66 THE JOINT PROTECTION MECHANISMS
(g)
Figure 4.6 (continued).
Table 4.1 Average age, height and weight for
40 subjects without a history of low back pain
Variable Males Females Total
No. 13 27 40
Heig ht
mean em 181.85 164.85 170.37
SO 5.49 5.59 9.75
Weight
mean kg 82.92 60.11 67.52
SO 16.49 6.27 15.06
Age
mean years 28.92 28.11 28.37
SO 5.75 5.83 5.74
SO, standard deviation.
cross-sectional ilrea at five vertebrill levels in sub
jects without il history of low back pain. Table 4.1
sl10ws the a v erage age, height ilnd weight of the 13
mille and 27 female subjects measured. Table 4.2
shows the multifidus cross-sectional areilS for the.
d ifferent lumbosacral vertebral levels 0. Hi�ies et ill
20l)3, u n published dMa). It was foy,nd tha t males
have significantly larger multifidus mu s cles than
females at all vertebrill levels (Hides et ill 1992,
S. Kelley et al 2003, wlpublished data) ilnd thilt tl1€
cross-sectional ilrea of the multi fidus is significantly
different for each vertebral level (H ides et al 1995,
S. Kelley et al 2003, unpublishe9 datil). In subjects
without il history of low back pilin, multifidus
cross-sectionill area is symmetricill between sides
(Hides et aI1994). The m u lti fidus is the largest mus
de spillU1ing the lumbosilcral junction (Macintosh
(h)
,
Table 4.2 The multifidus cross-sectional areas
for the different lumbosacral vertebral levels in
40 subjects without a history of low ba.ck pain
- ",
Vertebral level Cross-secti Confidel'te
(mean em .. interval-
Female
2.1 (0.13( S
L2 1.85-2.35
L3 3.35 (0,17) \3.0-3.7
L4 4.78 (0.2). .. . (37-5.18
L5 6.38 (0.18) 6.01-6.75
Male
L2 3.01 (0.18) 2.65-3.38
L3 4.31 (0.25) 3.81-4.82
L4 6.27 (0.29) 5.68-6.85
L5 6.79 (0.27) 6.25-7.32
SE, standard error.
J. Hides et ai, unpublished data.
et aI1986). In contrilst, the cross-sectional area of the
lumbar longissimus and iliocostali� decreases on
progression caudally. The large size of the multi
fidus muscle at the lumbosacral j u n ction, compared
\vith the adja cent lumbar erector spinae muscles,
also suggests thilt it is the muscle most capabk of
providing support at this level. Notably, it is the
L4-L5 and LS-Sl segme nts that have the highest
incidence of pathology in low back pilin. The multi
fidus has a close relationship to the zygapophyseal
jOints (Fig. 4.2) ilnd by controlling the sliding move
ment of the zygilpophyscill joints in the cranioca u
dal direction it controls the distribution of stresses
and IOilding on the vcrtebral triad. It is considered
that the multifidus is the only muscle·the primary
function of which is to protect the vertebral triad
(L.ewin d aI1962).
Biomechanical factors
Control of the neutral zone
Several studies have in'\!estigated the lumbar mus
cles' capa city to increase the spinal segmental stiff
ness and, iii particular, the control of neutral zone
rnotion in line with Panjabi's (lYY2b) hypothes .\ of
clinical instability. Stud ies have been done on vari
ous combinahons of muscles tQ.investigate their
in fluence on these parameters. Kaigle et al (1995)
developed an in vivo animal model of lumbar seg
m en tal instabilitY. Passive stabilizing structures
(disc, zygapophyscal joints and ligaments) were
tmnsccted. and the effects of active �usculature
on spinal kinematics were e xami ned in 33 pigs.
Musd� ;> , surrounding the spine, inc l u ding the
mult ift�us, t.
quadratus IU.lpb
,
orum and psoas major and minor,
werc eXJmine¢ The injured segments were sub
jected to muscle s.timulation using wire electrodes.
ResuHs showed that increased, combined muscle
activation stabilised the injured motion segment·
by reducing aberrant patterns of moti on in the
neutra I zone.
Coel et al (lYY3) used a combined finite-element
and optimization <lpproach to study the effects of
the actions of the interspinilJcs and intertransver
sarii, the lumbar multifidus and the quadratus
lumborum. The introduction of muscle forces led
to a decrease in displacements in the sagittJI plane,
anteroposterior translation and anterior rotation.
It was shown that these muscles imparted stability
to the ligamentous system. The load bearing of
tht' zygapophyseill joints was found to i.ncrease,
indicating that these joints playa significant role
in transmitting loads in a normal intact spine.
Musck dysfunc tion (simulated by decreasing the
computed force ill tlw muscles) destabilised the
motion segment. This led to a shift of loads to
the disc and ligaments and decreased the role of
the I.ygapophysl'<ll jllints in transmitting loads
(Cod and CilbertsonlY95).
l'ari.jabi et aI (19HY), in an in vitro stu dy of intact
,md se q uenti a l ly injured fresh lumbar spinal units,
Paraspinal mechanism and support of the lumbar spine 67
more specifically simulated the effect of interseg
m ental muscle forces on spina l instability. Simu
lated forces represented the multifidus (deep,
shorter fascicles), interspinales and rotatores
muscles . The segments were subjected to three
dimensional loads with increasing muscle forces.
This study, like the one by Goel et al (1993), exam
ined the effect of the segmental muscles without
the influence orthe larger lumbar longissimus and
iliocostalis. Panjabi et <11 (1989) conclu ded that
the intersegmental nature of the deep multifidus
fibres gave a tremendous advantage to the neuro-
. muscular system for controlling the stability of the
lumbar segment. '
Wilke et al (1995) investigated the influence of
five different muscle groups on the monosegmental
motion of the l4-L5 segment. The muscles exam
ined were the multifidus (caudal cll1d cranial direc
tions), lumbar longiSSim u s, lumbar iliocostalis and
psoas major. Seven hum a n lumbosacral spines
were tested on a spine tester that allowed simula
tion of muscle forces. The combined muscle action
of the muscles tested was found to decrease the
total range of motion and neutral zone motion of
the L4-L5 segment. The total neutral zone motion
in flexion and extension was decreased by 83°;" •.
Tn lateral flexion, the total range of motion was
d ecreased by 55'1., and the neutral zone by 76°1. ..
Under axial ro tation, the total range was reduced
by 35%, but there was nc) significa.nt change in
neutral zone motion. Muscle forces were found
to stiffen the motion segment. The strongest influ
ence was created by the lumbar multifidus, \.vhich
was responsible for more than tvvo-thirds of the
increase in segmental stiffness. The multifidus
action was responsible for a significant decrease
ill the range of motion of all movements except
rotation. These results supported those ubtaillL'd
by Steffen et al (lY94) in another in vitro study,
who also found that the influence of lumbar mul
tifidus decreased the neutral zone in flexion ,md
extension.
The la teral stabilizing potential of thl' Il)mbar
intersegmental and polysegnwntal muscles has also
been investigated by Crisco and l'anjabi (19Yl).
They fOWld that the polysegmental fascicles of
multifidus and lumbar longissmus and iliocosta lis
fascicles were more efficient in this direction
than were the short deep multifidus fascicles and
68 T H E J O I N T P R OT E CT I O N M E C H A N I S M S
in te rtra nsversa ri i a nd i n terspi n il les m u s c l e s . I t can
be s u r m ised th a t t h e role of the m u l tifidus i n l u m
bar s p i ne stab i liza tion is complex . The m u ltifi dus
i s capable o f con trolling the ne u tr a l zone in the
sagittal plane w i th i ts deeper, in tersegmen tal fibres,
b u t it requ i res the assistance of the l u m b a r longis
s i m u s a n d i l i o cos ta l i s in the l u mba r muscles' con
t r i b u t i o n to the control o f neu tra l zone motion i n
t h e fro n t il l plane.
S o m e biomec han ica l s tu d ies have investiga ted
the effe c ts of co-activ a tion o f m u scles on sp i n a l
stabi li ty. A recent s tudy i n vestiga tin g co-ac t i v a t i o n
o f t h e mu l t i fi d u s a n d the p s o a s showed t h a t the
s ta b i l i ty of the L4-LS segment was enh a n ced in
la teral tl e x i on and rota tion b u t it w a s d e s ta b i l ised
i n fl exion (Qu i n t et al 1 998). I t was conclu d e d th a t
the i m p o r tan ce o f the neura l control s t r a tegy o f the
s tab i l i z a tion of the spine c a rm o t be overempha
s i zed . The n e u r a l con trol l e r m u s t n o t only select
the a p prop ria te muscles to s t i m u l a te b u t a l so
d ec i d e on their a pp rop ri a te activa tion level (Qu in t
e t a I 1 998 ) .
The d eep m u l ti fi d u s fi bres i n p a r ticu l a r a re
placed close to the cen t res of rota tion of spi n il l
mov e m e n ts a n d connect a dj a ce n t v e rtebrile a t
appropria te a n gles. M c G i l l ( 19 9 1 ) confi rmed the
role of l u m b a r m ul t i fid u s in a three-dimension a l
study of l u m b a r s p i n e mechanics a nd concl u d ed
tha t the u ncha n g i ng geo m e try of the m u l t i fi d u s
th ro u g h a ril nge o f postures i n d ica ted th a t t h e p ur
pose o f t h i s mu s c l e is f i n e l y to a d j u s t v e r tebrae
with s m il l l mov emen ts rather tha n to fu n c t i o n il S a
p rime m o v e r. Th i s s t u d y sho wed t h a t the m u lti
fidus could func tion i n this way in a n y p h ysio
l og i ca l postu re .
Con trol o f the lo rdosis
It h a s been well acceptl' d i n t h e a rea o f b iomech
a n i c s a n d ergon o m i cs tha t th e sp i n a l c u rves a re
a n e ffi c ient Wily for the body to deil l w i th forces o f
gra v i ty. W h e n t h e spina l c urves M e m a i n ta ined,
this is the most energy-effi c i e n t position fo r the
bod y to s til y up right aga i n s t the forces o f grilv i ty
a n d to w i thsta n d f u r t h e r fo rces th a t Me a p p l ied to
the s p i n e . For the l u mba r region, it has been p ro
posed th a t the m u l t i f i d u s can contribu te to s ta b i l i ty
o f the s p i n e v i a con tro l of thl' l or d o s i s , a l low
ing e q u il l d is t ri b u ti o n o f forces ( As p d e n J 9Y2) .
Contril c t i o n of the polysegmentil l m u l ti fi d us fasc i
cles can resto re the l u m ba r l ord osis. Rece n t s t u d ies
h a ve con fi rmed this fi nd ing; the c o m p ressive loa d
bearing cap acity o f the passive thoracolumbar spine
was s igni fi c a n tl y en h a nced by pel v i c ro t a t ion
ca used by m i n i ma l muscle forces in the sagi tta l
p l a ne ( K e i fer et a I 1 997) . W h e n loca l m u s c l e s w e re
exa m i n e d in t h e m o d e l , m u l t i f i d u s w a s found
to con trib u te 80'Yu o f the re q u i red a c t i v i ty ( Ke i fe r
e t a l 1 998).
To m a in tain the spina l c u rves requ i res a b a l a nce
b e tween and i n tegra ti o n of loca l , m o n o - a r ti c u l a r
and glo b a l m u sc l e s . T h i s re l a ti o n sh i p h a s been
ex a m i ned i n biomec h a n i c a l s t u d i es ( K e i fer e t a l
1 997, 1 998) . Th e shldies used a fi n j te e l e m e n t model
(passive osseoligamen tou spi_ne) w i th o p t i m iza
tion o f the m u s c les ( a c t i v e a n d p a s s i v e c o m p o n
ents o f muscle fo rce ) . I n t h e fi rst s t u d y, Ke i fe r
et a l ( 1 997) IOilded t h e s p ine us i n g co m p ress i v e
a x i a l forces . T h e y showed tha t the thoracol u m b a r
s p ine trans lil ted i n to hype rmobil i ty u n der a x i a l
l o ads less tha n p h y s i o logica l loa d s w i t h o u t the
muscles ( i n d i c a ted by d i s p l acemen t o f the Tl ver
tebra ) . A d d i tion of l o c a l and globa l m u scles i n t o
t h e mod e l i n c reased the a b i l i ty of the s p i ne to
withstand compressive fo rces w i th o u t b u c k l i ng.
P e l v ic ro tation ( a n te r i or t i lt) s t i ffened the s p i rle,
and only 2 degrees of a n te r ior ro ta tion a l lowed the
spine to carry a xi a l com p ression o f LI p to 400 N,
wi th on l y 5' m m o f a n tl' r i o r d i sp i a cl' m e n t of T i .
There w a s kss a n te r i o r d i s p l ilcement o f T I w i t h
local a n d gl o b a l 111 u sc iL' s incorpor a ted i n to the
model thil n w i th glob'l l m Llsc iL's a l one, h i g h l i g h t
ing the i m p o r ta n ce o f i n tegra t i o n of t ill' two
m u scle syste m s . I n t h e second s t u dy, Kl' i tc r l't a l
(1 998:-) investiga ted th e synergy o f thl' s p in e i n ll L'U
tra ! positions. Using the s a me m od d , tl1l'Y d is p laced
T1 40 m m a n terimty and 20 mm posterio r iy. Res ults
s h owed thil t the gl obil l m u sc i L's an d p a s s i v e s t r u c
t u r e s a rc s u ffi c i e n t 1\ ) s t ,l b i l i z l' th e s p i ne fo r v e r y
sl1l.all d i sp i<lcl' m e n t s . H o wL'vcr, thl' sys iL' m i s fa r
m o re effi c i e n t w i tb inc l u s i o n of t h e I OGl l m u scles.
A c t i v a t i o n o f the l o c a l m u sc ks decrea sed mu scle
fmc es i n the global s y s t e m , p rov i d ed s t i ffnl'ss,
i n c reased sta b il i ty a n d i n c reased comp ress ion .
Consi d er i n g thl' con t r i b u t i o n of t h l' l oca l musc les,
HO% was 'prov i d L'�1 b y m u l t i fi d us, v\' i th s o m e c o n
tribLi ti on by i l i oc o s ta l i s . A n o t h n find i n g 'vV ,l S t h a t
t h e p os i t i o n o f th L' t h o r a c o l u m bil r j u n c t i u n is
i m porta n t . I f T1 2 w a s held ba ck, then the u ppe r
lu mba r spine w a s forced into fl exion, a n d the syn
ergy w a s d i sturbed . Th is h a d J reslI l ta nt marked
e ffect on the d i s tr i b u tion of i n terseg m e n tal rota
tions and lessened the capa c ity o f the passive sys
tem to ca rry sagitta l moments.
Tensio n ing th e th oracolumbar fascia
Mu scles enh J n c e s pi n a l s t a b i l i ty.. by i n crea sj g the
sti ffness of the spi n a l segment. It has a lre ady been
shown tha t the mult i f i dus acts to sti ffen th e motion
segment ( W i l ke et a l 1 995 ) . From a mec ha n i ca l
perspective, the be n d i ng sti ffness o f the spine w i l l
a lso b e i n fl u enced b y other factors. One struc t u re
th a t c a n contribute to l umba r sta b i l iza tion by
i nc re a s i ng the be n d i ng sti ffness of the spine is the
th o racol u m b a r fa sciJ . The thora columl?ar fa sc ia is
a strong tiss u e w ith a well-developed la tt ice o f col
l a gen f i bres, suggesting tha t its fu n ction m a y be
tha t o f an extensor muscl e reti n a cul u m ( Bogd u k
a n d M a c i ntosh 1 984) . In a d d it i o n , t h e fa s c i a does
con ta i n both R u ffi n i J n d Paci nia n corpusc l es
together w i th d i ffuse in n e rv ation (Ya h ia et aI 1 988) .
It is imp orta nt to recogn ize t h a t it is a musculo
fa sc i J I system that p rotects the lumbosJcral reg ion,
and the in fluenc e o f the m us c l es on tensi oning
fa s c i J s i s vi ta l l y i m po rtil nt. The tra nsve rs u s abd o
m i n i s (plus or m i nu s the intt'mJ I o b l iqu e muscle)
mJY te n s i on til L' thor,l C o l u m b J r fJ sciJ (see Ch . 3 ) .
I n a d d it i o n , thL' t h o r J co l u m ba r fa sci J con strJ in s
the rJ d i J I e x p a n s ion o f th e three lum b a r b ,l C k
m u scle s ( A spden 1 99 2 ) . It h a s been p roposed t h J t
con trac t i o n of t h e s e m u scll's e xer ts a p u sh i ng force
on the fa scia ( F J rfa n 1 973) . The i n fl uence o f
the mu l t i f i d u s a n d t h e l u m b il r l ong i ss i mu s and
iIi coco sta l i s on t h t' tho racolum b a r fJ sc i a w a s
i nvcstigJ tl'd b y C ra c ove tsky e t a l ( 1 977) us ing a
m at h e m a ti c il l mod e l . l t w a s p roposed that the tho
rilcol u mb a r fa scia co u l d serve to bracl' the back
muscles bec a u s e it su rro u nded them . The a u thors
c a l l1'd thi s til l' ' hy d ra u lic a mp l ifier mecha n i sm ' .
W h i l e t h i s mechanism h a s been dism issed o n the
g roun d s tha t it is un l i kL'iy to genera te extensor
forces ( M cC i I l 2002 b ) , it i s m o re tha n l i kely that the
thoracolum b a r fas c i a , through i ts m u sc u l i H J ttach
l1len ts, contr ibutl's to i n creased l u mba r sti ffn ess
and l u mb a r s t<1 b i l i zatiol1 . It may w e l l a l so h a v e J
p ropriocept i v L' wk.
Paraspinal' mec hanism and support of the lumbar spine 69
Control of she a r forces
Sheilr fo rtes a re those th a t cause two ve rtebrae to
s l i d e w i th respec t to one another ( Bogduk ] 997 ) .
D u r i n g fle x i on of the lumbar spine, a fo rward o r
anterior shear i s exerted o n the i n ter ve rtebral j oint.
Particu l il r a ttention has been given to these shea r
fo rces, w h ich a re i.nduced by ben d i ng a n d l i fting
tasks (Cholew i c k i et a l 1 99 1 , Porterfield and De Rosa
1 991 ) . The con trol o f a nte r i o r shear forces i s esse n
ti a l fo r the pro tec ti on o f t h e i n te rvertebr J l j o i n t,
especi ally at the lower lumbJ r lev e l s where these
forces are greatest. This cont rol is p ro v i d e d not
only by the passive elements a nd a rtic ula r config
ura tion o f the vertebra l column but a lso by the
mu sc l e syste m .
TrJ d i t i o n a l l y, th e lumb a r ex tensor muscl es hJV '
been assigned this ro l e , When con tf(1 c ti n g b i lat
e ra l l y, on the one ha nd , the l u mb a r l ongiss i m u s
a n d t h e lumba r i l i ocostJ l i s c a n d ra w the i r vertebra
of or i gi n poste r i o rl y, owing to their posterior il n d
c a u d a l dire c tion, a nd hence oppose the a n terior
shea r. On the other hand , cont ra ct ion of the mu l ti
fidus fascicles prod uces pos terior s a g i tta l rota tion
of the vertebra of origin ra ther than poste r i o r
trans l a ti o n . I t i s l i kely that d u r i n g a c tivities such
as fo rwa rd bendi ng a n d l i fting the i n d uced forces
a re control l ed by the lum bar erector s p i nil e
m uscles J n d the multifidus toge th e r. McG i l l d al
(2000) propose tb<l t, fo r l i ft ing , the lumba r longis
simus an d l u mbar i l ioc osta l i s lose their obli que
l i n e o f action if the lu mba r s p i n e flexes a n d , as <l
result, t h e l u m b a r sp i ne is unab l e to res i s t d a ma g
i ng s h e il r forces . This wo u l d h ighligh t the impor
ta nce of the neutril l sp ine posit ion and may show
how the lum b a r longi ss i m us, l u m ba r i l iocosta l i s
a n d m u l tifidus could work togcthet� a s one o f thl'
known roles of the m u lti fid us i s a dj u s t m e n t a n d
co ntro l o f lumb a r lordos i s .
Howev er, the con tro l of s h e a r fo rces ap pe,Hs
to bt' J fa r more com p le x iss u e . A model of ba c k
muscles th ilt m J p p e d thL' a ctions o f i n d i v i d u a l
fa scic les ( B ogduk c t a l l YSl2a) showed th,l t shea r
fo rces C il n be i nduce d by these musc l es on ma x i
m J I exerti o n . F ro m L 1 t o lA, th e n e t res u l t vv a s a
posterior shea r forc e . However, the net ba lance
was a n Jnterior shear force at the L5 leve l . This
would suggest th J t vi1 rious musc l es i n ad d i ti o n to
the back ex tensors m a y be involved in t h e con t m l
7 0 TH E J O I N T P ROTECT I ON M E CHANIS M S
of a n te r i o r shea r fo rces d u r i n g l i fting a n d be n d in g
ta s k s . I n teres t i ngl y, Fa rfan (1 975) p roposed th a t
anterior shear forces w e re resisted m ore b y the
zygapophyseal joi nts, with cou n tering forces in
the reverse d irec tion being p rovided by the abdom
i n a l m u s c u l a t u re .
T h e back m u s c l e s i n post u re
a n d moveme n t
I t i s possib l e t h a t there a re d i ffere n t pri m a r y fu nc
tions for the d i ffe rent fasc i c les of multi fidus. The
longer fascicles, w h ich origina te fro m the s p in o u s
p rocesses, h a v e a mechanic a l a d v a n tage o v er the
shorter, deeper fibres . The longer fa scicles m a y con
tri bu te m o re to ex tensor to rque, w h i l e the s h o rter
d eeper f i b res, w h i c h have l i ttle leve rage fo r torque
p ro d uction, may be m ore inv olved in a ton k s t a b i
l i z i n g ro l e . There i s e v i dence to sugges t trus from
elec tromyograp h i c (EMC) stud ies, where toni c acti
vation of th · deeper fibres has been examined d u ri n g
t h e m a in tenance o f uprig h t postures a n d d ur ing
a c tive tru n k movements. EMC ana lys is has a L lowed
the fu n c t i on of the m u l ti fi d u s to be s t u d ied in vivo .
Ma n y c l a ss i c s t u d i es h a v e been performed using
in-d w e l l ing electrodes to access the a c tivation of
t he d eeper fascicles, w hi c h a re l i ke l y to be i n v olved
in a sta b i l i z i n g role. A tonic or a l most cont i n u o us
level of activation of the mu l ti fidus ha s been demon
strated in m a n y o f these s t udies of u p righ t pos
t u res a nd p r i m a ry a c t ive movements. M oseley et a l
(2()02) studied the respon se o f the m u l tifid u s d u r
i n g p e r t u rb a t i o n p roduced by m o v e m e n t of a l i m b
i n response t o a s t i m u lus. Res u l ts con.firmed t h a t
t h e d e e p fi b res o f til l' m u l ti fi d u s beh a v e di ffere n tl y
from th e supe r fic i a l fi bres a n d the lumba r erec tor
s p i n al' muscles. Du ring both flexion a n d ex te n s i o n
o f t h e a rm , t h e onsl' t of EMC in the d e e p m u l ti fid u s
occ u r red p r i o r t o a rm movem e n t a nd a t the sa m e
ti m e J S d e l toid. [ n cnntrast, t h e other m uscles
beha ved d i fferen tly d u r ing flex i o n and e x te n s i o n ,
s u gges t i n g th a t, while m o s t of the posterior t r u n k
m u sc l es J P p e a r t o cowl te ract pertu rba tion o f glob'l l
sp inal orientation, th e deep m u l tifid u s may be more
in volved in segme n ta l s ta b i l iza tion ( Moseley et a l
2002) . F u r t h ermore, d u ring rep e ti ti v e upper l i mb
mo v e m e nt, to n i c a c t iv i ty of the deep m u l t i fi d u s
w a s observed, i n l i n e \v i th i ts role i n s tab i l i z a tion o f
t h e s p i n e ( Moseley et al 2002) . TI1L' re i s e v i d e n c e
tha t t h e m u l ti fi d u s m uscle i s conti n u o u s l y a c ti v e i n
u p righ t postu res, com p a red w i th re la xed rec u m
b e n t pos i tions. A l on g w i t h t h e l u m b a r longissi m u s
a n d i li ocosta lis, t h e m u l t ifid us p ro v i des a n tigrav
i ty support to the spLne w i th almost continuous acti
v i t y (Asmussen a nd K l a u sen 1 9 62). T h e mu l ti f i d u s
is p robably a c t i v e i n a l l a n t i g ra v i ty a c ti v i ty (Morris
et al 1 962, Don isch and B a s m a j i an ] 972, Va l e n c i a
a n d Mu nro 1 985) a n d i s to n ica l l y a c t i v e d u r i_ng
w a l king (Morris et al 1 9 62) . I n the s ta nd i ng posi
tion, sbght to m o d e r a te a c t i v i ty o f the m u l t i fi d u s
ha s been d e monstrated ( J onsson 1970, Do n i sc h a nd
Basmaj i a n 1 972, Va lencia and M u nro 1 985), e x e m
p l i fy ing i ts tonic pos t u ra l rol e . I n fa c t, ma inta ining
a neutr a l spine may be the one element tha t is cru
cial to m a i n ta i n t h i s ton i c a c t i v a t i o n . Evidence has
been provided by a neu rophys iologica l i n vivo shl d y
\v h e re l u mb a r s p i nes of c a ts w e re p b ced i n to m o d
era te s u s ta i n ed fle x i o n a n d i n tramusc u l a r E M C of
t h e mu l ti fi d u s was reco rded ( Wi l l i a ms e t a l 2(00) .
Prolonged flex i o n o f the l um bar s p i ne I:es u l ted
i.n tens ion-re l a x a tion and la x i ty of its v i sco-ela stic
structu res, loss of p rotective re flexive m u scular
activ i ty w i th j n 3 m i n u tes fo l l owed by EMG s p a sms
in the m u l ti fi d u s and o t h e r posterior muscles.
A l though pe rformed on ca ts, th i s study h o l d s i m p J j
ca tions for h u m a n s a n d w o u ld len d s u p port t o p ro
ponents of the i mporta nce of the n e u t ra l spine
concept.
Res u lts of s t u d i e s perfo r m ed i n thl' sitting pos i
t i o n h a v e va ried . I t h a s been reported that tht'
m u l t i fi d us w a s i n a c t i v e i n re l a x e d s i t t i n g as wel l
as w h e n s u bj e c ts Wl're i n s tr u c t e d to ' s i t u p r i g h t '
( Va lenc i a a n d M u n ro I 9H5) . [n con tf<l s t, Do n i sch
and B a smajia n ( 1 972) repo r ted th a t the m u l ti fid u s
w a s a c ti v e i n stra i g h t u n supported s i t t in g , w h i ch
accords w i th a ton ic a n ti g r a v i ty f u n c t i o n . The d i f
fe rence in res u l t s between the two s t u d ies m a y
re l a te to the w a y i n w h ic h s u b j ec t s a ss u m ed
a n up r i g h t s i t t i n g pos tu rl', a n d t h i s beco m e s a n
i mp o rta n t point i.n the c l i n i c a l re-e d u c a t i o n o f
u p r i g h t pos tu ra l posi t i o n .
A c t i v a t i on of the m u l ti fid us h a s been e x a mined
i n fo r w a rd trun k fl ex i o n a n d e x tension from the
fle xed p os i t i o n , trunk e x tcil sion in thl' p rone pos
i ti o n and tru rlk rota t i on . An arg ume n t c a n be p rc
sen kd t h a t the func tion of th � a c t iv i ty a p pea rs to
i n c l u d e pri m a r i l y one of s ta b i liza t io n . As till' s p i n e
b e n d s forw ,i rd f r o m the shl n d i n g p o s i t i o n , there i s
a n i nc rease i n m u l t i fid u s a c t i v i ty ( F l oyd a n d Silver
-1 95 1 , M or r i s et al 1 9 62, Pau l y 1966, ' Va lencia and
M u n ro 1985 ) . At a certa in p o i n t d u ring flex i o n , the
ac ti v i ty of the b a c k m u s c l es cea ses; t h i s is k n o w n
as the 'cri tica l poin t' ( Floyd a n d S i l v e r 1 95 1,
Morris et a 1 1 962, K i p pers a n d P a r ker 1 9 84, 1 985) .
The E M G a c t i v i t y of the l u m b a r erector spin a e
ceases a t about 90°/., o f l u mbar spine fl exion . The
c r i tica l poi n t for t h e m u l ti fi dus is not s u ch a char
ac te risti c ' fea ture as i t i s fo r the erector spinae
m u scles . A l though a d ec rease in a c ti v i ty is e v i d e n t ,
E M G s o f t he m u l tifi d u s show s i l ence i n freque n t l y,
i n contra s t to those for the lumb a r l o n g i s s i m u s
a n d ili ocosta l i s ( Va l encia and M u n ro 1 985) .
Ex ten si on of t he tru n k from the flexed posi t i o n
pred i c tably evokes h igh l e v e l s of mu l t i fidus a c t i v
i ty ( F l o y d a n d S i l v er 1 95 1 , Morr is et a l 1962, Pa u l y
1 966, D o n i s c h an d Basmaj ia n 1972) . M a rked a c tiv
ity o f the m u l t i fi d u s a l so occu rs when the tru n k i s
ex ten ded or h y perextended i n t h e p rone posi tion
( Pa u l y 1 966, J onsson 1 970, Don i s c h and Basmaj i a n
1972, Va l e n c i a a n d M u n ro 1 985) . Eve n t h o u g h , a s
h as been m e n t i o n e d , a c t i v i ty i n the m u l tif i d u s is
marked ill ex te ns i on , the maj o r i ty o f the a c t u a l
tru n k e x t e n s i o n torq u e ( 8 0% a t t h e L4 a n d L 5 ver
. tebra l levels) is p rov i d e d by the th ora cic com pon
ents o f th e erector s p inae m usc les (Bogduk e t a l
1 l)92b ) . The m u l t i fid us h a s been sho w n to be
il c tive bilateril i l y in both i p s i i J teril l a n d c o n t ra i J t
er,i1 ro tilt i on of the trun k i n s i t t i n g J nd sta nd i n g
(Morris l ' t a l 1 962, Pa uly 1 9 hh, J on sson 1 970,
L)o n i sc h il ll d Ba smaj i ,m I 972) . For this rcason, it
h,lS bcell s u ggested tlla t the 111 ul ti fid u s acts a s a
s til b i l i zL'r ril tlw r t h a n as a pri me mov e r d u ri ng
ro tl tion ( Va lenci,l J n d M un ro 1 985) .
A s a ge n e ra l observ a t i o n i n m o v e m e n t s t u d ies,
Uonisch and Basmaj i a n ( I Y72) re ported tha t J c t i v
i ty o f the m u l t i fi d u s W,l S re l ated to i t s pro posed
J c t i o n fo r on l y .')()0j., of the t i me. PJ u l y ( 1 966) a l so
s h o w L'd a l most contin u o u s a c t ivity d u rin g the
m a j or i ty of the d i fferen t d i rec tiona l activ ities
tes ted . These fi nd i ngs ca n be i n terpreted a s evi
dellCl' for a stabi l i zi ng role o f the mu l ti fi d u s rather
tha n a p r i m a ry rol e i n torq u e p rod u c t i on o n l y.
The o t h er i mporta n t role of t h e m ulti fid us i n
postu rl' a n d m o v e m e n t i s a p ropri oce p t i ve on e . In
s t u d ies on h U m JllS, thL' i mportance o f the m u l t i
fi d us as J m u sc l e thJ t p ro v i d e s fee d b a ck o n s p i nal
pos i ti o n has bcell high ligh ted . Sol omonow et al
Pa rasp i n a l mechan i sm a n d s u pport of the lumbar spine 7 1
( 1 998) have p ro p osed the ex i stence o f a fee d b a c k
system fro m the sp i n a l li gaments to the m u l ti fi d u s .
Certainly, t h e suprasp i n o u s a nd i n tersp i no u s l i ga
m e n ts con t a i n free nerve end i ngs, R u ffi n i cor
p u scles and Paci n i a n corpuscles ( Ya h i a et al 1 988,
J i ang e t al ] 995). Positi on-rep osition tasks h a v e
b e e n u s e d fo r e v a l u a t i n g the pos i t i o n sense of the
s p ine, and it has been shown tha t young h e a l thy
ind ivid ua l s a re capab le of re p osi t i o n i n g the p e l v i s
and back (abso l u te error o f ap prox ima tely 2 d egrees)
both d u r i n g s ta n d i ng a n d s i t t i n g ( B r u m a gne et a l
1 999a,b ) . Both e x e rcise- i n d u ced fa tig ue and mech
anical l o w b a c k pain h ave a de leterio us e ffect o n
l u m bosacra l position ing a c c u r a c y ( B r u m agne e t a l
20(0). I t is tho ugh t th a t the degraded pos i tion
sense res u l ts from a ltered m u l ti fidus mu scle sp i n
d l e a fference a n d centra l processing of this sensory
i n p u t, and tha t d e fi c i ts i n the sp i n a l re fl ex s y s tem
migh t a ls o contri b u te .
Summary
The l u mb a r m u l t i fidus, l u mb a r longiss i mu s a nd
iliocos t a l i s p lay an i m porta n t rol e in l u m b a r spine
s ta b i l i ty. Bec a u se o f its uni que m orp h ol ogy a n d
segm en tal innerv a tion, the m u l tifid u s wou ld a p pea r
to be a musc le well su i ted to thjs role of segmen tal
support and controL B i o mecha nic al resea rch has
con fi rm ed this important ro l l' . The b i omecha n i c a l
s tud y by Wi l ke e t al ( 1 995), w h ich incl uded b o t h
the m u l tifi d u s and the erec tor spinae mu scles i n the
model, fo u n d tha t the mu l ti fi d u s had the s trongest
influence on l u m ba r seg m e n t a l s t a b i l i ty. T h e mor
phology of the mu l ti fi d us, o u r c l i ni ca l fi n d i n gs of
a d y s fu n c ti o n i n the segme n ta l m u l t i fid u s ( H ides
e t al 1 994, 1 996b) and work such a s th a t by Wi l kc
l't al ( 1 YL)5) J I l p rovi d e a b a s i s for focusing spec i fi
ca lly o n the l u m bar m u l tifi d u s i n pa tie n ts w i th l o w
b a c k p'l i n .
Q U A D RAT U S L U M B O R U M
( M E D I A L F I B R E S)
ae=w __���____________=_ �__�__
The q U J d r Cl t u s l u mb o r u m co nsists of sev erJ I l a m
i n a e and is e n c l osed by the a n t e r i o r a nd m i d dle
la yers o f the thoracol umbilr f" sc i a ( 13ogduk I 9Y7,
Wi ll i a m s et a1 1 9H9) (Fig. 4 . 5 d ) . The med i a l p o r t i on
of the muscle runs from the i lium to t h e a n lL'r i o r
surfa ce of the tra nsve rse processes of the l u m b J r
7 2 T H E JOINT PROTECTION M E C H AN I SM S
ver tebra e, a nd other fib res tra v el f r o m t h e trans
verse processes to a nchor o n to t he twel fth rib. The
la tera l portion of the mu scle, which belongs to the
gl oba l system, spans the lumbar a rea, a ttach ing on
the l a tera l i l i u m to i nser t into the twel fth rib with
ou t a ttach men t to any vertebrae. The l a tera l fib res
p rod uce p ri m a r i ly a la tera l bending momen t . The
m ed ia l portion, whi le unlike ly to make a s ubstan
tia l contribu tion to lateral flex ion ( Bogd uk 1 997),
i s c a pa b le of p rov i d i ng segmen ta l s tab i l ity via i ts
seg m e n tal attac h ments ( McG il l et a I 1 996). Stud ies
i nv es tiga t i ng the p a ttern of acti v a tion of the qu ad
ratus l u m b orum in functional tasks h a ve been
l i m i ted beca use the depth o f this musc l e means
tha t i nvas ive EMG tech n iqu es a re requ i red
( A n dersson et a l 1 996, McG ill et al 1 996). In ad d
i tion, needle inser tion for fine-wire EMG is both
u np lea sa n t and p a i n fu l bec a u se of the thic kness o f
the fascia s u rrou n d i n g t h e muscle (P. W. Hod ges,
M. Comer ford , C. A . Rich ardson, unpublished
observa tions 1 995) . I n two recen t s t u dies, wh i c h
did u se fine- w i re E M G , record ings were made
fro m a m id portion of the m uscle, but the re was no
c l ea r indication of whether a c t i vity was recorded
from t h e l a te ra l or medial portion o f the muscle.
McG i l l et a l ( 1 996) provid ed evidence t h a t the
qu a d ra tus l umb orum p l ays a si gn ifican t role i n the
stab i li ty of the spine . Muscl e activ ity was mea s u red
dur i ng a symmetricdl b u cket-holding task. Activi ty
i nc reased wi th in c rea sing spinal compression
p ro v id ed th rough progressive a x i a l loa d i n g . F u r
ther evidence for the generil l s tabilizing role of the
q u a d rLl t us l u mboru m was p rovi d ed by Ande rsson
et al ( 1 996), w ho found tha t, u n l i ke the e rec t o r
sp i na e ( Kipper a nd Parker 1 984), there w a s no
elec tric a l s ilence of th e m u scle i n full-fo rw a rd
fl e x ion .
W h i l e th e res u l ts of these two st udies su pport
the thesis for a s tab il iz i ng role for the qu ad ra tu s
l u m b o r u m , we regilrd this m u sc le dS a gl oba l sta
bi l i z i ng m u scle, cdpilble of con troll i n g the extern a l
IOdds p laced o n the spine . I n teres ting ly, i n pa ti en ts
w i t h bdck p a in , overactiv ity, tightness a n d tr i gge r
po in ts il re o f ten rep or ted by c l in icians (Jan da 1996,
Trave U a n d Simons 1 983). Tre<l t m e n t is foc used on
decreil sing activ i ty in the q uadra tus l u mboru m
rilthe[' th a n i nc reilsi ng i t wi th exerc i s e . The m ed i a l
p o r t i o n o f the q u a d ra t u s l u m bo ru m m a y i n t h e
fu tu re be s how n t o be fun cti ona l l y se para te to the
l a te ra l p a rt o f the muscle a n d con tribu te d i rec t l y to
the segm en tdl s u p p o r t of the s p i n e .
PARAS P I N AL MECHAN I S M AN D SU P P O RT
O F T H E LU MBAR S P I N E : CL I N I CAL
R E LEVA N CE
The key morphologica l fea tures of the m u l ti fi d u s
tha t can be rela ted to a ro l e i n l u m b a r sta b i l i ty a re
the segm en ta l a rrangement of the m u l ti fid u s fas
cicles ( M a c intosh e t al 1986), the l a rge size of
the mul ti fid us a t the lu m bosac ra l j u n c tion a n d the
c lose re l a tions h ip between the m u l tifi d u s a n d the
zyga pophysea l j o i nts . The segmenta l mo r p h o l ogy
of the m u l t i fidus h a s im pl ica tions for rehabi l i ta
tion . As i t is a m uscle w i th se g m e n t a l i nn e rv a tion
a n d con trol , reh a bi l i ta tion can be a i med il t speci fic
segm en ts .
The relationsh ip b e tw een the m u l t i fi d u s and the
thoracolumba r fascia is a lso importa n t for rehabili
ta tion, a s i t has been proposed tha t con tra c tion
of the m u l t i fidus a n d the l u mb a r e rector s p i nae '
tensions the fa scia. I t is ou r contention tha t the
muscles of the abdomin a l corse t tha t ten sion the
fasc i as fu l fil l a v i ta l role in p rotectio n of the l u m
bopelvic region . Th js mecha n jsm is descri bed a n d
illustra ted in Cha p te r 5 .
The mul ti fid u s has a high pro por tion of type I
mu scle fibres and is w e l l perfu sed, mean i n g thJt
it is we l l su i ted to a tonic, hold i n g fu n c t i o n . EMG
a na l ysis has p rov i d ed fur ther evid ence o f a ton ic
or a l most conti n u o u s a c t i v a t i on o f the m u l t i fi d u s ,
partic u larly the deep fibres. Exerc ise therapy should
be a imed a t restorin g th i s function, a n d a n empha
sis on a c tiv a tion of the d eep fi bres for sta b i l i ty i s
j us tified . Re-education o f the mu l t i f i d u s t o p ro m o te
segmenta l s ta bi l i zation is ex tremely subtle. It is a n
i m p orta n t p r i nc i ple f o r thera pe u tic exercise that
isometric con tra ction o f the d eep m u l ti fi d us occ u rs
w i thout spinal or pl'lvic move men t. Patients sho u l d
be ab le to contrilct � h e m u s c l e slowly.
It has been proved tha t the m u l ti fid us con t rols
the neu tra l zone a n d increasl' s segmental sti ffness.
It h a s been prop osed tha t contractions o f only a
very sma l l percen tage o f the maxi m u m vo l u n ta ry
contrac tion are req u i red to i n c rease segmenta l
sti ffness (C hol ew ick i a n d McC; i l l 1 9%). F ur th e r
more, the d eep ftb res of m u l t i fi d u s a re p l a ced
c l ose to the cen tres of ro tation of s p i n a l movement.
Th i s mean� that the length of the muscle fibres
rem a ins un c h a n ged in any physiological posture .
The i m p l ica t i o n fo r reha b i l i tation is that iso
m e t r i c contractions o f a very s m a l l percen tage of
the m a x i m u m v ol u n ta ry contrac tion a re a p prop ri
a te , a s only sma l l a m ounts of muscle c o n t r a c ti on
w i l l be req ui red to p rotect the verteb ra l segme n t .
Per i p h e r a l feed b a c k from t h e j o i n t a n d l i g a m e n t
a fferents a ffec ts t h e ga m m a s p i nd l e system
U o h ans�on e t al 1 99 1 a ), and the gamma system
b i a ses t h e sp i n d l e towards a n increased sensi tiv
i ty. For th is reason, sensory techniques tha t affe c t
t h e j o i n t a n d m u s c l e receptors are i m p ortan t i n
reha b i l itation of the m u l t i fi d u s . F u rtherm ore, the
role o f the m u l t i fid us i n m a i n tenance of the
Pa rasp i n a l mech a n i s m and su pport of the l u m b a r s p i n e 7 3
normal lordosis p rov ides a no ther i m porta n t goa l
of reh abi l i ta tion, as i t is i m porta n t th a t the spine
c a n be a dj usted p recisely to m a tch the a p plied
loadin g . M a intena nce of the l u m b a r lord osis
req u i res i n tegration o f loca l, one joint (a ntigra v i ty)
a n d. m u l tijoi n t / multi fu n c ti o n m usc les . Activa tion
of glob a l m u scles alone p ro v i des ins u ffi cient con
trol o f the lord osis. As activa tion o f the loca l m u s
c l e s (80'10 of the con t r i b u t i on from the m u l t i fi d us)
decreases fo rces i n the global system, p rov ides
s tiffness, increases sta b i l i ty and increases compres
sion (Keifer et a 1 1 997, 1 998), the m u l t i fi d u s m u s t
b e re-ed uca ted i n t h i s ro le fo r success f u l 'rehabili
tation. Integra tion of loca l , one j o i n t ( a n ti grav i ty )
a nd m ultijoin t / m u l tifu nc tion m uscles m u s t be
u n der taken fo r o p t i m a l res ults.