Key-words:
Minimally invasive spine surgery - regional sagittal angle - thoracolumbar fracture
Introduction
Traumatic spinal fractures are common, with a constantly increasing rate of referrals
to neurosurgical tertiary centers in the United Kingdom. These injuries can result
in potentially devastating consequences including pain, deformity, and neurological
deficits.[[1]],[[2]],[[3]],[[4]],[[5]] In addition to the physical impairment, the long-term effects of spinal injuries
may also have a significant psychological, economic, and social impact.[[6]],[[7]],[[8]],[[9]] Surgical fixation is a well-established treatment option for spinal fractures,
with the aim of achieving spinal stability and preservation of neurological function.
Open pedicle screw fixation is the conventional technique to promote fusion and restore
stability. This is sometimes used in combination with other procedures such as interbody
fusion or vertebroplasty. Open fixation techniques are an effective treatment for
spinal fractures across cervical, thoracic, and lumbar spine and for degenerative
conditions. However, they are often associated with considerable morbidity such as
significant postoperative back pain, increased duration of surgery, and prolonged
hospital stay. Damage to the surrounding neurological structures, incomplete or suboptimal
stabilization, and instrumentation failure have been described as well, together with
high infection rates and high blood loss.[[2]],[[3]],[[4]],[[6]],[[10]],[[11]]
Most recently, pedicle screw fixation using a minimally invasive surgical (MIS) approach
has emerged as an alternative approach for the treatment of thoracolumbar fractures,
aiming to minimize soft-tissue injury and perioperative morbidity.[[7]],[[8]],[[10]] MIS utilizes fluoroscopic imaging or computed tomography to guide percutaneous
placement of pedicle screws and instrumentation in order to achieve the reduction
and fixation of spinal fractures. This technique reduces healing times, postoperative
analgesia requirements, and intraoperative blood loss while achieving a comparable
level of fixation and stabilization in the majority of cases. Although more studies
are required, so far the results of MIS fixation for the management of thoracolumbar
fractures are promising.[[1]],[[9]],[[10]],[[11]]
The aim of this study is to collect data regarding epidemiology, management, and outcomes
of patients treated with MIS pedicle screw fixation for traumatic thoracolumbar fractures
in our neurosurgical department in the United Kingdom.
Materials and Methods
All the patients who underwent MIS surgery for traumatic thoracolumbar fractures at
Royal Preston Hospital from March 2013 to March 2017 were included in this study.
Patients' data, operation notes, and clinical details including intraoperative and
postoperative complications were collected retrospectively from patients' notes and
our online database. We exclusively included patients who were not deemed suitable
for conservative management of the fracture, in particular, either patients with three-column
injury or new neurological deficit on presentation. Stable osteoporotic fractures
were not included in this study as they are normally treated conservatively in our
center.
Causes of injury were grouped into six categories: fall from a height, fall from standing
height, fall downstairs, road traffic accident, fall from a horse, and blunt trauma.
The fracture level was identified from radiological images and broadly divided into
(a) thoracic and (b) lumbar. Thoracic fractures were further subdivided into upper-thoracic
(T1–T4), mid-thoracic (T5–T8), and lower-thoracic (T9–T12); lumbar fractures were
grouped into L1, L2, L3, and L4.
The Frankel Grading system was used to classify the degree of neurological deficit.
The degree of preoperative pain was assessed using the Visual Analog Scale (VAS).
Patients were asked to grade their pain at the time of admission on a scale from 0
to 10. Zero is equivalent to no pain, 1–3 is mild pain, 4–6 is moderate pain, and
7–10 is considered severe pain.
All the pedicle screws were placed with fluoroscopy guidance, using the standard technique
for percutaneous pedicle screw placement. All the cases were performed by a group
of senior spinal neurosurgeons. Depending on the number of levels fixed during MIS
surgery, the procedure was categorized as: (A) 1 level above and below the fracture,
(B) 2 levels above and below the fracture, and (C) short pedicle screws at fracture
level in addition to 1 level above and below.
The length of stay (LOS) in the hospital was calculated in days from the day of admission
to the date of discharge from the neurosurgical unit. All patients who were transferred
to spinal rehabilitation or were repatriated to the local hospital were considered
as discharged. Outpatient follow-up was conducted at 2, 6, 12, and 24 months after
surgery in the majority of patients. A few patients were discharged earlier than 24
months based on surgeons' preference and patient condition. The overall functional
outcome was assessed for each patient at the end of the follow-up period. Patients
were divided into three groups: good functional outcome (able to perform daily activities
independently and return to functional baseline), restriction in the daily activities,
and poor functional outcome (unable to perform daily activities independently).
The regional sagittal angle (RSA) was measured preoperatively in the immediate postoperative
period and on the final weight-bearing X-ray. The difference in angle between the
first and last radiographs was tabulated in order to assess the degree of correction
of spinal deformity. The lateral radiographs of all the patients were analyzed using
the digital radiography software on our picture archiving and communication system
using a magnified image. A positive RSA indicates lordosis, and a negative angle indicates
kyphosis as described in previous studies [[Figure 1]].[[12]]
Figure 1: A - Local sagittal angle; B - Regional sagittal angle 12
Results
A total of 125 patients underwent MIS fixation with percutaneous pedicle screws between
March 2013 and March 2017. Sixty-one patients were male (49%) and 64 were female (51%);
the mean age was 59 years with the youngest patient being 17 years old and the oldest
79.
The majority of injuries were from falls. Forty-five patients fell from a height (36%),
26 fell downstairs (20.8%), 25 were involved in a road traffic accident (20%), eleven
fell from standing height (8.8%), ten patients fell from a horse (8%), and in eight
cases, the cause of fracture was blunt trauma (6.4%).
All the patients had a single-level traumatic thoracic or lumbar fracture. In 48 cases,
the fracture involved a thoracic vertebra (38.4%) and in 77 cases a lumbar vertebra
(61.6%). More than half of the fractures (53%) were at the thoracolumbar junction
(either T12 or L1). Among the thoracic fractures, 4 (3.2%) were upper-thoracic, 13
(10.4%) were mid-thoracic and 31 (24.8%) were lower-thoracic fractures. Forty-four
patients had a fracture at L1 (35.2%), twenty at L2 (16%), nine at L3 (7.2%), and
four at L4 (3.2%). No patient had a fracture at L5 level [[Figure 2]].
Figure 2: Distribution according to fracture level
Sixteen patients (12.8%) presented with a neurological deficit on admission on our
neurosurgical unit. Two patients were grade A according to the Frankel Grading classification,
nine were Grade C, and five were Grade D. All patients with neurological deficits
underwent posterior decompression at the same time of the MIS fixation. The five patients
who presented with Grade D all improved to normal power. Out of the nine patients
with Grade C, two improved to normal, five improved to Grade D, and two remained unchanged.
All the patients who presented with Grade A deficit failed to show any neurological
improvement [[Table 1]].
Table 1: Neurological deficits according to Frankel Grading system
Eighty-five patients (68%) had severe preoperative pain, 33 patients (26.4%) had moderate
pain, and 7 patients had mild pain (5.6%). Postoperatively, 82 patients (65.6%) had
no pain and 13 (10.4%) reported mild pain. Twenty-four (19.2%) patients had persistent
moderate back pain; three patients (2.4%) had severe back pain. Patients with moderate
and severe pain were investigated with further imaging. Most patients did not have
any pathology amenable to surgical intervention and were, therefore, referred to the
pain specialists. One patient needed a vertebrectomy with cage placement. One patient
had pain over the screw site and was offered an operation to remove the screw but
declined it [[Figure 3]].
Figure 3: Pre and Post-operative pain status
Ninety-five patients (76%) achieved a good functional outcome and were able to go
back to work and their normal daily activities. All the patients with absence of pain
or mild postoperative pain belong to this group. The patients with moderate pain had
some restriction of their daily activities, and all the patients with severe pain
belong to the poor functional outcome group. Three patients were lost to follow-up
at their last clinic review, so the final outcome was not measured in their case.
Sixty-four patients (51.2%) underwent fixation one level above and one level below
the fracture; 48 (38.4%) had fixation two levels above and two levels below, and 13
(10.4%) had short pedicle screws inserted at the fracture level along with fixation
one level above and one below. Four patients underwent vertebroplasty at the fractured
levels or augmentation of pedicle screws in addition to fixation. The procedures were
performed by either a consultant or a senior spinal fellow in all cases, using a standard
technique as previously described in the literature. [[Figure 4]] shows postoperative standing X-rays of patients at their second follow-up at 6
months.
Figure 4: Examples of post-operative X-rays at 6 months. Source: Royal Preston Hospital, Department
of Neurosurgery
The mean preoperative RSA of thoracic fractures was −13.5°, the mean postoperative
RSA was −5.7°, and at the end of the follow-up period, it was −6.5°. The mean preoperative
RSA of lumbar fractures was −4.2°, the mean postoperative RSA was +3.1°, and at the
end of the follow-up period, it was +2.8° [[Table 2]].
Table 2: Regional sagittal angle preoperatively, postoperatively, and at the end of the follow-up
period
Table 2: Contd...
Ninety-three patients were discharged home, 19 were sent to their local hospital for
short-term rehabilitation prior to discharge, and 13 required long-term spinal rehabilitation
at a dedicated spinal cord injury center. The average LOS in the hospital was 14 days
(2–118 days).
Six patients (4.8%) had a wound infection; 5 of them were treated conservatively with
antibiotics and one patient required surgical debridement. Three patients had pneumonia.
In two patients, the upper-level screws pulled out, and in two cases, one screw was
misplaced. These four patients underwent surgical intervention with either replacement
or removal of the screws.
Discussion
The aim of surgical intervention following traumatic spinal fractures of the thoracolumbar
spine is to achieve spinal stability, prevent further loss of neurological function,
facilitate early mobilization, and obtain the best possible outcome while minimizing
pain and morbidity.[[7]],[[8]],[[10]] Open pedicle screw fixation is an effective treatment option for these fractures;
however, it is associated with a significant degree of postoperative pain and prolonged
hospital stay. In our single-center retrospective study, we collected evidence to
support that MIS fixation is also a safe and less invasive option to manage these
types of fractures.
We include a significant number of patients presenting with traumatic thoracolumbar
fractures caused by a variety of mechanisms of injury. Our cohort of patients includes
a wide age range, an equal distribution of gender, and different degrees of preoperative
fitness, and it is overall a good representation of the general population, making
our results relevant to other neurosurgical centers. We excluded patients deemed suitable
for conservative management, for example, osteoporotic fractures in the elderly. However,
11 patients were elderly and sustained a fall from standing; hence, it is possible
that they had undiagnosed osteoporosis.
Our outcomes following surgery are good. More than 10% of the patients presented with
a neurological deficit on admission. Among these patients, 75% showed postoperative
improvement in their neurology, including 7 patients who regained normal function.
No patient had worse neurology postoperatively. No patient died within 30 days after
surgery, and as a result of good pain control and early mobilization, only a very
small number of patients (3) developed medical complications such as pneumonia. The
infection rate was comparable to open surgery infection rate and other MIS studies
previously published in the literature (around 4%). The average length of hospital
stay was 14 days. Ten patients were medically fit for discharge few days after surgery,
but they stayed in the hospital for more than 100 days each due to social issues.
Some of the patients had prolonged hospital stay due to additional occupational therapy
requirements. If we exclude these ten patients, the average length of hospital stay
for 115 patients was 6.3 days which is better than reported hospital stay in open
surgery. It would be interesting to collect further data and differentiate between
hospital stay related to the surgical procedure and hospital stay related to social
issues. Spinal stability was achieved in every patient at the end of the follow-up
period. We have not collected data related to fusion as the aim of the fixation was
mainly neurological decompression and stabilization.
MIS fixation achieved a satisfactory RSA angle postoperatively in all patients. In
fact, the natural kyphosis of the thoracic spine and lordosis of the lumbar spine
were maintained, while improving the initial pathological RSA. The vast majority of
patients had no or mild postoperative pain (76%) and achieved a good functional outcome,
being able to go back to their preinjury baseline. Our study demonstrates that MIS
fixation achieves adequate correction in RSA, which is linked with good surgical outcome,
excellent pain relief, and early return to normal activities. Minimally invasive approaches
dramatically decrease paraspinal musculature stripping, leaving the attachment of
the muscle to the bone intact, reducing ligament disruption, providing direct access
to the transverse processes and pedicles, and hence decreasing bleeding and postoperative
pain.[[13]],[[14]],[[15]] Other previous studies, for example, Jian et al.[[15]] have shown that percutaneous fixation of thoracolumbar fractures results in equivalent
or better clinical outcomes when compared to the open group.
Indications for minimally invasive spinal fixation have expanded in the recent years
and are now comparable to those for open surgery.[[16]],[[17]],[[18]],[[19]] Surgeons are becoming increasingly experienced in this field; however, we must
acknowledge that most MIS spinal techniques have a steep learning curve and other
authors suggest that the operating surgeon must have adequate experience with open
procedures before attempting minimally invasive approaches.[[19]],[[20]],[[21]] For the same reason, depending on the surgeon's experience, MIS fixation may in
some cases take more time to perform than open surgery and may result in inadequate
restoration of vertebral body height and RSA.[[16]],[[20]],[[22]] It is also worth mentioning that MIS techniques often require a not irrelevant
amount of fluoroscopy; consequently, appropriate training to reduce patient and occupational
radiation exposure is necessary.[[23]],[[24]],[[25]]
According to previous studies assessing the adequacy of screw placement in MIS, up
to 98% of screws were reported to be in a good or excellent position.[[26]],[[27]] Similarly, in our study, we found that 96.8% of screws were in an excellent position.
The screws that we placed had less tendency to violate the lateral wall of the pedicle
compared to other MIS studies, giving similar results to open fixation. In 2013, a
study by Dong et al., and in 2016 McAnany et al, on short-segment percutaneous screw
fixation demonstrated that the curative effect achieved by both open and percutaneous
approaches are similar, with no difference in radiological outcomes.[[13]],[[28]]
The strength of our study is the high number of patients and screws when compared
to other publications available in the literature. The retrospective nature of the
study is a limiting factor. Another limitation of our study is the lack of a comparison
group. We included exclusively patients who were already considered unsuitable for
conservative management, and we did not collect data regarding open fixation in our
center. Blood loss, surgical time, and vertebral height postoperatively have not been
assessed in our research project, despite being relevant variables when comparing
open surgery to minimally invasive techniques.[[28]] New prospective studies with a higher number of patients and screws, potentially
including cases with fractures at multiple levels and quantifying the above-mentioned
variables (blood loss, surgical time, and vertebral height) may be useful to further
assess the outcomes of minimally invasive techniques.
Conclusions
Our findings are comparable to other studies available in the literature and confirm
that MIS fixation is a safe surgical option with outcomes comparable to open surgery
and a potential reduced morbidity. MIS surgery achieves a rapid and significant improvement
in VAS score, functional outcome, Frankel Grade, and RSA. With technological advancements,
we expect that MIS fixation will become the predominant technique in the management
of unstable traumatic thoracolumbar fractures.
