abstract_cme abstract_cme abstract_cmeabstract_cme abstract_cme abstract_cmeabstract_cme abstract_cme

abstract_source abstract_source abstract_source abstract_source abstract_source abstract_source abstract_sourceabstract_source abstract_source abstract_source

italic_ab- stractsource

bold_abstractkeywordsabstract_keywordsabstract_keywords abstract_keywords abstract_keywords abstract_keywords

Bisphosphonates and low-impact femoral fractures:Current evidence on alendronate-fracture risk

Jennifer P. Schneider, MD, PhD

Dr Schneider

practices internal medicine and painmanagement in Tucson, Arizona.

Disclosure:

As she was the patient in a related 2006

Geriatrics

case report, the author discloses that shehas a personal interest in understanding the possiblecausative role of alendronate and atypical femoral frac-tures. She states that she has no financial interests inany pharmaceutical product used to treat osteoporosis.

66-year old, previously healthy woman developed aspontaneous stress fracture of her right foot, whicheventually healed. Nine months later she took a stepin her bedroom and collapsed to the floor. An x-ray re-vealed a nontraumatic fracture of her right femur. Sheunderwent surgery with placement of an intramedullaryrod. Her physician told her she had most likely had a stressfracture, which became a completed fracture. A bone scandone shortly after her surgery revealed a stress fracture of her left femur. Some months later she underwent prophy-lactic rodding of the left femur. The patient had been onalendronate for 7 years.A 65-year woman visiting Europe stepped off the bottomstep of a van and collapsed. An x-ray revealed a nontrau-matic fracture of her left femur. She had been experiencinga dull ache in her left femur for some months. The patientunderwent placement of an intramedullary rod. One yearlater she developed a dull ache in her right femur. A bonescan showed a stress fracture in the right femur. A bonespecialist recommended prophylactic rodding of the rightfemur, which was done. The patient had been on alendro-nate for 9 years.A 59-year-old-woman took a step, her right leg gave out,and she fell to the ground as she heard her leg break. Herfemur was fractured. The orthopedic surgeon on call toldher, “We don’t usually see this type of fracture withouttrauma.” For the preceding year she’d experienced pain inher right thigh, which was severe enough to cause limping.An x-ray had been negative, and her primary care physicianthought she had fibromyalgia. She had been on alendronatefor more than 5 years.These unpublished case reports, and several other simi-lar ones, were sent to the author following publication of a 2006 report in

Geriatrics

of a 59-year-old, previouslyhealthy woman who, while riding on a subway train, suf-fered a comminuted spiral fracture of the right femur whenthe train jolted (see figure, page 20). The patient had beenexperiencing pain in her right thigh for 3 months. A bonescan a week before the fracture showed a stress fracture of the right femur. The patient had been taking alendronate, 70mg/week, for approximately 7 years for osteopenia, as wellas calcium plus hormone replacement therapy. Despite pro-

Several recent medical articles have describedmultiple cases of unusual low-impactsubtrochanteric stress fractures or completedfractures of the femur in patients who have beenon the bisphosphonate alendronate for severalyears for osteoporosis or osteopenia. Somepatients have experienced such fractures in bothfemurs. The fractures are often preceded by painin the affected thigh, may have a typical x-rayappearance, and many have delayed healing. Ithas been hypothesized that in some patients,long-term alendronate causes oversuppressionof bone turnover, resulting in bones that arebrittle despite improved bone density. In patientswith atypical or low-impact fractures of thefemoral shaft, consider the possible connectionwith alendronate use. Some bone specialistsnow recommend stopping alendronate in mostpatients after 5 years.

Schneider JP. Bisphosphonates and low-impact femoral frac-tures: Current evidence on alendronate-fracture risk.

Geriatrics

.2009;64(1):18-23.

Key words:atypical fracture, femoral shaft, low-impact fracture, oversuppression, subtrochantericDrugs discussed:alendronate, ibandronate, pamidro-nate, risedronate, teriparatide, zoledronic acid

Geriatrics

www.geri.com

January 2009 Volume 64, Number 1

Geriatrics

www.geri.com

ALENDRONATE-FRACTURE RISK

longed use of an electrical bone stim-ulator, and cessation of alendronateuse, the fracture did not unite. After 9months, the patient had a second surgi-cal procedure to replace the originalrod with a larger one. After a delay,the bone finally united. The authorsuggested a possible causal relation-ship between long-term alendronateand the femoral fracture.Fragility fractures of the proximalfemur are rare. However, in the past 3years, multiple additional cases likethose above have been published andthe evidence continues to grow thatin a small subpopulation of patients,long-term alendronate use may berelated to low-impact, nontraumatic,or “atypical” fractures of the femur,often with delayed healing. This paperreviews the older evidence for a con-nection between bisphosphonates andbone fragility, and summarizes recentreports and recommendations.

Femoral fracturesand alendronate

Bisphosphonates are considered first-line treatment for postmenopausalosteoporosis. They are prescribed formillions of geriatric patients. Bisphos-phonates—alendronate (Fosamax),risedronate (Actonel), ibandronate(Boniva), and zoledronic acid (Zo-meta, Reclast)—inhibit bone resorp-tion by decreasing the activity of osteoclasts. Extensive studies haveshown that therapy with bisphospho-nates improves bone density and de-creases fracture risk.

2-6

When discon-tinued after 5 years, the physiologiceffect of alendronate continues for atleast 5 years, with no increase in mor-phometric vertebral fracture risk or inthe risk of nonvertebral fractures com-pared with patients who continued totake alendronate for the full 10 years.

This result is consistent with the factthat alendronate is incorporated intobone matrix and has a biological half-life of more than 10 years.Bone turnover is a natural part of maintaining bone health. When boneturnover is inhibited by bisphospho-nates, microdamage that occurs regu-larly in bone but is normally repairedmight accumulate after long-termuse. There have long been concernsabout the long-term safety of bisphos-phonates because of their potential tocause oversuppression of bone turn-over.

8-13

The first reportsuggestive of the clinicalrelevance of these hy-pothetical concerns waspublished in 2005 by Od-vina et al,

describing 8postmenopausal womenand a man who sustainedunusual nontraumatic nonspinal frac-tures while on alendronate therapy for3-8 years. All 9 continued taking alen-dronate after the fracture. Six of the 9patients had delayed or absent fracturehealing for 3 months to 2 years duringcontinued alendronate therapy. All 9patients underwent iliac crest biopsyof trabecular bone. All the specimensshowed markedly suppressed bone for-mation. The authors concluded thatlong-term alendronate therapy mayresult in severe suppression of boneturnover, with increased susceptibil-ity to nonspinal fractures along withdelayed healing.In 2007 a group from Singaporepublished a retrospective review of patients admitted with a low-energysubtrochanteric fracture (defined asone in the region of the femur thatextended from the lesser trochanterto the junction of the proximal andmiddle third of the femoral shaft.)

Of 13 women identified, 9 were onlong-term alendronate therapy (mean4.2 years, range 2.5-5). Their averageage was 67 years, versus 80 years inthe non-alendronate group. Four of the 9 patients in the alendronate groupreported that the fracture had occurredin the absence of a fall. Five patientsreported experiencing prodromalpain in the fractured limb, starting 2-6 months before the injury; none of the patients in the non-alendronategroup had prodromal symptoms. In6 patients in the alendronate group,cortical hypertrophy was identifiedon the lateral side of the subtrochan-teric region of the femur, and 3 of thesealso had cortical hypertrophy on thecontralateral femur.The Singapore group recentlyelaborated on its findings with a ret-rospective review of postmenopausalpatients with subtrochanteric insuffi-ciency fractures admitted to their hos-pital over a 20-month period.

Theyfound 17 patients, whose mean age was66 years, and all had been taking alen-dronate, for a mean of 4.4 years (range2.2-8), except for one patient whowas on risedronate for 6 years after4 years of alendronate. All fractureswere low-energy, typically sustainedafter tripping. Seven of the patients re-ported experiencing acute pain beforethey fell, suggesting that the fracturepreceded the fall. Thirteen of the 17patients (76%) had experienced pro-dromal pain in the affected thigh rang-ing from 1 week to 2 years before thefracture. Often these patients hadbeen treated for referred pain from aspinal origin, without improvement.Three patients had sustained priorcontralateral femoral fractures 2-4years earlier but had been continuedon their bisphosphonate; the patientwho was switched to risedronate wasone of these. Five other patients hadstress reactions seen on plain x-rays inthe contralateral femurs; a bone scanof one of these patients showed abnor-mal uptake in that femur. Pointing to

Long-term alendronatetherapy may suppressbone turnover.

Geriatrics

www.geri.com

January 2009 Volume 64, Number 1

ALENDRONATE-FRACTURE RISK

the incidence of bilateral stress reac-tions and fractures in more than half of their patients, the authors concludethat these patients have a systemicdisorder reflecting oversuppression of bone turnover rather than localized pa-thology. They advise cautious admin-istration of alendronate in osteoporosismanagement, and “in situations wherethe characteristic subtrochanteric frac-tures have already developed, physi-cians should strongly consider discon-tinuing the drug.”In 2008 a group from the Hospitalfor Special Surgery, an orthopedichospital in New York City, published2 reports of patients taking alendro-nate who had atypical fractures of the femur. One report focused on aspecific radiographic pattern.

Theauthors described 15 postmenopausalwomen who had been receiving alen-dronate for a mean of 5.4 years andwho presented with atypical low-en-ergy fractures, defined as fractures oc-curring in a fall from a standing heightor less. Ten of the 15 had a unique ra-diographic pattern, a simple transverseor oblique fracture with beaking of thecortex and diffuse cortical thickeningof the proximal femoral shaft. All thepatients with this pattern also had cor-tical thickening of the contralateralfemur and 3 had had a prior femoralfracture; none of the patients had a his-tory of vertebral fractures. The au-thors conclude that these 10 womenmay represent a subgroup of the popu-lation that is more susceptible to the ef-fects of prolonged suppression of boneturnover. They call for a prospectivestudy to characterize this subgroup.The same group also reported on aretrospective review of patients withfemoral shaft fractures (including the15 in the prior report) admitted betweenJanuary 2002 and March 2007.

Sev-enty low-energy fractures were identi-fied, in 59 females and 11 males, withan average age of 74.7 years. Osteo-porosis was present in 44% of the 70patients. Twenty-five patients (36%),all women, were being treated withalendronate, 84% for osteoporosis;none of the patients had used any otherbisphosphonate. Of the 25, 19 (76%)had a specific pattern to the fracture—it was transverse, with a one-sided beak in an area of thickening of the cortex.This fracture pattern was seen in only1 patient (2%) of patients not beingtreated with alendronate. The odds ra-tio for this pattern was 139.33 for alen-dronate users, and was 98% specificto identifying alendronate users. Thepatients with this pattern had been us-ing alendronate for a mean of 6.9 years.The authors concluded that althoughthey have not established a causal re-lationship, such fractures may resultfrom propagation of a stress fracturewhose repair is retarded by decreasedosteoclast activity and impaired mi-crodamage repair resulting from theprolonged use of alendronate. Minimaltrauma is then required to produce acompleted fracture.

Discussion

Fragility or insufficiency fractures area type of stress fracture that occurs inosteoporotic bone subjected to normallevels of stress. They typically occurin the vertebrae, hip, distal radius,and the proximal humerus follow-ing minimal or no trauma, but onlyrarely in the proximal femur.

Thesubtrochanteric region of the femur isone of the strongest parts of the fe-mur and it is unlikely to fracture inlow low-energy trauma unless ex-treme osteoporosis is present.

Thereports of multiple cases of low-impactfemoral fractures in patients who weretaking alendronate for several years, apreviously rare event, have thereforecalled for further study of the possibleconnection between alendronate andsuch fractures, as has been suggestedby several authors.Alendronate is stored in the bone formany years and is reactivated as bone isturned over and the drug re-enters thecirculation. Patients on long-term alen-dronate who experienced completedfractures of the femur with minimal

A case is madefor discontinuingalendronate.