Keywords
metacarpal bones - boxer's fracture - fracture fixation, intramedullary - minimally
invasive surgical procedures - clinical trials, randomized
Introduction
Despite the high prevalence (20% of the hand fractures) of unstable neck metacarpals
fractures (boxer's fractures), there is still no consensus concerning the preferred
method and ideal moment of treatment, especially in active patients[1]
[2] for whom the time or type of management can have a strong psychological impact on
the outcomes.[3]
[4]
The use of the intramedullary technique (headless screws or bouquet technique fixation)
as the definitive treatment of unstable boxer's fractures in active patients in the
first week may be a good choice of treatment. This technique is a fast, safe, minimally
invasive, and easily performed reproducible method, without addressing the extensor
tendon to prevent tendon adhesion and joint stiffness, enabling earlier functional
recovery and shortening the return-to-work time of these patients. Choose a reproducible
and effective method, which presents a cost / benefit compatible with our reality.[5]
[6]
[7]
The goal of the present study is to compare the return-to-work time, visual analogue
scale (VAS) score, disability of the arm, shoulder, and hand (QuickDASH) score, complications
rate, and radiographic outcomes of two methods of definitive internal fixation in
active patients with boxer's fractures, operated in the first week.
Methods
A double center, parallel group, prospective, randomized clinical trial was conducted
at the department of hand surgery of our institution. Two implants were used for fixation
in closed reduction of boxer's fractures ([Fig. 1]): the headless cannulated screws (Synthes, Davos) and Kirschner wires (k-wires)(bouquet
technique) (Synthes, Davos). The protocol was approved by the local research ethics
committee (CAAE 127.59813.4.0000.0082). All patients received, signed, and are aware
of what was recommended in the informed consent (IC). The inclusion criteria were:
presence of an acute (0–7 days), closed, and simple fracture of the metacarpal neck;
absence of an ipsilateral injury or deformity; presence of an angulation of over 40
degrees in oblique plane imaging; being adults or ≥ 18 years old; having acquired
a master's degree or being a student; and being an employee in a social-professional
environment with mild physical activities.
Fig. 1 Boxer's fracture mechanism and anatomic features—schematic drawing.
Rotational deformity was clinically assessed based on the extension of the axis of
the finger toward the scaphoid tubercle during flexion and orientation of the nail
of the finger during extension. Angulation of the fracture was evaluated by measuring
the angulation in the continuity of the dorsal cortical line of the metacarpal in
a 30° oblique X-ray image.
The trial was registered at Clinical Trials. gov No. 32925713.9.0000.0082.
Sample Size Calculation
Patients that received an intervention based on the diagnosis of boxer's fracture
between 2016 and 2017 were included in the present study after their consents were
obtained. For good results, 70 degrees (standard deviation [SD]: 5) of angulation,
as reported in this study, was accepted as the radiological threshold. Accordingly,
the power of the study was 80%, with an α value of 0.05, and each group was comprised
of 12 subjects.[8]
We Got a Similar Sample
During the study period, January 2016 to December 2017, a total of 45 patients met
the inclusion criteria and did not have any of the exclusion criteria ([Fig. 2]; Consolidated Standards of Reporting Trial [CONSORT] flowchart).[9] Five eligible patients were operated on by orthopedic surgeons not participating
in the study and were not included. Forty eligible patients and 41 fingers were operated
in the 1st week. A posthoc analysis showed that the 40 patients included in the study did not
differ regarding age, gender, or fracture type compared with the 5 not randomized
but eligible patients. Of these, 4 were women and 36 men, with an average of 30.46
years (range 16–54 years).
Fig. 2 Consolidated Standards of Reporting Trial (CONSORT) flowchart.
The patients were randomized by drawing lots (heads = A - treatment with k-wires;
tails = B - treatment with headless screws), which were printed and placed in 50 sealed
envelopes before the study started. Randomization was blinded to all examiners. Simple
randomization was used, and the envelopes were opened in the operating theater, immediately
prior to the surgery. Surgical instruments for both procedures were available in a
single box. Twenty patients were randomized to each group ([Tables 1] and [2]).
Table 1
|
ID
|
Age
|
Follow-up
|
Side
|
Final ROM
%
|
QuickDASH
|
Pain VAS
|
Return to work (days)
|
Complications
|
Other surgery
|
|
1
|
25
|
19
|
R
|
100
|
11.36
|
1
|
90
|
Y
|
Y
|
|
2
|
23
|
25
|
R
|
100
|
0
|
0
|
60
|
N
|
N
|
|
3
|
32
|
21
|
L
|
100
|
0
|
0
|
10
|
N
|
N
|
|
4
|
21
|
17
|
R
|
100
|
2.27
|
1
|
30
|
N
|
N
|
|
5
|
35
|
8
|
R
|
100
|
0
|
0
|
15
|
N
|
N
|
|
6
|
54
|
13
|
R
|
100
|
4.54
|
2
|
30
|
N
|
N
|
|
7
|
25
|
19
|
L
|
100
|
0
|
0
|
60
|
N
|
N
|
|
8
|
27
|
25
|
L
|
100
|
0
|
0
|
15
|
N
|
N
|
|
9
|
28
|
17
|
R
|
100
|
0
|
0
|
30
|
N
|
N
|
|
10
|
35
|
11
|
R
|
100
|
0
|
0
|
20
|
N
|
N
|
|
11
|
21
|
16
|
R
|
100
|
2.27
|
0
|
20
|
N
|
N
|
|
12
|
36
|
19
|
R
|
100
|
0
|
0
|
15
|
N
|
N
|
|
13
|
26
|
7
|
L
|
100
|
2.27
|
0
|
30
|
N
|
N
|
|
14
|
30
|
13
|
R
|
100
|
0
|
0
|
30
|
N
|
N
|
|
15
|
51
|
8
|
L
|
75
|
0
|
0
|
5
|
N
|
N
|
|
16
|
27
|
11
|
R
|
100
|
0
|
0
|
20
|
N
|
N
|
|
17
|
42
|
9
|
L
|
100
|
0
|
1
|
60
|
N
|
N
|
|
18
|
34
|
13
|
R
|
100
|
0
|
0
|
45
|
N
|
N
|
|
19
|
21
|
10
|
R
|
100
|
2.27
|
0
|
15
|
N
|
N
|
|
20
|
31
|
11
|
R
|
100
|
0
|
0
|
30
|
N
|
N
|
Table 2
|
ID
|
Age
|
Follow-up
|
Side
|
Final ROM
%
|
QuickDASH
|
Pain VAS
|
Return to work (days)
|
Complications
|
Other surgery
|
|
1
|
36
|
15
|
L
|
100
|
4.54
|
1
|
82
|
N
|
N
|
|
2
|
28
|
12
|
R
|
100
|
0
|
0
|
13
|
N
|
N
|
|
3
|
52
|
12
|
R
|
100
|
0
|
1
|
37
|
N
|
N
|
|
4
|
41
|
11
|
R
|
100
|
2.27
|
0
|
43
|
N
|
N
|
|
5
|
18
|
11
|
R
|
100
|
0
|
0
|
11
|
N
|
N
|
|
6
|
28
|
11
|
R
|
100
|
2.27
|
1
|
71
|
N
|
N
|
|
7
|
28
|
11
|
R
|
100
|
4.54
|
1
|
64
|
N
|
N
|
|
8
|
27
|
10
|
R
|
100
|
0
|
1
|
44
|
N
|
N
|
|
9
|
16
|
9
|
R
|
100
|
0
|
0
|
14
|
N
|
N
|
|
10
|
42
|
8
|
L
|
100
|
0
|
1
|
28
|
N
|
N
|
|
11
|
38
|
8
|
R
|
100
|
0
|
0
|
11
|
N
|
N
|
|
12
|
24
|
7
|
R
|
100
|
0
|
1
|
37
|
N
|
N
|
|
13
|
34
|
6
|
L
|
100
|
2.27
|
0
|
51
|
N
|
N
|
|
14
|
29
|
6
|
R
|
100
|
0
|
0
|
48
|
N
|
N
|
|
15
|
19
|
6
|
L
|
100
|
0
|
|
37
|
N
|
N
|
|
16
|
21
|
6
|
R
|
75
|
15.9
|
3
|
54
|
Y
|
N
|
|
17
|
29
|
6
|
L
|
100
|
0
|
0
|
10
|
N
|
N
|
|
18
|
29
|
6
|
R
|
100
|
0
|
0
|
7
|
N
|
N
|
|
19
|
41
|
6
|
R
|
100
|
0
|
0
|
6
|
N
|
N
|
|
20
|
23
|
6
|
R
|
100
|
0
|
0
|
13
|
N
|
N
|
|
21
|
22
|
6
|
R
|
100
|
0
|
0
|
9
|
N
|
N
|
The groups were similar. Thus, that patients were available for the intention to treat
analysis at the 6 months follow-up (minimum). ([Figs. 3] and [4]) ([Figs. 5] and [6])
Fig. 3 Kirschner wire (K-wire) antegrade intramedullary “bouquet” fixation technique. (A)
Implants and clinical view of operated limb (B) Radioscopic confirmation K-wires entry
point (C) Entry point confection (D) Intramedullary K-wire advancement through fracture
site. (E) Patient, surgical team, and radioscope positioning (F) Instrumentation of
K-wire through subcutaneous entry point. Traction is being applied.
Fig. 4 Kirschner wire antegrade intramedullary “bouquet” fixation technique and radioscopic
final images. (A) Boxer's fracture radiographic features. (B and C) Final radioscopic
aspect—intramedullary bouquet (D) Final post operatory clinical aspect. (E) Schematic
drawing—intramedullary bouquet technique.
Fig. 5 Percutaneous headless screws intramedullary fixation technique. (A) Radiographic
features—Boxer's fracture (B) Radioscopic features—Boxer's fracture (C and D) Headless
screws retrograde intramedullary insertion.
Fig. 6 Percutaneous headless screws intramedullary fixation final clinical and radioscopic
images. (A–C) Radioscopic final aspect—headless screws fixation technique. (D and
E) Postoperative clinical aspect.
Surgical Technique Description
In all patients, surgery was performed with the use of general anesthesia. The surgical
method of treatment used was closed fracture reduction (the mechanism used was longitudinal
traction associated with a Jahss maneuver) and stabilization.
-
Group A (control): All the patients were operated with an antegrade k-wire stabilization technique.
A small incision was performed proximal to the base of the metacarpal; following subcutaneous
dissection, a hole was made through the ulnar or radial cortex of the metacarpal,
directed distally to open the canal, avoiding perforation of the contralateral cortex;
one or two 1.2- mm k-wires were bent at one end to control the direction of introduction.
The fracture was then reduced and the k-wires are introduced longitudinally, from
the metacarpal base up to the metacarpal head. Hardware positioning was controlled
intra operatively with an image intensifier ([Figs. 3 ]and [4]).
-
Group B (tested): A 0.5 cm incision was performed at the level of the metacarpal head, and the extensor
tendon was not approached, only avoided and separated, longitudinally to a similar
extent. Two 1.0-mm guide wires were inserted along the longitudinal axis of the metacarpal
bone under fluoroscopic guidance. The k-wires were over drilled and replaced with
either two 2.4 or 3.0-mm cannulated headless compression screws, based on preoperative
templating. The first screw was inserted until all of the distal screw threads surpassed
the fracture site, and after the second screw was placed, also surpassing the fracture
site. After hardware positioning was controlled with an image intensifier ([Figs. 5] and [6]).
All patients were radiographically and clinically assessed at 1, 2, and 8 weeks, and
at 6 months. Total joint (metacarpophalangeal + proximal interphalangeal + distal
interphalangeal) ROM was evaluated using a standard goniometer. The patient-reported
outcome was recorded using the QuickDASH questionnaire (range, 0–100, with 0 as best
result),[10]
[11] VAS (range, 0–10, with 0 as best result) for pain and return-to-work time.
Complications were treated and assessed at 6 months. In addition, shortening, rotation
and angulation values were measured on the 30th day follow-up for both groups. Data
are presented as mean or median according to type of data and distribution.
Statistical Methods
The Microsoft Excel spreadsheet in its version of MS-Office 2013 (Microsoft Corp.,
Redmond, WA, USA) for the organization of the data, and the statistical package IBM
SPSS Statistics for Windows, Version 24.0 (IBM Corp., Armonk, NY, USA) for obtaining
the results. In the statistical analyses, the level of significance of 5% (0.050)
was adopted. The Fisher exact test was used to verify possible differences between
both groups in terms of categorical variables.
The Mann-Whitney test was used to verify possible differences between both groups
in terms of scalar variables.
Results
Categorical variables regarding side, complications, and other surgeries are found
in [Table 3]. Scalar variables regarding age, ROM, quickDASH, VAS, and return-to-work time are
found in [Table 4].
Table 3
|
|
GROUP
|
Sig. (p)
|
|
A (K-wire)
|
B (screw)
|
|
Freq.
|
Perc.
|
Freq.
|
Perc.
|
|
Side
|
L
R
|
6
14
|
30%
70%
|
5
16
|
23.8%
76.2%
|
0.655
|
|
Complications
|
Y
N
|
1
19
|
5%
95%
|
1
21
|
4.76%
95.25%
|
0.627
|
|
Other surgery
|
Y
N
|
1
19
|
5%
95%
|
0
21
|
0%
100%
|
0.300
|
Table 4
|
Variable
|
Group
|
n
|
Man
|
Standard deviation
|
Min
|
Max
|
Perc.
25
|
Perc. 50 (median)
|
Perc. 75
|
Sig. (p)
|
|
Age (years)
|
A
B
Total
|
20
21
41
|
31.20
29.76
30.46
|
9.23
9.22
9.14
|
21
16
16
|
54
52
54
|
25
22.5
23.5
|
29
28
28
|
35
37
35.5
|
0.715
|
|
Follow-up (months)
|
A
B
Total
|
20
21
41
|
14.6
8.52
11.49
|
5.49
2.77
5.25
|
7
6
6
|
25
15
25
|
10.25
6
7
|
13
08
11
|
19
11
14
|
< 0.001
|
|
ROM (% opposite side)
|
A
B
Total
|
20
21
41
|
98.75
98.81
98.78
|
5.59
5.46
5.45
|
75
75
75
|
100
100
100
|
100
100
100
|
100
100
100
|
100
100
100
|
0.972
|
|
Quick DASH
|
A
B
Total
|
20
21
41
|
1.25
1.51
1.38
|
2.71
3.62
3.17
|
0
0
0
|
11.36
15.9
15.9
|
0
0
0
|
0
0
0
|
2.27
2.27
2.27
|
1
|
|
VAS
|
A
B
Total
|
20
21
41
|
0.25
0.52
0.39
|
0.55
0.75
0.67
|
0
0
0
|
2
3
3
|
0
0
0
|
0
0
0
|
0
1
1
|
0.14
|
|
Return to work (days)
|
A
B
Total
|
20
21
41
|
31.5
32.86
32.20
|
21.41
23.16
22.05
|
5
6
5
|
90
82
90
|
15
11
13.5
|
30
37
30
|
41.25
49.5
46.5
|
0.865
|
There are complications: a patient (A group) showed impingement because of the k-wires,
and they were removed after 3 months. A patient (B group) showed loss of reduction
and he did not undergo another surgery.
Discussion
The treatment of isolated metacarpal fractures with k-wire pinning has a long and
proven track record. This treatment is based on the concept of flexible fixation introduced
by Foucher,[7] who described the results of the antegrade pinning technique; on their series of
66 patients with 68 fractures, all of their patients returned to their previous activities;
6 patients had a 10° extension lag; and 6 other patients had a 15° extension lag;
however, only one patient, an auto mechanic, complained of this decrease in ROM. Intramedullary
fixation with headless cannulated screws follows the principles of rigid stable fixation;
it allows for early mobilization and decreases the need for postoperative casting.
Boulton et al.[12] described the use of the intramedullary headless compression screw technique for
the fixation of a fifth metacarpal comminuted neck fracture. The patient's metacarpophalangeal
joint flexion at the latest follow-up was 80°; her extension was full.
Our trial included only unstable, simple, or complex fractures of the metacarpal neck
(Boxer's fractures), and we had a control group with similar fractures treated with
antegrade k-wire pinning versus tested group and patients treated with headless cannulated
screws. Del Pinal et al.[6] and Couceiro et al.[13] showed retrospective studies, case series, and patients with shaft metacarpal and
proximal phalanx fractures were included in these studies. In the present study, there
were no differences between the two in terms of ROM, postoperative pain (VAS), or
QuickDASH score.
A criticism of traditional conservative treatment is the inability to use cast immobilization
to maintain reduction of the lateral inclination of the metacarpal bone. Ruchelsman
et al.[14] and the present study demonstrated that only 2.43% of the patients showed loss of
initial reduction. The current study is a prospective, randomized clinical trial,
and all patients were operated by two surgeons, upper limb trauma specialists, in
a uniform group, and with complete follow-up. There are limitations of the study,
the sample size was small for QuickDASH and VAS analysis, but sufficient for statistical
analysis for the evaluation of the ROM.
We did not find differences in terms of the mean return-to-work time or time back
to their regular activities between the groups. We found no differences in terms of
function or patient-related outcomes between the two techniques. We have been unable
to conclude that there were any benefits in the application of one particular technique
when compared with the other. The use of cannulated screws must be carefully weighed
by the surgeon. The potential downsides include higher implant costs, the production
of an injury to the articular cartilage, and the retention of metallic hardware.[13]
There is a strong trend toward the use of headless cannulated screws in the treatment
of boxer's fractures; although the results found were similar, new comparative studies
are needed for electing the best method.
Conclusion
In the treatment of active patients with unstable boxer's fractures, headless screws
and bouquet fixation proved to be a safe and reliable treatment. The outcomes were
similar in both groups, with satisfactory postoperative ROM, QuickDASH score, VAS
results, and with quick return to daily living activities.
Radiographic consolidation was observed in all of the cases.
