A Cochrane Review

COCHRANE COLLABORATION

SPINE Volume 38, Number 1, pp 24–36©2012, Lippincott Williams & Wilkins

24 www.spinejournal.com January 2013

Total Disc Replacement for Chronic Discogenic Low Back Pain

A Cochrane Review

Wilco C.H. Jacobs , MSc , * Niels A. van der Gaag , MD , * Moyo C. Kruyt , MD, PhD , † Alexander Tuschel , MD , ‡ Marinus de Kleuver , MD, PhD , § Wilco C. Peul , MD, PhD , *¶ Abraham J. Verbout , MD, PhD , † and F. Cumhur Oner , MD, PhD †

Study Design. Systematic literature review. Objective. To assess the effect of total disc replacement for chronic low back pain due to lumbar degenerative disc disease compared with fusion or other treatment options. Summary of Background Data. There is an increasing use in disc replacement devices for degenerative disc disease, but their effectiveness compared with other interventions such as fusion of the motion segment or conservative treatment remains unclear. Methods. A comprehensive search in PubMedCentral, MEDLINE, EMBASE, BIOSIS, ClinicalTrials.gov, and FDA trials register was conducted. Randomized controlled trials comparing total disc replacement with any other intervention for degenerative disc disease were included. Risk of bias was assessed using the criteria of the Cochrane Back Review Group. Quality of evidence was graded according to the GRADE approach. Two review authors independently selected studies, assessed risk of bias, and extracted data. Results and upper bounds of confi dence intervals were compared with predefi ned clinically relevant differences.

The development of artifi cial joint prostheses has almost obviated the need for fusion of painfully and degener-ated major peripheral joints. For the lumbar spine, the

fi rst disc arthroplasty model was described by Fernstrom 1 but was quickly abandoned because of poor outcome. There are currently several disc replacement devices on the market, of which CHARITÉ (DePuy Spine, Raynham, MA) and ProDisc (DePuy Synthes Spine, Oberdorf, Germany) are approved by the FDA and the Maverick (Medtronic, Memphis, TE) and the Flexicore (Stryker Spine, Allendale, NJ) are in the investi-gational stage (phase III clinical trial). CHARITÉ has the most extensive (least constrained) movement possibilities, ProDisc and Maverick have more constraints and Flexicore has the least movement.

From the * Department of Neurosurgery, Leiden University Medical Center, Leiden, the Netherlands ; † Department of Orthopedics, University Medical Center Utrecht, Utrecht, the Netherlands ; ‡ Orthopaedic Hospital Vienna Speising, Vienna, Austria ; § Department of Orthopedic Surgery, Sint Maartenskliniek, Nijmegen, the Netherlands and Department of Orthopedic Surgery, Free University Medical Center (VUmc) Amsterdam, the Netherlands; and ¶ Department of Neurosurgery, Medical Center Haaglanden, The Hague, the Netherlands.

Acknowledgment date: September 10, 2012. Acceptance date: September 11, 2012.

This review is adopted from the Cochrane Review: “Jacobs W, van der Gaag NA, Tuschel A, de Kleuver M, Peul W, Verbout A, Oner FC. Total disc replacement for chronic discogenic low- back pain. The Cochrane Database of Systematic Reviews 2012, Issue 9. Art. No.: CD008326. DOI: 10.1002/14651858.CD008326 ”. Copyright Cochrane Library, reproduced with permission.

The device(s)/drug(s) is/are FDA-approved or approved by corresponding national agency for this indication.

No funds were received in support of this work.

Relevant fi nancial activities outside the submitted work: board membership, consultancy, employment, grants, payment for lectures.

Address correspondence and reprint requests to Wilco Jacobs, MSc, Department of Neurosurgery, Leiden University Medical Center, PO Box 9600, 2300 RC Leiden, the Netherlands; E-mail: [email protected] ; [email protected]

Results. We included 7 randomized controlled trials with a follow-up of 24 months. There is risk of bias in the included studies due to sponsoring and absence of any kind of blinding. One study compared disc replacement with rehabilitation and found a signifi cant advantage in favor of surgery, which, however, did not reach the predefi ned threshold. Six studies compared disc replacement with fusion and found that the mean improvement in visual analogue scale score of back pain was 5.2 mm higher (2 studies; 95% confi dence interval 0.2–10.3) with a low quality of evidence. The improvement of Oswestry disability index score at 24 months in the disc replacement group was 4.3 points more than in the fusion group (5 studies; 95% confi dence interval 1.85–6.68) with a low quality of evidence. Both upper bounds of the confi dence intervals were below the predefi ned clinically relevant difference. Conclusion. Although statistically signifi cant, the differences in clinical improvement were not beyond generally accepted boundaries for clinical relevance. Prevention of adjacent level disease and/or facet joint degeneration was not properly assessed. Therefore, because we think that harm and complications may occur after some years, the spine surgery community should be prudent to adopt this technology on a large scale, despite the fact that total disc replacement seems to be effective in treating low back pain in selected patients, and in the short term is at least equivalent to fusion surgery. Key words: degenerative disc , arthroplasty , meta-analysis , arthrodesis , prostheses and implants. Spine 2013 ; 38 : 24 – 36

DOI: 10.1097/BRS.0b013e3182741b21

Copyright © 2012 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

BRS205292.indd 24BRS205292.indd 24 04/12/12 9:25 AM04/12/12 9:25 AM

COCHRANE COLLABORATION Total Disc Replacement for Chronic Discogenic Low Back Pain • Jacobs et al

Spine www.spinejournal.com 25

TAB

LE 1

. Sea

rch

Stri

ngs

Use

d fo

r D

iffer

ent D

atab

ases

D

atab

ase

Sear

ch S

trin

gs*

MED

LIN

ER

ando

miz

ed c

ontro

lled

trial

Pros

pect

ive

stud

ies

Con

trol

Dis

citis

Dis

cM

aver

ick

Con

trolle

d cl

inic

al tr

ial

Cro

ss-o

ver s

tudi

esPr

ospe

ctiv

eD

orsa

lgia

Pros

thes

es a

nd im

plan

tsM

obid

isc

Ran

dom

ized

Epid

emio

logi

c st

udie

sV

olun

teer

Bac

kach

eIm

plan

ts, e

xper

imen

tal

Flex

icor

e

Plac

ebo

Cas

e-co

ntro

l stu

dies

Coh

ort s

tudi

esSp

ondy

losi

sPr

osth

esis

impl

anta

tion

Nub

ac

Dru

g th

erap

yC

ohor

t stu

dies

Coh

ort

Lum

bago

Arth

ropl

asty

, rep

lace

men

tA

crofl

ex

Ran

dom

lyC

ross

-sec

tiona

l st

udie

sEp

idem

iolo

gic

met

hods

Lum

bar p

ain

Arth

ropl

asty

Artifi c

ial

Tria

lC

ohor

t ana

lysi

sC

ase

serie

sSc

iatic

neu

ropa

thy

Dis

c re

plac

emen

tFl

exib

le

Gro

ups

Follo

w u

p st

udy

Spin

al d

isea

ses

Scia

tica

Kin

emat

icM

obile

Com

para

tive

stud

yO

bser

vatio

nal s

tudy

Bac

k pa

inD

isc

adj d

egen

erat

ion

Act

ive

lK

inem

atic

Eval

uatio

n st

udie

sLo

ngitu

dina

lIn

terv

erte

bral

dis

cSp

ine

Cha

rite

Pres

tige

Follo

w-u

p st

udie

sR

etro

spec

tive

Inte

rver

tebr

al d

isc

Spin

alPr

odis

cPo

rous

coa

ted

mot

ion

Cro

ss s

ectio

nal

Dis

plac

emen

tV

erte

bra

Kin

efl e

x

EMBA

SEC

linic

al a

rticl

ePl

aceb

oR

ando

mLo

w b

ack

pain

Dis

cA

ctiv

e l

Clin

ical

stu

dyA

lloca

teSi

ngle

or d

oubl

e or

trip

leLu

mba

r pai

nIn

terv

erte

bral

dis

cC

HA

RIT

É

Clin

ical

tria

lA

ssig

nO

r trip

leIs

chia

lgia

Deg

ener

atio

nK

inefl

ex

Con

trolle

d st

udy

Blin

dM

ask

Spon

dylo

sis

Pres

tige

Mav

eric

k

Ran

dom

ized

con

trolle

dC

linic

stu

dy o

r tria

lV

ersu

sLu

mba

goPo

rous

coa

ted

mot

ion

Mob

idis

c

Tria

lC

ompa

reSp

ine

Scia

tica

Prod

isc

Flex

icor

e

Maj

or c

linic

al s

tudy

Con

trol

Spin

e di

seas

eB

ack

pain

Pros

thes

isN

ubac

Dou

ble

blin

d pr

oced

ure

Cro

ss-o

ver

Bac

kach

eD

orsa

lgia

Impl

ant

Acr

ofl e

x

Mul

ticen

ter s

tudy

Fact

oria

lIn

terv

erte

bral

dis

cD

isci

tisIm

plan

tatio

nA

rtifi c

ial

Sing

le b

lind

proc

edur

eFo

llow

-up

Inte

rver

tebr

al d

isc

Dis

c de

gene

ratio

nA

rthro

plas

tyFl

exib

le

Phas

e 3

clin

ical

tria

lPl

aceb

oIn

terv

erte

bral

dis

c he

rnia

Ver

tebr

aIn

terv

erte

bral

dis

kect

omy

Kin

emat

ic

Phas

e 4

clin

ical

tria

l cr

osso

ver p

roce

dure

Pros

pect

ive

Dis

citis

Kin

emat

ic

BIO

SIS

Dis

cA

ctiv

e l

Prod

isc

Bac

kach

eSp

ine

Follo

w-u

p st

udy*

Rep

lace

men

tC

HA

RIT

ÉPr

estig

eD

orsa

lgia

Clin

ical

tria

l*Pr

ospe

ctiv

e st

udy*

Kin

emat

icM

aver

ick

Lum

bar v

erte

bra*

Dis

citis

*R

esea

rch

desi

gnR

ando

m*

( Con

tinue

d )





TAB

LE 1

. ( C

ontin

ued

) D

atab

ase

Sear

ch S

trin

gs*

Artifi c

ial

Flex

icor

eSc

iatic

aIn

terv

erte

bral

dis

c*C

ompa

rativ

e st

udy*

Plac

ebo*

Flex

ible

Arth

ropl

asty

Lum

bago

Low

bac

k pa

inEv

alua

tion

stud

y*Si

ngle

blin

d*

Mob

ilePo

rous

coa

ted

mot

ion

Spon

dylo

sis

Bac

k pa

inC

ontro

lled

trial

*D

oubl

e bl

ind*

PubM

edC

entra

lB

ack

pain

Scia

tica

Dis

c de

gene

ratio

nFa

iled

back

Dis

plac

e* d

isc

Arth

ropl

asty

Dor

salg

iaSp

ondy

losi

sD

isc

prol

apse

Cau

da e

quin

aPr

olap

se*

disc

Dis

c re

plac

emen

t

Bac

kach

eSc

iatic

aD

isc

hern

iatio

nLu

mba

r ver

tebr

a*Pr

osth

eses

and

impl

ants

Pres

tige

Low

bac

k pa

inSp

ine

Spin

al fu

sion

Spin

al s

teno

sis

Pros

thes

is im

plan

tatio

nPr

odis

c

Lum

bar p

ain

Spin

al d

isea

ses

Face

t joi

nts

Slip

ped

disc

Impl

ants

, exp

erim

enta

lPo

rous

coa

ted

mot

ion

Coc

cyx

Lum

bago

Inte

rver

tebr

al d

isc

Deg

ener

ativ

e* d

isc

Arth

ropl

asty

Kin

emat

ic d

isc

Coc

cydy

nia

Dis

citis

Post

lam

inec

tom

y ar

achn

oidi

tisSp

inal

ste

nosi

sA

rthro

plas

ty, r

epla

cem

ent

*For

rea

dabi

lity

we

omitt

ed c

onne

ctor

s, s

earc

h fi e

lds,

exp

lode

opt

ions

, and

mul

tiple

ver

sion

s of

the

sam

e se

arch

term

in 1

dat

abas

e

A possible advantage of total disc replacement is main-tained mobility at the operated level, thus preventing adja-cent segment degeneration. However, Leivseth et al 2 evaluated ProDisc II and found decreased mobility at the surgical level. SariAli et al 3 found normal kinematics with SB CHARITÉ prostheses, but only for 1-level procedures. Huang et al 4 found only increased adjacent segment mobility for cases with decreased mobility at the surgical level. In a literature review on adjacent segment degeneration, Harrop et al 5 show more adjacent segment degeneration after fusion procedures than after total disc replacement, but the studies supporting this were of low quality. Park 6 found progression of degeneration in the facet joints with ProDisc II and Rousseau et al 7 showed an increased facet joint loading in human cadavers. Chung et al 8 found that when constrained devices are used, the load-ing on the facet joints increases.

Salvage options are important when deciding in favor of new procedures such as total disc replacement. Punt et al 9 shows that removal of the damaged disc, with anterior and posterior fusion gives better outcomes than posterior fusion alone in salvage procedures. Complications after removal have been identifi ed, 10 and indeed salvage procedures are being investigated. 11 In a literature review, Patel et al 12 point to the need for proper patient selection for initial disc replace-ment to avoid revision.

Before large-scale implementation of disc replacement can be accepted we need clarity short- and long-term per-formances. Therefore, we conducted a systematic literature review to obtain answers for the following:

❑ What is the effectiveness of disc replacement versus fu-sion or other treatment options?

❑ What is the safety of disc replacement with regard to: loosening, subsidence, wear, adjacent segment degenera-tion, facet joint degeneration, and perioperative compli-cations?

❑ Is there an acceptable and safe salvage procedure in case of failure?

MATERIALS AND METHODS

Search and Selection The following sources were searched (by Rachel Couban) up to December 22, 2011:

❑ PubMedCentral ❑ MEDLINE (from 1966) ❑ EMBASE (from 1980) ❑ BIOSIS (from 2004) ❑ FDA register ❑ ClinicalTrials.gov ❑ Citation tracking through ISI Thompson ❑ References of included studies

Search strings are given in Table 1 . Two reviewer authors (W.J., A.T.), with referee (M.K.) whenever necessary, inde-pendently selected the articles from the list of identifi ed ref-erences.





No restriction was made with regard to language or date.

❑ We included studies that include patients scheduled for surgery for chronic (lasting longer than 12 wk) degenera-tive disc disease (DDD).

Criteria for Considering Studies for This Review

❑ Studies for the review were randomized controlled tri-als, published in a peer-reviewed journal as a full article, excluding grey literature and conference proceedings.

TABLE 2. Criteria for Risk of Bias Assessment Question Criteria for “Yes” Judgment

A 1. Was the method of randomization adequate?

A random (unpredictable) assignment sequence. Examples of adequate methods are coin toss, rolling a dice, drawing of ballots with the study group labels from a dark bag, computer-generated random sequence, preordered sealed envelops, and sequentially ordered vials.

Yes/No/Unsure

Examples of inadequate methods are alternation, birth date, social insurance/security number, and hospital registration number.

B 2. Was the treatment allocation concealed?

Assignments are generated by an independent person not responsible for determining the eligibility of the patients. This person has no information about the persons included in the trial and has no infl uence on the assignment sequence or on the eligibility decision of the patient.

Yes/No/Unsure

C 3. Was the patient blinded to the intervention?

The index and control groups are indistinguishable for the patients. Yes/No/Unsure

4. Was the care provider blinded to the intervention?

The index and control groups are indistinguishable for the care providers. Yes/No/Unsure

5. Was the outcome assessor blinded to the intervention?

For patient-reported outcomes with adequately blinded patients for outcome criteria that supposes a contact between participants and outcome assessors: the blinding procedure is adequate if patients are blinded, and the treatment or adverse effects of the treatment cannot be noticed during examination for outcome criteria that do not suppose a contact with participants: the blinding procedure is adequate if the treatment or adverse effects of the treatment cannot be noticed during the assessment for outcome criteria that are clinical or therapeutic events that will be determined by the interaction between patients and care providers, in which the care provider is the outcome assessor: the report needs to be free of selective outcome reporting.

Yes/No/Unsure

D 6. Was the dropout rate described and acceptable?

The number of participants who were included in the study but did not complete the observation period or were not included in the analysis are described and reasons are given and are < 20% for short-term and < 30% for long-term follow-up.

Yes/No/Unsure

7. Were all randomized participants analyzed in the group to which they were allocated?

All randomized patients are reported/analyzed in the group they were allocated to by randomization for the most important moments of effect measurement (minus missing values) irrespective of noncompliance and cointerventions.

Yes/No/Unsure

E 8. Are reports of the study free of suggestion of selective outcome reporting?

In general, we expected studies comparing interventions for low back pain to assess at least pain, Oswestry (or similar) scores and to evaluate mobility / fusion of the motion segment. If there were too many studies without these parameters, reporting bias was suspected, unless confi rmed by a full protocol.

Yes/No/Unsure

F 9. Were the groups similar at baseline regarding the most important prognostic indicators?

The groups have to be similar at baseline regarding demographic factors, duration, and severity of complaints, percentage of patients with neurological symptoms, and value of main outcome measure(s).

Yes/No/Unsure

10. Were cointerventions avoided or similar?

There were no cointerventions or they were similar between the index and control groups.

Yes/No/Unsure

11. Was the compliance acceptable in all groups?

The compliance with the interventions is acceptable, based on the reported intensity, duration, number, and frequency of sessions for both the index intervention and control intervention(s). For single-session interventions ( e.g., surgery), this item is irrelevant.

Yes/No/Unsure

12. Was the timing of the outcome assessment similar in all groups?

Timing of outcome assessment was identical for all intervention groups and for all important outcome assessments.

Yes/No/Unsure





❑ The interventions evaluated in the trials were total disc replacement compared with any other treatment for lum-bar DDD.

❑ Patient-centered outcomes were of primary interest in this review. We predefi ned pain, overall improvement, patient satisfaction, and specifi c or general functional measures to be of primary interest (Oswestry, RMDQ, SF-36, EuroQOL), but we made no exclusions on the type of outcome measure. The minimal required dura-tion of follow-up was 6 months.

Assessment of Risk of Bias in Included Studies The 2 review authors (W.J., N.vd.G.), with referee (M.K.), if necessary, independently assessed the risk of bias of the selected articles with the 12 criteria recommended by the Cochrane Back Review Group. 13 Criteria and operationaliza-tion are given in Table 2 . Studies were categorized as having a “low risk of bias” when at least 6 of the 12 criteria were met and the study had no serious methodological fl aws.

Data Extraction and Management Data were extracted onto separate, predeveloped forms. From each study, basic information was gathered concern-ing authors (affi liation, sponsoring), methods (study design, sample size), patients (selection criteria and diagnoses, pain location, age, sex), treatments (implant models, constraints, materials, levels involved), control treatments (nonsurgical vs . surgical, fusion vs. nonfusion, and anterior column vs . posterior or circumferential), and outcome variables with results. Data were extracted by 1 review author (W.J.). Prin-cipal investigators of included trials were contacted with the request for clarifi cations about items with an unclear or high risk of bias. Missing clinical data in trials were accepted when this was less than 20%, otherwise, the trial was excluded from the specifi c analysis of the outcome parameters and fol-low-up moments. Missing information about parameter vari-ability was estimated from ranges if provided or estimated from comparable trials.

Figure 1. Flow chart of search and selection process.

TAB

LE 3

. Cha

ract

eris

tics

of In

clud

ed S

tudi

es

Ref

eren

ceSa

mpl

e Si

ze%

Fe

mal

eA

vera

ge A

ge

(Ran

ge, S

D)

Part

icip

ants

(Exc

lusi

on

Cri

teri

a N

ot G

iven

)In

terv

entio

nsO

utco

mes

Follo

w-u

p

Ber

g et

al 18

15

259

39.4

(21–

55,

8.0)

DD

D, 1

or 2

leve

l, M

RI

dege

nera

tion

CH

AR

ITÉ,

Pro

Dis

c, o

r Mav

eric

kV

AS

leg

pain

Preo

pera

tive

LBP

> 1

yea

r; LB

P >

leg

pain

Fusi

on w

ith lo

cal o

r ilia

c cr

est

bone

gra

ft, p

edic

le s

crew

s,

post

erol

ater

al o

r int

erbo

dy

fusi

on w

ith B

rant

igan

cag

e

VA

S ba

ck p

ain

1 an

d 2

yr

OD

I > 3

0 or

bac

k pa

in V

AS

> 5

0SF

-36

20–5

5 yr

Osw

estry

dis

abili

ty in

dex

scor

e

> 3

mo

cons

erva

tive

treat

men

tEQ

-5D

Wor

k st

atus

Blu

men

thal

et a

l 20

304

4839

.6 (1

9–60

, 8.

5)D

DD

, sin

gle

leve

l, L4

–S1,

di

scog

raph

y co

nfi r

med

CH

AR

ITÉ

VA

S ba

ck p

ain

Preo

pera

tive

( Con

tinue

d )





TAB

LE 3

. ( C

ontin

ued

)

Ref

eren

ceSa

mpl

e Si

ze%

Fe

mal

eA

vera

ge A

ge

(Ran

ge, S

D)

Part

icip

ants

(Exc

lusi

on

Cri

teri

a N

ot G

iven

)In

terv

entio

nsO

utco

mes

Follo

w-u

p

Bac

k an

d or

leg

pain

BAK

(2 c

ages

) with

ilia

c cr

est

auto

graf

tO

swes

try d

isab

ility

inde

x sc

ore

6 w

k

OD

I > 3

0; V

AS

> 4

0SF

-36

neur

olog

ical

sta

tus

patie

nt s

atis

fact

ion

RO

M d

isc

heig

ht

neur

ofor

amin

al h

eigh

t se

gmen

tal t

rans

latio

n

3 an

d 6

mo

18–6

0 yr

1 an

d 2

yr

> 6

mo

cons

erva

tive

treat

men

t

Gor

net e

t al 23

57

750

40.0

(18–

70, ?

)D

DD

, L4–

S1, s

ingl

e le

vel

MA

VER

ICK

dis

cO

vera

ll su

cces

s

Bac

k pa

in w

/wo

leg

pain

Ant

erio

r fus

ion

with

LT-

CA

GE

lum

bar t

aper

ed fu

sion

dev

ice

and

INFU

SE b

one

graf

t (fu

sion

)

SF-3

6

Osw

estry

> 30

, bac

k pa

in >

20B

ack

pain

sta

tus

> 6

mo

cons

erva

tive

treat

men

tLe

g pa

in s

tatu

s

18–7

0 yr

Patie

nt s

atis

fact

ion

Patie

nt g

loba

l per

ceiv

ed

effe

ct re

turn

to w

ork

doct

or’s

perc

eptio

n

Hel

lum

et a

l 16

173

5341

.0 (?

, 7.1

)C

hron

ic lo

w b

ack

pain

Dev

ice:

tota

l dis

c pr

osth

esis

Osw

estry

dis

abili

ty in

dex

scor

e

Deg

ener

ativ

e di

sc d

isea

seB

ehav

iora

l: m

ultid

isci

plin

ary

reha

bilit

atio

n pr

ogra

mC

ost-e

ffect

iven

ess

Side

effe

cts

Tim

e ou

t of w

ork

Fear

-avo

idan

ce b

elie

fs

Self-

effi c

acy

for p

ain

Hop

kins

sym

ptom

s

Dru

g co

nsum

ptio

n

EQ-5

D

SF-3

6

( Con

tinue

d )





TAB

LE 3

. ( C

ontin

ued

)

Ref

eren

ceSa

mpl

e Si

ze%

Fe

mal

eA

vera

ge A

ge

(Ran

ge, S

D)

Part

icip

ants

(Exc

lusi

on

Cri

teri

a N

ot G

iven

)In

terv

entio

nsO

utco

mes

Follo

w-u

p

Mor

eno

and

Bou

lot 19

32

4442

(33–

55, ?

)D

DD

, sin

gle

leve

l, L4

–S1,

no

inst

abili

ty, M

odic

1 o

r 2 o

n M

RI

CH

AR

ITÉ

IIIV

AS

Preo

pera

tive

Chr

onic

LB

PA

LIF

with

KLA

cag

e, il

iac

cres

t au

togr

aft a

nd a

nter

ior p

late

Osw

estry

dis

abili

ty in

dex

scor

e6

mo

< 55

yr

Ret

urn

to w

ork

Late

st (2

3 or

26

mo)

> 6

mo

cons

erva

tive

treat

men

t, pr

evio

us d

isce

ctom

y, o

r ch

emon

ucle

olys

is a

llow

ed

Satis

fact

ion

Sass

o et

al 17

67

5137

.7 (?

,?)

DD

D, s

ingl

e le

vel,

L1–S

1Fl

exic

ore

Osw

estry

dis

abili

ty in

dex

scor

ePr

eop

MR

I/CT/

FEX

inst

abili

ty

(tran

slat

iona

l > 2

mm

, rot

atio

nal

> 5º

, hei

ght l

oss

> 2

mm

)

Circ

umfe

rent

ial f

usio

n w

ith

fem

oral

ring

allo

graf

t, pe

dicl

e sc

rew

s an

d ili

ac c

rest

aut

ogra

ft

VA

S ba

ck p

ain

6 w

k

Dis

coge

nic

pain

Blo

od lo

ss3

and

6 m

o

Axi

al p

ain

> ra

dicu

lar p

ain

Ope

ratio

n tim

e1

and

2 yr

18–6

0Le

ngth

of s

tay

Skel

etal

ly m

atur

eR

OM

on fl e

xion

ext

ensi

on

x-ra

ys

BM

I < 4

0

Con

serv

ativ

e tre

atm

ent f

aile

d

Zig

ler e

t al 21

23

651

39.2

(?, 7

.9)

DD

D, s

ingl

e le

vel,

L3–S

1To

tal d

isc

repl

acem

ent:

ProD

isc-

LV

AS

Preo

pera

tive,

po

stop

erat

ive

CT/

MR

I /di

scog

raph

y/FE

X

(inst

abili

ty >

3 m

m tr

ansl

atio

n,

> 5º

ang

ulat

ion;

dis

c he

ight

de

crea

se >

2 m

m; s

carr

ing/

thic

keni

ng a

nnul

us; H

NP;

va

cuum

phe

nom

enon

)

Con

trol:

Circ

umfe

rent

ial f

usio

n w

ith fe

mor

al ri

ng a

llogr

aft w

ith

pedi

cle

scre

ws

and

iliac

cre

st

auto

graf

t

Osw

estry

dis

abili

ty in

dex

scor

e6

wk

Bac

k an

d/or

leg

(radi

cula

r) pa

inSF

-36

3, 6

mo

OD

I > 4

0Ph

ysic

al e

xam

inat

ion

1, 1

.5, a

nd 2

yr

18–6

0 yr

Neu

rolo

gica

l exa

min

atio

n

> 6

mo

cons

erva

tive

ther

apy

DD

D in

dica

tes

dege

nera

tive

disc

dis

ease

; FE

X, F

lexi

on/E

xten

sion

rad

iogr

aphy

; H

NP,

Her

niat

ed N

ucle

us P

ulpo

sis;

MR

I, m

agne

tic r

eson

ance

imag

ing;

CT,

com

pute

d to

mog

raph

y; L

BP,

low

bac

k pa

in;

OD

I, O

s-w

estr

y D

isab

ility

Inde

x; V

AS,

v is

ual a

nalo

gue

scal

e; R

OM

, ran

ge o

f mot

ion;

SF,

Sho

rt-F

orm

.





Analysis Clinical heterogeneity was evaluated for study design, patient characteristics, device design, and control intervention vari-ability. When studies were judged to be clinically homoge-neous, statistical homogeneity was also tested with an I 2 test. When heterogeneity existed, post hoc subgroup analyses and sensitivity analyses were performed to explore the reason for heterogeneity. The results from individual studies were pooled when the studies were judged to be suffi ciently homogeneous after the assessment of heterogeneity (clinical and statistical). For dichotomous outcomes odds ratios were calculated. For continuous outcomes a weighted mean difference was cal-culated. For each outcome, a 95% confi dence interval (95% CI) was computed. A random-effects meta-analysis was used for all analyses. Clinically important changes were evaluated in accordance with the guideline given by Ostelo et al . 14 The minimal clinical important differences of the primary outcome measurements were defi ned as a mean difference of 15 for visual analogue scale ( VAS) pain score (0–100) and 10 of the Oswestry Disability Questionnaire (0–100). In case of absence of a clinical relevant point estimate difference, we also evalu-ated whether the upper bound of the 95% CI was smaller than this difference. The quality of evidence for all primary out-come parameters was evaluated using the GRADE approach 15 for the following criteria (adapted from Furlan et al 13 ):

❑ No risk of bias : 75% of studies had a low risk of bias (6 or more items met, no serious methodological fl aws).

❑ No inconsistency : Studies have consistent fi ndings. ❑ No indirectness : Results are based on direct comparison. ❑ No imprecision : Estimate of effect is suffi ciently precise

(resulting from more than 1 trial, CI narrow and con-clusive, and more than 75% of studies contributing to analysis).

❑ No publication bias : Analysis is free of reporting or pub-lication bias.

On the basis of the number of criteria met, the quality of evidence was judged to be “high” (all criteria), “moderate” (all but 1), “low” (all but 2), or “very low” (all but 3). Impor-tant outcomes for which there are no trials were considered to have “no evidence.” An outcome with only one trial was automatically low quality and if it also had a high ROB, it dropped to very low quality.

RESULTS For results of the selection process, see fl owchart in Figure 1 . Seven studies were included, see Table 3 for characteristics. Four types of total disc replacement were used in the studies, CHARITÉ, ProDisc-L, Maverick and Flexicore. One study, by Hellum et al , 16 compared ProDisc with a rehabilitation protocol with cognitive treatment and physical therapy. Six studies 17–21 , 23 compared total disc replacement with fusion. Three studies were noninferiority trials 20 , 21 , 23 and 1 study was a superiority trial. 16 In 2 studies, design and hypothesis were not described regarding superiority or noninferiority, but these studies were tested for superiority, 18 , 19 or the clinical outcomes were not statistically tested. 17 All studies used the TA

BLE

4. R

isk

of B

ias

Ass

essm

ent o

f Inc

lude

d St

udie

s C

ompa

ring

Dis

c R

epla

cem

ent W

ith F

usio

n or

Oth

er In

terv

entio

ns

Com

pari

son

A1

B2

C3

C4

C5

D6

D7

E8F9

F10

F11

F12

Ref

eren

ceR

ando

m-

izat

ion

Allo

catio

n C

once

alm

ent

Patie

nt

Blin

ding

Surg

eon

Blin

ding

Out

com

e B

lindi

ngD

rop-

outs

Inte

ntio

n to

Tre

atSe

lect

ive

Rep

ortin

gB

asel

ine

Co

Inte

rven

tions

Com

plia

nce

Out

com

e Ti

min

g

Ber

g et

al 18

Ye

sYe

sN

oN

oN

oYe

sYe

s?

No

Yes

Yes

Yes

Blu

men

thal

et a

l 20

Yes

Yes

No

No

No

Yes

Yes

?Ye

sYe

sYe

sYe

s

Gor

net e

t al 23

Ye

sYe

s?

No

?Ye

sYe

s?

Yes

Yes

Yes

Yes

Hel

lum

et a

l 16

Yes

?N

oN

oYe

sYe

sYe

s?

No

Yes

Yes

Yes

Mor

eno

and

Bou

lot 19

?

?N

oN

oN

oYe

sYe

s?

?Ye

sYe

sN

o

Sass

o 17

??

?N

o?

No

??

?Ye

sYe

sYe

s

Zig

ler e

t al 21

Ye

sN

oN

oN

o?

Yes

Yes

?Ye

sYe

sYe

sYe

s





Figure 2. Forest plot of improvement in back pain at 24 months comparing disc arthroplasty versus fusion.

Figure 3. Forest plot of improvement in Oswestry disability index score at 24 months comparing disc arthroplasty versus fusion.

Oswestry Disability Index score and a VAS pain score, speci-fi ed as VAS back pain in 3 studies and completed with a VAS leg pain in 1 study. Follow-up was up to 24 months for all studies with 1 study with a report of 5 year results. 20 All stud-ies included adult patients with back and/or leg pain due to DDD unresponsive to conservative treatment for at least 3 or 6 months. All studies included patients with 1 affected level, except for Berg et al , 18 who have 49% 2-level procedures, and Hellum et al , 16 who included an unclear amount of 1 and 2 affected levels. Zigler et al 21 separately reported 1 and 2 level procedures. Five studies 17 , 18 , 20 , 21 , 23 mentioned facet joint arthritis/degeneration as an exclusion criterion.

Risk of Bias in Included Studies Risk of bias of the included studies is shown in Table 4 . Except for outcome assessor blinding in 1 study, 16 blinding of any party is probably not used in any of the randomized controlled trials, although description of blinding was insuffi -cient. In general, the large FDA-IDE trials have acceptable loss to follow-up between 10% and 12% at 2 years. Four studies were sponsored solely by commercial parties, being the man-ufacturers of the total disc replacement devices. 17 , 18 , 20 , 23 For

one study, 21 funding and confl ict of interest statements were inconsistent across publications. One study was funded by noncommercial parties. 16 One study did not disclose informa-tion about funding or confl ict of interest. 19

Effects of Interventions If available, the analyses were performed on improvements of outcomes as these account for possible baseline differ-ences between and within the studies. We evaluated the pri-mary outcomes at the longest follow-up term available, in this case 24 months. The only report with 5-year follow-up 20 was excluded from analysis because of extensive loss to fol-low-up. Where possible the results were presented separately for 1- and 2-level procedures from the studies of Berg et al 18 and Zigler et al . 21 Pooled analysis and quality of evidence is reported in Table 5 .

Disc Arthroplasty Versus Fusion Five studies 18 – 21 , 23 with 1301 patients were found that com-pared total disc replacement (n = 865) with fusion (n = 436). Only 2 studies provided data that could be used for meta-analysis of back pain. The higher improvement in back pain

Figure 4. Forest plot of improvement patient satisfaction at 24 months comparing disc arthroplasty versus fusion.





TAB

LE 5

. Sum

mar

y of

Fin

ding

s W

ith Q

ualit

y of

Evi

denc

e fo

r R

epor

ted

Out

com

es

Out

com

e (2

4 m

o U

nles

s O

ther

wis

e In

dica

ted)

Stud

ies

Patie

nts

Gra

de

Lim

itatio

ns

Sum

mar

y of

Fin

ding

sQ

uant

itativ

e

Effe

ctQ

ualit

yPo

oled

Eff

ect

Clin

ical

Rel

evan

t D

iffer

ence

Clin

ical

R

esul

t

Dis

c re

plac

emen

t ver

sus

fusi

on: 6

stu

dies

Impr

ovem

ent i

n V

AS

back

pai

n2

676

PB, I

MP

Dis

c re

plac

emen

t bet

ter t

han

fusi

onLo

wM

D 5

.2

(0.2

–10.

3)15

Equi

vale

nt

Impr

ovem

ent i

n V

AS

leg

pain

267

6PB

, IM

P, IN

CN

o st

atis

tical

diff

eren

ceV

ery

low

…15

Uns

ure

Patie

nt s

atis

fact

ion

495

8IN

CD

isc

repl

acem

ent b

ette

r tha

n fu

sion

Mod

erat

eO

R 0

.52

(0.3

6–0.

74)

-U

nsur

e

Impr

ovem

ent i

n O

swes

try d

isab

ility

inde

x sc

ore

512

07IN

CD

isc

repl

acem

ent b

ette

r tha

n fu

sion

Mod

erat

eM

D 4

.3

(1.9

–6.7

)10

Equi

vale

nt

Osw

estry

dis

abili

ty in

dex

scor

e (%

impr

oved

)5

1207

INC

, IM

PD

isc

repl

acem

ent b

ette

r tha

n fu

sion

Low

OR

1.4

5 (1

.06–

1.98

)–

Uns

ure

Wor

king

sta

tus

413

88IN

C, I

MP,

PB

No

stat

istic

al d

iffer

ence

Ver

y lo

w…

…U

nsur

e

Ran

ge o

f mot

ion

481

4IN

CN

o m

eta-

anal

ysis

pos

sibl

e,

but d

ram

atic

effe

ct in

all

stud

ies,

dis

c re

plac

emen

t be

tter t

han

fusi

on

Mod

erat

e…

…Su

perio

r

Blo

od lo

ss (p

erio

pera

tive)

513

01IN

C, I

MP,

PB

Con

fl ict

ing

evid

ence

Ver

y lo

w…

…U

nsur

e

Reo

pera

tions

513

97IN

C, I

MP,

PB

No

stat

istic

al d

iffer

ence

Ver

y lo

w…

…U

nsur

e

Adj

acen

t seg

men

t deg

ener

atio

n1

152

IMP

No

stat

istic

al d

iffer

ence

Ver

y lo

w…

…U

nsur

e

Face

t joi

nt d

egen

erat

ion

115

2IM

PN

o st

atis

tical

diff

eren

ceV

ery

low

……

Uns

ure

Cos

t-effe

ctiv

enes

s1

152

IMP

No

stat

istic

al d

iffer

ence

Ver

y lo

w…

…U

nsur

e

Dis

c re

plac

emen

t ver

sus

cons

erva

tive

trea

tmen

t: 1

stud

y

Impr

ovem

ent i

n ba

ck p

ain

(12

mo)

115

2IM

PD

isc

repl

acem

ent b

ette

r tha

n re

habi

litat

ion

Ver

y lo

wM

D 1

4.0

(5.0

–23.

0)15

Uns

ure

Impr

ovem

ent i

n ba

ck p

ain

115

2IM

PD

isc

repl

acem

ent b

ette

r tha

n re

habi

litat

ion

Ver

y lo

wM

D 1

2.3

(3.1

–21.

3)15

Uns

ure

Patie

nt s

atis

fact

ion

115

2…

Dis

c re

plac

emen

t bet

ter t

han

reha

bilit

atio

nLo

wO

R 2

.65

(1.4

2–4.

96)

…U

nsur

e

Impr

ovem

ent i

n O

swes

try d

isab

ility

inde

x sc

ore

(12

mo)

115

2…

Dis

c re

plac

emen

t bet

ter t

han

reha

bilit

atio

nLo

wM

D 1

0.0

(5.0

–15.

0)10

Uns

ure

Impr

ovem

ent i

n O

swes

try d

isab

ility

inde

x sc

ore

115

2…

Dis

c re

plac

emen

t bet

ter t

han

reha

bilit

atio

nLo

wM

D 8

.0

(3.6

–13.

2)10

Uns

ure

Reo

pera

tions

115

2…

No

stat

istic

al d

iffer

ence

Low

……

Uns

ure

Wor

king

sta

tus

115

2…

No

stat

istic

al d

iffer

ence

Low

……

Uns

ure

VAS

indi

cate

s v i

sual

ana

logu

e sc

ale;

IMP,

impr

ecis

ion;

PB,

pub

licat

ion

or r

epor

ting

bias

; IN

C, i

ncon

sist

ency

; M

D, m

ean

diffe

renc

e; O

R, o

dds

ratio

.





for total disc replacement (2 studies, 5.2 of 100 mm; 95% CI, 0.2–10.3; see Figure 2 ) did not exceed the predefi ned clinically relevant difference (15 mm) with a low quality of evidence. For improvement on the Oswestry disability index score there was moderate quality of evidence from 5 studies (4.3 points; 95% CI, 1.9–6.7; see Figure 3 ) in favor of disc replacement, which also did not exceed the predefi ned clinically relevant difference of 10 points. There is also moderate quality evi-dence from 4 studies that patient satisfaction at 24 months is more prevalent in the total disc replacement group with an odds ratio of 1.93 (958 patients; 95% CI, 1.36–2.76; see Figure 4 ) than in the fusion group. For the other outcome parameters see Table 5 .

Disc Arthroplasty Versus Rehabilitation One low risk of bias study 16 compared disc replacement (n = 86) with rehabilitation (n = 87). The average effect for improvement in back pain at 24 months (12.3 mm) is below the predefi ned clinical relevant difference of 15 mm although there is a chance (95% CI, 3.1–21.3) that the clinical outcome is superior for total disc replacement. This conclusion could also apply on the improvement in Oswestry disability index score (8.4 points; 95% CI, 3.6–13.2). For other outcome parameters see Table 5 .

DISCUSSION

Effectiveness Most of the included studies show clinically relevant improvement 14 from baseline for both interventions, being total disc replacement on one hand and fusion or rehabilita-tion on the other. However, between the interventions, the differences on the primary outcome parameters, including their 95% CIs, were small and did not exceed clinical rel-evance and/or the quality of evidence was low or very low. Patients reported higher satisfaction with total disc replace-ment and the disc replacement devices seemed to succeed in maintaining more natural motion characteristics. None of the included studies, however, allowed the assessment of the predicted clinical benefi ts such as the prevention of adjacent segment degeneration.

Safety Radiological results are poorly reported across the studies and if reported, defi nitions are not provided. Although it can be argued that clinical studies evaluating the effectiveness of disc arthroplasty should always include the assessment of mobil-ity, motion assessment was only presented in 3 of the stud-ies. 20 , 21 , 23 However, the data presentation pertaining to motion were poor, limiting the possibility for meta-analysis. These studies found that total disc replacement showed range of motion comparable to preoperative range of motion, whereas fusion surgery lead to (nearly) absence of motion as would be expected. Two excluded studies 3 , 24 and 1 secondary publi-cation 21 confi rmed our fi ndings. A comparable mobility with healthy volunteers was found by SariAli et al 3 in patients with

disc arthroplasty at 1 level, but with more than 1 level, the mobility was abnormally increased. The study of Zigler 21 also fi nds physiological motion profi les for fl exion-extension for operative and proximal levels. Unfortunately we did not fi nd conclusive evidence of a benefi cial effect of total disc replace-ment versus fusion on adjacent segment degeneration at rele-vant long-term follow-up. Only 1 study assessed adjacent seg-ment degeneration, 18 but could not make a precise estimate of the effect because of a limited sample size and low incidence of adjacent segment degeneration at the short-term follow-up of 2 years. Ultimately because the population receiving total disc replacement is relatively young, adjacent segment degen-eration needs to be assessed in the long term. Interestingly, Putzier et al 22 found 17% adjacent segment degeneration after 17.3 years in 53 patients in a cohort of the CHARITÉ disc replacement. On the development of facet joint degeneration after implantation of a disc replacement, there is insuffi cient evidence. Only 1 study reports facet joint degeneration 19 with an imprecise estimate due to sample size and low incidence at 2-year follow-up. The assessment of facet joint degeneration seems relevant as the fl exion-extension seems to be compa-rable to normal, but especially if the axis of rotation is altered the motion might confl ict with the facet joints.

Perioperative complications are not adequately and con-sistently reported between the studies. Blood loss showed great heterogeneity and thromboembolic complications were only reported in 1 study. From these studies it is not clear how the perioperative complication rate of total disc replace-ment compares to fusion procedures. Gornet et al 23 men-tion 2 removals of disc arthroplasty due to epidural abscess and allergic reaction. Berg et al 18 mention fusion of 4 disc arthroplasty procedures close to the 24 month follow-up. It was diffi cult to link the failures of the device to salvage procedures and outcome. Defi nition of “failure” varies and it was diffi cult to classify failures as “device related.” One secondary study reports on the revisability of the total disc replacement from their trial 20 and concluded that one-third of the patients could be revised to a new total disc replace-ment and that two-thirds of patients could be revised to a fusion. Recent studies on revisability 10 show that revision is possible, but not without (vascular) complications and with highly variable outcome. 9

Overall Completeness and Applicability of Evidence Although clearly presented, the extensive selection criteria in the FDA-IDE trials 17 , 20 , 21 , 23 result in a very select patient sam-ple regarding height loss, age, and presence of any comor-bidity. This limits the external validity of the results of these trials, and precludes extrapolation to the application of total disc replacement in all patients with DDD. There were few comparable control groups between the studies. Because of the superiority of any technique in lumbar fusion for DDD has not yet been established, these results were pooled when statistical heterogeneity was absent. However, future evi-dence about relative effectiveness of these interventions could warrant subgroup analyses.





CONCLUSION

Implications for Practice Disc replacement compared with conventional fusion sur-gery for DDD seems to result in clinically irrelevant supe-riority with respect to pain relief, Oswestry, and Quality of Life in a selected population, so far only for the short term. Currently available trials did not assess adjacent level dis-ease and facet joint degeneration properly, which is a short-coming as this is the reason the product was manufactured. Therefore, because we think that harm and complications may occur years afterwards, the spine surgery community should be prudent about adopting this technology on a large scale, despite the fact that total disc replacement seems to be effective in treating low back pain in selected patients, and in the short term is at least equivalent to fusion surgery. Otherwise, if equivalence can be assumed along the proposed methodology, use of these expensive implants should remain limited including informed consent and mandatory quality assessment in registries similar to the recommendations for metal-on-metal hip prostheses. 25

Implications for Research Despite the publication of several recent studies on the effec-tiveness of total disc replacement, there is a strong need for high-quality studies, with less confl ict of interest. Most impor-tant are long-term follow-up studies because differences in adjacent segment degeneration will be identifi ed only then.

➢ Key Points

We found 7 studies reported in 40 publications com-paring total disc replacement with fusion (6 studies with 1301 patients) or rehabilitation (1 study with 173 patients).

Diff erences in clinical improvement were not beyond generally accepted clinical relevance.

Long-term eff ects such as adjacent segment degen-eration or facet joint degeneration were not properly assessed.

Acknowledgments The authors are grateful for being supported by the Cochrane Back Review Group, in particular Victoria Pennick and Teresa Marin for support and feedback, Rachel Couban for search strategies and results, and for her efforts with FDA requests.

The authors thank the principal investigators of the included studies, Dr. J. Zigler of the Texas Back Institute (Plano, TX); Dr. S. Berg of the Stockholm Spine Center (Stockholm, Sweden); Dr. C. Hellum of the Department of Orthopedics of the Oslo University Hospital and Oslo University (Oslo, Norway); and Dr. M. Gornet of The Orthopedic Center of St. Louis (St. Louis, MO) for providing additional information about unclear risk of bias items.

The authors also thank Wilco Jacobs, MSc, for contribu-tion in protocol development and preparation, selection, risk

of bias assessment, RevMan data entry, manuscript prepa-ration, and overall co-ordination; Moyo Kruyt, MD, PhD, for clinical interpretation and draft review; Niels A. van der Gaag, MD, for risk of bias assessment, clinical interpreta-tion, and draft review; Alexander Tuschel, MD, for protocol review and selection; Marinus de Kleuver, MD, PhD, for pro-tocol development, clinical interpretation, and draft review; Cumhur Oner, MD, PhD, for clinical interpretation; Wilco Peul, MD, PhD, for clinical interpretation and draft review; and Ab Verbout, MD, PhD, for clinical interpretation.

References 1. Fernstrom U . Arthroplasty with intercorporal endoproth-

esis in herniated disc and in painful disc . Acta Chir Scand Suppl 1966 ; 357 : 154 – 9 .

2. Leivseth G , Braaten S , Frobin W , et al. Mobility of lumbar segments instrumented with a ProDisc II prosthesis: a two-year follow-up study . Spine (Phila Pa 1976) 2006 ; 31 : 1726 – 33 .

3. SariAli E , Lemaire JP , Pascal-Mousselard H , et al. In vivo study of the kinematics in axial rotation of the lumbar spine after total inter-vertebral disc replacement: long-term results: a 10–14 years follow up evaluation . Eur Spine J 2006 ; 15 : 1501 – 10 .

4. Huang RC , Tropiano P , Marnay T , et al. Range of motion and adja-cent level degeneration after lumbar total disc replacement . Spine J 2006 ; 6 : 242 – 7 .

5. Harrop JS , Youssef JA , Maltenfort M , et al. Lumbar adjacent segment degeneration and disease after arthrodesis and total disc arthroplasty . Spine (Phila Pa 1976) 2008 ; 33 : 1701 – 7 .

6. Park CK , Ryu KS , Jee WH . Degenerative changes of discs and facet joints in lumbar total disc replacement using ProDisc II: minimum two-year follow-up . Spine (Phila Pa 1976) 2008 ; 33 : 1755 – 61 .

7. Rousseau MA , Bradford DS , Bertagnoli R , et al. Disc arthroplasty design infl uences intervertebral kinematics and facet forces . Spine J 2006 ; 6 : 258 – 66 .

8. Chung SK , Kim YE , Wang KC . Biomechanical effect of constraint in lumbar total disc replacement: a study with fi nite element analy-sis . Spine (Phila Pa 1976) 2009 ; 34 : 1281 – 6 .

9. Punt IM , Visser VM , van Rhijn LW , et al. Complications and reop-erations of the SB Charite lumbar disc prosthesis: experience in 75 patients . Eur Spine J 2008 ; 17 : 36 – 43 .

10. de Maat GH , Punt IM , van Rhijn LW , et al. Removal of the Charite lumbar artifi cial disc prosthesis: surgical technique . J Spinal Disord Tech 2009 ; 22 : 334 – 9 .

11. Cunningham BW , Hu N , Beatson HJ , et al. Revision strategies for single- and two-level total disc arthroplasty procedures: a biome-chanical perspective . Spine J 2009 .

12. Patel AA , Brodke DS , Pimenta L , et al. Revision strategies in lumbar total disc arthroplasty . Spine (Phila Pa 1976) 2008 ; 33 : 1276 – 83 .

13. Furlan AD , Pennick V , Bombardier C , et al. 2009 updated method guidelines for systematic reviews in the Cochrane Back Review Group . Spine (Phila Pa 1976) 2009 ; 34 : 1929 – 41 .

14. Ostelo RW , Deyo RA , Stratford P , et al. Interpreting change scores for pain and functional status in low back pain: towards interna-tional consensus regarding minimal important change . Spine (Phila Pa 1976) 2008 ; 33 : 90 – 4 .

15. GRADE Working G . Grading quality of evidence and strength of recommendations . Br Med J 2004 ; 328 : 1490 – 4 .

16. Hellum C , Johnsen LG , Storheim K , et al. Surgery with disc pros-thesis versus rehabilitation in patients with low back pain and degenerative disc: two year follow-up of randomised study . Br Med J 2011 ; 342 : d2786 .

17. Sasso RC , Foulk DM , Hahn M . Prospective, randomized trial of metal-on-metal artifi cial lumbar disc replacement: initial results for treatment of discogenic pain . Spine (Phila Pa 1976) 2008 ; 33 : 123 – 31 .

18. Berg S , Tullberg T , Branth B , et al. Total disc replacement compared to lumbar fusion: a randomised controlled trial with 2-year follow-up . Eur Spine J 2009 ; 18 : 1512 – 9 .





22. Putzier M , Funk JF , Schneider SV , et al. Charite total disc replacement–clinical and radiographical results after an average follow-up of 17 years . Eur Spine J 2006 ; 15 : 183 – 95 .

23. Gornet MF , Burkus JK , Dryer RF , et al. Lumbar disc arthro-plasty with maverick disc versus stand-alone interbody fusion: a prospective, randomized, controlled, multicenter investiga-tional device exemption trial . Spine (Phila Pa 1976) 2011 ; 36 : E1600 – 11 .

24. Tournier C , Aunoble S , Le Huec JC , et al. Total disc replacement: consequences for sagittal balance and lumbar spine movement . European Spine Journal 2007 ; 16 : 411 – 21 .

25. Sedrakyan A , Normand SL , Dabic S , et al. Comparative assess-ment of implantable hip devices with different bearing surfaces: systematic appraisal of evidence . British Journal of Medicine 2011 ; 343 : d7434 .

19. Moreno P , Boulot J . Comparative study of short-term results between total artifi cial disc prosthesis and anterior lumbar inter-body fusion. in French . Revue de Chirurgie Orthopedique et Repa-ratrice de l Appareil Moteur 2008 ; 94 : 282 – 8 .

20. Blumenthal S , McAfee PC , Guyer RD , et al. A prospective, random-ized, multicenter Food and Drug Administration Investigational Device Exemptions study of lumbar total disc replacement with the CHARITE artifi cial disc versus lumbar fusion - Part I: Evaluation of clinical outcomes . Spine (Phila Pa 1976) 2005 ; 30 : 1565 – 75 .

21. Zigler J , Delamarter R , Spivak JM , et al. Results of the prospec-tive, randomized, multicenter Food and Drug Administration investigational device exemption study of the ProDisc-L total disc replacement versus circumferential fusion for the treatment of 1-level degenerative disc disease . Spine (Phila Pa 1976) 2007 ; 32 : 1155 – 62 .



A Cochrane Review

Documents

Transcript of A Cochrane Review