To The Editor:
Regarding "Percutaneous Screw Fixation or Cast Immobilization
for Nondisplaced Scaphoid Fractures" (2001;83:483-8) by
Bond et al., I wish to comment on the methods used in this study.
The patients in Group II (cast immobilization) had the fractures
treated with a long-arm thumb-spica cast followed by a short-arm
thumb-spica cast. This regimen seems excessive and unnecessary, as
a previous study by Clay et al.1 showed
that a below-the-elbow (short-arm) Colles-type plaster cast is adequate
for treatment of fresh nondisplaced fractures of the scaphoid.
Perhaps the longer time taken for rehabilitation and return to
full-duty status by patients in Group II was due to the excessively
prolonged immobilization, which, as previously mentioned, is unnecessary.
Bond et al. stated that they reevaluated all patients at one
week after the start of treatment and subsequently at two-week intervals
until the fracture healed. At each follow-up visit, the wrist was examined
for snuffbox tenderness and radiographs were made. Do the authors mean
to state that, for patients in Group II, at each visit, the cast
was removed, the wrist was clinically examined, and the cast was
reapplied?
The authors judged fracture union with use of radiographs, which
is a fallacious and inaccurate method of assessment of these fractures.
A previous study by Dias et al.2 showed
that radiographs "cannot be used reliably and reproducibly
to assess union of a fractured scaphoid. Radiographic criteria should
not be used as an objective assessment of union in clinical studies."
C.D. Bond, A.Y. Shin, M.T. McBride, and K.D. Dao reply:
We appreciate the comments made by Mr. Agarwal regarding the
methods of our study. He stated that the use of a long-arm thumb-spica
cast followed by a short-arm thumb-spica cast was excessive and
unnecessary for the treatment of nondisplaced scaphoid fractures.
We chose to use this regimen on the basis of a study by Gellman
et al.3 that addressed the need
for immobilization of the elbow. In a prospective and randomized
study comparing the immobilization of nondisplaced scaphoid fractures
with a short-arm thumb-spica cast versus treatment with a long-arm
thumb-spica cast for six weeks followed by application of a short-arm
thumb-spica cast, Gellman et al. demonstrated a statistically shorter time
to union when the fractures were initially treated with the long-arm thumb-spica
cast3. The study by Clay et al.1, cited by Mr. Agarwal, focused on
the need for inclusion of the thumb in the short-arm cast used for
treatment of scaphoid fractures.
Mr. Agarwal suggested that the longer times needed for rehabilitation
and return to full-duty status were a result of the excessively
prolonged immobilization. While we and other authors agree that
the longer period of immobilization affects the time needed to return
to full-duty status4, the assumption
that the inclusion of the elbow in the cast for six weeks was the
cause of the delay is unsubstantiated. Gellman et al. demonstrated
that there was no difference in elbow motion whether or not the
elbow was included in the cast3.
Mr. Agarwal questioned the examination of the wrist for snuffbox
tenderness in the cast immobilization group. As described in the
Materials and Methods section, all patients in our study were evaluated
at one week after the start of treatment and then at two-week intervals
until fracture union. At each follow-up visit, the cast was removed,
a five-view radiographic series was made, and the fracture was assessed
for snuffbox tenderness. The cast was then reapplied after each
evaluation until the criteria for fracture union—that is, bridging
trabeculae on all radiographic views and a nontender fracture site upon
clinical examination—were met.
Mr. Agarwal’s final concern was in regard to the radiographic
evaluation of fracture union. Dias et al. evaluated twenty sets
of radiographs of scaphoid wrist fractures made at twelve weeks after
the injury2. Each set of radiographs
included a posteroanterior, lateral, semipronated, and semisupinated
view of the wrist. On the basis of radiographic appearance alone,
there was poor agreement regarding the achievement of fracture union.
The authors concluded that "the decision to discontinue
immobilisation after 8 to 12 weeks should be based on the absence
of marked tenderness in the scaphoid region and the absence of a clear
radiographic gap at the fracture line."2 Additionally,
the authors recommended that follow-up be extended for an additional
three to six months. In our study, the criteria for fracture union were
bridging trabeculae on all five radiographic views of the scaphoid
in addition to clinical appreciation of a nontender fracture site.
A radiograph of the long axis of the scaphoid (made with the wrist
in ulnar deviation and pronation) was included, as this view is specific
for the evaluation of waist fractures5.
Also, all patients were followed clinically and radiographically
for a minimum of two years. The criteria for fracture union were
strict and paralleled those used in other studies of fracture-healing6-8. We agree with Mr. Agarwal that
radiographic criteria alone should not be used
to assess fracture union; that assessment should be based on a nontender
fracture site, clinical examination, and radiographic evidence of
bridging trabeculae across the fracture site.