close

Вход

Забыли?

вход по аккаунту

?

00004623-197355040-00003

код для вставкиСкачать
Osteogenesis
Imperfecta:
Osteotomy
Treatment
and Intramedullary
REPORT ON THIRTEEN
BY FREDERICK
TILEY,
M.D.t,
by Multiple
Rod Insertion
PATIENTS*
AND JAMES
A. ALBRIGHT,
M.D.t,
NEW HAVEN, CONNECTICUT
From the Yale University School of Medicine, The Newington Children's Hospital,
Newington,
and the Yale-New Haven Hospital, New Haven
ABSTRACT: In thirteen
children
with osteogenesis
imperfecta,
129 operations
were done, starting at ages eleven months to ten years. All hut two of the patients
ultimately
were able to walk, nine of them with braces,
two without.
The operations
were insfrumental in preventing fractures to a degree, but none of the patients were
followed long enough to determine what level of activity was achieved by the pro
cedures.
In children with osteogenesis imperfecta the technique of multiple osteotomy,
realignment, and intramedullary rod fixation for deformities of the long bones has
become well established since Sofield, Page, and Mead originally introduced the con
cept in the early I 950's.
Since I 956 this procedure has become an integral part of the management of
children with osteogenesis imperfecta at the Newington Children's Hospital and the
Yale-New Haven Medical Center. Initially, the procedure was adopted for the pur
pose of decreasing or eliminating established deformities, thereby facilitating ambu
lation. This has continued to be the major objective of treatment, but surgery has also
been performed to interrupt the cycle of fractures and progressive deformity. Surgical
intervention was thought to be justified only when there was a reasonable chance of
ambulation, and in no case was facilitation of nursing care an indication.
Material
Between
June
1956 and August
1968,
129 Sofield procedures
were performed
in thirteen patients with osteogenesis imperfecta and, of these, sufficient information
was available to analyze 112 procedures. The bones operated on were the femur,
tibia, humerus, radius and ulna (Table I).
At the time of the initial procedure nine of the patients were over three years
old; the other four were under two. The youngest patient was eleven months old, and
the oldest ten years old. In all of the patients the initial procedure was performed on
one of the lower extremities,
and in nine patients
posite extremity within six to eight weeks.
surgery
was performed
on the op
The intelligence of our patients was normal, but motor development was de
layed. One patient began to stand at fourteen months, another at eighteen months,
two at twenty-eight months, two at thirty-seven months, and the remainder after fifty
months of age, except for one who has never been able to stand. Nine patients had
* Supported
by
the
National
Easter
Seal
Society
for
Crippled
Children
and
Adults,
R-655.
USPHSTI AM 5416-07.
t Yale University School of Medicine, New Haven, Connecticut 06510.
VOL. 55-A, NO. 4, JUNE 1973
701
702
FREDERICK TILEY AND J. A. ALBRIGHT
not started to stand before the initial procedure
was performed.
Most of the ambula
tory patients had stopped walking of their own accord prior to surgery because of the
degree of deformity and its consequences.
At the time of review nine patients were ambulatory
bulatory without braces, and two were non-ambulatory.
walked
without
braces
was the oldest
in the group
with braces, two were am
One of the patients who
and had the mildest
involvement.
The other patient who did not need braces had moderately severe disease and prob
ablywillrequirebracesin thefuture.
Of thepatients
who were non-ambulatoryat
final
reviewone had neverbeen abletostand,whiletheotherhad been abletostand
periodically, but had extremely severe disease complicated by recurrent respiratory
problems and almost certainly will be confined to a wheel chair or stretcher in the fu
ture. It remains
to be seen whether
all patients
with severe disease
will elect to remain
ambulatory once they are older; some may prefer to lead a wheel-chair
cause of the effort required to walk.
IAge
existence be
TABLE
ProceduresFemurTibiaHumerusPatientSexFollow-UpSurgeryFracturesR
at
(Yrs.)
InitialNo.
atAge
ofNo.
LR
RadiusUlnaTotalAD.M84yrs.100+1
14D.H.F103½
of
LR
L
11
yrs.624
313W.L.M84yrs.153
33
25G.M.M102yrs.113
36Y.NF54yrs.922D.P.F17lOyrs.131
34KR.F164yrs.334
22210R.R.M1211
2318L.R.F618
215J.S.M114yrs.163
mos.515
mos.152
44
5317D.T.M63
13P.Y.F139yrs.13213T.Z.M1014 yrs.122
33
mos.463
122Total33
45
2414
42
176
3
7
65
Methods
The study
included
review
of the medical
records,
and preoperative
and post
operative roentgenograms for each procedure. The indications for surgery, the details
of the procedure, the duration of correction, and the probable cause of eventual fail
ure (when
present)
were evaluated.
To grade the result for each procedure,
bone
were considered
evaluation
The
the proximal
primary
technical
and distal ends of each
factors
included length of the rod, adequacy of correction
tion, and placement
contributing
separately.
of the rod in the ends of the bone (Figs.
factors such as adequacy of postoperative
considered
of pre-existing
1-A through
cast treatment,
in the
angula
1-D). Other
activity level of
the patient, type and quality of braces, and so forth, were often of importance.
The time interval between operations
was used as an objective method for de
termining
the period of effectiveness
of each procedure
to eliminate subjective judg
ments as to when the rod was no longer adequate.
However, the rod often was made
quate long before the secondary procedure was done. The indication for the second
ary procedure generally was failure offixation at either the proximal or the distal end
of the bone.
In those cases
period of effectiveness
Since the operating
only the site of failure
was evaluated
to determine
the
of the procedure.
surgeons
included
many residents
THE JOURNAl.
and attending
physicians,
OF BONE AND JOINT SURGERY
OSTEOGENESIS
703
IMPERFECTA
0
FIG. 1-A
FIG. 1-B
Incomplete reduction with poor position of rod. The rod is peripherally placed and is anterior
which
does not provide
maximum
support.
It can be expected
to migrate
the anteroposterior plane the epiphyseal plate is still markedly tilted.
further
anteriorly.
In
0
FIG. 1-C
FIG. l-D
Complete reduction of end fragment and good position of rod.
there
was considerable
variation
in technique,
but the surgical
procedures
followed
the generally recognized principles. The minimum number of osteotomies necessary
for adequate positioning of the rod was used. In most cases standard incisions were
utilized and were placed to facilitate subsequent procedures, as well as to avoid un
due trauma to adjacent soft tissues. Because the osteotomies near the ends of the
bones were more difficult to stabilize than those at the mid-shaft, incisions were
placed as far from the epiphyseal plates as possible, and rods of maximum length
were used.
Frequently
the rod was driven
across
the epiphyseal
plate
into the sub
chondral bone. In most cases the diameter of the rod was the largest the medullary
canal would accept, but occasionally, the canal was drilled for a larger rod. Drilling
was seldom performed
in patients with severe disease since reaming them was apt to
cause fracture. When severe bowing was present the bone was purposely shortened.
In these patients good correction
was not possible without shortening
since the soft
tissues on the concave side became tight when correction
was attempted and excessive
stresson the bone ends increasedthe danger of displacementof the rod.When a seg
VOL. 55-A, NO.4,
JUNE 1973
704
FREDERICK TILEY AND J. A. ALBRIGHT
ment of bone was removed, the adjacent bone ends were frequently incompatible in
size, with a resulting difficulty in positioning the rod properly. Roentgenograms were
normally obtained since it was often impossible grossly to evaluate accurately the
angulation
of the segments.
Postoperative
casts were considered
essential
for most
patients
since
stability
was often unsatisfactory until the bone had healed, and even when stability appeared
to be adequate a cast was used to control rotation. The casts incorporated corrective
forces to maintain the desired position.
The operative technique for the femur involved a standard posterolateral in
cision
in the majority
ofcases,
and the lateral
intramuscular
septum
was followed
the femur. However, when severe bowing was present, a direct approach
quadriceps
muscle
was occasionally
used, although
this complicated
to
through the
subsequent
pro
cedures, especially when the incision had to be lengthened for better exposure. After
a sufficient number of osteotomies had been made in the diaphysis, the rod was in
troduced through the greater trochanter and then advanced through the diaphyseal
fragments
into the distal fragment.
When a Steinmann
pin was used as a rod, it was
passed retrograde
through the proximal
fragment and then distally through the re
maining fragments.
When a Rush pin was used, a Steinmann
pin was first passed ret
rograde through the proximal fragment, to be used as a guide for introduction of the
Rush pin. This technique was easier if the blunt end of the Steinmann pin was con
cave to receive
the point
of the Rush
pin. The required
length
of the pin was esti
mated prior to insertion. We attempted to place the tip adjacent to the epiphyseal
plate or through the plate into subchondral bone. When coxa vara was present an at
tempt was made to position the rod in the medial portion of the greater trochanter or
in the base of the neck.
After the end of the rod reached the distal osteotomy site under direct vision the
distal fragment was reduced and the rod was advanced as far as necessary to achieve
stability, at which point a roentgenogram
was made. When the position was adequate
the rod was advanced to its final position, but when the position was inadequate, the
rod was withdrawn, redirected and another roentgenogram made. Then the new rod
tended to follow the old channel in the distal fragment, in which case a new channel
had to be drilled. Estimation of the proper direction to drill was one of the most dif
ficult parts of the procedure. The configuration of the leg was not a good reference
line because the proximal part of the tibia usually was angulated in a compensatory
direction. Palpation of the medial and lateral borders of the knee joint provided the
most dependable landmarks. We tried to position the distal end ofthe rod either cen
trally, or slightly posteromedially.
Decisions concerning adequate correction of de
formity and placement of the rod in the distal fragment were influenced by such fac
tors as deformity of the knee joint, angulation of the proximal part of the tibia, and
future plans for surgery to the tibia. For instance, the correction of the accentuated
anterior bowing of both the femur and tibia could result in genu recurvatum because
it is difficult to position a straight
of the femur without eliminating
rod centrally or slightly posteriorly
in the distal end
the normal degree of anterior bowing. When it was
desirable to maintain some degree of anterior bowing, the rod could still be posi
tioned centrally by making the osteotomy closer to the epiphyseal plate or by bend
ing the rod slightly.
The operative technique for the tibia involved an anterior incision extending
almost the entire length of the tibia. To insert the tibial rod, a longitudinal cortical
slot was made in the end of the bone showing less deformity. After the osteotomies
were done, the rod was introduced through the osteotomy cut closest to the slot, and
passed through all the intervening segments, and positioned in the end of the bone
farthest from the slot. Then the other end of the rod was placed in the slot for final
THE JOURNAL
OF BONE AND JOINT SURGERY
OSTEOGENESIS
positioning
inside the metaphysis.
IMPERFECTA
705
The rod could not be placed
across the epiphyseal
plate in the slotted segment because the removal of a portion of the plate might well
interfere with its growth. The slot was usually placed posteromedially
in both the
proximal
and distal
containing
parts of the tibia to avoid
the fibula.
Since the end of the bone
the slot was relatively (sometimes grossly) unstable, the proper corrective
forces were determined
prior to closure
of the wound
and extreme
care was taken
to
maintain proper position until the cast was applied.
The operative procedure for the humerus usually involved introduction of the
rod through the greater tuberosity. Angulation in the middle and distal parts of the
humerus could be corrected then, but good correction and stabilization of lateral
bowing in the proximal end would have required insertion of a rod through the ar
ticular surface, which never was done. In two patients with severe deformity prox
imally, the rod was introduced through the distal part of the humerus proximal to
the epiphyseal plate.
The operative procedure in the forearm involved introduction
ofthe rod through
the olecranon
in the ulna, using either a Rush pin or a Steinmann
pin with a hook
bent at the proximal end, while in the radius the rod was introduced
dorsally at the
distal end.
Results
In general the rod was remarkably successful in maintaining good diaphyseal
alignment. Preoperatively, the major deformity was usually in the diaphysis, but if
an angulation
developed
postoperatively
taphysis, usually at the end of the rod.
In general, growth was associated
it almost
universally
with a progressive
occurred
at the me
increase in angulation
at
the end of the rod, and in most patients the type of angulation
was predictable
and
was uniform
for each bone. The angulation
usually was associated
with rotation
which, although difficult to interpret quantitatively,
was clinically insignificant
unless
the axis of motion of the joint was grossly disturbed.
In the femur an external rota
tional deformity
was most common, causing the knee to assume a position of mod
erate external rotation.
The ankle often assumed a position of external rotation at
rest, but tended to be internally
were least predictable
rotated
in the most severely
in reference
involved
to the knee. The deformities
patients,
especially
in those who
were unabletowalk.
Of the nine patients who had never walked prior to the initial procedure,
began walking
soon after surgery.
rect relationship
walk. However,
surgical
correction
As regards
the other
six patients,
three
there was no di
between the procedure and the time when the patient started to
once a patient had started, there was a close correlation between
ofdeformities
and the ability
to walk.
Except in one patient who had non-union of an osteotomy site in each tibia, no
difficulty was encountered in bone healing. Postoperatively the majority of osteot
omies were sufficiently
stable
by four to eight weeks to discontinue
cast immobiliza
tion.Even so,thebone did not alwaysremodel normallyso thatyearslatermany
osteotomy
sites were clearly
visible
on roentgenograms.
In one patient
a fracture
oc
curred through an old osteotomy site three months after the rod had been removed.
When recurrent fractures were the principal indication for the procedure, the
operation
almost
uniformly
interrupted
the cycle of fracture
and deformity.
There
were surprisingly
few postoperative
fractures
ofany kindwhich couldbe ascribedto
the surgical procedure itself or to such sequelae as osteoporosis or muscle weakness.
In the bones operated on, no fractures occurred in the first four months postopera
tively, and there were three fractures between four and six months, five at about six
to nine months, and a total of fifteen within the first year (the thirteen patients studied
VOL. 55-A, NO.4, JUNE 1973
706
FREDERICK TILEY AND J. A. ALBRIGHT
FIG. 2-A
FIG. 2-B
R. R. at twelve years old in June, 1968. The rod had been inserted two and one-half years
previously. Preoperative
femur
with protrusion
roentgenograms
show anterolateral
angulation of the distal part of the
of the rod.
had had a combined total of over 350 fractures preoperatively).
The effect that the
presence of the rod itself might have had on the cortical bone could not be deter
mined objectively. At times it appeared as if there was more osteoporosis and that
the shaft was smaller
In the proximal
in diameter
because
of the rod.
part of the femur, coxa vara was the most common deformity
and of all the deformities
which
occurred
in the lower extremities,
difficult to correct. Ten of the twenty femora operated
with a neck-shaft
angle close to, or frequently
it was the most
on had significant coxa vara,
less than,
90 degrees.
The deformity
was often greater following surgery. Correction by subtrochanteric osteotomy was at
tempted only once unsuccessfully; in this patient intramedullary rod fixation had to
be abandoned temporarily. No patient was found in whom an intramedullary rod had
been placed
in a position
which
would
have resulted
in more valgus.
Consequently,
significant improvement in the coxa vara was never attained from the Sofield pro
cedure.
The most common deformity in the distal part of the femur was anterolateral
angulation, with protrusion of the rod anterolaterally
as a complication (Figs. 2-A
through 2-D). This was one of the major indications for secondary surgery—70 per
cent of all the secondary femoral procedures, with an additional 13 per cent per
formed following fracture in patients in whom anterolateral
angulation had been
present prior to fracture.
THE JOURNAL
OF BONE AND JOINT SURGERY
OSTEOGENESIS
IMPERFECTA
FIG. 2-C
707
FIG. 2-D
October, 1968. Postoperative roentgenograms show good position of the rod. Angulation of
the distal part of the femur has been corrected, the rod is centrally placed and penetrates the
epiphyseal plate. However, the distal osteotomy site should be farther from the end of the bone
for maximum stability.
The initial position of the rod in the distal fragment was studied in relation to
the time
interval
between
procedures.
Eighteen
rods had initially
been
placed
an
teriorly or laterally and all subsequently protruded anterolaterally.
Of the thirteen
rods that had initially been placed in a central position, eight remained central, four
migrated laterally, and one migrated posteriorly; for these thirteen procedures an av
erage of 44.8 months elapsed until a secondary
procedure
was done,
with 22.6 months for those patients in whom the rods had been placed
as compared
anteriorly
or
laterally. One of the centrally placed rods remained in satisfactory position for eight
years but this case was not included in the previous calculations because the second
ary operation
was performed
for reasons
irrelevant
to the subject
under
study.
Three
rods had initially been placed in the posterolateral quadrant and all protruded poster
olaterally after an average of 26.2 months. Only two rods had been placed medially
to start with and one of these
had not required
adjustment
after forty-three
months
while the other did not require reoperation at the time of review, eleven months post
operatively. One rod had initially been placed in the posteromedial quadrant and had
protruded
posteriorly
after thirty-nine
months.
No rod migrated
medially,
and in all
but one instance, there was a tendency for lateral drifting of the rod, generally with
some anterior drift as well.
Anterolateral
angulation
compensatory
to anteromedial
VOL. 55-A, NO. 4, JUNE 1973
of the distal part of the femur often appeared
to be
angulation
of the proximal part of the tibia. This usu
708
FREDERICK
TILEY AND J. A. ALBRIGHT
FIG. 3
FIG. 4
Fig. 3: J. S. at five years old, August 1962. Lateral angulation of the distal part of the femur
is compensatory
with medial
angulation
of the proximal
part of the tibia, especially
on the right.
Alignment of the tibia with the femur is fair, but the plane of the joint axis is markedly ab
normal.
Fig. 4: J. S. at eight years old, May 1965. Typical
protrusion of the rod anteriorly.
deformity
of the distal
part of the tibia with
ally resulted in a fair alignment of mid-diaphysis of the tibia with that of the femur,
even when the plane of the axis of the knee joint was markedly abnormal (Fig. 3).
The anterior angulation of the femur and tibia produced an apparent flexion contrac
ture of the knee, compensated for, in part, by a hyperextension deformity in the joint.
Invariably, both the angulation of the proximal part of the tibia and the pro
trusion of the rod placed in the tibia were anteromedial. Most of the rods had initial
ly been placed medially (sometimes anteromedially or posteromedially).
In several
instances the rod had been placed posteriorly or posterolaterally,
but angulation and
protrusion still occurred anteromedially.
In the distal part of the tibia all of the rods
migrated anteriorly or anterolaterally, and the most common area for the rod to come
through the bone was directly anterior (Fig. 4). The rod had initially been placed an
teriorly or anterolaterally
in eleven cases, anteromedially
in two cases, centrally in
five cases, and posteromedially
in one case. Since correction was more difficult to
maintain in the distal than in the proximal part of the tibia, fixation was best when a
window was made at the proximal end and the rod was placed as far posteriorly as
possible. When a window was used in the distal part of the tibia, it was most success
ful if placed posteromedially.
The proximal part of the humerus tended to angulate laterally or anterolateral
ly. Five rods had initially been placed anteriorly or laterally and all migrated lat
erally. Two rods had been placed centrally and did not migrate, but fractures occurred
after eight months in one case and forty-five months in the other. Lateral angulation
occurred in the distal part of the humerus but tended to be posterolateral rather than
anterolateral, as in the proximal part of the humerus. Five of the six rods which had
been placed laterally, migrated further laterally. Two of the rods had been placed
centrally and neither migrated; in one patient fracture occurred at the distal end of
THE JOURNAL
OF BONE AND JOINT SURGERY
OSTEOGENESIS
IMPERFECTA
709
the rod six years after insertion and in the other patient
the time of review five and one-half years after insertion.
the rod remained
central
at
Forearm rotation was severely limited in all of the patients reviewed, and the
hand usually was in a position of mild pronation. The angulation present in the ulna
was determined
using the proximal
part of the ulna (and elbow)
as a frame
of refer
ence and the distal part of the radius (and wrist) was used similarly as a frame of ref
erence for angulation of the radius. Angulation was usually greater in the proximal
part of the forearm than in the distal and when present distally tended to be compen
satory. In most patients angulation
of the humerus was also present and appeared
to
be related. The most common deformity
in the upper extremity included lateral an
gulation of the humerus,
varus of the elbow with an apparent
flexion contracture,
posterolateral
bowing ofthe proximal part ofthe forearm (in reference to the elbow),
some compensatory
bowing of the distal part of the forearm, and limited rotational
movement
of the forearm on supination-pronation
with the forearm held in slight
pronation.
The proximal part of the ulna angulated
posteriorly
(usually posterolat
erally) in all patients reviewed, while the radius angulated
medially (usually antero
medially); bowing of the distal parts of the radius and ulna occurred in a direction
opposite to the bowing in the proximal portion. In most patients good bone align
ment was present
nificantly
immediately
improved.
time of review
ment.
Recurrent
following
Progressive
the deformities
deformity
surgery,
angulation
were generally
but forearm
developed
similar
rotation
postoperatively
to those present
was not sig
and at the
prior to treat
of the forearm was partly related to growth, and, as in the
lower extremities,
the angulation
tended to increase more rapidly as the rods became
relatively shorter. However, the severity of disease appeared
to be the factor of pri
mary importance.
In contrast to the deformities
in the lower extremities,
severe de
formities in the forearm only occurred in patients with severe disease; the four pa
tients who had operations
on the forearm
had the most severe form of the entire
group. At the time of review three of the seven rods had been removed while the
other four were relatively short and some protruded
through the cortex. These rods
undoubtedly
will require removal in the future.
Postoperative
roentgenograms
showed that the rod crossed the epiphyseal
plate
in thirty-nine
of the I I 2 cases reviewed without evidence of growth disturbance.
However, a growth disturbance
did occur in two other cases—not closure of the plate,
but a decreased
In both cases there were special circumstances.
a fracture
rate ofgrowth.
of the metaphysis
occurred
other a severe postoperative
infection
additional
operative procedures.
adjacent
to the epiphyseal
(the only one in the series)
In one,
plate, and in the
required
several
Complications
Longitudinal
migration
of the rod occurred
after twenty
of the I 12 procedures.
It was a major complication and often required an additional minor procedure
either reinsertion or extraction of the rod. Migration of the rod always jeopardized
the operative result. When the rod was relatively short, whether due to growth or mi
gration, increasing
angulation
generally supervened
(at the time of reoperation
the
rod was short in 90 per cent of the cases). Of the fifty-seven femoral procedures evalu
ated, nine showed proximal migration of the rod from the trochanter
and one showed
distal migration of the rod into the knee joint. Thirty-one
of the femoral rods did not
have a proximal hook or any device to prevent migration; eight (26 per cent) of these
showed migration. Only two (8 per cent) of the twenty-six with a proximal hook
showed migration.
In the humerus,
proximal
migration
occurred
in two of the six
rods without a hook and one of the five rods with a hook. Two of the rods with hooks
VOL. 55-A, NO.4. JUNE 1973
7 10
FREDERICK TILEY AND J. A. ALBRJGHT
had been introduced
through the distal end of the humerus and neither showed mi
gration. Two ulnar rods with hooks migrated proximally.
No hook had been used on
any of the thirty-one tibial rods and only one had migrated distally into the foot.
The other complications were as follows. One girl, fourteen and one-half years
old, had non-union at an osteotomy site in each tibia which required bone-grafting.
The two patients with delayed growth of one distal femoral epiphysis, already men
tioned, had considerable shortening of the extremity. In one patient fragmentation
of the proximal part of the tibia occurred during a secondary procedure. In this pa
tient epiphyseal
growth
was not disturbed,
but considerable
shortening
of the tibia
resulted. One patient had a cardiac arrest which responded to treatment and left no
residuum. This patient was the most severely diseased in the group; no further sur
gery was performed because he had progressive respiratory problems, and also be
cause his function did not improve as a result of the surgical procedures
he had had
previously.
One postoperative
infection
resulted in one of the two growth distur
bances found.
Discussion
Numerous
problems
face the orthopaedic
surgeon
who applies
the Sofield
pro
cedure to children with osteogenesis imperfecta. Success depends on attention to
many details besides the operative procedure. Inattention to any one detail may be
critical and may cause the procedure to fail or to have a short-lived success. Unneces
sary repetition of the operation may be the result. Although many of the principles
of treatment appear to be self-evident and should not require comment, our analysis
of this series of cases has brought out important, less obvious pitfalls.
The severity of disease should be such that the operation can be reasonably ex
pected to obtain its objective. Very severe disease, where the patient has never been
able to stand, is a relative contraindication.
In our series all four patients who were
able to stand prior to the initial procedure became ambulatory; in contrast, two of
the nine patients who had not been able to stand preoperatively did not stand after
the operation. Four of the patients who became ambulatory were markedly disabled
and may elect a wheel-chair existence when older.
The ideal age to initiate a program of surgical correction is the age when the pa
tient attempts to stand, which was over four years in six of our patients. If the opera
tive treatment
is initiated prior to this time, the patient is unlikely to begin standing
as a result of the surgery. In fact, recurrent
deformities
may develop requiring
re
placement
of the rod before the patient begins to stand, as was necessary
in four of
our patients.
Since
a rod in the medullary
canal
provides
better
support
for the diaphysis
than for the metaphysis, particular care should be given to positioning the rod in the
ends of the bone. To provide maximum support the rod must be of adequate length
and must be placed centrally in the metaphysis
(preferably
slightly off-center,
but op
posite to the direction of eventual migration).
In our patients the results in the distal
part of the femur illustrate this. All eighteen rods placed anteriorly
or laterally sub
sequently protruded
anterolaterally
with reoperat ion after an average interval of 22.6
months, whereas eight of the thirteen centrally placed rods remained central, only
four migrated laterally and one posteriorly, and these thirteen patients had an aver
age interval of 48.7 months before secondary surgery was necessary.
Roentgenographic
control during surgery is essential since clinical evaluation of
correction by gross observation at the time of surgery is often erroneous. The posi
tion of the rod after its preliminary introduction was often different from that vis
ualized and redirection of the rod prior to final seating had to be done.
Postoperative casts should be used to ensure maintenance of position. A judi
THE JOURNAL OF BONE AND JOINT SURGERY
OSTEOGENESIS
ciously
applied
instability,
position
acceptable
alignment,
but it must be well molded and incorporate
will be lost early
adequate,
formity
cast may salvage
IMPERFECTA
in the postoperative
a cast is usually necessary
during
the initial
period.
7 11
even in the presence
corrective
Even when
to prevent the development
of gross
forces; otherwise,
stability
appears
of rotational
de
three to four week period.
If surgery is performed
prior to skeletal maturity,
the osteotomies
can be ex
pected to heal without difficulty.
Our young patients showed no non-union
even
though no bone grafts were used. However, bilateral non-union
of an osteotomy
site
in the tibia developed
in the only patient operated on after skeletal maturity. We sug
gest the advisability
of bone-grafting
in older patients.
It is not advisable
time,
because
to attempt correction
the stability
of the two groups
of the femur and the tibia at the same
of osteotomies
will be difficult
to main
tam. If both femora and both tibiae require surgery, a safe and yet reasonably ex
pedient schedule includes correction
taneously,
spacing
the procedures
ofeach
femur individually
four to six weeks apart
and the tibiae simul
to allow time for adequate
healing.
In patients with severe disease, proper external support of the lower extremities
can be of equal or greater importance than the surgery itself and may be preferable
to surgery.
In fact when the rod becomes
to be the single most important
procedures.
As with casts,
relatively
determinant
most braces
should
short,
in extending
employ
adequate
bracing
appears
the time period between
corrective
forces
at the knee
and ankle; the most satisfactory type for a severely diseased patient is a long brace,
with a pelvic band, double uprights, a thigh cuff, and a leg cuff sufficient to support
the entire tibia. Commonly overlooked is the need for adequate support of the distal
part of the tibia. When this is provided, the need for insertion of another rod may be
deferred.
However,
sary for all patients,
the use of braces must be individualized.
Braces are not neces
but they are particularly
useful for patients with recurrent frac
tures, or rapidly progressive deformities, or for patients who are unable to stand
otherwise.
Even though adequate reduction of the epiphysis and epiphyseal plate was
found to be one of the most important determinants of successful treatment, attain
ment of correction at the time of surgery was often very difficult, not only because of
the problem
in determining
proper
reduction,
but also because
of the soft-tissue
shortening secondary to long-standing deformity. In such cases, stability was usually
poor and the result often was compromised when multiple passes of the rod had to be
made, or when fragmentation
of bone occurred,
or when an osteotomy
was made too
close to the end of the bone.
Except in patients with extremely severe disease the presence of an intramedul
lary rod appears to provide significant mechanical support to the metaphysis, as well
as to the diaphysis;
in only 10 per cent of our cases was the rod adequate in length at
the time of a secondary procedure,
whereas it was short in 90 per cent of cases. A rod
of inadequate
length not only fails to support the bone in the area most susceptible
to
deformity
(the metaphysis),
but it probably acts as a stress raiser, thereby actually in
creasing the chance of deformity occurring (this also appeared to be true for rods in
troduced through the cortex near the end of the bone).
Early relative shortness of the rod usually was due to migration.
The frequency
with which this happened could have been decreased by better technique,
such as the
use of a hook on all femoral or humeral rods. Ultimately
the rod was relatively too
short because of growth. This represents
one distressing
shortcoming
of the Sofield
technique
in young children.
An extensible
rod designed to elongate with growth
might meet the problem 1, but additional
information
on long-term results with such
rods is needed, especially with regard to the harm inflicted on the epiphyseal
plate 2
VOL. 55-A, NO. 4. JUNE 1973
7 12
FREDERICK TILEY AND J. A. ALBRIGHT
In the present series one attempt to use a spring-loaded extensible rod was unsuccess
ful.
Maximum support is provided by a rod which penetrates the epiphyseal plate
with the end positioned in the subchondral bone. Placement of a single rod across an
epiphyseal
plate appears to be relatively safe, but each repetition
increases the possi
bility of fragmentation,
and if multiple passes of the rod are necessary to obtain sat
isfactory
position,
the chance of interfering
with growth increases.
This was the
probable cause of growth retardation
Since the type of deformity
perfecta is highly predictable,
in one of our patients.
which
develops
in a given bone in osteogenesis
the details of treatment,
im
whether surgical or non-sur
gical, should be planned accordingly.
For example, the best position for a window in
the distal part of the tibia is usually posterior (posteromedially),
since the rod can be
@
expected to migrate anteriorly (usually anterolaterally).
Introduction of the rod
through the knee or through the foot obviates the need for a cortical window
and
improves the holding strength of the rod in the immediate
item of technique was not used in the present series.
postoperative
period.
This
Attention to the details of treatment can help to prevent an increase in coxa
vara following insertion of a femoral rod, even though the rod is not designed to
treat coxa vara. If an osteotomy is made near the proximal end of the femur, the
angle between the rod and the proximal
fragment can be varied
range, so the exact angle of insertion assumes increased importance
is present.
within a certain
when coxa vara
The least satisfactory results were obtained in the forearm, possibly because of
the inherent nature of the disease, since severe deformity occurred only in the most
severely
diseased
patients.
It is unlikely
that better
surgical
technique
would
have
improved our results. At the time of surgery the patients ranged in age from three to
eleven years and perhaps corrective
surgery performed
maturation
would have been more satisfactory.
closer
to the age of skeletal
A high incidence of fracture might be expected following surgery, but fractures
never occurred in the early postoperative period. Possibly, the use ofcasts and braces
prevented
a higher incidence,
but a more likely explanation
was the very guarded
pacing of activity by the patients themselves,
often influencing the most carefully de
signed rehabilitation
plans. Most patients appeared
to sense innately their physical
capabilities
and rarely could be enticed to attempt activities beyond that point.
Over-all, while it would appear that even the most rigid adherence
to technical
detail would result in only temporary
improvement
because no form of surgery is
able to influence the disease process, the value of the procedure
must be kept in clear
perspective,
particularly
when a patient is severely involved. An elaborate
plan of
operative
treatment
may fail to produce the anticipated
long-term result, and is not
without hazard in itself, but at the same time, surgical treatment
limited
benefit,
especially
to a patient
with mild or moderately
can be of definite if
severe disease.
Summary
One hundred and twelve intramedullary
rod insertions
have been evaluated
in
thirteen children with osteogenesis
imperfecta.
A highly predictable
pattern of long
bone deformity
occurred
in all cases: coxa vara of the proximal
part of the femur,
anterolateral
angulation
of the distal part of the femur, anteromedial
angulation
of
the proximal part of the tibia, anterior or anterolateral
angulation
of the distal part
of the tibia, lateral or anterolateral
angulation
of the proximal part of the humerus,
posterolateral
angulation of the proximal part of the ulna, and medial or antero
medial angulation of the proximal part of the radius. The most important principles
of treatment
included
adequate
reduction
of the ends of the long bones,
proper
place
THE JOURNAL OF BONE AND JOINT SURGERY
OSTEOGENESIS IMPERFECTA
713
ment of the rod in the metaphysis and epiphysis, use of a hook on the femoral or
humeral rods to prevent migration, use of a rod of adequate length, and incorpora
tion of corrective forces in postoperative casts and braces.
References
1. BAILEY,R. W., and DUBOW, H. I.:Experimental
and Clinical
Studiesof Longitudinal
Bone
Growth Utilizing a New Method of Internal Fixation Crossing the Epiphyseal Plate. In Pro
ceedings of the American Orthopaedic Association. J. Bone and Joint Surg., 47-A: 1669,
Dec. 1965.
2. RODRIGUEZ, R. P., JR., and WICKSTROM, J. K.: Osteogenesis
Imperfecta:
A Preliminary
Re
port on Resurfacing
Long Bones with Intramedullary
Fixation by an Extensible
Intra
medullary Device. Southern Med. J., 64: 169-176, 1971.
3.SOFIELD,
H. A.,and MILLAR,E.A.:Fragmentation,
Realignment
and Intramedullary
Rod
Fixationof Deformitiesof the Long Bones in Children.A Ten-Year Appraisal.J. Bone and
JointSurg.,41-A: 1371-1391,Dec. 1959.
4. SOFIELD, H. A.; PAGE, M. A.; and MEAD, N. C.: MultipleOsteotomies and Metal Rod Fixa
tion for Osteogenesis
Imperfecta. Paper read at The American Academy of Orthopaedic Sur
geons. January 1952.
5. WILLIAMS, P. F.: Fragmentation
and Rodding in Osteogenesis
Imperfecta. J. Bone and Joint
Surg., 47-B: 23-3 1, Feb. 1965.
VOL. 55-A,NO.4, JUNE 1973
Документ
Категория
Без категории
Просмотров
1
Размер файла
1 589 Кб
Теги
197355040, 00003, 00004623
1/--страниц
Пожаловаться на содержимое документа