To The Editor:
We thoroughly enjoyed reading the article, "Chondroblastoma
of Bone" (82-A: 1140-5, Aug. 2000), by Ramappa et al. Coincidentally,
we have reviewed our own experience with this subject and have found
that nine (13%) of our sixty-nine patients had a local
recurrence, a figure nearly identical to that from the Boston study.
We were unable to identify any risk factors for local recurrence, although
two of eight patients with proximal femoral lesions had a local
recurrence. The only other frequent sites of recurrence were the
proximal part of the humerus (two of sixteen) and the proximal part
of the tibia (two of thirteen).
We have used curettage alone without bone graft or cement in
the vast majority of patients and have been quite happy with the
results. It was not clear from the Ramappa paper whether bone cement
was used at all in the immature skeleton. We would certainly be
very wary of using it in children because of the potential risk
of damaging the articular cartilage and the epiphyseal plate.
A challenge for us, and I daresay for Ramappa and his colleagues,
has been the treatment of proximal femoral lesions, particularly
in children. We would be most interested to hear their comments
about how they approached these lesions—specifically, whether
they used a very limited approach through the neck of the
femur under image-intensifer control or preferred a direct approach.
If the latter, how do they achieve this? We have been surprised
with the results in one of our patients in whom recurrent chondroblastoma
was treated by making a trap door in the femoral head and curetting
out the defect through it before replacing the articular cartilage. After
three years of follow-up, there was no sign of recurrence or developing
chondrolysis.
We would like to bring to these authors’ attention our
recent review of our experience with metastatic chondroblastoma1, which showed quite clearly that
the risk of metastasis is increased dramatically by local recurrence.
We currently recommend that any patient with a recurrent chondroblastoma
should have a radiograph of the chest and be followed with regular
radiographs because of the risk of metastatic disease.
H.J. Mankin, A.J. Ramappa, F.Y.I. Lee, P. Tang, J.R.
Carlson, and M.C. Gebhardt reply:
We would like to thank our colleagues in Birmingham for their
comments and discussion of our recently published paper detailing
our experience with chondroblastoma. It is interesting that their experience
in relation to recurrence is virtually identical to ours in terms
of rate and anatomical site (around the hip).
In response to their first query, we have not used bone cement
in children, and none of the eight patients in our series who were
so treated were younger than nineteen years of age. The correspondents
are correct in expressing concern about the use of polymethylmethacrylate
in proximity to an open epiphysis, and we would certainly agree
that this should not be done. However, we have not encountered any
difficulty with the articular cartilage in the eight patients in
our study who were treated with polymethylmethacrylate or in the ninety-three
patients with giant-cell tumor who have been so treated since 1988.
We believe that the articular cartilage is much more tolerant of
adjacent polymethylmethacrylate than is the epiphyseal plate.
With regard to the second query, we also have had difficulty
with proximal femoral lesions. As indicated in our article, one
of the proximal femoral lesions was associated with a pathological
fracture; another required an osteotomy, presumably because of epiphyseal
damage; and another was associated with the development of an angular
deformity and limb-length discrepancy following surgery. All three
complications responded to subsequent surgery, but it is clear that
proximal femoral lesions are more likely to be associated with such
complications and, indeed, to recur locally.
As to their third query, our approach to proximal femoral lesions
was always through the base of the femoral neck or trochanter and
never through the articular cartilage.
We certainly agree with their point about metastatic chondroblastoma,
which is much more likely to occur in association with recurrent
disease. We also agree that the same may be said for giant-cell
tumor and, as indicated in our recent article2,
for chrondrosarcoma as well.