Abstract
Background:
Nondisplaced scaphoid fractures treated with prolonged cast immobilization
may result in temporary joint stiffness and muscle weakness in addition
to a delay in return to sports or work. Fixation of scaphoid fractures
with a percutaneous cannulated screw has resulted in a shorter time
to union and to return to work or sports. The purpose of this prospective,
randomized study was to compare cast immobilization with percutaneous
cannulated screw fixation of nondisplaced scaphoid fractures with
respect to time to radiographic union and to return to work.
Methods:
Twenty-five full-time military personnel with an acute nondisplaced
fracture of the scaphoid waist consented to be randomized to either
cast immobilization or fixation with a percutaneous cannulated Acutrak
screw (Acumed, Beaverton, Oregon) for the purpose of this study.
Time to fracture union, wrist motion, grip strength, and return
to work as well as overall patient satisfaction at the time of a two-year
follow-up were evaluated.
Results:
Eleven patients were randomized to percutaneous cannulated screw
fixation, and fourteen were randomized to cast immobilization. The
average time to fracture union in the screw fixation group was seven
weeks compared with twelve weeks in the cast immobilization group
(p = 0.0003). The average time until the patients returned
to work was eight weeks compared with fifteen weeks in the cast
immobilization group (p = 0.0001). There was no significant
difference in the range of motion of the wrist or in grip strength
at the two-year follow-up evaluation. Overall patient satisfaction
was high in both groups.
Conclusions:
Percutaneous cannulated screw fixation of nondisplaced scaphoid
fractures resulted in faster radiographic union and return to military
duty compared with cast immobilization. The specific indications for
and the risks and benefits of percutaneous screw fixation of such
fractures must be determined in larger randomized, prospective studies.
Scaphoid fractures represent the most common carpal fracture,
with an annual estimated incidence of thirty-eight fractures per
100,000 men1. Nondisplaced fractures
involving the waist or the proximal pole of the scaphoid are commonly treated
with immobilization in a long-arm thumb-spica cast for several weeks,
followed by treatment in a short-arm thumb-spica cast until the
fracture unites. The average reported time to union of a nondisplaced
fracture of the middle third of the scaphoid has ranged between
eight and twelve weeks2,3. During
cast immobilization, joint stiffness and muscle-wasting can occur,
resulting in additional time before the patient can return to full
activity4,5.
Open reduction with rigid internal fixation of acute nondisplaced
scaphoid fractures has had favorable results in several nonrandomized,
retrospective studies4,6-10. The
technique of percutaneous cannulated screw fixation of scaphoid
fractures has been described in both anatomical and clinical studies
and has shown promising results10-12.
The purpose of this randomized, prospective study was to compare
the time to fracture union, time to return to full military duty,
and short-term functional outcome after treatment of acute nondisplaced
fractures of the scaphoid waist with either cast immobilization
or percutaneous cannulated screw fixation.
Between January 1997 and January 1998, all full-time military
personnel who had sustained an acute nondisplaced fracture of the
scaphoid waist were evaluated at the Naval Medical Center San Diego for
participation in this study, which was approved by our institution’s
Investigational Review Board and Committee for Protection of Human
Subjects. Patients who were not evaluated within two weeks after
the injury, had a history of an untreated injury of the wrist, or
had a fracture with >1 mm of displacement that did not
involve the waist of the scaphoid or that was associated with a
scapholunate angle of >60° were excluded.
Sixty-two patients who met the inclusion criteria were identified,
and twenty-five of them agreed to participate. After giving informed
consent, patients were randomly assigned to one of two treatment groups
by the patient’s blinded selection of an envelope containing
a group number (Group I or Group II). Group-I patients underwent
percutaneous internal fixation with a cannulated Acutrak screw (Acumed,
Beaverton, Oregon). In Group II, the fracture was immobilized initially
in a long-arm thumb-spica cast and then in a short-arm thumb-spica
cast.
Surgical Technique
Surgery was performed with the patient under general or regional
anesthesia. The patient was placed in the supine position with the
arm abducted on a radiolucent arm-board. A tourniquet was applied but
was not routinely utilized. A towel roll was used under the supinated
wrist to allow for adequate wrist extension. A guide-wire (a 0.035-in [0.889-mm] Kirschner
wire) from the Acutrak cannulated screw set was introduced volarly,
entering the distal scaphoid tuberosity, and directed proximally,
dorsally, and ulnarly (Figs. 1-A, 1-B, 1-C, and 2). The image intensifier was used
to confirm that the wire was placed along the longitudinal axis
of the scaphoid and across the fracture site. Extension of the wrist
assisted in translation of the trapezium dorsal to the path of the
guide-wire. Screw length was measured indirectly with a second guide-wire or,
alternatively, with the measuring device available in the screw
set. An anti-rotation wire was placed parallel to the first guide-wire,
so that it crossed the fracture site and was away from the initial
guide-wire and thus did not interfere with the drill or screw (Figs. 3-A and 3-B).
A 3-mm incision was made around the initial guide-wire to allow
passage of the drill and screw. Blunt dissection around the guide-wire
was accomplished with a fine hemostat. The scaphoid was then drilled
by hand with the graduated cannulated drill, while the depth was
monitored with fluoroscopy. The cannulated screw was placed under
fluoroscopic guidance to ascertain the quality of the fracture reduction
and screw position. The anti-rotation guide-wire was removed, and
the wound was irrigated and then closed with a nylon suture. A well-padded
short-arm thumb-spica plaster splint was applied.
The postoperative dressing was maintained for seven to ten days.
After that time, a removable custom-fabricated Orthoplast short-arm
thumb-spica splint was worn except during bathing or during the early
active range-of-motion protocol and strength-maintenance program.
Use of the splint was discontinued when the fracture united (Figs. 4-A, 4-B, and 4-C).
Cast Immobilization
A long-arm thumb-spica cast, applied with the thumb in palmar
abduction, the interphalangeal joint free, the wrist and forearm
in neutral rotation, and the elbow flexed to 90°, was worn for six weeks.
Then a short-arm thumb-spica cast was applied in an identical manner,
with the exception that the elbow joint was not incorporated. This
cast was worn until the fracture united. All casts were applied
in the orthopaedic clinic by orthopaedic cast technicians to ensure
consistency in technique.
Analysis
Power Analysis
Power analysis based on previous reports of the time to union
of scaphoid fractures2,3,10,11 demonstrated
that, in order to detect a difference of one week in the time to
fracture union between groups, six patients were needed in each
group for b to equal 0.8. To detect a difference of half a week in
the time to fracture union, twenty-five patients were required in
each group for b to equal 0.8. On this basis, the minimum number
of patients in each group was set at fifteen patients.
Patient Evaluation
All patients were reevaluated one week after the start of treatment
and then at two-week intervals until the fracture united. At each
follow-up visit, the wrist was examined for snuffbox tenderness, and
five radiographic views of the scaphoid (anteroposterior and lateral
radiographs, radiographs made with ulnar deviation with the forearm in
30° of pronation [scaphoid view] and with radial deviation,
and a clenched-fist radiograph) were evaluated for evidence of fracture
union. Union was considered to have occurred when there was no scaphoid
tenderness and when trabecular bone could be seen crossing the fracture
site on all five radiographic views13,14.
All radiographs were reviewed by a musculoskeletal radiologist and
by two of the authors (C.D.B. and A.Y.S.) to confirm union.
Once union was documented, the patients were reevaluated. Grip
strength was measured with a Jamar dynamometer (Sammons Preston,
Bolingbrook, Illinois) with use of the third ring with the elbow
flexed 90° and the forearm in neutral rotation; the average of three
trials was recorded for each hand. The range of motion of the wrist
(extension, flexion, radial deviation, and ulnar deviation) was
measured with a goniometer. The grip strength and the range of motion
were measured every three months for two years. The time until the
patient returned to full military duty—that is, when he
or she could perform his or her previous occupation without modification
or pain—was determined. Complications of either the cast
immobilization or the percutaneous screw fixation were recorded,
as was the patient’s final satisfaction with the outcome of
the treatment. Satisfaction was rated on a 5-point scale, with 4
points indicating very satisfied; 3 points, satisfied; 2 points,
neutral; 1 point, dissatisfied; and 0 points, very dissatisfied.
Statistical Analysis
Demographic data were analyzed with use of descriptive statistics.
Differences between groups with regard to time to fracture union
and return to duty were evaluated with use of a Wilcoxon rank-sum
test. Linear regression was used to determine whether there were
differences between groups with respect to age, gender, handedness,
injured extremity, or duty status. Chi-square and t tests were used
to assess differences between groups with respect to range of motion
and grip strength at the time of follow-up. Statistical analysis
was performed with use of SPSS for Windows (version 6.1; SPSS, Chicago,
Illinois) with significance set at a £ 0.05.
Twenty-five patients, twenty-two
men and three women with an average age of twenty-four years (range,
eighteen to thirty-four years) who had a nondisplaced fracture of
the scaphoid waist, were enrolled in this investigation. All patients
were active-duty military personnel who had been on a full-duty
status prior to injury. Sixteen of the injuries were the result
of a fall on the outstretched hand, eight were due to an axial load,
and one was a twisting injury. Eleven patients were randomized to percutaneous
screw fixation (Group I), and fourteen were randomized to cast immobilization
(Group II). In Group I seven scaphoid fractures were on the dominant
side, whereas in Group II six were on the dominant side. The average
duration of follow-up for each group was twenty-five months (range, twenty-four
months to twenty-seven months).
Group I (Percutaneous Screw Fixation)
The average age of the nine men and two women in Group I was
twenty-four years. The average time (and standard deviation) to
fracture union was 7 ± 0.5 weeks, and the
average time until the patients returned to a full-duty status was
8 ± 0.7 weeks. At the two-year follow-up
evaluation, the total arc of flexion and extension of the injured wrist
averaged 139° ± 6° compared with 150° ± 5° for the uninjured wrist. Grip strength averaged 40 ± 2.5 kg, which was 95% of the grip strength
of the uninjured hand (42 ± 3.0 kg). There
were no nonunions. There was one complication related to the operative
technique: a screw that was proud at the entrance site into the scaphoid
tuberosity had to be removed, because of pain, seven months after
fracture union. The patient had no more pain after screw removal.
Overall patient satisfaction was high, with an average satisfaction
score of 3.8 ± 0.2 points.
Group II (Cast Immobilization)
The average age of the thirteen men and one woman in this group
was twenty-four years. The average time (and standard deviation)
to fracture union was 12 ± 0.7 weeks, and
the average time until the patients returned to a full-duty status
was 15 ± 0.7 weeks. At the two-year follow-up
evaluation, the total arc of flexion and extension of the injured wrist
averaged 124° ± 7° compared with
145° ± 5° for the uninjured wrist. Grip strength
averaged 36 ± 3 kg, which was 84% of
the grip strength of the uninjured hand (43 ± 3
kg). There were no nonunions or complications related to the application
or wearing of the cast. One patient, in whom the scaphoid fracture
had healed, subsequently had a perilunate fracture-dislocation,
eighteen months after union of the scaphoid fracture. The perilunate
fracture was treated with open reduction and ligament reconstruction.
This patient’s final follow-up data with respect to range
of motion and grip strength were omitted from the analysis. Overall,
the patients were satisfied with the results of the cast immobilization,
with an average satisfaction rating of 3.1 ± 0.4
points.
Comparison of Groups I and II
Multiple linear regression revealed no preoperative significant
differences between Group I and Group II with respect to age, gender,
handedness, or injured extremity (p = 0.83). The scaphoid
fractures in Group I (percutaneous screw fixation) healed approximately
four to five weeks earlier than those in Group II (cast immobilization).
This difference was significant (p = 0.0003). Similarly,
Group-I patients returned to full duty approximately seven weeks
earlier than did patients in Group II, a difference that was also
significant (p = 0.0001). No significant differences in
range of motion (p = 0.152) or grip strength (p = 0.351)
at the two-year follow-up evaluation could be detected. Overall
satisfaction was similar in the two groups, with no significant
difference detected (p = 0.125).
Nondisplaced fractures of the scaphoid waist have a high union
rate when treated with cast immobilization2.
However, the average time to fracture union can range from eight
to twelve weeks, after which temporary stiffness of the wrist is
almost universal4,5. Although
cast immobilization is associated with low rates of both morbidity
and long-term disability, the time until the patient returns to
unrestricted sports or employment may be prolonged. In the military
setting, the prompt return of personnel to a full, unrestricted
capacity after injury can be of critical importance to mission readiness.
In this study, we compared two methods of treatment of nondisplaced
fractures of the scaphoid waist. This fracture type was selected
for several reasons: it is a fracture pattern treated commonly at our
institution, it has a predictable rate of healing and a low rate
of nonunion when treated with immobilization, the fracture pattern
is more amenable to percutaneous fixation than are other scaphoid fracture
patterns, and radiographic assessment of union is easier to accomplish
than is that of other scaphoid fracture patterns.
In this randomized, prospective series, on the average, fractures
treated with internal fixation healed nearly four and a half weeks
before those treated with a cast and the patients returned to work
more than seven weeks sooner. These results are consistent with
those of other reports. Rettig and Kollias reported that athletes
returned to competitive sports at an average of 5.8 weeks after
open reduction and internal fixation8.
Huene’s patients returned to sports six to eight weeks
after acute open reduction and internal fixation7,
whereas those of O’Brien and Herbert returned to work at
an average of 3.7 weeks after such treatment4.
In contrast, in the series reported by Gellman et al., fractures
treated with a cast healed at an average of 9.5 weeks; return-to-work
data were not reported3. Inoue
and Shionoya retrospectively reviewed the results of seventy-nine
acute scaphoid fractures treated with a cast or percutaneous placement
of a Herbert screw11. In the surgery
group the fractures healed at an average of six weeks, whereas those
in the cast group healed at an average of 9.7 weeks. Laborers returned
to work at an average of 5.8 weeks in the screw fixation group and
at an average of 10.2 weeks in the cast group. Despite treating
a variety of fracture types (nondisplaced, oblique, and displaced
fractures), Inoue and Shionoya demonstrated that acute screw fixation
was statistically superior to cast immobilization with respect to
time to union and return to work (p < 0.001).
Limitations of our study include the relatively small sample
size and the lack of blinded review of the radiographs. The sample
size did not permit an accurate assessment of the complication rate.
The study was designed to enroll fifteen patients in each of the
treatment groups. However, after the conclusion of the first year
of the study, significant differences in time to fracture union
and return to full duty were seen. Therefore, the study was closed
to new patient enrollment, and the twenty-five patients formed the
cohort of the study.
There was a potential for observer bias in the evaluation of
the radiographs because the observer could not be blinded with respect
to whether the patient had had surgical treatment or cast immobilization. However,
the criterion for fracture union (bridging trabeculae on five radiographic
views of the scaphoid) was strict and paralleled the criteria used for
fracture union by other authors11,13,14.
Ideally, weekly radiographs would have been made to evaluate fracture-healing
and to detect anticipated differences in time to union between Group
I and Group II. However, practical considerations allowed radiographs
to be made only every two weeks. The patient’s compliance
with follow-up also influenced the frequency and interval of radiographic
assessment.
These limitations notwithstanding, this prospective, randomized
study of percutaneous screw fixation of acute nondisplaced fractures
of the scaphoid waist demonstrated significantly faster fracture
union and earlier return to work compared with cast immobilization.
Still, cast immobilization remains a safe, effective, and time-tested
method of treatment of acute nondisplaced fractures of the scaphoid
waist. We do not advocate the surgical treatment of every one of
these fractures. Larger prospective, randomized series are needed
to assess the complication rates of surgical treatment. A critical
risk-benefit analysis is necessary to determine the optimal treatment
of acute nondisplaced fractures of the scaphoid waist.
The overall excellent results, high patient satisfaction, and
low morbidity associated with percutaneous internal fixation in
this study make it a reasonable alternative for treatment of acute
nondisplaced fractures of the scaphoid waist in patients desiring
rapid return to work or athletics.
Kuschner SH; Lane CS; Brien WW; and Gellman H: Scaphoid fractures and scaphoid nonunion. Diagnosis and
treatment. Orthop Rev,1994.23: 861-71, 23861Â
1994Â
[PubMed] Â
Cooney WP; Dobyns JH; and Linscheid RL: Fractures of the scaphoid: a rational approach to management. Clin Orthop,1980.149: 90-7, 14990Â
1980Â
[PubMed] Â
Gellman H; Caputo RJ; Carter V; Aboulafia A; and McKay M: Comparison of short and long thumb-spica casts for non-displaced
fractures of the carpal scaphoid. J Bone Joint Surg Am,1989.71: 354-7, 71354Â
1989Â
[PubMed] Â
O’Brien L, and Herbert T: Internal fixation of acute scaphoid fractures: a new approach
to treatment. Aust N Z J Surg,1985.55: 387-9, 55387Â
1985Â
[PubMed] Â
Skirven T, and Trope J: Complications of immobilization. Hand Clin,1994.10: 53-61, 1053Â
1994Â
[PubMed] Â
Herbert TJ, and Fisher WE: Management of the fractured scaphoid using a new bone
screw. J Bone Joint Surg Br,1984.66: 114-23, 66114Â
1984Â
[PubMed] Â
Huene DR: Primary internal fixaton of carpal navicular fractures
in the athlete. Am J Sports Med,1979.7: 175-7, 7175Â
1979Â
[PubMed] Â
Rettig AC, and Kollias SC: Internal fixation of acute stable scaphoid fractures in the
athlete.. Am J Sports Med.,1996.24: 182-6, 24182Â
1996Â
[PubMed] Â
Rettig AC; Weidenbener EJ; and Gloyeske R: Alternative management of midthird scaphoid fractures
in the athlete. Am J Sports Med,1994.22: 711-4, 22711Â
1994Â
[PubMed] Â
Wozasek GE, and Moser KD: Percutaneous screw fixation for fractures of the scaphoid. J Bone Joint Surg Br,1991.73: 138-42, 73138Â
1991Â
[PubMed] Â
Inoue G, and Shionoya K: Herbert screw fixation by limited access for acute fractures
of the scaphoid. J Bone Joint Surg Br,1997.79: 418-21, 79418Â
1997Â
[PubMed] Â
Kamineni S, and Lavy CB: Percutaneous fixation of scaphoid fractures. An anatomical
study. J Hand Surg,1999.24: 85-8, 2485Â
1999Â
Â
Daly K; Gill P; Magnussen PA; and Simonis RB: Established nonunion of the scaphoid treated by volar
wedge grafting and Herbert screw fixation. J Bone Joint Surg Br,1996.78: 530-4, 78530Â
1996Â
[PubMed] Â
Rajagopalan BM; Squire DS; and Samuels LO: Results of Herbert-screw fixation with bone-grafting for
the treatment of nonunion of the scaphoid. J Bone Joint Surg Am,1999.81: 48-52, 8148Â
1999Â
[PubMed] Â