To The Editor:
I read with great interest the article entitled "Influence of
Acetabular Coverage on Hip Survival After Free Vascularized Fibular Grafting
for Femoral Head Osteonecrosis," (2006;88:2152-8), by Roush et al., and
the commentary by Steinberg and Steinberg1. Indeed, this article
and the commentary by Marvin Steinberg1 represent the views and
thoughts of two senior authors with opposing opinions on how to treat
osteonecrosis. Urbaniak2 supports vascularized bone-grafting,
whereas Steinberg3 supports avascular cancellous bone-grafting.
While the intent of the article by Roush et al. was not to document the
clinical efficacy of either type of bone-grafting, the conclusions made
therein invariably influence one's ability to recognize a potential limitation
of free vascularized fibular grafting, and it is from this observation that I
would like to offer a few comments.
Roush et al. retrospectively reviewed a consecutive series of 200 hips in
160 patients with osteonecrosis of the femoral head who had undergone free
vascularized fibular grafting. They found that, of the hips with a center-edge
angle of =30°, 55% demonstrated progression of collapse and 45% were
converted to a total hip arthroplasty. In contrast, of the hips with a
center-edge angle of >30°, only 10% demonstrated progression of
collapse and only 6% were converted to a total hip arthroplasty. Roush et al.
encouraged the reader to consider acetabular dysplasia as an independent risk
factor with a negative influence on prognosis and cited a study of children
with congenital dislocation of the hip, by Gregosiewicz and Wosko4,
to support their position. However, one must carefully consider this
conclusion. Gregosiewicz and Wosko4 reported that the highest risk
for osteonecrosis in children is associated with a number of factors: (1) an
age less of than six months, (2) severe acetabular dysplasia, (3) the use of
an abduction apparatus such as the Frejka pillow by outpatients, and (4) a
"frog-leg" position afte reduction. The observation by
Gregosiewicz and Wosko4 implies increased contact forces on a soft,
predominantly cartilaginous femoral head after reduction. An age of less than
six months correlates well with the congenital nature of acetabular dysplasia.
In contrast, the mean age at the time of surgery in the article by Roush et
al. was 33.6 years, with the worst degree of collapse being only 3 mm (in
eleven patients). If the intent of Roush et al. was to imply a causative role
for acetabular dysplasia, then one would think that the adult hips in their
series would have been more arthritic, particularly after thirty-three years.
Clearly, one must contemplate how dysplastic hips with a center-edge angle of
=30° functioned for an average of thirty-three years and then had
development of a primary bone disease (i.e., osteonecrosis), with the
etiologic associations being known in 75% of the 200 hips, only to collapse
after free vascularized fibular grafting. Roush et al. seemed to suggest that
the failed femoral heads would have survived had it not been for the
acetabular dysplasia. Yet the acetabular dysplasia was present prior to free
vascularized fibular grafting. Could these dysplastic hips have benefited from
a different joint preservation procedure? Although Steinberg and
Steinberg1 suggest that perhaps the femoral heads with the lower
center-edge angles were deformed, implying a propensity to collapse, one must
recognize how free vascularized fibular grafting may potentiate the demise of
a femoral head with a dysplastic acetabulum. Thus, is it the dysplastic
acetabulum that portends a poor outcome, as suggested by Roush et al., or is
it the treatment chosen, i.e., free vascularized fibular grafting?
The surgical technique of free vascularized fibular grafting as described
by Urbaniak et al.2 comprises thorough dé-bridement of the
femoral head. The core tract, ranging in diameter from 16 to 19 mm, is
designed to avoid occlusion of the peroneal vessels and to prevent tension on
the anastomosis. This large core tract likely destabilizes the femoral head
and neck and potentiates collapse where contact forces are greatest, i.e., a
dysplastic acetabulum. Although Urbaniak et al.2 described passing
a guidewire into the necrotic lesion within the femoral head, it is far more
important that the starting point of the guidewire along the lateral cortex be
situated to prevent tension on the anastomosis once the large core tract is
created. This requirement likely determines the position of the fibula and may
prevent optimal placement in view of the acetabular dysplasia. Roush et al.
fell short of identifying this potential limitation of free vascularized
fibular grafting and concluded by asking the reader to preoperatively quantify
the extent of dysplasia for prognostic and possibly surgical planning
purposes. One wonders what other surgical plans exist when Roush et al.
comment that "the surgical procedure has remained essentially unchanged
since the publication of our original reports."
Mont et al.5 and Rosenwasser et al.6 demonstrated
that avascular bone-grafting combined with thorough débridement can be
successfully applied to select patients with osteonecrosis of the femoral head
and that good outcomes can be achieved. Continued emphasis on the role of the
vascularized fibula in the treatment of osteonecrosis might
invariably prevent one from recognizing the features that vascular techniques
(i.e., free vascularized fibular grafting) and avascular techniques (i.e., the
trapdoor procedure as described by Mont et al.5 and the lightbulb
procedure as described by Rosenwasser et al.6) have in common,
namely, thorough débridement. Importantly, the thorough
débridement of the trapdoor/lightbulb procedure leaves the femoral neck
substantially intact. Thus, when the acetabulum is dysplastic, could more
femoral heads be saved with use of avascular techniques that provide better
subchondral support after thorough débridement?
I commend Roush et al. for critically reviewing the failures of free
vascularized fibular grafting in a series of 200 hips, but I strongly believe
that the article would have been more helpful had the authors discussed how
the surgical technique of free vascularized fibular grafting, having not
changed in nearly twenty years, may have contributed to destabilizing a
femoral head with increased contact forces due to acetabular dysplasia. The
work of the senior authors, Urbaniak2 and Steinberg7, is
well recognized in the literature. However, as a new generation of
orthopaedists develops interest in this devastating disease, we must recognize
that perhaps free vascularized fibular grafting cannot be uniformly applied to
all hips as implied by Roush et al. More importantly, treatment protocols
should focus on the features that vascular and avascular bonegrafting
techniques have in common when such features are associated with good clinical
outcomes.