what-when-how
In Depth Tutorials and Information
individual decision for each patient and each osteotomy
that is made intraoperatively.
and humerus were the most likely sites to be involved
with a non-union, especially at sites of progressive defor-
mity and multiple fractures. In a similar study of a differ-
ent OI population, 44 patients followed for 14 years had
nine non-unions seen in eight patients.
14
Thus non-union
is a common problem in the OI population compared
to those without metabolic bone disease, and should be
considered as a possible occurrence following osteotomy
for deformity correction.
In a recent study of seven patients treated with
bisphosphonates, the authors found no difference
between the rate of pseudarthrosis in the cohort com-
pared with the estimated 14% non-union rate expected
from historical controls.
12
Although this study is reassur-
ing, the authors note increased numbers of patients are
needed to determine if bisphosphonate therapy plays a
role in non-union.
Non-Union in OI
Munns et al.
11
reported a difference in bone healing
at the site of osteotomies, compared with bone healing
at the site of a fracture in children with OI being treated
with bisphosphonates. The etiology of delayed union or
non-union remains unclear - medical management lead-
ing to reduction of osteoclast function can be implicated,
as can surgical technique. The effects of bisphosphonates
on osteoclast function are not completely understood.
One hypothesis is that bisphosphonates do not interfere
with initial callus formation and bone bridging, but do
inhibit remodeling of the callus and the final stages of
bone healing, resulting in persistence of the fracture or
radiolucent osteotomy line. While some surgeons favor
drill corticotomy to avoid use of a saw, others find use
of a saw to cut the bone to be effective, particularly when
a wedge of bone needs to be removed to facilitate defor-
mity correction. In addition, the saw allows for a more
defined cut, as splintering of the bone can occur with
use of an osteotome, leading to an oblique cut or a seg-
mental loss of a fragment of bone. However, the heat of
the saw can result in localized heat necrosis of the bone,
which may contribute to problems with healing at the
osteotomy site. No study has yet compared drill corti-
cotomy versus saw blade osteotomy in relation to bone
healing. Often the bone can be drilled to mark the level
of the osteotomy, and wedges of the soft bone can be
removed from the convexity by use of a small rongeur. In
cases with severe bowing, shortening of the bone is often
necessary to reduce tension on the soft tissues, and allow
complete correction of deformity. Bone that is removed
can be morselized and replaced in the osteotomy bed to
ill in any gaps. In patients receiving bisphosphonates,
the bone may be harder and more brittle, making use
of a saw cooled with irrigation the preferred method of
osteotomy. As noted above, restoration of a normal ana-
tomical axis with multiple osteotomies is preferred, as
residual bowing may increase the risk of non-union and
failure of rod telescoping.
Persistent non-union or remodeling of bone can lead
to fracture at the osteotomy site in patients who have
undergone rodding of the long bones. In a study by Cho
et al.
1
of 72 telescopic rods in children with OI, 33 frac-
tures occurred post-rodding. Of these 33 fractures, 29
were in long bones containing a radiolucent line at the
fracture site.
However, non-union may be a biological problem
unrelated to surgical intervention. Gamble reported
both atrophic and hypertrophic non-unions occurring in
patients with OI at a frequency of 12 non-unions in ten
patients from a population of 52 patients.
13
The femur
Specialized Rods for the Treatment of Long Bone
Deformity in OI
The availability of specialized growing rods for OI
began when the Bailey-Dubow telescopic rod replaced
the fixed length rods such as Enders or Rush rods. With
the advent of telescopic rods the rate of revision was sig-
nificantly lowered, with the interval to revision surgery
increasing from 2-2.5 years for fixed length rods and to
4-5 years for the telescopic rods
5,9,15
The Bailey-Dubow
rod allowed overlap of the rod at the center of the con-
struct, while the ends of the rods were inserted from the
distal and proximal joints. The main complication with
the Bailey-Dubow rods was related to the T-piece detach-
ing, so the Sheffield rodding system modified this by
making the rod and T-piece a solid piece. The Sheffield
rodding system has been shown to work effectively, and
long-term results are available for some patients into
skeletal maturity.
3
Minimally invasive techniques have
been described for exchanging telescopic rods that are
disengaged or close to becoming disengaged.
16
More
recently developed rods such as the Fassier-Duval rod
allow passage of both rods from a single entry point, thus
eliminating the need to enter the distal joint, the knee
joint when rodding the femur or the ankle when rodding
the tibia.
17
Telescopic rods will only function as such if
the ends of the rods remain engaged in the bone of the
epiphyses, on either side of the growth plate (
Figure
47.5
). When the rods disengage from the epiphyses or
if the rod bends, the telescopic properties of the rod are
lost. Currently, the surgeon has the choice whether to
interlock the telescopic rod distally or not. This interlock
is another form of epiphyseal anchor but does not confer
rotational stability on the cylindrical rods. Reports of suc-
cess with this technique are promising, and in one series
the authors reported that even when the locking device
cut through the physis there was no physeal closure and
