Keywords
femoral fractures - bisphosphonate - osteoporosis
Introduction
Osteoporosis is defined as gradual, progressive reduction of bone mass. The main risk
factors for osteoporosis include a sedentary lifestyle, alcoholism, smoking, prolonged
use of corticosteroids, and a reduced calcium and vitamin D bioavailability.[1] Hip, spinal, distal radial, and proximal humeral fractures are the most associated
with osteoporosis.[2] Medical treatment of osteoporosis is recommended as primary prevention for these
fractures, with bisphosphonates as first-line drugs.[3]
Prolonged bisphosphonates use can result in a progressive loss of bone elastic properties,[4] representing a risk factor for atypical femoral fractures. These injuries are defined
as both incomplete (affecting the lateral cortex alone) and complete (in a transverse
pattern) fractures occurring between the lesser trochanter and the supracondylar region
of the femur (subtrochanteric or diaphyseal fracture) after minimal trauma.[5]
[6]
[7]
[8]
The risk for atypical fracture increases in people using oral bisphosphonates for > 3
years (average treatment, 7 years). The absolute risk in females is three times higher
when compared with males, and alendronate is associated with the highest risk for
atypical fractures.[5]
The literature describes good outcomes from the treatment of these fractures using
intramedullary nails.[9] However, the consolidation time is often quite prolonged, potentially because bisphosphonates
acts on the soft callus to suppress its remodeling.[10]
The present study aims to describe outcomes from a series of surgically treated patients
with atypical femoral fractures due to bisphosphonates use, to correlate the time
of previous medication use with fracture consolidation time, and to compare the consolidation
time in complete and incomplete fractures.
Material and Methods
This is an observational, retrospective study with 66 patients (2 bilateral cases),
totaling 68 atypical femur fractures associated with chronic bisphosphonates use.
The patients underwent orthopedic surgical treatment at a referral hospital from January
2018 to March 2020.
Inclusion and exclusion criteria
The following inclusion criteria were applied: 1) symptomatic complete and incomplete
atypical femur fractures; 2) chronic (> 5 years) bisphosphonates use; 3) age > 60
years old.
The following exclusion criteria were applied: 1) previous or current ipsilateral
osteoarticular hip infection; 2) typical femoral osteoporotic fractures (femoral neck,
transtrochanteric, subtrochanteric injuries); 3) pathological fractures associated
with primary or metastatic tumors and other bone conditions (for example, Paget disease
or fibrous dysplasia).
Data Collection, Radiographic Evaluation, and Surgical Technique
Data were collected directly from the electronic medical records of the hospital after
approval by the ethics committee. Variables included gender, age, body mass index
(BMI), and time of previous continuous bisphosphonates use.
Radiographic images were obtained in a standardized manner on a digital device with
a predefined 100% magnification. Images were analyzed digitally with the Centricity
Universal Viewer Zero Footprint software (GE Healthcare, Barrington, IL, USA). Consolidation
was defined by the formation of bridge calluses in at least three cortical layers
in two radiographic views (anteroposterior and lateral views), using the reliability
of the radiographic union scale in tibial fractures (RUST) score expanded for femoral
fracture consolidation.[11]
The lack of radiographic signs of consolidation after ≥ 6 months postoperatively was
defined as pseudoarthrosis. All radiographic evaluations were performed independently
by 2 orthopedists with > 5 years of experience. If there was no consensus among the
evaluators, the final decision was taken by the senior researcher of the study.
All patients were submitted to surgical treatment with closed reduction on an orthopedic
table, using a long cephalomedullary nail, a cephalic screw with an antirotating device
and a distal locking screw ([Figure 1]).
Fig. 1 (A) Preoperative radiographic image of an atypical bilateral fracture. (B) Radiographic image 7 months after surgery.
The postoperative procedures were standardized. Partial load with a walker was allowed
as tolerated starting at the 1st postoperative day. The patients continued on motor physical therapy until fracture
consolidation. Bisphosphonate therapy was suspended after fracture, and teriparatide,
20 mcg per day subcutaneously for 2 years, was indicated according to the institutional
protocol. Follow-up was carried out at the outpatient clinic specialized in bone metabolism.
Statistical Analysis
Categorical variables were shown as proportions, and continuous variables as mean
and standard deviation (SD) values. The Wilcoxon test compared continuous nonparametric
variables after distribution assessment using the Shapiro test. All statistical evaluations
were performed with the open-source Stats package from the R software (R Foundation,
Vienna, Austria).[12] Variables were correlated using the Pearson test.
Results
All patients were females, and two cases were bilateral. The patients were followed-up
on an outpatient basis until fracture consolidation, which occurred in all subjects.
The average follow-up time was 5.8 months. The mean age was 78.4 years old, with a
an SD value of 6.5. Most fractures occurred on the left side (56.5%). The mean BMI
was 26.6, with an SD value of 2.4.
Regarding fracture location, 65% were diaphyseal injuries and 45% occurred at the
subtrochanteric region of the femur. Regarding the anatomical type of fracture, there
were 55 cases of atypical complete fractures (81%), mostly transverse, followed by
oblique fractures and 1 case of a simple wedge fracture. There were 13 cases of atypical
incomplete fractures, representing ∼ 19%.
According to the radiographic analysis using the RUST score, fracture consolidation
occurred in an average period of 2.3 months, with an SD value of 1.6. Incomplete fractures
consolidated on an average period of 1.4 months, while complete fractures consolidated
within 2.5 months; this difference was statistically significant (p <0.05), as shown in [Figure 2].
Fig. 2 Consolidation time (in months) in complete or incomplete fractures.
There was no evidence of a correlation between the time of previous bisphosphonate
use and consolidation time, since the Pearson test revealed a correlation index of
0.05 and p = 0.63.
Cortical thickening was identified in all cases. The most common bisphosphonate was
alendronate (84.0%), followed by risendronate (8.7%), ibandronate (4.3%), and pamidronate
and denosumab (1.5% each).
The trauma mechanism was fall from own height in most cases (82.6%); there was no
history of trauma in the remaining cases. Only 26% of the patients included in the
study reported pain prior to the fracture. The mean time of bisphosphonates use was
7.8 years, with an SD value of 5.7 years.
Discussion
The main finding of the present study is the lack of correlation between the time
of previous bisphosphonate use and the fracture consolidation time after treatment
with a long cephalomedullary nail. There was a statistically significant difference
in the consolidation time when complete and incomplete fractures were compared.
Intramedullary fixation has biomechanical and biological advantages over plate osteosynthesis
in atypical fractures.[13] In a study, the plate failure rate was higher in atypical fractures (30%) when compared
with typical fractures (0%). Patients treated with intramedullary nail presented a
98% consolidation rate, as in our study, with a long consolidation time (8.3 months).
Weil et al.[14] showed that 7 (46%) of the 17 fractures treated with a long milled intramedullary
nail required a revision surgery. The high failure rate was attributed to impaired
bone healing related to the prolonged bisphosphonate therapy, not to the surgical
technique. In addition, these authors observed differences in consolidation time between
atypical femoral fractures submitted to anatomical and nonanatomical reduction; anatomically
reduced fractures healed an average of 3.7 months faster than those fixed in a varus
position.
Incomplete atypical femoral fractures represent a diagnostic challenge, as patients
may only experience discomfort in the thigh when walking. Thus, diagnosis is often
made only in the context of a complete fracture with major associated functional limitation.
A plain radiograph of the contralateral femur is recommended in patients with complete
fracture.[15]
[16]
[17] As a result, prophylactic fixation is recommended to patients with persistent pain.
However, the decision to proceed with prophylactic surgery for an incomplete fracture
is based on several factors, including the presence of a bilateral incomplete fracture,
persistent pain, and a complete fracture on the opposite side. Prophylactic surgery
may not be warranted in an asymptomatic patient.[18] In the present study, incomplete fractures in symptomatic patients were surgically
fixed.
The literature lacks definitive evidence regarding the consolidation time in surgically
treated complete and incomplete fractures. Some studies report a longer consolidation
time both in conservatively treated incomplete fractures and surgically treated complete
fractures. Authors attribute these findings to the biological effect of bisphosphonate
and to a reduction leaving a small diastasis at the fracture focus in some cases.[17]
[19]
[20]
The average age described in the literature for patients with atypical femur fractures
is ∼ 75 years old,[6] as in our study. Nevertheless, despite affecting elderly patients, atypical femur
fractures are not associated with an increased mortality.[7]
Bisphosphonate treatment must be discontinued in patients with atypical femoral fractures.[4]
[5]
[15]
[21] In addition, there is evidence that antiosteoporotic agents with potent bone-forming
effects, such as strontium ranelate and especially teriparatide, can improve bone
turnover and microarchitecture.[22]
[23] After diagnosis, all patients in the present study stopped using bisphosphonate
and were prescribed teriparatide.
Our study has some limitations. As it is retrospective, it suggests a potential information
bias regarding data collection from medical records. In addition, the methodology
employed in a single group does not allow a statistical verification of risk factors
related to this type of fracture. For the future, the authors plan to assess risk
factors in a case-control design and to establish a protocol for the prospective follow-up
of patients with atypical fractures.
Conclusion
All patients presented consolidation after surgical treatment with a long cephalomedullary
nail. The consolidation time was longer for complete fractures when compared with
incomplete injuries. There was no correlation between the time of previous bisphosphonate
use and the consolidation time for atypical femoral fractures.
