The effect of spinal fusion on the long-term outcome of idiopathic scoliosis.

Full text

Page 1

306THE JOURNAL OF BONE AND JOINT SURGERY
Correspondence
©2003 British Editorial Society of Bone and Joint Surgery
doi.10.1302/0301-620X.85B2.14109 $2.00
J Bone Joint Surg [Br] 2003;85-B:306-9.
We welcome letters to the Editor concerning articles which
have recently been published. Such letters will be subject to the
usual stages of selection and editing; where appropriate the
authors of the original article will be offered the opportunity to
reply.
Letters should normally be under 300 words in length,
double-spaced throughout, signed by all authors and fully ref-
erenced. The edited version will be returned for approval
before publication.
The timing of reduction and stabilisation of the acute,
unstable slipped upper femoral epiphysis
Sir,
I read with interest the article by Phillips et al1 in the September
2001 issue entitled ‘The timing of reduction and stabilisation of
the acute, unstable, slipped upper femoral epiphysis’.
The unstable nature of the slip demanded that the stabilisation
be carried out urgently to prevent risk of compromise to the blood
supply of the epiphysis. Although 24 hours had been decided on as
being a safe period, I assume that these hips would have been sta-
bilised at the next available operating session. Therefore the period
of 24 hours is arbitrary. With a larger sample and more variable
prognosis a decision could be made as to whether the operation
was more urgent, e.g. within two hours of presentation, and thus be
graded with more practical consideration according to CEPOD sta-
tus.
The most interesting fact is that a reduction of 100% was
achieved in all unstable hips. This was probably because of the
acute nature of the slip. Two issues need to be resolved, namely,
the technique and amount of force used in achieving closed reduc-
tion and the point of abandoning this in favour of open reduction,
and secondly, the technique of open reduction used in order not to
stretch the posteriorly placed retinacular vessels during the process
of reducing the fragments which are often wedged against each
other.
J. K. BARBOSA, MS Ortho, FRCS
Hope Hospital
Manchester, UK.
1. Phillips SA, Griffiths WEG, Clarke, NMP. The timing of reduction
and stabilisation of the acute, unstable, slipped upper femoral epiphysis.
J Bone Joint Surg [Br] 2001;83-B:1046-9.
Author’s reply:
Sir,
We thank Mr Barbosa for his interest in our article.
Clearly, the imperative is to reduce the slip as soon as possible
but before 24 hours. The message in the paper was that a manipu-
lative reduction was acceptable up to 24 hours.
The reductions were achieved either by gentle manipulation or
repositioning. Certainly, no force was used and none of the hips
was over-reduced. Open reduction was reserved for those cases in
which a reduction could not be achieved and yet it was clear that
there was an acute slip.
N. M. P. Clarke, CRM, FRCS
Southampton General Hospital
Southampton, UK.
Subtalar distraction bone block arthrodesis
Sir,
We read with interest the article by Trnka et al1 in the August 2001
issue entitled ‘Subtalar distraction bone block arthrodesis’.
The authors mentioned 20 feet with avascular necrosis (AVN)
of either the talus or the calcaneus. How do they definite AVN? In
our series sclerotic bone, found in almost all cases, facilitated
stable fixation.2 Is it possible to expose the anterior aspect of the
subtalar joint using the described approach? Our approach allows a
view of all parts of the former subtalar joint. The three corticocan-
cellous grafts correct the deformity creating intrinsic stability even
without screw fixation.
Measurement of the talocalcaneal height without the use of a
heel-jig could lead to errors because of divergence of the x-ray
beam, since the distance between the cassette and the foot often
differs.3 Placing a radiopaque ruler behind the heel can help to cal-
culate the true heel height from a weight-bearing radiograph.2
How do the authors explain the loss of ankle movement? Were
there signs of arthritis of the tibiotalar joint? Our series2,4 showed
that subtalar or triple arthrodesis has little adverse influence on the
function of the tibiotalar joint, even after many years.
Although the authors discuss the prominence of internal fixa-
tion we believe that this not only affects the results, but is associ-
ated with disruption of the heel pad. The ‘smashed heel pad
syndrome’ has been described as a reason for late pain in the heel
after fracture of the os calcis by other authors.5,6 We think that
placing screws through this already ‘smashed’ heel pad could fur-
ther damage it. By using one lag-screw with a washer from the
neck of the talus to the calcaneus, we have not seen secondary set-
tling within the hindfoot during incorporation of the graft.2
R. W. POOLMAN, MD
R. K. MARTI, MD, PhD
Academic Medical Centre
Amsterdam, The Netherlands.
1. Trnka H-J, Easley ME, Lam PW-C, et al. Subtalar distraction bone
block arthrodesis. J Bone Joint Surg [Br] 2001;83-B:849-54.
2. Marti RK, de Heus JA, Roolker W, Poolman RW, Besselaar PP. Sub-
talar arthrodesis with correction of deformity after fractures of the os
calcis. J Bone Joint Surg [Br] 1999;81-B:611-6.
3. Daffner RH. Clinical radiology: the essentials. 2nd ed. Baltimore, etc:
Lippincott Williams & Wilkins, 1999:8.
4. de Heus JA, Marti RK, Besselaar PP, Albers GH. The influence of
subtalar and triple arthrodesis on the tibiotalar joint: a long-term follow-
up study. J Bone Joint Surg [Br] 1997;79-B:644-7.
5. Pozo JL, Kirwan EOG, Jackson AM. The long-term results of con-
servative management of severely displaced fractures of the calcaneus.
J Bone Joint Surg [Br] 1984;66-B:386-90.
6. Lim EV, Leung JP. Complications of intra-articular calcaneal fractures.
Clin Orthop 2001;391:7-16.

Page 2

CORRESPONDENCE307
VOL. 85-B, No. 2, MARCH 2003
Authors’ reply:
Sir,
We thank Dr Poolman and Dr Marti for their interest in our article.
We noted that avascular changes of the subtalar joint were not
uncommon after trauma. These do not amount to avascular necro-
sis as we understand it, for example, with respect to the body of the
talus. Nonetheless, in the 20 feet with avascular changes, this term
was applied specifically to the hard, dense, and sclerotic appear-
ance of the subchondral bone surface which did not bleed at opera-
tion. They found that sclerotic bone in their series aided fixation,
and this is an accurate observation because sclerosis of this region
of the subtalar joint does facilitate stable fixation. Using our
approach to the subtalar joint, the posterior facet and the sinus tarsi
were visible and accessible. The talar calcaneal height was meas-
ured in a standardised manner. While I agree that there could be
some error in assessment because of the distance between the x-
ray cassette and the foot, all radiographs were obtained in a stand-
ardised fashion, before and after operation. Therefore, the differ-
ences in measurement remain constant despite the slight variations
which may have been introduced because of diversions of the x-ray
beam.
Loss of ankle movement after subtalar distraction bone block
arthrodesis has been noted by us and other authors. This is proba-
bly a result of tightening of the gastrocnemius soleus complex
with insertion of the tricortical bone block used for distraction.
This did not correlate with any findings of arthritis in the tibiotalar
joint. The statement that triple arthrodesis has little adverse influ-
ence on the function of the tibiotalar joint is not consistent with
our findings or with the literature, which records an incidence of
54% of arthritic changes in the tibiotalar joint after the triple
arthrodesis. I agree with the concept of problems of internal fixa-
tion introduced under the plantar aspect of the heel pad for subta-
lar arthrodesis. In fact, the introduction of a screw from the
undersurface of the heel should be avoided for this very reason.
The screw is introduced from the posterior aspect of the heel
immediately inferior to the insertion of tendo Achillis. While
efforts are made to bury the head of the screw flush with the pos-
terior cortical surface of the calcaneus, this is not always applica-
ble, hence the need for removal of the hardware in certain
patients.
M. S. MYERSON, MD
Union Memorial Hospital
Baltimore, USA.
N. M. P. CLARKE, ChM, FRCS
Southampton General Hospital
Southampton, UK.
Tendinopathy of tendo Achillis
Sir,
With reference to the review article in the January 2002 issue by
Maffulli and Kader entitled ‘Tendinopathy of tendo Achillis’,1 I
developed bilateral fusiform tendinopathy of tendo Achillis at the
age of 41 years after running for fitness, including marathons, for
22 years.
After reviewing the results of surgery published in the litera-
ture, and based on my own experience, I decided that surgery for
this MRI-proven diagnosis was not an option for treatment.
I therefore stopped running and lost 8 kg in weight. There was
total resolution of all symptoms and signs 18 months later.
After three years and having gained some weight I began run-
ning for fitness once again. The same problem recurred and was
confirmed by MRI. I lost 8 kg in weight and stopped running. After
30 months there was complete resolution of all symptoms and
signs.
Six years later I am completely free from symptoms, my basal
metabolic index remains at 24, and alternate cardiorespiratory pro-
grammes leave me with a resting pulse of 52. My conclusion is
that surgery for fusiform tendinopathy induced by sport is proba-
bly only successful because of the resultant physical restriction
and rest which it enforces.
P. GIBLIN, FRACS, FA Orth A
Goulburn Street Medical Centre
Liverpool, UK.
1. Maffulli N, Kader D. Tendinopathy of tendo Achillis. J Bone Joint Surg
[Br] 2002;84-B:1-8.
Author’s reply:
Sir,
There is little doubt that if the level and intensity of the offending
activity are reduced, then attenuation or resolution of the symp-
toms of a tendinopathy can occur. However, active individuals,
training and competing either as amateurs or professionals at a
high level, cannot or do not wish to change their activities. In these
patients, after an appropriate trial of conservative management,
surgery is an option. Our selection criteria for operative manage-
ment are strict, and reflect our experience as orthopaedic surgeons
and as athletes. Finally, notwithstanding the fruitful experience
that Mr Giblin has had personally, the only way to produce the ulti-
mate answer to his experience is to perform a randomised, control-
led trial which would allow us to formulate evidence-based rather
than opinion-based decisions.
N. MAFFULLI, MD, MS, PhD, FRCS Orth
Keele University School of Medicine
Hartshill, UK.
The effect of spinal fusion on the long-term outcome of
idiopathic scoliosis
Sir,
We have read with interest the article in the November 2001 issue
by Parsch et al1 entitled ‘The effect of spinal fusion on the long-
term outcome of idiopathic scoliosis’. We agree with the need to
compare the outcomes of patients having spinal fusion with those
of patients managed non-surgically. Such outcomes are crucial for
the advice given to patients with scoliosis in relation to the advan-
tages and the disadvantages of surgery.
Their case-control study suggests that patients with and without
spinal fusion have similar long-term outcomes. The two groups,
however, differ not only in the pattern of the curve, but also in the
size of the curve at the time of management. Since both groups had
similar Cobb angles at follow-up, the fused group presumably had
much larger curves at the time of surgery (86˚ in 30 patients) and
would have had even larger curves had they not been treated surgi-
cally. Although there was no significant correlation between the
Cobb angle and a lower spinal score in either group, this may be
the case for larger curves.
The authors conclude that “multisegmental spinal fusion does
not have an adverse effect on the long-term functional outcome”.
The length of follow-up was a minimum of five years but the mean
duration of follow-up for each group is not mentioned. Examina-
tion of the mean age of the groups at the time of assessment shows
that the mean follow-up period is evidently much longer. Their
conclusion should be viewed with caution given the short follow-

Page 3

308CORRESPONDENCE
THE JOURNAL OF BONE AND JOINT SURGERY
up period in some patients and the differences between the two
groups (curve site and curve size at the time of management).
Likewise, when considering degenerative changes as outcome
measures the inclusion of patients with a follow-up as short as five
years may be misleading.
Another compounding factor which is not declared may be the
inclusion of patients with infantile and juvenile idiopathic scoliosis
in each of the groups. Early studies on the mortality in idiopathic
scoliosis included patients with infantile, juvenile and adolescent
idiopathic scoliosis and drew conclusions which were not applica-
ble to patients with adolescent idiopathic scoliosis.2,3 The pattern
of the curve may also affect the findings and it would be useful to
know whether levels of pain and functional activity varied by the
site of the curve.
A. A. COLE, BMedSci, FRCS (Trauma & Orth)
R. G. BURWELL, MD, FRCS
R. K. PRATT, MA, FRCS
J. K. WEBB, FRCS
University Hospital
Nottingham, UK.
1. Parsch D, Gaertner V, Brocai DRC, Carstens C. The effect of spinal
fusion on the long-term outcome of idiopathic scoliosis: a case-control
study. J Bone Joint Surg [Br] 2001;83-B:1133-6.
2. Nachemson A. A long term follow-up study of non-treated scoliosis.
Acta Orthop Scand 1968;39:466-76.
3. Nilsonne U, Lundgren KD. Long-term prognosis in idiopathic scolio-
sis. Acta Orthop Scand 1968;39:456-65.
Author’s reply:
Sir,
We thank Mr Cole and his colleagues for their interest in our paper.
They have raised some important points. There is a difference
between the effectiveness and the effect of an extended spinal
fusion. The latter is supposed to answer the question: How does a
fusion affect spinal function? We agree with Mr Cole and his col-
leagues that, since the curve will be improved after surgery, the
effectiveness and the effect of a fusion cannot be evaluated within
the same study, at least not in a matched-pairs analysis including
the curve size as matching parameter. Hence, for the purpose of
our study, outcome parameters, i.e. the Cobb angle at follow-up,
have to be matched. This implies on the other hand that our results
are not supposed to impact directly on the indication in favour or
against surgery, but to support the orthopaedic surgeon in counsel-
ling the patient. Multisegmental spinal fusion, if indicated and suc-
cessfully performed, will not have an adverse effect on the long-
term functional outcome. A 40˚ scoliotic spine at follow-up is
functioning comparably whether it was previously fused or not.
With regard to the mean duration of follow-up, this information
(23 years) was lost during the editing process.
We agree that degenerative changes as outcome measures with
a follow-up as short as five years may be misleading. That is why
we included the patients’ age as a matching parameter. As a conse-
quence, younger patients in both groups with a shorter follow-up
were only compared with each other.
Finally, Mr Cole raised an important point regarding different
types of scoliosis and the time of onset of idiopathic scoliosis. The
papers quoted1,2 included different types of scoliosis, so that only
45% actually had an idiopathic scoliosis.3 This is why their con-
clusions were not applicable to idiopathic scoliosis. In our study
only patients with idiopathic scoliosis were included. Neverthe-
less, the onset of idiopathic scoliosis, especially early onset, is of
importance when evaluating the long-term outcome.3 Here we had
to rely on the information given by the participating patients. We
excluded infantile idiopathic scoliosis, but did not differentiate
between juvenile and adolescent types. This must be considered
when interpreting our results.
The impact of the site of the curve on the long-term outcome
was not evaluated. The inclusion of a fourth matching parame-
ter would have markedly reduced the number of patients within
our groups. We agree, however, that this would have been of
interest.
D. PARSCH, MD
The Orthopaedic University Hospital
Heidelberg, Germany
1. Nachemson A. A long-term follow-up study of non-treated scoliosis.
Acta Orthop Scand 1968;39:466-76.
2. Nilssone U, Lundgren KD. Long-term prognosis in idiopathic scolio-
sis. Acta Orthop Scand 1968;39:456-65.
3. Weinstein SL. Natural history. Spine 1999;24:2592-2600.
The floating shoulder: a multicentre study
Sir,
Our journal club read with interest the paper by van Noort et al1 in
the August 2001 issue entitled ‘The floating shoulder: a multicen-
tre study’.
We note that 28 patients were treated non-operatively and seven
operatively, all with plate fixation of the clavicle. None had inter-
nal fixation of the scapular glenoid. The results for both groups
were similar. Non-operatively treated patients with caudal migra-
tion of the scapula did relatively badly.
In the discussion the authors state that: “for a fracture of the
clavicle and a caudally displaced fracture of the neck of the scap-
ula, we recommend operative treatment……plate fixation of the
clavicle alone does not always seem to be sufficient….in the
absence of caudal dislocation of the glenoid, conservative treat-
ment will give a good functional outcome”.
No patient underwent operative fixation of the scapula.
There was no comparison of non-operative and operative
treatment for patients with or without caudal migration of the
glenoid. We feel that the authors have not established that
fixing the scapula gives better results than non-operative treat-
ment for patients either with or without caudal migration of
the glenoid. We do not feel that the results shown justify the
implied recommendations to fix the scapula with caudal
migration.
S. COLERIDGE, MRCS
D. RICKETTS, FRCS Orth
Princess Royal Hospital
Haywards Heath, UK.
1. van Noort A, te Slaa RL, Marti RK, van der Werken C. The floating
shoulder: a multicentre study. J Bone Joint Surg [Br] 2001;83-B:795-8.
Authors’ reply:
Sir,
We thank Mr Coleridge and Mr Ricketts for their interest in our
article.
Our findings justify the recommendation of conservative treat-
ment for patients with a fracture of the clavicle and ipsilateral
scapular neck without caudal dislocation.
The results of conservative treatment were poor in the case of a
caudally displaced glenoid. We agree that operative treatment for
the latter groups, based on our study, is debatable. Although no
comparison of operative and conservative treatment is possible, we
recommend fixation of the clavicle in patients with a fracture of the
clavicle and ipsilateral scapular neck with caudal dislocation. If
there is persistent caudal dislocation of the glenoid after fixing the
clavicle, we suggest reduction and fixation of the scapular neck, in

Page 4

CORRESPONDENCE309
VOL. 85-B, No. 2, MARCH 2003
the hope that the results may be better. This advice was indeed
based on assumption and not on fact.
A. VAN NOORT, MD
University Hospital of Nijmegen
Nijmegen, The Netherlands.
J. S. MEHTA, MS Orth, D Orth, MCh Orth
Gosforth
Newcastle upon Tyne, UK.
Specificity of the Oxford knee status questionnaire
Sir,
I read with interest the article in the April 2001 issue by Harcourt
et al1 entitled ‘Specificity of the Oxford knee status questionnaire’.
Unfortunately, aspects of this article are confusing and clarification
is required.
The authors state that each question in the Oxford knee score is
scored from 0 to 4. The items are summated so that 0 represents
the worst possible status and 48 a normal knee. The Oxford Knee
Score is actually scored from 1 to 5.2 This in itself would not be a
major problem as a percentage of the score could just be added.
However, they have also reversed the interpretation of the scoring
since an increasing score reflects worsening symptoms according
to Dawson et al.2
The authors conclude that the coexistence of hip or spinal
pathology will significantly alter the absolute score and any
change in score after operation. Indeed, the absolute score is very
important and hence extreme clarity in the use of scoring systems
is required to prevent confusion and to confirm validity. Common
sense would suggest that the data presented could be plausible.
Unfortunately, since their conclusions are based on data which, if
scored correctly, reflect the opposite scenario, they cannot be
accepted as valid without clarification.
I realise that this may be a simple misunderstanding on their
part, but request that their scoring methods and how such data have
been interpreted be clarified, in order to substantiate their findings.
L. UNITT, MCSP, SRP
Birmingham Heartlands & Solihull NHS Trust
Birmingham, UK.
1. Harcourt WGV, White SH, Jones P. Specificity of the Oxford knee sta-
tus questionnaire: the effect of disease of the hip or lumbar spine on pa-
tients’ perception of knee disability. J Bone Joint Surg [Br] 2001;83-
B:345-7.
2. Dawson J, Fitzpatrick R, Murray D, Carr A. Questionnaire on the
perceptions of patients about total knee replacement. J Bone Joint Surg
[Br] 1998;80-B:63-9
Authors’ reply:
Sir,
Our main concern with the Oxford knee status questionnaire was
its lack of specificity for the joint but Ms Unitt’s letter concerns
another difficulty and that is the polarity and range of scores.
Dawson et al1 originally recommended that the highest score (60),
should go to the knee with the lowest function. The best knee
therefore would have the lowest score (12), and scores in the range
of 0 to 11 would be meaningless.
We prefer to use the scoring system along a conventional scale
in which each of the 12 items is scored from 0 to 4 and the scores
are summated, thus giving 0 for the worst possible status and 48
for a normal knee. In our Methods section we very clearly stated
this and Ms Unitt is quite correct in emphasising that authors using
the Oxford knee score should clearly describe how they use this
instrument to avoid any confusion. It should be noted that this
method of using the Oxford knee score is now widely used in this
format.
S. H. WHITE, DM, FRCS
P. JONES
W. G. V. HARCOURT, FRCS, FRCS Orth
The Robert Jones and Agnes Hunt Orthopaedic Hospital
Oswestry, UK.
1. Dawson J, Fitzpatrick R, Murray D, Carr A. Questionnaire on the
perceptions of patients about total knee replacement. J Bone Joint Surg
[Br] 1998;80-B:63-9.