmallbutsignificantincreaseinBMDofthelumbarspineand
total hip, whereas the delayed group lost bone at bothsites.At12months,BMDwas5.2%higheratthelumbarspineand3.5%higheratthehipintheearlyzoledronicacidgroup.Participantsinbothtreatmentarmshadsimi-larbaselineBMDandriskfactors,sothedifferencecanbeattributedtozoledronicacid.Althoughthesedatasuggestbenefits in BMD, it remains to be determined whetherthisinterventionwillleadtoareductioninfractures.
8. Gnant M, Mlineritsch B, Schippinger W, Luschin-EbengreuthG,PostlbergerS,MenzelC,etal.Endocrinetherapy plus zoledronic acid in premenopausal breastcancer.NEnglJMed2009;360:679–91.
ThisarticlepresentstheresultsoftheAustrianBreastand Colorectal Cancer Study Trial 12, which wasdesignedtoevaluatetheefficacyof3yearsoftreatmentwith ovarian suppression plus anastrozole or tamoxi-fen with or without zoledronic acid in premenopausalwomenwithearlybreastcancer.Theprimaryoutcomeofthisrandomizedcontrolledtrialistodeterminewhetherthe addition of zoledronic acid to adjuvant endocrinetherapy improves disease-free survival. However, thereisagaininterest intheeffectofzoledronicacidonbonehealth. A previous publication (Gnant M, MlineritschB, Luschin-Ebengreuth G, Kainberger F, Kässmann H,Piswanger-Sölkner JC, et al. Adjuvant endocrine ther-apy plus zoledronic acid in premenopausal womenwith early-stage breast cancer: 5-year follow-up of theABCSG-12 bone-mineral density substudy. LancetOncol 2008;9:840–9) reported the effect of zoledronicacidonBMD,showingsignificantimprovementsinbonedensity with zoledronic acid. However, fracture rateswere not reported. This article presents fracture data inthereviewofadverseeffects.Inthegoserelinplustamox-ifen group, fractures occurred in 1.3% of patients notreceiving zoledronic acid and in 0.9% with zoledronicacid.Resultsweresimilarforthegoserelinplusanastro-zole treatment arm, with 1.6% of patients experiencinga fracture without zoledronic acid compared with 0.9%without bisphosphonate. The absolute risk reduction infracturewassmallinbothendocrinegroups,at0.4%forthetamoxifengroupand0.7%fortheanastrozolegroup.Thisresultisnotunexpectedgiventhestudypopulationofpremenopausalwomenwithanaverageageofabout45years,mostofwhomhadbaselineBMDwithinthenor-mal range for their age. Because bone loss at this age isexpected to cause fractures later in life, 3 years is prob-ably insufficient to determine whether treatment withzoledronic acid reduces fracture risk in this population.However,thisagentmayhaveotherbeneficialeffectsforpremenopausalwomenwithearlybreastcancer.
This retrospective cohort study is the largest toexamine the risk of osteoporosis and fractures in menreceivingandrogendeprivationtherapyforprostatecan-cer. Two large databases were used to identify men 66yearsorolderwithadiagnosisofprostatecancerwhohadreceivedaGnRHagonistorhadundergoneorchiectomy
PSAP-VII • Women’s and Men’s Health 217 Drug-Induced Osteoporosis
6monthsormoreafterdiagnosis.Theprimaryoutcomewas fracture occurrence. Overall, for men receiving aGnRHagonist,theNNHwas28,translatingintoalmost3000excessfracturesperyearbasedontheannualinci-denceofprostatecancerandcurrentpatternsofGnRHuse. Dose-response and age effects were also observed,withfractureincidenceincreasingwithboththenumberofdosesreceivedandincreasingage.Althoughthisstudywas unable to differentiate between fractures related tobonemetastases,nodeclineinfractureriskwasobservedwhenearlyand latestagediseasewerecompared,whenfractures are more likely to occur. The findings of thisstudy support the results of smaller studies and show arelationship between the use of GnRH agonists and anincreasedfractureriskinmenwithprostatecancer,out-liningtheneedforassessingbonehealthinmenreceivingantiandrogentherapies.
10. Michaelson MD, Kaufman DS, Lee H, McGovern FJ,Kantoff PW, Fallon MA, et al. Randomized controlledtrialofannualzoledronicacidtopreventgonadotropin-releasing hormone agonist-induced bone loss in menwithprostatecancer.JClinOncol2007;25:1038–42.
Aprevioussmall,randomizedcontrolledtrialshowedthatzoledronicacidreducesGnRHagonist–relatedbonelosswhenadministeredevery3monthstomenwithnon-metastatic prostate cancer. This randomized controlledtrialcomparedtheeffectonBMDofa4-mgintravenousdosegivenannuallyversusplaceboin40menwithnon-metastaticprostatecancerandreceivingGnRHagonists.MeasurementsofBMDweretakenatbaselineandafter3months,6months,9months,and12monthsoftherapy.At12months,BMDofthelumbarspinehaddecreasedby 3.1% in the placebo group but increased by 4.0% inthe group receiving zoledronic acid. Total hip BMDdeclinedby1.9%intheplacebogroupcomparedwitha0.7% increase in those receiving zoledronic acid. Thesedata suggest that one dose of zoledronic acid preventsGnRHagonist–relatedbonelossinmenwithearlypros-tate cancer. The data are also consistent with data fromothersmallstudiesevaluatingtheeffectofzoledronicacidon BMD in men with more advanced prostate cancer.However,theeffectofzoledronicacidonfractureratesinmenreceivingGnRHagonistsremainstobedetermined.
11. Smith MR, Egerdie B, Hernandez Toriz H, Feldman R,TammelaTLJ,etal.Denosumabinmenreceivingandro-gen-deprivation therapy for prostate cancer. N Engl JMed2009;361:745–55.
TheDenosumabHormoneAblationBoneLossTrial(HALT)comparedtheeffectofdenosumab(60mgsub-cutaneously every 6 months) with placebo on fractureratesinmenreceivingandrogendeprivationtherapyforprostatecancer.Denosumabisafullyhumanmonoclonalantibody,activeagainstRANKL,which isan importantmediatorofosteoclastformationandfunction.Ofthe734menrandomizedtoeachtreatmentgroup,445(60.6%)and467(63.6%)completedthe3-yearstudyinthepla-ceboanddenosumabgroups,respectively.At24months,the placebo group showed a 1% decline in BMD at the
lumbar spine compared with a 5.6% increase amongthe denosumab group. Significant increases in BMDwere also observed in the total hip, femoral neck, anddistal radius with denosumab compared with placebo.Treatmentwithdenosumabalsoresultedinasignificantreduction in new vertebral fractures at 12 months, 24months,and36months.After3yearsof treatment, theabsolute risk reduction for new vertebral fractures was2.4%, translating into a number needed to treat of 42.Thatis,onefracturewillbepreventedforevery42menreceivingdenosumabtherapyduringa3-yearperiod.Nosignificantdifferences in fracture rateswereobservedatother anatomic sites, including the hip or distal radius,despitedemonstrated increases inBMD.Thesefindingssupporttheuseofdenosumabforthetreatmentofosteo-porosis related to androgen deprivation therapy in menwithprostatecancer.
This opinion piece from the ACOG Committee onAdolescent Health Care and Gynecologic Practiceaddresses the controversial issue of long-term DMPAuse in young women, particularly adolescents. The evi-dence regarding bone loss associated with DMPA useis thoroughly reviewed and concisely summarized, asare the findings of BMD recovery after discontinuationof DMPA and the limited data regarding fractures. TheFDArecommendationtolimituseto2yearsortocon-duct BMD testing by DEXA after that time is disputedonthebasisofevidencesuggestingthatbonelossslowswith longer-term DMPA use and is reversible on dis-continuation; DMPA use is therefore unlikely to placewomenatincreasedriskoffractureinfutureyears.Basedontheirreviewoftheliterature,theauthorsconcludethatDMPAisasafeandeffectivemeansoflong-termcontra-ceptioninwhichtherisksofDMPA-associatedbonelossmustbeweighedagainsttherisksofpregnancy,particu-larlyinadolescents.Theimportanceofcounselingaboutthe benefits of adequate calcium and vitamin D intakeandphysicalactivityisstressed.Theuseofestrogensup-plements to prevent DMPA-associated bone loss is notrecommended.
This meta-analysis pools data from 11 epidemiologicstudiestodifferentiatetheeffectofseizuresandanticon-vulsant-relatedbonelossonfracturerisk.Whenpatientswith epilepsy were compared with those without, largeandsignificantRRincreaseswereobservedforanyfrac-ture (220%; RR 2.2; 95% CI, 1.9–2.5), for hip fracture(530%;RR5.3;95%CI,3.2–8.8),andforlumbarspinefracture(620%;RR6.2;95%CI,2.5–15.5).Thisriskwasmuchhigherthanexpectedbasedonreportsofanticon-vulsant-related bone loss. Of note, however, is that theARIcouldnotbedeterminedfromthecase-controlstudyandmanyofthecohortstudiesinthisanalysis.RelativeriskincreaseisusuallymuchlargerthantheARIandmustbe considered in the appropriate context. Nevertheless,
PSAP-VII • Women’s and Men’s Health218Drug-Induced Osteoporosis
a significant association was identified, indicating thatpatients with epilepsy were at greater risk of experienc-ing a fracture. To address this issue, the proportion offracturesrelatedtoseizureactivitywasidentifiedinfourstudies,withanestimated35%offracturesrelatedtosei-zures.ThesedatasuggestthattheRRincreaseobservedamong patients taking anticonvulsants was partiallycausedbyfracturesoccurringsecondarytoseizures.Evenwhen this is considered, an increased risk of fracturescontinues to be observed among anticonvulsant usersandisconsistentwiththedegreeofbonelossassociatedwiththeuseofanticonvulsantdrugs.
14. RichardsJB,PapaioannouA,AdachiJ,JosephL,WhitsonH, Prior JC, et al. Effect of selective serotonin reup-take inhibitors on the risk of fracture. Arch Intern Med2007;167:188–94.
This prospective cohort study of 5008 commu-nity-dwelling adults age 50 or older determined theassociationbetweendailySSRIuseandincidentclinicalfragilityfractures.Otheroutcomesofinterestwerepreva-lentfallsreportedatthetimeofenrollmentandbaselinedifferencesinBMD.Incident clinical fracturesrefertothenumber of new fractures reported during the study fol-low-upperiod,whereasprevalent fallsrefertothenumberof patients reporting a fall in the month before studyenrollment.ParticipantsweredrawnfromtheCanadianMulticentre Osteoporosis Study, a large population-basedlongitudinalstudy,andweremonitoredfor5years.Use of SSRIs was identified by interview, and incidentclinical fracturesweretrackedthroughannualquestion-naires sent to all participants. All self-reported incidentclinical fragility fractures were confirmed radiographi-cally.DailySSRIusewaslow,reportedinonly137studysubjectswithanaverageageofabout65years.However,after adjusting for potential confounding factors, it wasassociatedwithasubstantiallyincreasedriskofincidentclinicalfragilityfractures(HR2.1).TheHRcanbeinter-pretedsimilarlytotheRR,sotheRRincreaseoffractureisabout210%withdailySSRIuse. This compareswiththeinsignificant20%RRincreaseestimatedforthe162participants reportingdailyTCAuse(HR1.2;95%CI,0.7–2.2). Of note, an RR increase should not be inter-preted in the same manner as the ARI. Depression,measuredbyvalidatedinstruments,wasnotfoundtobeassociatedwithfracturerisk ineitherunivariateormul-tivariate analysis. Baseline evaluation of prevalent fallsrevealedadoublingoftheadjustedoddsoffallingamongdailySSRIusers(OR2.2;95%CI,1.4–3.5).DailySSRIuse was also associated with lower BMD of about 4%at thehipand2.4%at the lumbarspinecomparedwithnonusers.ThesedataindicatethatSSRIuseisassociatedwith more falls, greater bone loss, and increased risk offractures.
Thisretrospectivecohortstudyofmorethan300,000patients is the largest to estimate the risk of fracture in
patients with mental illness compared with the generalpopulation. Specifically, age- and sex-specific fractureriskswereestimatedinpsychoticillnesses,nonpsychoticaffective disorders, and all other psychiatric conditionsusing theUKGeneralPracticeResearchDatabase. Thisdatabasecapturestheprimarycarerecordsofabout5%of the UK population, recording prescriptions, clinicalevents,specialistcare,andhospitaladmissions.Theriskofanyfracturewasobservedtobehighestamongwomen18–44yearsoldwithadiagnosisofpsychoticillness(RR2.53;95%CI,1.49–4.32).Likewise,theriskofanyfrac-ture among patients with a psychotic disorder using aprolactin-raisingantipsychoticdrugwashighestinyoungwomen(RR2.74;95%CI,1.50–4.98),althoughriskwassignificantly increased in women of all age groups. Notunexpectedly,theriskofhipfractureincreasedwithageand was highest in both women and men aged 45–74yearswithpsychoticdisorders(RR5.12;95%CI,2.73–9.63, and RR 6.41; 95% CI, 2.55–16.11, respectively).The same pattern was observed for users of prolactin-raisingantipsychoticagents.Despitealargesamplesize,thisstudywasnotabletoidentifyandaccountforimpor-tantconfoundingvariablessuchasalcoholuse,cigarettesmoking, or physical inactivity. In addition, duration oftreatment with prolactin-raising antipsychotic drugscouldnotbedetermined.However,thesedatapointtoanincreasedriskof fracture in individualswithpsychiatricillness,particularlythosewithdiagnosesofpsychoticdis-orders.Furthermore,itidentifiesarelationshipbetweenantipsychotic agents with prolactin-raising effects andfracturethatwarrantsfurtherstudy.
16. YangYX,LewisJD,EpsteinS,MetzDC.Long-termpro-ton-pump inhibitor therapy and risk of fracture. JAMA2006;296:2947–53.
This study is the largest of three case-control studiesexaminingtheassociationbetweenPPIsandhipfracture.UsingtheUKGeneralPracticeResearchDatabase,13,556hipfracturecasesoccurringatleast1yearafterinitiationofPPI therapywerematchedwith135,386controls forsex,yearofbirth, indexdate,anddurationof follow-up.Adjustmentforpotentialconfoundingfactors(e.g.,drugs,othermedicalconditionsthatcouldincreasehipfracturerisk)wasmadeinthestatisticalanalysis.Becausethiswasacase-controlstudy,theassociationbetweenPPIuseandhipfractureisexpressedastheadjustedOR.Overall,theoddsofexperiencingahipfracturewithPPIswassignifi-cantlyincreasedat1year(OR1.22;95%CI,1.15–1.3),with the magnitude of effect increasing with increaseddurationoftherapy(ORat2years1.41;95%CI,1.28–1.56,ORat3years1.54;95%CI,1.37–1.73,andORat4years1.59;95%CI,1.39–1.80).Sexdifferenceswerealsonoted,withtheassociationbetweenPPItherapyandhipfracturestrongerinmen(OR1.78;95%CI,1.42–2.22)thaninwomen(OR1.37;95%CI,1.22–1.53),althoughasignificantrelationshipwasobservedforboth.Thefind-ingofasignificantassociationbetweenPPIuseandhipfracture is consistent with other studies, although thereisinconsistencyregardingsexdifferencesandincreasingriskwithdurationoftherapy.Protonpumpinhibitorsareamong the drugs most commonly prescribed to older
PSAP-VII • Women’s and Men’s Health 219 Drug-Induced Osteoporosis
individuals, many of whom have several risk factors forhipfracture.TheresultsofthisstudyandothersoutlinetheneedforclinicianstoconsidertheindicationsforPPItherapyandwhetherlong-termuseiswarranted.
17. Loke YK, Singh S, Furberg CD. Long-term use of thi-azolidinediones and fractures in type 2 diabetes; ameta-analysis.CMAJ2009;180:32–9.
This meta-analysis pooled data from 10 randomizedcontrolled trials to determine the OR and absolute riskof fractures with thiazolidinedione therapy. There were13,715participantsinthe10randomizedcontrolledtri-als that compared the effect of thiazolidinediones onotheroralagentsinthemanagementoftype2diabetes.Overall, the odds of having a fracture were significantlyincreasedwiththiazolidinedionetherapy(OR1.45;95%CI,1.18–1.79).However,with3%ofpatientsinthethi-azolidinedione group experiencing fractures, comparedwith 2.4% in the control groups, the ARI increase wasonly0.6%.Datainfiverandomizedcontrolledtrialswereavailabletostratifybysex,withtheoddsofexperiencinga fracture doubled in women (OR 2.23; 95% CI, 1.65–3.01)butnotinmen(OR1.00;95%CI,0.72–1.39).TheARI increase among women was 2.8%, translating intooneexcessfractureforevery35womentreatedwithathi-azolidinedione.Basedoncurrentdrugusepatternsinanestimated2millionAmericanwomenusingthiazolidine-diones,about30,000excessfracturesinwomenwithtype2 diabetes could be expected. These data, taken in con-textwithconcernsregardingthecardiovasculareffectsofthiazolidinediones, suggest clinicians need to carefullybalance the benefits of this therapy with the apparentskeletalandcardiovascularrisks.
PSAP-VII • Women’s and Men’s Health220Drug-Induced Osteoporosis
Questions 1–3 pertain to the following case.Y.K. is a 65-year-old Asian woman (height 5′1′′, weight49.5kg).ShewasgivenadiagnosisofstageIIAestrogenreceptor–positive breast cancer 2 years ago and, sincethen,hasbeenreceivingadjuvanttherapywithanastro-zole1mgoncedaily.Hermedicalhistoryissignificantforapreviouswristfractureafterslippingontheiceabout8yearsago.Shehasalsoreceivedcitalopram20mgoncedaily for depression for the past 6 months, which shereportshasimprovedherappetite,sleep,andmood.Herother drugs include calcium carbonate 1250 mg (500mgofelementalcalcium)twotimes/dayandvitaminD1000units/day.Shequitsmoking7yearsago,butbeforethat, she had a 35 pack-year history. Y.K. reports occa-sional alcohol use of about one or two glasses of winepermonth.Shedeniesanyfallsinthepastyear.Herpar-ents have not experienced any fractures. All laboratoryvaluesarewithinnormallimits.Herserum25-hydroxyvi-taminDconcentrationwas45ng/mL.Y.K.wasrecentlyreferredforbonemineraldensity(BMD)testingbydual-energyx-rayabsorptiometry(DEXA),andherT-scoreswere as follows: lumbar spine −3.3, femoral neck −3.2,andtotalhip−3.1.
1. Based on the WHO fracture risk assessment tool (FRAX) (scroll to T-scores on item 12 on the FRAX tool), which one of the following best rep-resents Y.K.’s 10-year probability of experiencing a major fragility fracture?A. Noincreasedrisk.B. Lessthan10%.C. Between10%and20%.D. Greaterthan20%.
2. Which one of the following represents the most appropriate intervention to reduce Y.K.’s risk of fracture?A. Initiatetherapywithrisedronate
3. Which one of the following represents the most important factor in determining Y.K.’s risk of fra-gility fractures secondary to anastrozole?A. Concomitantuseofcitalopram.B. Currentcalciumintake.C. VitaminDdeficiency.D. BaselineT-score.
4. A41-year-oldpremenopausalwhitewomanhasbeengiven a diagnosis of stage IIIB breast cancer and isscheduledtostartadjuvanttherapywith leuprolidein addition to her anthracycline-based chemother-apy.BaselineBMDtestingrevealsaT-scoreof−1.5forthetotalhip.Which one of the following is the most appropriate intervention for this patient’s bone health?A. Initiatezoledronicacid4mgintravenously
5. Yourpatientisa58-year-oldwhiteman(height5′6″,weight 66.4 kg) who received a diagnosis of stageT3 prostate cancer 8 months ago. He began radia-tiontherapyandadjuvanttreatmentwithgoserelin,andhiscurrentdosageis10.8mgbydepotinjectionevery3months.Hismedicalhistoryissignificantfordyslipidemiaandhypertension,whicharewellcon-trolled with atorvastatin 40 mg/day and ramipril 5mg/day. His history is negative for oral glucocorti-coid use, and he has had no fragility fractures. Thepatientisanonsmokerandreportsoccasionalalco-holuseofnomorethanoneortwodrinkspermonth.He denies using recreational drugs. His family his-tory is significant for osteoporosis, and his motherhashadtwohipfractures.All laboratoryresultsarewithinnormal limits,andhis25-hydroxyvitaminDconcentrationisreportedas50ng/mL.ThepatientrecentlyunderwentBMDtestingbyHologicDEXA,andhisresultsareasfollows:lumbarspine0.759g/cm2;femoralneck0.683g/cm2;andtotalhip0.686g/cm2.Incorporating his risk based on the FRAX tool, which one of the following interventions is most appropriate for this patient?A. Ensurehiselementalcalciumintakeisatleast
Questions 6 and 7 pertain to the following case.E.K. is a 34-year-old African American woman (height5′8″, weight 72 kg) who presents for administrationof depot medroxyprogesterone acetate (DMPA) 150mg for contraception. She has received DMPA every 3months for the past 12 months. In addition to DMPA,
Self-Assessment Questions
PSAP-VII • Women’s and Men’s Health 221 Drug-Induced Osteoporosis
E.K.takes400unitsofvitaminDdailyandhasa12-yearhistoryofvalproicacidusetomanageherepilepsy.E.K.reports a lifelong lactose intolerance and has not toler-ated dairy products since she was a small child, usingsoymilkinstead.Shedoesnotsmokeoruserecreationaldrugs,andshereportsonlyoccasionalalcoholuse.E.K.isphysicallyactive,running5kmtwotimes/weekwithherrunninggroup,andtakespartinresistancetrainingtwotimes/week.Herparents,aliveandwellintheirearly70s,havenotexperiencedanyfragilityfractures.
6. Which one of the following places E.K. most at risk of developing an osteoporosis-related fracture?
A. UseofDMPA.B. AfricanAmericanethnicity.C. Lowcalciumintake.D. Excessivephysicalexercise.
7. Which one of the following is the most appropri-ate intervention to make with respect to E.K.’s bone health?
A. CounselE.K.toincreasehervitaminDintaketo2000units/day.
B. ReferE.K.forBMDtestingusingDEXA.C. Recommendswitchingfromvalproicacidto
Using these data, which one of the following represents the most accurate expression of the absolute increased risk of experiencing a fracture during the next 5 years with anastrozole com-pared with tamoxifen?A. 44%.B. 11%.C. 7%.D. 3.3%.
Questions 9 and 10 pertain to the following case.J.R.isa77-year-oldmanadmittedtotheorthopedicser-viceforarthroplastyofthehipafterafallathisnursing
home. His medical history is significant for Alzheimerdisease, pedal edema, esophageal reflux, hyperten-sion, and osteoarthritis. J.R.’s drug regimen and historyincludes pantoprazole 20 mg/day for 1 month; furose-mide120mgtwotimes/dayfor3years;ramipril2.5mg/dayfor3years;andacetaminophen500mgfourtimes/day for 5 years. Vital signs include the following: tem-perature 98oF, supine blood pressure 148/72 mm Hg,standingbloodpressure102/68mmHg,andpulserate58 beats/minute supine increasing to 72 beats/minuteonstanding.
9. Which one of the following drugs is most likely responsible for J.R.’s fall?
A. Pantoprazole.B. Acetaminophen.C. Furosemide.D. Ramipril.
10. Which one of the following patient-specific fac-tors is most likely to increase J.R.’s risk of hip fracture associated with pantoprazole use?
A. Age.B. Sex.C. Dose.D. Durationofuse.
Questions 11 and 12 pertain to the following case.S.D. is a postmenopausal 56-year-old white woman(height 5′4″, weight 77 kg) recently given a diagnosisof type 2 diabetes. She was initially treated with met-formin 500 mg three times/day but experienced severegastrointestinalupset requiringdiscontinuationof ther-apy.Considerationisbeinggiventostartingtherapywithrosiglitazone4mgoncedailywithtitrationasnecessary.Otherdrugsinherregimenincluderamipril10mg/dayfor hypertension and atorvastatin 80 mg/day for dys-lipidemia. S.D. is a nonsmoker, does not drink alcohol,andhasnotexperiencedany fractures.Hermotherwasgivenadiagnosisofosteoporosisandexperiencedahipfracturewheninher80s.S.D.recentlyhadBMDtestingbyDEXA,andherT-scoreswerereportedas−0.8atthelumbarspine,−0.7atthefemoralneck,and−0.7fortotalhip.
11. Which one of the following risk factors is most significant in determining S.D.’s risk of develop-ing fractures related to use of rosiglitazone?
A. LumberspineT-score.B. Femalesex.C. Familyhistoryofhipfracture.D. Postmenopausalstatus.
PSAP-VII • Women’s and Men’s Health222Drug-Induced Osteoporosis
12. Six months later, S.D. is doing well on rosigli-tazone. Her glycosylated hemoglobin and fastingblood glucose values are within target, and she hasnotexperiencedanyedema,shortnessofbreath,orother signs or symptoms suggestive of heart fail-ure.However,S.D.expressesconcernregardingherpotentialriskoffractureafterreadingseveralmediareportsanddoingsomereadingontheInternet.Onreviewoftheliterature,thefollowinginformationisobtainedfromameta-analysisexaminingtheassoci-ationbetweenfracturesandthiazolidinediones:
Which one of the following best represents the number of women needed to treat with a thiazolidinedione compared with other oral anti-hyperglycemic agents over 5 years to cause one excess fracture? A. 23.B. 35.C. 76.D. 111.
13. Which one of the following individuals is at great-est risk of developing a drug-related osteoporotic fracture?A. A28-year-oldwhitewomanreceivingDMPA
C. A46-year-oldHispanicwomanreceivingolanzapineandcitalopramforthepast3years.
D. A63-year-oldAfricanAmericanwomanreceivingletrozoleandpioglitazoneforthepast3years.
Questions 14 and 15 pertain to the following case.M.W. is a 68-year-old postmenopausal woman (height5′4″,weight76kg)witha32-yearhistoryofbipolardisor-der.Herdiseaseiscurrentlywellmanagedwithvalproicacid 500 mg three times/day and olanzapine 5 mg twotimes/day.Otherdrugsincluderisedronate35mgonceweekly,ramipril5mg/day,andcalciumcarbonate1250mg two times/day. Although M.W. has a 35 pack-yearhistoryofsmoking,shenolongersmokesordrinksalco-hol.Atarecentappointment,laboratorytestsrevealeda25-hydroxyvitaminDconcentrationof15ng/mL.
14. Which one of the following factors is most likely responsible for M.W.’s current vitamin D status?A. Valproicacid.B. InadequatevitaminD3intake.C. Inadequatecalciumintake.D. Olanzapine.
15. Which one of the following represents the most appropriate daily dose of vitamin D supplemen-tation for M.W.?A. 600units.B. 800units.C. 1000units.D. 4000units.
16. Your patient is a 65-year-old man with Down syn-drome who resides in a long-term care facility. HismedicalhistoryissignificantforAlzheimerdemen-tiaandageneralizedseizuredisorder, forwhichhereceives donepezil 10 mg/day, risperidone 1.0 mgtwo times/day, and phenytoin 250 mg/day. His25-hydroxyvitamin D concentration is measured at8ng/mL.Which one of the following vitamin D dosages is most appropriate for this patient?A. 1000unitsofvitaminDdaily.B. 50,000unitsofergocalciferoloncemonthly.C. 50,000unitsofergocalciferolonceweeklyfor
8weeks,followedby1000unitsofvitaminDdaily.
D. 50,000unitsofergocalciferoltwiceweeklyfor3months;then1000unitsofvitaminDdaily.
17. A 51-year-old woman with a 15-year history ofschizophreniapresentsforBMDtestingafteraCollesfracture of the wrist. Her T-scores are reported as−2.0 at the hip and −2.5 at the lumbar spine. Herdrugisdepotfluphenazinedecanoateforthepast8years, which has substantially improved her adher-encetotherapyandreducedhernumberofhospitaladmissions.Which one of the following interven-tions is most appropriate to manage this patient’s bone health?A. Discontinuefluphenazinedecanoateandstart
D. InitiatemonthlyoralbisphosphonatetherapyanddailycalciumandvitaminDsupplementation.
PSAP-VII • Women’s and Men’s Health 223 Drug-Induced Osteoporosis
18. Your patient is a 56-year-old woman treated withl-thyroxine150mcg/dayforthepast6yearsaftertheremovalofher thyroidglandsecondary toamalig-nancy. Her recent thyroid-stimulating hormoneconcentrationwasmeasuredat1.0milli-unit/L.HerBMD,measured1monthago,producedT-scoresof−1.2atthespineand−0.8forthetotalhip.Which one of the following is most appropriate for the management of this patient’s bone health?A. AddcalciumandvitaminDsupplementation
anddecreasethel-thyroxinedoseto137mcg/day.
B. AddcalciumandvitaminDsupplementation,decreasethel-thyroxinedoseto137mcg/day,andaddrisedronate150mg/month.
C. AddcalciumandvitaminDsupplementationanddiscontinuel-thyroxine.
D. AddcalciumandvitaminDsupplementation,discontinuel-thyroxine,andaddrisedronate150mg/month.
19. Your patient, a 54-year-old man with rheumatoidarthritis, has required many courses of prednisonefor several months during the past 5 years. Herecently lost his job and has no health insurance.RecentBMDtestingrevealsaT-scoreof−3.0atthespineandaT-scoreof−2.6forthetotalhip.Which one of the following represents the most appro-priate treatment for this patient?A. CalciumandvitaminDsupplementation.B. Genericalendronate70mg/week;calciumand
D. Teriparatide20mcgsubcutaneouslydaily;calciumandvitaminDsupplementation.
20. Yourpatientisa72-year-oldwhiteman(height5′11″,weight72kg)withearlyprostatecancertreatedwithgoserelin depot injection 10.8 mg every 3 months.Hismedicalhistorywasinsignificantbeforehisdiag-nosis of prostate cancer. He recently experienceda vertebral fragility fracture, and BMD revealed aT-score of −3.0. Denosumab is being consideredfor treatment of his osteoporosis. Data from theHormone Ablation Bone Loss Trial (HALT) showthefollowingregardingthecumulativeincidenceofvertebralfracturesat36months:
Using these data, which one of the following rep-resents the most accurate expression of the RR decrease for experiencing a fracture during the next 3 years with denosumab compared with placebo?A. 1.5%.B. 2.4%.C. 62%.D. 38%.
PSAP-VII • Women’s and Men’s Health224Drug-Induced Osteoporosis
![(9/21) Thambi Lecture: Drug-Induced Renal Disease ... · (9/21) Thambi Lecture: Drug-Induced Renal Disease Identifying Drug-Induced AKI [Drug-Induced] to be shorted to DI - Sx: Decreased](https://static.documents.pub/doc/80x56/5ee17160ad6a402d666c535c/921-thambi-lecture-drug-induced-renal-disease-921-thambi-lecture-drug-induced.jpg)
