-
British Journal of Ophthalmology, 1989, 73, 68-75
Congenital adduction palsy and synergisticdivergence: a clinical
and electro-oculographic studyJOHAN R M CRUYSBERG,l ALI T MTANDA
,'KIRSTI U DUINKERKE-EEROLA,' AND PATRICK L M HUYGEN2
From the 'Institute of Ophthalmology and the 2Institute of
Otolaryngology, University ofNijmegen, Nijmegen,The Netherlands
SUMMARY We studied two patients with a peculiar congenital
disturbance of ocular motility inwhich the horizontal movements of
the left eye were always opposite the normal expecteddirection. The
common features were: (1) congenital monocular adduction palsy and
exotropia ofthe left eye; (2) simultaneous abduction of both eyes
(divergence) on attempted dextroversion; (3)ocular torticollis,
head turned to the right; and (4) inverse nystagmus of the left
eye, occurringspontaneously as well as during optokinetic and
vestibular testing. Clinical and electro-oculographic findings
suggested a close relationship to Duane's retraction syndrome and
supportedthe concept that innervational mechanisms were responsible
for the phenomenon.
Congenital unilateral paralysis of adduction associ-ated with
simultaneous bilateral abduction onattempted gaze into the field of
action of the pareticmedial rectus muscle is a rare entity. We have
beenable to find nine reported cases.'-7 Although themajor defect
is a congenital failure of adduction, themost striking clinical
finding is that on attemptedadduction the affected eye moves
further into abduc-tion and thus causes extreme divergence.
In this communication we briefly review thefeatures of the
reported cases and analyse the clinicaland electro-oculographic
findings in two personallyexamined cases.
Case reports
CASE 1A 3-year-old girl was referred to the Institute
ofOphthalmology because of congenital adductionparalysis and
variable exotropia of the left eyeassociated with torticollis to
the right. Neurologicalexamination at the age of 8 months revealed
nosignificant abnormalities except for borderline
skullmeasurements.We examined the child in 1981. The visual
acuity
was RE 20/25 and LE 20/80. Cycloplegic refractionCorrespondence
to Dr J R M Cruysberg, Institute of Ophthal-mology, Sint Radboud
Hospital, PO Box 9101, 6500 HBNijmegen, The Netherlands.
68
was +3-5 D in both eyes. There was a permanenthead turn to the
right; this kept the eyes straight (Fig.1A). With forced
straightening of the head, the righteye abducted beyond the midline
while the left eyeremained exotropic (Fig. 1B). Attempted
dextrover-sion resulted in first, extreme exotropia due
todivergence of both eyes (Fig. IC); secondly, simul-taneous
abduction nystagmus in both eyes with thefast phase beating
temporally; and, thirdly, narrow-ing of the left palpebral fissure.
On levoversion theright eye adducted normally and the extropia of
theleft eye decreased, so that the eyes became parallel(Fig. 1D).
Adduction of the left eye beyond themidline could not be provoked
by the oculocephalicmanoeuvre or optokinetic stimulation.No
pupillary reactions accompanied any of these
ocular movements. Under general anaesthesia theforced ductions
confirmed the increased resistance toadduction of the left eye.
Abduction of the left eyeand the forced ductions of the right eye
were normal.Audiometry, x-ray films of the skull and cervicalspine,
and cerebral computed tomography werenormal.
CASE 2A 23-year-old male with multiple congenital abnor-malities
described as arthrogryposis multiplexcongenita was referred to the
Institute of Ophthal-mology in 1983 because of congenital exotropia
and
on April 5, 2021 by guest. P
rotected by copyright.http://bjo.bm
j.com/
Br J O
phthalmol: first published as 10.1136/bjo.73.1.68 on 1 January
1989. D
ownloaded from
http://bjo.bmj.com/
-
Congenital addiction palsy and synergistic divergence
.
.:::
IC B A D
Fig. 1 Case l. A: Lookingforward with permanent head turn to the
right. B: Exotropia on forced straightening ofthe head.C: Extreme
exotropia on attempted gaze to the right. Note that the left eye
movesfurther into abduction, and there is narrowingofthe left
palpebralfissure. D: Decreased exotropia ofthe left eye on gaze to
the left. Note that the exoptropia ofthe left eye ismaximal on gaze
to the right and minimal on gaze to the left.
adduction paralysis of the left eye. At previousstrabismus
surgery, under general anaesthesia, hewas found to have a short and
tight left lateral rectusmuscle and a thin and atrophic left medial
rectusmuscle. Following maximal recession of the leftlateral rectus
muscle and 10 mm resection of the leftmedial rectus muscle the eyes
were parallel for ashort time, while the adduction palsy
persisted.
C ~~~~B
The visual acuity was 20/20 in the right eye and20/200 in the
left eye. Cycloplegic refraction was RE+ 1-0 D, and LE + 1.25 D.
Slit-lamp examination andfunduscopy were normal in both eyes. There
was ahead turn to the right (Fig. 2A). In the primaryposition there
was a left exotropia of 400 andhypotropia of 150 with absent
adduction of the lefteye (Fig. 2B). On attempted dextroversion both
eyes
I
I
D
A .
a:A'~
P ToA
Fig. 2 Case 2. A: Lookingforward with permanent head turn to the
right, exotropia and hypotropia ofthe left eye.B: Increasing
exotropia onforced straightening ofthe head. C: Extreme exotropia
on attempted gaze to the right. Note that theleft eye moves further
into abduction, and there is narrowing ofthe left palpebralfissure.
D: Decreased exotropia ofthe left eyeon gaze to the left. Note that
the exotropic position ofthe left eye is maximal on gaze to the
right and minimal on gaze to the left.
69
on April 5, 2021 by guest. P
rotected by copyright.http://bjo.bm
j.com/
Br J O
phthalmol: first published as 10.1136/bjo.73.1.68 on 1 January
1989. D
ownloaded from
http://bjo.bmj.com/
-
Johan R M Cruysberg, Ali TMtanda, Kirsti UDuinkerke-Eerola, and
Patrick L M Huygen
rig. 3 rlexion contractures ortepirst ana seconadngers inpatient
2.
abducted and became extremely divergent (Fig.2C); this was
associated with retraction of the leftglobe and narrowing of the
palperbral fissure. Onlevoversion the right eye adducted normally
and theexotropia of the left eye decreased (Fig. 2D). Theamounts of
exotropia and hypotropia of the left eyewere clearly less in
levoversion (Fig. 2D) than indextroversion (Fig. 2C). There was
underaction ofthe depressor and levator muscles of the left
eye.
General examination disclosed flexion con-tractures of the first
and second fingers of both hands(Fig. 3), a short atrophic
club-footed right leg, and adeformed left foot. The face was narrow
and the
Fig. 4 Absence ofleft lateral rectus muscle insertion, as
aresult ofprevious strabismus surgery, demonstrated duringthe
second surgical procedure in patient 2.
palate high. Otolaryngological examination dis-closed
deformation of the external ears. Audiometrywas normal. X-ray films
of the cervical spine andskull and cerebral CT scans were all
normal. Therewere no cardiopulmonary abnormalities and
theintelligence was normal.Because of the severe exotropia and the
ugly head
posture the patient requested another surgical inter-vention. A
large loop recession of the left lateralrectus and resection of the
left medial rectus wereplanned. Under general anaesthesia forced
ductiontesting of the left eye showed moderate restriction
ofadduction; forced ductions of the right eye were
Fig. 5A Fig. 5BFig. 5 Electro-oculographic recordings in case 1.
Tracings from top to bottom in Figs. 5, 6, 7A, 8 time (seconds),
right andleft eye position. Upward deflection signals eye movement
to the right; downward deflection signals eye movement to the
left.Position calibration was notfeasible in the case. Note that
the movement ofthe left eye is always opposite to the movement
ofthe right eye, and that the divergent-convergent nystagmus in the
primary position (A) is enhanced by gaze to the right
(righthalfofpanel B) and disappears on gaze to the left (left
halfofpanel B).
70
on April 5, 2021 by guest. P
rotected by copyright.http://bjo.bm
j.com/
Br J O
phthalmol: first published as 10.1136/bjo.73.1.68 on 1 January
1989. D
ownloaded from
http://bjo.bmj.com/
-
Congenital addiction palsy and synergistic divergence
110 t
Fig. 6A (left) Fig. 6A (right)
Fig. 6B
normal. It was also noted that there was severescarring of the
conjunctiva and Tenon's capsule as aresult of previous surgery.
After a limbal incision andcareful dissection the site of the left
lateral rectus wascarefully inspected. Despite a thorough search
deepinto the orbit no lateral rectus muscle was seen (Fig.4).
Inspection of the medial rectus muscle revealed avery thin,
atrophic muscle, on which a 6mm resectionwas performed.
Postoperatively there was practicallyno change in either the
deviation of the eye or thehead posture.
Methods
The procedure for recording and analysis of eyemovements has
previously been described.8 Tonniesequipment was used for both
optokinetic androtatory stimulation. Saccade velocity and
accelera-tion profiles were calculated from monocular eyepositions
sampled at 250 Hz during a 20° horizontalcalibration of saccades by
means of a method ofnumerical smoothing and differentiation.
Results
With forced fixation in the primary position case 1displayed
dysconjugate nystagmus with divergent
Fig. 6 Responses to evokednystagmus in case 2: right eye
givesnormal responses, left eye givesabnormal (opposite) responses.
A:Optokinetic stimulation; black-and-white stripes, each of
7-50width, moving at 40" per second, tothe left (left panel) or to
the right(rightpanel). B: Vestibularstimulation by
sinusoidalstimulation in darkness (maximumvelocity 28"/s at 0 05
Hz).Calibration applies to right eyeonly.
fast phases (Figs. 5A, B). The nystagmus increased inintensity
on gaze to the right and was virtually absenton gaze to the left.
Spontaneously left gaze up to 200was achieved through a
compensatory torticollis. Incase 2 a similar nystagmus was noted in
primary gaze.As in case 1, the nystagmus increased on gaze to
theright, but there was no neutral zone. In completedarkness no
spontaneous nystagmus occurred ineither patient.We used evoked
nystagmus to find out which eye
was correctly controlled by the ocular motor plant inthe brain
stem. In both patients this appeared to bethe right eye, for with
this eye both the slow and thefast phases of optokinetic and
vestibular nystagmuswere in the expected direction (Figs. 6A, B).
On theother hand the left eye moved in the oppositedirection, that
is, it adducted when the right eyeadducted and abducted when the
right eye abducted.Adduction in the present context implied only
amedially directed movement, since in both patientsadduction of the
left eye beyond the midline waslacking. Vertical optokinetic
nystagmus was normalin both patients.
Voluntary saccade analysis was possible only incase two (Figs.
7A, B). Correct calibration wasfeasible for the right eye only; the
calibration valuewas assigned also to the left eye channel because
in
71
on April 5, 2021 by guest. P
rotected by copyright.http://bjo.bm
j.com/
Br J O
phthalmol: first published as 10.1136/bjo.73.1.68 on 1 January
1989. D
ownloaded from
http://bjo.bmj.com/
-
Johan RM Cruysberg, Ali TMtanda, Kirsti U Duinkerke-Eerola, and
Patrick LM Huygen
I I.1 £
I I
Fig. 7AFig. 7 Voluntary saccade testing, case 2. A:
Electro-oculographic recording; the position step ofthe right eye
(2ndtracingfrom top) presumably is 200.
monocular viewing the left eye could not attain anamplitude of
200. Abduction of the right eye was
04 7 b30 1 .rI..T
R
L
Fig. 7B
normal in the sense that the velocity and accelerationprofiles
had the expected shapes and the saccadeparameters were within
normal limits.9 On theassumption that calibration is also
applicable to theleft eye, adduction of both eyes was slow, that
ofthe left eye being more severely impaired. Theappearances of the
saccade profiles resembled thatencountered in cases of bilateral
internuclearophthalmoplegia (INO). This applied particularly tothe
acceleration profiles.During intense fixation both patients had
saccadic
back-to-back oscillations. This interesting phenome-non was more
prominent in the right eye of case 2.This eye showed frequency
sweeps, that is, a progres-sive increase of frequency of saccades
reaching afrequency of 12 Hz. The saccadic frequency increasewas
associated with a decrease in amplitude (Figs.8A, B).
Discussion
As reviewed in Table 1, congenital adduction palsywith
synergistic divergence is usually noted during
L
R.
C H 1
L
0.1 0. 1.~ 0.4 0. ---H1Cffi. TIME SEC
Fig. 7 B: Superposition display of20° saccadesfor right eye
(CHI) and left eye (CH2). Correct calibration applies to the
righteye only (see text). Adduction is clearly impaired in both
eyes relative to abduction. Allowing for an overallfilter effect
ofapproximately second-order low-pass with a cut-offfrequency of 15
Hz, the maximum abduction velocity ofthe right eye isnormal.
.AL A.
72
E E, I.- -4. 1 L. E D .-- --. i-, D E -4. N -4, L Y '--. I
'::,
on April 5, 2021 by guest. P
rotected by copyright.http://bjo.bm
j.com/
Br J O
phthalmol: first published as 10.1136/bjo.73.1.68 on 1 January
1989. D
ownloaded from
http://bjo.bmj.com/
-
Congenital addiction palsy and synergistic divergence
I!I 1I I I 1
Fig. 8AI ;
I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
_. I .I - 1 IoH
Fig. 8 Electro-oculographicrecordings during intensefixation
incase I (A) and case 2 (B). Noteback-to-back saccadic
oscillationsespecially in case 2, which alsoshows afrequency sweep
ofsaccades.~~~~~~~~A4~1100
Fig. 8B
childhood as an adduction deficit, involving the lefteye in 75%
of the cases, and associated with syner-gistic divergence during
attempted adduction of theapparently paretic eye. In the primary
position thereis always an exotropia of the affected eye and,
inthree of the 11 reported cases, a spontaneousabduction nystagmus.
Evoked inverse nystagmus ofthe affected eye has been demonstrated
in all casestested. The disorder shows a moderate male
pre-ponderance.The clinical findings in our two cases conform
to
this presentation. However, to the best of ourknowledge our case
2 is the first patient in whomarthrogryposis multiplex has been
recorded as anassociated congenital disorder.
Arthrogryposismultiplex is, on the other hand, seen with the
Mobiussyndromes a disorder which shows a close rela-tionship to
Duane's retraction syndrome both inpathogenesis," 12 and in its
association with seg-mental developmental anomalies. 13 Our
secondpatient is also unique in the sense that
synergistricdivergence was noted both clinically and
electro-oculographically, and yet at subsequent surgicalexploration
the lateral rectus muscle was found to beabsent. Although it is
most likely that this was a caseof a 'lost muscle' following the
first surgical interven-
tion, it is also equally clear that the observedsynergistic
abduction did not result from lateralrectus muscle innervation
alone. We believe that inthis case the abduction movement was
carried out asa secondary action of the inferior rectus
and/orinferior oblique muscles. We draw our assumptionfrom
anatomical evidence' and from evidencederived from two
histopathologically examined casesof Duane's retraction syndrome,
in which anomalousinnervation between the oculomotor nerve and
thelateral rectus muscle has usually involved the inferiordivision
of the oculomotor nerve.' 16 Such a mecha-nism is in accord with
the innervational hypothesisadvanced to explain synergistic
divergence.
Previously the apparent medial rectus muscle palsywas attributed
to internuclear ophthalmoplegia orselective destruction in the
medial rectus musclenucleus, whereas the abduction movement
wasattributed to anomalous connections between theabducens
nuclei.23 Our electro-oculographic study ofthese two patients
showed impairment of adductionassociated with bilateral abduction
nystagmus, andthe affected eye showed an inverse nystagmus
duringoptokinetic and vestibular testing. We were unable,however,
to demonstrate dysmetric waveforms of theabducting saccades.8'7 It
seems unlikely therefore
73
on April 5, 2021 by guest. P
rotected by copyright.http://bjo.bm
j.com/
Br J O
phthalmol: first published as 10.1136/bjo.73.1.68 on 1 January
1989. D
ownloaded from
http://bjo.bmj.com/
-
Johan RM Cruysberg, Ali TMtanda, Kirsti U Duinkerke-Eerola, and
Patrick LM Huygen
Table 1 Congenital unilateral adduction palsy and synergistic
divergence
Authors Patients Primary position Adduction Synergistic Evoked
Associated Aetiology: Proposedsex and age palsy divergence
nystagmus features mechanism for divergence
BArAny' F, 3 ±Straight, bilateral Left eye On right Inverse in -
Birth trauma connectionsabduction nystagmus gaze LE of L+R nucleus
VI
Worth and M, 3 Exotropia left eye Left eye On right - - Medial
rectus palsy allowsChavassel gaze superior+inferior
oblique to abduct theeye further
Burian and M, 5 Exotropia left eye Left eye On right Inverse in
- Anomalous innervationCahill3 gaze LE
Burian et al.4 F, 9 Exotropia right eye Right eye On right and
Covergent Mental Related to Duane'sleft gaze nystagmus retardation
syndrome
Znajda and M, 4 Exotropia left eye Left eye On right - -
Congenital anomaly ofKrill' gaze innervation
Wilcox et al.,6 M, 7 Exotropia right eye Right eye On left gaze
Inverse in - Related to Duane'sRE syndrome
Wagner et al. F, 5 Exotropia left eye Left eye On right Inverse
in - Variant of type 2 Duane'sgaze LE syndrome
F, 8 Exotropia left eye Left eye On right - -gaze
M, 4 Exotropia right eye Right eye On left gaze - -Cruysberg et
al. F, 4 Exotropia left eye, Left eye On right Inverse in - Related
to Duane's
(present bilateral abduction gaze LE syndrome. Oculomotorstudy)
nystagmus innervation intended
for the medial rectusmuscle supplies thelateral rectus muscleand
other extraocularmuscles
M, 23 Exotropia left eye, Left eye On right Inverse in
Arthrogryposisbilateral abduction gaze LE multiplexnystagmus
congenita
that internuclear disturbances were responsible forthese
patients' movement disorder. This is corrobor-ated by the finding
that the adduction had a normalduration, whereas in internuclear
ophthalmoplegia itis significantly prolonged.Wilcox et al.6 studied
by electromyography a case
of congenital adduction palsy with synergisticdivergence. They
found that limitation of adductionin their patient resulted from
anomalous and greaterinnervation of the antagonistic lateral rectus
muscle.This mechanism placed this disorder in the samecategory of
innervational disturbances as Duane'sretraction syndrome. Like
Huber's'8 Duane type IIcase, failure of adduction was regarded as
the majordefect. However, in the present cases, instead
ofretracting on adduction the eye moves further intoabduction.
Wilcox et al.6 proposed an innervational mecha-nism whereby the
abducens nerve was supposed to beeither hypoplastic or absent and
most of the oculo-motor innervation intended for the medial
rectusmuscle supplied the lateral rectus muscle. The motorsupply to
the medial rectus muscle arises from theinferior branch of the
oculomotor nerve, which alsoinnervates the inferior oblique and the
inferior rectusmuscles. This anatomical fact would explain the
maintenance of synergistic divergence by the inferiormuscles in
the absence of the lateral rectus muscle, asnoted in our second
patient.
In this patient, the slowing of adduction of the lefteye is
sufficiently explained by the finding of a thin,atrophic medial
rectus muscle, but the apparentlysimilar slowing of adduction
observed in the right eyecalls for another explanation. It might be
that, on theright side also, some of the innervation intended
forthe medial rectus muscle aberrantly reaches theinferior oblique
and the inferior rectus muscles. Thiswould then result in
cocontraction impairing theadduction of that eye to some extent,
but apparentlyless than in the left eye.Because synergistic
divergence shows such close
relationship to Duane's retraction syndrome, itshould also be
considered a developmental anomalycharacterised by absence of the
abducens nucleus andsubsequent innervation of the lateral rectus
muscleby the inferior branch of the oculomotor nerve,"' 16probably
owing to a teratogenic disturbance duringthe second month of
pregnancy.'9High frequency saccadic oscillations with back-to-
back saccades occur in voluntary nystagmus2" and inocular
flutter,2' an ocular movement disorder indicat-ing
pontine-cerebellar dysfunction. The occurrence
74
on April 5, 2021 by guest. P
rotected by copyright.http://bjo.bm
j.com/
Br J O
phthalmol: first published as 10.1136/bjo.73.1.68 on 1 January
1989. D
ownloaded from
http://bjo.bmj.com/
-
Congenital adduction palsy and synergistic divergence
of these movements in both our patients may beinterpreted as an
indication of the brain stem originof their movement disturbance.
However, it is alsopossible that the saccadic oscillations are
actually atype of fixation instability caused by the
peculiarinnervation which results in conflicting retinal
infor-mation being sent to the brain.
References
1 BArAny R. Ein Fall von monocularer Lahmung aller
seitlichenwillkurlichen Blickbewegungen, bei Intaktheit der
vertikalenBlickbewegungen, mit horizontalem Konvergenz-
undDivergenznystagmus im Bereich des fur die
Wilikurbewegungengelahmten Abducens. Arch Klin Exp Ohren
NasenKehlkopfheilkd 1930; 26: 237-44.
2 Worth C, Chavasse FB. Isolated paralysis of the medial
rectus.In: Lyle TK, ed. Squint. Philadelphia: Blakiston, 1950:
106.
3 Burian HM, Cahill JE. Congenital paralysis of medial
rectusmuscle with unusual synergism of the horizontal muscles.
TransAm Ophthalmol Soc 1952; 50: 87-102.
4 Burian HM, Van Allen MW, Sexton RR, Baller RS. Substitu-tion
phenomena in congenital and acquired supranuclear dis-orders of eye
movement. Ophthalmology 1965; 69:1105-14.
5 Znajda JP, Krill AE. Congenital medial rectus palsy
withsimultaneous abduction of the two eyes. Am J Ophthalmol
1969;68:1050-2.
6 Wilcox LM, Gittinger JW, Breinin GM. Congenital adductionpalsy
and synergistic divergence. Am J Ophthalmol 1981; 91:1-7.
7 Wagner RS, Caputo AR, Frohman LP. Congenital
unilateraladduction deficit with simultaneous abduction.
Ophthalmology1987; 94: 1049-53.
8 Huygen PLM. Vestibular hyperreactivity in patients
withmultiple sclerosis. Adv Otorhinolaryngol 1983; 30: 141-9.
9 Baloh RW, Sills AW, Kumley WE, Honrubia V.
Quantitativemeasurement of saccade amplitude, duration and
velocity.Neurology 1975; 25: 1025-70.
10 Wishnick MM, Nelson LB, Huppert L, Reich EW. Mobiussyndrome
and limb abnormalities with dominant inheritance.Ophthalmic Pediatr
Genet 1983; 2: 77-81.
11 Pitner SE, Edwards JE, McCormick WF. Observations on
thepathology of Mobius syndrome. J Neurol Neurosurg Psychiatry1965;
28: 362-74.
12 Towfighi J, Marks K, Palmer E, Vannucci R. Mobius
syndrome:neuropathologic observations. Acta Neuropathol (Berl)
1979; 48:11-7.
13 Gadoth N, Biedner B, Torok G. Mobius syndrome and
Polandanomaly: case report and review of the literature. J
PediatrOphthalmol Strabismus 1979; 16: 374-6.
14 Hoyt WF, Nachtigaller H. Anomalies of ocular motor
nerves.Neuro-anatomic correlates of paradoxical innervation
inDuane's syndrome and related congenital ocular motor dis-orders.
Am J Ophthalmol 1965; 60: 443-8.
15 Hotchkiss MG, Miller NR, Clark AW, Green WR. BilateralDuane's
retraction syndrome: a clinical-pathologic case report.Arch
Ophthalmol 1980; 98: 870-4.
16 Miller NR, Kiel SM, Green WR, Clarck AW. UnilateralDuane's
retraction syndrome (Type 1). Arch Ophthalmol 1982;100:
1468-72.
17 Crane TB, Yee RD, Baloh RW, Helper RS. Analysis
ofcharacteristic eye movement abnormalities in
internuclearophthalmoplegia. Arch Ophthalmol 1983; 101: 206-10.
18 Huber A. Electrophysiology of the retraction syndromes. Br
JOphthalmol 1974; 58: 293-300.
19 Pfaffenbach DD, Cross HE, Kearns TP. Congenital anomalies
inDuane's retraction syndrome. Arch Ophthalmol 1972; 88:635-9.
20 Coren S, Komada MK. Eye movement control in
voluntarynystagmus. Am J Ophthalmol 1972; 74: 1161-5.
21 Cogan DG. Ocular dysmetria, flutter-like oscillations of
theeyes, and opsoclonus. Arch Ophthalmol 1954; 51: 318-35.
Acceptedfor publication 3 December 1987.
75
on April 5, 2021 by guest. P
rotected by copyright.http://bjo.bm
j.com/
Br J O
phthalmol: first published as 10.1136/bjo.73.1.68 on 1 January
1989. D
ownloaded from
http://bjo.bmj.com/