- Email: [email protected]
Lagged Syndesmotic Fixation: Our Clinical Experience
The Journal of Foot & Ankle Surgery xxx (2015) 1–9
Contents lists available at ScienceDirect
The Journal of Foot & Ankle Surgery journal homepage: www.jfas.org
Original Research
Lagged Syndesmotic Fixation: Our Clinical Experience Kwasi Yiadom Kwaadu, DPM 1, Justin James Fleming, DPM, FACFAS 2, 3, Trudy Salmon, DPM 4 1
Assistant Professor, Temple University School of Podiatric Medicine, Philadelphia, PA Fellowship Director, Philadelphia Foot and Ankle Fellowship, The Muscle, Bone, and Joint Center, Philadelphia, PA 3 Podiatric Residency Director, Aria Health Systems, Philadelphia, PA 4 Postgraduate Year-2 Resident, Aria Health Systems, Philadelphia, PA 2
a r t i c l e i n f o
a b s t r a c t
Level of Clinical Evidence: 4
Ankle fractures are very common, and although algorithms are in place for osseous management, consensus has not been reached regarding treatment of associated ligamentous injuries. Although tibiofibular syndesmotic stabilization can be done using different forms of fixation, the biomedical literature has long emphasized the risk of long-term restriction of ankle mobility with the use of lagged transfixation. However, when reduction cannot be maintained with positional fixation, we found that lagging the syndesmotic screw helped to maintain the reduction without causing functional restriction. In this report, we describe our experience with patients who had undergone lagged tibiofibular transfixation and were available for short- to intermediate-term follow-up to assess ankle function. A total of 31 patients (32.63% of 95 consecutive patients) were available at a mean of 34.87 (range 18 to 52) months to complete the American Orthopedic Foot and Ankle Society ankle-hindfoot questionnaire. The mean score was 88.38 (range 42 to 100) points at a mean follow-up interval of 34.87 (range 18 to 52) months. Of 31 patients, 19 had an AOFAS score of 90 points, 9 an AOFAS score of 80 to 89 points, 2 an AOFAS score of 60 to 69 points, and 1 an AOFAS score of <60 points. Because all syndesmotic screws were placed using the lag technique, unrestricted motion compared with the uninjured limb was used as the endpoint. All subjects had unrestricted motion compared with the uninjured limb, refuting the assertion that lagged syndesmotic screw fixation confers more restriction in ankle kinematics than positional syndesmotic fixation. Ó 2015 by the American College of Foot and Ankle Surgeons. All rights reserved.
Keywords: ankle fracture kinematics lag screw fixation ligamentous instability malleolus syndesmotic fixation tibiofibular syndesmosis
According to the U.S. National Trauma Center Database (2007 to 2011), ankle fractures accounted for 55.67% of all foot and ankle fractures, and 24.38% of these involved open fractures (1). The concurrent presence of injury to the deep deltoid and syndesmotic ligaments has been shown to further destabilize the ankle mortise, increasing the incidence and importance of operative reduction (2–4). The incidence of syndesmotic injuries in Weber B and C fractures has been reported to be as high as 66%, and improved functional outcome of the ankle joint after anatomic restoration of the unstable mortise has been elucidated (5). The ankle joint functions in a constrained system that tolerates misalignment and instability poorly, the presence of which results in accelerated degeneration of the joint (6). Ligamentous instability of the ankle mortise is addressed with anatomic reduction and fixation with positional syndesmotic screw fixation. Biomechanical studies have noted the physiologic motion
Financial Disclosure: None reported. Conflict of Interest: None reported. Address correspondence to: Kwasi Yiadom Kwaadu, DPM, Temple University School of Podiatric Medicine, 148 North 8th Street, Philadelphia, PA 19107. E-mail address: [email protected] (K.Y. Kwaadu). Video online only at http://www.jfas.org
present within the syndesmosis, and, as a result, concern for restriction has been pointed to as the prime reason for the use of nonlagged fixation across the syndesmosis (2,7). Despite evidence to the contrary that compression of the syndesmosis is not associated with restriction of ankle motion, the published data have continued to advocate against it (8). We found that in certain instances the reduction achieved with the bone tenaculums could not be maintained with positional fixation intraoperatively. We also found cases of late syndesmotic widening during the postoperative course in which the mortise appeared widened despite no obvious hardware failure. We thus began inserting all screws across the syndesmosis using the lag technique. Based on our prior clinical experience, we found our technique to be associated with the radiographic reduction maintained throughout the postoperative course with no late widening. In addition, we observed no functional loss or subjective complaints in patients who underwent this technique and thus present our outcomes with this technique. We specifically sought to demonstrate that from our clinical experience with this technique. In an effort to objectively evaluate our clinical experience, we reviewed the outcomes of 95 patients on whom we had performed lagged tibiofibular transfixation, and we were able to obtain ankle-related quality of life outcome measurements on 31 (32.63%) of these patients with the American
1067-2516/$ - see front matter Ó 2015 by the American College of Foot and Ankle Surgeons. All rights reserved. http://dx.doi.org/10.1053/j.jfas.2014.12.017
2
K.Y. Kwaadu et al. / The Journal of Foot & Ankle Surgery xxx (2015) 1–9
Fig. 1. Patient population of those initially identified and those who did not meet our search criteria.
Academy of Orthopaedic Surgeons (AOFAS) foot and ankle scoring system (9). Patients and Methods After receiving approval by the institutional review board, one co-author (T.S.), who was blinded to the results but took no part in patient care, performed a thorough medical record review of all consecutive ankle injuries requiring operative syndesmotic stabilization performed from January 2009 through December 2011 by the primary author (J.J.F.). These cases were identified using the Current Procedural Terminology (American Medical Association, Chicago, IL) code 27829, representing the open treatment of distal tibiofibular joint (syndesmosis). A total of 275 operations potentially eligible (for inclusion in our retrospective cohort) in 273 patients were initially identified using this search criterion. Our inclusion criteria were unstable syndesmotic injuries with or without operative fractures, syndesmotic stabilization only with screws placed using the lag technique, age 18 years, the ability to provide consent, closed injuries, and patient availability for evaluation using the AOFAS hindfoot questionnaire. Unstable syndesmotic injuries were defined as a tibiofibular clear space >6 mm and a medial clear space >5 mm (10–12). The exclusion criteria included any concurrent injuries beyond the identified ankle injury with the exception of osteochondral lesions identified intraoperatively, open fractures, syndesmotic screws placed using a nonlagged technique, any stabilization performed without screws, patients aged <18 years, patients with documented neuropathy, patients who were nonambulatory, patients lost to follow-up, patients with previous operatively or nonoperatively treated ankle fractures unrelated to the acute or index injury in question, and patients with previous complaints of ankle pain or instability as reported in the history and physical examination findings. Of the initial 275 operations, we excluded 12 (4.4%) patients who had sustained severe crushing highenergy injuries, 16 pediatric fractures (5.8%) in 16 patients, 7 fractures (2.5%) with concomitant fractures to the ipsilateral limb in 7 patients, 5 patients (1.8%) with 5 previous ankle fractures to the ipsilateral limb, and 10 open fractures (3.6%) in 10 patients (13). Four patients (1.4%) requiring 4 operations on the ipsilateral limb for concurrent injuries not involving the ankle joint were also excluded. Two patients requiring reoperation on postoperative day 0 and day 19 because of acute hardware failure, the first resulting from a fall and the second from gross ambulatory noncompliance, were included. These 2 patients, who had undergone 2 operations on the same limb, met our inclusion criteria, resulting in 97 operations in 95 patients. Thus, the total number of evaluated ankles equaled the same number of patients. An additional 124 patients (45%) were lost to follow-up, because they no longer resided at the addresses they had provided, the new residents had no recollection of the individual or how to reach them, and the telephone numbers provided had either been disconnected or assigned to another individual with no knowledge of the patient. The 97 operations (35.3%) in 95 patients of the initial 275 cases were available for medical record review and were included in the present study (Fig. 1). Finally, complications were defined as unplanned surgical intervention after the definitive open reduction and internal fixation.
Repair of the syndesmotic rupture was carried out in standard fashion. The surgical approach involved a lateral incision directly over the distal fibular and one over the medial malleolus when indicated (Fig. 2). In the presence of a concurrent operative posterior malleolar fracture, the standard lateral incision was moved posteriorly and placed halfway between the posterior border of the fibular and lateral border of the Achilles tendon to facilitate identification of the posterior malleolar fracture (Fig. 3). A posterior plafond fracture that extended medially into a medial malleolar fracture was approached posteromedially with a curvilinear J-shaped incision just posterior to the medial malleolus. Syndesmotic stabilization was performed with fully threaded cortical screws placed using the lag technique to 1 full turn above 2-finger tightness until the mortise was symmetric on fluoroscopy (Fig. 4). All intraoperatively identified osteochondral lesions were microfractured with a microfracture awl. None were >15 mm in diameter. For the purposes of this investigation, we contacted 31 (32.63%) patients to invite them to complete the American Orthopedic Foot and Ankle Society ankle-hindfoot questionnaire. We sought to evaluate the patients from a long-term functional perspective to determine the presence of any impairment that could be attributed to this technique. Of these 95 patients, 31 (32.63%) were available for evaluation using the AOFAS hindfoot clinical rating system and questionnaire. The remaining 64 patients (67.4%) had relocated and could not logistically participate in this portion of the evaluation. We prioritized the patient subjective reports, focusing on the uninjured contralateral limb as our control and previously described anatomic radiographic parameters (10–12). Postoperative management consisted of immobilization in a well-padded posterior splint with the ankle in neutral alignment. Range of motion exercises were begun when the wounds had “sealed” and the sutures and/or staples had been removed. Serial radiographs were obtained at weeks 2, 6, 10, 16, and 20 postoperatively unless the patient had been discharged from the practice before this time and then bimonthly until discharge if the patient required additional follow-up examinations. Progressive protected weightbearing was initiated when both radiographic and clinical union were present, as demonstrated by the absence of pain, edema, or erythema at the fracture sites.
Results The mean follow-up period for the 95 patients was 18 (range 10 to 46) months. Of the 95 patients, 55 were male, with a mean age of 49.58 (range 19 to 84) years, and 40 were female, with a mean age of 46.1 (range 19 to 81) years. Of our 95 patients, 39 (41%) had bimalleolar equivalent fractures, defined by the presence of an isolated fibular fracture with a medial clear space >5 mm, and 19 (20%) had trimalleolar equivalent fractures, defined by the presence of an isolated fibular and posterior malleolar fracture with a medial clear space >5 mm and without fracture of the medial malleolus. Twenty-six patients (27.4%) had trimalleolar
K.Y. Kwaadu et al. / The Journal of Foot & Ankle Surgery xxx (2015) 1–9
Fig. 2. Standard lateral incisional approach for anatomic restoration of the fractured fibula.
fractures, 5 (5.3%) had bimalleolar fractures, and 3 (3.2%) had Maisonneuve fractures, 1 of whom also presented with an operative posterior plafond fracture and a deltoid injury resembling a trimalleolar equivalent. No patient had bilateral injuries. Of the 95 patients, 46 (48.4%) had right-sided injuries and 49 (51.6%) left-sided injuries. Of our 95 patients, 7 (7.4%) underwent open reduction and internal fixation of their posterior malleolar fracture, because it involved 25% of the plafond evaluated on the sagittal computed tomography (CT) reconstruction scan. All fractures with a posterior malleolar component underwent CT evaluation. Of the 5 patients with a bimalleolar fractures, 2 (6.5%) were available for the AOFAS questionnaire and reported a score of 90 and 100. The patient with an AOFAS score of 90 complained of episodic weather-related uneasiness but functioned without restrictions. All 5 bimalleolar fractures in this AOFAS group had resulted from a lowenergy mechanical fall. At discharge for the 3 patients with a bimalleolar fracture who were not available for the AOFAS questionnaire, 2 had reported no complaints whatsoever. The third, however, had reported subjective and objective stiffness and had presented for intra-articular cortisone ankle injections at 8, 10, and 17 months postoperatively that ultimately resulted in symptomatic relief.
Fig. 3. Marked incision for the posterolateral approach in the presence of an operative posterior malleolar fracture.
3
Of the 39 patients with a bimalleolar equivalent fracture (41%), 11 (35.5%) were available for the AOFAS questionnaire. Their demographics are listed in the Table. Of this AOFAS subgroup, 1 fracture had resulted from an unknown mechanism. This patient recorded an AOFAS score of 42 at the 43-month follow-up visit. She was subsequently diagnosed with multiple sclerosis, which progressed during her postoperative course. Four fractures (36.4%) in this AOFAS group were high-energy injuries and six (45.4%) were low-energy rotational injuries. Only 1 of the 4 patients (25%) with a high-energy injury reported an AOFAS score of 90. In contrast, only 1 of the 6 patients (16.7%) with a lowenergy rotational injury reported an AOFAS score of 90. Of the remaining 28 patients (71.8%) with a bimalleolar equivalent fracture who were not available for the AOFAS questionnaire, 5 (17.8%) had sustained high-energy injuries involving falls 6 ft and 23 (82.1%) had sustained low-energy rotational injuries. The 5 patients with highenergy injuries who were not available for the AOFAS questionnaire had been discharged at a mean of 6.4 (range 2.5 to 8) months with clinical motion comparable to that of the uninjured contralateral limb and no subjective complaints. The 23 patients with a low-energy injury who were not available for the AOFAS questionnaire were discharged at a mean of 7.69 (range 2 to 35) months. Of these same 23 patients, 1 required a reoperation 3 weeks after the index procedure that involved reinsertion of the syndesmotic screw, which had moved retrograde as a result of early weightbearing. This patient proceeded with an uneventful convalescence otherwise. Another required hardware removal at 19 months because of prominence and irritation of the lateral plate. A third patient received an intra-articular cortisone ankle injection that resulted in symptomatic relief. One patient, however, complained of stiffness on discharge with less motion clinically than available in the uninjured contralateral limb. Of the initial 20 patients (21% of the 95 patients) with a trimalleolar equivalent fracture (19 true trimalleolar equivalent fractures by our definition and the third Maisonneuve with a mortise that presented as a trimalleolar equivalent), 8 (25.8%) were available for the AOFAS questionnaire. The mean follow-up period for these 8 patients was 39 (range 31 to 52) months. The mean AOFAS score for these 8 patients was 88.87 (range 64 to 100). Of these 8 patients, 3 had sustained a high-energy injury and 5 a low-energy rotational injury. Despite their scores, the subjective reports early in the postoperative course were less optimistic. Of the low-energy group, 1 patient (20%), who had undergone operative repair of a concurrent posterior malleolar fracture, developed stiffness and pain 6 months postoperatively that was unresponsive to an intra-articular cortisone injection. However, the patient experienced some symptomatic relief after hardware removal at 17 months postoperatively. Another who had concurrently sustained an operative posterior malleolar fracture complained of stiffness at 8 months, but it had resolved at their latest follow-up examination at 42 months. The remaining 3 patients (60%) who had sustained low-energy injuries and were available for the AOFAS questionnaire had had benign and good outcomes as defined by our criteria. In the high-energy group, 2 (66%) of the 3 patients were available for our AOFAS questionnaire. One recorded a score of 85 and complained of episodic weather-related stiffness that resolved with increased activity. The other recorded a score of 89 at the latest follow-up visit at 32 months after undergoing hardware removal and ankle arthroscopy at 7 months postoperatively (Fig. 5). Of the remaining 12 patients with a trimalleolar equivalent injury who were not in the AOFAS group, 5 (41.7%) had sustained highenergy injuries and 7 (58.3%) low-energy rotational injuries. In this subgroup of non-AOFAS high-energy fractures, 2 patients (20%) presented with residual complaints at their latest follow-up visit at 26 and 32 months, respectively. The former had sustained a concurrent operative posterior malleolar fracture after he was thrown off a motorcycle. He had been grossly noncompliant during the
4
K.Y. Kwaadu et al. / The Journal of Foot & Ankle Surgery xxx (2015) 1–9
Fig. 4. The sequential placement of a fully threaded lagged syndesmotic screw demonstrated on the anteroposterior fluoroscopic projection in which both cortices of the fibula were overdrilled. The lateral and medial tibial cortices were subsequently underdrilled. Only the lateral cortex in this instance was underdrilled because of the blocking effect of the medial malleolar screws. The syndesmotic screw was inserted to a 2-finger tightness and up to a full turn beyond if the mortise was still asymmetric.
postoperative course and presented at month 5 for his first postoperative follow-up visit. This patient subsequently developed radiographic changes of post-traumatic arthritis found 14 months postoperatively. The second patient continued to complain of residual stiffness at 10 months postoperatively. In the non-AOFAS lowenergy subgroup, 1 patient (14.3%) with a concurrent operative posterior malleolar fracture presented with clinical stiffness at 18 months postoperatively. Another was diagnosed with ankle synovitis and received intra-articular cortisone injections that offered symptomatic relief at 8 and 30 months postoperatively. An additional 2 patients (28.6%) reported weather-related changes that resolved with activity; however, on examination, both had hindfoot and ankle motion comparable to that of the uninjured contralateral limb. Of the 31 patients available for the AOFAS hindfoot questionnaire evaluation,18 (58.1%) were male and 13 (41.9%) were female. Of the 31 fractures, 16 (51.6%) were right-sided and 15 (48.4%) left-sided. Additional demographic data for this group are listed in the Table. Of
the 31 patients in the AOFAS group, 2 (6.4%) had had operative posterior malleolar fractures, both of which were associated with lowenergy injury mechanisms. Of the 64 patients in the non-AOFAS group, 2 (3.1%) had had operative posterior malleolar fractures, 1 each in the high-energy and low-energy groups. Of the initial 26 patients with a trimalleolar fracture, 8 (25.8%) were available for the AOFAS questionnaire. The mean score was 90.5 (range 60 to 100) for these 8 patients, 6 (75%) of whom had a score of 90. These 8 patients had all sustained low-energy rotational injuries, none of whom had sustained an operative posterior malleolar fracture. One patient (12.5%), however, had had a concurrent osteochondral lesion that was visualized intraoperatively and complained of stiffness at 7 months postoperatively. However, at the 38-month follow-up visit, the patient complained of mild and episodic pain with no restrictions otherwise. The demographics of the remaining patients can be reviewed in the Table. Of the 18 patients in the non-AOFAS trimalleolar group, 15 (83.3%) had had low-energy injuries and 1 (5.5%) had had an injury due to an unknown mechanism. This patient complained of pain
Table Demographic information for 31 stabilizations in 31 patients available to complete AOFAS questionnaire Age (y)
Mechanism
AO Fracture (14)
Fracture Details
Follow-Up (mo)
M M M
43 65 31
Rotational Rotational Rotational
44C 44B 44B
Right Maisonneuve fracture, trimalleolar equivalent Right bimalleolar equivalent fracture Left trimalleolar fracture
41 28 38
AOFAS Score 83 90 96
F M M
57 37 29
Thrown from horse Rotational Rotational
44B 44B 44C
Left trimalleolar equivalent fracture Left trimalleolar equivalent fracture Left trimalleolar fracture
38 52 32
85 64 60
F M
40 27
Rotational Fall >6 feet
44B 44C
Right trimalleolar fracture Left bimalleolar equivalent fracture
38 43
96 81
M
34
Rotational
44B
Left trimalleolar equivalent fracture
42
100
F M
37 64
Rotational Motor vehicle accident
44B 44B
Right bimalleolar fracture Right bimalleolar equivalent fracture
39 32
90 90
M
42
Rotational
44B
Left bimalleolar equivalent fracture
28
81
F
46
Fall >6 feet
44B
Left bimalleolar equivalent fracture
38
87
M M F F M M M F
32 62 33 49 64 46 62 63
Rotational Rotational Rotational Rotational Rotational Rotational Rotational Fall >6 feet
44B 44C High Ankle Sprain 44B 44B 44B 44B 44B
Right trimalleolar equivalent fracture Right bimalleolar equivalent fracture Left isolated syndesmotic injury Left bimalleolar equivalent fracture Right trimalleolar fracture Left trimalleolar equivalent fracture Right trimalleolar equivalent fracture Left bimalleolar equivalent
41 22 24 46 34 35 31 35
90 100 90 100 88 100 100 81
F M F
56 84 53
Rotational Rotational Rotational
44B 44B 44B
Right trimalleolar fracture Left trimalleolar fracture Right bimalleolar fracture
37 18 26
90 94 100
M M F
52 56 49
Assault Motor vehicle accident Rotational
44C 44B 44B
Right maisonneuve fracture Right trimalleolar equivalent fracture Left bimalleolar equivalent fracture
23 32 48
80 89 93
F F F M
47 35 55 51
Unknown Rotational Rotational Rotational
44B 44C 44A 44B
Right bimalleolar equivalent fracture Left trimalleolar fracture Right bimalleolar equivalent fracture Right trimalleolar fracture
43 36 32 29
42 100 100 100
Comments Hardware removal at 17 months postoperative Minimal aches with swelling Crepitus during postoperative month 7 Osteochondral injury Occasional stiffness at 38 months postoperative Occasional pain and stiffness on uneven terrain Severe pain, pain in toes, abnormal gait Constant but moderate pain Moderate restriction Occasional pain Pain with prolonged activity Cannot run at same pace Stiffness in ankle dorsiflexion during postoperative month 8; resolved at latest follow-up None Postoperative ankle arthritis and joint effusion at 1 year postoperative resolved with injection Episodic pain at 32 months postoperative Noncompliance Good range of motion at 4 months postoperative Syndesmotic ossification at 13 months postoperative Moderate pain, edema, and stiffness with running Ambulates with brace Edema and episodic fatigue Mild occasional pain. Removal of 1 of 2 syndesmotic screws at 5 weeks postoperative None None None None Excellent range of motion Stiffness in ankle dorsiflexion in postoperative week 2, resolved to excellent motion in postoperative month 2 None None Nondisplaced ipsilateral calcaneus fracture following a fall during postoperative month 5 None Retrograding syndesmotic screw removed during postoperative month 7 Radiolucency surrounding threads of syndesmotic screws during postoperative month 8 None None Delayed union at 2.5 months postoperative, with screw loosening None
K.Y. Kwaadu et al. / The Journal of Foot & Ankle Surgery xxx (2015) 1–9
Sex
Abbreviations: AOFAS, American Orthopaedic Foot and Ankle Society; F, female; M, male.
5
6
K.Y. Kwaadu et al. / The Journal of Foot & Ankle Surgery xxx (2015) 1–9
and stiffness on discharge at 6 months. The remaining 2 patients (11.1%) in the non-AOFAS trimalleolar fracture group had sustained vehicularmodulated injuries. Of these 18 patients, 3 (16.6%) had had an operative posterior malleolar fracture, none in the high-energy injury group. Of the 3 with an operative posterior malleolar fracture, 1 presented with residual stiffness and restriction of motion on discharge. Another patient (6.7%) in this low-energy, non-AOFAS trimalleolar fracture group sustained an osteochondral lesion. This patient continued to complain of pain at 1 year postoperatively and had clinically documented restriction. The third reported no complaints. The mean follow-up period for the trimalleolar non-AOFAS group was 11.13 (range 2.5 to 34) months. The only deep venous thrombus in the entire study population occurred in this group. Of the initial 95 patients, 3 (3.1%) had had a Maisonneuve injury, 1 of whom was described in the trimalleolar equivalent group, because the patient had also had an operative posterior malleolar fracture. Of the remaining 2 patients, 1 was available for evaluation with the AOFAS questionnaire. He recorded a score of 80 at 23 months and had moderate restriction demonstrated clinically compared with that of the uninjured limb. At the latest follow-up visit at 20 months, the second of these 2 patients presented with no subjective complaints but had documented restriction in range of motion. Of the initial 95 patients, 3 patients (3.1%) had had an isolated syndesmotic injury, with only 1 (3.2%) represented in the entire AOFAS questionnaire group. This patient had sustained a rotational injury and at the latest follow-up visit had complained of incisional pain. Another physician had removed his hardware 13 months after the index procedure. Of the remaining 2 patients, 1 had had completely normal examination findings at 11 months postoperatively but had sustained a fall and began complaining of pain thereafter. That patient underwent complete hardware removal at 22 months postoperatively and at the latest follow-up visit at 24 months reported no complaints. The final patient underwent syndesmotic stabilization and at the last follow-up visit at 11 months had presented with no symptoms and normal physical examination findings, with clinical motion comparable to that of the uninjured contralateral limb. Two of the 95 patients (2.1%) required acute reoperation and were included in the present study. The first reoperation occurred on postoperative day 0 after the patient had fallen on ice, bending the syndesmotic screws. The second underwent reoperation on postoperative day 19 because of frank hardware failure due to gross ambulatory noncompliance. A total of 7 patients (7.4%) underwent hardware removal. One patient underwent hardware removal 6 weeks after the initial stabilization with 2 syndesmotic screws. Radiographs at 3 to 5 weeks postoperatively demonstrated a distal syndesmotic screw moving progressively retrograde, with the proximal syndesmotic screw remaining intact along an anatomic mortise. At the latest follow-up visit, this patient had an AOFAS score of 100. Another patient, not in the AOFAS group, underwent removal of a lateral plate 4 months after the index procedure. The remaining 5 patients required hardware removal well after the postoperative course at a mean of 14.2 (range 7 to 22) months. None of our patients developed complex regional pain syndrome or deep infections. One patient experienced a deep venous thrombotic episode. Also, 1 patient experienced delayed wound healing as a result of a superficial infection that had resolved completely with local care and oral antibiosis at 28 days. Discussion Many factors contribute to ankle stiffness and restricted motion following injury. Tornetta et al (8) compared ankle range of motion in open chain before and after placement of a 4.5-mm lag screw across the syndesmosis in order to assess ankle restriction associated with
syndesmotic screw placement with the ankle in dorsiflexion. Despite their findings that ankle dorsiflexion was not restricted with syndesmotic compression, the investigators still discouraged the use of lagged fixation across the syndesmosis. The presence of posterior malleolar fractures has been associated with poorer outcomes (15). Stufkens et al (16) demonstrated that only 58% of ankle fractures with posterior malleolar fractures had good to excellent outcomes 4 years after injury. Furthermore, the retromalleolar approach, involving dissection and retraction of the flexor hallucis longus, can result in an indeterminable amount of scar tissue formation during convalescence that can result in ankle stiffness. The ankle capsule is torn during the rotational mechanism of the injury and at times with an iatrogenic insult. The healing of capsule occurs under a reparative pathway that involves replacement with scar tissue, which histologically lacks the elasticity of the native tissue. The generation and abundance of this inelastic scar tissue is under gene expression and individualized (Brigido S, personal communication, Greater Penn Education Foundation National Fellows Meeting, Pittsburgh, PA, May 17, 2013). It is not unreasonable to believe that this variability can confer undesired stiffness. Fractures and syndesmotic injuries occur in closed chain. Under these circumstances, the plafond internally rotates and torques on the fully loaded trochlear of the talus. The incidence of cartilaginous injury, even if macroscopically invisible, is likely greater than reported (Fig. 6). Hepple and Guha (5) reported on the use of arthroscopy and reported an incidence of intra-articular injury of 60% to 75%. Yoshimura et al (17) reported chondral injury in all 4 of their patients. Of these 4 patients, 1 presented with partial thickness fissuring of the cartilage of <50% of its thickness, and 3 presented with full chondral injury down to the subchondral bone. Loren and Ferkel (18) similarly reported a 63% rate of traumatic articular injuries in their cohort, 19 of which were located on the talus and 11 on the tibia. Also, the relative unpredictability of chondral healing of the talus has been elucidated (19). However the report by Mologne and Loren (20) of chondral tibial lesions points to another interesting confounder in outcomes, because tibial lesions have been notoriously associated with poor outcomes. Even beyond the chondral injury itself, arthrofibrosis native to the reparative process can similarly worsen ankle stiffness (21). Syndesmotic malreduction is another variable that can influence the outcome of these ankle injuries. Of their 68 patients, Sagi et al (22) reported that of the 13 who underwent direct open visualization and the 55 who underwent closed and indirect repair of their syndesmotic injury, 2 and 24, respectively, presented with malreduction on CT visualization. They also found that patients with malreduction presented with poor functional outcomes and further recommended CT evaluation of the contralateral limb and the postoperative reduction to help improve the outcomes. Gardner et al (23) reported a syndesmotic malreduction rate of 52% when subsequently evaluated using CT. Despite our best efforts, the 2-dimensional radiographic evaluation of the syndesmosis has had little bearing on the true anatomic reduction because subtle sagittal plane malreduction and, most difficult to evaluate, coronal plane rotation of the reduction can routinely occur (22,24–30). In cases in which the transmalleolar axis is not appreciated, the syndesmosis can also be malreduced (31). Darwish et al (32) applied a relatively novel approach to the effects of compression across the syndesmosis. The study investigators transected the syndesmotic ligaments and inserted a pressure sensor within the incisura. When the bone tenaculums were then placed along the transmalleolar axis and clamped to simulate reduction, the pressure monitor registered 61 N. After placement of a positional 3.5-mm cortical screw and release of the bone tenaculum, the pressure sensor reading decreased to 23 N. In contrast, the 3.5- and 4.5-mm cortical lag screws placed maintained the compressive effect initially achieved. The physiologic implications of this approach are
K.Y. Kwaadu et al. / The Journal of Foot & Ankle Surgery xxx (2015) 1–9
7
Fig. 5. (A) Right bimalleolar equivalent ankle fracture after a motor vehicle accident. (B) Axial computed tomography scan of a bimalleolar equivalent ankle fracture demonstrating a large posterior malleolar fracture with small intra-articular fracture fragments. (C) Sagittal reconstruction of the posterior malleolar fracture with impaction slightly anterior to the concomitant posterior malleolar fracture. (D) Sagittal reconstruction of a bimalleolar equivalent ankle fracture with fibular comminution and impaction. (E) Postoperative anteroposterior and lateral radiographs of a bimalleolar equivalent ankle fracture with double plating of the comminuted fibular fracture, open reduction with internal fixation of a posterior malleolar fracture, and lagged syndesmotic fixation. (F) Eight-month postoperative anteroposterior and lateral radiographs with second-look arthroscopy of a bimalleolar equivalent ankle fracture after a motor vehicle collision that underwent surgical repair with evidence of significant arthrofibrosis and residual chondral injury.
8
K.Y. Kwaadu et al. / The Journal of Foot & Ankle Surgery xxx (2015) 1–9
Fig. 6. Weber B fracture with hemarthrosis and chondral delamination along the entire lateral shoulder of the talus.
theoretical; however, the crux of their investigation demonstrated that the positional screw was unable to maintain even one third of the initial compression achieved by the bone tenaculum (1). In our observational study, we found that lag screw fixation of the tibiofibular syndesmosis did not appear to adversely influence the outcome of ankle fracture repair. Of the 31 patients available for longterm follow-up, the mean AOFAS score was 88.38 13.2 (range 42 to 100) (Table). Our study had multiple weaknesses. We performed no null hypothesis, because we were unable to evaluate a significant percentage of our initial population. Furthermore, we included no control group to compare the effects of positional transfixation. We did not investigate the effects of stainless versus titanium. However, the published data elucidate no statistically significant difference in this regard (12). No recent follow-up radiographs were available for our patient population who completed our AOFAS questionnaire to ascertain the fate of the ankle joint. Furthermore, the reliability of the objective
components of the AOFAS scoring system itself is not been validated or standardized (33). However, because it has been used for many previous studies, we also used it to allow for relative comparisons. Nonetheless, the rationale behind this thought and approach is fundamentally flawed. Also, although we referred to cases of clinically good motion, we included no actual measurements in the study. We were unable to objectively standardize the amount of force needed to reproduce a measurable and quantifiable amount of dorsiflexion; thus, we focused on the clinical, macroscopic, and patients’ subjective accounts regarding their perceived restrictions and motion. With this assumption, we also sought to prioritize patient subjective reports because the objective absence of 10 of dorsiflexion might be normal for 1 patient and not for another patient. Despite our best attempts, a comparison of motion in the contralateral uninjured limb was based on the educated assumption that the involved limb, before injury, would have had the exact same kinemantics, an educated guess at best, but a guess nonetheless. Most importantly, a significant portion of our population was lost to follow-up. We were not unaware of the outcomes of the study, which could potentially lend significant bias. We performed no comparison or additional statistical analysis of the AOFAS scores regarding the use of lagged syndesmotic fixation, because it was not ultimately our attempt to necessarily coerce our readers into necessarily adopting this technique. A host of reasons can cause stiffness and confound the outcomes after these injuries. The deleterious effects of syndesmotic compression have been theorized but never truly been substantiated. In our cohort, we sought to demonstrate that because all our patients underwent lag syndesmotic compression, according to the historical premise, none should have had a good range of motion. These were certainly not our findings (Fig. 7, Supplemental Videos S1 and S2). In conclusion, although we saw no specific association with the injury mechanism and outcome, we similarly noted no specific detrimental outcome associated with the lagged syndesmotic screw. Our study population was a mixture of patients with good and poor outcomes, despite the use of lagged syndesmotic fixation. The results of our observational investigations could help guide the development of additional randomized controlled trials and prospective studies on appropriate recommendations on tibiofibular syndesmotic stabilization when indicated.
Fig. 7. (A and B) Weightbearing dorsiflexion view of a patient who underwent open reduction and internal fixation of a Weber B bimalleolar equivalent fracture with lag syndesmotic screw fixation.
K.Y. Kwaadu et al. / The Journal of Foot & Ankle Surgery xxx (2015) 1–9
Supplementary Material Supplementary material associated with this article can be found in the online version at www.jfas.org http://dx.doi.org/10.1053/j.jfas. 2014.12.017. References 1. Shibuya N, Davis ML, Jupiter DC. Foot and ankle fractures in the United States: an analysis of the National Trauma Data Bank (2007 to 2011). J Foot Ankle Surg 53:606–608, 2014. ~ a F, Coetzee JC. Ankle syndesmosis injuries. Foot Ankle Clin 11:35–50, 2006. 2. Pen 3. Tornetta P III. Competency of the deltoid ligament in bimalleolar ankle fractures after medial malleolar fixation. J Bone Joint Surg Am 82:843–848, 2000. 4. Ogilvie-Harris DJ, Reed SC, Hedman TP. Disruption of the ankle syndesmosis: biomechanical study of the ligamentous restraints. Arthroscopy 10:558–560, 1994. 5. Hepple S, Guha A. The role of ankle arthroscopy in acute ankle injuries of the athlete. Foot Ankle Clin 18:185–194, 2013. 6. Summers HD, Sinclar MK, Stover MD. A reliable method for intraoperative evaluation of syndesmotic reduction. J Orthop Trauma 27:196–200, 2013. 7. den Daas A, van Zuuren WJ, Pelet S, van Noort A, van den Bekerom MPJ. Flexible stabilization of the distal tibiofibular syndesmosis: clinical and biomechanical considerations: a review of the literature. Strategies Trauma Limb Reconstr 7:123– 129, 2012. 8. Tornetta P III, Spoo JE, Reynolds FA, Lee C. Overtightening of the ankle syndesmosis: is it really possible? J Bone Joint Surg Am 83:489–492, 2001. 9. Kitaoka HB, Alexander IJ, Adelaar RS, Nunley JA, Myerson MS, Sanders M. Clinical rating system for the ankle-hindfoot, midfoot, hallux, and lessers toes. Foot Ankle Int 15:349–353, 1994. 10. Lin CF, Gross MT, Weinhold P. Ankle syndesmosis injuries: anatomy, biomechanics, mechanism of injury, and clinical guidelines for diagnosis and intervention. J Orthop Sports Phys Ther 36:372–384, 2006. 11. Metzler AV, Johnson DL. Dynamically unstable syndesmosis injuries. Orthopedics 36:201–211, 2013. 12. Zalavras C, Thordarson D. Ankle syndesmotic injury. J Am Acad Orthop Surg 15:330–339, 2007. 13. Kwaadu KK, Fleming JJ, Lin D. Management of complex fibular fracture: double plating of fibular fractures. J Foot Ankle Surg, 2014 Oct 29. http://dx.doi.org/ 10.1053/j.jfas.2013.08.002. Epub ahead of print]. 14. AO/OTA fracture and dislocation classificationdlong bones. AO Foundation Website. Available at: https://www.aofoundation.org/Structure/resource/muellerclassification/Pages/mueller-cassification.aspx. Accessed January 15, 2014. 15. Irwin TA, Lien J, Kadakia AR. Posterior malleolus fracture. J Am Acad Orthop Surg 21:32–40, 2013.
9
16. Stufkens SAS, van de Bekerom MPJ, Kerkhoffs GMMJ, Hintermann B, van Dijk CN. Long-term outcome after 1822 operatively treated ankle fractures: a systematic review of the literature. Injury 42:119–127, 2011. 17. Yoshimura I, Naito M, Kanazawa K, Takeyama A, Ida T. Arthroscopic findings in Maisonneuve fractures. J Orthop Sci 13:3–6, 2008. 18. Loren GJ, Ferkel RD. Arthroscopic assessment of occult intra-articular injury in acute ankle fractures. Arthroscopy 18:412–421, 2002. 19. Deol PPS, Cuttica DJ, Smith WB, Berlet GC. Osteochondral lesions of the talus: size, age, and predictors of outcomes. Foot Ankle Clin 18:13–34, 2013. 20. Mologne TS, Ferkel RD. Arthroscopic treatment of osteochondral lesions of the distal tibia. Foot Ankle Int 28:865–872, 2007. 21. Utsugi K, Sasaki H, Hiraoka H, Yashiki M, Mogi H. Intra-articular fibrous tissue formation following ankle fracture: the significance of arthroscopic debridement of fibrous tissue. Arthroscopy 23:89–93, 2007. 22. Sagi HC, Shah AR, Sanders RW. The functional consequence of syndesmotic joint malreduction at a minimum 2-year follow-up. J Orthop Trauma 26:439–443, 2012. 23. Gardner MJ, Demetrakopoulos D, Briggs SM, Helfet DL, Lorich DG. Malreductions of the tibiofibular syndesmosis in ankle fractures. Foot Ankle Int 27:788–792, 2006. 24. Marmor M, Kandemir U, Matityahu A, Jergesen H, McClellan T, Morshed S. A method for detection of lateral malleolar malrotation using conventional fluoroscopy. J Orthop Trauma 27:e281–e284, 2013. 25. Marmor M, Hansen E, Han HK, Buckley J, Matityahu A. Limitations of standard fluoroscopy in detecting rotational malreduction of the syndesmosis in an ankle. Foot Ankle Int 32:616–622, 2011. € tzner A, Wendl K. Intraoperative three26. Franke J, von Recum J, Suda AJ, Gru dimensional imaging in the treatment of acute unstable syndesmotic injuries. J Bone Joint Surg Am 94:1386–1390, 2012. rbert-Davies J, Laflamme GY, Leduc S. CT scan assessment of the 27. Nault ML, He syndesmosis: a new reproducible method. J Orthop Trauma 27:638–641, 2013. 28. Grenier S, Benoit B, Rouleau DM, Leduc S, Laflamme GY, Liew A. APTF: anteroposterior tibiofibular ratio, a new reliable measure to assess syndesmotic reduction. J Orthop Trauma 27:207–211, 2013. 29. Ebraheim NA, Lu J, Yang H, Mekhail AO, Yeasting RA. Radiographic and CT evaluation of tibiofibular syndesmotic diastasis: a cadaver study. Foot Ankle Int 18:693–698, 1997. 30. Miller AN, Carroll EA, Parker RJ, Boraiah S, Helfet DL, Lorich DG. Direct visualization for syndesmotic stabilization of ankle fractures. Foot Ankle Int 30:419–426, 2009. 31. Nimick CJ, Collman DR, Lagaay P. Fixation orientation in ankle fractures with syndesmosis injury. J Foot Ankle Surg 52:315–318, 2013. 32. Darwish HH, Glisson RR, DeOrio JK. Compression screw fixation of the syndesmosis. Foot Ankle Int 33:893–899, 2012. 33. Ibrahim T, Beiri A, Azzabi M, Best AJ, Taylor GJ, Menon DK. Reliability and validity of the subjective component of the American Orthopaedic Foot and Ankle Society clinical rating scales. J Foot Ankle Surg 46:65–74, 2007.
Contents lists available at ScienceDirect
The Journal of Foot & Ankle Surgery journal homepage: www.jfas.org
Original Research
Lagged Syndesmotic Fixation: Our Clinical Experience Kwasi Yiadom Kwaadu, DPM 1, Justin James Fleming, DPM, FACFAS 2, 3, Trudy Salmon, DPM 4 1
Assistant Professor, Temple University School of Podiatric Medicine, Philadelphia, PA Fellowship Director, Philadelphia Foot and Ankle Fellowship, The Muscle, Bone, and Joint Center, Philadelphia, PA 3 Podiatric Residency Director, Aria Health Systems, Philadelphia, PA 4 Postgraduate Year-2 Resident, Aria Health Systems, Philadelphia, PA 2
a r t i c l e i n f o
a b s t r a c t
Level of Clinical Evidence: 4
Ankle fractures are very common, and although algorithms are in place for osseous management, consensus has not been reached regarding treatment of associated ligamentous injuries. Although tibiofibular syndesmotic stabilization can be done using different forms of fixation, the biomedical literature has long emphasized the risk of long-term restriction of ankle mobility with the use of lagged transfixation. However, when reduction cannot be maintained with positional fixation, we found that lagging the syndesmotic screw helped to maintain the reduction without causing functional restriction. In this report, we describe our experience with patients who had undergone lagged tibiofibular transfixation and were available for short- to intermediate-term follow-up to assess ankle function. A total of 31 patients (32.63% of 95 consecutive patients) were available at a mean of 34.87 (range 18 to 52) months to complete the American Orthopedic Foot and Ankle Society ankle-hindfoot questionnaire. The mean score was 88.38 (range 42 to 100) points at a mean follow-up interval of 34.87 (range 18 to 52) months. Of 31 patients, 19 had an AOFAS score of 90 points, 9 an AOFAS score of 80 to 89 points, 2 an AOFAS score of 60 to 69 points, and 1 an AOFAS score of <60 points. Because all syndesmotic screws were placed using the lag technique, unrestricted motion compared with the uninjured limb was used as the endpoint. All subjects had unrestricted motion compared with the uninjured limb, refuting the assertion that lagged syndesmotic screw fixation confers more restriction in ankle kinematics than positional syndesmotic fixation. Ó 2015 by the American College of Foot and Ankle Surgeons. All rights reserved.
Keywords: ankle fracture kinematics lag screw fixation ligamentous instability malleolus syndesmotic fixation tibiofibular syndesmosis
According to the U.S. National Trauma Center Database (2007 to 2011), ankle fractures accounted for 55.67% of all foot and ankle fractures, and 24.38% of these involved open fractures (1). The concurrent presence of injury to the deep deltoid and syndesmotic ligaments has been shown to further destabilize the ankle mortise, increasing the incidence and importance of operative reduction (2–4). The incidence of syndesmotic injuries in Weber B and C fractures has been reported to be as high as 66%, and improved functional outcome of the ankle joint after anatomic restoration of the unstable mortise has been elucidated (5). The ankle joint functions in a constrained system that tolerates misalignment and instability poorly, the presence of which results in accelerated degeneration of the joint (6). Ligamentous instability of the ankle mortise is addressed with anatomic reduction and fixation with positional syndesmotic screw fixation. Biomechanical studies have noted the physiologic motion
Financial Disclosure: None reported. Conflict of Interest: None reported. Address correspondence to: Kwasi Yiadom Kwaadu, DPM, Temple University School of Podiatric Medicine, 148 North 8th Street, Philadelphia, PA 19107. E-mail address: [email protected] (K.Y. Kwaadu). Video online only at http://www.jfas.org
present within the syndesmosis, and, as a result, concern for restriction has been pointed to as the prime reason for the use of nonlagged fixation across the syndesmosis (2,7). Despite evidence to the contrary that compression of the syndesmosis is not associated with restriction of ankle motion, the published data have continued to advocate against it (8). We found that in certain instances the reduction achieved with the bone tenaculums could not be maintained with positional fixation intraoperatively. We also found cases of late syndesmotic widening during the postoperative course in which the mortise appeared widened despite no obvious hardware failure. We thus began inserting all screws across the syndesmosis using the lag technique. Based on our prior clinical experience, we found our technique to be associated with the radiographic reduction maintained throughout the postoperative course with no late widening. In addition, we observed no functional loss or subjective complaints in patients who underwent this technique and thus present our outcomes with this technique. We specifically sought to demonstrate that from our clinical experience with this technique. In an effort to objectively evaluate our clinical experience, we reviewed the outcomes of 95 patients on whom we had performed lagged tibiofibular transfixation, and we were able to obtain ankle-related quality of life outcome measurements on 31 (32.63%) of these patients with the American
1067-2516/$ - see front matter Ó 2015 by the American College of Foot and Ankle Surgeons. All rights reserved. http://dx.doi.org/10.1053/j.jfas.2014.12.017
2
K.Y. Kwaadu et al. / The Journal of Foot & Ankle Surgery xxx (2015) 1–9
Fig. 1. Patient population of those initially identified and those who did not meet our search criteria.
Academy of Orthopaedic Surgeons (AOFAS) foot and ankle scoring system (9). Patients and Methods After receiving approval by the institutional review board, one co-author (T.S.), who was blinded to the results but took no part in patient care, performed a thorough medical record review of all consecutive ankle injuries requiring operative syndesmotic stabilization performed from January 2009 through December 2011 by the primary author (J.J.F.). These cases were identified using the Current Procedural Terminology (American Medical Association, Chicago, IL) code 27829, representing the open treatment of distal tibiofibular joint (syndesmosis). A total of 275 operations potentially eligible (for inclusion in our retrospective cohort) in 273 patients were initially identified using this search criterion. Our inclusion criteria were unstable syndesmotic injuries with or without operative fractures, syndesmotic stabilization only with screws placed using the lag technique, age 18 years, the ability to provide consent, closed injuries, and patient availability for evaluation using the AOFAS hindfoot questionnaire. Unstable syndesmotic injuries were defined as a tibiofibular clear space >6 mm and a medial clear space >5 mm (10–12). The exclusion criteria included any concurrent injuries beyond the identified ankle injury with the exception of osteochondral lesions identified intraoperatively, open fractures, syndesmotic screws placed using a nonlagged technique, any stabilization performed without screws, patients aged <18 years, patients with documented neuropathy, patients who were nonambulatory, patients lost to follow-up, patients with previous operatively or nonoperatively treated ankle fractures unrelated to the acute or index injury in question, and patients with previous complaints of ankle pain or instability as reported in the history and physical examination findings. Of the initial 275 operations, we excluded 12 (4.4%) patients who had sustained severe crushing highenergy injuries, 16 pediatric fractures (5.8%) in 16 patients, 7 fractures (2.5%) with concomitant fractures to the ipsilateral limb in 7 patients, 5 patients (1.8%) with 5 previous ankle fractures to the ipsilateral limb, and 10 open fractures (3.6%) in 10 patients (13). Four patients (1.4%) requiring 4 operations on the ipsilateral limb for concurrent injuries not involving the ankle joint were also excluded. Two patients requiring reoperation on postoperative day 0 and day 19 because of acute hardware failure, the first resulting from a fall and the second from gross ambulatory noncompliance, were included. These 2 patients, who had undergone 2 operations on the same limb, met our inclusion criteria, resulting in 97 operations in 95 patients. Thus, the total number of evaluated ankles equaled the same number of patients. An additional 124 patients (45%) were lost to follow-up, because they no longer resided at the addresses they had provided, the new residents had no recollection of the individual or how to reach them, and the telephone numbers provided had either been disconnected or assigned to another individual with no knowledge of the patient. The 97 operations (35.3%) in 95 patients of the initial 275 cases were available for medical record review and were included in the present study (Fig. 1). Finally, complications were defined as unplanned surgical intervention after the definitive open reduction and internal fixation.
Repair of the syndesmotic rupture was carried out in standard fashion. The surgical approach involved a lateral incision directly over the distal fibular and one over the medial malleolus when indicated (Fig. 2). In the presence of a concurrent operative posterior malleolar fracture, the standard lateral incision was moved posteriorly and placed halfway between the posterior border of the fibular and lateral border of the Achilles tendon to facilitate identification of the posterior malleolar fracture (Fig. 3). A posterior plafond fracture that extended medially into a medial malleolar fracture was approached posteromedially with a curvilinear J-shaped incision just posterior to the medial malleolus. Syndesmotic stabilization was performed with fully threaded cortical screws placed using the lag technique to 1 full turn above 2-finger tightness until the mortise was symmetric on fluoroscopy (Fig. 4). All intraoperatively identified osteochondral lesions were microfractured with a microfracture awl. None were >15 mm in diameter. For the purposes of this investigation, we contacted 31 (32.63%) patients to invite them to complete the American Orthopedic Foot and Ankle Society ankle-hindfoot questionnaire. We sought to evaluate the patients from a long-term functional perspective to determine the presence of any impairment that could be attributed to this technique. Of these 95 patients, 31 (32.63%) were available for evaluation using the AOFAS hindfoot clinical rating system and questionnaire. The remaining 64 patients (67.4%) had relocated and could not logistically participate in this portion of the evaluation. We prioritized the patient subjective reports, focusing on the uninjured contralateral limb as our control and previously described anatomic radiographic parameters (10–12). Postoperative management consisted of immobilization in a well-padded posterior splint with the ankle in neutral alignment. Range of motion exercises were begun when the wounds had “sealed” and the sutures and/or staples had been removed. Serial radiographs were obtained at weeks 2, 6, 10, 16, and 20 postoperatively unless the patient had been discharged from the practice before this time and then bimonthly until discharge if the patient required additional follow-up examinations. Progressive protected weightbearing was initiated when both radiographic and clinical union were present, as demonstrated by the absence of pain, edema, or erythema at the fracture sites.
Results The mean follow-up period for the 95 patients was 18 (range 10 to 46) months. Of the 95 patients, 55 were male, with a mean age of 49.58 (range 19 to 84) years, and 40 were female, with a mean age of 46.1 (range 19 to 81) years. Of our 95 patients, 39 (41%) had bimalleolar equivalent fractures, defined by the presence of an isolated fibular fracture with a medial clear space >5 mm, and 19 (20%) had trimalleolar equivalent fractures, defined by the presence of an isolated fibular and posterior malleolar fracture with a medial clear space >5 mm and without fracture of the medial malleolus. Twenty-six patients (27.4%) had trimalleolar
K.Y. Kwaadu et al. / The Journal of Foot & Ankle Surgery xxx (2015) 1–9
Fig. 2. Standard lateral incisional approach for anatomic restoration of the fractured fibula.
fractures, 5 (5.3%) had bimalleolar fractures, and 3 (3.2%) had Maisonneuve fractures, 1 of whom also presented with an operative posterior plafond fracture and a deltoid injury resembling a trimalleolar equivalent. No patient had bilateral injuries. Of the 95 patients, 46 (48.4%) had right-sided injuries and 49 (51.6%) left-sided injuries. Of our 95 patients, 7 (7.4%) underwent open reduction and internal fixation of their posterior malleolar fracture, because it involved 25% of the plafond evaluated on the sagittal computed tomography (CT) reconstruction scan. All fractures with a posterior malleolar component underwent CT evaluation. Of the 5 patients with a bimalleolar fractures, 2 (6.5%) were available for the AOFAS questionnaire and reported a score of 90 and 100. The patient with an AOFAS score of 90 complained of episodic weather-related uneasiness but functioned without restrictions. All 5 bimalleolar fractures in this AOFAS group had resulted from a lowenergy mechanical fall. At discharge for the 3 patients with a bimalleolar fracture who were not available for the AOFAS questionnaire, 2 had reported no complaints whatsoever. The third, however, had reported subjective and objective stiffness and had presented for intra-articular cortisone ankle injections at 8, 10, and 17 months postoperatively that ultimately resulted in symptomatic relief.
Fig. 3. Marked incision for the posterolateral approach in the presence of an operative posterior malleolar fracture.
3
Of the 39 patients with a bimalleolar equivalent fracture (41%), 11 (35.5%) were available for the AOFAS questionnaire. Their demographics are listed in the Table. Of this AOFAS subgroup, 1 fracture had resulted from an unknown mechanism. This patient recorded an AOFAS score of 42 at the 43-month follow-up visit. She was subsequently diagnosed with multiple sclerosis, which progressed during her postoperative course. Four fractures (36.4%) in this AOFAS group were high-energy injuries and six (45.4%) were low-energy rotational injuries. Only 1 of the 4 patients (25%) with a high-energy injury reported an AOFAS score of 90. In contrast, only 1 of the 6 patients (16.7%) with a lowenergy rotational injury reported an AOFAS score of 90. Of the remaining 28 patients (71.8%) with a bimalleolar equivalent fracture who were not available for the AOFAS questionnaire, 5 (17.8%) had sustained high-energy injuries involving falls 6 ft and 23 (82.1%) had sustained low-energy rotational injuries. The 5 patients with highenergy injuries who were not available for the AOFAS questionnaire had been discharged at a mean of 6.4 (range 2.5 to 8) months with clinical motion comparable to that of the uninjured contralateral limb and no subjective complaints. The 23 patients with a low-energy injury who were not available for the AOFAS questionnaire were discharged at a mean of 7.69 (range 2 to 35) months. Of these same 23 patients, 1 required a reoperation 3 weeks after the index procedure that involved reinsertion of the syndesmotic screw, which had moved retrograde as a result of early weightbearing. This patient proceeded with an uneventful convalescence otherwise. Another required hardware removal at 19 months because of prominence and irritation of the lateral plate. A third patient received an intra-articular cortisone ankle injection that resulted in symptomatic relief. One patient, however, complained of stiffness on discharge with less motion clinically than available in the uninjured contralateral limb. Of the initial 20 patients (21% of the 95 patients) with a trimalleolar equivalent fracture (19 true trimalleolar equivalent fractures by our definition and the third Maisonneuve with a mortise that presented as a trimalleolar equivalent), 8 (25.8%) were available for the AOFAS questionnaire. The mean follow-up period for these 8 patients was 39 (range 31 to 52) months. The mean AOFAS score for these 8 patients was 88.87 (range 64 to 100). Of these 8 patients, 3 had sustained a high-energy injury and 5 a low-energy rotational injury. Despite their scores, the subjective reports early in the postoperative course were less optimistic. Of the low-energy group, 1 patient (20%), who had undergone operative repair of a concurrent posterior malleolar fracture, developed stiffness and pain 6 months postoperatively that was unresponsive to an intra-articular cortisone injection. However, the patient experienced some symptomatic relief after hardware removal at 17 months postoperatively. Another who had concurrently sustained an operative posterior malleolar fracture complained of stiffness at 8 months, but it had resolved at their latest follow-up examination at 42 months. The remaining 3 patients (60%) who had sustained low-energy injuries and were available for the AOFAS questionnaire had had benign and good outcomes as defined by our criteria. In the high-energy group, 2 (66%) of the 3 patients were available for our AOFAS questionnaire. One recorded a score of 85 and complained of episodic weather-related stiffness that resolved with increased activity. The other recorded a score of 89 at the latest follow-up visit at 32 months after undergoing hardware removal and ankle arthroscopy at 7 months postoperatively (Fig. 5). Of the remaining 12 patients with a trimalleolar equivalent injury who were not in the AOFAS group, 5 (41.7%) had sustained highenergy injuries and 7 (58.3%) low-energy rotational injuries. In this subgroup of non-AOFAS high-energy fractures, 2 patients (20%) presented with residual complaints at their latest follow-up visit at 26 and 32 months, respectively. The former had sustained a concurrent operative posterior malleolar fracture after he was thrown off a motorcycle. He had been grossly noncompliant during the
4
K.Y. Kwaadu et al. / The Journal of Foot & Ankle Surgery xxx (2015) 1–9
Fig. 4. The sequential placement of a fully threaded lagged syndesmotic screw demonstrated on the anteroposterior fluoroscopic projection in which both cortices of the fibula were overdrilled. The lateral and medial tibial cortices were subsequently underdrilled. Only the lateral cortex in this instance was underdrilled because of the blocking effect of the medial malleolar screws. The syndesmotic screw was inserted to a 2-finger tightness and up to a full turn beyond if the mortise was still asymmetric.
postoperative course and presented at month 5 for his first postoperative follow-up visit. This patient subsequently developed radiographic changes of post-traumatic arthritis found 14 months postoperatively. The second patient continued to complain of residual stiffness at 10 months postoperatively. In the non-AOFAS lowenergy subgroup, 1 patient (14.3%) with a concurrent operative posterior malleolar fracture presented with clinical stiffness at 18 months postoperatively. Another was diagnosed with ankle synovitis and received intra-articular cortisone injections that offered symptomatic relief at 8 and 30 months postoperatively. An additional 2 patients (28.6%) reported weather-related changes that resolved with activity; however, on examination, both had hindfoot and ankle motion comparable to that of the uninjured contralateral limb. Of the 31 patients available for the AOFAS hindfoot questionnaire evaluation,18 (58.1%) were male and 13 (41.9%) were female. Of the 31 fractures, 16 (51.6%) were right-sided and 15 (48.4%) left-sided. Additional demographic data for this group are listed in the Table. Of
the 31 patients in the AOFAS group, 2 (6.4%) had had operative posterior malleolar fractures, both of which were associated with lowenergy injury mechanisms. Of the 64 patients in the non-AOFAS group, 2 (3.1%) had had operative posterior malleolar fractures, 1 each in the high-energy and low-energy groups. Of the initial 26 patients with a trimalleolar fracture, 8 (25.8%) were available for the AOFAS questionnaire. The mean score was 90.5 (range 60 to 100) for these 8 patients, 6 (75%) of whom had a score of 90. These 8 patients had all sustained low-energy rotational injuries, none of whom had sustained an operative posterior malleolar fracture. One patient (12.5%), however, had had a concurrent osteochondral lesion that was visualized intraoperatively and complained of stiffness at 7 months postoperatively. However, at the 38-month follow-up visit, the patient complained of mild and episodic pain with no restrictions otherwise. The demographics of the remaining patients can be reviewed in the Table. Of the 18 patients in the non-AOFAS trimalleolar group, 15 (83.3%) had had low-energy injuries and 1 (5.5%) had had an injury due to an unknown mechanism. This patient complained of pain
Table Demographic information for 31 stabilizations in 31 patients available to complete AOFAS questionnaire Age (y)
Mechanism
AO Fracture (14)
Fracture Details
Follow-Up (mo)
M M M
43 65 31
Rotational Rotational Rotational
44C 44B 44B
Right Maisonneuve fracture, trimalleolar equivalent Right bimalleolar equivalent fracture Left trimalleolar fracture
41 28 38
AOFAS Score 83 90 96
F M M
57 37 29
Thrown from horse Rotational Rotational
44B 44B 44C
Left trimalleolar equivalent fracture Left trimalleolar equivalent fracture Left trimalleolar fracture
38 52 32
85 64 60
F M
40 27
Rotational Fall >6 feet
44B 44C
Right trimalleolar fracture Left bimalleolar equivalent fracture
38 43
96 81
M
34
Rotational
44B
Left trimalleolar equivalent fracture
42
100
F M
37 64
Rotational Motor vehicle accident
44B 44B
Right bimalleolar fracture Right bimalleolar equivalent fracture
39 32
90 90
M
42
Rotational
44B
Left bimalleolar equivalent fracture
28
81
F
46
Fall >6 feet
44B
Left bimalleolar equivalent fracture
38
87
M M F F M M M F
32 62 33 49 64 46 62 63
Rotational Rotational Rotational Rotational Rotational Rotational Rotational Fall >6 feet
44B 44C High Ankle Sprain 44B 44B 44B 44B 44B
Right trimalleolar equivalent fracture Right bimalleolar equivalent fracture Left isolated syndesmotic injury Left bimalleolar equivalent fracture Right trimalleolar fracture Left trimalleolar equivalent fracture Right trimalleolar equivalent fracture Left bimalleolar equivalent
41 22 24 46 34 35 31 35
90 100 90 100 88 100 100 81
F M F
56 84 53
Rotational Rotational Rotational
44B 44B 44B
Right trimalleolar fracture Left trimalleolar fracture Right bimalleolar fracture
37 18 26
90 94 100
M M F
52 56 49
Assault Motor vehicle accident Rotational
44C 44B 44B
Right maisonneuve fracture Right trimalleolar equivalent fracture Left bimalleolar equivalent fracture
23 32 48
80 89 93
F F F M
47 35 55 51
Unknown Rotational Rotational Rotational
44B 44C 44A 44B
Right bimalleolar equivalent fracture Left trimalleolar fracture Right bimalleolar equivalent fracture Right trimalleolar fracture
43 36 32 29
42 100 100 100
Comments Hardware removal at 17 months postoperative Minimal aches with swelling Crepitus during postoperative month 7 Osteochondral injury Occasional stiffness at 38 months postoperative Occasional pain and stiffness on uneven terrain Severe pain, pain in toes, abnormal gait Constant but moderate pain Moderate restriction Occasional pain Pain with prolonged activity Cannot run at same pace Stiffness in ankle dorsiflexion during postoperative month 8; resolved at latest follow-up None Postoperative ankle arthritis and joint effusion at 1 year postoperative resolved with injection Episodic pain at 32 months postoperative Noncompliance Good range of motion at 4 months postoperative Syndesmotic ossification at 13 months postoperative Moderate pain, edema, and stiffness with running Ambulates with brace Edema and episodic fatigue Mild occasional pain. Removal of 1 of 2 syndesmotic screws at 5 weeks postoperative None None None None Excellent range of motion Stiffness in ankle dorsiflexion in postoperative week 2, resolved to excellent motion in postoperative month 2 None None Nondisplaced ipsilateral calcaneus fracture following a fall during postoperative month 5 None Retrograding syndesmotic screw removed during postoperative month 7 Radiolucency surrounding threads of syndesmotic screws during postoperative month 8 None None Delayed union at 2.5 months postoperative, with screw loosening None
K.Y. Kwaadu et al. / The Journal of Foot & Ankle Surgery xxx (2015) 1–9
Sex
Abbreviations: AOFAS, American Orthopaedic Foot and Ankle Society; F, female; M, male.
5
6
K.Y. Kwaadu et al. / The Journal of Foot & Ankle Surgery xxx (2015) 1–9
and stiffness on discharge at 6 months. The remaining 2 patients (11.1%) in the non-AOFAS trimalleolar fracture group had sustained vehicularmodulated injuries. Of these 18 patients, 3 (16.6%) had had an operative posterior malleolar fracture, none in the high-energy injury group. Of the 3 with an operative posterior malleolar fracture, 1 presented with residual stiffness and restriction of motion on discharge. Another patient (6.7%) in this low-energy, non-AOFAS trimalleolar fracture group sustained an osteochondral lesion. This patient continued to complain of pain at 1 year postoperatively and had clinically documented restriction. The third reported no complaints. The mean follow-up period for the trimalleolar non-AOFAS group was 11.13 (range 2.5 to 34) months. The only deep venous thrombus in the entire study population occurred in this group. Of the initial 95 patients, 3 (3.1%) had had a Maisonneuve injury, 1 of whom was described in the trimalleolar equivalent group, because the patient had also had an operative posterior malleolar fracture. Of the remaining 2 patients, 1 was available for evaluation with the AOFAS questionnaire. He recorded a score of 80 at 23 months and had moderate restriction demonstrated clinically compared with that of the uninjured limb. At the latest follow-up visit at 20 months, the second of these 2 patients presented with no subjective complaints but had documented restriction in range of motion. Of the initial 95 patients, 3 patients (3.1%) had had an isolated syndesmotic injury, with only 1 (3.2%) represented in the entire AOFAS questionnaire group. This patient had sustained a rotational injury and at the latest follow-up visit had complained of incisional pain. Another physician had removed his hardware 13 months after the index procedure. Of the remaining 2 patients, 1 had had completely normal examination findings at 11 months postoperatively but had sustained a fall and began complaining of pain thereafter. That patient underwent complete hardware removal at 22 months postoperatively and at the latest follow-up visit at 24 months reported no complaints. The final patient underwent syndesmotic stabilization and at the last follow-up visit at 11 months had presented with no symptoms and normal physical examination findings, with clinical motion comparable to that of the uninjured contralateral limb. Two of the 95 patients (2.1%) required acute reoperation and were included in the present study. The first reoperation occurred on postoperative day 0 after the patient had fallen on ice, bending the syndesmotic screws. The second underwent reoperation on postoperative day 19 because of frank hardware failure due to gross ambulatory noncompliance. A total of 7 patients (7.4%) underwent hardware removal. One patient underwent hardware removal 6 weeks after the initial stabilization with 2 syndesmotic screws. Radiographs at 3 to 5 weeks postoperatively demonstrated a distal syndesmotic screw moving progressively retrograde, with the proximal syndesmotic screw remaining intact along an anatomic mortise. At the latest follow-up visit, this patient had an AOFAS score of 100. Another patient, not in the AOFAS group, underwent removal of a lateral plate 4 months after the index procedure. The remaining 5 patients required hardware removal well after the postoperative course at a mean of 14.2 (range 7 to 22) months. None of our patients developed complex regional pain syndrome or deep infections. One patient experienced a deep venous thrombotic episode. Also, 1 patient experienced delayed wound healing as a result of a superficial infection that had resolved completely with local care and oral antibiosis at 28 days. Discussion Many factors contribute to ankle stiffness and restricted motion following injury. Tornetta et al (8) compared ankle range of motion in open chain before and after placement of a 4.5-mm lag screw across the syndesmosis in order to assess ankle restriction associated with
syndesmotic screw placement with the ankle in dorsiflexion. Despite their findings that ankle dorsiflexion was not restricted with syndesmotic compression, the investigators still discouraged the use of lagged fixation across the syndesmosis. The presence of posterior malleolar fractures has been associated with poorer outcomes (15). Stufkens et al (16) demonstrated that only 58% of ankle fractures with posterior malleolar fractures had good to excellent outcomes 4 years after injury. Furthermore, the retromalleolar approach, involving dissection and retraction of the flexor hallucis longus, can result in an indeterminable amount of scar tissue formation during convalescence that can result in ankle stiffness. The ankle capsule is torn during the rotational mechanism of the injury and at times with an iatrogenic insult. The healing of capsule occurs under a reparative pathway that involves replacement with scar tissue, which histologically lacks the elasticity of the native tissue. The generation and abundance of this inelastic scar tissue is under gene expression and individualized (Brigido S, personal communication, Greater Penn Education Foundation National Fellows Meeting, Pittsburgh, PA, May 17, 2013). It is not unreasonable to believe that this variability can confer undesired stiffness. Fractures and syndesmotic injuries occur in closed chain. Under these circumstances, the plafond internally rotates and torques on the fully loaded trochlear of the talus. The incidence of cartilaginous injury, even if macroscopically invisible, is likely greater than reported (Fig. 6). Hepple and Guha (5) reported on the use of arthroscopy and reported an incidence of intra-articular injury of 60% to 75%. Yoshimura et al (17) reported chondral injury in all 4 of their patients. Of these 4 patients, 1 presented with partial thickness fissuring of the cartilage of <50% of its thickness, and 3 presented with full chondral injury down to the subchondral bone. Loren and Ferkel (18) similarly reported a 63% rate of traumatic articular injuries in their cohort, 19 of which were located on the talus and 11 on the tibia. Also, the relative unpredictability of chondral healing of the talus has been elucidated (19). However the report by Mologne and Loren (20) of chondral tibial lesions points to another interesting confounder in outcomes, because tibial lesions have been notoriously associated with poor outcomes. Even beyond the chondral injury itself, arthrofibrosis native to the reparative process can similarly worsen ankle stiffness (21). Syndesmotic malreduction is another variable that can influence the outcome of these ankle injuries. Of their 68 patients, Sagi et al (22) reported that of the 13 who underwent direct open visualization and the 55 who underwent closed and indirect repair of their syndesmotic injury, 2 and 24, respectively, presented with malreduction on CT visualization. They also found that patients with malreduction presented with poor functional outcomes and further recommended CT evaluation of the contralateral limb and the postoperative reduction to help improve the outcomes. Gardner et al (23) reported a syndesmotic malreduction rate of 52% when subsequently evaluated using CT. Despite our best efforts, the 2-dimensional radiographic evaluation of the syndesmosis has had little bearing on the true anatomic reduction because subtle sagittal plane malreduction and, most difficult to evaluate, coronal plane rotation of the reduction can routinely occur (22,24–30). In cases in which the transmalleolar axis is not appreciated, the syndesmosis can also be malreduced (31). Darwish et al (32) applied a relatively novel approach to the effects of compression across the syndesmosis. The study investigators transected the syndesmotic ligaments and inserted a pressure sensor within the incisura. When the bone tenaculums were then placed along the transmalleolar axis and clamped to simulate reduction, the pressure monitor registered 61 N. After placement of a positional 3.5-mm cortical screw and release of the bone tenaculum, the pressure sensor reading decreased to 23 N. In contrast, the 3.5- and 4.5-mm cortical lag screws placed maintained the compressive effect initially achieved. The physiologic implications of this approach are
K.Y. Kwaadu et al. / The Journal of Foot & Ankle Surgery xxx (2015) 1–9
7
Fig. 5. (A) Right bimalleolar equivalent ankle fracture after a motor vehicle accident. (B) Axial computed tomography scan of a bimalleolar equivalent ankle fracture demonstrating a large posterior malleolar fracture with small intra-articular fracture fragments. (C) Sagittal reconstruction of the posterior malleolar fracture with impaction slightly anterior to the concomitant posterior malleolar fracture. (D) Sagittal reconstruction of a bimalleolar equivalent ankle fracture with fibular comminution and impaction. (E) Postoperative anteroposterior and lateral radiographs of a bimalleolar equivalent ankle fracture with double plating of the comminuted fibular fracture, open reduction with internal fixation of a posterior malleolar fracture, and lagged syndesmotic fixation. (F) Eight-month postoperative anteroposterior and lateral radiographs with second-look arthroscopy of a bimalleolar equivalent ankle fracture after a motor vehicle collision that underwent surgical repair with evidence of significant arthrofibrosis and residual chondral injury.
8
K.Y. Kwaadu et al. / The Journal of Foot & Ankle Surgery xxx (2015) 1–9
Fig. 6. Weber B fracture with hemarthrosis and chondral delamination along the entire lateral shoulder of the talus.
theoretical; however, the crux of their investigation demonstrated that the positional screw was unable to maintain even one third of the initial compression achieved by the bone tenaculum (1). In our observational study, we found that lag screw fixation of the tibiofibular syndesmosis did not appear to adversely influence the outcome of ankle fracture repair. Of the 31 patients available for longterm follow-up, the mean AOFAS score was 88.38 13.2 (range 42 to 100) (Table). Our study had multiple weaknesses. We performed no null hypothesis, because we were unable to evaluate a significant percentage of our initial population. Furthermore, we included no control group to compare the effects of positional transfixation. We did not investigate the effects of stainless versus titanium. However, the published data elucidate no statistically significant difference in this regard (12). No recent follow-up radiographs were available for our patient population who completed our AOFAS questionnaire to ascertain the fate of the ankle joint. Furthermore, the reliability of the objective
components of the AOFAS scoring system itself is not been validated or standardized (33). However, because it has been used for many previous studies, we also used it to allow for relative comparisons. Nonetheless, the rationale behind this thought and approach is fundamentally flawed. Also, although we referred to cases of clinically good motion, we included no actual measurements in the study. We were unable to objectively standardize the amount of force needed to reproduce a measurable and quantifiable amount of dorsiflexion; thus, we focused on the clinical, macroscopic, and patients’ subjective accounts regarding their perceived restrictions and motion. With this assumption, we also sought to prioritize patient subjective reports because the objective absence of 10 of dorsiflexion might be normal for 1 patient and not for another patient. Despite our best attempts, a comparison of motion in the contralateral uninjured limb was based on the educated assumption that the involved limb, before injury, would have had the exact same kinemantics, an educated guess at best, but a guess nonetheless. Most importantly, a significant portion of our population was lost to follow-up. We were not unaware of the outcomes of the study, which could potentially lend significant bias. We performed no comparison or additional statistical analysis of the AOFAS scores regarding the use of lagged syndesmotic fixation, because it was not ultimately our attempt to necessarily coerce our readers into necessarily adopting this technique. A host of reasons can cause stiffness and confound the outcomes after these injuries. The deleterious effects of syndesmotic compression have been theorized but never truly been substantiated. In our cohort, we sought to demonstrate that because all our patients underwent lag syndesmotic compression, according to the historical premise, none should have had a good range of motion. These were certainly not our findings (Fig. 7, Supplemental Videos S1 and S2). In conclusion, although we saw no specific association with the injury mechanism and outcome, we similarly noted no specific detrimental outcome associated with the lagged syndesmotic screw. Our study population was a mixture of patients with good and poor outcomes, despite the use of lagged syndesmotic fixation. The results of our observational investigations could help guide the development of additional randomized controlled trials and prospective studies on appropriate recommendations on tibiofibular syndesmotic stabilization when indicated.
Fig. 7. (A and B) Weightbearing dorsiflexion view of a patient who underwent open reduction and internal fixation of a Weber B bimalleolar equivalent fracture with lag syndesmotic screw fixation.
K.Y. Kwaadu et al. / The Journal of Foot & Ankle Surgery xxx (2015) 1–9
Supplementary Material Supplementary material associated with this article can be found in the online version at www.jfas.org http://dx.doi.org/10.1053/j.jfas. 2014.12.017. References 1. Shibuya N, Davis ML, Jupiter DC. Foot and ankle fractures in the United States: an analysis of the National Trauma Data Bank (2007 to 2011). J Foot Ankle Surg 53:606–608, 2014. ~ a F, Coetzee JC. Ankle syndesmosis injuries. Foot Ankle Clin 11:35–50, 2006. 2. Pen 3. Tornetta P III. Competency of the deltoid ligament in bimalleolar ankle fractures after medial malleolar fixation. J Bone Joint Surg Am 82:843–848, 2000. 4. Ogilvie-Harris DJ, Reed SC, Hedman TP. Disruption of the ankle syndesmosis: biomechanical study of the ligamentous restraints. Arthroscopy 10:558–560, 1994. 5. Hepple S, Guha A. The role of ankle arthroscopy in acute ankle injuries of the athlete. Foot Ankle Clin 18:185–194, 2013. 6. Summers HD, Sinclar MK, Stover MD. A reliable method for intraoperative evaluation of syndesmotic reduction. J Orthop Trauma 27:196–200, 2013. 7. den Daas A, van Zuuren WJ, Pelet S, van Noort A, van den Bekerom MPJ. Flexible stabilization of the distal tibiofibular syndesmosis: clinical and biomechanical considerations: a review of the literature. Strategies Trauma Limb Reconstr 7:123– 129, 2012. 8. Tornetta P III, Spoo JE, Reynolds FA, Lee C. Overtightening of the ankle syndesmosis: is it really possible? J Bone Joint Surg Am 83:489–492, 2001. 9. Kitaoka HB, Alexander IJ, Adelaar RS, Nunley JA, Myerson MS, Sanders M. Clinical rating system for the ankle-hindfoot, midfoot, hallux, and lessers toes. Foot Ankle Int 15:349–353, 1994. 10. Lin CF, Gross MT, Weinhold P. Ankle syndesmosis injuries: anatomy, biomechanics, mechanism of injury, and clinical guidelines for diagnosis and intervention. J Orthop Sports Phys Ther 36:372–384, 2006. 11. Metzler AV, Johnson DL. Dynamically unstable syndesmosis injuries. Orthopedics 36:201–211, 2013. 12. Zalavras C, Thordarson D. Ankle syndesmotic injury. J Am Acad Orthop Surg 15:330–339, 2007. 13. Kwaadu KK, Fleming JJ, Lin D. Management of complex fibular fracture: double plating of fibular fractures. J Foot Ankle Surg, 2014 Oct 29. http://dx.doi.org/ 10.1053/j.jfas.2013.08.002. Epub ahead of print]. 14. AO/OTA fracture and dislocation classificationdlong bones. AO Foundation Website. Available at: https://www.aofoundation.org/Structure/resource/muellerclassification/Pages/mueller-cassification.aspx. Accessed January 15, 2014. 15. Irwin TA, Lien J, Kadakia AR. Posterior malleolus fracture. J Am Acad Orthop Surg 21:32–40, 2013.
9
16. Stufkens SAS, van de Bekerom MPJ, Kerkhoffs GMMJ, Hintermann B, van Dijk CN. Long-term outcome after 1822 operatively treated ankle fractures: a systematic review of the literature. Injury 42:119–127, 2011. 17. Yoshimura I, Naito M, Kanazawa K, Takeyama A, Ida T. Arthroscopic findings in Maisonneuve fractures. J Orthop Sci 13:3–6, 2008. 18. Loren GJ, Ferkel RD. Arthroscopic assessment of occult intra-articular injury in acute ankle fractures. Arthroscopy 18:412–421, 2002. 19. Deol PPS, Cuttica DJ, Smith WB, Berlet GC. Osteochondral lesions of the talus: size, age, and predictors of outcomes. Foot Ankle Clin 18:13–34, 2013. 20. Mologne TS, Ferkel RD. Arthroscopic treatment of osteochondral lesions of the distal tibia. Foot Ankle Int 28:865–872, 2007. 21. Utsugi K, Sasaki H, Hiraoka H, Yashiki M, Mogi H. Intra-articular fibrous tissue formation following ankle fracture: the significance of arthroscopic debridement of fibrous tissue. Arthroscopy 23:89–93, 2007. 22. Sagi HC, Shah AR, Sanders RW. The functional consequence of syndesmotic joint malreduction at a minimum 2-year follow-up. J Orthop Trauma 26:439–443, 2012. 23. Gardner MJ, Demetrakopoulos D, Briggs SM, Helfet DL, Lorich DG. Malreductions of the tibiofibular syndesmosis in ankle fractures. Foot Ankle Int 27:788–792, 2006. 24. Marmor M, Kandemir U, Matityahu A, Jergesen H, McClellan T, Morshed S. A method for detection of lateral malleolar malrotation using conventional fluoroscopy. J Orthop Trauma 27:e281–e284, 2013. 25. Marmor M, Hansen E, Han HK, Buckley J, Matityahu A. Limitations of standard fluoroscopy in detecting rotational malreduction of the syndesmosis in an ankle. Foot Ankle Int 32:616–622, 2011. € tzner A, Wendl K. Intraoperative three26. Franke J, von Recum J, Suda AJ, Gru dimensional imaging in the treatment of acute unstable syndesmotic injuries. J Bone Joint Surg Am 94:1386–1390, 2012. rbert-Davies J, Laflamme GY, Leduc S. CT scan assessment of the 27. Nault ML, He syndesmosis: a new reproducible method. J Orthop Trauma 27:638–641, 2013. 28. Grenier S, Benoit B, Rouleau DM, Leduc S, Laflamme GY, Liew A. APTF: anteroposterior tibiofibular ratio, a new reliable measure to assess syndesmotic reduction. J Orthop Trauma 27:207–211, 2013. 29. Ebraheim NA, Lu J, Yang H, Mekhail AO, Yeasting RA. Radiographic and CT evaluation of tibiofibular syndesmotic diastasis: a cadaver study. Foot Ankle Int 18:693–698, 1997. 30. Miller AN, Carroll EA, Parker RJ, Boraiah S, Helfet DL, Lorich DG. Direct visualization for syndesmotic stabilization of ankle fractures. Foot Ankle Int 30:419–426, 2009. 31. Nimick CJ, Collman DR, Lagaay P. Fixation orientation in ankle fractures with syndesmosis injury. J Foot Ankle Surg 52:315–318, 2013. 32. Darwish HH, Glisson RR, DeOrio JK. Compression screw fixation of the syndesmosis. Foot Ankle Int 33:893–899, 2012. 33. Ibrahim T, Beiri A, Azzabi M, Best AJ, Taylor GJ, Menon DK. Reliability and validity of the subjective component of the American Orthopaedic Foot and Ankle Society clinical rating scales. J Foot Ankle Surg 46:65–74, 2007.