ORIGINAL ARTICLE

Less invasive reduction and fusion of fresh A2 and A3 traumaticL1–L4 fractures with a novel vertebral body augmentation implantand short pedicle screw fixation and fusion

Panagiotis Korovessis • Konstantinos Vardakastanis •

Thomas Repantis • Vasilios Vitsas

Received: 1 June 2013 / Accepted: 12 October 2013

� Springer-Verlag France 2013

Abstract The aim of this clinical study was to report on

the efficacy in reduction and safety in PMMA leakage of a

novel vertebral augmentation technique with PEEK and

PMMA, together with pedicle screws in the treatment of

fresh vertebral fractures in young adults. Twenty consec-

utive young adults aged 45 ± 11 years with fresh burst A3/

AO or severely compressed A2/AO fractures underwent via

a less invasive posterior approach one-staged reduction

with a novel augmentation implant and PMMA plus

3-vertebrae pedicle screw fixation and fusion. Radiologic

parameters as segmental kyphosis (SKA), anterior (AV-

BHr) and posterior vertebral body height ratio (PVBHr),

spinal canal encroachment (SCE), cement leakage and

functional parameters as VAS, SF-36 were measured pre-

and post-operatively. Hybrid construct restored AVBHr

(P \ 0.000), PVBHr (P = 0.02), SKA (P = 0.015), SCE

(P = 0.002) without loss of correction at an average fol-

low-up of 17 months. PMMA leakage occurred in 3

patients (3 vertebrae) either anteriorly to the fractured

vertebral body or to the adjacent disc, but in no case to the

spinal canal. Two pedicle screws were malpositioned (one

medially, one laterally to the pedicle at the fracture level)

without neurologic sequelae. Solid posterolateral spinal

fusion occurred 8–10 months post-operatively. Pre-opera-

tive VAS and SF-36 scores improved post-operatively

significantly. This study showed that this novel vertebral

augmentation technique using PEEK implant and PMMA

reduces and stabilizes via less invasive technique A2 and

A3 vertebral fractures without loss of correction and leak-

age to the spinal canal.

Keywords Traumatic lumbar fractures � Minimal

invasively � KIVA � Pedicle screw � PMMA

Introduction

Posterior short-segment spinal instrumentation for trau-

matic causes decreases number of immobilized vertebrae,

surgical time, blood loss and pain, while it preserves

motion segments [1]. These benefits allow quicker reha-

bilitation in young adults and are of major importance for

elderly patients because they decrease complications due to

potential medical co-morbidities.

However, a high prevalence of early instrumentation

failure and subsequent progressive loss of intraoperatively

achieved reduction has been reported [2], particularly in

high vertebral body comminution associated with short-

segment pedicle screw instrumentation. The possible

explanation for this failure is that short posterior instru-

mentation alone is not adequate for sufficient anterior

structural spinal support. The fractured vertebral body

cannot heal due to this anterior column insufficiency, and a

supplemental load sharing through anterior column recon-

struction and support should be added.

Combined anterior plus posterior surgery for lumbar burst

fractures, with short pedicle screw fixation and placement of

mesh cage after anterior corpectomy, was associated with

excellent radiologic results. However, the medium-term

clinical results of combined anteroposterior surgery are not

superior to those achieved with short-segment pedicle screw

fixation alone [3]. Furthermore, traditional anterior surgery

in elderly patients with osteoporosis is fraught with com-

plications due to medical co-morbidities.

A sufficient and safe anterior column reconstruction via

a posterior transpedicular approach has been made possible

P. Korovessis (&) � K. Vardakastanis � T. Repantis � V. Vitsas

Orthopaedic Department, General Hospital ‘‘Agios Andreas’’

Patras, Charalambi Str. 65-67, Patras, Greece

e-mail: korovess@otenet.gr

123

Eur J Orthop Surg Traumatol

DOI 10.1007/s00590-013-1339-2

in the recent few years, thanks to the development of

transpedicular, inflatable balloons, implantable stents or

other reinforcement devices [4] combined with injectable

PMMA and bone substitutes. Thus, risks associated with

anterior surgery have been diminished.

However, some serious complication (lung embolism,

cement leakage, etc.) associated with bone cement injec-

tion has been reported in balloon kyphoplasty (BK) and

vertebroplasty (VP). Despite the fact that use of BK has

become more prevalent, the occurrence of cement leakage

remains a small yet significant problem with rates ranging

from 7.0 to 10 % [4, 5].

The hypothesis of this study is that a novel vertebral

body augmentation implant (KIVA) with PMMA, along

with short pedicle screw fixation and posterolateral fusion,

would provide and maintain short-term reduction and sta-

bilization of fresh traumatic lumbar fractures in young

adults, with reduced PMMA leakage.

Patients and methods

A consecutive series of 20 patients (11 women, 9 men),

suffering from one-level lumbar (L1–L4) fresh traumatic

burst (AO/A3 type) or severely compressed (AO/A2 type)

fracture, were included in the study. The average age of the

patients was 45 ± 11 years at surgery (range 25–61). The

exclusion criteria were as follows: osteoporotic fractures,

polytraumatized patients, pre-existed spinal deformity,

malignancy, history of vertebral fracture in the same or

adjacent vertebrae and previous spinal operation. Eligible

patients were required to sign an informed consent form,

and the internal review board approved this study.

All patients were evaluated both clinically and radio-

logically and were treated operatively within 24 h follow-

ing trauma. Non-serious associated injuries were recorded

in 3/20 patients. Neurologic examination was made on

admission in each patient using a rating system based on

the American Spine Injury Association (ASIA) impairment

scale [6] (Table 1). Back pain intensity was recorded on

VAS (10-point scale). Functional outcome was measured

using the SF-36 questionnaire (role physical and bodily

pain domains). On admission, there were 2 patients with

incomplete paraplegia: one with ASIA grade C, and one

with ASIA Grade D, while the remaining patients were

ASIA E [6].

Pre-operative imaging assessment was performed using

plain films, computed tomography (CT) and/or MRI. There

were 15 type A3 with load-sharing score (LSS) [7] C6

(Table 2) and 5 type A2 fractures. All patients were post-

operatively followed, by standing anteroposterior and lat-

eral roentgenograms of the whole spine. Supine oblique

roentgenograms and sitting lateral flexion–extension

roentgenograms as well as CT-scans, including sagittal

reconstruction images and axial views, were made

8 months post-operatively to assess spinal fusion. Two

independent senior orthopaedic radiologists were asked to

evaluate the evolution of the intertransverse bony fusion

after surgery with the Christiansen’s radiologic method [8].

The observers re-evaluated the plain roentgenograms

within 3 weeks to test interobserver reliability.

The well-established radiographic variables which were

evaluated on plain radiographs on admission and post-

operative were as follows: Gardner segmental kyphotic

deformity (SKA = angle formed from the lines drawn on

the lower endplate of the fractured and the upper endplate

of the above adjacent vertebrae), anterior vertebral body

height ratio (AVBHr) and posterior vertebral body height

ratio (PVBHr). Spinal canal encroachment (SCE) was also

measured in axial CT views.

Patients were encouraged to walk wearing a 3-point

fixation brace on the following day after surgery for

4 weeks.

Statistical analysis was performed with Chi-squared test

for non-parametric data comparisons, paired/unpaired t test

Table 1 ASIA (American Spinal Injury Association) classification of

Spinal Cord Injury (SCI) [5]

Category Motor function Sensory

function

Level

A = complete

SCI

No No S4–S5 includeda

B = incomplete

SCI

No Yes Below the

neurologic level

of the damage,

S4–S5 level

included

C = motor

incomplete

SCI

Yes, more than

half of key

musclesb have a

muscle grade

\3/5

Below the

neurologic level

of the damage

D = motor

incomplete

SCI

Yes, at least half

of the key

musclesb have a

muscle grade

[3/5

Below the

neurologic level

of the damage

E = normal Normal Normal

a The anal sphincter is innervated by the S4–S5 cord and is a critical

part of the spinal cord injury examination. If the person has any

voluntary anal contraction, regardless of any other finding, that person

is by definition a motor incomplete injuryb Arm and hand key muscles: elbow flexors (biceps), wrist extensors,

elbow extensors (triceps), finger flexors and little finger abductor

(outward movement of the pinky finger). Leg and foot key muscles:

hip flexors (psoas), knee extensors (quadriceps), ankle dorsiflexors

(anterior tibialis), long toe extensors (hallucis longus) and ankle

plantar flexors (gastrocnemius)

Eur J Orthop Surg Traumatol

123

for changes of each radiographic parameter and correlation

coefficient (R) for comparison changes of different

parameters.

Surgical technique

All patients were operated under general endotracheal

anaesthesia, by a senior orthopaedic surgeon (first author)

and continuous intraoperative neuromonitoring and image

intensifier.

Pedicle screws were inserted in the two non-fractured

adjacent vertebrae above and below the fracture. The

fractured vertebra was reduced and augmented with the

KIVA implant and PMMA. Intermediate short screws were

added afterwards in the fractured vertebra to complete the

hybrid construct.

A targeting cannulated needle (Abbot, Austin, TX,

USA) is inserted in each pedicle (Fig. 1a). Then, 2-mm K-

wires are passed through the cannulated needle. A straight

short paramedian skin incision connecting the entry points

of the K-wires on each side is made (Fig. 1b) and followed

by a fascial incision. The plane between multifidus and

longissimus muscles was identified (Fig. 1c) [9]. Blunt

dissection of the intermuscular plane between multifidus

and longissimus with the index fingers down to the facet

joints. Using the K-wires as guide, 4 multiaxial fully

cannulated pedicle screws (5.5–7.5 mm, Pathfinder, Abbot,

Austin, TX, USA) with extender sleeves on are placed

down into the pedicles (Fig. 1).

The vertebra augmentation was subsequently performed

unilaterally with the KIVA device (Benvenue Medical,

Santa Clara, CA, USA) [10] and low-viscosity PMMA. The

KIVA system is a sterile, single-use device in which an

external delivery handle is used to deploy the PEEK implant

over a ninitol coil guidewire. The coil is first advanced

through the deployment cannula into the cancellous portion

of the vertebral body. The KIVA implant, which is

comprised of PEEK–OPTIMA (Invibio Inc., West Cons-

hohocken, PA, USA), is incrementally advanced over the

coil to form a nesting, cylindrical column with an in situ outer

diameter of 20 mm. Up to four loops of the implant may be

inserted into the vertebral body for a maximum coil stack

height of 12 mm. This cylindrically shaped PEEK implant

re-elevates the endplate, thereby providing desired sufficient

vertebral endplate reduction. After the coil is retracted,

radiopaque PMMA cement (usually 1–2 cc, depending on

the number of PEEK loops inserted) is injected through the

lumen of the PEEK implant, thereby interlocking the implant

to the vertebral body cancellous bone. Additionally, the

injected PMMA into the PEEK implant creates a uniform

cylindrical column in the anterior and partially middle ver-

tebral column (Figs. 2d, 3d). PEEK cylinder safeguards

PMMA containment and has been shown to significantly

reduce cement leakage [10].

Immediately, after PMMA injection, a 35-mm-long

cannulated 5.5-mm pedicle screw (Pathfinder, Abbot,

Austin, TX, USA) is inserted into each pedicle of the

augmented vertebra. Two rods, after appropriate rod-length

selection and contouring, are then inserted through the

screw extender sleeves (Fig. 1d). Reduction of the rods into

the screw tulips is finally done using forceps reducer. Facet

joints and transverse processes are exposed out with Cobb

elevator, and bone graft substitute is applied. No drain was

applied in any patient since there was no visible bleeding

during wound closure.

Results

The operative time averaged 40 min (range 30–50 min).

The volume of injected PMMA was 1.5 cc (range

1–2 cc).

The total blood loss was 45 mL (range 30–105 mL).

The hospital stay averaged 2 days (range 1–3 days).

Table 2 Load-sharing classification (LSC) of spine fractures [6]

Factors Points

1 2 3

Communition on sagittal plane

section CT

Little B 30 % Medium = 30–60 % Gross C 60 %

Apposition of fragments on axial

CT cut

Minimal Spread = at least 2 mm displacement of

\50 % cross section of body

Wide = at least 2 mm displacement of

[50 % cross section of body

Reducibility of sagittal

deformation on lateral plain films

Kyphotic

correction B

3�

Kyphotic correction 4�–9� Kyphotic correction C 10�

The load-sharing classification grades the extent of vertebral body comminution, the amount of fracture displacement and the amount of

correction of kyphotic deformity

Total load-sharing score (LSS) is the sum of the points of the three factors

Eur J Orthop Surg Traumatol

123

All patients were followed for 17 months (range

12–22 months).

Radiologic results (Table 3)

Pre-operative SKA and AVBHr were correlated signifi-

cantly (r = -0.77, P \ 0.001).

Hybrid fixation reduced pre-operative SKA value at

5� ± 4� (P \ 0.000).

Spinal canal encroachment (SCE) was reduced follow-

ing surgery at an average 78 % (P = 0.02).

No measurable loss of correction in SKA, AVBHr,

PVBHr and SCE was identified at the latest observation

after surgery.

Oblique/bending radiographs and CT-scans taken

8–10 months post-operatively showed solid posterolateral

fusion, bone healing within the fractured vertebral body,

and bridging bone trabeculae around the PMMA–PEEK–

PMMA cylinder (Fig. 2d). The average radiographic score

Fig. 1 Pathfinder technique: a targeting cannulated needles. b Short paramedian skin incision connecting the entry points of the K-wires.

c Splitting of the fascia and identification of the plane between the multifidus and the longissimus muscles. d Schematic representation of the

pathfinder device in situ with the precontoured longitudinal rod being implanted

Fig. 2 a Lateral CT-scan image of a 46-year-old man with a burst A3

fracture of L2 vertebra and neurologic impairment. b Axial CT-scan

showing spinal canal encroachment. c Lateral-standing roentgeno-

gram 8 months post-operatively showing excellent restoration of the

anterior and posterior vertebral height with well-positioned PEEK

implant and PMMA containment. d Post-operative axial CT-scan

image showing excellent PMMA containment, spinal canal clearance

and fusion of the posterior elements

Eur J Orthop Surg Traumatol

123

regarding dorsolateral and intertransverse fusion according

to the Christiansen’s radiologic method [5] was 2.7 (range

2–3).

Additionally, the axial CT-scan images revealed a

remarkable containment of PMMA within the constructed

PEEK-cement cylinder.

Functional results

Pre-operative low back pain VAS score of 7.2 ± 3

improved post-operatively to 2.6 ± 2 (P \ 0.002). No

further significant changes in VAS score were reported

thereafter.

Role physical (SF-36) improved from 38 ± 17 (aver-

age ± SD) before surgery to 77 ± 18 after surgery

(P \ 0.001), and the domain bodily pain (SF-36) from

22 ± 10 (average ± SD) before surgery to 81 ± 13 after

surgery (P \ 0.001), at the 6 months follow-up. No further

changes in both SF-36 domains scores were reported in

both groups.

Both patients with incomplete neurologic impairment on

admission improved after surgery from ASIA C & D to

ASIA D & E, respectively. They both had a severely

comminuted fracture (LSS score 8 and 7, respectively) with

great canal encroachment ([30 %) in which post-operative

had a spinal canal clearance of 80 % (Fig. 2).

Complications

No intraoperative hypotension, cardiogenic side-effects, or

pulmonary embolism were clinically recorded.

In 3 patients, bone cement leakage was documented on

CT-scan after surgery, anteriorly to the augmented verte-

bral body and into the adjacent disc (Fig. 3b), without any

sequelae. All these fractures were A3/AO type with severe

comminution (LSS score [7] 7,8) (Fig. 3).

In two obese patients, 2 pedicle screws were misplaced;

one medially and one laterally without, however, clinical

sequelae (Fig. 3d).

No patient deteriorated neurologically after surgery. No

implant failure was observed in any case.

In the short-term follow-up evaluation, no adjacent new

vertebral fractures were observed.

Discussion

The hypothesis of this study was justified by the achieved

clinical and radiologic results associated with the hybrid

fixation of fresh lumbar vertebral fractures: the novel

KIVA implant with PMMA reduced and simultaneously

reinforced the fractured vertebral body immediately after

PMMA hardening thus restoring both anterior and posterior

vertebral body height and simultaneously reduced post-

Fig. 3 a Supine lateral roentgenogram of 67-year-old male patient

who sustained a L3 type A3 fracture after fall from a height. There is

loss of segmental lordosis; b Lateral-standing roentgenogram

6 months following KIVA plus short fixation. The segmental lordosis

is well restored following KIVA implantation; c Pre-operative axial

CT showing a comminution of the vertebral body associated with

spinal canal encroachment; d Axial CT-scan 6 months following

hybrid surgery showing spinal canal clearance and sufficient PMMA

containment. Note the spinal canal clearance

Table 3 Radiologic data pre-to post-operatively in 20 patients with

lumbar fractures

Pre-

operative

Post-

operative

P value Change %

AVBHr 0.66 ± 0.17 0.85 ± 0.13 0.000027 37.3 ± 35.5

PVBHr 0.88 ± 0.1 0.94 ± 0.05 0.02 10.1 ± 13.8

SKA 11.2 ± 7.2 6 ± 4.7 0.015 51.4 ± 24

(5.38� ± 4.2�)

SCE 30.7 ± 28.5 11.8 ± 17 0.02 78 ± 23 spinal

canal clearance

All values are shown as average ± standard deviation

AVBHr anterior vertebra body height ratio, PVBHr posterior vertebra

body height ratio, SKA segmental kyphosis angle, SCE spinal canal

encroachment

Eur J Orthop Surg Traumatol

123

traumatic kyphosis. The segmental short pedicle screw

construct finally safeguarded the long-term stability and

maintained the intraoperatively obtained correction.

Similar results with us were reported in comparable

historical series with open and percutaneous approach [10–

15]. The immediate spinal stability, particularly in A3-type

fractures, provided by the hybrid KIVA construct, allowed

for early and safe mobilization of our patients, thus

reducing any immobilization-related complications.

Balloon kyphoplasty (BK) and VP procedures may be

associated with PMMA extravasation [16, 17] that could

result in pulmonary embolism, deep-seated infection, post-

procedure acute respiratory distress syndrome, palsy and

death [18–21]. None of these complications occurred in our

patients.

Our explanation regarding the efficacy of the PEEK

implant in vertebral body reduction and maintenance of

achieved reduction and the role of the PMMA in support-

ing the vertebral body, without significant cement leakage,

speculated the following: the controlled PMMA injection

into the PEEK cylinder reduced cement leakage [22]. In

Kyphoplasty, after balloon deflation, some void collapse

may occur, and PMMA leakage may occur since there are

no bony boundaries to create a safe nest for PMMA within

this void [23]. In contrary, the PEEK–PMMMA construct

holds the achieved correction immediately after its

implantation. This PEEK column limits the rate of cement

leakage. Extracanal PMMA leakage was reported as high

as 25 %, following short pedicle screw fixation combined

with BK for thoracolumbar burst fractures [8]. In our ser-

ies, extracanal cement leakage was evident just in 3 of 20

cases (15 %) without any clinical sequelae.

There is some controversy as if a posterolateral fusion is

necessary additionally to short pedicle screw fixation and

vertebral augmentation for thoracolumbar A3 fractures [10,

11, 13, 22–24]. A recent study [24] compared on a pro-

spective basis two series of patients operated with posterior

short-segment fixation with/without addition of iliac bone

graft and postulated that fusion is not necessary. In con-

trary, another study [11] compared short-segment pedicle

screw fixation with/without fusion for A3 burst fractures

with LSS \6 and reported equally good results. In all 20

patients included in our study, the addition of posterolateral

fusion to short instrumentation was justified because there

were 15 type A3 with LSS C6 and 5 type A2 fractures with

severe compression. The lack of instrumentation failure

until the final short-term observation justified our decision

for additional posterolateral fusion in these fractures.

The reported kyphosis correction with transpedicular

augmentation (bone, bone graft substitute, PMMA, tita-

nium spacer, hydroxyapatite sticks) and short pedicle fix-

ation techniques ranged from 64 to 105 % [10–13, 25]. In

our study, the post-operatively achieved Gardner kyphosis

correction of 51 % was less than that of previously

reported because in those studies, either solid grafts (tita-

nium block, HA stick) or BK was used. Furthermore, the

method used for kyphosis measurement was not clearly

described in those articles.

Following transpedicular vertebral body grafting, spinal

canal clearance averaged 65 % [10–13, 25] and this was

mostly due to bone remodelling with the time lapsed from

index surgery. In our series, the spinal canal clearance

following vertebral fracture reduction and vertebral aug-

mentation was 78 %, and this depicts the stable fixation

achieved by the hybrid construct.

Our study, in accordance with a previous similar [9] one,

clearly showed the unique advantage of KIVA to safely

reduce not only severely compression fractures but also

burst fractures with moderate SCE and LSS C6.

There is some controversy regarding bone displacement

following vertebral body augmentation. Some anterior bone

displacement (at both thoracic and lumbar level) and pos-

terior bone displacement (at thoracic level) can occur after

inflation of balloons in burst fractures treated with balloon

VP and pedicle screw instrumentation [26]. The absolute

amount of posterior bone displacement is probably not of

clinical relevance [26]. In contrary, a cadaveric study [26]

showed that BK after pedicle screw instrumentation may

safely be used, in terms of bone displacement and cement

leakage, even in fracture types where damage to longitudinal

ligaments is to be expected. In our study, KIVA PEEK [10]

implant was introduced via a coil within the crushed can-

cellous bone of the vertebral body elevating the endplate

without simultaneously pushing bone fragments posteriorly

to the spinal canal [10]. Anterior or most importantly pos-

terior to the spinal canal vertebral body displacement was not

measured even in A3 fractures in our series.

Furthermore, the use of two intermediate screws at the

level of KIVA implantation was proved in this series to be

safe without retropulsion of the fragments into the canal.

The latter is in accordance with the previous relative lit-

erature [9, 27].

No instrumented released distraction was necessary to

reduce post-traumatic kyphosis because as it was previ-

ously shown the local post-traumatic kyphosis in fresh

lumbar fractures is corrected combined by positioning of

the patient on the frame and the action of KIVA [10].

Although serious drawbacks associated with the long-

term results after use of PMMA in cemented total joint

arthroplasties have been reported, the complications with

the use of PMMA in kyphoplasty and VP focus solely on

cement leakage. Self-hardening calcium phosphate bone

cements have been developed as an alternative to overcome

the short-term and possible long-term side-effects of

PMMA [9]. Marco et al. used PMMA in patients with age

ranged from 15 to 73 years and reported no side-effects up

Eur J Orthop Surg Traumatol

123

to 55 months after index surgery [28]. However, the long-

term side-effects of the use of PMMA in spines of young

adults, to our knowledge, have not yet been studied.

Although minimal invasive techniques reserve back

muscles, they may be associated with disastrous neurologic

complications. A CT study disclosed on axial post-opera-

tive CT views a 23 % pedicle penetration rate with 3.3 %

severe frank screw penetration. Others recently showed

that percutaneous placement of pedicle screws was asso-

ciated with 14 % misplacement in normal weight versus

22.6 % in overweight patients [29]. In our series, 3 of 120

screws were medially misplaced on post-operative CTs

without, however, any clinical complaints. In accordance

with the published literature [29], all malpositioned screws

in our series occurred in overweighed patients.

The rate of new adjacent and remote vertebral fractures

following BK and pedicle screw fixation is reported up to

18.8 %, depending on the duration and nature of the follow-

up [25]. No short-term adjacent fractures were observed in

our series. We speculate that the reason of lack of new

fractures was the reduction in mechanical stresses through

the sufficient restoration of vertebral body height and local

kyphosis and the supplementary use of instrumentation.

Following restoration of collapsed vertebral body, mainte-

nance of as many as possible free motion segments could be

the ideal surgical goal. However, loss of correction has been

reported after hardware removal because the correction loss

often occurs primarily through disc space collapse. [30–32]

Furthermore, degeneration of the facet joints within the

immobilization area may occur [30–32].

The limitations of this study were as follows: its retro-

spective nature and the lack of studies regarding the long-

term effects of PMMA in spine of young patients. Since

extracanal PMMA leakage in our series occurred solely in

A3 fractures, this kind of fractures may be a potential

theoretical limitation for KIVA augmentation [4, 5].

Although the short-term clinical and radiologic results

of this comparative study are promising, clinical and

radiologic studies with longer observation periods and

comparisons with conservatively treated patient series are

required to support the longevity of this novel method of

hybrid fixation.

Conflict of interest The authors of the present study received no

external funding resources, either directly or indirectly, in order to

complete the present work. No other relationships/conditions/cir-

cumstances that present a potential conflict of interest exist.

References

1. Olerud S, Karlstrom G, Sjostrom L (1988) Transpedicular fixa-

tion of thoracolumbar vertebral fractures. Clin Orthop Relat Res

227:44–51

2. McLain RF, Sparling E, Benson DR (1993) Early failure of short-

segment pedicle instrumentation for thoracolumbar fractures. A

preliminary report. J Bone Joint Surg Am 75:162–167

3. Korovessis P, Baikousis A, Zacharatos S et al (2006) Combined

anterior plus posterior stabilization versus posterior short-seg-

ment instrumentation and fusion for mid-lumbar (L2–L4) burst

fractures. Spine 31:859–868

4. Waiter J, Haciyakupoglu E, Waschke A et al (2012) Cement

leakage as a possible complication of balloon kyphoplasty. Is

there a difference between osteoporotic compression fractures

(AO type A1) and incomplete burst fractures (AO type A3.1)?

Acta Neurochir (Wien) 154:313–319

5. Bohner M, Gasser B, Baroud G, Heini P (2003) Theoretical and

experimental model to describe the injection of a polymethac-

rylate cement into porous structure. Biomaterials 24:230–272

6. Marino RJ, Barros T, Burns SP, Donovan WH, Graves DE,

Haak M, Hudson LM, ASIA Neurological Standards Commit-

tee 2002 et al (2003) International standards for neurological

classification of spinal cord injury. J Spinal Cord Med

26(Suppl 1):S50–S56

7. McCormack T, Karaikovic E, Gaines RW (1994) The load-

sharing classification of spine fractures. Spine 19:1741–1744

8. Christensen BF, Laursen M, Gelineck J et al (2001) Interobserver

and intraobserver agreement of radiograph interpretation with and

without pedicle screw implants. Spine 26:538–544

9. Korovessis P, Hadjipavlou A, Repantis T (2008) Minimal inva-

sive short posterior instrumentation plus balloon kyphoplasty

with calcium phosphate for burst and severe compression lumbar

fractures. Spine 33:658–667

10. Korovessis P, Repantis T, Miller LE et al (2011) Initial clinical

experience with a novel vertebral augmentation system for

treatment of symptomatic vertebral compression fractures: a case

series of 26 consecutive patients. BMC Musculoskelet Disord

22(12):206–209

11. Verlaan JJ, Dhert WJ, Verbout AJ et al (2005) Balloon verteb-

roplasty in combination with pedicle screw instrumentation: a

novel technique to treat thoracic and lumbar burst fractures. Spine

30:E73–E79

12. Kramer DL, Rodgers WB, Mansfield FL (1995) Transpedicular

instrumentation and short-segment fusion of thoracolumbar

fractures: a prospective study using a single instrumentation

system. J Orthop Trauma 9:499–506

13. Dai LY, Jiang LS, Jiang SD (2009) Posterior short-segment fix-

ation with or without fusion for thoracolumbar burst fractures. A

five to seven-year. J Bone Joint Surg Am 91:1033–1041

14. Toyone T, Tanaka T, Kato D et al (2006) The treatment of acute

thoracolumbar burst fractures with transpedicular intracorporeal

hydroxyapatite grafting following indirect reduction and pedicle

screw fixation: a prospective study. Spine 31:208–214

15. Alanay A, Acaroglu E, Yazici M et al (2001) Short-segment

pedicle instrumentation of thoracolumbar burst fractures: does

transpedicular Intracorporeal grafting prevent early failure? Spine

26:213–217

16. Wardlaw D, Cummings SR, Van Meirhaeghe J, Bastian L, Till-

man JB, Ranstam J, Eastell R, Shabe P, Talmadge K, Boonen S

(2009) Efficacy and safety of balloon kyphoplasty compared with

non-surgical care for vertebral compression fracture (FREE): a

randomised controlled trial. Lancet 373(9668):1016–1024

17. Taylor RS, Taylor RJ, Fritzell P (2006) Balloon kyphoplasty and

vertebroplasty for vertebral compression fractures: a comparative

systematic review of efficacy and safety. Spine (Phila Pa 1976)

31(23):2747–2755

18. Choe DH, Marom EM, Ahrar K et al (2004) Pulmonary embolism

of polymethylmethacrylate during percutaneous vertebroplasty

and kyphoplasty. Am J Roentenol 183:1097–1102

Eur J Orthop Surg Traumatol

123

19. Yu SW, Chen WJ, Lin WC et al (2004) Serious pyogenic spon-

dylitis following vertebroplasty: a case report. Spine 29:E209–

E211

20. Yoo KY, Jeong SW, Yoon W, Lee J (2004) Acute respiratory

distress syndrome associated with pulmonary cement embolism

following percutaneous vertebroplasty with polymethylmethac-

rylate. Spine 29:E294–E297

21. Chen HL, Wong CS, Ho ST et al (2002) A lethal pulmonary

embolism during percutaneous vertebroplasty. Anesth Analg

95:1060–1062

22. Korovessis P, Vardakastanis K, Repantis T, Vitsas V (2013)

Balloon kyphoplasty versus KIVA vertebral augmentation–com-

parison of 2 techniques for osteoporotic vertebral body fractures:

a prospective randomized study. Spine (Phila Pa 1976)

38(4):292–299

23. Werner CM, Osterhoff G, Schlickeiser J, Jenni R, Wanner GA,

Ossendorf C, Simmen HP (2013) Vertebral body stenting versus

kyphoplasty for the treatment of osteoporotic vertebral com-

pression fractures: a randomized trial. J Bone Joint Surg Am.

95(7):577–584

24. Benson DR, Burkus JK, Montesano PX et al (1992) Unstable

thoracolumbar and lumbar burst fractures treated with the AO

fixateur interne. J Spinal Disord 5:335–343

25. Li KC, Li Anna F-Y, Hsieh CH et al (2007) Transpedicle body

augmenter in painful osteoporotic compression fractures. Eur

Spine J 16:589–598

26. Marco RA, Kushwaha VP (2009) Thoracolumbar burst fractures

treated with posterior decompression and pedicle screw instru-

mentation supplemented with balloon-assisted vertebroplasty and

calcium phosphate reconstruction. J Bone Joint Surg Am

91:20–28

27. Verlaan JJ, van de Kraats EB, Oner FC et al (2005) Bone dis-

placement and the role of longitudinal ligaments during balloon

vertebroplasty in traumatic thoracolumbar fractures. Spine

30:1832–1839

28. Mahar A, Kim C, Wedemeyer M et al (2007) Short-segment

fixation of lumbar burst fractures using pedicle fixation at the

level of the fracture. Spine 32:1503–1507

29. Park Y, Won HJ, Lee TY, Sung YN (2011) Percutaneous

placement of pedicle screws in overweight and obese patients.

Spine J 11:919–924

30. Wang XY, Dai LY, Xu HZ, Chi YL (2008) Kyphosis recurrence

after posterior short-segment fixation in thoracolumbar burst

fractures. Neurosurg Spine 8:246–254

31. Dekutoski MB, Conlan ES, Salciccioli GG (1993) Spinal

mobility and deformity after Harrington rod stabilization and

limited arthrodesis of thoracolumbar fractures. J Bone Joint Surg

Am 75:168–176

32. Chen WJ, Niu CC, Chen LH, Chen JY, Shih CH, Chu LY (1995)

Back pain after thoracolumbar fracture treated with long instru-

mentation and short fusion. J Spinal Disord 8:474–478

Eur J Orthop Surg Traumatol

123