Keywords
scaphoid fractures - delayed union - percutaneous fixation - consolidation
Scaphoid fractures are the most common carpal fractures and carry a high risk of delayed union and pseudarthrosis. The incidence of scaphoid nonunion is between 5 and 25%.[1]
[2]
[3] Untreated nonunion will lead to degenerative changes and scaphoid nonunion advanced collapse.[4]
Treatment options for delayed union of the scaphoid consist of cast immobilization[5] or surgical treatment. Surgical options vary from percutaneous fixation to open reduction and internal fixation with or without bone grafting. Open reduction and internal fixation of the scaphoid has a higher complication rate, including scar problems, infection, nerve injury, and complex regional pain syndrome,[6] compared with percutaneous fixation that is much safer and a minimally invasive technique. However, only limited data are available for the use of percutaneous fixation in cases of delayed union of the scaphoid,[7]
[8] and it is often considered to be contraindicated.[4]
The definition of delayed presentation, delayed union, and nonunion are slightly variable in the literature. In this article, delayed presentation was defined as a lack of any sign of union 1 to 3 months postinjury. The cutoff between delayed union and nonunion was set at 6 months postinjury.[9]
[10]
[11]
[12]
The purpose of this study is to evaluate the use of transtrapezial percutaneous fixation with a headless bone screw without bone graft in a selected group of patients with delayed presentation, delayed union, or nonunion of a scaphoid fracture.
Material and Methods
A retrospective study was performed to analyze the outcome of the percutaneous fixation of selected cases of delayed presentation, delayed union, or nonunion of the scaphoid waist in the period between 2006 and 2011. Sixteen patients were included in this study: fifteen males and one female with an average age of 27 years (range: 17–61 years). The average time between trauma and surgery was 5 months (range: 2–27 months). A delayed presentation was present in six patients, a delayed union in eight, and a nonunion in two. The scaphoids nonunion was graded according to the revised classification of scaphoid nonunion.[13] Two patients were categorized as grade 1, 2 as grade 2, and 12 as grade 3 nonunion. All cases were untreated late presentations of scaphoid fractures. In seven patients, computed tomography (CT) scans of the scaphoid were obtained to confirm the diagnosis and to decide on further treatment. Patients who needed bone graft, with sclerotic bone changes at the fracture site or cortical bone resorption with a displacement of more than 2 mm, were excluded from the study based on the preoperative radiographic assessment. According to the Slade and Geissler scaphoid nonunion classification system, all class IV to VI nonunions were excluded from this study[13] as well.
A percutaneous, transtrapezial approach was used in all cases to obtain central placement of the screw. The surgical technique was previously described already.[14] The central axis of the scaphoid was determined with a guidewire under fluoroscopy and marked on the skin. Then, a guidewire was drilled through the trapezium, along the central axis of the scaphoid into the proximal cortex of the scaphoid. To determine the appropriate length of the screw, a second guidewire was drilled trough the trapezium into the distal cortex of the scaphoid. The difference in length between these two guidewires was set as the length of the screw needed. Finally, the guidewire was overdrilled with a 2-mm drill, and the headless Headless Bone Screw screw (KLS Martin, Tütlingen, Germany) was inserted.
Postoperatively, all patients were immobilized with a splint and were reviewed every 4 weeks until clinical and radiographic union was observed ([Figs. 1] and [2]). Between June 2010 and May 2013, they were invited to visit the outpatient clinic for a clinical examination (including grip strength and range of motion [ROM]) and radiographic examination; DASH (Disabilities of the Arm, Shoulder and Hand) and PRWHE (Patient-Rated Wrist and Hand Evaluation) outcome scores were also obtained. All preoperative and final follow-up radiographs were taken to analyze degenerative changes on scaphotrapezial arthritis using the modified Eaton and Glickel classification.[15]
Fig. 1 Preoperative computed tomography scan and X-ray images of a delayed union of a left scaphoid.
Fig. 2 X-ray image after percutaneous fixation of the scaphoid showing union of the fracture site.
Statistics
Nonparametric statistical tests were applied to account for small sample size. Spearman's rho test was used to investigate if the time to surgery was related to the time to union. A p-value less than 0.05 was considered statistically significant.
We analyzed the grip strength and ROM (flexion, extension, pronation, supination, radial deviation, and ulnar deviation) using the paired samples Wilcoxon signed-rank test. The seven parameters were analyzed using Bonferroni-adjusted α levels of 0.007 per parameter (0.05/7) to correct for multiple comparisons.
Results
Sixteen patients with late presentation of a scaphoid fracture were treated with percutaneous screw fixation without bone graft ([Appendix]). All included patients attended the final review appointment. Union was obtained in 15 (94%) of the 16 patients at a mean of 4 months following surgery (range: 1–9 months). There was no correlation between the time to surgery and the time to union (R = –0.298; p = 0.281). In one patient, the scaphoid classified as a grade 3 nonunion 27 weeks after initial injury failed to unite. The revision procedure in this patient was performed 19 months after the initial surgery, and union of the scaphoid was achieved in 6 months with CT scan assessment.
Average follow-up time was 36 months (range: 12–98 months). The average movement arcs and grip strength are summarized in [Table 1]. No significant statistical difference was found in ROM or grip strength between the operated and contralateral wrist.
Table 1
Average movement arcs (degrees) and grip strength (kg)
|
Injured
|
Uninjured
|
p-Values
|
|
Average flexion arc
|
83 degrees (range: 68–110)
|
87 degrees (range: 68–110)
|
0.138
|
|
Average extension arc
|
71 degrees (range: 36–100)
|
77 degrees (range: 52–110)
|
0.083
|
|
Average pronation arc
|
85 degrees (range: 75–92)
|
87 degrees (range: 80–100)
|
0.114
|
|
Average supination arc
|
89 degrees (range: 80–100)
|
90 degrees (range: 86–92)
|
0.577
|
|
Average radial deviation arc
|
27 degrees (range: 10–42)
|
28 degrees (range: 19–38)
|
0.798
|
|
Average ulnar deviation arc
|
54 degrees (range: 36–70)
|
54 degrees (range: 38–70)
|
0.838
|
|
Average grip strength
|
43 degrees (range: 23–58)
|
45 degrees (range: 25–59)
|
0.244
|
Note: This table includes the average movements arcs and grip strength of the injured and uninjured wrist.
Appendix
This table includes the ROM, grip strength, and DASH and PRWHE scores of all patients
|
Patient
|
Age (y)
|
Gender
|
Side of injury
|
Nonunion grade
|
Time to surgery (mo)
|
Time to union (mo)
|
Degenerative changes on ST joint preoperatively
|
Degenerative changes at final follow-up?
|
Grip strength (kg) (injured/uninjured)
|
ROM wrist (degrees) (injured/uninjured)
|
PRWHE score (points)
|
DASH score (points)
|
|
|
|
|
|
|
|
|
|
|
Flexion
|
Extension
|
Pronation
|
Supination
|
Radial deviation
|
Ulnar deviation
|
|
|
|
1
|
29
|
M
|
Right
|
3
|
5
|
2
|
2
|
2
|
57/53
|
90/100
|
50/68
|
82/80
|
86/86
|
28/24
|
42/58
|
4
|
3
|
|
2
|
30
|
M
|
Left
|
3
|
2
|
3
|
1
|
1
|
47/48
|
80/80
|
80/70
|
90/90
|
90/90
|
15/19
|
60/50
|
2
|
0
|
|
3
|
61
|
F
|
Right
|
2
|
3
|
4
|
1
|
1
|
23/25
|
84/92
|
52/52
|
90/100
|
90/90
|
26/26
|
64/68
|
6
|
4
|
|
4
|
26
|
M
|
Left
|
2
|
5
|
2
|
1
|
1
|
56/56
|
68/74
|
56/64
|
75/90
|
80/90
|
32/32
|
44/40
|
2
|
1
|
|
5
|
22
|
M
|
Right
|
3
|
2
|
1
|
1
|
1
|
58/57
|
90/80
|
90/90
|
90/90
|
90/90
|
20/20
|
60/60
|
0
|
0
|
|
6
|
21
|
M
|
Left
|
1
|
2
|
4
|
1
|
1
|
38/42
|
70/85
|
100/110
|
80/80
|
90/90
|
35/30
|
60/60
|
9
|
3
|
|
7
|
24
|
M
|
Right
|
3
|
2
|
2
|
1
|
1
|
44/44
|
90/90
|
50/62
|
92/90
|
88/92
|
18/34
|
58/60
|
2
|
7
|
|
8
|
36
|
M
|
Right
|
3
|
4
|
4
|
1
|
1
|
44/37
|
70/80
|
78/74
|
86/90
|
100/90
|
24/30
|
70/68
|
21
|
10
|
|
9
|
24
|
M
|
Left
|
3
|
2
|
7
|
1
|
1
|
44/43
|
85/74
|
75/68
|
90/90
|
90/90
|
42/38
|
40/40
|
0
|
0
|
|
10
|
17
|
M
|
Left
|
3
|
3
|
4
|
1
|
1
|
36/41
|
90/90
|
70/70
|
80/80
|
90/90
|
25/25
|
40/40
|
10
|
0
|
|
11
|
25
|
M
|
Right
|
3
|
4
|
4
|
1
|
1
|
56/59
|
75/90
|
90/90
|
90/90
|
90/90
|
34/32
|
36/40
|
2
|
3
|
|
12
|
17
|
M
|
Left
|
1
|
2
|
9
|
1
|
1
|
39/43
|
88/84
|
62/70
|
80/90
|
86/90
|
24/22
|
50/42
|
1
|
2
|
|
13
|
20
|
M
|
Left
|
3
|
10
|
3
|
1
|
1
|
51/49
|
100/100
|
85/90
|
80/80
|
90/90
|
30/30
|
70/70
|
4
|
3
|
|
14
|
24
|
M
|
Left
|
3
|
5
|
2
|
1
|
1
|
32/31
|
110/110
|
90/90
|
90/90
|
90/90
|
30/30
|
60/60
|
10
|
8
|
|
15
|
36
|
M
|
Right
|
3
|
4
|
4
|
1
|
1
|
36/47
|
70/90
|
80/80
|
80/80
|
90/90
|
10/25
|
60/65
|
35
|
18
|
|
16
|
20
|
M
|
Right
|
3
|
27
|
−
|
1
|
1
|
28/52
|
68/68
|
36/68
|
90/90
|
90/90
|
32/24
|
50/38
|
56
|
39
|
Abbreviations: DASH, Disabilities of the Arm, Shoulder and Hand; PRWHE, Patient-Rated Wrist and Hand Evaluation; ROM, range of motion; ST, scaphotrapezial.
The average postoperative DASH score was 6 (range: 0–39). The average postoperative PRWHE score was 10 (range: 0–56). According to the DASH questionnaire, 12 patients did not report any wrist pain at rest. The patient with failure of scaphoid union complained of moderate wrist pain, and one other patient reported mild wrist pain. No complications from the percutaneous technique were encountered. Scaphotrapezial arthritis progression was never noted in this series.
Discussion
Volar percutaneous screw fixation without bone graft was commonly used for acute scaphoid fractures with good reported outcomes.[1]
[16]
[17]
[18] However, limited data of the volar percutaneous screw fixation were described in the late presentations of the scaphoid fractures. Kim et al[8] included 12 patients and found a 100% union rate at 12 months follow-up with an average time to union after surgery of 3 months and the average DASH score of 9 (range: 0–21). Capo et al[7] used a volar percutaneous approach for scaphoid nonunion in 8 of 12 patients and obtained a union rate of 92% with an average time to union of 3 months and the average DASH score of 6 (range: 0–16) at final follow-up.
In this study, a union rate of 94% was achieved using a volar percutaneous transtrapezial screw fixation technique without bone graft.[14] There was no significant difference in ROM or grip strength between the operated side and the contralateral side at the final follow-up. The average time to union was 4 months and the average DASH score was 6 (range: 0–39). The technique allows placement of the screw exactly along the central axis of the scaphoid, which may correlate to biomechanical superiority.[19] This technique may not require dividing the volar carpal ligaments nor damage the blood supply of the scaphoid, and may result in a smaller scar.
Drilling of the scaphotrapezial joint did not lead to symptomatic osteoarthritis in this study, which was consistent with a short- to medium-term follow-up by Geurts et al.[17]
Careful evaluation of late presentations of scaphoid fractures, and patient selection for the procedure are important. CT scan has been shown to be superior to radiographs for the evaluation of scaphoid fracture displacement and union.[20]
[21] Therefore, this is often indicated when radiographs are not clear enough for final decision making. In this study, CT scans of the scaphoid were deemed necessary in eight patients to evaluate cyst formation, sclerosis, and fracture displacement and to decide on further treatment. Patients with delayed union of the scaphoid with or without cyst formation are good candidates for percutaneous screw fixation, as long as there are no signs of sclerosis or cortical bone resorption with fracture displacement.
Slade and Geissler proposed a revised classification of scaphoid nonunion in an effort to match the healing potential of a nonunion to a specific treatment algorithm.[13] Nonunion of grade I (fibrous union), grade II (fibrous nonunion with minimal sclerosis, <1 mm), and grade III (minimal bone resorption of the anterior cortical bone and limited fracture sclerosis, <2 mm) were found appropriate for rigid fixation without bone grafting. In this study, excellent clinical outcome was obtained with the technique for grades I to III.
This study has several limitations. First, our sample size was small because late presentations of scaphoid fractures were not encountered very often. Second, the time to union was assessed clinically and radiographically in 4 week intervals. Third, CT scans were not performed to confirm union of all scaphoids. In the postoperative follow-up, we will try to avoid unnecessary exposure to radiation. We will therefore only ask for postoperative CT scans when radiographs do not show progressive healing, when signs of loosening around the screw are present, or when patient symptoms do not resolve. Fourth, it is a retrospective study and no control group was available. For clinical evaluation, the contralateral hand was used to assess functional outcome.
Percutaneous screw fixation of the scaphoid was successful in carefully selected cases of scaphoid delayed presentation, delayed union, or nonunion. Cyst formation, less than 5 mm, is not a contraindication as long as there are no signs of fracture displacement. We do not recommend this technique if there is sclerosis or collapse at the scaphoid delayed presentation, delayed union, or nonunion site.
