|Home | About | Journals | Submit | Contact Us | Français|
Bisphosphonates are widely used for treatment of postmenopausal osteoporosis. Although short-term safety and efficacy of these drugs have been investigated and documented, an increasing number of recent reports draw attention to the possible correlation between long-term alendronate therapy and the occurrence of insufficiency fractures in the proximal femur owing to what is known as severely suppressed bone turnover. We describe two femoral insufficiency fractures in two women receiving long-term alendronate therapy. The first woman sustained a periprosthetic fracture at the tip of the femoral stem whereas the other woman had a fracture in the subtrochanteric region. We analyze the characteristics and natural course of these two unique fractures, and emphasize the importance of being aware of the possible correlation between long-term alendronate therapy and insufficiency femoral fractures.
Bisphosphonates are considered the first-line therapy of postmenopausal osteoporosis. They improve bone density and decrease fragility fractures mainly in the spine and hip [1–3, 6, 26]. They reduce bone resorption by inhibiting osteoclast function and promoting apoptosis [12, 18]. Alendronate was the first drug of this family to be approved by the US Food and Drug Administration in 1995. Since then, a large number of postmenopausal women and a smaller number of men with idiopathic, steroid-induced, or hypogonadal osteoporosis have been treated with alendronate.
The short-term efficacy and safety have been widely studied and documented [1, 2, 18, 26]. However, there have been concerns regarding possible adverse effects of long-term therapy. Although the drug is excreted extensively by the kidneys, the small amount that remains in the body may attach to the osteoid tissue for decades [9, 20]. Histomorphometric analysis of bone taken from patients receiving long-term alendronate shows changes consistent with severely suppressed bone turnover suggesting an adynamic bone disorder . These changes include reduction in osteoid thickness and volume, reduced osteoblastic/osteoclastic surface, and diminished bone matrix. Consequently, this may give rise to failure of microfracture repair and secondary mineralization with increased brittleness and deterioration of the bone biomechanical properties. Currey  reported increasing bone mineralization gave high Young modulus of bone elasticity but low values of work of fracture (which is a measure of fracture toughness).
There have been an increasing number of reports regarding insufficiency femoral fractures in patients receiving long-term therapy with alendronate [2, 5, 11, 14, 15, 23, 24]. Clinical presentation of these fractures has been similar. The patient, typically a postmenopausal woman, sustains a fracture in the subtrochanteric region or the femoral shaft after low-energy trauma such as a gentle fall or twisting movement. Many of these patients experience prodromal pain in the affected limb several months before the occurrence of these insufficiency fractures. The radiographs show thickening of the cortex and transverse fracture line with medial spiking. Lateral cortical reaction is encountered on the contralateral side in some patients and is considered the first stage of these fractures. Delayed union or nonunion is common in these fractures. The etiology of the delayed union or nonunion in these patients is not clearly known but might be related to severely suppressed bone turnover.
In this report, we present two cases of femoral insufficiency fractures in two women receiving long-term alendronate therapy. In the first case, a woman sustained a periprosthetic femoral fracture, and in the second, another woman sustained a subtrochanteric femoral fracture.
A 78-year-old woman presented to the emergency room in March 2007 with pain of 1 week in the right thigh. The pain started after a same-level fall while walking. Six years previously, the patient had a Charnley prosthesis implanted for primary osteoarthritis of the right hip and she had no complaint thereafter until she fell. The patient was treated with 70 mg alendronate per week since 1998 for postmenopausal osteoporosis. No dual-energy xray absorptiometry measurement was obtained before starting alendronate therapy. The diagnosis of postmenopausal osteoporosis was made clinically when the patient sustained two distal radial fractures during 1998.
Her physical examination showed diffuse tenderness over the thigh muscles with normal range of motion (ROM) of the hip and knee. The radiographs revealed no signs of fracture or loosening such as a radiolucency at the bone-cement interface or changes in the stem position (Fig. 1). Complete blood picture (CBP) and C-reactive protein (CRP) were normal. Analgesics were prescribed for pain relief.
In July 2007, the patient returned with increasing pain on weightbearing. New radiographs were taken, which showed thickening of the femoral shaft cortex with a fracture line below the tip of the femoral stem (Fig. 2). The new CBP, CRP, and serum calcium, phosphate, and vitamin D, were normal. No bone formation or resorption markers were investigated.
Reviewing radiographs from the time of her hip arthroplasty in 2001 revealed no cortical thickening of the femoral shaft at the time of surgery, whereas there was apparent cortical thickening with a lateral cortical reaction seen on the radiographs taken when the patient presented to us in March 2007 (Fig. 1).
At this time, a decision was made to treat the fracture nonoperatively with nonweightbearing. Clinical and radiographic followup after 6 months showed no improvement. Alendronate therapy then was stopped and the patient was scheduled for surgery. Preoperative radiographs showed a complete fracture going through both cortices (Fig. 3). The patient underwent open reduction and internal fixation with an angle-stable plating system with cable grip (Fig. 4). Postoperatively, the patient showed an improvement with minimal thigh pain on weightbearing with good ROM. At 5 months followup, radiographs showed callus formation bridging the fracture gap (Fig. 5), there was no tenderness over the fracture site, and the patient could bear weight without walking aids.
A 55-year-old woman with chronic alcoholic liver disease, hypertension, and a 10-year history of postmenopausal osteoporosis presented to the emergency room in January 2008 after falling on the floor at home. The patient’s osteoporosis had been diagnosed clinically (no dual-energy xray absorptiometry measurement was performed) when she sustained a vertebral compression fracture in 1999. Since that time, the patient was treated with vitamin D supplement and 70 mg alendronate weekly. At presentation, she had severe pain, swelling, and tenderness in the left thigh and her whole leg was shortened and externally rotated. Several months before the fall, she reported diffuse pain in both thighs with walking difficulties. Radiographs revealed a left-sided transverse subtrochanteric fracture (Type 32-A3.1 according to the AO classification) with cortical hypertrophy and medial spiking (Fig. 6). At the subtrochanteric region on the contralateral femur, a cortical hypertrophy with lateral cortical reaction was seen.
The next morning, the patient underwent insertion of a long gamma nail (Fig. 7) and received antibiotics and thrombosis prophylaxis. No tissue specimen was obtained during the operation because we could not identify any osteolytic lesions at the fracture site. There were no immediate postoperative complications and the patient was mobilized with the help of a walker. Followup of the patient 3 and 6 months postoperatively revealed gradual improvement with mild to moderate thigh pain on weightbearing. Her radiographs at those times showed increasing callus formation. Nine months postoperatively, the patient had minimal thigh pain and tenderness and the radiographs showed decreased fracture gap with bridging callus (Fig. 8). Our two patients were informed that data concerning their cases would be submitted for publication.
Insufficiency fractures of the femur have been reported either as secondary to metastatic osteolytic lesions or in relation to osteomalacia [4, 8], pycnodysostosis [13, 22], primary hyperparathyroidism , and in patients receiving fluoride therapy . The occurrence of insufficiency femoral fractures in patients receiving long-term alendronate therapy was described by Neviaser et al. , Kwek et al. , and Goh et al. . Neviaser et al.  identified insufficiency femoral fractures in 25 patients receiving long-term (average, 6.2 years) alendronate therapy. Nineteen of these patients exhibited a fracture pattern of a transverse fracture line, cortical hypertrophy, and medial spiking. Furthermore, Goh et al.  and Kwek et al.  reported 17 patients receiving long-term alendronate therapy (average, 4.8 years) with low-energy subtrochanteric fractures of a similar configuration. Nine patients had either a contralateral fracture or cortical reaction and 13 patients experienced thigh pain preceding the fracture occurrence.
Periprosthetic femoral fractures are uncommon, but the incidence seems to be increasing . Gill et al.  reported two cases of lateral insufficiency periprosthetic femoral fractures caused by osteopenia and varus angulation of the stem with subsequent increased mechanical loading applied to the lateral cortex, which was extremely thinned on radiographs. They recommended revision of the stem and used noncemented Wagner implants with successful outcomes. In our Patient 1, the cortex was thickened and the stem was not angulated.
Localization of the fractures might be the result of the increased bending forces applied on the subtrochanteric area, as described by Pauwels , in our Patient 2 and below the tip of the implant in Patient 1. The delayed union in Patient 1 and involvement of the contralateral femur in Patient 2 were consistent with previously reported patients receiving long-term alendronate therapy and might indicate the pathology is not confined just to the site of the fractures. In Patient 2, we believe the fracture was related to severely suppressed bone turnover secondary to long-term alendronate therapy and not to the chronic liver disease because the patient’s overall liver function was well compensated despite chronicity and the typical presentation and characteristics of the fracture.
These two cases add another report emphasizing a possible relationship between long-term therapy with alendronate and severely suppressed bone turnover resulting in insufficiency fractures. In a patient receiving such therapy, groin or thigh pain should raise the suspicion of an insufficiency fracture and this type of fracture should be ruled out with repeated radiographs. Withdrawing alendronate therapy should be considered if investigations show early signs of such a fracture. We do not know if this possible adverse effect on bone is unique to alendronate or if it is shared with other bisphosphonates. Additional research is needed to study the long-term effect of this class of drugs on bone strength.
Each author certifies that he/she has no commercial associations (eg, consultancies, stock ownership, equity interest, patent/licensing arrangements, etc) that might pose a conflict of interest in connection with the submitted article.
Each author certifies that his or her institution has approved the reporting of these case reports, that all investigations were conducted in conformity with ethical principles of research, and that informed consent for participation in the study was obtained.