guideline update for the performance of fusion procedures

J Neurosurg Spine 21:106–132, 2014

106

©AANS, 2014

J Neurosurg: Spine / Volume 21 / July 2014

RecommendationsThere is no evidence that conflicts with the previous

recommendations published in the original version of the “Guidelines for the performance of fusion procedures for degenerative disease of the lumbar spine” regarding the use of hydroxyapatite (HA), various calcium-based prep-arations, and recombinant human bone morphogenetic protein–2 (rhBMP-2) as bone graft extenders and substi-tutes for lumbar fusion.48

No prior recommendations regarding the use of rhBMP-7 for lumbar fusions were published in the origi-nal Lumbar Fusion Guidelines.

Guideline update for the performance of fusion procedures for degenerative disease of the lumbar spine. Part 16: Bone graft extenders and substitutes as an adjunct for lumbar fusion

Michael G. Kaiser, M.D.,1 Michael W. Groff, M.D.,2 WilliaM c. Watters iii, M.D.,3 Zoher GhoGaWala, M.D.,4 Praveen v. MuMManeni, M.D.,5 anDreW t. Dailey, M.D.,6 tanvir f. chouDhri, M.D.,7 Jason c. ecK, D.o., M.s.,8 aloK sharan, M.D.,9 Jeffrey c. WanG, M.D.,10 sanJay s. Dhall, M.D.,5 anD Daniel K. resnicK, M.D.11

1Department of Neurosurgery, Columbia University, New York, New York; 2Department of Neurosurgery, Brigham and Women’s Hospital, Boston, Massachusetts; 3Bone and Joint Clinic of Houston, Houston, Texas; 4Alan and Jacqueline Stuart Spine Research Center, Department of Neurosurgery, Lahey Clinic, Burlington, and Tufts University School of Medicine, Boston, Massachusetts; 5Department of Neurological Surgery, University of California, San Francisco, California; 6Department of Neurosurgery, University of Utah, Salt Lake City, Utah; 7Department of Neurosurgery, Icahn School of Medicine at Mount Sinai, New York, New York; 8Center for Sports Medicine and Orthopaedics, Chattanooga, Tennessee; 9Department of Orthopaedic Surgery, Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, New York; 10Department of Orthopaedic Surgery, Keck School of Medicine, University of Southern California, Los Angeles, California; and 11Department of Neurosurgery, University of Wisconsin, Madison, Wisconsin

In an attempt to enhance the potential to achieve a solid arthrodesis and avoid the morbidity of harvesting au-tologous iliac crest bone (AICB) for a lumbar fusion, numerous alternatives have been investigated. The use of these fusion adjuncts has become routine despite a lack of convincing evidence demonstrating a benefit to justify added costs or potential harm. Potential alternatives to AICB include locally harvested autograft, calcium-phosphate salts, demineralized bone matrix (DBM), and the family of bone morphogenetic proteins (BMPs). In particular, no option has created greater controversy than the BMPs. A significant increase in the number of publications, particularly with respect to the BMPs, has taken place since the release of the original guidelines. Both DBM and the calcium-phosphate salts have demonstrated efficacy as a graft extender or as a substitute for AICB when combined with local autograft. The use of recombinant human BMP-2 (rhBMP-2) as a substitute for AICB, when performing an interbody lumbar fusion, is considered an option since similar outcomes have been observed; however, the potential for het-erotopic bone formation is a concern. The use of rhBMP-2, when combined with calcium phosphates, as a substitute for AICB, or as an extender, when used with local autograft or AICB, is also considered an option as similar fusion rates and clinical outcomes have been observed. Surgeons electing to use BMPs should be aware of a growing body of literature demonstrating unique complications associated with the use of BMPs.(http://thejns.org/doi/abs/10.3171/2014.4.SPINE14325)

Key WorDs • lumbar spine • bone graft • bone substitute • fusion • bone morphogenetic protein • practice guidelines

Abbreviations used in this paper: ACS = absorbable collagen sponge; AICB = autologous iliac crest bone; ALIF = anterior lumbar interbody fusion; b-TCP = b-tricalcium phosphate; BMA = bone marrow aspirate; CHA = coralline hydroxyapatite; CRM = compression-resistant matrix; DBM = demineralized bone matrix; FRA = femoral ring allograft; HA = hydroxyapatite; ICBG = iliac crest bone graft; IDE = investigational device exemption; mJOA = modified Japanese Orthopaedic Association; NRS = numeric rating scale; ODI = Oswestry Disability Index; OP-1 = osteogenic pro-tein–1; PLIF = posterior lumbar interbody fusion; RCT = random-ized controlled trial; rhBMP = recombinant human bone morphoge-netic protein; SF-36 = 36-Item Short Form Health Survey; TLIF = transforaminal lumbar interbody fusion; VAS = visual analog scale.

Unauthenticated | Downloaded 02/13/22 06:58 AM UTC

Part 16: Bone graft extenders and substitutes

107J Neurosurg: Spine / Volume 21 / July 2014

Demineralized Bone MatrixGrade C (Single Level III and Single Level V Studies)

The use of demineralized bond matrix (DBM) as a bone graft extender is an option for 1- and 2-level instru-mented posterolateral fusions.

Hydroxyapatite/Calcium ExtendersGrade C (Single Level II Study)

The use of b-tricalcium phosphate (b-TCP)/local autograft as a substitute for autologous iliac crest bone (AICB) is an option for single-level instrumented postero-lateral fusion due to comparable fusion rates and clinical outcomes.

Grade C (Single Level II Study)The use of HA with local autograft/bone marrow as-

pirate (BMA) as a substitute for AICB in an option for instrumented posterolateral fusion due to comparable fu-sion rates and clinical outcomes.

Grade C (Multiple Level V Studies)The use of HA can be considered an option as a graft

extender when mixed with AICB for instrumented pos-terolateral fusions.

Grade C (Single Level IV and Multiple Level V Studies)The use of calcium sulfate preparations mixed with

local autograft, as a substitute for AICB, is an option for instrumented posterolateral fusions.

Grade I (Single Level V Study)There is insufficient evidence to recommend for or

against the use of a HA-glass/BMA composite as an au-tograft substitute for posterolateral fusion.

rhBMP-2: Interbody FusionGrade B (Multiple Level II Studies)

The use of rhBMP-2 as a substitute for AICB for ALIF with threaded interbody cages is an option due to similar fusion rates and clinical outcomes.

Grade C (Single Level II Study)The use of rhBMP-2 as a substitute for AICB for

single-level PLIF is an option due to similar fusion rates and clinical outcomes; however, formation of heterotopic bone has been observed.

Grade C (Single Level IV and Multiple Level V Studies)The use of rhBMP-2 as a bone graft extender can be

considered as an option when performing a TLIF proce-dure with a structural interbody graft.

Grade I (Single Level III Study)There is insufficient evidence to make a recommen-

dation regarding the use of rhBMP-2 as a supplement for

stand-alone ALIF procedures using femoral ring allograft (single Level III study) or with a resorbable spacer when performing TLIF procedures (single Level V study).

rhBMP-2: Posterolateral FusionGrade B (Multiple Level II Studies)

The use of rhBMP-2 supplemented with 15% HA/85% b-TCP matrix as a substitute for AICB is an option in single-level posterolateral instrumented fusions given the consistent observation of comparable fusion rate and clinical outcomes.

Grade C (Single Level II and Single Level IV Studies)The use of rhBMP-2 supplemented with graft extend-

ers as an alternative to AICB is an option for single-level, instrumented posterolateral fusions in patients older than 60 years.

Grade C (Single Level III and Single Level V Studies)The use of rhBMP-2 as a graft extender with either

AICB or local bone is an option in patients undergoing either instrumented or noninstrumented posterolateral fu-sions.

Grade IThere is insufficient evidence to formulate a recom-

mendation regarding the use of rhBMP-2/local bone as a substitute for AICB when performing revision postero-lateral fusions (single Level III study) or the use of rh-BMP-2/calcium-based extenders for single level postero-lateral fusions in patients who smoke and elect to undergo surgery for lumbar spondylosis (single Level III study).

rhBMP-2: ComplicationsGrade C (Multiple Level IV and V Studies)

The use of rhBMP-2 as a graft option has been as-sociated with a unique constellation of complications that the surgeon should be aware of when considering the use of this graft extender/substitute.

rhBMP-7Grade C (Single Level II Study)

The use of rhBMP-7 when combined with local auto-graft as an alternative to AICB/local autograft is an option for single-level instrumented fusions based on equivalent clinical and radiographic outcomes. The use of rhBMP-7 has not been approved by the FDA for spinal fusions and currently requires a humanitarian device exemption.

Grade I (Conflicting Level II Studies)No recommendation regarding the use of rhBMP-7/

absorbable collagen sponge (ACS) as a substitute for AICB in posterolateral fusions can be made due to con-flicting evidence from studies of equal strength.


M. G. Kaiser et al.

108 J Neurosurg: Spine / Volume 21 / July 2014

RationaleThe objective of a lumbar fusion is to create an en-

vironment that will allow bone to form a solid osseous bridge across the involved spinal segments. Autologous iliac crest bone has been considered the gold standard because of its ideal graft characteristics, including osteo-conduction, osteoinduction, and osteogenesis.4,28,39 The harvesting of AICB, however, is commonly associated with increased postoperative pain, which may be under-estimated by the treating surgeon.31,55 Additional draw-backs of AICB include limited supply and increased op-erative time and blood loss.

Allograft bone, one of the original substitutes for AICB, may avoid some of these drawbacks; however, when used alone, it is commonly associated with an in-creased pseudarthrosis rate.28 For this reason, and to avoid the morbidity of harvesting AICB, a great deal of time and expense has been dedicated to investigate and promote extenders and/or substitutes of AICB. Potential candidates include locally harvested autograft, calcium-phosphate salts, such as HA or b-TCP, and DBM. Howev-er, no material has received more attention and generated more controversy than the family of BMPs. There are numerous papers that demonstrate the fusion potential of BMPs;2,7,18,19 however, complications associated with their use have been reported.29,47,50 Whether the benefits of BMPs justify the costs remains to be determined. Pos-sibly more alarming than the potential complications and costs have been questions related to bias and conflict of interest associated with the reporting of results from tri-als investigating the potential of BMPs.9 This escalating controversy prompted the editors of The Spine Journal to dedicate the June 2011 issue to concerns regarding the use of BMPs in spinal fusion surgery.

The objective of this update is to build upon the previous recommendations formulated in the original guidelines publication.48 A review of the recent medical literature was conducted to determine the utility of these materials with respect to their clinical efficacy, fusion potential, and complication risk. It is beyond the scope of the current update to comment on cost utility of these materials or the ethics of investigational reporting.

Search CriteriaA computerized search of the National Library

of Medicine MEDLINE database, utilizing the on-line search engine PubMed, was conducted from 2003 through December 2011 utilizing the following search terms (((“Lumbosacral Region”[MeSH] OR “Lumbar Vertebrae”[MeSH]) AND “Spinal Fusion”[MeSH]) OR “lumbar fusion”[All Fields] OR (“lumbar”[title] AND “fusion”[title])) AND (((“Bone Substitutes”[MeSH] OR “Calcium Phosphates”[MeSH]) OR “Hydroxyapatites” [MeSH]) OR “Bone Morphogenetic Proteins”[MeSH]) AND ((“2003”[PDAT]: “3000”[PDAT]) AND “humans” [MeSH Terms] AND English[lang]). The search was limited to the English language and human subjects and yielded a total of 151 papers. The titles and abstracts of these articles were reviewed and those specifically inves-tigating the fusion potential, clinical efficacy, and poten-

tial complications of bone graft substitutes and extenders were selected. Of these papers a secondary review of the bibliographies was conducted to identify any additional relevant papers. A total of 79 articles were selected and reviewed in detail. Studies supporting similar conclusions of equivalent strength were grouped together. Those pro-viding the best evidence from these compilations were included in the evidentiary tables. A detailed description of high-level studies or a representative of lower-level studies of similar conclusions serve as the scientific foun-dation for this update.

Scientific Foundation

Demineralized Bone Matrix

Since the publication by Urist, the osteoinductive properties of demineralized bone matrix (DBM) have been well recognized and extensively studied as both a substitute and extender of autograft bone.57 Cammisa et al. conducted a multicenter, prospective, controlled trial to investigate the potential of DBM as a graft extender for AICB when performing a posterolateral instrumented lumbar fusion (Table 1).8 One hundred twenty patients with a variety of degenerative disorders were enrolled and underwent up to a 3-level lumbar fusion. An inde-pendent, blinded radiologist, utilizing static and dynamic radiographs, performed fusion assessment at 3, 6, 12, 18, and 24 months. The clinical outcome of these patients was not recorded. All patients served as his/her own con-trol receiving AICB within one intertransverse space and an equal volume of DBM/AICB to the contralateral in-tertransverse space. The follow-up rate at 24 months was 68%. A comparable fusion rate was observed on both sides (52% with DBM/AICB and 54% with AICB). Sev-enty-five percent of patients demonstrated fusion on both sides. Based on these observations the authors concluded that DBM could serve as an effective graft extender, de-creasing the amount of autograft required and potentially reducing the risk and severity of donor site morbidity. Due to the utilization of internal controls, one cannot ex-clude a possible interaction between the investigational and control groups that could affect the outcome. This is therefore considered a case series when determining baseline level of evidence. Additional limitations include a heterogeneous patient population with respect to pre-senting diagnosis and the inclusion of a variety of fusions methods, including various interbody techniques. A large percentage of patients were lost to follow-up at 24 months. In the presence of pedicle screw stabilization, assessment of fusion with plain radiographs may be compromised. Secondary to these limitations, the case series was down-graded to Level V evidence in support of DBM as a graft extender for AICB in posterolateral lumbar fusion.

Schizas et al. conducted a pilot study comparing the clinical and radiographic outcome of patients undergoing 1- and 2-level posterolateral instrumented lumbar fusion using a novel DBM as a graft extender for autograft (Table 1).51 Fifty-nine consecutive patients were divided into the 2 treatment groups; 33 received DBM mixed with autograft/BMA and 26 received only autograft. Fusion assessment




TABL

E 1:

Dem

iner

alize

d bo

ne m

atrix

: sum

mar

y of e

viden

ce*

Auth

ors &

Ye

arLe

vel o

f Ev

idenc

eDe

scrip

tion o

f Stu

dyCo

mmen

ts

Schiz

as et

al.,

20

08III

The o

bjecti

ve of

this

pilot

coho

rt stu

dy w

as to

comp

are t

he cl

inica

l & ra

diogr

aphic

perfo

rman

ce of

a nov

el DB

M as

a gr

aft e

xtend

er fo

r auto

graf

t in 1-

& 2-

level

poste

rolat

instr

umen

ted fu

sions

.

59 co

nsec

utive

pts w

ere d

ivide

d into

the 2

trea

tmen

t gro

ups:

33 re

ceive

d DBM

mixe

d w/ a

uto-

gr

aft/B

MA

(trea

tmen

t) &

26 re

ceive

d only

auto

graf

t (co

ntrol)

. Fus

ion as

sess

ment

was b

linde

d

& de

termi

ned w

/ plai

n rad

iogra

phs.

Valid

ated o

utcom

e mea

sure

s wer

e use

d to d

eterm

ine

cli

nical

status

. At 1

2 mos

after

surg

ery,

the f

usion

rate

for t

he in

terve

ntion

al gr

oup w

as 6

9.7%

& th

at fo

r the

contr

ol gr

oup w

as 76

.9%

(p =

0.57

). Th

ere w

as no

diffe

renc

e in t

he cl

inica

l out-

come

btwn

grou

ps. T

he au

thor

s con

clude

d tha

t DBM

is a

safe

& ef

fectiv

e gra

ft ex

tende

r for

1- &

2-lev

el po

stero

lat fu

sions

.

This

study

is lim

ited b

y the

relat

ively

small

stud

y pop

ulatio

n w/ v

aryin

g

pres

entin

g diag

nose

s. In

adeq

uate

base

line d

emog

raph

ic da

ta ar

e

prov

ided.

The a

utho

rs fa

il to s

tand

ardiz

e the

DBM

/auto

graf

t com

-

posit

e. Du

e to t

hese

limita

tions

this

study

prov

ides L

evel

III ev

i-

dence r

egardin

g the efficacy o

f DBM

to ac

t as a

graft exte

nder.

Camm

isa et

al., 2

004

VTh

e authors co

nducted

a mu

lticenter

prospective co

ntrolled trial to de

termine the e

fficacy o

f DBM

as a

graf

t exte

nder

for p

oster

olat in

strum

ented

lumb

ar fu

sions

. 120

pts w

ere e

nroll

ed w

/ var

y-

ing de

gene

rativ

e diag

nose

s & un

derw

ent u

p to 3

-leve

l lumb

ar fu

sions

. Rad

iogra

phic

asse

ss-

me

nt of

fusio

n was

perfo

rmed

at 3,

6, 12

, 18,

& 24

mos

by in

depe

nden

t blin

ded r

adiol

ogist

s

using

stati

c & dy

nami

c rad

iogra

phs.

No cl

inica

l outc

ome m

easu

res w

ere u

sed.

Each

pt se

rved

as hi

s/her

own c

ontro

l & re

ceive

d equ

al vo

ls of

eithe

r AIC

B or

a DB

M/A

ICB

comp

osite

to th

e

inter

trans

vers

e spa

ces.

The f

ollow

-up r

ate at

24 m

os w

as 6

8%. T

he fu

sion r

ate fo

r the

DBM

/

AICB

comp

osite

side

was

52%

& 5

4% on

the c

ontro

l side

, w/ a

75%

agre

emen

t btw

n the

2

sid

es. B

ased

on th

ese o

bser

vatio

ns th

e aut

hors

conc

luded

that

DBM

can s

erve

as an

effec

tive

gr

aft e

xtend

er, de

crea

sing t

he am

ount

of au

togr

aft r

equir

ed &

poten

tially

redu

cing t

he ri

sk &

seve

rity o

f don

or si

te co

mplic

ation

s.

This stu

dy be

nefits

from

a relative

ly large no

. of pts wh

o were follow

ed

fo

r an e

xtend

ed tim

e; ho

weve

r, this

stud

y is c

onsid

ered

a ca

se se

-

ries s

ince c

ontro

l & tr

eatm

ent c

ohor

ts we

re co

nduc

ted w

/in th

e

same

pt &

ther

efore

poss

ible i

ntera

ction

btwn

the i

nves

tigati

onal

&

contr

ol sites ca

nnot be ex

cluded. A sig

nificant los

s to follow

-up r

ate

(3

2%) w

as en

coun

tered

. A va

riety

of dia

gnos

es w

ere i

nclud

ed &

the

su

rgica

l tech

nique

was

not s

tand

ardiz

ed. T

he us

e of p

lain r

adio-

grap

hs m

ay be

an in

adeq

uate

mean

s of f

usion

asse

ssme

nt in

the

pr

esen

ce of

pedic

le sc

rew

instru

ment

ation

. The

auth

ors f

ailed

to

inc

lude c

linica

l outc

ome d

ata.

Due t

o the

se lim

itatio

ns th

e stu

dy

wa

s dow

ngra

ded t

o Lev

el V

in su

ppor

t of D

BM as

a gr

aft e

xtend

er

fo

r AIC

B in

pts un

derg

oing i

nstru

mente

d pos

terola

t fus

ions.

* AI

CB =

auto

logou

s ilia

c bon

e gra

ft; B

MA

= bo

ne m

arro

w as

pirate

; DBM

= de

mine

raliz

ed bo

ne m

atrix;

pt =

patie

nt.


M. G. Kaiser et al.


was performed 12 months after surgery by a blinded in-dependent observer utilizing plain radiographs. Validated outcome measures, including the Oswestry Disability In-dex (ODI) and visual analog scale (VAS) pain score, were used to determine clinical status. The fusion rate for the in-terventional group was 69.7% and the control group 76.9% (p = 0.57). There was no difference in the clinical outcome between the 2 groups. The authors concluded that DBM is a safe and effective graft extender for single- and 2-level pos-terolateral fusions. The relatively small number of patients, varying diagnoses, inadequate baseline demographic data, and short clinical follow-up all limit the conclusions of this study. The authors failed to standardized the volume and type of autograft used, whether AICB or locally harvested autograft. Due to these limitations the study is considered to provide only Level III evidence in support of DBM as a graft extender.

Calcium Phosphate SaltsThis class of graft extenders and substitutes consists of

calcium phosphate salts of varying composition that pro-vide a lattice framework for in growth of new bone. These materials provide an osteoconductive matrix, having little if any osteoinductive or osteogenic properties. Examples include b-TCP, HA, and coral-based materials (Table 2).

β-Tricalcium Phosphate. Dai and Jiang performed a prospective, randomized, controlled trial to determine the efficacy of b-TCP as a bone graft substitute for AICB in single-level posterolateral instrumented fusions for degen-erative spinal stenosis (Table 2).16 Sixty-two patients were randomized to one of 2 cohorts, receiving either b-TCP (n = 32) or AICB (n = 30), both supplemented with local autograft. An independent observer assessed clinical out-come with validated outcomes measures, including mJOA, 36-Item Short Form Health Survey (SF-36), and VAS for donor site pain, at 6 weeks and 3, 6, 12, 24 and 36 months after surgery. Two independent observers evaluated plain radiographs to assess fusion status at 3, 6, 12, 24, and 36 months after surgery. No patients were lost to follow-up at 36 months. The reported fusion rate for all study partici-pants at 36 months was reported to be 100%, and signifi-cant improvement in the clinical outcome was observed in all patients. No significant differences were observed be-tween the treatment cohort and control group with respect to fusion rate or clinical outcome. A postoperative hemato-ma was reported in 3 patients undergoing harvest of AICB. All patients from the AICB group reported donor site pain after surgery, and only 20% reported no pain at 6 weeks after the operation. In some patients (data not provided) pain was still present 36 months after surgery. The authors concluded that b-TCP, supplemented with local autograft, could serve as an effective substitute for AICB when local bone is insufficient. This was a well-designed and executed trial of a homogeneous group of patients with excellent clinical follow-up over an extended time. There were sev-eral limitations, however, including a failure to describe the randomization process and failure to use a disease-specific outcome measure. The utilization of plain radiographs to determine fusion status may also be considered subopti-mal, particularly in the presence of an instrumented fusion.

Due to these limitations the study is downgraded to Level II evidence in support of utilizing b-TCP/local autograft as a substitute for AICB.

Hydroxyapatite. In 2005, Korovessis et al. conducted a prospective randomized controlled trial (RCT) to deter-mine the fusion potential of coralline hydroxyapatite (CH) in multilevel, instrumented posterolateral lumbar fusions (Table 2).38 Sixty patients were randomized to one of 3 cohorts: bilateral application of AICB, AICB on the left and CH/local bone/BMA on the right, and CH/local bone/BMA bilaterally. Validated outcome measures, including the ODI, VAS, Roland-Morris score, and SF-36, were ob-tained preoperatively and at 6, 12, 24, and 48 months after surgery. Two blinded, independent radiologists evaluated plain radiographs at 3, 6, 12, 24, and 48 months after sur-gery, supplemented with CT imaging at 12 and 24 months, to assess fusion status. Ninety-five percent of patients were available for follow-up at a minimum of 3 years. The fu-sion rate was 100% for all 3 groups at 1 year after sur-gery, based on CT and plain radiographs; however, in the CH/local bone/BMA cohort the fusion was limited to the facet joint and lamina. Reliability of radiographic assess-ment was adequate with an intraobserver and interobserver correlation coefficient (r) of 0.71 and 0.69, respectively. Improvement in all clinical outcome parameters was ob-served; however, no statistical analysis was performed to determine if any intergroup differences existed. The au-thors concluded that CH when combined with local bone/BMA is an appropriate AICB substitute when placed over the lamina and facet; however, this is inappropriate for in-tertransverse fusion. This was a comprehensive study of adequate design; however, the authors fail to provide ad-equate baseline demographic data to determine if pretreat-ment differences existed in the study groups. The authors avoided any limitations of utilizing an internal control by creating 2 additional study cohorts, those receiving only AICB and those receiving CH/local bone/BMA. Differ-ences in clinical outcome are difficult to determine since no statistical analysis was performed. Due to these limita-tions the study is considered to provide Level II evidence in support of the authors’ conclusions.

The objective of the single-center prospective cohort study conducted by Lee et al. was to determine the efficacy of a HA as a graft extender in instrumented posterolateral fusion (Table 2).41 Thirty-three patients with varying diag-noses underwent either 1- or 2-level circumferential fusion with an HA/AICB mixture or AICB randomly applied to either the right or left intertransverse space. Equal volumes of graft material were used with the control arm receiving twice as much AICB as the investigational side, 6 versus 3 ml, respectively. Radiographs, obtained at 3, 6, and 12 months after surgery, and 3D thin-cut CT scans, obtained at 12 months, were independently reviewed to determine fusion status. Clinical outcome was not objectively re-corded. The fusion status at 12 months was 86.7% in the investigational group and 88.9% in the control group. The volume of fusion was measured with CT and considered significantly greater on the investigational intertransverse space. The authors concluded that HA is a safe and ef-fective graft extender for posterolateral lumbar fusion. A




TABL

E 2:

Cal

cium

pho

spha

te sa

lts: s

umm

ary o

f evid

ence

*

Auth

ors &

Ye

arLe

vel o

f Ev

idenc

eDe

scrip

tion o

f Stu

dyCo

mmen

ts

Dai &

Jian

g,

20

08II

The o

bjective of this sin

gle-center

prospective RCT

was to de

termine the e

fficacy o

f β-TCP

as a

bone

graf

t sub

stitut

e for

AIC

B in

single

-leve

l pos

terola

t instr

umen

ted fu

sions

for d

egen

erati

ve

spina

l sten

osis. 62

pts w

ere r

andomized to o

ne of 2 cohorts

, receiv

ing either β-TCP

(n = 32

) or

AI

CB (n

= 3

0), b

oth su

pplem

ented

w/ lo

cal a

utogr

aft. C

linica

l dat

a wer

e coll

ected

by an

inde

-

pend

ent o

bser

ver a

t 6 w

ks &

3, 6,

12, 2

4, &

36 m

os. V

alida

ted ou

tcome

s mea

sure

s wer

e utili

zed

(m

JOA,

SF-

36, &

VAS

for d

onor

site

pain)

. Fus

ion st

atus w

as as

sess

ed by

2 ind

epen

dent

obse

rver

s eva

luatin

g plai

n rad

iogra

phs a

t 3, 6

, 12,

24, &

36 m

os. T

he 3

6-mo

follo

w-up

rate

was

10

0%. A

ll stu

dy pt

s wer

e con

sider

ed to

have

achie

ved a

solid

arth

rode

sis &

demo

nstra

te

sig

nificant improvem

ent com

pared w

/ their pr

eoperative c

linica

l status

. There was no

difference

in

clinic

al ou

tcome

btwn

the 2

coho

rts. 3

comp

licati

ons (

posto

p hem

atoma

s) we

re as

socia

ted

w/

AIC

B ha

rves

ting,

& do

nor s

ite pa

in wa

s still

reco

rded

in so

me pt

s at th

e 36-

mo fo

llow-

up. T

he

au

thor

s con

clude

d tha

t b-TCP

can b

e used a

s an e

ffective su

bstitu

te for A

ICB wh

en local bone

is ins

ufficien

t.

This

is a r

elativ

ely ho

moge

neou

s, sm

all gr

oup o

f pts

w/ ad

equa

te

inf

orma

tion r

egar

ding b

aseli

ne de

mogr

aphic

s. Th

e foll

ow-u

p

inter

val is

adeq

uate

w/ an

exce

llent

rate

of fo

llow-

up. H

owev

er,

th

e ran

domi

zatio

n pro

cess

was

not d

escr

ibed.

Alth

ough

valid

ated

ou

tcome

mea

sure

s wer

e use

d, th

e aut

hors

faile

d to u

tilize

a

dis

ease-specifi

c instru

ment to me

asure r

esponse to s

urgery. The

me

thod

of fu

sion a

sses

smen

t, par

ticula

rly in

the p

rese

nce o

f an

ins

trume

nted s

pine,

is su

bopti

mal. I

t is no

t clea

r if th

e ass

esso

rs of

clinic

al &

radio

grap

hic da

ta w

ere b

linde

d to t

reatm

ent. D

ue to

thes

e lim

itatio

ns, th

e stu

dy is

down

grad

ed to

Leve

l II ev

idenc

e in

support of β-TCP

/local bone a

s a su

bstitu

te for A

ICB/loc

al bone in

1-

level

instru

mente

d pos

terola

t fus

ions.

Koro

vess

is et

al., 2

005

IITh

is pu

rpos

e of th

is pr

ospe

ctive

RCT

was

to de

termi

ne th

e fus

ion po

tentia

l of C

HA in

mult

ileve

l in-

str

umen

ted po

stero

lat fu

sions

. 60 p

ts we

re ra

ndom

ized i

nto on

e of 3

grou

ps: r

eceiv

ing A

ICB

bilate

rally

, rec

eiving

AIC

B on

the l

eft &

CHA

/loca

l bon

e/BM

A on

the r

ight, &

CHA

/loca

l bon

e/

BMA

bilate

rally

. Vali

dated

outco

mes i

nstru

ments

(ODI

, VAS

, Rola

nd-M

orris

, & S

F-36

) wer

e

obta

ined p

reop

erati

vely

& at

6, 12

, 24,

& 48

mos

after

surg

ery.

Fusio

n ass

essm

ent w

as pe

r-

form

ed w

/ plai

n rad

iogra

phs a

t 3, 6

, 12,

24, &

48 m

os po

stope

rativ

ely, s

upple

mente

d w/ C

T

im

aging

at 12

& 24

mos

. Imag

es w

ere e

valua

ted by

2 ind

epen

dent

radio

logist

s blin

ded t

o the

inter

venti

on. T

he fo

llow-

up ra

te wa

s 95%

at a

minim

um of

3 yr

s. Re

liabil

ity of

radio

grap

hic as

-

sess

ment

of pla

in ra

diogr

aphs

was

perfo

rmed

by re

peati

ng th

e eva

luatio

n afte

r 3 w

ks. T

he

fu

sion r

ate w

as 10

0% fo

r all 3

grou

ps at

1 yr

after

surg

ery,

base

d on C

T &

plain

radio

grap

hs;

ho

weve

r, the

fusio

n obs

erve

d w/ a

pplic

ation

of C

HA/lo

cal b

one/

BMA

was l

imite

d to t

he fa

cet

joi

nt &

lamina

w/o

bridg

ing bo

ne in

the i

ntertr

ansv

erse

spac

e. Th

e intr

a- &

inter

obse

rver

coef-

ficien

t valu

es (r) for fusio

n assessm

ent w

ere 0

.71 & 0.69, respectively. Im

provem

ent in

all clinica

l

outco

me pa

rame

ters w

as ob

serv

ed; h

owev

er, no

stati

stica

l ana

lysis

was p

erfo

rmed

to de

termi

ne

if a

ny in

tergr

oup d

iffere

nces

in cl

inica

l outc

ome o

ccur

red.

The a

utho

rs co

nclud

ed th

at CH

A

wh

en co

mbine

d w/ lo

cal b

one/

BMA

is an

appr

opria

te AI

CB su

bstitu

te wh

en pl

aced

over

the

lam

ina &

face

t but

is ina

ppro

priat

e for

inter

trans

vers

e fus

ion.

This

is a r

elativ

ely sm

all, h

etero

gene

ous p

opula

tion o

f pts

w/ re

spec

t

to pr

eop d

iagno

sis &

no. o

f leve

ls fu

sed.

The p

roce

ss of

rand

om-

iza

tion w

as no

t ade

quate

ly de

scrib

ed. A

lthou

gh th

e aut

hors

state

that

the t

reatm

ent g

roup

s wer

e sim

ilar, a

dequ

ate pr

eop d

emo-

grap

hic da

ta ar

e not

prov

ided,

& th

ey fa

iled t

o dem

onstr

ate

eq

uipois

e btw

n the

trea

tmen

t gro

ups.

The u

se of

an in

terna

l con

-

trol is

subo

ptima

l, sinc

e one

cann

ot ru

le ou

t inter

actio

n btw

n the

2

int

ertra

nsve

rse s

pace

s; ho

weve

r, the

auth

ors c

ontro

lled f

or th

is

lim

itatio

n by c

reati

ng an

isola

ted A

ICB

& CH

A/loc

al bo

ne/B

MA

grou

ps. T

he au

thor

s fail

ed to

perfo

rm ad

equa

te sta

tistic

al an

alysis

of th

e clin

ical re

sults

. Due

to th

ese l

imita

tions

the s

tudy

was

down

-

graded to Le

vel II ev

idence s

upporting the e

fficacy o

f CHA

com-

bin

ed w

/ loca

l bon

e/BM

A as

an A

ICB

subs

titute.

Chan

g et a

l.,

2008

IVTh

e aut

hors

perfo

rmed

a re

trosp

ectiv

e rev

iew of

resu

lts fr

om pt

s und

ergo

ing si

ngle-

segm

ent

po

stero

lat lu

mbar

fusio

n usin

g eith

er lo

cal a

utogr

aft e

xpan

ded w

/ calc

ium su

lfate

or ili

ac cr

est

au

togr

aft. 1

15 pt

s wer

e divi

ded i

nto tr

eatm

ent g

roup

s & ev

aluate

d ove

r 1 yr

. Sim

ilar f

usion

rates

were

obse

rved

, 92.

3% fo

r the

trea

tmen

t gro

up (n

= 6

6) &

92.9

% in

the c

ontro

l (n =

49),

as w

ell

as

clini

cal o

utcom

e, as

deter

mine

d by V

AS sc

ore &

ODI

. The

auth

ors c

onclu

ded t

hat lo

cal a

uto-

gr

aft s

upple

mente

d w/ c

alcium

sulfa

te is

a safe

& ef

fectiv

e alte

rnati

ve to

auto

geno

us ili

ac cr

est

au

togr

aft.

The i

nves

tigato

rs fa

iled t

o des

cribe

how

pts w

ere a

lloca

ted in

to tre

at-

me

nt gr

oups

, whe

ther

the e

valua

tion o

f pts

was b

linde

d, pe

r-

form

ed an

inad

equa

te de

scrip

tion o

f the r

esult

s, ina

dequ

ately

desc

ribed

the s

tatis

tical

analy

sis of

the d

ata,

& ma

de st

ateme

nts

w/

in th

e con

clusio

n tha

t can

not b

e acc

ounte

d for

by th

eir ob

serv

a-

tio

ns. T

his st

udy p

rovid

es Le

vel IV

evide

nce i

n sup

port

of loc

al

au

togr

aft w

/ calc

ium su

lfate

as a

subs

titute

for A

ICB

in sin

gle-

se

gmen

t pos

terola

teral

fusio

ns.

(cont

inued

)


M. G. Kaiser et al.


TABL

E 2:

Cal

cium

pho

spha

te sa

lts: s

umm

ary o

f evid

ence

* (co

ntin

ued)

Auth

ors &

Ye

arLe

vel o

f Ev

idenc

eDe

scrip

tion o

f Stu

dyCo

mmen

ts

Lee e

t al.,

2009

VTh

e obje

ctive

of th

is sin

gle-c

enter

pros

pecti

ve co

ntroll

ed co

hort

study

was

to de

termi

ne th

e ef-

ficacy o

f a co

mmercia

lly av

ailable HA

as a graft exte

nder in instrum

ented

poste

rolat fusio

n. 33

pts

w/ v

aryin

g diag

nose

s und

erwe

nt eit

her 1

- or 2

-leve

l circ

umfer

entia

l fusio

n w/ a

n HA/

AICB

mixtu

re or

AIC

B ra

ndom

ly ap

plied

to ei

ther

the r

ight o

r left

poste

rolat

spac

e. Ra

diogr

aphs

, per

-

form

ed at

3, 6,

& 12

mos

after

surg

ery,

& 3D

thin-

cut C

T at

12 m

os, w

ere i

ndep

ende

ntly r

e-

vie

wed t

o dete

rmine

fusio

n sta

tus.

Clini

cal o

utcom

e was

not o

bjecti

vely

reco

rded

. The

fusio

n

statu

s of th

e pos

terola

t spa

ce at

12 m

os w

as 8

6.7%

in th

e inv

estig

ation

al gr

oup &

88.

9% in

the

contr

ol group. Th

e volu

me of fusio

n, me

asured w/ C

T, wa

s consid

ered signific

antly gr

eater

in the

inv

estig

ation

al gr

oup.

The a

utho

rs co

nclud

ed th

at HA

gran

ules a

re a

safe

& ef

fectiv

e gra

ft

ex

tende

r for

poste

rolat

fusio

n.

This

study

is co

nside

red a

case

serie

s sinc

e con

trol &

trea

tmen

t co-

horts

wer

e con

ducte

d w/in

the s

ame p

t, at th

e sam

e lev

el, &

ther

e-

fo

re ca

nnot

be co

nside

red i

ndep

ende

nt va

riable

s. It i

s pos

sible

that

an in

terac

tion o

ccur

red b

twn t

he in

vesti

gatio

nal &

contr

ol sit

es

th

at co

uld ha

ve an

impa

ct on

the o

utcom

e. Th

e outc

omes

of th

e

poste

rolat

fusio

ns m

ay al

so be

alter

ed by

the p

lacem

ent o

f the

interbody gr

aft; therefore, it is diffic

ult to de

termine the true

efficacy o

f HA as a graft exte

nder w/in the p

oster

olat space. T

he

au

thor

s fail

ed to

inclu

de an

objec

tive m

easu

re of

clini

cal o

utcom

e.

Th

e stu

dy po

pulat

ion w

as sm

all &

heter

ogen

eous

w/ r

espe

ct to

pres

entin

g diag

nosis

. This

case

serie

s is t

here

fore

down

grad

ed to

Leve

l V ev

idenc

e in s

uppo

rt of

HA as

a gr

aft e

xtend

er.Ac

hary

a et

al.

, 200

8V

The a

im of this stu

dy was to de

termine the e

fficacy o

f an H

A–bio

active g

lass c

eram

ic comp

osite as

a gra

ft su

bstitu

te fo

r auto

logou

s bon

e in l

umba

r pos

terola

t fus

ions.

24 co

nsec

utive

pts u

nder

go-

ing

poste

rolat

instr

umen

ted lu

mbar

fusio

n wer

e ente

red i

nto th

e stu

dy. T

he le

ft int

ertra

nsve

rse

sp

ace r

eceiv

ed a

stand

ard m

ixtur

e of B

MA/

HA-g

lass c

eram

ic co

mpos

ite &

the r

ight s

ide re

-

ceive

d an e

qual

volum

e of lo

cally

harv

ested

auto

graf

t. The

prim

ary o

utcom

e mea

sure

was

graf

t

cons

olida

tion a

s dem

onstr

ated o

n ante

ropo

sterio

r rad

iogra

phs a

t 12 m

os af

ter su

rger

y, ev

alu-

ate

d by a

n ind

epen

dent

orth

oped

ic su

rgeo

n. At

1 yr

after

surg

ery,

22/24

pts w

ere a

vaila

ble fo

r

evalu

ation. T

he co

ntrol sid

e dem

onstrate

d a fusio

n rate

of 73

%; how

ever, no

definitiv

e fusion

was o

bser

ved o

n the

HA-

glass

/BM

A co

mpos

ite si

de in

any p

t, w/ 7

7% of

pts d

emon

strati

ng

co

mplet

e gra

ft re

sorp

tion.

The s

tudy

was

term

inated

at th

e req

uest

of th

e prin

cipal

inves

tigato

r.

The a

utho

rs co

nclud

ed th

at th

ere w

as no

role

for t

his H

A-gla

ss co

mpos

ite/B

MA

as a

stand

-

alone

graf

t sub

stitut

e in p

oster

olat f

usion

of th

e lum

bar s

pine.

Alth

ough

this

study

attem

pts to

comp

are 2

trea

tmen

t alte

rnati

ves,

a

major

limita

tion i

s the

poten

tial in

terac

tion b

twn t

he co

ntrol

&

int

erve

ntion

al gr

oups

sinc

e both

occu

rred i

n the

same

pt &

ther

e-

fo

re w

ere n

ot ind

epen

dent

of ea

ch ot

her. F

or th

is re

ason

this

study

is co

nside

red a

case

serie

s & no

t a co

mpar

ative

coho

rt stu

dy. T

his

is

a hete

roge

neou

s coh

ort, w

/ res

pect

to pr

eop d

iagno

sis &

no. o

f

levels

fuse

d. A

subo

ptima

l ass

essm

ent o

f fus

ion w

as pe

rform

ed.

Th

e foll

ow-u

p inte

rval

may b

e con

sider

ed br

ief; h

owev

er, 91

% of

enro

lled p

ts we

re av

ailab

le. T

he dr

amati

c diffe

renc

e in f

usion

outco

me is

noted

. Due

to th

ese l

imita

tions

the s

tudy

is do

wngr

ad-

ed

to Le

vel V

evide

nce a

gains

t the u

se of

HA-

glass

/BM

A co

mpos

-

ite as

a fu

sion s

ubsti

tute;

howe

ver, i

t is in

clude

d due

to th

e dra

-

matic

diffe

renc

e in f

usion

rates

.Hs

u et a

l.,

20

05V

The p

urpo

se of

this

pros

pecti

ve co

ntroll

ed co

hort

study

was

to de

termi

ne th

e effe

ctive

ness

of

loc

al ha

rves

ted au

togr

aft &

CHA

as gr

aft e

xtend

ers/s

ubsti

tutes

for A

ICB

w/ in

strum

ented

poste

rolat

fusio

ns. 5

8 pts

were

divid

ed in

to 3 g

roup

s rec

eiving

diffe

rent

graf

t mixt

ure i

n the

righ

t

inter

trans

vers

e spa

ce, A

ICB/

local

bone

(n =

20),

AICB

/CHA

(n =

19),

& CH

A/loc

al bo

ne (n

= 19

).

All p

ts re

ceive

d AIC

B in

the l

eft in

tertra

nsve

rse s

pace

. Fus

ion as

sess

ment

using

radio

grap

hs

wa

s ind

epen

dentl

y per

form

ed by

2 re

viewe

rs at

4, 6,

& 12

mos

after

surg

ery.

If unc

erta

inty e

x-

isted

a CT

was

perfo

rmed

. The

fusio

n rate

s bet

ween

the i

nves

tigati

onal

& co

ntrol

grou

ps at

12

mo

s wer

e 90%

for A

ICB/

local

bone

vs 9

5% co

ntrol,

78.9

% fo

r AIC

B/CH

A vs

84.

2% co

ntrol,

&

57.9% for C

HA/local bone v

s 89.5

% co

ntrol. The on

ly sta

tistically s

ignific

ant diffe

rence in r

e-

po

rted f

usion

rate

was b

twn A

ICB/

local

& CH

A/loc

al. T

he au

thor

s con

clude

d tha

t eith

er lo

cal

bo

ne or

CHA

can s

erve

as a

graf

t exte

nder

w/ A

ICB;

howe

ver, C

HA/lo

cal b

one w

as no

t an a

c-

ce

ptab

le su

bstitu

te fo

r AIC

B w/

instr

umen

ted po

stero

lat fu

sions

.

Ther

e wer

e a re

lative

ly sm

all no

. of p

ts w/

in ea

ch st

udy s

ubgr

oup.

Heter

ogen

eity e

xisted

btwn

stud

y gro

ups w

/ res

pect

to dia

gnos

is

&

surg

ery &

inco

mplet

e dem

ogra

phic

data

are p

rovid

ed. U

se of

the c

ontra

lat in

tertra

nsve

rse s

pace

as a

contr

ol is

subo

ptima

l

since

one c

anno

t rule

out a

n inte

racti

on bt

wn th

e con

trol &

inves

tigati

onal

sides

w/in

the s

ame p

t. The

auth

ors f

ailed

to pe

r-

form

stati

stica

l ana

lysis

btwn

the i

nves

tigati

onal

& co

ntrol

arms

w/

in

each

stud

y sub

grou

p. Du

e to t

hese

limita

tions

the s

tudy

was

down

grad

ed to

Leve

l V ev

idenc

e in s

uppo

rt of

CHA

as a

graf

t

exten

der.

(cont

inued

)




major limitation of the study is the use of the contralat-eral intertransverse space as the control arm in the study subjects. Interaction between treatment and control groups cannot be excluded; therefore, this is considered equivalent to a case series. Additional limitations include the small, heterogeneous population of patients and failure to include an objective assessment of clinical outcome. The study is considered to provide Level V evidence in support of HA acting as a graft extender for AICB.

Chang et al. performed a retrospective comparative study to determine if calcium sulfate mixed with local autograft could serve as an alternative to AICB for sin-gle-segment posterolateral fusions (Table 2).12 One hun-dred fifteen patients were divided between treatment and control groups and were observed for longer than 1 year. Fusion was assessed through static and dynamic radio-graphs as well as reformatted CT images. Outcome was assessed utilizing VAS scores and ODIs. Similar fusion rates (92.3% for the treatment cohort [n = 66] and 92.9% for the control group [n = 49]) and clinical scores were observed between groups. This study benefitted from an accurate assessment of fusion and use of objective out-come measures, but it is limited by the investigators’ failure to describe how patients were allocated to study cohorts, whether the evaluations were blinded, and an in-adequate description of the statistical analysis. Because of these limitations, the study was downgraded to Level IV evidence in support of using a mixture of calcium sul-fate and local autograft as a substitute for AICB.

Acharya et al. conducted a prospective cohort study to determine the efficacy of an HA-bioactive glass ceramic composite as a substitute for autologous bone (Table 2).1 Twenty-four consecutive patients undergoing posterolater-al instrumented lumbar fusion were entered into the study. Each patient served as his or her own control, with the left intertransverse space receiving a standard mixture of BMA and HA-glass ceramic composite and the right side receiving an equal volume of locally harvested autograft. Anteroposterior radiographs obtained 12 months after sur-gery were evaluated by an independent orthopedic surgeon and were used to assess fusion status. The follow-up rate at 1 year after surgery was 91%. A definitive fusion was demonstrated on the control side in 73% of patients; how-ever, there was no clear evidence of a fusion on HA/BMA side for any patient, with 77% of patients demonstrating complete resorption of the HA-bioactive glass/BMA graft. Given the dramatic difference in outcome, the principal in-vestigator terminated the study and the authors concluded that this HA-glass composite/BMA was ineffective as a graft substitute in posterolateral lumbar fusion. Although this study suffers from a relatively small, heterogeneous patient cohort, the outcome assessment was performed in an objective manner with an adequate follow-up rate at the end point of the study. However, the use of an internal control is considered inadequate, as one cannot exclude an interaction between the control and treatment sides. Such an interaction would likely bias in favor of the treatment arm. Despite this inference, this study was considered equivalent to a case series and not a comparative cohort study. Although the observed results demonstrate a sig-nificant difference in fusion potential, the study design and TA

BLE

2: C

alciu

m p

hosp

hate

salts

: sum

mar

y of e

viden

ce* (

cont

inue

d)

Auth

ors &

Ye

arLe

vel o

f Ev

idenc

eDe

scrip

tion o

f Stu

dyCo

mmen

ts

Chen

et al

.,

2005

V

The a

utho

rs pe

rform

ed a

pros

pecti

ve co

ntroll

ed co

hort

study

to de

termi

ne th

e effe

ctive

ness

of

artificia

l calc

ium su

lfate mixed w

/ local auto

graft vs A

ICB for fusion

form

ation in 1- or

2-lev

el

ins

trume

nted p

oster

olat f

usion

s. Bt

wn S

eptem

ber 2

000 &

Sep

tembe

r 200

1, 74

cons

ecut

ive pt

s

were

enro

lled w

/ a m

ean f

ollow

-up o

f 32.

5 mos

. 39 p

ts un

derw

ent a

1-lev

el fu

sion &

35 a

2-lev

el

fu

sion.

Each

pt re

ceive

d AIC

B to

one s

ide &

a mi

xture

of ca

lcium

sulfa

te/lo

cal a

utogr

aft to

the

oth

er po

stero

lat gu

tter. 2

inde

pend

ent b

linde

d orth

oped

ic su

rgeo

ns ev

aluate

d rad

iogra

phs f

or

br

idging

inter

trans

vers

e bon

e at 3

-mo i

nterv

als fo

r the

1st y

ear &

then

on an

annu

al ba

sis. T

he

proto

col fo

r clinica

l evalua

tion w

as no

t well de

fined, but it a

ppears that OD

Is we

re ob

taine

d in

a r

etros

pecti

ve fa

shion

. The

fusio

n rate

for 1

-leve

l pro

cedu

res w

as 87

.2%

on th

e inv

estig

ation

al

sid

e & 8

9.7%

on th

e con

trol s

ide (p

= 1.

0). F

or 2-

level

proc

edur

es th

e fus

ion ra

te wa

s 82.

9% fo

r

the i

nves

tigati

onal

side &

85.7

% fo

r the

contr

ol sid

e (p =

1.0)

. Clin

ical o

utcom

e was

judg

ed as

exce

llent

or go

od in

78.3%

. The

auth

ors c

onclu

ded t

hat c

alcium

sulfa

te/lo

cal b

one i

s as e

ffecti

ve

as

AIC

B in

achie

ving a

solid

fusio

n for

instr

umen

ted po

stero

lat fu

sions

.

This

is a h

etero

gene

ous s

tudy

popu

lation

w/ r

espe

ct to

pres

entin

g di-

ag

nosis

& th

e aut

hors

fail t

o pro

vide a

dequ

ate ba

selin

e dem

o-

gr

aphic

data

. Sinc

e eac

h pt a

cted a

s his/

her o

wn co

ntrol,

it is

impo

ssibl

e to r

ule ou

t a po

ssibl

e inte

racti

on bt

wn th

e inv

esti-

gatio

nal &

contr

ol sid

es. A

pplic

ation

of gr

aft w

as no

t ran

domi

zed;

ther

efore

bias

cann

ot be

exclu

ded w

/ res

pect

to fu

sion b

ed pr

epa-

ratio

n. Th

e meth

od of

fusio

n ass

essm

ent w

as su

bjecti

ve. T

his

stu

dy is

down

grad

ed to

Leve

l V ev

idenc

e in s

uppo

rt of

calci

um

su

lfate

/loca

l auto

graf

t as a

graf

t sub

stitut

e for

AIC

B.

* AICB

= auto

logous iliac b

one graft; β-TC

P = β-tricalcium

phosphate; BMA = bone marrow aspirate

; CHA

= co

rallin

e hydroxyapatite

; HA = hydroxyapatite

; mJO

A = mo

dified Ja

panese Orth

opaedic

As

socia

tion;

ODI =

Osw

estry

Disa

bility

Inde

x; RC

T =

rand

omize

d con

trolle

d tria

l; SF-

36 =

36-

Item

Shor

t For

m He

alth S

urve

y; VA

S =

visua

l ana

log sc

ore.


M. G. Kaiser et al.


limitations necessitate an assignment of Level V evidence that HA-bioactive glass when combined with BMA cannot serve as a substitute for autologous bone.

A number of lower-quality studies and case series, providing Level IV and V evidence, have also been pub-lished investigating the utility of various calcium-based graft extenders21,22,32 or substitutes for AICB.11,13,15,33,45 These materials were generally mixed with either local autograft or BMA (Table 2). Given the nature of the study design, control groups were poorly designed, historical, or absent; therefore, direct comparisons to AICB are dif-ficult and lack appropriate validity. When compared with previous published results, investigators considered the fusion rates acceptable; however, many of these studies are unable to provide data with respect to the actual clini-cal benefit since baseline demographic data are not pro-vided. Due to these limitations, these studies only dem-onstrate the feasibility of utilizing these calcium-based materials as graft extenders or substitutes.

Bone Morphogenetic ProteinsSince the introduction of BMPs by Marshall Urist in

1965,57 the application of these fusion agents, intended to induce bone formation from surrounding tissue, has dra-matically increased. In 2002, the US FDA granted premar-ket approval for the use of InFUSE (rhBMP-2, Medtronic Sofamor Danek) for single-level ALIF procedures from L-4 to S-1 when used in conjunction with the LT-CAGE Lumbar Tapered Fusion Device (Medtronic Sofamor Danek).58 Under a humanitarian device exemption, the FDA subsequently approved osteogenic protein–1 (OP-1; rhBMP-7, Stryker) for revision posterolateral lumbar spine fusion, where harvesting of autograft was not possible or not expected to achieve solid arthrodesis.60 Although the FDA had granted a similar approval for InFUSE/MAS-TERGRAFT (rhBMP-2, Medtronic Sofamor Danek) for revision of symptomatic, posterolateral lumbar pseudar-throsis, at the request of the sponsor, this device was with-drawn in 2010.59 The use of these agents extends well be-yond the FDA-approved applications, with approximately 85% of primary spine procedures utilizing BMP consid-ered off label.46 Although the off-label use of BMP for spine has met with radiographic and clinical success, con-cern has been raised due to reports of rare but significant neurological or structural complications following the use of BMPs, particularly with interbody fusions.42,50 In addi-tion, whether the routine use of BMPs is cost-effective has yet to be demonstrated. This uncertainty requires a careful evaluation of the literature investigating the various appli-cations of the available BMPs.

rhBMP-2: Interbody Fusion. The utilization of rh-BMP-2 as a substitute for AICB with threaded interbody cages for single-level ALIF procedures has been investi-gated in 2 randomized control trials.5,6

The larger of the 2 trials was previously evaluated in the original Lumbar Fusion Guidelines and was desig-nated as a Level I study.48 These higher-level studies were reevaluated for the purposes of this update since different criteria were used to determine levels of evidence and dif-ferent recommendation grades formulated from the evi-

dence. After reviewing the paper by Burkus et al.,6 several limitations were identified including a failure to perform a power calculation to determine sample size, incomplete description of presenting demographic data (specifically no mention of comorbid medical conditions), and failure to perform an appropriate statistical analysis regarding outcomes between study cohorts (Table 3).

Burkus et al. also performed a prospective RCT in 42 patients to determine fusion progression of rhBMP-2 in a threaded titanium cage compared with AICB for single-level ALIF procedures.5 The investigational cohort (n = 22) received rhBMP-2 and the control arm received AICB (n = 20). Fusion status was determined by 2 independent blinded radiologists evaluating both radiographs and CT images at 2 days and 6, 12, and 24 months after surgery. The fusion rate was 100% in the investigational cohort and 95% in the control group. The patients receiving rhBMP-2 demonstrated a greater average increase in bone density as demonstrated by Hounsfield units. The authors concluded that use of rhBMP-2 is associated with a high fusion rate and is a promising method to facilitate fusion in ALIF pro-cedures. Given the dates of recruitment, these patients may have been included in a previous publication presented by the same authors and reviewed in the original Lumbar Fu-sion Guidelines.6 This study focuses solely on the radio-graphic outcome of these patients without any inclusion of clinical data. The number of patients in each cohort is relatively small and varies with respect to presenting diag-nosis. Although this is an RCT, inadequate baseline demo-graphic data are included and the authors failed to deter-mine appropriate sample size prior to initiating the study. This study therefore provides Level II evidence in support of rhBMP-2 as a substitute for AICB for single-level ALIF procedures with threaded interbody cages (Table 3).

Haid et al. conducted a multicenter prospective ran-domized controled study to investigate the clinical and radiographic outcomes of patients undergoing single-level PLIF utilizing either iliac crest bone graft (ICBG) or rhBMP-2/ACS.29 Sixty-seven patients with single-level degenerative disc disease were randomized. Clinical out-come was assessed utilizing validated outcome measures at 6 weeks and at 3, 6, 12, and 24 months after surgery. Radiographs and CT scans were obtained at 6, 12, and 24 months after surgery. The follow-up rate at all time points was at least 89.6%. At 24 months after surgery the investigational group demonstrated a 92.3% fusion rate while only 77.8% were considered fused in the control group; this difference did not prove to be statistically sig-nificant. Significant clinical improvement was observed in both cohorts, with the investigational group demon-strated superior improvement in the back pain score at 24 months. Although considered to be clinically irrelevant, a significantly greater percentage of patients in the in-vestigational group (71% vs 12%), had heterotopic bone formation posterior to the interbody cage. Sixty percent of controls continued to complain of donor site pain at 24 months. Due to concern regarding the significant increase in heterotopic bone formation, the authors terminated the study but concluded that these results were encouraging. Despite the lack of an observed consequence of this ex-cessive bone formation, the authors elected not to con-




TABL

E 3:

rhBM

P-2 i

n in

terb

ody f

usio

n: su

mm

ary o

f evid

ence

*

Auth

ors &

Ye

arLe

vel o

f Ev

idenc

eDe

scrip

tion o

f Stu

dyCo

mmen

ts

Haid

et al.

,

2004

IITh

is mu

lticen

ter pr

ospe

ctive

rand

omize

d non

blind

ed st

udy w

as in

tende

d to e

valua

te the

clini

cal &

radio

-

graphic

outco

mes a

t 24 m

os of pts u

ndergoing

1-lev

el PL

IF us

ing eithe

r ICB

G or rh

BMP-2/A

CS. 67

pts

w/ 1

-leve

l deg

ener

ative

disc

dise

ase w

ere r

ando

mize

d to s

tudy c

ohor

ts. C

linica

l outc

ome w

as as

-

sess

ed us

ing va

lidate

d outc

ome m

easu

res a

t 6 w

ks &

3, 6,

12, &

24 m

os af

ter su

rger

y. Ra

diogr

aphs

&

CT

scan

s wer

e obta

ined a

t 6, 1

2, &

24 m

os af

ter su

rger

y. Th

e foll

ow-u

p rate

at al

l time

point

s was

at

lea

st 89

.6%. T

he fu

sion r

ates a

t 24 m

os w

ere 9

2.3%

& 77

.8% in

the i

nves

tigati

onal

& co

ntrol

grou

ps,

respectively

; the d

ifference w

as no

t statistica

lly signific

ant. A

signific

antly greater pe

rcenta

ge of pts in

the

inve

stiga

tiona

l gro

up (7

1% vs

12%)

had h

etero

topic

bone

form

ation

poste

rior t

o the

inter

body

cage

.

All clinica

l param

eters improved signific

antly in bo

th stu

dy co

horts co

mpared w/ preop statu

s. The

improvem

ent in

back pa

in score w

as signific

antly greater in the treatm

ent cohort vs c

ontrols at 24 mos.

60

% of

contr

ols co

ntinu

ed to

comp

lain o

f don

or si

te pa

in at

24 m

os. T

he au

thors

conc

luded

that

the

results were e

ncouragin

g regardin

g the us

e of rhB

MP-2 through the po

sterior in

terbody ap

proach;

howe

ver, furthe

r stud

ies incorporating more refined s

urgic

al tec

hniqu

e are required.

This

study

is lim

ited b

y the

nonb

linde

d ass

essm

ent o

f clin

ical

ou

tcome

& re

lative

ly sm

all pt

coho

rts. T

he pr

oces

s of r

an-

do

miza

tion i

s not

adeq

uatel

y des

cribe

d. Th

e stu

dy w

as

ter

mina

ted du

e to c

once

rn re

gard

ing he

teroto

pic bo

ne fo

rma-

tion p

oster

ior to

the i

nterb

ody c

ages

. Alth

ough

this

prov

ed to

be clinically insig

nificant, the s

tudy was no

t restarte

d sinc

e

the u

se of stand-alo

ne PLIF

cages h

ad fallen o

ut of fav

or.

Th

is stu

dy di

d pro

vide L

evel

II evid

ence

in su

ppor

t of

rhBM

P-2 a

s a su

bstitu

te for A

ICB in sin

gle-level P

LIF

w/

thre

aded

cage

s.

Burk

us et

al.,

20

03II

The a

utho

rs co

nduc

ted a

pros

pecti

ve ra

ndom

ized c

ontro

lled t

rial to

deter

mine

fusio

n pote

ntial

of

rhBM

P-2 in a

threaded titanium

cage co

mpared w/ A

ICB for 1-level A

LIF pr

ocedures. B

twn A

ugust

19

98 &

Mar

ch 19

99, 4

5 pts

were

rand

omize

d, w/

42 (9

3%) a

vaila

ble fo

r foll

ow-u

p at 2

4 mos

after

surg

ery.

Ther

e wer

e 22 p

ts in

the i

nves

tigati

onal

coho

rt &

20 in

the c

ontro

l arm

. 2 in

depe

nden

t blin

ded

ra

diolog

ists e

valua

ted bo

th ra

diogr

aphs

& C

T im

ages

at 2

days

& 6,

12, &

24 m

os af

ter su

rger

y.

Ba

sed o

n rad

iogra

ph &

CT

crite

ria, th

e fus

ion ra

te wa

s 100

% in

the i

nves

tigati

onal

coho

rt &

95%

in

the c

ontro

l group. T

he pts receiv

ing rh

BMP-2 d

emonstrate

d a gr

eater

average increase in b

one

density as

demo

nstra

ted by

Hounsfield units; how

ever, the r

ate of follow-up w/ respect to these d

ata is

not consis

tent w

/ the o

verall follow

-up r

ate. T

he au

thors c

onclu

ded that use of rh

BMP-2 is a

ssociated

w/

a hig

h fus

ion ra

te &

is a p

romi

sing m

ethod

to fa

cilita

te fu

sion i

n ALIF

proc

edur

es.

This

study

is lim

ited b

y the

relat

ively

small

pt po

pulat

ion th

at

va

ries w

/ res

pect

to pr

esen

ting d

iagno

sis. T

he au

thor

s fail

to

pr

ovide

adeq

uate

base

line d

emog

raph

ic da

ta. T

he de

scrip

-

tion o

f sta

tistic

al an

alysis

is lim

ited &

no po

wer c

alcula

tions

were

perfo

rmed

to de

termi

ne sa

mple

size.

Due t

o the

se

lim

itatio

ns th

is stu

dy w

as do

wngr

aded

to Le

vel II

evide

nce

that rhBM

P-2 c

an se

rve a

s a su

bstitu

te for A

ICB for 1-level

sta

nd-a

lone A

LIF pr

oced

ure w

/ thre

aded

cage

s.

Burk

us et

al.,

20

02II

The o

bjective of this mu

lticenter

ed pr

ospective ra

ndom

ized trial w

as to co

mpare the efficacy o

f rhB

MP-2

vs

AIC

B fo

r 1-le

vel A

LIF pr

oced

ures

using

tape

red c

ylind

rical

cage

s. 27

9 pts

w/ sy

mptom

atic 1

-leve

l

degenerative d

isease w

ere e

nrolled, randomized in a

1:1 fashio

n to e

ither re

ceive

rhBM

P-2 (n =

143)

or

AIC

B (n

= 13

6). R

adiog

raph

ic ou

tcome

was

inde

pend

ently

asse

ssed

using

stati

c & dy

nami

c rad

io-

gr

aphs

& C

T sc

ans a

t 6, 1

2, &

24 m

os. C

linica

l ass

essm

ent w

as co

nduc

ted at

6 wk

s & 3,

6, 12

, & 24

mos &

inclu

ded w

ork s

tatus

, neu

rolog

ical o

utcom

e, pa

in qu

estio

nnair

es, &

ODI

s. At

24 m

os, th

e

follo

w-up

rate

for b

oth gr

oups

was

>90

%. C

linica

l outc

ome w

as co

mpar

able

btwn

grou

ps. T

he

rhBM

P-2 g

roup de

monstra

ted a fusio

n rate

of 94.5

% at 24

mos while the fusio

n rate

was for 8

8.7%

of

the c

ontro

l gro

up. D

onor

site

pain

was r

epor

ted in

32%

of co

ntrols

at 24

mos

. The

rate

of se

cond

-

ary p

rocedures w

as similar btwn

groups. B

ased on

these r

esults the au

thors c

onclu

ded that rhB

MP-2

wh

en us

ed w

/ a ta

pere

d cag

e is a

n effe

ctive

tech

nique

for A

LIF pr

oced

ures

& av

oids t

he ad

vers

e

effec

ts of

harv

estin

g AIC

B.

This is a w

ell-design

ed clinical trial & be

nefits

from

a large sa

mple

siz

e, va

lidate

d outc

omes

instr

umen

ts, st

anda

rdize

d sur

gical

techn

ique,

comp

rehe

nsive

outco

me an

alysis

, & ex

celle

nt

fol

low-u

p rate

. The

re ar

e num

erou

s lim

itatio

ns in

cludin

g a

fai

lure t

o per

form

powe

r calc

ulatio

n to d

eterm

ine sa

mple

size,

incom

plete

desc

riptio

n of p

rese

nting

demo

grap

hics (

no

me

ntion

of co

morb

id me

dical

factor

s), fa

ilure

to de

scrib

e inc

lu-

sio

n/exc

lusion

crite

ria, &

failu

re to

inde

pend

ently

colle

ct cli

nical

data.

The

autho

rs do

not c

omple

tely d

escri

be th

e dist

ributi

on of

comp

licati

ons b

twn t

reatm

ent &

contr

ol gr

oups

. No s

tatist

ics

we

re pe

rform

ed to

comp

are o

utcom

e btw

n tre

atmen

t & co

ntrol

grou

ps. D

ue to

the n

umer

ous m

inor li

mitat

ions t

he st

udy w

as

down

graded to Le

vel II ev

idence in s

upport of rhBM

P-2 a

s a

su

bstitu

te for

AIC

B w/

1-lev

el AL

IF pr

oced

ures

.

(cont

inued

)


M. G. Kaiser et al.


TABL

E 3:

rhBM

P-2 i

n in

terb

ody f

usio

n: su

mm

ary o

f evid

ence

* (co

ntin

ued)

Auth

ors &

Ye

arLe

vel o

f Ev

idenc

eDe

scrip

tion o

f Stu

dyCo

mmen

ts

Slos

ar et

al.,

20

07III

The o

bjective of this prospective co

hort stu

dy was to de

termine if rhBM

P-2 c

an sa

fely accelerate

al-

log

raft

inter

body

fusio

ns &

redu

ce th

e no.

of no

nunio

ns w

hen c

ompa

red w

/ allo

graf

t alon

e. 75

pts w

/

varying

diagnoses &

undergoin

g 1–3 level fu

sions were e

nrolled; 30 r

eceiv

ed an

FRA

w/ allograft

crouton

s (contr

ol) & 45 r

eceiv

ed an

FRA

supplem

ented

w/ rhB

MP-2 (

investigational); both

groups

re

ceive

d pos

terior

pedic

le sta

biliza

tion w

/o fu

sion.

Fusio

n ass

essm

ent, p

erfo

rmed

at 6,

12, &

24 m

os,

wa

s blin

ded &

perfo

rmed

using

radio

grap

hs &

CT

imag

es. O

bjecti

ve ou

tcome

instr

umen

ts we

re

us

ed to

deter

mine

clini

cal o

utcom

e; ho

weve

r, ass

essm

ent w

as no

t blin

ded.

At 24

mos

after

surg

ery,

the c

ontro

l gro

up fo

llow-

up ra

te wa

s 97%

& th

e inv

estig

ation

al gr

oup r

ate w

as 9

6%. S

tatis

ticall

y sig-

nificant inc

rease in fusion

rate wa

s observed a

t all time p

oints for the investigational group (9

4%,

10

0%, &

100%

) com

pare

d w/ c

ontro

ls (6

6%, 8

4%, &

89%

). Th

e inv

estig

ation

al gr

oup d

emon

strate

d

signifi

cantly b

etter

outco

mes a

t 6 mos, but both groups de

monstra

ted signific

ant im

provem

ent at 12

&

24 m

os co

mpar

ed w

/ bas

eline

. The

auth

ors c

onclu

ded t

hat a

llogr

aft in

terbo

dy fu

sion s

upple

mente

d

w/ rh

BMP-2 s

ignific

antly im

proved the fusion

rate comp

ared w/ allograft alon

e & the m

ore r

apid fusio

n

observed w/ rhB

MP-2 led to more r

apid clinic

al improvem

ent.

The s

tudy

popu

lation

inclu

des p

ts w/

vary

ing di

agno

ses &

the

au

thor

s fail

to de

scrib

e the

distr

ibutio

n of th

ese d

iagno

ses

bt

wn th

e 2 tr

eatm

ent g

roup

s. Th

e pote

ntial

impa

ct of

differ

-

ence

s in t

he ba

selin

e dem

ogra

phic

data

& no

. of le

vels

oper

-

ated b

twn t

he co

horts

cann

ot be

deter

mine

d sinc

e the

auth

ors f

ailed

to pe

rform

stati

stica

l ana

lysis

of th

ese d

ata.

Alth

ough

radio

grap

hic da

ta w

ere i

ndep

ende

ntly r

eview

ed, it

is no

t clea

r if th

e clin

ical a

sses

smen

t was

blind

ed. D

ue to

thes

e lim

itatio

ns th

is stu

dy w

as do

wngr

aded

to Le

vel II

I

w/ re

spect to

rhBM

P-2 a

s a su

pplem

ent to

anter

ior inter

body

fu

sion u

sing f

emor

al rin

g allo

graf

t & pe

dicle

scre

w

sta

biliza

tion.

Mum

mane

ni

et

al.,

20

04

IVTh

is retro

spective c

ohort study was inten

ded to c

ompare the e

fficacy o

f rhB

MP-2 w

/ AICB wh

en placed

in

an in

terbo

dy sp

acer

durin

g a T

LIF pr

oced

ure.

44 pt

s und

erwe

nt th

e TLIF

proc

edur

e btw

n Sep

tem-

be

r 200

2 & D

ecem

ber 2

003.

Follo

w-up

data

wer

e ava

ilable

for 4

0 pts

(90%

). 19

pts r

eceiv

ed A

ICB

in

the inte

rbody s

pace, &

21 pts receiv

ed rh

BMP-2 s

upple

mente

d w/ either AICB (n = 12

) or lo

cal auto

-

graf

t (n =

9). R

adiog

raph

ic ev

aluati

on w

as pe

rform

ed w

/ sta

tic &

dyna

mic r

adiog

raph

s at 6

-wk &

3-m

o

intervals. Clinica

l outc

ome w

as as

sessed at 3-m

o inte

rvals

using

the P

rolo scale

& VAS

for d

onor site

morb

idity.

The

mea

n foll

ow-u

p was

9 mo

s. Th

e fus

ion ra

te in

the A

ICB

grou

p was

95%

. A 10

0%

fusio

n rate

was ob

served in pts receiv

ing rh

BMP-2 w

/ at le

ast 6 mos of follow-up; how

ever, on

ly 76%

of

pts fr

om th

is co

hort

were

avail

able.

At 6

mos

after

surg

ery,

58%

of pt

s con

tinue

d to c

ompla

in of

donor site pa

in. The au

thors c

onclu

ded that th

e use of rh

BMP-2 &

local auto

graft is

an ex

cellent op-

tio

n whe

n per

form

ing a

TLIF

proc

edur

e & av

oids A

ICB

dono

r site

mor

bidity

.

This

is a h

etero

gene

ous,

small

coho

rt of

pts th

at va

ried w

/ res

pect

to

diagn

osis

& su

rger

y. It i

s not

clear

how

many

pts u

nder

went

a sup

pleme

ntal p

oster

olat fu

sion.

The n

o. of

pts is

too s

mall t

o

perfo

rm an

y mea

ningfu

l stat

istica

l ana

lysis

to de

termi

ne di

ffer-

en

ces w

/in th

e tre

atmen

t gro

ups.

The r

adiog

raph

ic as

sess

ment

of fus

ion was no

t clea

rly de

fined. C

linica

l outc

ome w

/ respect

to treatm

ent effic

acy w

as no

t conducte

d w/ validated

outco

mes

me

asur

es. T

he fo

llow-

up in

terva

l was

relat

ively

shor

t w/ a

signifi

cant no. of rhB

MP-2 pts n

ot available. The ev

aluation of

bo

th cli

nical

outco

me &

radio

grap

hs w

as no

t blin

ded.

Due t

o

these

limita

tions

the s

tudy i

s dow

ngra

ded t

o Lev

el IV

evide

nce,

supporting the role of rhBM

P-2 a

s a su

pplem

ent to

interbody

fus

ions w

hen p

erfor

ming

a TL

IF.Ge

ibel et al.,

2009

VThis case se

ries d

escribes the clinical & radio

graphic

outco

me of 48

pts u

ndergoing

1- & 2-lev

el PL

IF

& poste

rolat instrum

ented

fusio

ns w/ rhB

MP-2. 37 (77.1%

) underwe

nt a 1

-level procedure & 11

(22.9

%)

un

derw

ent a

2-lev

el pr

oced

ure.

ODIs

were

obtai

ned a

fter s

urge

ry &

CT

imag

es w

ere o

btaine

d to a

sses

s

fusion

. The

aver

age f

ollow

-up w

as 17

mos

. 2 in

depe

nden

t rad

iolog

ists e

valua

ted th

e CT

imag

es &

de-

termined a

fusio

n rate

of 10

0%. P

t satisfa

ction, recorded a

s a willingness to repeat surgery, wa

s

observed in 89

%. The au

thors conclud

ed that rhBM

P-2 c

an se

rve a

s a su

bstitu

te for AICB & effec

tively

& safely a

ccom

plish fusio

n through the P

LIF ap

proach.

This

is a h

etero

gene

ous p

opula

tion o

f pts

w/ re

spec

t to di

agno

-

sis &

surg

ical p

roce

dure

. No p

reop

clini

cal d

ata w

ere

re

cord

ed; th

erefo

re, th

e tre

atmen

t effe

ct of

surg

ery c

anno

t be

de

termi

ned.

Due t

o the

se lim

itatio

ns th

e cas

e ser

ies w

as

down

graded to Le

vel V ev

idence r

egardin

g the efficacy &

safety o

f rhB

MP-2 a

s a su

bstitu

te for A

ICB w/ a PL

IF. (cont

inued

)




TABL

E 3:

rhBM

P-2 i

n in

terb

ody f

usio

n: su

mm

ary o

f evid

ence

* (co

ntin

ued)

Auth

ors &

Ye

arLe

vel o

f Ev

idenc

eDe

scrip

tion o

f Stu

dyCo

mmen

ts

Burk

us et

al.,

20

09V

This

repo

rt re

pres

ents

an up

date

of th

e clin

ical &

radio

grap

hic ou

tcome

s of p

ts fro

m 2 s

epar

ate st

udies

undergoin

g stand-alon

e 1-level A

LIF w/ tapered ca

ges &

rhBM

P-2. No

comp

ariso

n w/ th

e contro

l

arm

(pts

rece

iving

auto

logou

s ilia

c cre

st) w

as pe

rform

ed. A

t 6 yr

s afte

r the

inde

x pro

cedu

re, 1

46 pt

s

prov

ided c

linica

l follo

w-up

, & ra

diogr

aphic

evalu

ation

was

perfo

rmed

on 13

0 pts.

A so

lid ar

thro

desis

was identified in 9

8% of pts. Th

ere w

as no

signific

ant diffe

rence in a

ny of the c

linica

l outc

ome m

ea-

sures a

t 6 yrs c

ompared w

/ the v

alues ob

served at 2 yrs a

fter surgery. Sign

ificant im

provem

ent com

-

pare

d w/ p

reop

scor

es w

as m

ainta

ined a

t 6 yr

s afte

r sur

gery.

25 re

vision

surg

eries

wer

e per

form

ed

ov

er th

e 6-y

r foll

ow-u

p per

iod &

7 bt

wn th

e 2- &

6-y

r tim

e poin

ts. T

he au

thor

s con

clude

d tha

t this

techniqu

e was an

effec

tive m

ethod of ob

taining an

ALIF

& main

taining long-te

rm, sign

ificant clinica

l

impr

ovem

ents.

The e

fficacy o

f rhB

MP-2 a

s a su

bstitu

te for a

utolog

ous b

one w

/

tape

red c

ages

in 1-

level

ALIF

was

prev

iously

esta

blish

ed w

/

the p

rior s

tudie

s. Th

e res

ults r

epor

ted in

this

case

serie

s are

limited b

y the signific

ant no. of pts

lost to follow-up, 48%

drop

out r

ate fo

r clin

ical d

ata,

& 55

% fo

r rad

iogra

phic

evalu

-

ation

& th

erefo

re w

as do

wngr

aded

to Le

vel V

evide

nce.

This

case se

ries d

oes n

ot retra

ct or co

ntribu

te sig

nificantly to the

pr

ior st

udies

.

Rihn

et al

.,

2009

49V

The o

bjecti

ve of

this

retro

spec

tive r

eview

of a

case

serie

s was

to de

termi

ne th

e clin

ical &

radio

grap

hic

outco

mes o

f pts undergoin

g 1-level TLIF

w/ rhB

MP-2. 48 of 53 p

ts w/ va

rying

diagnoses o

f lumb

ar

degenerative d

isease w

ere identified o

ver a 2-yr interval. S

tatic & dy

namic radiog

raphs w

ere r

eview

ed

by

an in

depe

nden

t blin

ded s

pine s

urge

on at

an av

erag

e foll

ow-u

p of 1

9.4 m

os. O

dom’

s crit

eria,

pt

sa

tisfac

tion,

& NR

S of

leg &

back

pain

were

obta

ined t

hrou

gh te

lepho

ne in

tervie

w at

an av

erag

e

follo

w-up

of 27

.4 mo

s. Th

e fus

ion ra

te wa

s 95.

8%. 7

1% ac

hieve

d exc

ellen

t or g

ood o

utcom

e, w/

84%

of pts

satisfied w

/ their ou

tcome

. Persis

tent back &

leg p

ain was re

porte

d in 6

0.4% & 41

.7%, respec-

tively

. An o

vera

ll com

plica

tion r

ate of

21.7%

was

repo

rted,

w/ ap

prox

imate

ly 25

% at

tribu

ted to

the u

se

of rhBM

P-2. Th

e authors co

nclud

ed that adequate clinic

al & radio

graphic

outco

mes c

an be

obtaine

d.

This

is a r

etros

pecti

ve re

view

of a c

ase s

eries

of pt

s und

ergo

-

ing a 1-lev

el TL

IF. The study b

enefits fro

m an independent

as

sess

ment

of fu

sion u

sing a

ppro

priat

e rad

iogra

phic

meth

ods.

Valid

ated o

utcom

e mea

sure

s wer

e inc

orpo

rated

to

de

termi

ne re

spon

se to

trea

tmen

t. The

heter

ogen

eous

pt

population

is co

nside

red a

signific

ant limitation, & therefo

re

th

e stu

dy is

down

grad

ed to

Leve

l V ev

idenc

e in s

uppo

rt of

rhBM

P-2 u

sed a

s an a

djunct to

inter

body fusio

n.

Villa

vicen

cio

et

al., 2

005

VTh

is re

trosp

ectiv

e cas

e ser

ies de

scrib

es th

e clin

ical &

radio

grap

hic ou

tcome

s of 7

4 pts

unde

rgoin

g TLIF

w/ allog

raft supplem

ented

w/ rhB

MP-2/AC

S for a va

riety of degenerative d

isorders through a variety

of surgica

l approaches. Pts w

ere d

ivided into

subgroups b

ased on

the n

o. of lev

els fused &

whether

an

open

or m

inima

lly in

vasiv

e app

roac

h was

used

. A si

ngle

indep

ende

nt ra

diolog

ist ev

aluate

d rad

io-

gr

aphs

at 3,

6, 12

, & 24

mos

& C

T im

ages

at 12

& 24

mos

after

surg

ery.

Inde

pend

ent c

linica

l ass

ess-

ment

was p

erfo

rmed

at 12

mos

after

surg

ery u

tilizin

g Mac

Nab c

riter

ia. 9

6% of

pts w

ere a

vaila

ble at

12

mo

s after surgery. There were n

o adverse ev

ents directly re

lated to the u

se of rh

BMP-2. Th

e fusion

rate

for t

he en

tire c

ohor

t was

100%

by 10

mos

after

surg

ery.

Impr

oved

clini

cal o

utcom

e was

obse

rved

in all

grou

ps; h

owev

er, th

ose u

nder

going

mini

mally

inva

sive p

roce

dure

s ten

ded t

o hav

e a be

tter o

ut-

come

. The au

thors c

onclu

ded that rhB

MP-2 is a

safe & effec

tive b

one g

raft exten

der w

hen u

sed in

co

njunc

tion w

/ allo

graf

t for t

he T

LIF pr

oced

ure.

This stu

dy su

ffers from

signific

ant hete

rogeneity w/ respect

to dia

gnos

is &

surg

ical p

roce

dure

. The

no. o

f pts

in ea

ch

su

bgro

up w

as re

lative

ly sm

all, &

inco

mplet

e des

cript

ion of

demo

grap

hic ch

arac

terist

ics is

prov

ided.

Nonv

alida

ted ou

t-

come

mea

sure

s wer

e use

d to a

sses

s clin

ical o

utcom

e. Th

e

study pr

ovide

s only

Level V ev

idence in s

upport of rhBM

P-2

as

a gr

aft e

xtend

er fo

r allo

graf

t for T

LIF pr

oced

ures

.

Lanm

an &

Hopk

ins,

20

04

VTh

is ca

se se

ries d

escr

ibes t

he cl

inica

l & ra

diogr

aphic

resu

lts of

42 pt

s und

ergo

ing th

e TLIF

proc

edur

e

using

rhBM

P-2 &

a bio

resorbable interbody gr

aft. C

linica

l & ra

diographic

follow-up was co

nducted

at

3,

6, &

12 m

os af

ter su

rger

y usin

g the

ODI

, plai

n rad

iogra

phs,

& CT

imag

es. T

he ra

diogr

aphic

follo

w-

up ra

te at 6 m

os was 98%

& at 12

mos was 26

%. P

reop clinical data w

ere o

btain

ed in on

ly 59% of pts

w/

a fo

llow-

up ra

te of

92%

at 6

mos &

36%

at 12

mos

. The

fusio

n rate

was

98%

at 6

mos &

100%

at

12 mos. C

linica

l improvem

ent w

as ob

served at ea

ch study e

nd po

int, although signific

ance was no

t

deter

mine

d. Th

ere w

ere n

o dev

ice-re

lated

comp

licati

ons.

The a

utho

rs co

nclud

e tha

t thes

e res

ults

ind

icate that the c

ombin

ation of rh

BMP-2 w

/ a re

sorbable spacer may be

an ap

propriate alternative for

int

erbo

dy fu

sion &

dese

rves

furth

er in

vesti

gatio

n.

This

is a r

elativ

ely po

orly

cond

ucted

case

serie

s w/ li

ttle in

for-

ma

tion r

egardin

g the efficacy o

f rhB

MP-2 u

sed w

/ a bio-

reso

rbab

le sp

acer.

The

stud

y pop

ulatio

n is s

mall &

varie

s w/

re

spec

t to di

agno

sis, n

o. of

levels

fuse

d, &

op ap

proa

ch. A

n

exce

ssive

no. o

f pts

were

lost

durin

g the

follo

w-up

perio

d to

ma

ke an

y judgm

ents regarding

clinical effic

acy. It is n

ot cle

ar

if t

he ra

diogr

aphic

or cl

inica

l follo

w-up

was

blind

ed. T

he

au

thor

s fail

ed to

perfo

rm an

y sta

tistic

al an

alysis

. This

stud

y

prov

ides o

nly Le

vel V

evide

nce.

* AC

S = absorbable collagen sponge; A

ICB = autolog

ous iliac c

rest bone; A

LIF = ante

rior lum

bar inte

rbody fusion

; FRA

= femo

ral ring

allograft; ICBG

= iliac crest bone graft; NR

S = nume

ric ra

ting

scale

; ODI = Os

westr

y Disa

bility Index; PL

IF = poste

rior lu

mbar inter

body fusio

n; pt = p

atient; r

hBMP-2 =

recomb

inant huma

n morphogenetic p

rotein–2; TL

IF = tra

nsforaminal lu

mbar inter

body fusio

n.


M. G. Kaiser et al.


tinue since the use of threaded titanium cages through the PLIF approach had fallen out of favor. The relatively small number of patients (< 50) in each cohort and the lack of a blinded clinical assessment limit conclusions formulated from this study. The study provides Level II evidence in support of using rhBMP-2 with threaded cages through the PLIF approach (Table 3).

Additional studies of lesser quality have explored the potential of rhBMP-2 as a graft extender for lumbar inter-body fusion. Slosar et al. performed a prospective cohort study to determine the impact of rhBMP-2 on fusion rate and clinical outcome following ALIF with femoral ring allograft (FRA).54 Seventy-five patients with varying di-agnoses and undergoing up to a 3-level fusion were en-rolled; 30 control patients (n = 30) received an FRA with allograft croutons and an investigational group (n = 45) received an FRA supplemented with rhBMP-2/ACS. A sta-tistically significant increase in fusion rate was observed at all time points for the investigational group compared with controls. Both groups demonstrated significant clini-cal improvement at 12 and 24 months over baseline, but no significant difference was observed between treatment groups. A heterogeneous patient population with respect to presenting diagnosis and number of levels fused, an inadequate statistical analysis of potentially confounding baseline demographics, and failure to perform an indepen-dent, blinded clinical assessment requires that the study be downgraded to Level III in support of rhBMP-2 as an ad-junct to FRA interbody fusion (Table 3).

Several retrospective cohort studies and case series have investigated the use of rhBMP-2 as a graft extender when performing TLIF with an interbody graft. Rihn et al. performed a retrospective review of 48 patients receiving rhBMP-2 during TLIF procedures and observed a fusion rate of 95.8% at the 2-year follow-up with 71% reporting excellent or good outcomes.49 The complication rate was 21.7% with one-quarter of these complications attributed to the use of rhBMP-2. Villavicencio et al. reviewed the data from 74 patients undergoing either open or minimally invasive TLIF for varying diagnoses using rhBMP-2 and allograft.67 The fusion rate for the entire cohort was 100%; however, a trend toward improved clinical outcome was observed for patients undergoing less invasive procedures. Mummaneni et al. conducted a retrospective cohort study intended to compare the efficacy of rhBMP-2 with AICB for TLIF.44 Forty-four patients underwent a TLIF, with 40 patients (90%) available for a mean follow-up of 9 months. The control arm (n = 19) consisted of patients receiving AICB in an interbody spacer, while the investigational group (n = 21) received rhBMP-2 supplemented with either AICB (n = 12) or local autograft (n = 9). With at least 6 months of follow-up, the fusion rate in the AICB group was 95% and 100% in patients receiving rhBMP-2; however, only 76% of patients receiving rhBMP-2 were available for follow-up. At 6 months after surgery, 58% of patients con-tinued to complain of donor site pain. The authors conclud-ed that rhBMP-2 and local autograft is an excellent graft option and avoids donor site morbidity when performing a TLIF procedure. The relatively small, heterogeneous popu-lation of patients with respect to diagnosis and surgery per-formed limits these studies. Nonvalidated clinical outcome

measures were used, and the method of fusion assessment is questionable given the presence of pedicle screw instru-mentation. In the Mummaneni et al. study, neither the ra-diographic evaluation nor assessment of clinical outcome was performed in an independent, blinded fashion. Due to the baseline study designs and various limitations, these studies provide at best Level IV or V evidence in support of rhBMP-2 as a supplement for interbody fusion through the TLIF approach (Table 3). Additional case series have been published exploring the potential of rhBMP-2 as a graft extender or substitute.7,24,40 Burkus et al. published a long-term clinical and radiographic companion study to their previous published report of patient undergoing single level ALIF procedures with stand-alone tapered cages and rhBMP-2/ACS.7 No significant difference in outcome at 6 years was observed when compared with the previously published data obtained at 2 years after surgery. Geibel et al. reported a 100% fusion rate with an 89% patient satis-faction rate in 48 patients undergoing 1- and 2-level instru-mented PLIF with rhBMP-2 and posterolateral fusion.24 Lanman and Hopkins published the only case series inves-tigating the use of rhBMP-2 in conjunction with a biore-sorbable cage.40 This study was limited by 64% of patients lost to follow-up at 12 months after surgery, compromis-ing any attempt at a meaningful interpretation of the data. Since these studies are all case series with limitations, at best they provide only Level V evidence.

rhBMP-2: Posterolateral Fusion. In 2006, Dimar et al. reported the 24-month radiographic and clinical re-sults of patients enrolled in an FDA investigational device exemption (IDE) study comparing rhBMP-2 combined with a compression-resistant matrix (CRM; bovine type I collagen carrier containing 15% HA and 85% b-TCP) with AICB in instrumented posterolateral fusions (Table 4).19 Ninety-eight of 150 randomized patients were avail-able for review. Clinical outcome, assessed using vali-dated outcomes instruments (SF-36, ODI, and back/leg pain scores), was performed at 6 weeks and 3, 6, 12, and 24 months. Independent assessment of radiographs and CT images was performed at 6, 12, and 24 months. Op-erative parameters, including the surgical time and blood loss, were significantly less in the rhBMP-2/CRM cohort. Both groups demonstrated a significant clinical improve-ment compared with baseline, but not between treatment groups. The rhBMP-2/CRM cohort demonstrated a sta-tistically higher fusion rate, 90.6% compared with 73.3%. At final follow-up, 16% of patients in the AICB cohort continued to complain of donor site pain. The authors concluded that rhBMP-2/CRM demonstrated similar clinical outcomes and improved fusion rates compared with AICB for single-level instrumented posterolateral fusions. Thirty-five percent of patients from the original cohort of randomized patients were lost to follow-up. This study included a heterogeneous patient population with respect to diagnosis. Due to these limitations the study was considered to provide level II evidence in support of rhBMP-2/CRM as a substitute for AICB.

Dimar et al. later reported the 2-year radiographic and clinical outcomes of a multicenter prospective random-ized controlled IDE trial to investigate the use of rhBMP-2




TABL

E 4:

rhBM

P-2 i

n po

ster

olat

eral

fusio

n: su

mm

ary o

f evid

ence

*

Auth

ors &

Ye

arLe

vel o

f Ev

idenc

eDe

scrip

tion o

f Stu

dyCo

mmen

ts

Dima

r et a

l.,

20

09II

The p

urpo

se of

this

study

was

to re

port

the 2

-yr r

adiog

raph

ic &

clinic

al ou

tcome

s of a

mult

icente

r,

prospective, randomized c

ontro

lled IDE

trial to investigate the u

se of rh

BMP-2 m

atrix (bovine

type I co

llagen c

arrier containing 15

% HA & 85% β-TCP

) as a

substitu

te for a

utolog

ous iliac

cr

est fo

r 1-le

vel p

oster

olat in

strum

ented

fusio

ns. W

ell-d

escr

ibed r

adiog

raph

ic cr

iteria

wer

e use

d

to as

sess

fusio

n in a

blind

ed fa

shion

, & va

lidate

d outc

ome m

easu

res w

ere c

omple

ted to

de-

ter

mine

clini

cal o

utcom

e. Cl

inica

l follo

w-up

was

perfo

rmed

at 6

wks &

at 3,

6, 12

, & 24

mos

&

ra

diogr

aphic

follo

w-up

at 6,

12, &

24 m

os. 4

63 pt

s wer

e ran

domi

zed b

twn t

he tr

eatm

ent c

ohor

ts,

w/ a 2-year follow-up ra

te of 89%. T

he co

ntrol group d

emonstrate

d sign

ificantly longer op

times

& blo

od loss. T

he clinical outc

ome m

easures imp

roved s

ignific

antly co

mpared w/ preop sc

ores

in both cohorts

w/ no s

ignific

ant diffe

rence n

oted b

twn treatm

ent groups. 60% of the c

ontro

l

grou

p con

tinue

d to c

ompla

in of

dono

r site

pain

at th

e 24-

mo fo

llow-

up ev

aluati

on. T

he fu

sion

rate, ba

sed o

n CT assessme

nt, de

monstra

ted a sta

tistically s

ignific

ant diffe

rence in fusion

rates

at all tim

es po

ints; the

final fus

ion rate wa

s 96%

in the rhB

MP-2 group &

89% in the c

ontrol group.

Th

ere w

as no

signific

ant diffe

rence in a

dverse ev

ents except that the c

ontro

l group su

ffered 1

7

graft site–rela

ted ev

ents. The au

thors c

onclu

ded that th

e use of rh

BMP-2 m

atrix improved op

para

meter

s, ac

hieve

d a hi

gher

fusio

n rate

& co

mpar

able

clinic

al ou

tcome

s, &

ther

efore

can

eli

mina

te th

e nee

d for

auto

logou

s ilia

c cre

st bo

ne fo

r 1-le

vel p

oster

olat in

strum

ented

fusio

ns.

This

is a w

ell-d

esign

ed tr

ial. T

he au

thor

s fail

ed to

perfo

rm a

powe

r

analy

sis to

deter

mine

samp

le siz

e & fa

iled t

o acc

ount

for p

ts los

t

to fo

llow-

up. It

is no

t clea

r if th

e clin

ical a

sses

smen

t was

blind

ed;

ho

weve

r, give

n tha

t the o

utcom

e mea

sure

s wer

e pt s

elf-a

sses

s-

me

nt ins

trume

nts, th

is fac

t does n

ot detra

ct sig

nificantly from the

ob

serv

ation

s & co

nclus

ions o

f the s

tudy

. Due

to th

ese l

imita

tions

the s

tudy

was

down

grad

ed to

Leve

l II ev

idenc

e to s

uppo

rt th

e use

of rhBM

P-2 a

s a su

bstitu

te for A

ICB in 1-lev

el poste

rolat fusio

ns.

Daws

on et

al., 2

009

IITh

e pur

pose

of th

is mu

lticen

ter pr

ospe

ctive

rand

omize

d con

trolle

d tria

l was

to in

vesti

gate

the u

se

of rhBM

P-2 o

n an A

CS re

inforced w

/ 15%

HA/85% TCP

ceramic g

ranules

as a repla

ceme

nt for

ilia

c cre

st au

togr

aft in

poste

rolat

instr

umen

ted fu

sions

. Well

-des

cribe

d rad

iogra

phic

crite

ria w

ere

us

ed to

asse

ss fu

sion i

n a bl

inded

fash

ion, &

valid

ated o

utcom

e mea

sure

s wer

e com

pleted

to

de

termi

ne cl

inica

l outc

ome.

Over

all su

cces

s was

deter

mine

d by c

ombin

ing th

e res

ults o

f thes

e

meas

ures

. Clin

ical fo

llow-

up w

as pe

rform

ed at

3, 6,

12, &

24 m

os &

radio

grap

hic fo

llow-

up at

6,

12

, & 24

mos

. 50 p

ts we

re ra

ndom

ized t

o the

coho

rts w

/ a 24

-mo f

ollow

-up r

ate of

88%

in th

e

treatm

ent g

roup

& 8

6% in

the c

ontro

l gro

up. Im

prov

emen

ts we

re ob

serv

ed in

all c

linica

l outc

ome

me

asur

es in

both

grou

ps w

/ a tr

end t

owar

d sup

erior

impr

ovem

ent in

the e

xper

imen

tal g

roup

. At

each tim

e poin

t, there was a tre

nd for m

ore s

uccessful fu

sion in the treatment group, w

/ final

fu

sion r

ates o

f 95%

in th

e inv

estig

ation

al gr

oup &

70%

in th

e con

trol g

roup

. No d

iffere

nce i

n the

radio

graphic

or clinical fo

llow-up pr

oved to be

statistically signific

ant. T

he au

thors c

onclu

ded

that the c

ombin

ation of rh

BMP-2 &

HA/TC

P ma

y be a

n effe

ctive alter

native to a

utolog

ous b

one

gr

aft fo

r 1-le

vel p

oster

olat in

strum

ented

fusio

ns.

This

is a w

ell-d

esign

ed st

udy b

ut wa

s dow

ngra

ded t

o Lev

el II e

vi-

de

nce d

ue to

the r

elativ

ely sm

all pt

coho

rts (<

50 pt

s/coh

ort),

in-

co

mplet

e des

cript

ion of

base

line p

t dem

ogra

phic

data

, failu

re to

desc

ribe i

f the c

linica

l ass

essm

ent w

as bl

inded

, & an

over

all ou

t-

come

mea

sure

that

has n

ot be

en va

lidate

d.

(cont

inued

)


M. G. Kaiser et al.


TABL

E 4:

rhBM

P-2 i

n po

ster

olat

eral

fusio

n: su

mm

ary o

f evid

ence

* (co

ntin

ued)

Auth

ors &

Ye

arLe

vel o

f Ev

idenc

eDe

scrip

tion o

f Stu

dyCo

mmen

ts

Glassm

an et

al.

, 200

8II

The p

urpo

se of

this

pros

pecti

ve R

CT w

as to

comp

are t

he cl

inica

l & ra

diogr

aphic

outco

mes o

f

paten

ts >6

0 yrs un

dergoin

g instru

mente

d poster

olat fusion

s w/ either rh

BMP-2/AC

S or AICB.

Va

rious

bone

graf

t exte

nder

s wer

e app

lied a

t the d

iscre

tion o

f the s

urge

on in

both

coho

rts. V

al-

ida

ted ou

tcome

mea

sure

s (SF

-36,

ODI, &

NRS

for b

ack &

leg p

ain) w

ere a

dmini

stere

d. 2-

yr

follow-up da

ta were c

ollected

in 94%

of the p

ts (49 in the rh

BMP-2 c

ohort &

51 in the A

ICB

group). Baseline NRS

leg p

ain was re

porte

d w/ greate

r frequency in the r

hBMP-2/AC

S cohort

(p =

0.03

1); ho

weve

r, the

re w

ere n

o oth

er di

ffere

nces

in ba

selin

e dem

ogra

phics

. The

re w

as a

statis

ticall

y gre

ater n

o. of

comp

licati

ons i

n the

AIC

B co

hort

(20 v

s 8, p

= 0.

014),

alth

ough

none

of

the c

omplications were d

irectly a

ttribu

ted to either the h

arvest of AICB

or the u

se of rh

BMP-2/

ACS. Statistically s

ignific

ant im

provem

ent w

as ob

served in all clinica

l outc

ome m

easures in b

oth

cohorts

comp

ared w/ baseline, although no

ne of the d

ifferences btwn

the c

ohorts wa

s sign

ifi-

ca

nt. C

T im

ages

wer

e obt

ained

in 9

9 pts

at 24

mos

after

surg

ery (

93%

follo

w-up

rate)

. The

fusio

n rate

in the r

hBMP-2/AC

S cohort wa

s 86.3%

& 70

.8% in the A

ICB group. Th

e average CT

grade w

as signific

antly higher in the r

hBMP-2/AC

S group (4.3

vs 3.8 [p =

0.03]). Revision su

r-

gery for n

onunion

was re

quired in 1

pt in the r

hBMP-2 c

ohort &

5 in the A

ICB group. An

estim

ation of total cost over the 2 yrs w

as no

t sign

ificantly different btwn

the 2

groups ($

42,574

for the AICB cohort & $40,1

31 for the rh

BMP-2/AC

S cohort). T

he au

thors c

onclu

ded that th

e

study pr

ovide

d Level I evid

ence su

pporting the us

e of rhB

MP-2/AC

S as an

AICB repla

ceme

nt

fo

r lumb

ar fu

sion i

n the

olde

r pt.

This

is a h

etero

gene

ous p

opula

tion o

f pts

w/ re

spec

t to th

e pre

sent-

ing di

agno

sis, le

vel o

f invo

lveme

nt, &

no. o

f leve

ls inc

luded

in th

e

surg

ery.

No po

wer a

nalys

is wa

s per

form

ed to

deter

mine

samp

le

siz

e. Th

e sur

gical

proc

edur

e was

not s

tand

ardiz

ed, w

/ bon

e gra

ft

ex

tende

rs ad

ded t

o both

grou

ps at

the d

iscre

tion o

f the s

urge

on.

It i

s not

clear

if th

e “gr

ading

” sch

eme o

f fus

ion as

sess

ment

has

be

en va

lidate

d. No

stati

stics

wer

e per

form

ed w

/ res

pect

to ra

te of

fusio

n. Du

e to t

hese

limita

tions

the s

tudy

was

down

grad

ed to

Level II ev

idence in s

upport of rhBM

P-2 a

s a su

bstitu

te for A

ICB

when

supp

lemen

ted w

/ a gr

aft e

xtend

er fo

r pos

terola

t instr

umen

t-

ed fu

sions

in pt

s >60

yrs.

Dima

r et a

l.,

20

06II

The p

urpo

se of

this

pros

pecti

ve ra

ndom

ized n

onbli

nded

repo

rt wa

s to p

rese

nt the

24-m

o rad

iogra

phic

&

clinic

al re

sults

of pt

s enr

olled

in an

FDA

IDE

study

unde

rgoin

g 1-le

vel in

strum

ented

poste

rolat

fusion

w/ either rh

BMP-2/C

RM or AICB. 98

of 15

0 pts, a follow

-up rate

of 65

%, w

ere a

vailable

at

24

mos

after

surg

ery.

Clini

cal a

sses

smen

t w/ v

alida

ted ou

tcome

s ins

trume

nts (S

F-36

, ODI

, & ba

ck/

leg

pain

scor

es) w

as pe

rform

ed at

6 wk

s & 3,

6, 12

, & 24

mos

. A bl

inded

radio

logist

& 2

ortho

pedic

surg

eons

inde

pend

ently

evalu

ated r

adiog

raph

s & C

T im

ages

at 6,

12, &

24 m

os. T

he su

rgica

l time

& blo

od loss were s

ignific

antly less in the rhB

MP-2/CRM

cohort. Sign

ificant im

provem

ents in all

cli

nical

outco

me m

easu

res c

ompa

red w

/ bas

eline

value

s wer

e obs

erve

d in b

oth gr

oups

, but

no

difference ex

isted btwn

groups. A

signific

antly higher fusio

n rate

was ob

served in the rhB

MP-2/

CR

M co

hort

(90.6

% vs

73.3%

). 16%

of pt

s in t

he A

ICB

coho

rt co

ntinu

ed to

comp

lain o

f don

or si

te

pain at 24 mos. The au

thors conclud

ed that rhBM

P-2/C

RM de

monstrated

similar clinical re

sults &

im

prov

ed fu

sion r

ates a

s AIC

B for

1-lev

el ins

trume

nted p

oster

olat fu

sions

.

The h

etero

gene

ous p

t pop

ulatio

n w/ r

espe

ct to

diagn

osis

limits

this

study & the s

ignific

ant lo

ss to follow-up at the 2

-yr study en

d poin

t.

Valid

ated c

linica

l outc

omes

wer

e utili

zed &

an ef

fectiv

e meth

od

of

fusio

n ass

essm

ent w

as pe

rform

ed in

a bli

nded

fash

ion. D

ue to

the s

ignific

ant lo

ss to follow-up the s

tudy is do

wngraded to Le

vel

II e

vidence in su

pport of rhB

MP-2/CR

M as a substitu

te for A

ICB in

1-

level

poste

rolat

instr

umen

ted fu

sions

.

(cont

inued

)




TABL

E 4:

rhBM

P-2 i

n po

ster

olat

eral

fusio

n: su

mm

ary o

f evid

ence

* (co

ntin

ued)

Auth

ors &

Ye

arLe

vel o

f Ev

idenc

eDe

scrip

tion o

f Stu

dyCo

mmen

ts

Tagh

avi e

t

al., 2

010

IIITh

e obje

ctive of this retro

spective c

omparative s

tudy was to de

termine the e

fficacy o

f rhB

MP-2/

local

bone

to ei

ther

allog

raft

or au

togr

aft in

revis

ion in

strum

ented

poste

rolat

fusio

ns. 6

2 pts

w/

varying

initia

l diag

noses w

ere includ

ed w/ m

inimu

m follow-up of 2 yrs. Pts w

ere d

ivided into

3

groups: G

roup 1 (n = 24

) receiv

ed rh

BMP-2, Group 2

(n = 18

) BMA/allograft, & Group 3 (n = 20

)

auto

graf

t. All r

eceiv

ed su

pplem

enta

l loca

l bon

e. Fu

sion a

sses

smen

t, thr

ough

stati

c & dy

nami

c

radio

grap

hs, w

as pe

rform

ed by

3 bli

nded

inde

pend

ent r

eview

ers w

/ a di

agno

sis of

nonu

nion

based o

n either su

rgica

l exploration if revis

ion was pe

rform

ed or

radio

graphic

findin

gs. C

linica

l

outco

me was de

termined through VAS

scores. G

roup 1 demo

nstra

ted a fusio

n rate

of 10

0%,

Group 2

demo

nstra

ted a 77.8%

fusio

n rate

, & Group 3 had a

100%

fusio

n rate

. Pts undergoin

g

multil

evel

proc

edur

es w

/ BM

A/all

ogra

ft de

mons

trated

a sta

tistic

ally l

ower

fusio

n rate

. No d

iffer-

ence in VAS

scores was ob

served. T

he au

thors c

onclu

ded that rhB

MP-2 c

ould be an

appropri-

ate al

terna

tive t

o AIC

B in

revis

ion po

stero

lat fu

sion.

Alth

ough

an ad

equa

te de

scrip

tion o

f bas

eline

demo

grap

hics i

s

prov

ided w

/ app

ropr

iate s

tatis

tical

analy

sis. th

ere r

emain

s the

pos-

sibilit

y of s

electi

on bi

as du

e to t

he po

tentia

l diffe

renc

es in

pres

entin

g diag

nose

s tha

t the a

uthor

s do n

ot de

scrib

e. Th

e auth

ors

inc

luded

an ad

equa

te as

sess

ment

of fu

sion w

/ acc

epta

ble ra

dio-

gr

aphic

crite

ria &

inco

rpor

ated v

alida

ted ou

tcome

mea

sure

s. Du

e

to th

e retr

ospe

ctive

natur

e of th

e stu

dy de

sign,

but la

ck of

sig-

nificant lim

itations, th

e study is co

nside

red L

evel III ev

idence in

support of rhB

MP-2/loc

al bone as

an alter

native to A

ICB for revi-

sio

n pos

terola

t fus

ions.

Sing

h et a

l.,

20

06III

This is a p

rospective c

ase-ma

tched co

hort stu

dy to de

termine if rhBM

P-2 e

nhances fusion

rate w/in

a s

horte

r tim

e inte

rval

for pt

s und

ergo

ing in

strum

ented

poste

rolat

fusio

n usin

g AIC

B. 52

pts

presenting w

/ sten

osis & spondylolisthesis

were e

valua

ted: 41 receiv

ed rh

BMP-2/A

ICB/loc

al bone

&

11 re

ceive

d AIC

B/loc

al bo

ne. F

usion

asse

ssme

nt wa

s per

forme

d w/ r

eform

atted

CT

imag

es &

evalu

ated i

n a bl

inded

, inde

pend

ent m

anne

r by 2

surg

eons

& 1

radio

logist

. 2 pt

s wer

e los

t to

follow

-up a

t the 2

-yr tim

e poin

t. The fusio

n rate

s in the rh

BMP-2 &

AICB cohorts were 9

7% & 77

%,

respectively

. Pts receiving rh

BMP-2 w

ere judged to fuse faster & de

monstrate mo

re robust fus

ions.

Fusio

n rate

& qu

ality of fus

ion proved to be

statistically su

perior in

the rhB

MP-2 co

hort. No c

ompli-

cation w

as attrib

uted to the us

e of rhB

MP-2. The au

thors conclud

ed that rhBM

P-2 m

ay se

rve a

s a

sa

fe &

effec

tive s

upple

ment

to AI

CB fo

r pos

terola

t instr

umen

ted fu

sion.

Limita

tions

of th

is stu

dy in

clude

the p

otenti

al for

selec

tion b

ias si

nce a

n

inade

quate

desc

riptio

n of b

aseli

ne de

mogr

aphic

s is p

rovid

ed &

the

autho

rs failed

to ev

aluate

these d

ata for sign

ificant diffe

rences, e.g.,

it i

s not

know

n if th

e no.

of lev

els fu

sed w

as co

mpar

able

btwn

tre

atmen

t gro

ups.

The n

o. of

pts w

/in th

e con

trol g

roup

is re

lative

ly

sm

all. It

is no

t clea

r whe

ther t

he su

rgica

l tech

nique

was

stan

dard

-

ized b

etwee

n coh

orts,

w/ n

o men

tion r

egar

ding t

he am

ount

of AI

CB

us

ed. T

he va

lue of

the o

utcom

e par

amete

rs, in

parti

cular

the s

ub-

jec

tive a

sses

smen

t of fu

sion q

uality

, is of

limite

d valu

e & ha

s not

been

show

n to i

mpac

t clin

ical o

utcom

e. Du

e to t

hese

limita

tions

the st

udy w

as do

wngr

aded

to Le

vel II

I evid

ence

in su

ppor

t of

rhBM

P-2 a

s a graft exte

nder for A

ICB in ins

trume

nted p

osterola

t

fusion

s.

(cont

inued

)


M. G. Kaiser et al.


TABL

E 4:

rhBM

P-2 i

n po

ster

olat

eral

fusio

n: su

mm

ary o

f evid

ence

* (co

ntin

ued)

Auth

ors &

Ye

arLe

vel o

f Ev

idenc

eDe

scrip

tion o

f Stu

dyCo

mmen

tsGlassm

an et

al.

, 200

727III

The o

bjecti

ve of

this

retro

spec

tive r

eview

of da

ta co

llecte

d dur

ing a

pros

pecti

ve, r

ando

mize

d, unblinded trial was to de

termine the influence o

f smo

king o

n fusion

rate & outco

me of pts receiv

ing

either A

ICB or rh

BMP-2 for single-lev

el poste

rolateral fus

ions. All pts we

re ev

aluate

d at a minimu

m of

2 yrs

after

surg

ery.

Clini

cal o

utcom

e was

mea

sure

d utili

zing v

alida

ted ou

tcome

s ins

trume

nts

(ODI

, SF-

36, b

ack &

leg p

ain sc

ores

) at 6

wks

& 3,

6, 12

, & 24

mos

after

surg

ery.

An in

depe

nden

t, bli

nded

asse

ssme

nt of

fusion

statu

s was

perfo

rmed

w/ s

tatic

& dy

nami

c rad

iogra

phs &

CT

imag

-ing

at 6,

12, &

24 m

os af

ter su

rger

y. Th

e rec

ords

of 14

8 pts

were

revie

wed,

42 sm

oker

s & 10

6 no

nsmo

kers.

The

smok

ers w

ere e

quall

y dist

ribute

d btw

n the

2 co

horts

, w/ 5

5 non

smok

ers i

n the

rhBM

P-2 c

ohort &

51 in the A

ICB group. Fusio

n rate

at 24

mos ba

sed o

n radiog

raphs w

as 10

0%

in the

rhBM

P-2 n

onsm

okers, 95.2%

in the rhB

MP-2 sm

okers, 94.1%

in the A

ICB nonsmo

kers, &

76.2%

in the A

ICB sm

okers. A sig

nificant difference was ob

served btwn

all smo

kers (85.7

%) &

all no

nsmo

kers

(97.2%

) as w

ell as

btwn

smok

ers &

nons

moke

rs re

ceivi

ng A

ICB.

Sim

ilar r

ates o

f fus

ion w

ere d

eterm

ined w

/ CT

imag

ing. C

linica

l outc

ome w

as im

prov

ed in

all p

aram

eters

in all

4 cohorts, w

/o a s

ignific

ant diffe

rence in the de

gree of im

provem

ent btwn g

roups. When c

onsid

ered

colle

ctive

ly, th

e non

smok

ers c

onsis

tently

demo

nstra

ted be

tter O

DI &

SF-

36 sc

ores

. The

autho

rs conclud

ed that rhBM

P-2/C

RM may en

hance fusion

rate in cig

arette s

mokers undergoin

g sing

le-lev

el fus

ion &

that

smok

ing is

detrim

ental

to cl

inica

l outc

ome r

egar

dless

of fu

sion s

tatus

.

The r

etros

pecti

ve na

ture o

f this

study

does

limit i

nfere

nces

form

u-lat

ed fr

om th

e res

ults d

espit

e the

fact

that

the d

ata w

ere c

ollec

ted

durin

g a pr

ospe

ctive

RCT

. Sinc

e pts

were

not r

ando

mize

d w/ r

e-spect to

smoking

statu

s, the b

enefit of th

e randomization p

rocess

is nullifi

ed. G

iven the study d

esign

, this report provide

s Level III

evide

nce r

egardin

g the im

pact of rhBM

P-2/CR

M on fusio

n rate

in

smok

ers &

the i

mpac

t of s

mokin

g on c

linica

l outc

ome.

Lee e

t al.,

2009

IVThe o

bjective of this retrospective c

omparative s

tudy w

as to de

termine the e

fficacy o

f rhB

MP-2 in pts

>65 y

rs un

derg

oing a

n ins

trume

nted p

oster

olat fu

sion.

127 p

ts w/

lumb

ar de

gene

rativ

e dise

ase

were divid

ed into 3 g

roups: pts

>65 yrs receiv

ed bo

th rhBM

P-2 &

allog

raft (Group A

, n = 34), p

ts <65 y

rs receive

d rhB

MP-2 & allog

raft (Group B

, n = 52), &

pts >

65 yrs receiv

ed au

tograft (Group

C, n

= 41).

Fusio

n ass

essm

ent w

as pe

rform

ed by

an in

depe

nden

t blin

ded r

adiol

ogist

using

stati

c &

dyna

mic r

adiog

raph

s, su

pplem

ented

w/ C

T wh

en fu

sion w

as qu

estio

nable

. Kirk

aldy-W

illis &

VAS

sc

ores

wer

e use

d to d

eterm

ine cl

inica

l outc

ome.

The a

vera

ge fo

llow-

up w

as ~3

6 mos

, alth

ough

VA

S scores were o

btaine

d up to 2

4 mos after su

rgery. The fusion

rates

were 8

2.4% in Group A,

94.2%

in Group B, &

78.1%

in Group C. A

t 2 yrs, Group B

demo

nstrated

a sta

tistically s

uperior

VAS score c

ompared w

/ Group A. C

linica

l outc

ome w

as no

t statistica

lly different btw

n the 3

groups. The au

thors conclud

ed that rhBM

P-2/a

llograft yields eq

uivale

nt outco

mes to A

ICB in pts

>6

5 yrs,

but is

not a

ble to

over

come

all n

egati

ve co

morb

iditie

s ass

ociat

ed w

/ age

.

Ther

e are

limite

d bas

eline

demo

grap

hic da

ta pr

ovide

d, w/

an in

ad-

eq

uate

desc

riptio

n of c

omor

biditie

s tha

t wou

ld af

fect fu

sion

po

tentia

l. App

aren

t diffe

renc

es ex

ist w

/ res

pect

to se

vera

l of th

ese

co

morb

iditie

s, e.g

., no.

of lev

els fu

sed,

& the

autho

rs fai

l to de

ter-

mine if the

se differences were s

ignific

ant. T

he ex

istence of se

lection

bia

s the

refor

e can

not b

e exc

luded

. The

surg

ical te

chniq

ue is

not d

e-

sc

ribed

. An a

dequ

ate as

sess

ment

of fus

ion w

as pe

rform

ed, a

l-

thoug

h the

autho

rs inc

lude t

ime t

o fus

ion, w

hich i

s of q

uesti

onab

le

sig

nificance. Du

e to these lim

itations the s

tudy w

as do

wngraded

to Level IV

evide

nce in s

upport of rhBM

P-2/a

llograft as a

substitu

te

for

AIC

B in

pts >

65 yr

s und

ergo

ing 1-

level,

instr

umen

ted, p

oster

olat

fus

ions.

Hami

lton e

t

al., 2

008

VTh

e auth

ors p

erfor

med a

retro

spec

tive r

eview

of pt

s und

ergo

ing no

ninstr

umen

ted po

stero

lat fu

sion

using

local harveste

d auto

graft supple

mente

d w/ rhB

MP-2/ACS

. 47 o

f 55 p

ts, median

age 6

8.2

ye

ars,

were

evalu

ated.

Fusio

n ass

essm

ent w

as in

depe

nden

tly pe

rform

ed us

ing ra

diogr

aphs

& C

T

sc

ans a

t var

ious t

ime p

oints

up to

36 m

os af

ter su

rger

y. An

inde

pend

ent c

linica

l eva

luatio

n usin

g

valid

ated i

nstru

ments

was

perfo

rmed

at le

ast 2

9 mos

& up

to 36

mos

after

surg

ery.

The f

usion

rate wa

s 80%

. Greate

r than 8

5% of pts d

emonstrate

d an imp

roved c

linica

l outc

ome &

pain relief.

The a

uthors c

onclu

ded that th

e use or rh

BMP-2 a

s a su

pplem

ent to

nonin

strum

ented

poste

rolat

fus

ions l

eads

to im

prov

ed pa

in &

a com

para

ble fu

sion r

ate in

elde

rly pt

s.

This

case

serie

s con

tains

a he

terog

eneo

us st

udy p

opula

tion w

/

respect to

fusio

n justification &

no. of le

vels inv

olved. A

lthough the

cli

nical

outco

me w

as pe

rform

ed by

an in

depe

nden

t ass

esso

r, pts

were

requ

ired t

o “re

call”

their

pre-

& po

stop p

ain &

healt

h sta

tus.

Th

e description

of clinical re

sults is diffic

ult to inter

pret. CT evalu

-

ation

was

only

perfo

rmed

in 8

5% of

stud

y pts.

Due

to th

ese l

imita

-

tions

, the c

ase s

eries

is do

wngr

aded

to Le

vel V

evide

nce t

hat

rhBM

P-2 is a

n effe

ctive bo

ne gr

aft exte

nder.

* AC

S = absorbable collagen s

ponge; AICB

= au

tolog

ous iliac c

rest bone; β-TCP

= β-trica

lcium

phosphate

; BMA = bone marrow aspirate

; CRM

= co

mpression

-resis

tant ma

trix; HA

= hy

droxyapatite

; IDE = investigational devic

e exemp

tion; NR

S = n

umeric rating s

cale; ODI = Os

westr

y Disa

bility Index; pt = p

atient; R

CT = random

ized c

ontro

lled trial; rhB

MP-2 =

recomb

inant huma

n bone m

orphoge-

netic pr

otein–

2; SF

-36 =

36-Item Sh

ort Form He

alth S

urvey; VA

S = vis

ual analog

scale

.




matrix as a substitute for AICB for single-level postero-lateral instrumented fusion (Table 4).20 Fusion assessment was performed in a blinded fashion, and clinical status was evaluated through validated outcome measures. Clinical follow-up was performed at 6 weeks and at 3, 6, 12, and 24 months and radiographic follow-up at 6, 12, and 24 months. Four hundred sixty-three patients were randomized, with a 2-year follow-up rate of 89%. Significantly longer op-erative times and greater blood loss was observed in the control group. There was no statistical difference in clini-cal outcome between groups, although both demonstrated significant improvement compared with baseline scores. Donor site pain was reported in 60% of the control group at final follow-up. Based on CT imaging, the interventional group demonstrated a statistically superior fusion rate at 6 months (79% with rhBMP-2 and 65% with AICB [p = 0.002]) and at 24 months [96% with rhBMP-2 and 89% with AICB [p = 0.014]). The overall rate of adverse events was statistically similar; however, 17 graft-related compli-cations were recorded in the control group. The authors concluded that the use of rhBMP-2 matrix improved op-erative parameters, led to a higher fusion rate, and achieved comparable clinical outcomes to AICB and therefore can be considered an acceptable substitute for single-level pos-terolateral instrumented fusion. This was a well-designed and well-executed study. It is not clear if the clinical as-sessment was blinded; however, utilization of patient self-assessment questionnaires decreases the likelihood of bias with reporting. The authors did not perform an appropriate power analysis to determine sample size and failed to pro-vide information regarding patients lost to follow-up. Due to these limitations the study was downgraded to Level II evidence in support of rhBMP-2 matrix as a substitute for AICB.

Dawson et al. conducted a multicenter prospective RCT to investigate if rhBMP-2/ACS supplemented with an HA/TCP extender could serve as an appropriate sub-stitute for AICB in instrumented posterolateral fusions (Table 4).17 Fifty patients were randomized; clinical fol-low-up was performed at 3, 6, 12, and 24 months, and radiographic follow-up at 6, 12, and 24 months. At 24 months, the follow-up rate was 88% in the treatment group and 86% in the control group. Both groups demonstrated improvements in all clinical outcome measures with the rhBMP-2 cohort demonstrating a trend toward better out-comes. The fusion rate was higher at all time points in the rhBMP-2 cohort, with final fusion rates of 95% in the investigational group and 70% in the control group. No difference in the radiographic or clinical outcome proved to be statistically significant. The authors concluded that the combination of rhBMP-2 and HA/TCP could be an effective alternative to AICB for single-level posterolat-eral instrumented fusions. The relatively small numbers of patients (< 50 patients per treatment arm), failure to provide adequate baseline demographic data, and utiliza-tion of a nonvalidated composite score to assess overall success are limitations of the study. The study was there-fore considered to provide Level II evidence in support of rhBMP-2/ACS and HA/TCP as a graft substitute for instrumented posterolateral fusions.

Glassman et al. conducted a prospective RCT to com-

pare the clinical and radiographic outcomes of 106 pa-tients older than 60 years of age undergoing instrumented posterolateral fusions with either rhBMP-2/ACS or AICB (Table 4).26 The method of grafting was not standardized, with various graft extenders added at the discretion of the surgeon. Clinical outcome was determined with validated outcome measures, including SF-36, ODI, and numeric rat-ing scale (NRS) for back and leg pain. Computed tomog-raphy scans were used to assess fusion. At 24 months after surgery the clinical and radiographic follow-up was 94% and 93%, respectively. At baseline, the patients in the rh-BMP-2 cohort reported leg pain with greater frequency (p = 0.031); there were no other differences in baseline demo-graphics. The complication rate was significantly greater in the AICB cohort (20 vs 8, p = 0.014), although none of the complications were directly attributed to either the harvest of AICB or the use of rhBMP-2/ACS. Both cohorts dem-onstrated a statistically significant clinical improvement over baseline; however, there was no difference in clini-cal outcome between treatment groups. An 86.3% fusion rate was observed in the rhBMP-2/ACS cohort, compared with 70.8% in the AICB group. The authors provided a CT “grade” for the observed fusion with the rhBMP-2 cohort demonstrating a significantly higher score (4.3 vs 3.8 [p = 0.03]). Nonunion requiring revision was reported in 1 pa-tient in the rhBMP-2 cohort and 5 in the AICB group. An estimation of total cost over 2 years was calculated, and the difference between the 2 groups was not significantly dif-ferent ($42,574 for the AICB cohort and $40,131 for the rh-BMP-2/ACS cohort). The authors concluded that the study provided Level I evidence supporting the use of rhBMP-2/ACS as an AICB replacement for lumbar fusion in the older patient. This study suffers from several limitations, including a heterogeneous patient cohort with respect to presenting diagnosis, failure to account for patients lost to follow-up, lack of a standard surgical protocol, question-able “grading” scheme to assess fusion, failure to deter-mine sample size through a power analysis, and failure to perform an adequate statistical analysis. Due to these limi-tations the study was downgraded to Level II evidence in support of rhBMP-2 for patients older than 60 years of age undergoing posterolateral lumbar fusions.

Singh et al. published a prospective cohort study to compare outcome of patients receiving a mixture of rhBMP-2/local bone/AICB (n = 41) to those receiv-ing only local bone/AICB (n = 11).53 Fusion assessment was performed in an independent, blinded manner with reformatted CT images. The fusion rate with rhBMP-2 was 97% while the control cohort demonstrated a fusion rate of 77%. Those receiving rhBMP-2 were thought to achieve fusion faster and demonstrate a more robust fu-sion. However, this study failed to provided an adequate description of baseline demographics (for example, the number of levels fused in each group). This is a small and heterogeneous population of patients; it is not clear if the surgical procedure was standardized between cohorts, and no objective clinical outcomes were reported. The study was therefore downgraded to Level III evidence in support of utilizing rhBMP-2 as an extender.

Hamilton et al. published a retrospective case se-ries of patients undergoing noninstrumented posterolat-


M. G. Kaiser et al.


eral fusions utilizing local autograft supplemented with rhBMP-2.30 An 80% fusion rate was observed; 85% of patients were felt to demonstrate clinical improvement. This study provides only Level V evidence in support of rhBMP-2 as an extender with local bone for noninstru-mented fusions due to the heterogeneous patient popula-tion, failure to collect clinical outcomes in a prospective manner, and a radiographic follow-up rate of only 85%.

Taghavi et al. performed a retrospective cohort study to determine the efficacy of rhBMP-2/local bone to either allograft combined with bone marrow aspirate or autograft, in revision instrumented, posterolateral fusions. Sixty-two patients with varying diagnoses were included with a mini-mum follow-up of 2 years.56 Patients were divided into 3 groups: Group 1 (n = 24) received rhBMP-2, Group 2 (n = 18) received BMA/allograft, and Group 3 (n = 20) received autograft. The exact source of autograft bone for Group 3 was not clearly defined. All 3 cohorts received supplemen-tal local bone. Static and dynamic radiographs were used to assess fusion and were reviewed by 3 blinded indepen-dent reviewers with a diagnosis of nonunion based on either surgical exploration if revision performed or radiographic findings. Clinical outcome was determined through VAS scores. A fusion rate of 100% was observed for Groups 1 and 3; however, Group 2 demonstrated a 77.8% fusion rate. Patients undergoing multilevel procedures with BMA/al-lograft demonstrated a statistically lower fusion rate. No difference in VAS scores was observed. The authors con-cluded that rhBMP-2 could be an appropriate alternative to AICB in revision posterolateral fusion. Although this study was well executed with appropriate follow-up, vali-dated outcome measures, and appropriate assessment of fusion, due to the retrospective nature of the study design it provides Level III evidence. As this was the only study identified to investigate the use of rhBMP-2 in revision sur-gery, evidence is insufficient to formulate a recommenda-tion. Additional studies of similar or lesser quality, such as retrospective reviews or case series, promoting the use of rhBMP-2 for various clinical scenarios, such as in patients who use tobacco, have been published.3,25,27,36 Due to an in-sufficient number of such studies no formal recommenda-tions could be constructed regarding the use of rhBMP-2 under the specific clinical circumstances.

rhBMP-2: Complications. Between 2003 and 2007, the annual number of procedures utilizing BMPs increased by 4.3-fold (from 23,900 to 103,194 cases), with spinal fu-sions accounting for almost 93% of these cases.46 Although it is difficult to deny a positive impact on fusion rate, sur-geons must be aware of the potential risks and complica-tions related to the use of BMPs, particularly since the ma-jority of procedures would be considered off label.

Rihn et al. performed a single-center retrospective co-hort study to specifically identify complications associated with the use of rhBMP-2 for single-level TLIF procedures and to determine if these complications differed compared with the use of AICB.50 Between January of 2004 and May of 2007, 130 patients underwent a single-level TLIF using either AICB or rhBMP-2 (Table 5). One hundred nineteen patients were available for review, 33 receiving AICB and 86 receiving rhBMP-2, with an average radiographic fol-

low-up of 19.1 months and an average clinical follow-up of 27.6 months. A combination of plain radiographs and CT images were used to assess fusion status. Those patients receiving AICB demonstrated a 96.5% fusion rate, and the fusion rate in the rhBMP-2 cohort was 97% (p = 0.09). The overall complication rate was higher in the autograft cohort (45.5% vs 29.1%), but the difference was not significant. Donor site morbidity was the most common complication associated with AICB, and postoperative radiculitis was more often observed in the rhBMP-2 cohort (14% vs 3% [p = 0.08]). A significant decrease in radiculitis was observed after 2006 (20.4% to 5.4% [p = 0.047]), following the uti-lization of a hydrogel sealant intended to shield the exiting root. Additional complications thought to be related to the use of rhBMP-2 included osteolysis and heterotopic bone formation. The authors concluded that the TLIF proce-dure, regardless of graft, is associated with a relatively high complication rate (33.6% for the entire cohort). Although rhBMP-2 eliminates donor site morbidity, the surgeon should be aware of additional complications, such as radic-ulitis, osteolysis, and heterotopic bone formation, that may be associated with its use. The authors failed to disclose if the assessment, clinical or radiographic, was performed in a blinded fashion. No validated outcome measures were used, and the fusion criteria were not adequately described. This study provides Level IV evidence supporting the use of rhBMP-2; however, more importantly, it highlights sev-eral of the more common complications thought to be as-sociated with the interbody application of rhBMP-2.

Pradhan et al. observed an increased rate of graft re-sorption, fracture, or collapse in patients undergoing stand-alone ALIF with femoral ring allograft.47 Lewandrowski et al. observed osteolysis of the vertebral endplate follow-ing minimally invasive TLIF and speculated that endplate violation during interbody decortication may have been a contributing factor.42 Although not specific for lumbar pro-cedures, Vaidya et al. observed a higher incidence of graft subsidence when rhBMP-2 was used with an allograft inter-body spacer.66 Joseph and Rampersaud observed a greater incidence of heterotopic bone formation following the use of rhBMP-2 for minimal access interbody lumbar fusion, but no clinical sequelae associated with this excessive bone formation were identified.34 Mindea et al. observed a high-er incidence of postoperative radiculitis of a nonstructural cause associated with the use of rhBMP-2 during minimal access TLIF procedures.43 Garrett et al. reported on the formation of painful postoperative seromas following the use of rhBMP-2 during posterolateral fusions.23 Finally, Carragee et al. identified a higher incidence of retrograde ejaculation in patients receiving rhBMP-2 during ALIF procedures.10 There have been a number of additional ret-rospective reviews and case series that have corroborated the findings from these reports.3,14,37,52

Although a direct cause and effect relationship be-tween the use of rhBMP-2 and these complications can-not be formulated based on these studies, the potential association should not be ignored. The surgeon should carefully consider the off-label utilization of rhBMP-2, or any osteobiologic, and make sure that the patient has been adequately informed regarding these risks.




rhBMP-7: Posterolateral Fusion. Osteogenic pro-tein–1, also known as rhBMP-7, is another member of the transforming growth factor–b superfamily and plays a key role in osteoblast differentiation. Compared with rhBMP-2, there have been relatively few clinical stud-ies investigating rhBMP-7 as an agent of spinal fusion, with the majority, if not all, focusing on posterolateral fu-sion techniques. The FDA has not approved the use of rhBMP-7 for spinal fusions. Currently, its use requires a humanitarian device exemption. There were no recom-mendations pertaining to the use of rhBMP-7 in the origi-nal guidelines publication (Table 6).48

Delawi et al. published a prospective multicenter RCT comparing the efficacy of local autograft supple-mented with either rhBMP-7 or AICB in single-level in-strumented posterolateral fusions for isthmic or degen-erative spondylolisthesis.18 Clinical and fusion assessment was performed at 6 weeks, and 3, 6, and 12 months after surgery. Fusion assessment through CT imaging was per-formed in a blinded fashion. Validated clinical measures (ODI and VAS) were used to determine response to sur-gery. The follow-up rate at 12 months was 89%. There was no statistical difference in the fusion rate between the groups (63% in the rhBMP-7 group and 67% in the AICB cohort). Both groups demonstrated clinical improvement compared with baseline scores, but there was no statisti-cal difference regarding clinical outcome between groups. At 12 months after surgery, 64% of AICB patients com-plained of at least “mild” donor site pain (2.7 ± 2.8 VAS). No specific adverse event was related to AICB harvesting or use of rhBMP-7. The authors concluded that rhBMP-7 is an effective alternative to AICB for supplementing lo-cal autograft for single-level posterolateral fusions. The small sample size (< 50 patients), incomplete statistical analysis, and potential impact of differences in baseline characteristics requires that the study be downgraded to Level II evidence in support of utilizing rhBMP-7/local autograft as a substitute for AICB/local autograft in sin-gle-level instrumented posterolateral fusions.

In 2008, Vaccaro et al. conducted a multicenter pro-spective RCT to further investigate the safety and effi-cacy of rhBMP-7/ACS and to demonstrate noninferiority as a replacement for AICB for noninstrumented, single-level posterolateral fusion.62 Three hundred thirty-five patients were randomized in a 2:1 fashion, but only 293 were treated (208 patients received rhBMP-7/ACS and 87 received AICB). Independent blinded clinical and radio-graphic evaluations were performed at 6 weeks and at 3, 6, 12, 24, and longer than 36 months utilizing validated outcome measures, including ODI, SF-36, and VAS, and radiographs. Fusion assessment after 36 months was sup-plemented with CT scans. The primary overall success was reported as a composite measure, intended for FDA submission, and required a 20% improvement in ODI, ab-sence of treatment-emergent adverse events related to the device, absence of a decline in neurological status, and ra-diographic successful fusion. At 24 months with a follow-up rate of 87%, the investigational group did not achieve statistical equivalence with respect to the overall success rate compared with controls (38.7% compared with 49.4% [p = 0.33]). The investigational group demonstrated a TA

BLE

5: C

ompl

icat

ions

and

the u

se o

f rhB

MP-2

: sum

mar

y of e

viden

ce*

Auth

ors &

Ye

arLe

vel o

f Ev

idenc

eDe

scrip

tion o

f Stu

dyCo

mmen

ts

Rihn

et al

.,

2009

50IV

This

retro

spec

tive c

ohor

t stu

dy w

as in

tende

d to e

valua

te th

e com

plica

tions

asso

ciated

w/ 1

-leve

l TLIF

proc

e-

dures &

deter

mine if comp

lications d

iffered b

twn p

ts receiving AICB or rh

BMP-2. 119 o

f 130 pts w

ere a

vail-

able for follow

-up, 33 re

ceiving au

tograft &

86 r

hBMP-2. After

2006, a hy

drogel seala

nt wa

s used in 3

7 pts to

shield the ne

rve r

oot from exposure to the r

hBMP-2. Fusio

n assessm

ent, through a comb

ination o

f radio-

grap

hs &

CT

scan

s, wa

s per

form

ed on

aver

age a

t 19.1

mos

& cl

inica

l follo

w-up

at 27

.6 mo

s afte

r sur

gery.

The fusion

rate for the au

tolog

ous c

ohort w

as 96.5%

& it wa

s 97%

for the rh

BMP-2 c

ohort (

p = 0.09). A

highe

r com

plica

tion r

ate w

as ob

serv

ed in

the a

utogr

aft c

ohor

t (45

.5% vs

29.1%

), bu

t the d

iffere

nce w

as no

t

signifi

cant. Donor site mo

rbidity wa

s the most com

mon c

omplication in the au

tolog

ous c

ohort. P

ostop

radi-

culitis w

as more c

ommo

nly ob

served in the r

hBMP-2 c

ohort (1

4% vs

3% [p = 0.08]); ho

wever, this

incid

ence

sig

nificantly d

ecreased after

the a

pplication of the h

ydrogel sealan

t (20.4%

to 5.4%

[p = 0.047]). Ad

dition

al

comp

lications thought to be re

lated to the u

se of rh

BMP-2 includ

ed os

teoly

sis & he

teroto

pic bo

ne form

ation.

Th

e aut

hors

conc

luded

that

the T

LIF pr

oced

ure,

rega

rdles

s of g

raft,

is as

socia

ted w

/ a re

lative

ly hig

h com

-

plication r

ate (33.6%

for the en

tire c

ohort), & although rh

BMP-2 e

limina

tes do

nor site morbid

ity, additio

nal

co

mplic

ation

s suc

h as r

adicu

litis,

oste

olysis

, & he

teroto

pic bo

ne fo

rmati

on ar

e ass

ociat

ed w

/ its u

se.

The s

tudy

is lim

ited d

ue to

its re

trosp

ectiv

e des

ign &

failu

re

to

desc

ribe i

f the a

sses

smen

t of c

ompli

catio

ns &

fusio

n

statu

s was

perfo

rmed

in a

blind

ed fa

shion

. No v

alida

ted

ou

tcome

instr

umen

ts we

re us

ed. T

he cr

iteria

for f

usion

asse

ssme

nt we

re no

t ade

quate

ly de

scrib

ed, &

it is

not

cle

ar if

simila

r meth

ods w

ere p

erfo

rmed

equa

lly fo

r both

study

grou

ps. T

his re

trosp

ectiv

e coh

ort s

tudy

is do

wn-

gr

aded

from

Leve

l III to

Leve

l IV w

/ res

pect

to th

e com

pli-

cations as

socia

ted w/ th

e use of rh

BMP-2 in T

LIF

pr

oced

ures

.

* AICB

= au

tolog

ous iliac c

rest bone; pt = pa

tient; rh

BMP-2 =

recomb

inant huma

n bone m

orphogenetic p

rotein–2; TL

IF = transfo

raminal lu

mbar inter

body fusio

n.


M. G. Kaiser et al.


TABL

E 6:

rhBM

P-7 i

n po

ster

olat

eral

fusio

n: su

mm

ary o

f evid

ence

*

Auth

ors &

Ye

arLe

vel o

f Ev

idenc

eDe

scrip

tion

Comm

ents

Delaw

i et a

l.,

20

10II

The o

bjecti

ve of

this

pilot,

pros

pecti

ve, r

ando

mize

d, co

ntroll

ed m

ultice

nter t

rial w

as to

comp

are t

he

efficacy o

f rhB

MP-7 w

/ AICB, bo

th supplem

ented

w/ lo

cal bone, in 1-lev

el ins

trume

nted p

oster

olat

fus

ions f

or is

thmic

or de

gene

rativ

e spo

ndylo

listhe

sis. C

linica

l & fu

sion a

sses

smen

ts we

re pe

rform

ed

at

6 wks

& 3,

6, &

12 m

os af

ter su

rger

y. Fu

sion a

sses

smen

t thro

ugh C

T im

aging

was

perfo

rmed

in a b

linde

d fas

hion.

Valid

ated c

linica

l mea

sure

s (OD

I & V

AS) w

ere u

sed t

o dete

rmine

resp

onse

to

su

rger

y. Th

e foll

ow-u

p rate

at 12

mos

was

89%

. The

re w

as no

stati

stica

l diffe

renc

e in t

he fu

sion

rate btwn

the g

roups (63% in the r

hBMP-7 g

roup & 67

% in the A

ICB cohort). B

oth gr

oups de

m-

on

strate

d clin

ical im

prov

emen

t com

pare

d w/ b

aseli

ne sc

ores

, but

ther

e was

no st

atisti

cal d

iffere

nce

re

gard

ing cl

inica

l outc

ome b

twn g

roup

s. At

12 m

os af

ter su

rger

y, 64

% of

AIC

B pts

comp

laine

d of a

t

least “m

ild” d

onor site pain (2.7 ± 2.8 V

AS). No

specific

adverse e

vent wa

s rela

ted to AICB harvest-

ing

or us

e of rhB

MP-7. Th

e authors co

nclud

ed that rhBM

P-7 is a

n effe

ctive alter

native to A

ICB

when

supp

lemen

ting l

ocal

auto

graf

t for 1

-leve

l pos

terola

t fus

ions &

avoid

s the

mor

bidity

asso

ciated

w/ A

ICB

harv

estin

g.

This

study

popu

lation

is re

lative

ly sm

all (<

50 pt

s), bu

t rep

rese

nts

a h

omog

eneo

us gr

oup o

f pts.

The

rand

omiza

tion p

roto

col is

adeq

uatel

y des

cribe

d & th

e tre

ating

surg

eon w

as bl

inded

as

be

st as

poss

ible r

egar

ding i

nterv

entio

n. De

spite

rand

omiza

-

tion t

here

was

a dif

feren

ce bt

wn gr

oups

w/ r

espe

ct to

the s

pinal

se

gmen

t und

ergo

ing fu

sion.

Alth

ough

stati

stics

wer

e ad-

equately d

escribe

d, it is d

ifficult to d

eterm

ine the s

ignific

ance

due to the sm

all sa

mple siz

e & no

confidence inte

rvals

were

ap

plied

. The

small

samp

le siz

e, inc

omple

te sta

tistic

al an

alysis

,

& po

tentia

l impa

ct of

differ

ence

s in b

aseli

ne ch

arac

terist

ics

re

quire

that

the s

tudy

be do

wngr

aded

to Le

vel II

evide

nce i

n

support of rhB

MP-7 a

s a su

bstitu

te for A

ICB to supplem

ent

loc

al au

togr

aft in

1-lev

el ins

trume

nted p

oster

olat f

usion

s.Va

ccar

o et a

l.,

2008

62II

The p

urpo

se of

the p

rosp

ectiv

e ran

domi

zed c

ontro

lled m

ultice

nter s

tudy w

as to

deter

mine

the s

afety

&

efficacy o

f rhB

MP-7/AC

S & to demo

nstrate nonin

feriority as

a repla

ceme

nt for AICB for un

instru

-

mente

d, 1-

level

poste

rolat

fusio

n. 33

5 pts

were

rand

omize

d, bu

t only

293 w

ere t

reate

d. 20

8 pts

re-

ceive

d rhB

MP-7 & 87 co

ntrol pts

receive

d AICB. Independent blinded c

linica

l & radio

graphic

evalu

-

ation

s wer

e per

forme

d at 6

wks

& at

3, 6,

12, 2

4, &

>36 m

os af

ter su

rger

y usin

g vali

dated

outco

me

me

asur

es &

radio

grap

hs. F

usion

asse

ssme

nt af

ter 36

mos

was

supp

lemen

ted w

/ CT

scan

s. At

the

24

-mo e

nd po

int th

e inv

estig

ation

al gr

oup d

id no

t ach

ieve s

tatist

ical e

quiva

lence

w/ r

espe

ct to

the

overall su

ccess rate

comp

ared w/ controls (38.7

% vs

49.4%

[p = 0.3

3]). The investigational group

de

mons

trated

a low

er fu

sion r

ate (6

1.7%

vs 83

.1%). N

oninf

erior

ity of

the o

vera

ll suc

cess

was

dem-

onstrate

d after 36 m

os (47.2

% in the investigational group &

46.8%

in the c

ontrol cohort [p

= 0.0

25]).

CT

evalu

ation

demo

nstra

ted a

statis

ticall

y gre

ater p

erce

ntage

of pt

s in t

he co

ntrol

arm

demo

nstra

ting

bridg

ing bo

ne ac

ross the inte

rtransverse sp

ace (83% vs

56% [p = 0.0

01]). There w

as no

t a statistically

sig

nificant difference in the rate

of treatment-rela

ted se

rious ad

verse

events btw

n the investigational

& control groups (85.6

% & 84

.7%, respectively [p = 0

.863]). Do

nor site pa

in wa

s reported

in 44

% of

co

ntrols

at 12

mos

& 35

% af

ter 36

mos

, w/ V

AS sc

ores

of 1.

6 at 1

2 mos

& 1.

1 afte

r 36 m

os. O

p tim

e

& blo

od loss were s

ignific

antly less in the investigational cohort. No s

ignific

ant im

munolog

ical re

ac-

tion rela

ted to the a

pplication of rh

BMP-7 w

as recorded. The au

thors conclud

ed that rhBM

P-7/c

olla-

gen c

omposite is a

safe & effec

tive a

lternative to IC

BG.

This

is a w

ell-d

esign

ed la

rge p

rosp

ectiv

e ran

domi

zed t

rial b

ut

se

vera

l limi

tatio

ns ex

ist. A

lthou

gh th

e aut

hors

perfo

rmed

a

powe

r calc

ulation to d

eterm

ine sa

mple siz

e, the b

enefits of ran-

domi

zatio

ns w

ere c

ompr

omise

d sinc

e 13%

of pt

s wer

e not

treate

d & th

e aut

hors

faile

d to d

escr

ibe th

ese p

ts. T

he fo

llow-

up ra

te at

24 m

os w

as 87

% of

pts t

reate

d (76

% of

pts r

ando

m-

ize

d), d

ropp

ing to

69%

(60%

of ra

ndom

ized p

ts) af

ter 3

6 mos

.

The a

ssessm

ent of overall s

uccess was modifie

d after 3

6 mos

co

mpar

ed w

/ the 2

4-mo

evalu

ation

. Due

to th

ese l

imita

tions

,

this stu

dy pr

ovide

s Level II e

vidence that at 24 mos, rhB

MP-7

did

not a

chiev

e a no

ninfer

ior ou

tcome

to A

ICB,

as m

easu

red

by the a

priori overall s

uccess criteria de

fined by

the a

uthors,

at

24 m

os co

mpar

ed w

/ AIC

B. T

he re

sults

wer

e com

para

ble

af

ter 3

6 mos

; how

ever,

this

conc

lusion

is co

mpro

mise

d due

to

the s

ignific

ant dropout rate.

(cont

inued

)




TABL

E 6:

rhBM

P-7 i

n po

ster

olat

eral

fusio

n: su

mm

ary o

f evid

ence

* (co

ntin

ued)

Auth

ors &

Ye

arLe

vel o

f Ev

idenc

eDe

scrip

tion

Comm

ents

Kana

yama

et

al.

, 200

6II

The o

bjecti

ve of

this

pros

pecti

ve R

CT w

as to

comp

are t

he cl

inica

l & ra

diogr

aphic

outco

mes o

f pts

unde

r-

going

1-lev

el po

stero

lat in

strum

ented

fusio

ns fo

r deg

ener

ative

spon

dylol

isthe

sis. 2

0 pts

were

rand

om-

ize

d into

2 groups, rhB

MP-7/collagen (n =

10) &

local auto

graft/H

A/TC

P (n = 1

0). Static & dynamic

ra

diogr

aphs

& C

T im

ages

wer

e obta

ined a

t 3 &

6 wk

s as w

ell as

at 3,

6, &

12 m

os to

asse

ss fu

sion

sta

tus. V

alida

ted cl

inica

l outc

ome m

easu

res (

ODI)

were

obtai

ned a

t 3, 6

, 9, &

12 m

os af

ter su

rger

y. Al

l

pts w

ho de

mons

trated

radio

grap

hic fu

sion u

nder

went

remo

val o

f instr

umen

tation

, reg

ardle

ss of

clini

cal

sta

tus. The follow-up rate at 1 y

r was 90

%. B

oth co

horts de

monstrated

signific

ant im

provem

ent in

the

OD

I scores a

t 3 mos after su

rgery; howe

ver, n

o sign

ificant diffe

rence w

as ob

served btwn

the g

roups.

It is n

ot cle

ar if the

signific

ance of clinical im

provem

ent w

as main

tained a

fter this

time p

oint. T

he

fus

ion rate based o

n radiog

raphic assessme

nt wa

s 78%

in the rhB

MP-7 cohort & 90% in the c

ontrol

group. Those p

ts conside

red fused (7 in the rhB

MP-7/collagen c

ohort &

9 controls

) underwe

nt surgica

l

explo

ration a

t an a

verage of 15

.3 mo

s after the index p

rocedure. S

olid a

rthrodesis

was identified in

57% of rh

BMP-7 p

ts explo

red (calcu

lated fusio

n rate

for all p

atients receiving rh

BMP-7/c

ollagen =

44

%) &

in 78

% of

the c

ontro

l gro

up (c

alcula

ted fu

sion r

ate fo

r enti

re co

ntrol

grou

p = 70

%). N

o stat

isti-

cal analys

is regarding

fusio

n rate

was pe

rform

ed. The au

thors conclud

ed that rhBM

P-7 w

as ca

pable

of ind

ucing

new bone grow

th howe

ver the fusio

n rate

was no

t encouragin

g & they su

ggested

modific

a-

tio

ns in

eithe

r sur

gical

techn

ique o

r car

rier m

ay be

requ

ired.

This is a s

mall p

t popula

tion, but it be

nefits

from

the h

omoge-

neou

s diag

nosis

. The

auth

ors f

ail to

desc

ribe t

he ra

ndom

iza-

tio

n sch

eme.

It is n

ot cle

ar if

the r

adiog

raph

ic as

sess

ment

was p

erfo

rmed

in a

blind

ed fa

shion

. The

auth

ors p

rovid

e

curs

ory b

aseli

ne de

mogr

aphic

data

. No p

ower

calcu

lation

was

perfo

rmed to de

termine sa

mple siz

e & a superficia

l description

of sta

tistic

al an

alysis

was

perfo

rmed

. The

intra

op de

termi

na-

tio

n of f

usion

statu

s is g

ener

ally c

onsid

ered

the “

gold

stand

ard”

for f

usion

asse

ssme

nt &

a clea

r adv

anta

ge in

this

study

. Due

to lim

itatio

ns th

is stu

dy w

as do

wngr

aded

to Le

vel II

evide

nce,

not supporting the u

se of rh

BMP-7 for 1-

level ins

trume

nted

po

stero

lat fu

sions

in de

gene

rativ

e spo

ndylo

listh

esis.

Vacc

aro e

t al.,

20

04

IITh

e pur

pose

of th

e pro

spec

tive r

ando

mize

d con

trolle

d mult

icente

r pilo

t stu

dy w

as to

deter

mine

the

safety &

efficacy o

f rhB

MP-7/A

CS as

a repla

ceme

nt for A

ICB for u

ninstrum

ented

poste

rolat fusio

n.

36

pts w

/ deg

ener

ative

spon

dylol

isthe

sis w

ere r

ando

mize

d & fo

llowe

d at 6

wks

& 3,

6, &

12 m

os

af

ter su

rger

y. Sa

fety w

as de

termi

ned b

y com

parin

g the

natur

e & fr

eque

ncy o

f adv

erse

even

ts.

Ra

diogr

aphs

wer

e ana

lyzed

by in

depe

nden

t rad

iolog

ists,

blind

ed to

the i

nterv

entio

n, &

valid

ated

outco

mes instru

ments

were u

sed to d

eterm

ine clinical effic

acy. Th

e follow

-up r

ate at 12

mos was

79

% fo

r the

inve

stiga

tiona

l gro

up &

83%

for t

he co

ntrols

. The

rate

of ad

vers

e eve

nts di

d not

differ

btwn

groups & no

specific

adverse e

vent wa

s rela

ted to the r

hBMP-7. Th

e fusion

rate for the

rhBM

P-7 c

ohort w

as 74

% & for the co

ntrol group 6

0% (p

= 0.675). T

he clinical success ra

te wa

s

86% for the rh

BMP-7 c

ohort &

73% for the co

ntrol group (

p = 0.39). Th

e authors co

nclud

ed that

rhBM

P-7 h

as an

accepta

ble sa

fety p

rofile

, & the c

omparable

results to AICB jus

tify furth

er investiga-

tion to d

efine the e

fficacy o

f rhB

MP-7 a

s a bo

ne gr

aft substitute.

This

is a w

ell-d

esign

ed pi

lot st

udy b

ut is

limite

d by t

he sm

all

sa

mple

size (

<50 p

ts) &

pts l

ost to

follo

w-up

at th

e stu

dy en

d

point

. No p

ower

calcu

lation

was

perfo

rmed

to de

termi

ne th

e

samp

le siz

e. Th

e 1-y

r end

point

may

also

be co

nside

red t

oo

br

ief fo

r fus

ion pr

oced

ures

. The

mea

sure

of ov

erall

clini

cal s

uc-

ce

ss w

as a

nonv

alida

ted co

mpos

ite sc

ore o

f both

fusio

n & cl

ini-

ca

l mea

sure

s. Du

e to t

hese

limita

tions

the s

tudy

was

down

-

graded to Le

vel II, finding

that the u

se of rh

BMP-7 a

s a bo

ne

gr

aft s

ubsti

tute i

n unin

strum

ented

poste

rolat

fusio

ns is

safe

&

effec

tive.

Vacc

aro e

t al.,

20

05II

The p

urpo

se of

this

study

was

to re

port

the l

ong-

term

follo

w-up

data

from

the p

revio

usly

repo

rted

pilot stu

dy de

scribed ab

ove. Pts w

ere follow

ed up

to 24

mos after

surgery u

sing the sa

me clinical &

ra

diogr

aphic

outco

me m

easu

res a

s pre

vious

ly re

porte

d. Th

e clin

ical &

radio

grap

hic ra

tes of

follo

w-

up

wer

e 86%

& 8

3%, r

espe

ctive

ly. T

he fu

sion r

ates f

or th

e con

trol &

inve

stiga

tiona

l coh

ort w

ere

40

% &

55%

, res

pecti

vely.

Clin

ical s

ucce

ss, w

/ res

pect

to th

e vali

dated

outco

mes i

nstru

ment,

was

85%

in th

e inv

estig

ation

al co

hort

& 64

% in

the c

ontro

l gro

up. N

o lon

g-ter

m ad

vers

e eve

nts w

ere

specific

ally rela

ted to the u

se of rh

BMP-7. Th

e authors co

nclud

ed that safety &

efficacy o

f rhB

MP-7

is

comp

arab

le to

AICB

for a

t leas

t 24 m

os af

ter su

rger

y.

The i

nitial

stud

y (fro

m 20

04) w

as do

wngr

aded

to Le

vel II

evide

nce

du

e to l

imita

tions

outlin

ed in

the c

omme

nts. A

t the 2

4-mo

time

period n

o additio

nal sign

ificant limitations were identified;

ther

efore

the s

tudy

main

taine

d a Le

vel II

desig

natio

n sup

port-

ing rh

BMP-7 a

s a bo

ne gr

aft substitute.

(cont

inued

)


M. G. Kaiser et al.


TABL

E 6:

rhBM

P-7 i

n po

ster

olat

eral

fusio

n: su

mm

ary o

f evid

ence

* (co

ntin

ued)

Auth

ors &

Ye

arLe

vel o

f Ev

idenc

eDe

scrip

tion

Comm

ents

Vacc

aro e

t al.,

20

0865

III

The p

urpo

se of

this

study

was

to re

port

the lo

ng-te

rm fo

llow-

up da

ta fro

m the

prev

iously

repo

rted p

ilot

stu

dy de

scribed ab

ove. Pts w

ere follow

ed up

to 48

mos after su

rgery u

sing the sa

me clinical & radio

-

grap

hic ou

tcome

mea

sure

s as p

revio

usly

repo

rted.

The c

linica

l & ra

diogr

aphic

rates

of fo

llow-

up w

ere

69

% &

61%

, res

pecti

vely.

The

fusio

n rate

s for

the c

ontro

l & in

vesti

gatio

nal c

ohor

t wer

e 50%

& 68

.8%,

re

spec

tively

. Clin

ical s

ucce

ss, w

/ res

pect

to the

valid

ated o

utcom

es in

strum

ent, w

as 73

.7% in

the i

n-

ve

stiga

tiona

l coh

ort &

57.1%

in th

e con

trol g

roup

. The

over

all su

cces

s was

62.5%

in th

e inv

estig

ation

al

cohort & 33.3%

in the c

ontrol group. N

o long-term adverse

events we

re sp

ecific

ally rela

ted to the u

se

of rhBM

P-7. The a

uthors c

onclu

ded that safe

ty & efficacy o

f rhB

MP-7 is comp

arable to AICB

for at

lea

st 48

mos

after

surg

ery.

The i

nitial

stud

y (fro

m 20

04) w

as do

wngr

aded

to Le

vel II

evide

nce

du

e to l

imita

tions

outlin

ed in

the c

omme

nts. F

urth

er lim

itatio

ns

we

re identified du

ring the follow-up pe

riod that sign

ificantly lim

it

conc

lusion

s dra

wn fr

om th

is stu

dy, in

parti

cular

the l

arge

no. o

f

pts lost to follow-up. D

ue to this sig

nificant lim

itation this

follow-

up re

port provide

s only

Level III evid

ence su

pporting r

hBMP-7

as

a bo

ne gr

aft s

ubsti

tute.

Vacc

aro e

t al.,

20

03IV

The p

urpose of this pilot stu

dy was to de

termine the s

afety of rhBM

P-7 c

ombin

ed w/ auto

graft fo

r

poste

rolat

unins

trume

nted f

usion

s in p

ts w/

symp

tomati

c deg

ener

ative

spon

dylol

isthe

sis. 1

2 pts

un-

derwent a 1-

level fusio

n receiv

ing rh

BMP-7 m

ixed w

/ AICB. Independent radiog

raphic assessme

nt

of

dyna

mic r

adiog

raph

s was

perfo

rmed

at 6

wks &

3, 6,

9, &

12 m

os, &

valid

ated o

utcom

es in

stru-

ments

, ODI

scor

es, w

ere u

sed t

o ass

ess c

linica

l outc

ome a

t 6 &

12 m

os. R

esult

s wer

e com

pare

d

w/ a

histor

ical c

ohor

t of p

ts wh

o only

rece

ived A

ICB.

75%

of pt

s ach

ieved

clini

cal s

ucce

ss &

55%

attaine

d a so

lid fusio

n. Th

ere w

ere n

o adverse ev

ents specific

ally a

ssociated w/ th

e use of

rhBM

P-7. Th

e authors co

nclud

ed that rhBM

P-7 d

emonstrate

d an a

cceptable

safety p

rofile

when

us

ed as

an ad

junct

to AI

CB.

The s

mall p

t pop

ulatio

n, co

mpar

ed w

/ a hi

storic

al co

hort,

&

re

lative

ly sh

ort fo

llow-

up fo

r a fu

sion p

roce

dure

limit t

his co

hort

stu

dy. T

he study d

esign

is diffic

ult to ca

tegorize b

ut falls btwn

a pro

spec

tive c

ase s

eries

& re

trosp

ectiv

e coh

ort s

tudy

. It w

as

ini

tially

cons

idere

d a Le

vel II

I stu

dy bu

t due

to th

e lim

itatio

ns

wa

s dow

ngra

ded t

o Lev

el IV

evide

nce s

uppo

rting

the s

afety

&

efficacy o

f rhB

MP-7 a

s a bo

ne gr

aft exte

nder for u

ninstrum

ent-

ed

1-lev

el po

stero

lat fu

sions

in pt

s w/ s

ympto

matic

spon

dylo-

listh

esis.

* AC

S = a

bsorbable

collagen s

ponge; AICB

= autolog

ous iliac c

rest bone; H

A = h

ydroxyapatite; IC

BG = ilia

c crest bone graft; O

DI = Os

westr

y Disa

bility Index; pt = p

atient; R

CT = random

ized c

ontro

lled

trial; rh

BMP-7 =

recomb

inant huma

n bone m

orphogenetic p

rotein–7; TC

P = tricalcium

phosphate

; VAS

= visual analo

g scale.




lower fusion rate (61.7% vs 83.1%). Noninferiority of the overall success was demonstrated after 36 months (47.2% in the investigational group and 46.8% in the control co-hort [p = 0.025]). Computed tomography evaluation dem-onstrated a statistically greater percentage of patients in the control arm demonstrating bridging bone across the intertransverse space (83% vs 56% [p = 0.001]). However, at 36 months after surgery, only 69% of the original 293 patients were available for evaluation. There was no sig-nificant difference in the rate of treatment-related serious adverse events between the investigational and control groups, 85.6% and 84.7% respectively (p = 0.863). No sig-nificant immunological reaction related to the application of rhBMP-7 was recorded. The authors concluded that rh-BMP-7/collagen composite is a safe and effective alterna-tive to ICBG. This was a well-designed and executed trial; however, limitations exist. The randomization process was compromised as 13% of those originally randomized were not treated, and no data are provided regarding these pa-tients or those lost at final follow-up. The method of fu-sion assessment was altered at the final follow-up, with the addition of CT images, and a significant number of patients not available for evaluation (> 30%). Due to these limitations the study was downgraded to Level II evidence that rhBMP-7/ACS is noninferior to AICB for noninstru-mented posterolateral lumbar fusion.

Kanayama et al. conducted a prospective RCT to com-pare the clinical and radiographic outcomes in 20 patients with degenerative spondylolisthesis receiving either rh-BMP-7/ACS (n = 10) or local autograft/HA/TCP (n = 10) for single-level posterolateral instrumented fusions.35 Fu-sion status was assessed using plain radiographs and CT images at 3 and 6 weeks as well as at 3, 6, and 12 months after surgery. Validated clinical outcome measures (ODI) were obtained at 3, 6, 9, and 12 months after surgery. Re-gardless of clinical status, all patients underwent a second surgery to remove their instrumentation if a solid arthrod-esis was diagnosed based on radiographic imaging, on av-erage 15.3 months following the index procedure. At 1 year after surgery the follow-up rate was 90%. The ODI scores significantly improved at 3 months in both cohorts; how-ever, it is difficult to determine if the significance of this improvement was maintained beyond 3 months. The fu-sion rate based on radiographic assessment was 78% in the rhBMP-7 cohort; however, only 57% of these patients dem-onstrated a solid arthrodesis during direct surgical explora-tion. The radiographic fusion rate of the control group was 90%, with 78% of controls demonstrating a fusion at the time of implant removal. No statistical analysis regarding fusion rate was performed. The authors concluded that uti-lization of rhBMP-7 was feasible, but the observed fusion rate was not encouraging, suggesting that modifications of the surgical technique or carrier were required. The study cohort was small yet homogeneous with respect to surgical procedure and presenting diagnosis. It is not clear if the ra-diographic assessment was performed in a blinded fashion; however, confirmation of fusion through direct operative exploration is considered the gold standard for fusion as-sessment. Limitations of the study design include failure to describe the randomization scheme or perform a power calculation to determine sample size. The authors failed

to perform an adequate statistical analysis; this may be a secondary consequence of the small sample size. Despite the randomized nature of this study, the study was down-graded to Level II evidence suggesting that rhBMP-7 is an inadequate substitute for AICB in instrumented posterolat-eral fusion.

Vaccaro et al. published 3 separate studies over a 4-year period reporting the radiographic and clinical results from a prospective randomized controlled multicenter clinical pilot study investigating the efficacy and safety of OP-1 compared with AICB in noninstrumented posterolateral lumbar fusions.61,64,65 The original pilot study, published in 2004, randomized 36 patients with degenerative spondy-lolisthesis to receive either rhBMP-7/ACS or AICB.64 The initial study followed patients at 6 weeks and at 3, 6, and 12 months after the index procedure. An independent blinded radiologist evaluated plain radiographs to assess fusion and validated outcome measures; ODI and SF-36, were used to assess clinical status. At 12 months after surgery, the follow-up rate was 79% for the rhBMP-7 cohort and 83% for the control group. No short-term adverse events directly related to the use of rhBMP-7 were reported. At 12 months after surgery, the fusion and clinical success rates were 74% and 86%, respectively, for the rhBMP-7 co-hort and 60% and 73%, respectively, in the control group, with no statistically significant difference between groups. From this initial study the authors concluded that rhBMP-7 has an acceptable safety profile and comparable results to AICB to justify further investigation. This initial study was limited by the lack of a power calculation, a small sample size (< 50 patients), and significant loss to follow-up within the interventional group. A nonvalidated overall clinical outcome composite score is included that is difficult to ob-jectify. Given these limitations, and those inherent with a pilot study, the initial publication is downgraded to Level II evidence in support of an acceptable safety profile and comparable efficacy of rhBMP-7 to AICB.

The same authors published 2 follow-up studies in 2005 and 2008 to report the 2- and 4-year outcomes from the same study population.61,65 As the initial report provid-ed Level II evidence, the subsequent reports were started at this level and downgraded further if additional limitations were identified. At the 24-month follow-up end point, the follow-up rate for the investigational and control groups were 86% and 83%, respectively.61 Radiographic fusion occurred in 55% of patients receiving OP-1 and 40% of AICB patients. Clinical success was recorded in 85% of OP-1 patients and 64% of control patients. No additional limitations were identified in this study; therefore, a Level II designation is maintained. However, at 48 months only 69% and 61% were available for clinical and radiographic evaluations, respectively.65 The fusion and clinical success rates were reported as 68.8% and 73.7% in the rh-BMP-7 group and 50% and 57.1% in the control cohort. Due to the small number of patients available at 48 months, formal statistical analysis was not performed. The authors con-cluded that rhBMP-7 had an acceptable safety profile and comparable results to AICB. Due to the additional attrition of patients at the 48-month follow-up time point, this study was downgraded to Level III evidence in support of com-parable efficacy between rhBMP-7 and AICB.


M. G. Kaiser et al.


In 2003, Vaccaro et al. conducted a pilot study to de-termine the safety of rhBMP-7 combined with AICB for posterolateral uninstrumented fusions in patients with symptomatic degenerative spondylolisthesis.63 Seventy-five percent of patients achieved clinical success and 55% at-tained a solid fusion. There were no adverse events spe-cifically associated with the use of rhBMP-7. The authors concluded that rhBMP-7 demonstrated an acceptable safe-ty profile when used as an adjunct to AICB. The small pa-tient population, comparison with a historical cohort, and relatively short follow-up for a fusion procedure limit the study. Due to these limitations the study is downgraded to Level IV evidence. As this was the only study investigating this specific application of rhBMP-7, there is insufficient evidence to formulate a recommendation.

SummaryA wide variety of bone graft extenders and substitutes

are currently available. Enhanced fusion rates and the ability to avoid complications associated with iliac crest harvesting are the intended benefits of their use. Many, if not all, of these extenders and substitutes evaluated in this review have demonstrated a positive effect on fusion rate with clinical outcomes comparable to AICB. Convincing evidence exists that calcium-based composites cannot be considered substitutes for AICB due to inferior fusion rates. These materials, along with allograft-derived grafts (DBM), function primarily as extenders, requiring some form of autograft to achieve adequate fusion rates. There has been little if any risk associated with the use of these extenders.

Bone morphogenetic proteins have dramatically al-tered the landscape of spinal fusion surgery. These pow-erful osteoinductive agents have demonstrated excellent potential as substitutes for AICB with both interbody and posterolateral fusions. The vast majority of investigations have evaluated the effect of rhBMP-2. Although rhBMP-2 has shown a positive effect on fusion rate, complications have been reported related to its use. As a result, careful consideration is required when utilizing these products.

Despite the beneficial effect on fusion, the current literature has also not adequately addressed the issue of whether these improved fusion rates justify the cost, es-pecially for treatment of routine degenerative lumbar dis-ease. Although it is likely that certain patient populations would benefit from the addition of BMPs when perform-ing spinal fusion surgery, the current literature has failed to adequately identify such patient populations.

Key Issues for Future InvestigationThere has already been an extensive amount of re-

search investigating the potential impact of these graft extenders and substitutes. Further investigations should focus on improving study design to validate the conclu-sions formulated from previous publications and a com-prehensive evaluation of risks and complications will be necessary to properly inform our patients. Identification of patient populations at risk for pseudarthrosis would also better define patient populations where the benefits of utilizing BMPs justify the risks. Potentially more rel-

evant than defining the clinical impact of these materi-als is to determine their cost utility. Comprehensive cost analyses, not simply a superficial quantification of upfront costs, will ultimately be required. Such an endeavor will require the concerted effort of a multidisciplinary panel of experts from the clinical, epidemiological, and admin-istrative disciplines.

Acknowledgments

We would like to acknowledge the AANS/CNS Joint Guide-lines Committee (JGC) for their review, comments, and sugges-tions; Laura Mitchell, CNS Guidelines Project Manager, for her organizational assistance; and Linda O’Dwyer, medical librarian, for assistance with the literature searches. We would also like to acknowledge the following individual JGC members for their con-tributions throughout the review process: Timothy Ryken, M.D.; Kevin Cockroft, M.D.; Sepideh Amin-Hanjani, M.D.; Steven N. Kalkanis, M.D.; John O’Toole, M.D., M.S.; Steven Casha, M.D., Ph.D.; Aaron Filler, M.D., Ph.D., F.R.C.S.; Daniel Hoh, M.D.; Steven Hwang, M.D.; Todd McCall, M.D.; Jeffrey J. Olson, M.D.; Julie Pilitsis, M.D., Ph.D.; Joshua Rosenow, M.D.; and Christopher Winfree, M.D.

Disclosure

Administrative costs of this project were funded by the Con-gress of Neurological Surgeons and the Joint Section on Disorders of the Spine and Peripheral Nerves of the American Association of Neurological Surgeons and Congress of Neurological Surgeons. No author received payment or honorarium for time devoted to this project. Dr. Ghogawala receives grants from the Patient Centered Outcomes Research Institute (PCORI) and the National Institutes of Health (NIH). Dr. Groff is a consultant for DePuy Spine and EBI Spine. Dr. Mummaneni owns stock in Spinicity and receives hono-raria from DePuy Spine and Globus and royalties from DePuy Spine, Quality Medical Publishers, and Thieme Publishing. Dr. Wang owns stock in Bone Biologics, AxioMed, Amedica, CoreSpine, Expand-ing Orthopedics, Pioneer, Syndicom, VG Innovations, PearlDiver, Flexuspine, Axis, FzioMed, Benvenue, Promethean, Nexgen, Elec-troCore, and Surgitech and holds patents with and receives royalties from Biomet, Stryker, SeaSpine, Aesculap, Osprey, Amedica, Syn-thes, and Alphatec. The authors report no other potential conflicts of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author contributions to the study and manuscript preparation include the following. Acquisition of data: all authors. Analysis and interpretation of data: all authors. Drafting the article: Kaiser. Criti-cally revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Kaiser. Study supervision: Kaiser.

References

1. Acharya NK, Kumar RJ, Varma HK, Menon VK: Hydroxyap-atite-bioactive glass ceramic composite as stand-alone graft substitute for posterolateral fusion of lumbar spine: a prospec-tive, matched, and controlled study. J Spinal Disord Tech 21:106–111, 2008

2. Agarwal R, Williams K, Umscheid CA, Welch WC: Osteoin-ductive bone graft substitutes for lumbar fusion: a systematic review. Clinical article. J Neurosurg Spine 11:729–740, 2009

3. Balseiro S, Nottmeier EW: Vertebral osteolysis originating from subchondral cyst end plate defects in transforaminal lumbar interbody fusion using rhBMP-2. Report of two cases. Spine J 10:e6–e10, 2010

4. Banwart JC, Asher MA, Hassanein RS: Iliac crest bone graft harvest donor site morbidity. A statistical evaluation. Spine (Phila Pa 1976) 20:1055–1060, 1995




5. Burkus JK, Dorchak JD, Sanders DL: Radiographic assess-ment of interbody fusion using recombinant human bone mor-phogenetic protein type 2. Spine (Phila Pa 1976) 28:372–377, 2003

6. Burkus JK, Gornet MF, Dickman CA, Zdeblick TA: Anterior lumbar interbody fusion using rhBMP-2 with tapered inter-body cages. J Spinal Disord Tech 15:337–349, 2002

7. Burkus JK, Gornet MF, Schuler TC, Kleeman TJ, Zdeblick TA: Six-year outcomes of anterior lumbar interbody arthrod-esis with use of interbody fusion cages and recombinant hu-man bone morphogenetic protein-2. J Bone Joint Surg Am 91:1181–1189, 2009

8. Cammisa FP Jr, Lowery G, Garfin SR, Geisler FH, Klara PM, McGuire RA, et al: Two-year fusion rate equivalency between Grafton DBM gel and autograft in posterolateral spine fusion: a prospective controlled trial employing a side-by-side compari-son in the same patient. Spine (Phila Pa 1976) 29:660–666, 2004

9. Carragee EJ, Ghanayem AJ, Weiner BK, Rothman DJ, Bono CM: A challenge to integrity in spine publications: years of living dangerously with the promotion of bone growth factors. Spine J 11:463–468, 2011

10. Carragee EJ, Mitsunaga KA, Hurwitz EL, Scuderi GJ: Retro-grade ejaculation after anterior lumbar interbody fusion using rhBMP-2: a cohort controlled study. Spine J 11:511–516, 2011

11. Carter JD, Swearingen AB, Chaput CD, Rahm MD: Clinical and radiographic assessment of transforaminal lumbar inter-body fusion using HEALOS collagen-hydroxyapatite sponge with autologous bone marrow aspirate. Spine J 9:434–438, 2009

12. Chang CH, Lin MZ, Chen YJ, Hsu HC, Chen HT: Local au-togenous bone mixed with bone expander: an optimal option of bone graft in single-segment posterolateral lumbar fusion. Surg Neurol 70 Suppl 1:S47–S49, 2008

13. Chen CL, Liu CL, Sun SS, Han PY, Lee CS, Lo WH: Postero-lateral lumbar spinal fusion with autogenous bone chips from laminectomy extended with OsteoSet. J Chin Med Assoc 69:581–584, 2006

14. Chen NF, Smith ZA, Stiner E, Armin S, Sheikh H, Khoo LT: Symptomatic ectopic bone formation after off-label use of recombinant human bone morphogenetic protein-2 in trans-foraminal lumbar interbody fusion. Report of 4 cases. J Neu-rosurg Spine 12:40–46, 2010

15. Chen WJ, Tsai TT, Chen LH, Niu CC, Lai PL, Fu TS, et al: The fusion rate of calcium sulfate with local autograft bone compared with autologous iliac bone graft for instrumented short-segment spinal fusion. Spine (Phila Pa 1976) 30:2293–2297, 2005

16. Dai LY, Jiang LS: Single-level instrumented posterolateral fu-sion of lumbar spine with beta-tricalcium phosphate versus autograft: a prospective, randomized study with 3-year fol-low-up. Spine (Phila Pa 1976) 33:1299–1304, 2008

17. Dawson E, Bae HW, Burkus JK, Stambough JL, Glassman SD: Recombinant human bone morphogenetic protein-2 on an absorbable collagen sponge with an osteoconductive bulk-ing agent in posterolateral arthrodesis with instrumentation. A prospective randomized trial. J Bone Joint Surg Am 91:1604–1613, 2009

18. Delawi D, Dhert WJ, Rillardon L, Gay E, Prestamburgo D, Garcia-Fernandez C, et al: A prospective, randomized, con-trolled, multicenter study of osteogenic protein-1 in instru-mented posterolateral fusions: report on safety and feasibility. Spine (Phila Pa 1976) 35:1185–1191, 2010

19. Dimar JR, Glassman SD, Burkus KJ, Carreon LY: Clinical outcomes and fusion success at 2 years of single-level instru-mented posterolateral fusions with recombinant human bone morphogenetic protein-2/compression resistant matrix versus iliac crest bone graft. Spine (Phila Pa 1976) 31:2534–2540, 2006

20. Dimar JR II, Glassman SD, Burkus JK, Pryor PW, Hardacker JW, Carreon LY: Clinical and radiographic analysis of an op-timized rhBMP-2 formulation as an autograft replacement in posterolateral lumbar spine arthrodesis. J Bone Joint Surg Am 91:1377–1386, 2009

21. Epstein NE: Beta tricalcium phosphate: observation of use in 100 posterolateral lumbar instrumented fusions. Spine J 9:630–638, 2009

22. Epstein NE: Efficacy of different bone volume expanders for augmenting lumbar fusions. Surg Neurol 69:16–19, 2008

23. Garrett MP, Kakarla UK, Porter RW, Sonntag VKH: Forma-tion of painful seroma and edema after the use of recombinant human bone morphogenetic protein-2 in posterolateral lum-bar spine fusions. Neurosurgery 66:1044–1049, 2010

24. Geibel PT, Boyd DL, Slabisak V: The use of recombinant hu-man bone morphogenic protein in posterior interbody fusions of the lumbar spine: a clinical series. J Spinal Disord Tech 22:315–320, 2009

25. Glassman SD, Carreon L, Djurasovic M, Campbell MJ, Puno RM, Johnson JR, et al: Posterolateral lumbar spine fusion with INFUSE bone graft. Spine J 7:44–49, 2007

26. Glassman SD, Carreon LY, Djurasovic M, Campbell MJ, Puno RM, Johnson JR, et al: RhBMP-2 versus iliac crest bone graft for lumbar spine fusion: a randomized, controlled trial in pa-tients over sixty years of age. Spine (Phila Pa 1976) 33:2843–2849, 2008

27. Glassman SD, Dimar JR III, Burkus K, Hardacker JW, Pryor PW, Boden SD, et al: The efficacy of rhBMP-2 for posterolat-eral lumbar fusion in smokers. Spine (Phila Pa 1976) 32:1693–1698, 2007

28. Goldberg VM, Stevenson S: Natural history of autografts and allografts. Clin Orthop Relat Res (225):7–16, 1987

29. Haid RW Jr, Branch CL Jr, Alexander JT, Burkus JK: Poste-rior lumbar interbody fusion using recombinant human bone morphogenetic protein type 2 with cylindrical interbody cag-es. Spine J 4:527–539, 2004

30. Hamilton DK, Jones-Quaidoo SM, Sansur C, Shaffrey CI, Osk-ouian R, Jane JA Sr: Outcomes of bone morphogenetic pro-tein-2 in mature adults: posterolateral non-instrument-assisted lumbar decompression and fusion. Surg Neurol 69:457–462, 2008

31. Heary RF, Schlenk RP, Sacchieri TA, Barone D, Brotea C: Persistent iliac crest donor site pain: independent outcome as-sessment. Neurosurgery 50:510–517, 2002

32. Hsu CJ, Chou WY, Teng HP, Chang WN, Chou YJ: Coralline hydroxyapatite and laminectomy-derived bone as adjuvant graft material for lumbar posterolateral fusion. J Neurosurg Spine 3:271–275, 2005

33. Jenis LG, Banco RJ: Efficacy of silicate-substituted calcium phosphate ceramic in posterolateral instrumented lumbar fu-sion. Spine (Phila Pa 1976) 35:E1058–E1063, 2010

34. Joseph V, Rampersaud YR: Heterotopic bone formation with the use of rhBMP2 in posterior minimal access interbody fu-sion: a CT analysis. Spine (Phila Pa 1976) 32:2885–2890, 2007

35. Kanayama M, Hashimoto T, Shigenobu K, Yamane S, Bauer TW, Togawa D: A prospective randomized study of postero-lateral lumbar fusion using osteogenic protein-1 (OP-1) versus local autograft with ceramic bone substitute: emphasis of sur-gical exploration and histologic assessment. Spine (Phila Pa 1976) 31:1067–1074, 2006

36. Katayama Y, Matsuyama Y, Yoshihara H, Sakai Y, Nakamura H, Imagama S, et al: Clinical and radiographic outcomes of posterolateral lumbar spine fusion in humans using recombi-nant human bone morphogenetic protein-2: an average five-year follow-up study. Int Orthop 33:1061–1067, 2009

37. Knox JB, Dai JM III, Orchowski J: Osteolysis in transforami-nal lumbar interbody fusion with bone morphogenetic pro-tein-2. Spine (Phila Pa 1976) 36:672–676, 2011


M. G. Kaiser et al.


38. Korovessis P, Koureas G, Zacharatos S, Papazisis Z, Lambiris E: Correlative radiological, self-assessment and clinical anal-ysis of evolution in instrumented dorsal and lateral fusion for degenerative lumbar spine disease. Autograft versus coralline hydroxyapatite. Eur Spine J 14:630–638, 2005

39. Kurz LT, Garfin SR, Booth RE Jr: Harvesting autogenous iliac bone grafts. A review of complications and techniques. Spine (Phila Pa 1976) 14:1324–1331, 1989

40. Lanman TH, Hopkins TJ: Lumbar interbody fusion after treatment with recombinant human bone morphogenetic pro-tein-2 added to poly(L-lactide-co-D,L-lactide) bioresorbable implants. Neurosurg Focus 16(3):E9, 2004

41. Lee JH, Hwang CJ, Song BW, Koo KH, Chang BS, Lee CK: A prospective consecutive study of instrumented posterolateral lumbar fusion using synthetic hydroxyapatite (Bongros-HA) as a bone graft extender. J Biomed Mater Res A 90:804–810, 2009

42. Lewandrowski KU, Nanson C, Calderon R: Vertebral osteoly-sis after posterior interbody lumbar fusion with recombinant human bone morphogenetic protein 2: a report of five cases. Spine J 7:609–614, 2007

43. Mindea SA, Shih P, Song JK: Recombinant human bone mor-phogenetic protein-2-induced radiculitis in elective minimally invasive transforaminal lumbar interbody fusions: a series re-view. Spine (Phila Pa 1976) 34:1480–1485, 2009

44. Mummaneni PV, Pan J, Haid RW, Rodts GE: Contribution of recombinant human bone morphogenetic protein-2 to the rapid creation of interbody fusion when used in transforaminal lum-bar interbody fusion: a preliminary report. J Neurosurg Spine 1:19–23, 2004

45. Neen D, Noyes D, Shaw M, Gwilym S, Fairlie N, Birch N: Healos and bone marrow aspirate used for lumbar spine fu-sion: a case controlled study comparing healos with autograft. Spine (Phila Pa 1976) 31:E636–E640, 2006

46. Ong KL, Villarraga ML, Lau E, Carreon LY, Kurtz SM, Glassman SD: Off-label use of bone morphogenetic proteins in the United States using administrative data. Spine (Phila Pa 1976) 35:1794–1800, 2010

47. Pradhan BB, Bae HW, Dawson EG, Patel VV, Delamarter RB: Graft resorption with the use of bone morphogenetic protein: lessons from anterior lumbar interbody fusion using femoral ring allografts and recombinant human bone morphogenetic protein-2. Spine (Phila Pa 1976) 31:E277–E284, 2006

48. Resnick DK, Choudhri TF, Dailey AT, Groff MW, Khoo L, Matz PG, et al: Guidelines for the performance of fusion pro-cedures for degenerative disease of the lumbar spine. Part 16: bone graft extenders and substitutes. J Neurosurg Spine 2: 733–736, 2005

49. Rihn JA, Makda J, Hong J, Patel R, Hilibrand AS, Anderson DG, et al: The use of RhBMP-2 in single-level transforaminal lumbar interbody fusion: a clinical and radiographic analysis. Eur Spine J 18:1629–1636, 2009

50. Rihn JA, Patel R, Makda J, Hong J, Anderson DG, Vaccaro AR, et al: Complications associated with single-level trans-foraminal lumbar interbody fusion. Spine J 9:623–629, 2009

51. Schizas C, Triantafyllopoulos D, Kosmopoulos V, Tzinier-is N, Stafylas K: Posterolateral lumbar spine fusion using a novel demineralized bone matrix: a controlled case pilot study. Arch Orthop Trauma Surg 128:621–625, 2008

52. Shah RK, Moncayo VM, Smitson RD, Pierre-Jerome C, Terk MR: Recombinant human bone morphogenetic protein 2-in-duced heterotopic ossification of the retroperitoneum, psoas muscle, pelvis and abdominal wall following lumbar spinal fusion. Skeletal Radiol 39:501–504, 2010

53. Singh K, Smucker JD, Gill S, Boden SD: Use of recombinant human bone morphogenetic protein-2 as an adjunct in pos-terolateral lumbar spine fusion: a prospective CT-scan analy-sis at one and two years. J Spinal Disord Tech 19:416–423, 2006 (Erratum in J Spinal Disord Tech 20:185, 2007)

54. Slosar PJ, Josey R, Reynolds J: Accelerating lumbar fusions by combining rhBMP-2 with allograft bone: a prospective analy-sis of interbody fusion rates and clinical outcomes. Spine J 7: 301–307, 2007

55. Summers BN, Eisenstein SM: Donor site pain from the ilium. A complication of lumbar spine fusion. J Bone Joint Surg Br 71:677–680, 1989

56. Taghavi CE, Lee KB, Keorochana G, Tzeng ST, Yoo JH, Wang JC: Bone morphogenetic protein-2 and bone marrow aspirate with allograft as alternatives to autograft in instrumented revi-sion posterolateral lumbar spinal fusion: a minimum two-year follow-up study. Spine (Phila Pa 1976) 35:1144–1150, 2010

57. Urist MR: Bone: formation by autoinduction. Science 150:893–899, 1965

58. US Food and Drug Administration: InFUSE™ Bone Graft/LT-CAGE™ Lumbar Tapered Fusion Device - P000058. http://www.fda.gov/MedicalDevices/ProductsandMedical Procedures/DeviceApprovalsandClearances/Recently-Ap provedDevices/ucm083423.htm) [Accessed April 21, 2014]

59. US Food and Drug Administration: Listing of CDRH Hu-manitarian Device Exemptions. (http://www.fda.gov/Medical Devices/ProductsandMedicalProcedures/DeviceApprovals andClearances/HDEApprovals/ucm161827.htm) [Accessed April 21, 2014]

60. US Food and Drug Administration: OP-1 Putty - H0200008. (http://www.fda.gov/MedicalDevices/ProductsandMedical rocedures/DeviceApprovalsandClearances/Recently-App rovedDevices/ucm081181.htm) [Accessed April 21, 2014]

61. Vaccaro AR, Anderson DG, Patel T, Fischgrund J, Truumees E, Herkowitz HN, et al: Comparison of OP-1 Putty (rhBMP-7) to iliac crest autograft for posterolateral lumbar arthrodesis: a minimum 2-year follow-up pilot study. Spine (Phila Pa 1976) 30:2709–2716, 2005

62. Vaccaro AR, Lawrence JP, Patel T, Katz LD, Anderson DG, Fis-chgrund JS, et al: The safety and efficacy of OP-1 (rhBMP-7) as a replacement for iliac crest autograft in posterolateral lumbar arthrodesis: a long-term (>4 years) pivotal study. Spine (Phila Pa 1976) 33:2850–2862, 2008

63. Vaccaro AR, Patel T, Fischgrund J, Anderson DG, Truumees E, Herkowitz H, et al: A pilot safety and efficacy study of OP-1 putty (rhBMP-7) as an adjunct to iliac crest autograft in posterolateral lumbar fusions. Eur Spine J 12:495–500, 2003

64. Vaccaro AR, Patel T, Fischgrund J, Anderson DG, Truumees E, Herkowitz HN, et al: A pilot study evaluating the safety and ef-ficacy of OP-1 Putty (rhBMP-7) as a replacement for iliac crest autograft in posterolateral lumbar arthrodesis for degenerative spondylolisthesis. Spine (Phila Pa 1976) 29:1885–1892, 2004

65. Vaccaro AR, Whang PG, Patel T, Phillips FM, Anderson DG, Albert TJ, et al: The safety and efficacy of OP-1 (rhBMP-7) as a replacement for iliac crest autograft for posterolateral lum-bar arthrodesis: minimum 4-year follow-up of a pilot study. Spine J 8:457–465, 2008

66. Vaidya R, Weir R, Sethi A, Meisterling S, Hakeos W, Wybo CD: Interbody fusion with allograft and rhBMP-2 leads to consistent fusion but early subsidence. J Bone Joint Surg Br 89:342–345, 2007

67. Villavicencio AT, Burneikiene S, Nelson EL, Bulsara KR, Fa-vors M, Thramann J: Safety of transforaminal lumbar inter-body fusion and intervertebral recombinant human bone mor-phogenetic protein-2. J Neurosurg Spine 3:436–443, 2005

Manuscript submitted March 27, 2014.Accepted April 9, 2014.Please include this information when citing this paper: DOI:

10.3171/2014.4.SPINE14325.Address correspondence to: Michael G. Kaiser, M.D., Columbia

University, Neurological Surgery, The Neurological Institute, 710 W. 168th St., New York, NY 10032. email: [email protected].


guideline update for the performance of fusion procedures

Documents