Keywords
femoral neck - femoral head necrosis - femoral neck fractures - pseudoarthrosis
Introduction
Proximal femoral fractures are a public health problem directly associated with increased
morbidity and mortality rates. Only 50% of the patients with proximal femoral fractures
are fully rehabilitated and can resume their previously performed activities.[1] In all age groups, ∼ 50% of the proximal femoral fractures occur in the region of
the femoral neck. Femoral head, trochanteric, and subtrochanteric fractures comprise
the other half, demonstrating the high prevalence and social and economic impact of
this injury in the spectrum of proximal femoral lesions.[2]
Femoral neck fractures have a bimodal distribution. In patients > 60 years old, these
lesions are usually associated with low-energy traumas and decreased bone mass in
the proximal femur. In patients < 60 years old, this fracture is related to higher
energy traumas and associated systemic injuries.[3]
The main objectives of the treatment of femoral neck fractures are regional anatomy
restoration, bone stock preservation, and rapid functional recovery of the limb.[4] The surgical treatment can be total or partial arthroplasty, or osteosynthesis based
on the fracture pattern and on the characteristics of the patient. Osteosynthesis
options include screw fixation, fixation with hip sliding screw associated with tube
plate, or angled plate associated with antirotation screws.
The use of osteosynthesis in the treatment of this fracture is still debated. A recent
study showed better outcomes in patients treated with arthroplasty, especially in
those > 65 years old.[5] Fixation would be indicated for the treatment of nondeviated fractures and in cases
in which, despite the deviation, the patient is < 60 years old. In this situation,
the preservation of the cephalic segment of the femur would avoid the need for arthroplasty,
as well as its complications, in a young individual. However, the incidence of femoral
neck osteosynthesis complications, such as avascular femoral head necrosis (AFHN),
lack of fracture consolidation, and fixation failure, made this subject controversial
in the literature.[6] Another frequent complication is femoral shortening due to fracture focus impingement,
which may alter the offset of the hip. Zielinski et al[7] performed a randomized, multicenter study with 76 patients submitted to femoral
neck fracture osteosynthesis who presented some degree of shortening after the consolidation
of the fracture. These authors studied the gait pattern and the motor force of the
hip compared with the contralateral side. They concluded that young patients are able
to compensate the shortening resulting from the consolidation of the femoral neck
fracture, and that few cases required the use of insole compensation.
One factor related to the poor prognosis in the treatment of this lesion through osteosynthesis
is the long time elapsed between the time of the fracture and the surgical procedure.
It is believed that earlier fracture fixations result in better functional outcomes
and in a lower tendency for synthesis failure and for AFHN. The favorable prognosis
of early fixation would be associated with the rapid restoration of local blood flow
and with the reduction of secondary damage. Fracture reduction and fixation within
12 hours are deemed ideal. Surgical stabilization after 24 hours would be associated
with worse results and with a higher incidence of surgical failure, which, in this
case, would be attributed to the late fixation of the lesion.[8] Other factors, such as initial fracture deviation, the reduction obtained during
surgery, and the type of implant used, also seem to directly influence the final outcome
of the surgical treatment.[9] Thus, information that helps surgeons in the therapeutic management of this fracture
is fundamental for the reduction of complication and failure rates. In the present
study, we have identified patient factors that may directly influence the outcome
of femoral neck osteosynthesis, and we suggest criteria to support the surgeon when
deciding the best therapeutic approach for such fractures, particularly in young adults.
Material and Methods
Between August 2003 and August 2011, 845 patients were admitted with femoral neck
fractures to a large orthopedic hospital. Men and women ≤ 60 years old with femoral
neck fracture related to high- or moderate-energy traumas and who were treated with
closed reduction and internal fixation with 3 cannulated screws, arranged in an inverted
triangle configuration, were included in the study.[10] Patients < 15 years old, with femoral neck pathological fractures or septic arthritis,
with no radiographic documentation in at least 2 views (anteroposterior [AP] and lateral
views), or with low-quality radiographic documentation, with a diagnosis of AFHN previous
to the fixation of the fracture, time between fracture and surgical treatment > 42
days, loss of clinical follow-up before the end of the 36-month observation period,
as well as patients submitted to other osteosynthesis techniques than those used in
the present study, or submitted to hip arthroplasty, were excluded.
The application of the inclusion and exclusion criteria reduced the initial sample
of 845 patients to 81 individuals. From the initial amount, 632 patients (74.8%) were
excluded from the study because they had undergone an arthroplasty; 112 (13.25%) were
excluded because they had been submitted to other osteosynthesis techniques, and 20
(2.4%) patients did not have complete medical documentation, did not follow the outpatient
reassessment routine, or were outside the age range of the present study.
This is an observational, longitudinal, retrospective, analytical, and comparative
study. The present study was performed through the review of medical records with
an evolutionary analysis of the complementary exams. The lesions were analyzed by
radiographic images with good technical quality by three independent observers experienced
in the treatment of femoral fractures who work at the Trauma Specialized Care Center
(CAE Trauma) of the hospital. The lesions were classified according to the initial
deviation by serial visualization of simple, AP and lateral hip radiographs. The lesions
were stratified according to the Garden classification criteria[11] into types I and II if there was no deviation, and into types III and IV if there
was any deviation.
All of the patients underwent the same surgical procedure performed by the CAE Trauma
surgeons. The surgery consisted of closed fracture reduction in an orthopedic traction
table, under radioscopic control, followed by internal fixation with 3 7.0 mm cannulated
screws. No patient required intracapsular hematoma drainage. At the postoperative
period, the mobilization of the affected limb was stimulated early, while partial
load bearing was stimulated as soon as it was tolerated by the patient. Total load
bearing was only allowed after radiographic confirmation of the consolidation of the
fracture.
The quality of the reduction was evaluated in 2 anteroposterior and lateral radiographs,
and it was deemed satisfactory when the deviation of the focus of the fracture was < 2 mm
and the Garden angles were of 160° and 180°, respectively. The positioning of the
screws was considered adequate when the distance between their tip and the subchondral
bone was < 5 mm, the distal screw was at the small trochanter or above it, and the
angle between the parallel screws was < 10°.[12]
From the active search in the medical records, the demographic data and information
about the characteristics of the lesions (time elapsed between the occurrence of the
fracture and the treatment, mechanism of the trauma, and initial deviation grade of
the lesion), the treatment and the clinical evolution of the patient (consolidation,
AFHN diagnosis, or lack of consolidation associated with synthesis failure) were obtained.
All of the patients were followed-up with periodic outpatient visits for at least
36 months. The re-evaluation routine followed the CAE Trauma protocol, which consists
of serial consultations at 15 days, 1 month, 3 months, and 6 months after the procedure,
in addition to an annual review. The time elapsed between the treatment and the occurrence
of a complication (lack of consolidation or AFHN) was obtained for all of the patients,
and was expressed in months.
The consolidation of the fracture was determined in all of the patients through clinical
examination and analysis of AP and lateral proximal femoral radiographic images. This
evaluation was performed by three CAE Trauma surgeons, who independently reviewed
all of the radiographic documentation, separated into unidentified envelopes. Consolidation
was defined by the absence of symptoms during the clinical examination and by the
functional recovery of the patient associated with a radiographic image of bone formation
and the disappearance of the fracture line at the lesion site in the femoral neck.[13]
The clinical diagnosis of lack of consolidation was established in patients with clinical
symptoms characterized by progressive pain and functional disability of the operated
limb. The radiographic diagnosis of lack of consolidation was made through the observation
of a persistent fracture line in the femoral neck associated with signs of implant
failure and of loss of reduction.
The diagnosis of AFHN was confirmed based on the classification by Ficat et al[14] and on conventional radiographs in AP and lateral views. The clinical diagnosis
of AFHN was confirmed in patients with progressive inguinal pain aggravated by physical
exertion.
After a serial evaluation at 36 months postsurgery, the patients with consolidated
fractures who resumed daily activities were allocated at the Consolidation group.
The patients who needed surgical revision, either due to AFHN or to lack of consolidation,
constituted the Failure group.
The Research Ethics Committee from the Instituto Nacional de Traumatologia e Ortopedia
approved the terms of the present study, which was exempted from Informed Consent
Form signature requirements.
Numerical data (quantitative variables) were expressed as mean ± standard deviation
(SD), median, minimum, and maximum values. Categorical data (qualitative variables)
were expressed as frequency (n) and percentage (%). The distribution analysis of the numerical data (gaussian or
normal distribution) was determined by the Shapiro-Wilk test. The comparison of treatment
groups regarding age and time between the fracture and the surgical treatment was
performed by the Mann-Whitney test, because these variables presented a non-normal
distribution. The time elapsed between the surgery and the detection of failure was
compared by the Student t-test. The Fisher exact test was used to analyze categorical
data. The relative risk of development of failure was determined according to the
reference parameters. The statistical analysis was performed in the GraphPad Prism
version 5.00 for Windows software (GraphPad Software, San Diego, CA, USA).
The present project was authorized by the Research Ethics Committee of our institution
and it was approved under the number 919.66.
Results
The 81 patients were allocated into two groups: Consolidation or Failure; the latter
group was subdivided into AFHN or lack of consolidation, according to the reason for
therapeutic failure. Thus, 67 patients were included in the Consolidation group, and
14 in the Failure group ([Fig. 1]).
Fig. 1 Experimental design, final sample and groups according to the outcome of the treatment.
DHS, dynamic hip screw.
The median age was 44 years old (range: 16–60 years old) in the Consolidation group,
and 45 years old (range: 19–59 years old) in the Failure group, with no statistical
difference (p = 0.955).
The two groups were homogeneous regarding gender, diabetes mellitus, smoking, and
surgical risk according to the American Society of Anesthesiologists (ASA). These
variables did not influence the surgical outcome in the studied groups ([Table 1]).
Table 1
|
Variable
|
Category
|
Consolidation
(n = 67)
|
Failure (n = 14)
|
p-value
[a]
|
|
n
|
%
|
n
|
%
|
|
Gender
|
Male
|
|
|
10
|
71,4
|
1.000
|
|
Female
|
22
|
32.8
|
4
|
28.6
|
|
|
Diabetes mellitus
|
Yes
|
4
|
6.0
|
0
|
0
|
1.000
|
|
No
|
63
|
94.0
|
14
|
100
|
|
|
ASA[b]
|
I
|
28
|
53.8
|
6
|
50.0
|
1.000
|
|
II and III
|
24
|
46.1
|
6
|
50.0
|
|
|
Smoker
|
Yes
|
16
|
23.9
|
5
|
35.7
|
0.502
|
|
No
|
51
|
76.1
|
9
|
64.3
|
|
Time Elapsed between the Fracture and the Surgical Treatment
The median time elapsed between the fracture and the surgery in the Consolidation
group was of 17 days (range: 3–40 days). In the Failure group, the median was of 17
days (range: 6–42 days). No statistically significant difference was observed between
the groups (p = 0.648) ([Fig. 2A]).
Fig. 2 Follow-up of patients with surgically-treated femoral neck fractures who evolved
with therapeutic failure (avascular femoral head necrosis or lack of consolidation)
or with bone consolidation. A, time elapsed between fracture and surgery. B, hospital
length of stay. The horizontal lines at the ends of the rectangles represent the interquartile
range values, and the extremities of the bars represent the minimum and maximum values
found in each group.
There was no difference in the length of hospital stay between the groups. The median
period was of 6 days (range: 1–31 days) for the Failure group, and of 6 days (range:
2–41 days) for the Consolidation group ([Fig. 2B]).
Influence of the Initial Fracture Deviation on the Outcome
Only 3 patients (3.7%) had no fracture deviation. All (100%; 14/14) of the patients
from the Failure group presented an initial fracture deviation, whereas 95.5% patients
(64/67) from the Consolidation group presented a deviation; there was no statistical
difference (p = 1.000) ([Fig. 3]).
Fig. 3 Influence of femoral neck fracture deviation on the outcome of the surgical treatment
in young patients. Fisher exact test. RR, relative risk (confidence interval).
Quality of the Reduction of the Fracture
Regarding the quality of the reduction of the fracture, all of the patients evolving
with bone consolidation had a satisfactory reduction (67/67). In the Failure group,
the reduction was satisfactory in 5 patients (35.7%), and unsatisfactory in 9 individuals
(64.3%). The risk of failure was 8.37 times greater when the reduction was inadequate
(p < 0.0001) ([Fig. 4]).
Fig. 4 Surgical treatment outcome of femoral neck fractures in young patients as a function
of fracture reduction. The reduction was considered satisfactory when the alignment
was 160° in anteroposterior views and 180° in profile views. Fisher exact test. RR,
relative risk (confidence interval).
Positioning of the Implants
In the Consolidation group, 95.5% (64/67) of the patients presented adequate implant
placement; in contrast, in the Failure group, the positioning was inadequate in 85.7%
(12/14) of the cases (p < 0.001). The correct positioning was a protective factor for failure, as shown by
the relative risk of 0.04 (confidence interval [CI]: 0.01–0.15) ([Fig. 5]).
Fig. 5 Surgical treatment outcome of femoral neck fractures in young patients as a function
of implant positioning. Fisher exact test. RR, relative risk (confidence interval).
The fracture was classified as deviated in the six patients from the Failure group
who presented with AFHN. Among them, 2 patients presented satisfactory reduction with
adequate implant placement. In the remaining four patients, the reduction was unsatisfactory,
and the positioning of the implant was inadequate.
Time Elapsed between the Treatment of the Fracture and the Occurrence of Complication
The time elapsed between the surgical treatment and the detection of therapeutic failure
ranged from 5.7 to 49.1 months. Analyzing the outcomes separately, it was observed
that this period was higher for patients with AFHN (24.4 ± 11.4 months) compared to
those with no bone consolidation (11.3 ± 4.2 months) ([Fig. 6]). All of the six cases of AFHN were classified as stage III according to the classification
by Ficat et al,[14] that is, loss of femoral head sphericity with no acetabular lesion ([Fig. 7]).
Fig. 6 Time between surgery and diagnosis of therapeutic failure. AFHN, avascular femoral
head necrosis.
Fig. 7 Avascular femoral head necrosis. A, patient with satisfactory reduction and adequate
implant positioning. B, Patient with unsatisfactory reduction and inadequate implant
positioning.
All of the patients with no consolidation had a deviated fracture. Of these, reduction
was satisfactory in three patients, and it was unsatisfactory in five individuals.
The positioning of the implant was considered inadequate in all of these cases ([Fig. 8]).
Fig. 8 Radiographic evolution of a femoral neck fracture in a patient with no consolidation.
A, Deviated fracture. B, Surgical treatment with satisfactory reduction and inadequate
implant positioning. C, Evolution to lack of bone consolidation and osteosynthesis
failure.
Discussion
Preservation of the femoral head by osteosynthesis is desirable in young patients
with fractures of the femoral neck, because bone consolidation occurring in the absence
of AFHN leads to satisfactory results, preserving the anatomy and the function of
the joint.[15] However, due to fracture instability and to the difficulty in obtaining a stable
assembly for the maintenance of the reduction, it can be very challenging due to the
high incidence of failure.[16] The failure rates of therapeutic reduction depend on some factors that may be controlled,
such as the quality of the reduction and the positioning of the osteosynthesis implants.[9] Other factors are independent and could also influence the final outcome, such as
the initial fracture deviation and the time elapsed between the fracture and its surgical
treatment.[8]
The ideal timing for the surgical correction of fractures of the femoral neck is controversial.
Authors advocating early surgery suggest that the main advantage of immediate reduction
of the fracture is the decrease of the potential compression of retinacular vessels
by deviated fragments and the decompression of the hematoma, which increases the intracapsular
pressure, improving the blood flow to the femoral head and minimizing the risk of
the development of AFHN.[8] These authors reviewed retrospectively the early fixation of fractures of the femoral
neck within 12 hours and the delayed fixation after > 12 hours in 38 patients with
a mean age of 46 years old. Avascular femoral head necrosis occurred in 16% of the
patients submitted to late fixation. In contrast, Upadhyay et al,[17] in a prospective and randomized study with 92 patients < 50 years old presenting
with fractures of the femoral neck, did not observe a significant difference in the
consolidation rate and in the incidence of AFHN compared with surgeries performed
within 48 hours after the trauma.
In our study, the time elapsed between the occurrence of the fracture and its surgical
treatment did not influence the clinical outcomes. Similar results were obtained in
other series,[9]
[18] which also did not observe an association between early fixation and favorable lesion
evolution.
In a review by Papakostidis et al,[19] the authors concluded that late fixation, after > 24 hours, is related to an increase
in the incidence of pseudarthrosis, but they did not find a relation with the development
of AFHN. Our study did not observe a higher incidence of pseudarthrosis in patients
submitted to a later fixation. However, our sample consisted of patients whose fractures
occurred > 3 days previously, compromising the comparison with the results of the
aforementioned review.[19]
An important aspect to be discussed is hematoma drainage. The classic report by Swiontkowski
et al[20] advocates capsular decompression as an important practice to avoid AFHN. These authors
believed that an increased capsular pressure would lead to local vascular injury and
to subsequent bone tissue necrosis. However, Wong et al[21] showed that this type of decompression should be discouraged, since they did not
observe an association between previous hematoma drainage and a reduction in the rate
of necrosis. These authors related the incidence of necrosis to the initial trauma
and to the grade of deviation of the fracture, which account for the rupture of the
retinacular vessels at the moment of the injury.
Capsular hematoma drainage was not performed in our study; in addition, the AFHN rates
were not high, which was consistent with the literature. Kakar et al[22] published a multinational study, which corroborated our practice, in which they
evaluated the routine of surgeons in the treatment of fractures of the femoral neck.
In their study, 90% of the surgeons reported failure to aspirate a fracture hematoma
prior to the osteosynthesis procedure, emphasizing that the quality of the reduction
and that the implant used would be the most important factors for satisfactory outcomes.
Some authors believe that the initial deviation of the fracture, which is associated
with the energy grade of the trauma, is directly related to the operative outcome.[23]
[24]
[25] In our series, this factor could not be proven, since only 3 patients from the total
sample (n = 81) had nondeviated fractures. Since this sample is composed by young patients
with better bone quality, it is possible that a high- or moderate-energy trauma would
be necessary to cause a deviated fracture. We believe that the initial deviation may
be related to ischemia, and that the AFHN is due to the intimate regional anatomic
relationship with retinacular vessels, that is, the high energy of the trauma. However,
our study does not allow us to affirm or to dismiss this correlation.
The reduction within the patterns described by Garden[11] seems to be a determining factor in the postoperative evolution of fractures of
the femoral neck submitted to osteosynthesis. Several studies report higher success
rates when a good reduction is obtained.[8]
[26]
[27]
Our study found similar results, since a satisfactory reduction had a strong association
with consolidation of bone fractures. In patients with inadequate reduction, the chance
of failure was 8.3 times higher when compared with patients with adequate reduction.
We believe that the correct alignment of the fracture with varus correction and retroversion
is paramount in the treatment of fractures of the femoral neck, because it provides
a favorable mechanical environment for the consolidation of the bone.[9]
Conclusion
The most important factors influencing the final result of osteosynthesis in fractures
of the femoral neck in young patients are the quality of the reduction and the correct
application of the surgical technique. The time elapsed between the fracture and the
surgery does not seem to be related to the surgical outcome. Our study could not establish
if the initial deviation was a determining factor in the surgical outcome.
