This retrospective study was performed in a regional trauma hospital in Delémont, Switzerland. The study was approved by the local ethical committee and conducted following the STROBE guidelines. (19) Patients who underwent surgery using ORIF or MIPO for the treatment of distal fibula fractures between 2010 and 2014 were consecutively included. These patients were then retrospectively divided into two groups according to the operative technique used.
Fractures were classified using the Danis-Weber classification as recommended by both the Orthopaedic Trauma Association (OTA) and the Arbeitsgemeinschaft für Osteosynthesefragen (AO Foundation) (20, 21).
Inclusion criteria were all distal fibular fractures requiring an operative treatment (Danis-Weber type B ≙ AO type 44 B1, 2, 3 and Danis-Weber type C ≙ AO type 44 C1, 2). Exclusion criteria were complex pilon fractures (AO43C3), Maissoneuve fractures (AO44C3), bilateral leg fractures and patients who had undergone previous surgery at the fracture site. In addition, patients were excluded if they suffered from existing disorders that might affect healing process and function, such as congenital deformities or neurologic disorders.
In total, 70 patients matched the inclusion criteria and were finally included in the present study. The baseline demographic and clinical data are shown in table 1. All surgeries were done by one of the four senior surgeons of the institution, all of them with minimum 10 years of experience in trauma surgery.
For soft tissue evaluation, an empiric classification was used, correlating the local status with the energy of trauma and creating three different groups. “Low” for minor trauma with mild soft tissue lesion, “medium” for pro-supination trauma or direct contusion trauma with no open fracture associated and “high energy trauma” for motor vehicle collisions with massive soft tissue involvement or open fractures.
Table 1 Demographic and clinical data at baseline
Factor
|
MIPO (n=35)
|
ORIF (n=35)
|
t-test
|
p-value
|
Age at injury (yrs)
|
54.8±17.6
|
52.2±11.3
|
t=0.731
|
0.467
|
Male sex (n)
|
17 (49%)
|
12 (34%)
|
I²=1.472
|
0.225
|
Osteoporosis (n)
|
1 (3%)
|
2 (6%)
|
I²=0.348
|
0.555
|
Peripheral artery disease (n)
|
1 (3%)
|
1 (3%)
|
I²=0.000
|
1.000
|
Diabetes mellitus (n)
|
3 (9%)
|
2 (6%)
|
I²=0.215
|
0.643
|
Smoking (n)
|
6 (17%)
|
8 (23%)
|
I²=0.357
|
0.550
|
Fracture classification (n)
|
Weber B = 29
Weber C = 6
|
Weber B = 29
Weber C = 6
|
I²=0.000
|
1.000
|
Level of energy of trauma (n)
|
High = 8 (23%)
Medium = 4 (6%)
Low = 25 (71%)
|
High = 4 (11%)
Medium = 4 (11%)
Low = 27 (77%)
|
I²=2.077
|
0.354
|
Ex-Fix placement (n)
|
9 (26%)
|
11 (31%)
|
I²=0.280
|
0.597
|
All variables were reported in terms of counted cases and relevant percentages and compared with the I²-test, with the exception of age at trauma reported in terms of mean and standard deviations and compared by means of t-test.
Description of surgical technique
For both, open reduction and internal fixation (ORIF) and minimally invasive plate osteosynthesis (MIPO) the surgical recommendations of the AO Foundation were strictly followed (22).
Surgery was done after soft tissue swelling had settled. In case of high-energy trauma or subluxated fracture with prominent swelling, a two-step approach, using a temporary fracture stabilization with an external fixator, was performed.
Patients of both groups were placed in supine position (supine with a bump under the ipsilateral hip with the knee slightly flexed) on a radiolucent table. If an external fixator was in place, all the bars and pins were removed.
In the ORIF group, an open surgical approach was established. The skin incision was lateral to the fibula and slightly anterior if additional access to the anterior syndesmosis was required. The area of the fracture was uncovered and gently reduced with one or two Weber clamps. If required, a lag screw was inserted. Then a plate was placed according to the AO technique. Depending on the fracture morphology and bone quality either a 1/3 tubular plate (DePuy Synthes, Oberdorf, Switzerland), a 1/3 tubular locking compression plate (LCP) (DePuy Synthes, Oberdorf, Switzerland), a Sidewinder Plate System (Trimed, Santa Clarita, California) or a preformed distal fibula LCP (DePuy Synthes, Oberdorf, Switzerland) were used.
In the MIPO group, the correct dimension of the plate (in these cases only LCP 1/3 tubular plate or preformed distal fibula LCP) was chosen based on preoperative radiographic planning. A tourniquet was used for the whole time of operation with a pressure of 100 mmHg above the systolic arterial pressure of the patient. Under fluoroscopic control, the tip of the malleolus was identified, and a slightly curved incision of 2 cm length was made distally to the tip. A distal locking drill sleeve was placed in the plate and used as a grip. Then the plate was slid subcutaneously along the fibula in a retrograde fashion, care being taken not to create false pathways. Then a second locking drill sleeve was placed distally and centering the plate onto the fibula with good bone contact, a locking screw was inserted into the most distal plate hole.
In some cases, this maneuver already indirectly reduced the fracture. If not, closed reduction was accomplished with assistance of a toothed reduction forceps (Figure 1). Correct length and rotation of the fibula in relation to the talus and distal tibia were assessed under fluoroscopic control (Figure 2). Fractures on frontal plane were reduced with a 2,7 mm bicortical lag screw through a stab incision of the skin and placed perpendicular to the fracture. After reduction, the plate was set with locking head screws through a small 2 cm incision made over the proximal portion of the plate (Figure 3).
Correct position of plate and screws were fluoroscopically documented in mortise and lateral views (Figure 4). In both groups a stress testing on the ankle was performed to assess syndesmotic stability. In case of instability of the AITFL and PITFL (anterior and posterior inferior tibiofibular ligament) the syndesmosis was stabilized using two tricortical screws through plate holes. After syndesmotic stabilization, another stress radiograph was performed and in case medial clear space showed to be above 5 mm, we performed a deltoid ligament repair with a Corkscrew FT Suture Anchor (Arthrex Inc, Naples, USA). Skin closure was performed in a standard manner with non-absorbable sutures (Figure 5).
Postoperative management
Both groups and all patients underwent the same postoperative management and follow up protocol. All ankles were immobilized in a VACOPed walker (Oped, Cham, Switzerland) with the foot at 90 degrees flexion and partial weight bearing was allowed with a load of 15 kg for 6 weeks. Range of motion training was initiated after 2 weeks in case wound healing allowed it. Full weight bearing was allowed 6 weeks postoperatively for radiographically consolidated fractures. In case of trans-syndesmotic fixation, the screws were removed 10 to 12 weeks postoperatively.
Follow-up
Clinical and radiographic follow-up was done 6 weeks, 3 months, 6 months and 1 year after surgery. Operation time measured between incision and wound closure as well as length of stay after surgery were noted.
Pain scores were recorded at one-year follow up. All patients were assessed using a visual analogue scale (VAS) for pain (0-10). VAS was classified into 4 groups; “no pain” for VAS=0, “low” for VAS=1-3, “moderate” for VAS= 3-5 and “severe” for VAS= 5-10.
Postoperative complications were recorded and defined as postoperative skin necrosis, nonunion, fracture-related infections, wound healing disorders and vascular-nerve injuries. Nonunion was defined as a fracture that has not completely healed within 9 months after injury or did not show any signs of healing for 3 consecutive months. (23)
Fractures were independently assessed for union in plain radiographs by the operating surgeon as well as by a trained radiologist. Bone healing was defined as absence of pain during weight bearing and bridging of at least three out of four cortices on both the anteroposterior and lateral view. In the event of uncertainty, a computed tomography scan was performed. Any disagreement between the operating surgeon and the radiologist was resolved by consensus.
To assess radiological outcomes, angular and spatial factors, containing the talocrural angle, lateral and medial clear space, tibiofibular overlap and talar tilt angle were measured in mortise view (anteroposterior view with 15° internal rotation) and compared in both procedures (Table 4).
Statistical analysis
Data management and statistical analysis were performed using SPSS (IBM Corp. Released 2016. IBM SPSS Statistics for Windows, Version 24.0. Armonk, NY: IBM Corp.). Continuous measures were summarized in terms of mean values and standard deviations, and these values were compared between the two groups using two-tailed unpaired t-tests. Countable variables were reported as counted values and relevant percentages, and they were compared between groups using Chi-Square test (χ2-).
The alpha level for the statistically significant threshold, for all the tests, was set at 0.05, and when significant the p-values are reported in bold in tables. A power analysis was conducted on the percentage of complications in the two groups using an alpha level set at 5%, finding a power of 80% for the performed analysis.