Non-displaced scaphoid fractures: a clinical trial of cast immobilization including vs. excluding the thumb

Scaphoid fractures occur in young active patients and have become notorious for
their potential for nonunion. Non-displaced fractures of the waist (midportion)
of the scaphoid can be treated successfully in a cast. The relationship between
the extent of immobilization and the elimination of non-union has still to be
demonstrated. Clay et al found comparable results when comparing thumb
immobilization with a cast that allows the thumb to move freely. Our standard
care for nondisplaced fractures of the scaphoid is to offer the patient cast
immobilization where a below elbow cast including the thumb is applied for 10
weeks.

Plain radiographs are the most common method of assessment of fracture union
but their lack of reliability is notorious . In particular the study of Dias
and colleagues which showed poor agreement among eight observers and an
unacceptably low degree of reproducibility when analyzing the inter-observer
agreement and reproducibility of radiographs for the determination of union at
twelve weeks. They concluded that the final outcome of radiographic union
should be based on appearances after a minimal follow-up of 6 months and
probably a year.

The primary goal of this study is to determine the union as determined on both
radiographically and computed tomography. This will allow us to compare these
two modalities for the determination of union 10 weeks and six months after
fracture*our hypothesis based upon existing literature is that CT scan will be
a more reliable predictor of eventual radiographic union 6 months after
injury.

As a secondary goal of this study, we will compare the two different
configurations of below-elbow cast in the treatment for non-displaced fractures
of the waist of the scaphoid: one including the thumb (so-called thumb spica
cast) and one not including the thumb (a standard below-elbow or short arm
cast). We will use the DASH questionnaire to measure health status at six
months and compare the outcome in these two groups.

The subjects will complete a data sheet with demographic information at the
initial visit. The nondisplaced scafoid fracture needs to be confirmed on
CT-scan. After obtaining informed consent, the patients will be randomly
assigned by computer to either short arm casting with thumb immobilization
(Group 1) or short arm casting without thumb immobilization (Group 2). At ten
weeks (visit 1), an x-ray and a CT scan will be obtained. At final follow up at
six months (visit 2), subjects will have an x-ray. If the fracture appears to
be healing on x-rays, no further imaging studies (CT scan) will be done. In
addition, Range of motion and grip strength will be measured. Subjects will be
asked to fill out a DASH questionnaire. Also, ordinal scale for pain, Mayo
wrist score and Gartland /Werely score will be obtained on physical
examination.

Patients will be randomly assigned by computer to either short arm casting with
thumb immobilization (Group 1) or short arm casting without thumb
immobilization (Group 2).

We do not anticipate treatment related risks related to participation in this
study, but it is possible (though unlikely) that one casting method will prove
inferior to the other with respect to healing of the fracture. It is standard
in many parts of the world not to include the thumb in a cast. Studies
addressing the difference in outcome between the two cast variations show
comparable results.