Đặc điểm hình thái tổn thương và kết quả phẫu thuật điều trị gãy cột sống ngực, thắt lưng đa tầng tt tiếng anh

MINISTRY OF EDUCATION
AND TRAINING
DEFENSE

MINISTRY OF
NATIONAL

MILITARY MEDICAL UNIVERSITY

LE HUU TRI

MORPHOLOGICAL CHARACTERISTICS OF
MULTILEVEL THORACIC AND LUMBAR
VERTEBRAL FRACTURE AND THE
EFFECTIVENESS OF SURGICAL TREATMENT.

Speciality: Surgery
Code: 9720104

SUMMARY OF MEDICAL THESIS

HA NOI - 2020


THESIS IS COMPLETED AT
MILITARY MEDICAL UNIVERSITY

Supervisor:
1. Associate Prof. PhD. Vu Van Hoe
2. Associate Prof. PhD. Vo Van Nho

Reviewer 1: PhD. Hoang Gia Du
Reviewer 2: Associate Prof. PhD. Bui
Van Lenh
Reviewer 3: Associate Prof. PhD. Nguyen
Tho Lo

The thesis will be defended by University’s Council at:

The thesis can be found at:
1. National Library


3
INTRODUCTION
Multilevel noncontiguous spinal fractures (MNSF) are
defined as fractures of the vertebral column at more
than one level. According to many reports, multilevel
noncontiguous spinal fractures are found at from 3.2%
to

16.7%.

Multilevel

spine

injuries

are

often

uncommon, occur commonly as a result of high-speed
road traffic accidents or falls from a height. Multilevel
spine injuries tend to be more severe than unilevel
because of the other injuries combined. Rapid diagnosis of
MNSF is essential since a misdiagnosis or delayed diagnosis may
complicate the clinical picture. Therefore, the role of a
thorough physical and radiological examination is the
key to not to miss any lesions. Although there are
many studies evaluating and treating single level spine
fractures, there have not been many reports of
multilevel spine injuries. In Vietnam, there are not
many researches on this issue. So we work for:
“Morphological characteristics of multilevel thoracic and
lumbar vertebral fracture and the effectiveness of surgical
treatment” with these targets:
1. Describing the morphological characteristic of
multilevel
fractures

thoracic
in

which

and

lumbar

surgical

vertebral

treatments

are

indicated at Da Nang Hospital.
2. Evaluating the outcome of surgical treatment
for multilevel thoracic and lumbar vertebral
fractures at Da Nang Hospital.


4
-

New main scientific contributions of the thesis:
The research subjects are lesions in multilevel thoracic

-

and lumbar vertebral fractures
Providing the evidence of the effectiveness of surgical
treatment for multilevel thoracic and lumbar vertebral
fractures

- Structure of the thesis: the thesis includes 127
pages with 40 tables, 27 pictures and 10 charts.
Introduction (2 pages); Chapter 1: Overview (31

pages); Chapter 2: Subjects and Methods of the
study (26 pages); Chapter 3: The results of the
study (31 pages); Chapter 4: Discussion (33
pages); Conclusion (2 pages); Petition (1 page);
List of published articles related to the results of
the thesis (1 page); References (138 documents
including

33

documents

in

Vietnamese,

105

documents in English); The appendices.
SECTION 1
OVERVIEW
1.1.

A

brief

history

of

treatment

of

spine

fractures.
1.1.1. In the world
Multilevel spine injuries have been reported for a
long time. Griffith H.B., Gleave J. R. W, Taylor R. G.
(1966) reported 5 patients accounting for 3.2% of 155
cases of thoracic and lumbar spine fractures. Lizbeth


5
C. A. M. G. et al. (2018) reported 47 cases of multilevel
spinal surgery at the Center Dr. Manuel Dufoo Olvera
in Mexico. Thus, the multilevel spine fractures have
been studied for a long time, but they are still
separated. There are no single reports on multilevel
spinal injuries.
1.1.2. In Vietnam
Previous studies inside the country have only
studied one-level fractures, there were some authors
who mentioned several cases of multilevel spinal
injuries.

And

there

are

no

separate

studies

on

multilevel thoracic and lumbar spinal injuries.
1.2. Classification of multilevel thoracic and
lumbar veterbral fractures
1.2.1. Classification of Denis
In 1983, Denis introduced the definition of "threecolumn spine": the anterior, middle and posterior
column; and it has been widely applied.
1.2.2. Classification of Margel (AO)
In 1994, Margel proposed a classification according
to AO (Arbeitsgemeinschaft fur Osteosynthesefragen)
which mainly assess spinal morphological damage.
1.2.3. Load Sharing Classification (LSC)
In 1994, Mc Cormack and colleagues released a new
classification to assess vertebrae damage based on
three criteria: the amount of damaged vertebral body,
the spread of the fragments in the fracture site, the
amount of corrected traumatic kyphosis.

6
1.2.4. Classification of multilevel thoracic and
lumbar spinal injury based on the severity of
lesion.
In

2005

Vaccaro

A.R.

et

al.

proposed

a

Thoracolumbar Injury Classification and Severity Score
(TLICS) which is is based on three major categories,
known as parameters: injury morphology, posterior
ligamentous complex integrity, patient neurology.
1.4. Imaging Tests
1.4.1. X-ray
X-rays are the fastest technique to assess the
overall picture of the spine and bone anatomy.
- To locate lesions.
- To identify changes in the overall spine: angulation,

displacement,

spondylolisthesis,

and

rotation

by

evaluating over 3 lines in a straight posture (spinous
process line, pedicles line and lateral border).
- To measure local Kyphotic Angle (LKA) and Cobb
Angle.
1.4.2. Computed Tomography Scan.
Computed tomography (CT-scan) provides greater
details and resolution for evaluation of the bone
elements and assessment of the entire spinal canal.
CT-scan allows us to assess bone lesions, fracture
lines, fracture fragments, lesions of processes, joints,
holes, posterior arcs, damaged fragments moving into
the spinal canal, and spinal stenosis.
1.5. Methods of Spine Stabilization.
1.5.2. Posterior Stabilization Systems


7
Posterior stabilization systems can restore vertebral body height
by distraction forces. Furthermore, anterior and middle columns
maintained their normal length during correction of kyphosis.

Distractive forces provided by the posterior stabilization system
developed a tensile strength in the posterior longitudinal ligament
which pushes back the retro-pulsed bone fragments forward. This
process has been termed ligamentotaxis and it is beneficial
particularly if performed at the early period.
1.5.2.3 Posterior lumbar interbody fusion surgery
(PLIFs).
Transpedicular screw is now the standard of spine
fusion surgery for many spinal diseases. We use a
"free hand" technique with screws are parallel to the
joint of the vertebra and we use X-rays for guiding
during surgery.
1.6.

Indications

of

surgical

treatment

for

multilevel thoracic and lumbar fractures
Greenberg

M.S.

(2010)

refers to

a

treatment

indication based on a clear, complete description of
the morphological characteristics of lesions in three
columns, signs of neurological damages and their
associations. They also consider the characteristics of
the fracture group as well as the location of fractures
in assessing the instability of the injury. In this
document,

author

McAfee

presents

the

specified

surgery in the case.
1.7. Evaluation the results of surgical treatment
1.7.1. The assessment of the results of spinal
correction and stabilization surgery

8
Anteroposterior (AP) and oblique X-rays on all patients;
measuring local LKA and Cobb Angle on oblique X-rays
is the most commonly used method today.
1.7.2.

The

assessment

of

the

neurological

recovery after surgery
Based on the neurological damage classification
table according to Frankel: Pre-operative evaluation,
post-operative evaluation, and evaluations in follow-up
visits, monitoring changes in Frankel level between two
consecutive visits.
CHAPTER 2
SUBJECTS AND METHODS OF STUDY
2.1. Subjects

2.1.1.

The criteria for selecting research patients

- All patients diagnosed with multilevel vetebral

fractures (≥ 2 levels) from the fourth thoracic vertebra
to the fifth lumbar vertebra, having unstable fracture,
and caused by trauma. These include multilevel
vertebral fractures emphasized on thoracic-thoracic,
thoracic-lumbar, and lumbar-lumbar segments.
-All patients who have plain and oblique X-rays,
computerized

tomography

scan

and

posterior

stabilization of at least one fractured vertebra body.
2.1.2. Exclusion criteria
-Elderly patients with osteoporosis, bone tumors,
tuberculosis, cardiovascular disease, coagulopathy,


9
etc. Patients with brain sequelae or brain entity
pathologies, spinal cord sequelae or paralysis of
peripheral nerves in the lower extremities due to
pathology,

mental

disorders

that

distort

nerve

assessment at the spinal cord. Patients who had
surgery then lost information, did not come back for
examination, or did not cooperate with treatment.
2.2. Research method
2.2.1. Sample size
Convenience

sampling

method,

including

all

patients who match selection and exclusion criteria
during the research period.
2.2.2. Method of data collection
Clinical and radiological findings : routine X-ray, CTscan; the results of treatment are collected based on a
pre-set form. Direct clinical examination, assessment

of symptoms, analysis of results of X-rays, CT scans
before surgery.
Perform posterior stabilization surgery, evaluate the
postoperative results. All patients receive discharge
instruction in self-care and mobilization.
Re-examination after 06 months: examination of
clinical symptoms such as movement on ASIA scale,
sensation, reflexes, urination, incisions, neurological
recovery according to Frankel level, urinary tract
infections, pneumonia, muscle atrophy, regular X-ray


10
measuring KLA, height of fractured vertebra, screw
position.
2.3. Method of Data Analysis
Collected data were processed in the statistical
software 12.2.1.0.


11
CHAPTER 3
RESEARCH RESULTS

3.1.

Clinical

and

multilevel

radiological

thoracic

characteristics

and

lumbar

of

vertebral

fractures.
3.1.1. General characteristics
3.1.1.1. Age, gender, occupation
Age group 20-19 accounted for the majority of cases
with 19 patients (35.84%). The youngest age 16, the
oldest

age.

Average

age

was

37.47

±13.47.

Male/Female ratio: 4.88/1.
Workers accounted for the higest rate with 26
cases (49.05%). The majority were workers and
freelancers with 36 cases, accounting for 67.85%.
3.1.2.

Clinical

conditions

of

patients

on

admission
3.1.2.6. The degree of nerve damage according to
Frankel
Table 3.5. Patient classification according to Frankel
neurological damage

Frankel
A
B

C
D
E
Total

No.
Patients
(n)
10
2
7
4
30
53

Percentag
e (%)
18.86
3.77
13.20
7.54
56.63
100

p<
0.01


12

13
3.1.2.7. The associated injuries
Table 3.6. The associated injuries
Associated
lesions
No
Abdominal
trauma
Thoracic Trauma
Head Trauma
Extremity
Fractures

No.
Patients
(n)
17
2

Percenta
ge (%)

7
3
7

32.07
3.77

p<
0.01

13.20
5.66
13.20

3.1.3. Radiological findings
3.1.3.1. Detect vertebral lesions on regular X-ray and CTscan
In 53 cases of MNSF, CT-scans revealed 118
fractured vertebrae. The number of cases with 2
broken vertebrae accounted for the highest proportion
with 44 patients (83.01%), and with 4 fractures or 5
fractured vertebrae was 1 patient accounted for
1.88%.
3.1.3.2. Distribution of multilevel vertebral fractures
Chart 3.7. Distribution of multilevel vertebral fractures
3.1.3.4. Detecting contiguous or noncontiguous lesions
on CT-scans
Chart 3.8. Detecting contiguous or noncontiguous
lesions on CT-scans


14
3.1.3.6. The fragment of bone causing spinal stenosis
on CT-scans

Chart 3.9. Spinal stenosis

3.1.9.

The

correlation

between

clinical

and

radiological findings
Table 3.16. The correlation between Denis
classification of spinal fractures and
The Frankel Grade classification

Denis

Compressio
n fractures
Burst
fractures

Total
n, (%)

Frankel
A

B

C

D

E

0

0

0

0

4

4 (7.54)

4

2

5

4

23

38

(71.69)

p=
0.031
9


15
Seat-belt
fractures
FracturesDislocation
Total
n, (%)

0

0

0

0

2

2 (3.77)

6

0

2

0

1

9 (17.00)

10
2
7
4
30
(18.8 (3.7 (13.2 (7.5 (56.6 53 (100)
6)
7)
0)
4)
3)

3.1.10. The correlation between neurological
damage and degree of spinal stenosis on CTscan
Table 3.17. The correlation between degree of spinal
stenosis and degree of paralysis
Fragme
nt of
bone

Frankel

A
B
C
D
E
Non
spinal
0
0
0
0
8
stenosis
Spinal
stenosis
2
0
2
1
16
< 50%
Spinal
stenosis
8
2
5
3
6
≥ 50%
Total

10
2
7
4
30
n, (%) (18.8 (3.7 (13.2 (7.5 (56.6
6)
7)
0)
4)
3)

Total
n,
(%)

p<
0.05

8
(15.0
9)
21
(39.6
3)
24
(45.2
8)
53
(100)

3.2. Surgical results
3.2.1. Surgical factors
3.2.1.1. Time of surgery
Surgery within 3 days after accidents (72 hours)
only had 8 cases (15.08%), after 3 days accounting for


16
84.92%. The earliest time from admission to surgery
was within 1 day, the longest time from admission to
surgery was 21 days. Average time 7.37 ± 4.37 days.
3.2.1.6. Overlapping surgeries
Table 3.23. Overlapping surgeries
No.
Patients
(n)
1
4
6

Overlapping
surgeries
Cranial surgery
Thoracostomy
Fracture Fixation
Thoracostomy +
Fracture Fixation
Total

Percent
age (%)
1.88
7.54
11.32

3

5.67

14

26.41

p<
0.01

3.2.2. Surgical results
3.2.2.1.

Results

of

neurological

operation

recovery

after

according

to

The Frankel Grade classification
Table 3.24. Neurological recovery after operation

Frankel

A
B
After
Operatio C
n
D
E
Total

Total

Preopration

n, (%)

A
10

B

0

C
0

D
0

E
0

0

1

0

0

0

10
(18.86%)
1 (1.88%)

0

0

3

0

1

4 (7.54%)

0

1

2

1

0

4 (7.54%)

0

0

2

3

29

10

2

7

4

30

34 (64.18
%)
53
(100%)

n, (%) (18.8 (3.7 (13.2 (7.5 (56.
6%)

7%)

0%)

p<
0.000
1

4%) 63%
)

3.2.2.3. Results of neurological recovery after re-examination

17
The Denis Pain and Work Scale: there were 17
graded level 1 patients (32,07%) who were able to
return to work and 12 graded level 5 patients (22,66%)
completely lost working capacity.
Table 3.27. Neurological recovery after 6 months
Post operation

Total

n,
(%)

Frankel
A
10

B
0

C
0

D
0

E
0

0

1

0

0

0

0

0

1

0

0

0

0

3

0

0

0

0

0

4

34

10
(18.86
%)

1
(1.88
%)

4
(7.54
%)

4
(7.54
%)

34
(64.18
%)

A
B
Reexami
ned

C
D
E

Total

n, (%)

p<
0.00
01

10
(18.86%
)
1
(1.88%)
1
(1.88%)
3
(5.66%)
38
(71.72%
)
53

(100%)

3.2.2.4. The result of local kyphosis angle recovery
results
Table 3.28. Local kyphosis angle recovery results
Local
kyphosis
angle

Lowe
st

High
est

Mean

Preoperation

50

42

0

18.73 ± 7.380
(1)

Postoperation

10

27

0

10.56 ± 5.530
(2)

Reexamination

20

29

0

13.37 ± 5.940
(3)

Compare
(p)
p12<
0.0001
p13=
0.0001
p23=
0.0132

18


19
3.2.2.5. The results of vertebral body compression recovery after
surgery
Table 3.30. Anterior Vertebral Body Compression
Percentage Results

Index

Anterior
Vertebral
Body
Compressio
n
Percentage
(%)

Preoperation
(1)

Postsurgery
(2)

ReCompare
examinati
(p)
on (3)

38.52 ±
17.81

24.92 ±
16.03

25.30 ±
16.03

p12=
0.0001

(7.1689.39)

(0.0867.02)

(1.5976.31)

p13=
0.0001
p23=
0.9031

3.2.2.7. Overall results
Table 3.32. Overall Results
Result

No.
Patients

Percentage
(%)

Good

38

71.69

Fair

2

3.76

Average

12

22.67

Bad

1

1.88

Total

53

100

p<
0.0001


20

CHAPTER 4
DISCUSSION
4.1. Clinical and radiological characteristics of
multilevel

thoracic

and

lumbar

vertebral

fractures
4.1.1. General characteristics
4.1.2.6. The degree of neurological injuries according
to the Frankel Grade classification
In table 3.5 we find that patients with normal
sensory and motor function (Frankel E) accounted for
the highest proportion with 30 cases (56.63%). The
number of patients with symptoms of complete

paralysis (Frankel A) ranked second with 10 patients
(18.86%).
The rate of paralysis varies by author, depending on
research subjects.
4.1.2.7. Associated injuries
Patients with multilevel spinal injuries were mostly
caused by a complex, strong traumatic mechanism,
which often resulted in other traumas. Patients with
other accompanied injuries accounted for the highest
proportion with 36 cases (67.93%), in which multiple
injuries ( 2 types of combined injuries or more) had 11
patients (20.78%).
Our research had a high percentage of combined


21
injuries due to the strong traumatic mechanism of
multilevel spinal injuries that caused damage to
organs of the body.
4.1.3. Radiological findings
4.1.3.1. Detection of vertebral lesions on regular X-ray
and CT-scan
In all cases where only one fractured vertebra was
detected on X-rays, more than one fracture was
detected on CT-scans. CT-scans revealed 118 fractures
in 53 cases. (ratio: 2.22 fractures/1 patient).
Reasons for missing lesions on X-ray films: poor
image quality, not capturing the entire spine at our
request,

initial

examinations

by

inexperienced

physicians.
This shows that lesions will be more accurately
detected in CT-scanned patients than in patients with
x-rays only. This is considered the best method for
assessing lesions of bone today.
4.1.3.2. Distribution of lesions of multilevel spinal
fractures.
Figure 3.7 shows that the most common segment
among all patients suffering multilevel spinal fractures
were concentrated in thoracic+thoracic segment with
11 cases (20.76%), thoracic+lumbar segment and
lumbar+ lumbar segment with 21 cases (39.62%).
Most injuries emphasized on the thoracic+lumbar
segment and lumbar+ lumbar segment because
these segments are mobile vertebrae that are prone to


22
injury. And the thoracic segment has protection of the
ribs and back muscles so it is less vulnerable.
We found that the lesions in two separate sections
of lumbar and thoracic spine accounted for a relatively

high proportion (39.62%). Therefore, we recommend
that in case of multiple injuries, it is essential to have a
careful examination and X-ray of the entire spine to
avoid

missing

multilevel

noncontiguous

spinal

fractures to prevent complications from errors of
omission.
4.1.3.4. Detection of contiguous or noncontiguous
fractures on CT-scan
Figure 3.8. showed that there were 36 patients with
contiguous fractures, accounting for a high rate of
67.93% and 17 patients with non-contiguous fractures
(32.07%).
The rate of fracture detection in our study is
relatively high compared to other previous reports,
probably because our hospital is a general hospital,
not a specialty hospital. We not only treat spinal
injuries but also treat other injuries such as thoracic,
abdominal,

or

extremity

fractures;

and

more

importantly, our hospital has intensive unit care which
is essential for treatment of patients with severe
injuries. Therefore, most of traumatic cases were
admitted to our hospital.


23
4.1.3.6. The fragment of bone causing spinal stenosis
on CT-scan
Figure 3.9 shows that fragments of bones causing
spinal stenosis> 50% were 21 patients, accounting for
39.63%. Bone fragments causing spinal stenosis ≥ 50%
were 24 patients, accounting for 45.28%. This suggests
that the majority of patients with multilevel spinal
injuries due to a strong, complex trauma often have
severe spinal stenosis.
4.1.9.

The

correlation

between

clinical

and

radiological findings.
Table 3.16 shows the relationship between lesion
images on CT-scans and clinical features. The number
of patients with spinal fractures accounted for the
highest proportion with 38 cases (71.69%), followed by
fracture-dislocations with 9 cases (17%), in which
paralysis caused by fracture-dislocations accounted for
the highest rate with 8/9 cases ( 88.88%).
The number of cases of vertebral fractures showing
signs of neurological lesion were 15/38 patients,
accouting for 39.47%.
Neurological lesions were more common in the
fracture-dislocations of spine, which is also reasonable
due to the fact that the most common anatomical
lesions

in

the

compression

fractures

and

the

dislocations are worse than in the other two groups.
Evaluation

of

neurological

lesions

in

diagnostic

imaging was determined only on magnetic resonance


24
images. In our study, there was no pre-surgery MRI,
which was also the problem of the study.
4.1.10. The correlation between neurological
lesions and the degree of spinal stenosis on CTscan
Table 3.17 shows that 8 out of 10 patients with
paralysis (80%) had spinal stenosis of 50% or more, 2
patients with Frankel B (100%) had spinal stenosis of
50% or more.

5 out of 7 patients with Frankel C

(71.42%) had spinal stenosis of 50% or more. Thus,
the majority of patients with spinal stenosis of 50% or
more had more severe neurological lesions. There was
a correlation between the degree of spinal stenosis
and

the

degree

of

paralysis.

This

relevance

is

statistically significant.
This is a problem that many authors are interested
in finding the link between spinal stenosis and
neurological lesions. Several studies have shown that
fragments of bones narrowing the spinal canal are the
causes of neurological lesions.
This study also shows that CT has great value in the
diagnosis of spinal stenosis, but CT has certain

limitations, making it difficult to assess ligament and
disc injuries.
4.2. Surgical Results
4.2.1. Surgical factors
4.2.1.1. Time of surgery
Table 3.18 shows that there was only patient had


25
surgery within 1 day (24 hours), accounting for 1.88%.
The longest time from admission to surgery was 21
days. The majority of patients had surgery after 3
days, accounting for 84.92%. Average time 7.37 ±
4.37 days.
In our view, in multilevel spinal injuries, there are
many other injuries accompanied. The first hours or
the first day after injury is the stage of spinal shock if
the patient has spinal cord injury. At this stage it is
difficult to fully assess the neurological lesions, so the
prognosis and determination of surgical approaches
are difficult to avoid shortcomings, or errors. Surgery
should be performed after 3 days because at this time
the status of spinal shock and other injuries is stable.
Furthermore, at this stage, neurological lesions can be
clearly identified clinically, accompanied by imagining
findings and laboratory tests.
4.2.1.6. Overlapping surgeries
Table 3.23. showed that 14/53 cases (26.41%) had
overlapping trauma surgeries, of which 6 cases had
fixation surgeries, accounting for 11.32%; and 3 cases

were accompanied by surgeries for thoracostomy +
fixation surgery.
The rate of our overlapping surgeries is relatively
high, which means that the majority of multilevel
spinal injuries have other accompanied injuries and
even require surgery. Multilevel spinal injury is a
severe injury.