7/24/2019 BF L2-5 1996

1/11

Copyright 1996 by

The Journal of Bone and Joint Surgery Incorporated

Burst Fractures of the Second through Fifth Lumbar Vertebrae

C L I N I C A L A N D R A D I O G R A P H I C R E S U L T S *

BY DAVID A. AND REYC HIK, M.D.F . DIRK H. ALAN DER, M.D.T KAROLYN M. SEN1CA, M.D. ,

AND E. SHANNON STAU FFER, M.D. , SPRINGFIE LD, ILLINOIS

nvestigation performed at the Southern llinois University School

of Medicine

Springfield

ABSTRACT: Fifty-five patie nts w ho had sustain ed

a burst fracture of the lumbar spine were followed for

a mean of seventy-nine months (range, twenty-four to

192 months) after the injury. Thirty patients had been

managed non-operatively with a short period of bed

rest followed by protected mobilization. The remaining

twenty-five patients had been managed operatively:

eight, with posterior arthrodesis with long-segment

hook-and-rod fixation; eight, with posterior arthrodesis

with short-segment transpedicular fixation; six, with

posterior arthrodesis and instrumentation followed by

anterior decompression and arthrodesis; and three,

with anterior decompression and arthrodesis.

Thirty-six patients had bee n neur ological } intact

at the time of presentation and had remained so

throughout the follow-up period. No neurological de

terioration or symptoms of late spinal stenosis were

seen .

Isolated partial single-nerve-root deficits re

solved regardless of the m ethod of treatment. Patients

who had had a complete single or a multiple-nerve-root

paralysis seemed to have benefited from anterior de

compression.

Although the anatomical results as seen on the

most recent radiographs were superior for the group

that had been managed operatively with long posterior

fixation or anterior and posterior arthrodesis, the most

recent pain scores and the functional outcomes were

similar for all treatment groups. At the latest follow-up

evaluation, some loss of spinal alignment was noted in

the patients who had been managed with short trans-

pedicular fixation; the alignment at the most recent

follow-up examination was comparable with that in

the patients who had been managed non-operatively.

For the patients who had had non-operative treat

ment, we were unable to predict the deformity at the

time of follow-up on the basis of the initial diagnos

tic radiographs. The clinical outcome was not related

*No benefits in any form have been received or will be received

from a commercial party related directly or indirectly to the subject

of this article. No funds were received in support of this study.

tGeisinger Clinic, 100 North Academy, Danville, Pennsylvania

17822.

^Dickson Diveley Midwest Orthopaedic Clinic, 4320 Wornall

Road, Suite

610,

Kansas City, Missouri64111.

Division of Orthopaedics and Rehabilitation, Southern Illinois

University School of Medicine, P.O. Box 19230, Springfield, Illinois

62794-9230.

to the deformity at the latest follow-up evaluation.

On the basis of the results of our study, we recom

mend non-operative treatment for patients who do not

have neurological dysfunction or who have an isolated

partial nerve-root deficit at the time of presentation.

For patients who have a multiple-nerve-root paralysis,

anterior decompression is indicated.

T h e r e h a s b e e n c o n s i d e r a b l e c o n t r o v e r s y r e g a r d

i n g w h a t c o n s t i t u t e s t h e b e s t t r e a t m e n t f o r l u m b a r b u r s t

f r a c t u re s . P r o p o n e n t s o f n o n - o p e r a t i v e t r e a t m e n t h a v e

c l a im ed th a t , ex cep t in p a t i e n t s wh o h av e a l a rg e r r es id

u a l k y p h o s i s , t h e c li n i ca l o u t co m e co m p are s f av o rab ly

w i t h t h a t of o p e r a t i v e t r e a t m e n t , w i t h o u t th e a t t e n d a n t

r isk of an operat ive procedure

91 1

'

5

-

2627

'

3438454652

. H o w e v e r ,

a d v o c a t e s o f s t a b i l i z a t i o n p r o c e d u r e s h a v e c i t e d b e t t e r

r e s t o r a t i o n o f a n a t o m i c a l a l i g n m e n t , m o r e r a p i d m o b i

l i z a ti o n , a n d i m p r o v e d n e u r o l o g i c a l o u t c o m e s w i t h o p

erative intervention

2

'

4

'

5

-

7

'

10121417

'

19

-

23

-

25

'

28

'

32

-

33

'

37

'

39

-

40

-

42

'

43

'

4748

. A fear

o f n e u r o l o g i c a l d e t e r i o r a t i o n , o r t h e b e l i e f t h a t o p e r a

t i v e r e a l i g n m e n t a n d a r t h r o d e s i s p r e v e n t l a t e s y m p

to m at i c j o in t d eg en e ra t i o n an d sp in a l s t en o s i s , h as l ed

s o m e a u t h o r s t o r e c o m m e n d o p e r a t i v e r a t h e r t h a n n o n -

operative treatment

3

'

5

'

8

'

10

-

12

-

14

-

17

-

23

-

28

-

33

-

36

'

39

'

44

'

47

-

48

'

50

'

53

.

P r e v i o u s r e p o r t s h a v e d o c u m e n t e d c o n f l i c t i n g r e

s u l ts of n o n - o p e r a t i v e t r e a t m e n t o f p a t i e n t s w h o d o

not have a neuro logical def ici t

1

-

11

-'

7

-

26

-

38

-

44

-

52

. De n i s e t a l .

r ep o r t ed h ig h r a t es o f f a i l u re i n t h e fo rm o f sev ere d e

fo rm i ty , an i n c reas in g n eu ro lo g i ca l d e f i c i t , an d i n ca

p a c i t a t i n g p a i n

17

. T h e y c o n c l u d e d t h a t p a t i e n t s w h o

h a d b e e n m a n a g e d o p e r a t i v e l y f a r e d m u c h b e t t e r i n

t e r m s o f p a i n , d e f o r m i t y , r e t u r n t o w o r k , a n d n e u r o

l o g ic a l f u n c t i o n . M c E v o y a n d B r a d f o r d r e p o r t e d o n

t w e n t y - t w o p a t i e n t s w h o h a d i n i t i a l l y b e e n m a n a g e d

n o n - o p e r a t i v e l y

4 4

. A l t h o u g h n o n e u r o l o g i c a l d e t e r i o r a

t i o n was o b serv ed i n n eu ro lo g i ca l l y n o rm al p a t i en t s ,

t h e re wer e si x f a i l u res d u e t o an i n c rea s in g n e u ro lo g i

ca l d e f i c it i n p a r t i a l l y p a r a ly ze d p a t i en t s , i n c rea s in g d e

fo rm i ty , an d p er s i s t en t p a in .

D e s p i t e t h e s e p e s s i m i s t i c r e p o r t s , t h e r e i s a g r o w i n g

b o d y o f ev id en ce t h a t a b u r s t f r ac tu re wi th o u t a n eu ro

lo g i ca l d e f i c it can b e t r ea t e d n o n -o p era t i v e ly , wi th o n ly

r a r e n e u r o l o g i c a l d e t e r i o r a t i o n a n d w i t h a g o o d l o n g -

t e rm c l i n i ca l r esu l t . Wein s t e in e t a l .

52

, Ch an e t a l . , an d

M u m f o r d e t al .

45

r e p o r t e d t h e re s u l t s o f n o n - o p e r a t i v e

t r e a t m e n t o f t h o r a c o l u m b a r a n d l u m b a r b u r s t f r a c t u r e s .

1156

T H E J O U R N A L O F B O N E A N D J O IN T S U R G E R Y

7/24/2019 BF L2-5 1996

2/11

BURST FRAC TURE S OF THE SECOND THR OUG H FIFTH LUMBAR VERTE BRA E

1157

T A B L E I

DI S TRI BUTI ON OF THE-F IFT Y-FIVE PAT IE NT S AMONG THE F I V E T R E A T M E N T G R O U P S , A C C O R D I N G TO THE L E V E L OF THE F R A C T U R E

Level

of Fracture

G r o u pI:

Non-Op .

T rea tmen t

G r o u pII:

Posterior

Arthrodesis

(Long)

G r o u p

III:

Posterior

Arthrodesis

(Short)

G r o u pIV:

Ante r io rand

Posterior

Arthrodesis

G r o u p

V:

Anterior

Arthrodesis

Only

L2

L3

L4

L5

12

8

5

5

Of the 101pat ients in those three series, only one had

neurological deteriorat ion. All t h ree groups of investi

gators reported acceptable cl inical outcomes and the

lack of a relat ionship between the final anatomicalre

sultand the clinical sym ptoms.

A neurological defici t adds another component to

the process

of

making decisions regarding treatment .

No previous study, to our knowledge, has specifically

addressed the results of neurological injuries of the

cauda equina. Although neurological recovery

has

been

demonstrated after non-operat ive t reatment

of

t horaco

lumbar fractures

69153436

46

, there is current ly a t rend to

treat these injuries more intensively. Several reports

have documented improved neurological funct ion after

decompressive operative procedures

722

-

3742

.

The purpose

of the

presen t retrospect ive study

was

to compare the long-term radiographic and functional

resultsof t r ea t m en t of lumbar burst fractures with and

without operat ive decompression

and

arthrodesis .

aterials and ethod s

Seventy-four consecutive pat ientsinwhom alumbar

burst fractureat thesecond, third, fourth,orfifth lu mb ar

level

had

been t reated

at

Southern Illinois University

School of Medicine between 1976 and 1992 were iden

tified by a review of the records of the Or thopaedic

Spine Service.To be eligible for inclusion in the study,

the fracture had to have been non-pathological and

the pat ient had to have been fol lowed for more than

two years.Of the seventy-four patients,two haddiedof

causes unrelated to the injury of the spine, six were

contacted but refused to part icipate in the study,and

eleven had been lost to follow -up. This left fifty-five

pat ientsfor thereview.

There were for ty-one male and fourteen female

pat ien t s . The m ean age was thirty-one years (range,

four teen

to

s eventy-one years ) .

The

mechanisms

of

injury included a motor-vehicle accident (twenty-nine

pat ients),a fal l (seventeen pat ients),a motorcycle acci

dent (four pat ients),a crush injury (four patients), and

a small-aircraft a ccident (o ne patie nt) . Twenty-five frac

tures were at the second lumbar vertebra; seventeen,

at the third lumbar vertebra; eight ,at the fourth lum

bar ver tebra; and five, at the f if th lumba r v erteb ra

(Table I).Th ere w ere forty-five associated injuries in

twenty-three pat ients . Twenty-one pat ientshad a frac

tureof anextrem ity; six,apelvic fracture ; five,aclosed

head injury; four,a chest injury; four, another injury of

the spine; three, a blunt abdominal injury; and two, a

burn . Of the four patients who had an additional in

jury of the spine, three had a compression fracture at

the first, third, or fourth lumbar level and the fourth

had a burst fracture at the first lumbar level.Thephys

ical and neurological status at the t ime of the injury

was assessed for each pat ienton thebasisof thephysi

cal examinat ion reports ,the repor ted motor- t rauma in

dex

4

,

and the

grade according

to the

classification

of

Frankel eta l.

27

.

Thirty pat ients were managed non-operat ively

(Gro u p I).Twenty pat ients were mana ged with a body

cast and seven, with a custom-molded thorac olumb o-

sacral orthosis; these patients were allowed out of bed

te n to fourteen days after the injury. The remaining

three pat ients were managed w ith mor e than four w eeks

ofbed rest because of an associated pelvic fracture.

Twenty-five pat ients were manage d operat ively Ta

ble I). Sixteen had posterior instrumentat ion and ar

throdesis: eight of t hem had long-segment (mo re than

two motion segm ents) hook-and-rod fixat ion (G rou p II) ,

and eight

had

short-segment

two

motion segments)

transpedicular fixat ion (Group III). An addi t ional six

pat ien t s (Group IV) had an ter ior and pos ter ior ar

throdesis . Three pat ients one who had a fracture at

the second lumbar leveland two who had a fracture at

the fourth lumbar level had anterior decompression

and arthrodesis only (Group V). No anterior metallic

implants were used. Autogenous bone graft was used

for all arthrodeses.It isnotewo rthy that onlyone frac

ture caudad to the third lumbar levelwas treated with

instrumentat ion.

Both

the

initial

and the

follow-up radiographs were

reviewed.Of thefifty-five pa tien ts, forty-eight hadcom

plete radiographic da ta. Becauseof an institutional pol

icy

of

destroying

the

rad iographs

of

patients w hose files

are inactive,the ini t ial radiographs were not available

fortheremaining seven pat ients . The init ial radiograp hs

had been made with

the

pat ient supine because

of the

acuteness of the injury. Radiographs at the follow-up

evaluat ion were made with the pat ient s tanding.An

t eropos ter ior

and

lateral radiographs, with

a

d is tance

of forty-four inches (111.8 centimeters) from the x-ray

tube to the film cassette, were made for all patients.

VOL. 78-A, NO. 8, AUG US T 1996

7/24/2019 BF L2-5 1996

3/11

1158

D.

A. ANDREYCHIK, D. H. ALANDER, K. M. SENICA, AND E. S. STAUFFER

No radiog raphs w ere ma de with the spine in flexion or

extension.

The fractures were classified according to the sys

tem of Denis

16

, and m easure men ts of kyphosis , com

pression, scoliosis, and the vertebral wedge index were

recorded. Kyphosis and scoliosis were measured with

the Cobb

13

method, with the superior end plate of the

vertebra cephalad to the fracture and the inferior end

plate of the vertebra caudad to the fracture used as

references. Vertebral body compression was calculated

with the method of Willen et al.

M

. Th e wedg e index w as

expressed as a ratio of the anterior height of the body

to the posterior height , as measured on the lateral ra

diograph. The ini t ial computerized tomography scans

that had been made before t reatment were avai lable

for review for thirty-seven of the fifty-five patients.

Compromise of the spinal canal was calculated with

the method of Mumford et al.

45

, with the intact verte

brae cephalad and caudad to the fracture used as refer

ences. The p ostoperat ive and fol low-up computerized

tomography scans were also reviewed when they were

avai lable. Magnetic resonance imaging was not rou

tinely used in this population of patients, largely be

cause of the time-period of the study. The success of

the arthrodesis and the subsequent fusion were not

specifical ly addressed for the groups that had been

managed operat ively because of the inherent inaccu

racy of plain radiographs in defining a solid fusion.

However, the instrumentat ion did not fai l in any pa

tient. Evidence of injury of a facet joint or another pos

terior element was also recorded. For consistency, all

rad iographic measurements were made by one or tho

paedic surgeon (D. A. A.) who had not been involved in

the ini tial t rea tmen t .

The most recent follow-up evaluation consisted of

an office visit, during which the patient responded to

a quest ionnaire regarding pain and funct ional s tatus

and on e of us (D . A. A., D. H. A., or K. M. S.) perfor med

a complete clinical and radiographic examination. This

assessment was performed at a mean of seventy-nine

mo nths (r ange , twenty-four to 192 mo nths ) after the

injury. Each patient was assigned a pain score on the

basis of the frequency of pain and the use of medica

tions. According to this scoring system, 0 indicates no

pain; 1 point, intermittent mild pain not necessitating

use of medication; 2 points, frequent mild pain necessi

tat ing occasional use of non-narcotic medicat ion; 3

points , mo dera te pain necessi tat ing frequent use of non

narcotic medication; 4 points, severe pain necessitating

occasional use of narcotic medication; and 5 points, pro

found pain necessitating regular use of narcotic medica

t ion. If a pat ient needed a reconstruct ive procedure

because of pain or instability, the score immediately

before that operation was used for analysis.

The functional rating was based on a comparison of

the pat ient s occupational and recreat ional s tatus before

the injury and th at after it . Ac cord ing to this rating

system, 0 indicates a retu rn to the patien t s pre-injury

occup ation w ith no limitation; 1 point, a retu rn to the

pre-injury occupation but minor occupational and rec

reational limitations; 2 points, a return to the pre-injury

occupation but major occupational and recreat ional

limitations; 3 points, a return to work but in a job that

is less strenuo us th an that before the injury; and 4 points,

an inability to perform productive work or to partici

pat e in recre ationa l activities. The patie nt s e mp loym ent

status before and after the injury was recorded with use

of a 0 to 5-point work-sta tus scale, deve loped by the

United States Department of Training and Employment

Service

51

, to deter min e the effect of previou s em ploy

ment status on the functional outcome. According to

this scale, 1 point ind icates heavy labor that involves

constant lifting of more than fifty pounds (22.7 kilo

grams); 2 points, moderately heavy labor that involves

constant lifting of more than twenty-five pounds (11.3

kilograms); 3 points, light labor that involves constant

lifting of less than ten pounds (4.5 kilograms); 4 points,

a sedentary job that involves sitting or standing without

lifting; and 5 points, unemployment.

Each patient was assigned to one of five treat

ment groups: Gro up I was managed with bed rest and

use of a cast; G rou p II, poste rior arthro desis w ith long-

segment instrumentat ion; Group III , posterior arthro

desis with short-segment instrumentat ion; Group IV,

anterior and p osterior arthrodesis; and G rou p V, an

terior dec om pressio n and arthro desis only. Th e initial

and most recently seen deformities of the lumbar spine

were compared among the t reatment groups, and any

relat ionships among the method of t reatment , the de

formity and pain at the time of follow-up, and the func

t ional outcome were analyzed. In the pat ients who had

been managed non-operat ively, the ini t ial and most re

cent radiographic pat terns were compared to determine

whether the ini t ial radiographic parameters had been

predictive of the radiographic deformity at the time of

follow-up. The effectiveness of the opera tive fixation in

obtaining and maintaining correction was also evalu

ated. The results for the operatively and non-operatively

treated groups were compared with regard to the most

rece nt radiogra phic, neurolog ical, pain, and functional

outcomes.

A one-way analysis of variance with five levels

was used to compare the means of the five t reatment

groups for each of the outcome measures. The Du ncan

multiple-range follow-up tests were used to make pair-

wise comparisons. A result was considered significant

if the p value was less than 0.05.

esults

eurological Outcome

Thirty-six patients had been neurologically intact at

the t ime of presentat ion; twenty-three had been man

aged non-operat ively and thirteen had had posterior

arthrodesis . No neurological deteriorat ion was observed

THE JOURN L OF BONE ND JOINT SURGERY

7/24/2019 BF L2-5 1996

4/11

BURST FRACTURES OF THE SECOND THROUGH FIFTH LUMBAR VERTEBRAE

.1159

T ABL E I I

D A TA O N TH E N I N ETEEN P A TI EN TS W H O H A D A N EU RO LO G I CA L D EF I CI T A T TH E TI M E O F P RES EN TA TIO N

Case

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

Level of

Fracture

L2

L3

L3

L4

L5

L5

L5

L3

L 2

L4

L2

L2

L2

L 2

L2

L 3

L3

L3

L4

Treatment

Non-op.

Non-op.

Non-op.

Non-op.

Non-op.

Non-op.

Non-op.

Post, arthrodesis, short

Ant. arthrodesis

Ant. arthrodesis

Post, arthrodesis, long

Post, arthrodesis, long

Ant. and post, arthrodesis

Ant. and post, arthrodesis

Ant. and post, arthrodesis

Ant. and post, arthrodesis

Ant. and post, arthrodesis

Ant. and post, arthrodesis

Ant. arthrodesis

Initial

D

D

D

B

D

D

D

D

D

D

D

B

D

D

D

A

A

A

C

Grade

27

A t

Follow-up

E

E

E

D

E

E

E

D

E

E

E

D

D

D

E

D

D

D

E

Motor-Trauma Index

41

Initial

99

98

99

95

96

96

99

96

95

97

78

50

90

88

71

58

63

51

64

A t

Follow-up

100

100

100

99

100

100

100

99

100

100

100

74

95

99

100

65

80

63

100

regardless of the method of t reatment , and there were

no symptoms of spinal stenosis at the latest follow-up

evaluat ion.

Nineteen patients had had neurological loss at the

time of pre sen tatio n Table II). Of these patients, ten

had had an isolated single-nerve-root deficit: eight had

a partial paralysis and two, a complete paralysis. Seven

of these pat ients s ix who had had a part ial and one wh o

had had a complete paralysis) had been managed non-

operatively. All six patients who had had a partial pa

ralysis t reated non-operat ively had complete motor

recovery, although minor sensory deficits persisted in

three. On e pat ient Case 4) who had been m anage d

non-operat ively had had a complete uni lateral paralysis

of the fourth lumbar nerve root; nearly complete re

covery was note d at the time of follow-up chang e in

motor-trauma index

41

, from 95 to 99). An othe r pat ient

Case 8) who had had a part ial paralysis had bee n m an

aged with posterior arthrodesis. At the time of the in

jury, the left quadriceps had been noted to be grade 1

of 5. The patie nt had inc om plete reco very at the latest

fol low-up examination change in motor-trau ma index,

from 96 to

99).

Two additional patien ts Cases 9 and 10),

one who had had a part ial paralysis and the other who

had had a complete s ingle-nerve-root lesion, had had

anterior decom pression and arthrode sis; both had com

plete recov ery at the latest follow-up visit change in

mo tor-traum a indices,from 95 to

1

and from

97

to 100).

Nine patients had had diffuse multiple-nerve-root

dysfunct ion. Two had been managed with posterior

arthrodesis and seven, with anterior and posterior ar

throdesis or anterior decompression and arthrodesis

alone. Of the two pat ients who had been managed pos

teriorly, one Case 11) had comp lete mo tor reco very

change in motor-trauma index, from 78 to 100). This

patient had diffuse weakness of both lower extremities,

although there was some motor function of all muscle

groups. Soon after placem ent of Har rington instrumen

tation, there was a rapid return of motor function, mak

ing an ant icipated anterior procedure unnecessary. The

second pat ient Case 12) had had a complete m otor

paralysis grad e B, as defined by Frankel et al.

27

) sec

ondary to a fracture of the second lumbar vertebra. She

was subsequently managed with Harrington instrumen

tat ion, but despi te a s trong recommendation for an an

terior procedure she refused to have.it. She had partial

recovery change in motor -traum a index, from 50 to 74)

but, at the latest follow-up evaluation, no motor or sen

sory function was observed distal to the quadriceps. At

the time of writing, she was able to walk about the

community with use of an ankle-foot orthosis.

The remaining seven pat ients who had had mult iple-

nerve-root dysfunction at the time of presentation had

had anterior and posterior arthrodesis or anterior de

compression and arthrodesis alone. Thre e of these pa

tients had had gra de-D status

27

; one Case 15) of the thr ee

had complete recovery change in motor-trau ma index,

from 71 to 100), and two Cases 13 and 14) had pa rtial

recove ry change in mo tor-tra um a indices, from 90 to 95

and from 88 to 99). Sensory deficits persisted in all thre e

pat ients . Four pat ients had had nearly complete motor

paralysis grade C in one and grade A in three ). The

patient who had had grade-C paralysis Case 19) had

complete motor recovery change in motor-trau ma in-

VOL. 78-A, NO. 8, AUGUST 1996

http://localhost/var/www/apps/conversion/tmp/scratch_7/have.ithttp://localhost/var/www/apps/conversion/tmp/scratch_7/have.it

7/24/2019 BF L2-5 1996

5/11

1160

D.

A. ANDREYCHIK, D. H. ALANDER, K. M. SENICA, AND E. S. STAUFFER

T ABL E I I I

DA TA O N K Y P H O S I S A N D CO M P RES S I O N , A CCO RDI N G T O TH E TY P E O F TREA TM EN T*

L2 fracture

N o.

of patients

Kyphosis degrees)

Initial

At follow-up

Change

Compression per cent)

Initial

At follow-up

Change

L3 fracture

No .

of patients

Kyphosis degrees)

Initial

At follow-up

Change

Compression per cent)

Initial

At follow-up

Change

L4 fracture

N o.of patients

Kyphosis degrees)

Initial

At follow-up

Change

Compression per cent)

Initial

At follow-up

Change

L5 fracture

No .

of patients

Kyphosis

degrees)

Initial

At follow-up

Change

Compression per cent)

Initial

At follow-up

Change

Gro up 1:

Non-Op .

Treatment

12

3 -5 to +12)

12 5 to 26)

9 0 to 21)

29 10 to 52)

37 1.4 to 63)

8 0 to 27)

8

-2 -19 to +17)

8 -5 to +26)

10 2 to 17)

32 24 to 43)

39 24 to 66)

7 0 to 25)

5

-24 -42 to -9)

-12 -22 to +5)

12 5 to 22)

33 9 to 46)

47 36 to 63)

14 1 to 22)

5

-25 0 to -35)

-15 -26 to +10)

10 4 to 15)

47 34 to 58)

51 33 to 69)

4 0 to 11)

Group II :

Post . Arthrodesis

18

15

- 2

54

27

-27

Long)

8

6 to 40)

7 to 30)

-10 to +15)

33 to 90)

21 to 41)

-49 to +1)

0

0

0

Group III :

Post . Arthrodesis

Short) t

1

N A

30

N A

60

6

5 -15 to +13)

10 -6 to+21)

5 -16 to +22)

48 44 to 54)

43 32 to 59)

-5 -17 to +7)

1

5

5

0

52

33

-19

0

Group IV:

Ant. and Post.

Arthrodesis

3

13 6 to 19)

13 6 to 21)

0 -5 to +19)

56 52 to 58)

40 31 to 61)

-16 -25 to+ 9)

3

7 5 to 8)

9 5 to 14)

2 0 to 6)

40 30 to 56)

44 30 to 63)

4 -11 to +20)

0

0

*The values are given as the mean, with the range in parentheses.

tNA = not available.

dex, from 64 to 100), and the rem aining th ree Case s 16,

17,and 18) all had incomplete recovery moto r-trauma

indices at the time of follow-up, 65, 80, and 63). One of

these patients used an ankle-foot orthosis bilaterally for

walking about the community. The other two used a

wheelchair al though both had enough strength in the

lower extremities for transfers. Of the nine patients who

had had mult iple-nerve-root involvement , four had had

bowel and bladder dysfunction at the time of presenta

tion. Th ree of the four attained no rma l function after

anterior decompression and arthrodesis .

Radiographic Evaluation

The initial kyphosis and initial compression associ

ated with fractures at the second and third lumbar lev

els were compared between the group that had been

managed non-operat ively and the groups that had been

managed with posterior arthrodesis , to determine if

there had been a bias toward performing an operat ion

for fractures associated with a larger initial deformity

Table III). Only fractures at these two levels were in

cluded, as only one fracture caudad to the third lumbar

level had been stabilized with instrumentation.

The mean initial kyphosis was 2 degre es range , -1 9

to +17 degrees) for the pat ients who had been mana ged

non-operat ively Gro up I), com pared with 10 degrees

range, -15 to +40 degrees) for the pat ients who had had

posterior s tabi lizat ion Gro ups II and III) . The m ean

initial compre ssion was 29 per cen t range , 10 to 52 per

cent) for G rou p I, com pared with 50 per cent range, 33

to 90 per cent) for Groups II and III. Both of these

differenc es w ere significant p = 0.05 and 0.0006).

THE JOURNAL OF BONE AND JOINT SURGERY

7/24/2019 BF L2-5 1996

6/11

BURST FRACTURES OF THE SECOND THROUGH FIFTH LUMBAR VERT EBRAE

1161

T A B L E

IV

D A T A W IT H R E G A R D TO C O R R E C T I O N OF THE K Y P H O S I S , A C C O R D I N G TO THE T Y P E OF INST RUME NT ION*

Initial kyphosis

Correction obtained

Correction lost with

instrumentation

in

place

Correction lost after

removal

of

instrumentation

Total lossof correction

Total change

in

kyphosis

G r o u pII:

Post. Arthrodesis

Long)

N =8)

18 6 to 40)

9

0 to 20)

9 3 to 14)

- 2 -10 to +15)

G r o u pIII:

Post. Arthrodesis Short)

Steffee Plates

N =3)

1 -15 to +10)

19

15 to 27)

23 11to 38)

4 -16 to +22)

Fixateur Internet

N = 5)

5 5 to 13)

13 11

to 16)

13 7 to 18)

6 5 to 7)

19 12 to 26)

6 -1 to +14)

G r o u pIV:

Ant .andPost.

Arthrodesis

N =

6)

12 5 to 19)

17

8 to 25)

18 9 to 25)

1 -5 to +19)

*The values

are

given

as the

mean ,

in

degrees, with

the

range

in

parentheses.

The

mean duration

of

follow-up

for all

patients

was

eighty-six

months.

t F o u r

of

the five patients who were managed with

the

fixateur inter ne

had the

device removed

at a

mean

of

eleven months postoperatively.

The degree of kyphosis immediately after the in

jury

and

preoperat ively

was

com pared with that

at the

latest follow-up evaluation. The mean increase in ky

phosis during treatment

was 10

degrees range,

0 to 21

degrees)

for the

pat ients

who had

been managed

non-

operat ively Group

I)

compared with

-2

degrees range,

-1 0to +15degrees) for thosewho hadbeen managed

withalong hook -and-ro d system Gro up II) ,5degrees

range,

-16 to +22

degrees)

for

those

who had

been

treated with a short t ranspedicular construct Gro up

I I I ) ,and 0

degrees range,

-5 to +19

degrees)

for

those

who had been managed wi th an ter ior and posterior

ar throdes i s Group

IV). The

increase

in

kyphosis

for

the pat ients who had been manag ed non-operat ively

and for thosewho had been treated with short t rans-

pedicular fixation

was

similar. Both values w ere signifi

cant ly greater than thosefor thepat ientswho had been

managed with ei ther long posterior arthrodesis

or

an te

rior and posterior arthrodesis p =0.005).

The change

in

vertebra l body compression averaged

8per cent range,0 to 27 per cent) for the group that

had been managed non-operat ively, -27

per

cent range ,

- 4 9to +1 percent )for the group thathad been treated

with long posterior arthrodesis ,

-5 per

cent range,

-17

to +7 per cent) for the group thathad had short pos

terior arthrodesis ,and -10 per cent range,-25 to +20

per cent)

for the

group that

had had

anterior

and

pos

terior arthrodesis . The difference betw een the group

that

had

been managed non-operat ively

and the

groups

that had been treated operat ively wassignificant p =

0.0004).

The

change

in

kyphosis

did not

correspond

to

the change in compression in the group that had been

managed with short fixation.

The ini t ial radiographic measurements, including

the type of fracture according to the classification of

Denis

16

,

the

location

of the

fracture,

the

initial kyphosis,

the initial compression, the ini t ial vertebral wedge in

dex,

and the

presence

of

radiogra phically identifiable

injury of the posterior elements, were analyzed with

regard

to

progressive deformity

in the

form

of

increased

kyphosis. Sixteen fracture s we re classified asDenis type

Aandfourteen,astype

B.

Onlyonetype-B fra cturewas

caudad

to the

third lumbar level, indicating

a

predi lec

tion of type-A fractures for the more caudad lumbar

segments. With

the

numbers avai lable,

the

level

of the

fracture

p =

0.47),

the

initial kyphosis

p =

0.27),

the

initial compression p =0.07),anddisruptionofthe pos

terior elements p = 0.10)demons t ra tednorelat ionship

to increased kyphosis.

The

only significant relationship

was between Denis type-A fractures and progressive

kyphosis mean increase,12degrees [range,2 to 22 de

grees] ,

comp ared w ith

7

degrees [range, 0

to 15

degrees]

for the type-B fractures;p =0.02).The initial compres

sion associated with the type-A fractures was greater

than that associated with

the

type-B fractures

38

com

pared with28 per cent). This wedge compression re

mained constantand did notaccountfor theincreasein

kyphosis seen

at the

t ime

of

follow-up.

The effectiveness

of the

various types

of

operat ive

t reatment

in

achieving

and

maintaining correct ion

of

the deformity wasassessed Table IV). With the n u m

bers available,

the

mean values

for the

initial kyphosis

and

the

initial compression were

not

significantly

dif

ferent p = 0.10 and 0.94)amongthe t reatment groups.

For

the

pat ients

who had had

long posterior arthro

desis Group

II),

short posterior arthrodesis Gr oup

I I I ) ,andanteriorandposterior arthrodesis Gro upIV),

the mean initial kyphosis

was 18

degrees range,

6 to

40 degrees),

5

degrees range,

-15 to +13

degrees) ,

and

12 degrees range, 5 to 19 degrees), respectively.The

mean ini t ial compression

was 54 per

cent range,

33 to

90

per

cent),

48 per

cent range,

44 to 54 per

cent ) ,

and

50per cent range,30 to 58 per cent). With the num

bers available,

no

significant differences w ere obse rved

among

the

groups with respect

to

correct ion obtained

and correct ion lost throughout the follow-up perio d

p

= 0.15 and

0.09).

For the

pat ients

who had had

long

posterior arthrodesis Gro up

II), the

me an correct ion

was 9 degrees range,0 to 20 degrees) and the mean

correction lost

was 9

degrees range,

3 to 14

degrees).

VOL. 78-A, NO. 8, AUGUST 1996

7/24/2019 BF L2-5 1996

7/11

1162

D.

A. ANDREYCHIK, D. H. ALANDER, K. M. SENICA, AND E. S. STAUFFER

For the pat ients who had had short posterior arthro

desis (Group III) , the mean correct ion was 15 degrees

(range, 11 to 27 degrees) and the mean correct ion lost

was 19 degrees (range, 7 to 38 degrees). For the pa

t ients who had had anterior and posterior arthrodesis

(G roup IV), the mean correct ion was 17 degrees (range,

8 to 25 degrees) and the mean correct ion lost was 18

degrees (range, 9 to 25 degrees). Although transpedicu-

lar fixation and anterior and posterior arthrodesis dis

played the greatest capacity for correction, all constructs

tended to col lapse toward the preoperat ive deformity.

The radiographs of the three pat ients who had had an

terior arthrodesis only were excluded from the analysis,

as only one of these pat ients had comp lete radiographic

data.

The ini t ial comp uterized tom ography scans were an

alyzed for twenty-one of the thirty patients in Group I,

none of the eight in Group II, all eight in Group III, all

six in Gro up IV, and two of the thre e in Gro up V. The

mean reduction in the cross-sectional size of the canal

was 43 per cent (range, 10 to 90 per cent) for Group I,

68 per cent (range, 50 to 90 per cent) for Group III, and

87 per cent (range, 75 to 90 per cent) for Group IV. In

Gr ou p V, both pat ients for whom c omputerized tomog

raphy scans were available had a 90 per cent reduction

in the cross-sectional size of the canal.

Follow-up computerized tomography scans were in

terpreted for two pat ients (one fracture at the second

lumbar level and one at the third lumbar level) who had

been managed non-operat ively. Both pat ients had par

tial resorption of the fragments in the canal. The patient

who had had a fracture at the second lumbar level had

an initial reduction of 44 per cent in the cross-sectional

size of the canal; forty-four mo nths after the injury, the

canal had increased to 90 per cent of its expected size.

The pat ient who had had a fracture at the third lumbar

level had an initial reduction of 40 per cent in the cross-

sectional size of the canal; fifty months after the injury,

the can al had increa sed to 80 pe r cent of its expected

size.

Immediate postoperat ive scans were avai lable and

were reviewed for four of the eight patients who had

been managed with a short t ranspedicular construct

(Group III); all four had had a fracture at the third

lumbar level. The increase in the cross-sectional size of

the canal at the level of the injury was unpredictable;

the mean initial decrease was 78 per cent (range, 60 to

90 per cent), and the mea n postop erat ive size was 62 per

cent (range, 50 to 90 per cent).

Pain and unctional Results

At the latest follow-up evaluation, the pain scores

were a mean of 1.50 points for Group I, 1.13 points

for Group II, 1.38 points for Group III, 1.17 points

for G rou p IV, and 1.33 points for Gr ou p V; with the

num ber s available, these differences w ere not signifi

cant (p = 0.78). No relat ionship was demon strated be

tween the pain score and the level of the fracture, the

kyphosis at the time of follow-up, the compression at

the time of follow-up, the change in the kyphosis, or the

change in the compression (p = 0.27,0.15,0.23,0.09, and

0.09, respectively). The only variable that had a signifi

cant relationship (p = 0.03) to the pain score was the

type

16

of fracture. The pain score was a mean of 1.76

points for type-A fractures, 1.33 points for type-B, and

1.00 for type-D. Although we could not determine a

significant difference with the numbers available, there

was a trend for higher pain scores in association with

the fractures at the more caudad lumbar segments (a

mean of 1.33, 1.18, 1.50, and 2.20 points in association

with fractures at the seco nd, third, fourth, and fifth lum

bar levels, respec tively).

Most of the thirty pat ients who had been managed

non-operat ively had mild or moderate pain. Only three

of these pat ients were pain-free (a pain score of 0

points). Seventeen pat ients had intermit tent mild pain

(a pain score of 1 poin t); three , freque nt mild pain (a

score of 2 points); five, moderate pain (a score of 3

points); one, severe pain (a score of 4 points); and one,

profound pain (a score of points). The pat ient who h ad

a score of 4 points was a nineteen-year-old w oman who

had sustained a burst fracture at the fourth lumbar level

as a result of a motor-vehicle accident. She did well (a

pain score of 1 point) for two and one-half years, until

she was involved in another motor-vehicle accident. Af

ter the second accident, she had persistent back pain

al though she remained employed. She eventual ly had

anterior and posterior arthrodesis for rel ief of pain

associated with instability at a level cephalad to the

fracture. Eight months postoperatively, she had only

interm ittent mild discomfort (a pain score of 1 poin t).

The patient who had a pain score of 5 points was a

twenty-nine-year-old man who had sustained a fracture

at the second lumbar level secondary to a work-related

injury. He reported profound pain and eventually, at

twelve months, had anterior and po sterior arthrodesis

for the pain as well as disc degeneration and instability

(13 degrees on flexion and extension). Seven years after

the operat ion, the pat tern of pain remained unchanged

and the patient was totally disabled.

In the ope rat ive groups, three pa t ients had no pain;

fifteen, inter mitte nt mild pain; five, frequen t mild pain;

and two, mo dera te pain.

For the pat ients who did not have neurological

dysfunction, the functional outcome at the time of the

most recent follow-up was analyzed with respect to the

pa tien t s age , the level of the fra cture, the kypho sis

and compression at the time of follow-up, and the pa

tient s pre-injury oc cupa tion. With the nu mb ers avail

able, no significant relationships were found. Of the

thirty pat ients who had been managed non-operat ively,

ten had performed heavy labor before the injury, six

had performed moderately heavy labor, three had per

formed light labor, nine had been sedentary, and two

THE JOURN L OF BONE ND JOINT SURGERY

7/24/2019 BF L2-5 1996

8/11

BURST FRACTURES OF THE SECOND THROU GH FIFTH LUMBAR VERTEBRAE 11 63

had bee n unemp loyed. Of the twenty-eight pat ients who

had been employed at the time of the injury, twenty-five

(89 per cent) returned to their pre-injury occupational

status,

al though ten had minor and two had major l imi

tations (a functional score of 1 and 2 points, respec

tively). One patient, who had a functional score of 3

points , need ed to change to a less s trenuous occupa

tion. Two patients were disabled as a result of pain: the

twenty-nine-year-old man described previously as well

as a forty-eight-year-old man who had sustained a frac

ture at the fifth lumbar level and was unable to return

to work because of moderate pain; these pat ients had a

pain score of 5 and 3 points, respectively.

Of the twenty-five pat ients who had been managed

operat ively, seven had performed heavy labor before

the injury, six had performed moderately heavy labor,

three had engaged in light labor, five had been seden

tary, and four had bee n une mploye d. Of the tw enty-one

patients who had been employed at the t ime of the

injury, three were disabled secondary to neurological

dysfunct ion. O f the e ighteen re maining pa t ients , s ixteen

return ed to their pre-injury work status, one was em

ployed in a less s trenuous job, and one was unem ployed

secondary to moderate act ivi ty-related back pain. Of

the en tire grou p of twenty-five patients, fifteen ha d no

functional impairment and four had minor functional

limitations.

iscussion

Burst fractures of the lumbar spine have unique

biomechanical and neurological features as compared

with burst fractures in other regions of the spine. Deci

sions with regard to stability and treatment are different

than those made when there is a more cephalad frac

ture of the vertebral column. In the lumbar spine, the

body's center of gravity falls at or posterior to the ver

tebral axis

1

. Slight flexion decre ases the lordosis and

places the axial load force-of-injury vector throu gh th e

vertebral body. This dictates the specific pattern of in

jury as well as the tendency for progressive deformity.

Pure axial load injuries are more common in this spinal

region

16

-

24

. Bec ause of the rela tive stability provid ed by

the posterior elements, the risk of neurological d eterio

rat ion is iow

1

'

1U5

'

263

'

38

'

45