Abstract The authors report on 32consecutive patients with instabilityat the craniocervical, cervical andcervicothoracic regions sufferingfrom various pathologies, who weretreated with posterior instrumenta-tion and fusion using the posteriorhooks-rods-plate cervical compactCotrel-Dubousset (CCD) instrumen-tation alone or, in three patients, incombination with anterior operation.The patients were observed postoper-atively for an average of 31 months(range 25–44 months) and evaluatedboth clinically and radiographicallyusing the following parameters: spineanatomy and reconstruction, sagittalprofile, neurologic status, functionallevel, complications and status ofarthrodesis. All patients but one(who died) achieved a solid arthrode-sis based on plain and flexion/exten-sion roentgenograms. Cervical lordo-sis (skull–C7) and cervicothoracickyphosis (C7–T2) was improved byinstrumentation towards a physiolog-ical lateral curve by an average of33% (P<0.05) and 28% (P<0.05) re-spectively. Anterior vertebral olisthe-sis was reduced in the craniocervicaland cervicothoracic region, by 73%and 90% respectively. At final fol-low-up there was an improvement ofthe neurologic Frankel status by anaverage of 1.2 grades and of myelo-

pathy in 75% of the operated pa-tients. Good to excellent functionalresults were seen in 77% of the oper-ated patients, while acute andchronic pain was reduced by an aver-age of 2.4 grades, on a scale of 0–3,in operated patients. No neurovascu-lar or pulmonary complications arosefrom surgery. There was no signifi-cant change in lateral spine profileand olisthesis at the latest follow-upevaluation. There were no instru-ment-related failures. One patient re-quested hardware removal in thehope of reducing postoperative painin the cervicothoracic region. Thepoor and fair results were related tothe lack of improvement of neuro-logic impairment and myelopathy.The results of this study demonstratethat cervical CCD instrumentationapplied in the region of the skull tothe upper thoracic region for variousdisorders is a simple and safe instru-mentation that restores lateral spinealignment, improves the potential fora solid fusion and offers sufficientfunctional results in the vast majorityof the operated patients. However,the use of hooks in spinal stenosis iscontraindicated.

Keywords Cervical spine · Craniocervical · Cervicothoracic ·Cotrel-Dubousset · CCD

ORIGINAL ARTICLEEur Spine J (2001) 10 :385–394DOI 10.1007/s005860100245

Panagiotis Korovessis Pavlos Katonis Agisilaos Aligizakis Josef Christoforakis Andreas Baikousis Zisis Papazisis Giorgos Petsinis

Posterior compact Cotrel-Dubousset instrumentation for occipitocervical, cervical and cervicothoracic fusion

Received: 14 April 2000 Revised: 6 December 2000 Accepted: 18 December 2000 Published online: 19 June 2001© Springer-Verlag 2001

A comment to this paper is available athttp://dx.doi.org/10.1007/s005860100246.

P. Korovessis (✉ ) · P. Katonis ·A. Aligizakis · J. Christoforakis ·A. Baikousis · Z. Papazisis · G. PetsinisOrthopedic Department, General Hospital “Agios Andreas”, Patras,and Orthopedic Department, University of Crete, Greece e-mail: korovess@otenet.gr, Fax: +30-61-622227

P. KorovessisSpine Unit, General Hospital “Agios Andreas”, 26224 Patras, Greece

Introduction

Instability of the cervical spine, whether traumatic, degen-erative, rheumatic or neoplastic, may necessitate internalfixation. The location of the lesion and the pattern of thespinal instability should determine selection of anterior orposterior fixation procedure. From a biomechanical pointof view, posterior fixation devices have an advantage overanterior devices for fixation of posterior instability, suchas postlaminectomy instability, and for fixation of three-column instability, such as fracture-dislocation or circum-ferentially compromised vertebral metastases [2, 11, 24,29, 36, 39, 40].

Since the first occipitocervical fusion in 1927 by Foer-ster [17], a number of posterior techniques for fusing thecervical spine using wire, metal, or methacrylate, with orwithout the addition of bone have been described [1, 2, 6,8, 9, 12, 13, 16, 18, 19, 20, 21, 24, 25, 26, 27, 28, 30, 31,38].

Abumi and Kaneda [1] first reported pedicle screw fix-ation in the middle and lower cervical spine in traumaticlesions. Based on comparative biomechanical studies ofcervical fixation procedures, however, there is not muchdifference in stability between lateral mass screw-platefixation and conventional non-screw fixation methods [1,10, 22, 29, 37, 38].

Posterior upper cervical fusion with autogenous bonegraft and wires has been advocated in the management ofatlantoaxial instability in rheumatoid patients, with an im-provement rate of 30–40% [9, 35], failure of fusion in16–50% and a mortality rate ranging from 27 to 42% [12,16, 35].

There have been only a few reports in the literature onthe spinal disorders of and surgical approach to the cervi-cothoracic junction [1, 2, 7].

The purpose of this study is to report the use of theposterior cervical hook-rod compact Cotrel-Dubousset(CCD) instrumentation in the skull to the upper thoracicspine region in 32 patients suffering from different dis-eases in order to investigate the versatility and the suc-cessful use of the system. Specifically, the operated pa-tients were examined for changes in neurologic function,spine stability, maintenance of spine alignment, and com-plications after using this fixation technique.

Materials and methods

Thirty-two patients with acute and chronic spinal lesions (Table 1)located in the region from C1 to T2 were managed by the first twoauthors using posterior cervical CCD instrumentation and fusion intwo affiliated spine units between 1994 to 1997, using the samesurgical technique and evaluation protocol. The indications forsurgery were: presence of acute or chronic instability, presence ofneurologic deficit, and iatrogenic instability. The indications forsurgery are shown in Table 1. There were 24 men and 8 women,and their average age was 54±20 years (range 18–84 years). Four-teen patients suffered from chronic cervical spine diseases and 18

from fresh traumatic lesion of the cervical and upper thoracic spine(Table 1). In 7 of the 32 patients, fusion involved the occipito-cervical junction; in 11 of the 32 patients, the cervicothoracicjunction was included in the instrumentation; and in 14 of the 32,the instrumentation was limited to the cervical (C1–C7) region(Table 1). Impaired single or multiple vertebrae were variouslydistributed across the occipitocervical region to the cervicothoracicjunction. Single- or multisegmental instability was mostly causedby the cervical lesions themselves, and instability was obvious, so that functional roentgenograms were either not necessary (ver-tebral bone deficiency due to metastasis or primary tumor: cases 1,2, 3, 14) or theoretically dangerous for iatrogenic neurologic lesion(facet subluxation, dislocation, vertebral body fracture: cases 4, 6, 15–32). Alternatively, the instability was caused by exten-sive laminectomy for spinal cord decompression (cases 7–12)(Table 1). Only in one case (case 13) were preoperative flexion/extension roentgenograms needed to document postlaminectomyinstability.

In all patients, cervical CCD was used for cervicothoracic fu-sion combined with conventional CD. The patients were re-evalu-ated at 2, 6 and 12 months after surgery, and thereafter once ayear.

Implants

The cervical Compact Cotrel-Dubousset (Cervical CCD, Sofamor-Danek, USA) instrumentation (Fig.1) was inspired by the Com-pact CD and derived from cervical and occipitocervical applica-tions of the reduced-size CD implanted since 1983. This instru-mentation is of stainless steel and was designed for treatment fromthe occipitum to the thoracic junction, using the posterior ap-proach, in trauma, degenerative disease, instability, tumor, andiatrogenic instability. Cervical CCD instrumentation is not limitedto a single screw, plate, or rod, but may be adapted to a specificanatomy and diversity of pathologies using implants of differentanchorage, shape and dimensions. The implants have been de-signed to obtain increased mechanical resistance values. The con-struct comprises two rods and one transverse link device for im-mediate and long-term stability. A domino connector [4] is alsoavailable, in case of extension of the instrumentation. The trans-verse link device is user friendly, strong (bending strength 183 N)and of small size [4]. The axial slippage of the hooks on the rod isestimated to be 2225 N and the force required for hook rotation onthe rod is approximately 2.8 Nm [4]. The locking mechanism en-ables easy removal of the construct without bone damage duringexplantation. Each step and each maneuver during the operation isfacilitated by multifunctional instruments adapted for the cervicalspine. The screwdriver is universal. The hooks have accurateshape, low profile in the spinal canal, are compact and of minimalinventory (open-closed and large-small groove), their handling andinsertion are easy. The rods have a reduced volume (5 mm diame-ter) compared to the original Cotrel-Dubousset instrumentation,are resistant but easy to contour in an appropriate shape for the cer-vical, occipitocervical, cervicothoracic spine and are connectableto the thoracic level. The occipital fixation may be obtained usingscrews or hooks. Compression and/or distraction is possible usingappropriate instruments. Claw configuration between adjacent ver-tebrae is usually performed. The occipitocervical junction is in-strumented with a combination of hooks and 3.5-mm screws pene-trating both cortices of the skull [34].

Surgical procedure

All patients with dislocations underwent halo traction until reduc-tion was achieved. After intubation, patients were placed prone ona frame and a horseshoe-type head holder or halo traction device.The head was taped to the head holder with the cervical spine

386

387

Tab

le 1

C

lini

cal

data

on

30 p

atie

nts

who

und

erw

ent

post

erio

r ce

rvic

al f

usio

n us

ing

cerv

ical

com

pact

Cot

rel-

Dub

ouss

et (

CC

D)

inst

rum

enta

tion

(la

min

lam

inec

tom

y,ra

dicu

lop

radi

culo

path

y)

Pai

naN

euro

logi

c(F

rank

el g

rade

)F

ollo

w-u

p(m

onth

s)C

ompl

i-ca

tion

sR

esul

tC

ase

Age

Sex

Pat

holo

gic

cond

itio

nfo

r op

erat

ion

Pos

teri

orop

erat

ive

proc

edur

e

No.

of

segm

ents

fuse

d

Add

itio

nal

ante

rior

oper

atio

nP

reop

Pos

top

Pre

opP

osto

p

172

FC

2 he

man

gio-

thel

iom

aL

amin

C0–

C4

5N

o3

0D

E41

No

Goo

d

242

MA

neur

ysm

al b

one

cyst

T1–

T2

Lam

inC

5–T

710

C7–

T3

rese

ctio

n&

fus

ion

20

DE

35N

oG

ood

328

MO

steo

blas

tom

aT

1 la

min

aR

esec

tion

C6–

T3

5N

o3

0E

E29

No

Goo

d

444

FC

onge

nita

l dis

loca

-ti

on C

1/C

2 &

my-

elop

athy

C0–

C5

fusi

on 6

Tra

nsor

alde

ns r

esec

tion

30

CE

25N

oG

ood

554

FR

heum

atoi

dsp

ondy

liti

s C

1–C

4C

0–C

5fu

sion

6N

o3

0E

E26

No

Goo

d

661

MO

dont

oid

pseu

dar-

thro

sis

& m

yelo

p-at

hy

Lam

inC

0–C

7 8

No

30

CE

31N

oE

xcel

lent

768

MM

yelo

path

yL

amin

& C

4–C

7fu

sion

4N

o2

1C

C29

No

Poo

r

876

MM

yelo

path

yL

amin

& C

3–C

7fu

sion

5N

o2

1C

D37

No

Fai

r

966

MM

yelo

path

yL

amin

& C

3–C

6fu

sion

4N

o2

0D

D25

No

Goo

d

1072

MM

yelo

path

yL

amin

& C

3–C

7fu

sion

6N

o2

0C

D37

No

Goo

d

1175

MM

yelo

path

yL

amin

& C

4–C

6fu

sion

3N

o2

0D

E37

No

Goo

d

1271

MM

yelo

path

yL

amin

& C

3–C

6fu

sion

4N

o2

1C

D25

No

Fai

r

1348

FP

ostl

amin

ecto

my

inst

abil

ity

C4–

C7

fusi

on 4

No

31

EE

42N

oE

xcel

lent

1471

MM

etas

tati

c pr

osta

teT

1–T

3L

amin

& C

5–T

7fu

sion

14N

o3

0C

E44

No

Exc

elle

nt

1518

MC

1–C

2 di

sloc

atio

nC

1–C

2fu

sion

2N

o3

0D

E31

No

Exc

elle

nt

1625

MC

5–C

6 bi

late

ral

face

t dis

loca

tion

C5–

C6

fusi

on 2

No

30

DE

28N

oE

xcel

lent

388

Tab

le 1

(c

onti

nued

)

Pai

naN

euro

logi

c(F

rank

el g

rade

)C

ase

Age

Sex

Pat

holo

gic

cond

itio

nfo

r op

erat

ion

Pos

teri

orop

erat

ive

proc

edur

e

No.

of

segm

ents

fuse

d

Add

itio

nal

ante

rior

oper

atio

nP

reop

Pos

top

Pre

opP

osto

p

Fol

low

-up

(mon

ths)

Com

pli-

cati

ons

Res

ult

1719

FC

4–C

5 su

blux

atio

nC

4–C

5fu

sion

2N

o3

0E

E25

No

Exc

elle

nt

1831

MC

5–C

6 un

ilat

eral

disl

ocat

ion

C5–

C6

fusi

on 2

No

32

CD

33N

oF

air

1975

FJe

ffer

son

& t

ype

IIde

ns f

ract

ure

C0–

C4

fusi

on 4

No

30

EE

30N

oE

xcel

lent

2026

MC

6–C

7 su

blux

atio

n&

fac

et f

ract

ure

C6

C6–

T2

fusi

on 4

No

30

EE

25N

oE

xcel

lent

2174

MD

islo

cati

on C

5–C

6C

5–C

6fu

sion

2N

o3

deat

hA

A2

days

Dea

th,

vasc

.ce

rebr

al

deat

h

2236

FD

islo

cati

on C

5–C

6,la

min

a fr

actu

reC

4–C

7fu

sion

4N

o3

0ra

dicu

lop

reso

lved

27N

oE

xcel

lent

2354

MF

ract

ure/

disl

ocat

ion

C6–

C7

C5–

T2

fusi

on 5

No

32

EE

30P

ain

due

to h

ard-

war

e

Goo

d

2449

MD

islo

cati

on C

6–C

7,la

min

a fr

actu

reC

4–T

2fu

sion

6N

o3

0ra

dicu

lop

reso

lved

33N

oE

xcel

lent

2584

MD

ens

II, l

amin

afr

actu

re C

5–C

7C

0–T

2fu

sion

10N

o2

1C

C26

No

Fai

r

2655

MF

ract

ure

body

of

C6

C5–

C7

fusi

on 3

Tit

aniu

mm

esh

& p

late

30

EE

40N

oE

xcel

lent

2780

MD

islo

cati

on C

6–C

7C

6–C

7fu

sion

2N

o2

0ra

dicu

lop

reso

lved

36N

oG

ood

2825

FJe

ffer

son

& d

ens

III

C0–

C5

fusi

on 6

No

30

EE

28N

oE

xcel

lent

2941

MD

islo

cati

on C

7–T

1C

5–T

6fu

sion

9N

o3

1C

C25

No

Poo

r

3050

MF

ract

ure

T2

& d

is-

loca

tion

T2–

T3

C6–

T6

fusi

on 8

No

30

EE

29N

oE

xcel

lent

3143

MD

islo

cati

on T

1–T

2C

5–T

6fu

sion

9N

o3

0A

A32

No

Poo

r

3278

MD

islo

cati

on T

1–T

2C

5–T

6fu

sion

9N

o3

0D

E26

No

Goo

d

a Pai

n sc

ale

of 0

–3, g

rade

d ac

cord

ing

to t

he c

rite

ria

deve

lope

d by

the

Hos

pita

l fo

r S

peci

al S

urge

ry

maintained in a neutral position, and the shoulders were pulledcaudal by a heavy bandage for intraoperative lateral radiographicimaging of the lower cervical spine. With a median skin incision,the paravertebral muscles were dissected laterally to expose thelaminae and the articular masses in the cervical and upper thoracicvertebrae and the transverse processes at the upper thoracic region.The laminar hooks were inserted at sites of the laminae that wereprepared by blunt dissection or after partial excision of the liga-mentum flavum or partial laminotomy. They formed a claw be-tween adjacent vertebrae, and were assembled to two longitudinalrods. The longitudinal rods were augmented with the transfixationdevice (Fig.1, Fig.2C), representing a quadrilateral fixation thatcan counteract forces that act on different planes: axial loading,flexion-extension, and torsion. In case of instrumentation of theskull, it was subperiosteally prepared up to the occipital notch. Thecraniovertebral junction was instrumented using the premodeledCCD rods and four 3.5-mm screws that penetrate the cranial walland are bilaterally inserted about 1.5 cm below the occipital notchand at about 2 cm laterally with respect to the median line. Thetype of material used allows for “in situ” contouring of the rods bymeans of special benders. This CCD implant can be considered asa “dynamic” fixation device, because the hooks, in contrast toscrews, do not offer absolutely rigid fixation, since some move-ment at the lamina-hook interface can theoretically occur. It is alsopossible to apply limited force during surgery (distraction andcompression forces) to obtain a better realignment of the skeletalcomponents. In general, the reduction was performed before sur-gery by means of halo traction, and thereafter it was controlled in-traoperatively using lateral plain roentgenograms. However, aspointed out earlier, it is also possible to apply limited force duringsurgery to compact the metal construct. After meticulous decorti-cation of the posterior spinal elements and the skull (air drill witha diamond burr for preparation to craniovertebral fusion) posteriorfusion was carried out using iliac cortical cancellous autogenousbone graft. In the three patients (patients 2, 4, 26) where a com-bined approach was necessary, the posterior operation was per-formed first, for better alignment of the spine, and it was followed

by the anterior operation. Anterior operation was indicated in caseof extensive anterior pathology that caused immediate or potentialsevere instability which was not possible to be managed posteri-orly only. Postoperative immobilization varied according to thenumber of the spinal segments fixed, the patient’s general condi-tion, and the extent of osteoporosis. In principle, a soft collar or ashort Philadelphia collar was applied for 3 months. All patients ex-cept the neurologically impaired were permitted to ambulate or situp in the bed the day after surgery, unless contraindicated by theirgeneral condition.

Clinical and functional evaluation

The pain level of the patients, their neurologic status and functionand the complications after surgery were recorded in detail preop-eratively through to postoperative follow-up. Pain status (on ascale of 0–3) was determined using the criteria developed at theHospital for Special Surgery [33]. The pain grades were: none (0),mild (1), moderate (2), and severe (3). Neurologic status was de-termined using the Frankel classification (A to E); the preoperativeneurologic status was graded as A in two, C in ten, D in seven andE in ten patients. Three patients suffered from radiculopathy. Pre-operative pain level averaged 2.7±0.5 (range 2–3) and was gradedas 3 in 23 patients and as grade 2 in 9 patients (Table 1). Duringthe follow-up period the improvement of the neurologic function,the existence of pain and disability, the success of fusion, and therestoration of the normal alignment of the cervical spine (tested ra-diologically with static and dynamic radiographs) were taken intoaccount to determine the final outcome of each patient.

Follow-up information was obtained clinically and radiologi-cally, and a rating of excellent, good, fair, or poor was assigned. If a complete recovery and return to previous activities occurred,the result was considered excellent; good if occasional pain, returnto gainful activities, and intermittent use of mild analgesics oc-curred; fair, if partial recovery, frequent use of analgesics, andmodified activities occurred; and poor, if no relief of originalsymptoms, constant pain and need for full-time support occurred[2].

Radiographic evaluation

Radiographic assessments were made before and after surgery inall patients (Fig.2, Fig.3, Fig.4). Plain roentgenograms and bilat-eral oblique plain films were taken to evaluate the fusion process,and computed tomography (CT) and magnetic resonance imagingMRI were performed preoperatively. Fusion status was assessedradiographically by the presence of bridging trabeculae with ho-mogeneous fusion mass and by lateral flexion/extension radio-graphs to evaluate motion at the fusion site (as physiological seg-mental motion is considered to be less than 2° according to Chanet al. [9]). To estimate sagittal plane correction within the instru-mented spine and alignment of the cervical and cervicothoracicspine, we compared preoperative with postoperative values of (1)cervical lordosis [5] (skull–C7) in the 21 patients who underwentinstrumentation at any level from the skull to C7, and (2) cervi-cothoracic junctional kyphosis [5] (C7–T2) in the 11 patients whounderwent instrumentation and fusion at the cervicothoracic junc-tion. Measurements were made as follows.

1. Cervical lordosis (average normal =–44.2° according to Brid-well [5]) is the angle formed by a line drawn from the posterioroccipitum to the clivus and a second line on the lower endplateof C7 (Table 2).

2. Cervicothoracic junctional kyphosis (average normal =5° ac-cording to Bridwell [5]) is the angle formed by a line drawn onthe upper endplate of C7 and the line drawn on the lower end-plate of T2 (Table 3).

389

Fig.1 Schematic demonstration of the compact Cotrel-Dubousset(CCD) hardware in situ

3. Sagittal vertebral translation (Fig.2) was measured on lateralroentgenograms of the cervical spine, preoperatively, postoper-atively and at the latest follow-up, as the amount (mm) of slip-ping of any vertebral body on the adjacent lower vertebral body(Table 2, Table 3).

Statistical methods

Statistica software and the paired and unpaired t-test were used toevaluate and to compare the results.

Results

There was one death in the early postoperative period be-cause of acute vascular cerebral hemorrhage (Table 1).Thus 31 of the 32 patients were available for the follow-up evaluation. The follow-up averaged 31±6 months(range 25–44 months). The operated patients with pri-mary tumors and prostate metastasis were still alive at thelatest follow-up evaluation. Operation time and intraoper-ative blood loss for the posterior procedure were mea-sured exclusive of the anterior or combined surgery. Aver-age operation time for the posterior operation was 120 min, (range 90–180 min), including the time forlaminectomy and tumor resection, and average intraoper-ative blood loss was 300 ml, (range 60–1500 ml). Periop-erative blood transfusion was not needed. The number ofsegments fused averaged 5±3 (range 2–14). For fusionwithin the region from skull to C7, an average of 4±2 seg-ments were fused, whereas the average number for thecervicothoracic junction was 8±3 segments. There was no

390

Table 2 Sagittal plane align-ment in patients who under-went occipitocervical and cer-vical CCD instrumentation

Parameters Preoperatively Postoperatively Follow-up Mean (range) Mean (range) Mean (range)

Skull–C7 lordosis (degrees) –30.9 (–10 to –42) –40.9 (–36 to –44) –40.4 (–34 to –44)Vertebral translation (mm) 1.5 (0 to 3) 0.4 (0 to 2) 0.4 (0 to 2)

Table 3 Sagittal plane align-ment (C7–T2) in patients whounderwent cervicothoracicCCD instrumentation and fu-sion

Parameters Preoperatively Postoperatively Follow-up P-valueMean (range) Mean (range) Mean (range)

C7–T2 kyphosis (degrees) 7.6 (5–12) 5.3 (4–7) 5.3 (4–7) <0.05Vertebral translation (mm) 3.55 (0–12) 0.4 (0–12) 0.4 (0–1.2) <0.01

Fig.2A–C Case 17. A Lateral roentgenogram of the cervicalspine, showing a unilateral dislocation of C5-C6. B Lateral plainroentgenogram 25 months after surgery, demonstrating excellentreduction and fusion using CCD hooks. C Anteroposterior roent-genogram showing the CCD instrumentation in situ with the trans-verse connector

neurologic deterioration in this series. Thirteen patientsimproved their preoperative Frankel status by an averageof 1.2 grades, (range 0–2) (Table 1). Of the eight patientswho had preoperative myelopathy, six improved by atleast one Frankel grade (improvement rate of 75%) andthe other two patients remained at the same Frankel grade

(Table 1). Radiculopathy in three patients resolved com-pletely in the first 3 postoperative months. There was aneurologic improvement in 53% of the operated patients.At the latest follow-up, all 31 patients who had undergonebone grafting had solid bony union. There were no pa-tients with failure of implant components or connections,

391

Fig.3A,B Case 6. A Lateralroentgenogram showing odon-toid pseudarthrosis with poste-rior dislocation associated withmyelopathy. B Lateral roent-genogram of the cervical spine31 months postoperatively af-ter C1–C3 laminectomy andoccipitocervical fusion usingthe CCD hook-plate system

Fig.4A,B Case 2. A Com-puted tomography (CT) scanshowing severe destruction ofT2 on the right side. B Lateralroentgenogram of the cervi-cothoracic junction after stagedanterior tumor resection andplating-grafting and posteriorCCD instrumentation, 35months after surgery. Note thecombination of the cervicalCCD with the CD instrumenta-tion at the cervicothoracicjunction, using the special“domino” connectors

screw loosening, or lucent zone formation around thescrews penetrating the skull. One patient (patient 23,Table 1) had constant complaints in the midthoracic re-gion at the lower end of the rods, and he underwent a sec-ond operation 1 year after the index operation for removalof the hardware. The fusion area under and around thehardware was solid, without signs of pseudarthrosis.

The pain, on a scale of 0–3, improved in all patients byan average of 2.4 grades (range 1–3). The excellent andgood results (overall 77%) were recorded for 18 of the 21 patients with instrumentation within the regionskull–C7 and 8 of the 11 patients who underwent instru-mentation at the cervicothoracic junction. The fair andpoor results were due to associated neurologic impairmentor myelopathy, and were observed in 4 of the 21 patientswith fusion in the region skull–C7 and in 3 of the 11 pa-tients with stabilization of the cervicothoracic junction(Table 1).

The operation improved significantly (P<0.05) thesagittal profile of the cervical spine (Table 2) by reducingany kyphotic deformation, and offered a cervical lordosis(skull–C7) that was close to the physiological. Vertebraltranslation in the region skull–C7 was significantly (P<0.01) reduced postoperatively and remained unchanged atthe latest follow-up evaluation (Table 2). There was an av-erage of 73% reduction of the anterior displacement of thevertebrae in this region. At the cervicothoracic junction,the operation significantly (P<0.05) reduced pathologi-cally increased kyphosis, which remained unchanged atthe latest follow-up (Table 3). There was an average of90% reduction of the ventral olisthesis of any vertebra atthis junction.

Discussion

The present study showed that cervical CCD applied inthe craniovertebral, cervical and cervicothoracic spine forvarious conditions offered immediate stabilization of thespine, significant reduction of vertebral olisthesis, mainte-nance of its reduction until the latest observation, restora-tion of the lateral profile of the spine in the craniocervical,cervical and cervicothoracic region, relief of acute andchronic pain, improvement of neurologic impairment andmyelopathy and excellent and good functional results inthe vast majority of the operated patients. However, sev-eral cases included in the present series received multi-level fusions even when one- or two-level pathology ex-isted. This was performed because the CCD, as a hook androd system, requires longer instrumentation because of:

1. Use of the “claw” configuration at the two uppermostand two lowermost levels of fixation

2. Extension of instrumentation above and below anylevels of prior laminectomy(-ies), and

3. The need to avoid using hooks at the levels of signifi-cant spinal canal stenosis

These are theoretically the limitations and demerits of theCCD system in comparison to the recently developedspinal fixation devices [32, 37] of lateral mass screw andpedicular systems.

However, a recent clinical-neurophysiological study[4] in cadavers (without spinal stenosis) showed that cer-vical CCD hooks can be used with safety and reliabilitywithout neurologic complications during and after sur-gery. One cadaveric study [14] showed a close anatomicalrelationship with the dura and spinal cord, with a meancervical CCD hook intrusion into the spinal canal of 27%of the diameter; however, they found no evidence of de-formation of the spinal cord.

Bueff et al. [7] performed in 1995 a comparative bio-mechanical study between three instrumentation con-structs (posterior Synthes lateral mass plate, cervicalCotrel-Dubousset rod-hooks and anterior Synthes plate),applied at the destabilized cervicothoracic junction, andshowed that the posterior instrumentation had more stiff-ness than the anterior plate, and the stiffness of the poste-rior Synthes plate is the same as that of CCD.

The first preliminary report on successful use of a newrod-screw instrumentation, applied posteriorly for stabi-lization of the occipitocervical, cervical and upper tho-racic spine, was that of Jeanneret [23] in 1996.

There are only a few articles reporting on the use of theCotrel-Dubousset rod and hook system in the cervicalspine applied for different disorders, mostly for occipito-cervical instability [15, 22, 33], and all reported stablebony fusion in all cases, no surgery-related morbidity ormortality and no hardware failure.

Olerud et al. [32] recently published the results of theirscrew-rod posterior system, applied in 30 patients withvarious conditions. They reported good results in all butone patient, one neurologic deterioration, two cases ofloss of fixation, two infections and one hematoma.

Sasso et al. [37] reported their results in occipitocervi-cal fusions with posterior plate and screw instrumentationin 32 patients. They reported no pseudarthrosis, and animprovement or no change in the preoperative neurologicstatus in the operated patients.

In this series, there were no patients with pseudarthro-sis, no cases of skull screw loosening and no rod break-age, in spite of early ambulation without postoperativerigid external support.

Regarding postoperative immobilization, Chan et al.[9] recommended a halo vest, and Ranawat et al. [35] ob-served that the more stable the fixation obtained duringposterior cervical spinal fusion, the more satisfactory thefusion. Early series showed fusion rates ranging from 50to 94% [9, 16]. The fusion rate in this series was 100%without use of rigid postoperative orthoses, and it there-fore seems that cervical CCD is sufficiently stable to se-cure solid fusion.

The few poor and fair results were due to myelopathyor to lack of neurologic improvement postoperatively.

392

The anatomy at the cervicothoracic junction is uniquein that there is a change from cervical lordosis to thoracickyphosis, and thus the cervicothoracic junction poses spe-cial concerns for posterior spinal surgical procedures, be-cause it is an area of transitional anatomy [3, 7]. To theauthors’ knowledge there are no previous reports on theapplication of cervical CCD at the cervicothoracic junc-tion. In all previous reports on fusion in the cervicotho-racic region, no mention was made of anything that mightsuggest that the sagittal profile of the spine had been re-stored. In the present series the preoperative and postop-erative measurements showed that the appropriate con-touring of the Cotrel-Dubousset rods offers a sufficient

cervicothoracic transitional kyphosis that does not differfrom that of a normal spine.

In conclusion, cervical CCD hook-rod instrumentationprovided good clinical and radiological results in the vastmajority of traumatic and non-traumatic lesions extendingfrom the skull to the upper thoracic spine, without neuro-logic complications. This posterior instrumentation aloneor in combination with anterior reconstruction providesexcellent initial stability and safeguards solid fusion,maintaining the sagittal profile of the craniocervical, cer-vical and cervicothoracic spine. It is one of the potentialprocedures for posterior reconstruction of the cervicalspine with various kinds of disorders.

393

1.Abumi K, Kaneda K (1997) Pediclescrew fixation for nontraumatic lesionsof the cervical spine. Spine 16:1853–1863

2.Abumi K, Panjabi MM, Duranceau J(1989) Biomechanical evaluation ofspinal fixation devices. II. Stabilityprovided by six spinal fixation devicesand interbody body graft. Spine 14:1249–1255

3.An HS, Vaccaro A, Cotler JM, Lin S(1994) Spinal disorders at the cervi-cothoracic junction. Spine 19:2557–2564

4.Bancel PH, Vurpillat MD, Boulut Ph(1988) Can sub-laminar hooks besafely used in cervical spine surgery?A clinical, radiological and electricalstudy. The 65th Annual Meeting of theAAOS, New Orleans, March 12–22

5.Bridwell KH (1994) Normal sagittalalignment. Sagittal Spinal BalanceSymposium 1. Federation of SpineAssociations, Section IV, ScoliosisResearch Society, February 27

6.Brooks AL, Jenkins EB (1978) At-lantoaxial arthrodesis by the wedgecompression method. J Bone JointSurg Am 60:279–284

7.Bueff HU, Lotz JC, Colliou OK,Khapchik V, Asford F, Hu SS, BozicK, Bradford DS (1995) Instrumentationof the cervicothoracic junction afterdestabilization. Spine 15:1789–1792

8.Cantone G, Ciapperta P, Delfini R(1997) New steel device for occipito-cervical fixation: technical note. J Neu-rosurg 60:1104–1106

9.Chan KPD, Ngian SK, Cohen L (1992)Posterior upper cervical fusion inrheumatoid arthritis. Spine 17:268–272

10.Chapman JR, Anderson PA, Pepin C,Toomey S, Newell DW, Grady MS(1996) Posterior instrumentation of theunstable cervicothoracic spine. J Neu-rosurg 84:552–558

11.Coe JD, Warden KE, Surrerlin III CE,Mcfee PC (1989) Biomechanical eval-uation of cervical stabilization methodsin a human cadaveric model. Spine14:1223–1231

12.Crellin RQ, Maccabe JJ, Hailton EBD(1970) Severe subluxation of the cervical spine in rheumatoid arthritis. J Bone Joint Surg Br 52:244–251

13.Ebraheim NA, Rupp RE, SavolaineER, Brown JA (1995) Posterior platingof the cervical spine. J Spinal Disord 8:111–115

14.Fagerstroem T, Hedlund R, Bancel P,Robert R, Dupas B (1998) CotrelDubousset instrumentation in the cervi-cal spine. An experimental study onthe relation of hooks to the spinal cord.The 65th Annual Meeting of theAAOS, New Orleans, March 12–22

15.Faure A, Bord E, Monteiro da Silva R,Diaz Saldana A, Robert R (1998) Occi-pitocervical fixation with a single occi-pital clamp using inverted hooks. EurSpine J 7:80–83

16.Ferlic DC, Clayton ML, Leidholt JD,Gambie WE (1975) Surgical treatmentof symptomatic unstable cervical spinein rheumatic arthritis. J Bone JointSurg Am 57:349–354

17.Foerster O (1927) Die Leitungsbahnendes Schmerzgefuehls. Urban andSchwarzenburg, Berlin, p 266.

18.Gallie WE (1939) Fractures and dislo-cations of the cervical spine. Am JSurg 46:495–499

19.Gill K, Paschal S, Corin J, Ashman R,Bucholz RW (1988) Posterior platingof the cervical spine. Spine 13:813–816

20.Grob D, Dvorak J, Panjabi M,Froehlich M, Hayek J (1991) Posterioroccipitocervical fusion. A preliminaryreport of a new technique. Spine 16:517–524

21. Hamblen DL (1967) Occipito-cervicalfusion: indications, technique and re-sults. J Bone Joint Surg Br 41:33–45

22.Heidecke V, Rainov NG, Burkert W(1998) Occipito-cervical fusion withthe cervical Cotrel-Dubousset rod sys-tem. Acta Neurochir 140:969–976

23. Jeanneret B (1996) Posterior rod sys-tem of the cervical spine: a new im-plant allowing optimal screw insertion.Eur Spine J 5:350–356

24.Kotani Y, Cunningham BW, Abumi K,McAfee PC (1994) Biomechanicalanalysis of cervical stabilization sys-tems: an assessment of transpedicularscrew fixation in the cervical spine.Spine 19:2529–2539

25.Lieberman IH, Webb JK (1998) Occip-ito-cervical fusion using posterior tita-nium plates. Eur Spine J 7:308–312

26.Logroscino CA, Amadou D, Lavaste F(1996) Development of a new shortmetal construct for the treatment of se-vere craniovertebral instability. SpineState Art Rev 10 (2)

27.McAfee PC, Cassidy JR, Davis RF,North RB, Ducker TB (1991) Fusionof the occipitum to the upper cervicalspine. Spine 16S:S490–S494

28.McGraw, Rush RM (1973) Atlanto-ax-ial arthrodesis. J Bone Joint Surg Br55:482–489

29.Montesano PX, Jauch E, Johnsson H Jr(1992) Anatomic and biomechanicalstudy of posterior cervical spine platearthrodesis: an evaluation of two dif-ferent techniques of screw placement. J Spinal Disord 5:301–305

30.Mori T, Matsunaga S, Sunahara N,Sakou T (1998) 3- to 11-year follow-up of occipitocervical fusion forrheumatoid arthritis. Clin Orthop 351:169–179

31.Nagashima C (1973) Surgical treat-ment of irreducible atlanto-axial dislo-cation with spinal cord compression.Case report. J Neurosurg 38:374–378

32.Olerud C, Lind B, Sahlstedt B (1999)The Olerud Cervical Fixation System;a study of safety and efficacy. Ups JMed Sci 104:131–143

References

394

33.Paquis P, Lonjon M, Grellier P (1998)Craniovertebral stabilization using anew CCD technique. Neurochirurgie.44:101–104

34.Paquis P, Breuil V, Lonjon M, Eullor-Ziegler L, Grellier P, Crockard HA,Sonntag VKH, Benzel EC, Cooper PR(1999) Occipitocervical fixation usinghooks for upper cervical instability.Neurosurgery 44:324–331

35.Ranawat CS, O’Leary P, Pellicci P, etal (1979) Cervical spine fusion inrheumatoid arthritis. J Bone Joint SurgAm 61:1003–1010

36.Richman JD, Daniel TE, AndersonDD, Miller PL, Douglas RA. (1995)Biomechanical evaluation of cervicalspine stabilization methods using aporcine model. Spine 20:2192–2197

37.Sasso RC, Jeanneret B, Fischer K,Magerl F (1994) Occipitocervical fu-sion with posterior plate and screw in-strumentation. A long-term follow-upstudy. Spine 15:2364–2368

38.Stecken J, Boissonnet H, Manzo L, etal (1987) Mal de pot sous-occipital.Neurochirurgie 33:482–486

39.Sutterlin CE, McAfee PC, Warden KE,Rey RM, Farey ID (1988) A biomech-nical evaluation of cervical spine stabi-lization in a bovine model. Spine 13:795–802

40.Ulrich C, Woedsdoerfer O, Kalf R,Claes L, Wilke H-J (1991) Biomechan-ics of fixation systems to the cervicalspine. Spine 16:54–59