Original Article With Video Illustration

Concomitant Syndesmotic Instability and Medial Ankle Instability Are Risk Factors for Unsatisfactory Outcomes in Patients With Chronic Ankle Instability Jin Su Kim, M.D., Ph.D., Ki Won Young, M.D., Ph.D., Hun Ki Cho, M.D., Seok Min Lim, M.D., Young Uk Park, M.D., Ph.D., and Kyung Tai Lee, M.D., Ph.D.

Purpose: To determine whether concomitant chronic syndesmotic injury or chronic medial ankle instability is associated with unsatisfactory outcomes in patients with chronic lateral ankle instability (CAI). Methods: We conducted a retrospective review of prospectively collected data from 276 male military patients with CAI who had undergone a modified Broström procedure between May 2007 and March 2010 and had been followed up for a mean of 26.7 months (range, 22 to 41 months). We evaluated clinical outcomes using American Orthopaedic Foot & Ankle Society ankle-hindfoot, visual analog scale, and ankle functional satisfactory scale scores. Possible associations with concomitant instability were reconstructed and investigated using the Pearson correlation coefficient and multivariate logistic regression analysis. Results: Of the 236 patients with satisfactory outcomes, 19 (8%) had medial ankle instability and 13 (6%) had syndesmotic instability; in contrast, of the 40 patients with unsatisfactory outcomes, 14 (35%) had medial ankle instability and 12 (30%) had syndesmotic instability. The mean American Orthopaedic Foot & Ankle Society scores for patients with satisfactory and unsatisfactory outcomes increased from 68.1 to 92.8 (P < .001) and from 65.9 to 76.8 (P < .001), respectively. The mean visual analog scale scores for the groups with satisfactory and unsatisfactory outcomes decreased from 5.8 to 2.1 (P < .001) and from 6.2 to 4.8 (P < .001), respectively. The mean ankle satisfaction score was 27.8 for patients with satisfactory outcomes and 18.7 for those with unsatisfactory outcomes (P < .001). Multivariate logistic regression analysis showed a 4-fold higher risk of dissatisfaction (95% confidence interval [CI], 0.81 to 20.07; P ¼ .0880) with CAI and syndesmotic instability, a 3.8-fold higher risk (95% CI, 0.96 to 15.07; P ¼ .0576) with CAI and medial ankle instability, and an 11.7-fold higher risk (95% CI, 2.60 to 52.70; P ¼ .0014) with CAI and both syndesmotic and medial ankle instability. Conclusions: Chronic syndesmotic instability and chronic medial ankle instability are significantly associated with unsatisfactory outcomes in patients with CAI. Level of Evidence: Level IV, therapeutic case series.

O

perative treatments for chronic lateral ankle instability (CAI) have favorable long-term outcomes.1,2 However, there are a subset of patients who still have disability after undergoing the Broström procedure.3,4 Adverse sequelae occur in 10% to 20% of

From the Surgery of Foot and Ankle, Eulji Medical Center, College of Medicine, Eulji University (J.S.K., K.W.Y., H.K.C., S.M.L.), Seoul; the Department of Orthopedic Surgery, College of Medicine, Ajou University (Y.U.P.), Suwon; and KT Lee’s Orthopedic Hospital (K.T.L.), Seoul, Republic of Korea. The authors report that they have no conflicts of interest in the authorship and publication of this article. Received August 16, 2014; accepted February 19, 2015. Address correspondence to Jin Su Kim, M.D., Ph.D., Surgery of Foot and Ankle, Eulji Medical Center, College of Medicine, Eulji University, 68 Hangeulbiseok-ro, Nowon-Gu, Seoul, Republic of Korea 139-711. E-mail: [email protected] Ó 2015 by the Arthroscopy Association of North America 0749-8063/14701/$36.00 http://dx.doi.org/10.1016/j.arthro.2015.02.021

1548

patients with CAI and are commonly associated with other intra-articular problems, such as osteochondral lesions, bony impingement, syndesmotic ligament injuries, and deltoid ligament injuries, in addition to injuries of the lateral ligamentous complex.3,5,6 Recent studies have reported that the frequency of specific ankle ligament injuries is greater than previously thoughtdin the range of 1% to 20% of cases for syndesmotic injury and 2.8% to 45.3% for the tibionavicular band of the deltoid ligament.7,8 We have often identified combined ankle injuries. Various terms, including “global ankle instability,” “rotational ankle instability,” and “multidirectional ankle instability,” have been used to indicate CAI combined with other ankle instability, such as medial ankle instability and syndesmotic instability.6,9 These multidirectional chronic ankle instabilities (MCAIs) result in severe ankle instability in accordance with the ring theory,10 according to which an ankle with intact ligaments and

Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 31, No 8 (August), 2015: pp 1548-1556

MULTIDIRECTIONAL CHRONIC ANKLE INSTABILITY

1549

bones is stable and acts as an intact ring whereas 2 breaks in the ring make it unstable. Postoperative rehabilitation takes twice as long in patients with MCAIs and has uncertain treatment outcomes.11 Because medial instability and syndesmotic instability can be causes of treatment failure, they need to be identified preoperatively. The accuracy of diagnosis of medial ankle and syndesmotic instabilities should have improved with refinements in arthroscopic exploration and imaging methods. The results of treatment for medial ankle, syndesmotic, or lateral ankle instability in isolation have been reported8,12-15; however, there are few reports of simultaneous treatment of these instabilities.16 We evaluated CAI patients to determine whether concomitant chronic syndesmotic injury or chronic medial ankle instability is associated with unsatisfactory outcomes. We hypothesized that patients with CAI associated with chronic syndesmotic injury or chronic medial ankle instability would have unsatisfactory outcomes.

Methods This retrospective study of prospectively collected data, which received institutional review board approval, included 276 male military patients who underwent operative treatment for CAI between May 2007 and March 2010 at a single institution. Of the 276 patients, 129 had right ankle instability and 150 had left ankle instability (3 had bilateral instability). The mean age was 25.7 years (range, 18 to 55 years). The median duration of postoperative follow-up was 26.7 months (range, 22 to 41 months). Patients with generalized laxity, ankle osteoarthritis, or associated ankle fracture were excluded. All study patients had histories of multiple sprains and repeated bouts of lateral ankle instability symptoms for more than 6 months. All patients underwent a 6-week peroneus longusestrengthening exercise program using a rubber band and proprioceptive singleheel balance training.17 Surgery was recommended for patients who were still found to have mechanical ankle instability after completing the 6-week program. The radiographic mean varus talar tilt angle was 9.4  5.6 with maximum manual varus stress; 206 cases (75%) were checked while patients were awake. Lateral ankle instability was defined as greater than 8.9 of varus tilt or a greater than 3 difference in varus tilt angle compared with the contralateral side.18,19 Figure 1 presents a schematic summary of the preoperative and intraoperative decision plan. Diagnostic Assessment All participants were diagnosed and evaluated by the same orthopaedic foot and ankle specialist (J.S.K.). Diagnosis of chronic syndesmotic instability or chronic medial ankle instability was based on physical

Fig 1. Protocol for managing chronic ankle instability. (MRI, magnetic resonance imaging; OP, operative.)

examination, ankle magnetic resonance imaging (MRI),7 and intraoperative ankle arthroscopic findings. CAI was diagnosed when giving way and pain occurred with direct pressure on the anterior talofibular ligament (ATFL) and the lateral rotary drawer test for assessing lateral ligamentous laxitydin which the ankle is rotated and pulled anteriorly while the medial tibionavicular ligament is fixeddwas positive. (Video 1, available at www.arthroscopyjournal.org, depicts the anterior, lateral, and medial rotary drawer tests.) ATFL injury was identified on the oblique axial plane of the MRI study by assessing the plane of a line from the center of the navicular bone to the center of the talus. Thinning, thickening, irregularity, waviness, and discontinuity of the ATFL were considered abnormal findings. In patients with CAI, a modified Broström operation was performed through a curvilinear incision over the distal fibula after arthroscopic exploration.

1550

J. S. KIM ET AL.

Chronic syndesmotic instability was diagnosed by the presence of a teardrop sign20 on a coronal-plane MRI study (height of tibiofibular recess 10 mm) and arthroscopic findings of a greater than 4-mm diastasis measured by probing the tibiofibular distances with an obturator21 (Fig 2; Video 2, available at www. arthroscopyjournal.org). Syndesmotic instability was considered to be present when there was a diastasis of 4 mm or greater into which an obturator could be inserted, given the report that syndesmotic widening of 2 mm or less does not require syndesmotic screw fixation.22 The pass-through sign (obturator placed within tibiofibular syndesmosis, >4-mm diastasis) is a method that we have developed and have not found in the literature. Accurate assessment of a diastasis cannot be achieved by C-arm or other techniques because this instability is dynamic. Other authors have evaluated diastasis with dynamic sonography preoperatively but not in the operating room. We perform arthroscopic measurement with probes of varying thickness and other devices such as an obturator assembly measuring 2.9 mm (component of arthroscope sheath). Because its tip is blunt, it is safe to place it within a syndesmosis. A 2.9-mm diameter is suitable to measure a diastasis. We initially attempt to place a 2-mm probe within the syndesmosis: If we succeed, we take this as evidence of syndesmotic injury. We then try to place an obturator (2.9 mm) within the syndesmosis and see if it moves. If it does, indicating a diastasis of more than 4 mm, we conclude that the syndesmosis is unstable and fix it with a screw or suture button. Chronic syndesmotic instability was diagnosed in 25 of 279 ankles (9%). Chronic medial ankle instability was diagnosed based on the results of the medial rotary

drawer test, in which the ankle is pulled anteriorly to allow tibionavicular ligament laxity while the lateral ATFL is fixed posteriorly. In addition, talonavicular ligament injury on the oblique axial plane of an MRI study7 and arthroscopic findings of tibionavicular ligament laxity or the drive-through sign (i.e., a 2.9-mmdiameter arthroscope can easily drive through the medial tibiotalar joint) helped establish the diagnosis of medial ankle instability (Fig 3; Video 3, available at www.arthroscopyjournal.org). Hintermann et al.6 classified chronic medial ankle instability as normal (not enough to open the tibiotalar joint >2 mm), moderate (2 to 5 mm, measured by a 2-mm hook, not enough to introduce a 5-mm arthroscope), or severe (>5 mm or introduction of a 5-mm arthroscope). In our study medial ankle instability was defined when there was enough space to introduce a 2.9-mm arthroscope because of tibionavicular laxity without using a traction device. Chronic medial ankle instability was diagnosed in 33 of 279 ankles (12%). Because chronic medial ankle instability (type I deltoid insufficiency) is caused by laxity of the tibionavicular band, which supports limitation of external talar rotation (medial rotary drawer) without coronal tibiotalar stability,23 the tibionavicular band of the deltoid was repaired. Other concomitant lesions such as os subfibulare or osteochondral lesions of the talus (OLTs), anterior bony impingement, peroneal tendon problems, and sinus tarsi pain were also evaluated by radiographs, MRI, and arthroscopic examination. Sinus tarsi syndrome was diagnosed preoperatively by the lidocaine block test24 and confirmed with sinus tarsi arthroscopy as a partial rupture or degeneration of the cervical ligament in the sinus tarsi.

Fig 2. A 21-year-old man was diagnosed with left-sided multidirectional chronic ankle instability based on magnetic resonance imaging and arthroscopic findings. (A) Coronal T2-weighted magnetic resonance image showing syndesmotic injury with teardrop sign (arrow, prominent tibiofibular recess with height 10 mm). (B) We place a 2.9-mm obturator within the tibiofibular diastasis and swing it (Video 3, available at www.arthroscopyjournal.org). If the diastasis is greater than 4 mm, we fix it. (C) Syndesmotic instability has been reconstructed with suture button material. (T, tibia; F, fibula.)

MULTIDIRECTIONAL CHRONIC ANKLE INSTABILITY

1551

Fig 3. A 24-year-old man was diagnosed with medial and lateral chronic ankle instability based on magnetic resonance imaging and arthroscopic findings. (A) The injured tibionavicular ligament is shown as a complete discontinuity (arrow) on a T2-weighted magnetic resonance image. (B, C) Arthroscopic findings showing (B) widening of the medial tibiotalar joint (drive-through sign; i.e., a 2.9-mm arthroscope is easily introduced into the medial joint space through a medial portal) and (C) medial malleolar (MM) to talar (T) ligament laxity (Video 2, available at www.arthroscopyjournal.org). (D) These findings helped establish a diagnosis of left-sided chronic medial ankle instability, and subsequent repair of the deltoid talonavicular band was performed through a medial skin incision along the medial malleolus to the navicular bone with a suture anchor and nonabsorbable sutures. The talonavicular band was shortened and repaired by the pantsover-vest technique.

Surgical Procedures A modified Broström procedure was performed by making a curvilinear 5-cm incision at the lateral malleolar anterior margin. First, the extensor retinaculum was dissected as a sheet together with subcutaneous fat. Second, the anterior capsule, ATFL, and calcaneofibular ligament around the fibula were incised. The distal capsuloligamentous flap was repaired by attaching it with nonabsorbable sutures with the everted-type “pants-over-vest” technique to the fibular side of the ligament remnant. The extensor retinaculum was reinforced on the periosteum over the fibula. We rechecked the stability of the ankle and then performed skin closure. If there was persistent medial ankle instability, tibionavicular ligament repair was also performed. Chronic syndesmotic instability was treated by debridement of the syndesmotic hypertrophied tissue followed by fixation of the tibia and fibula using a suture button (19 cases) or a 5.0-mm cannulated screw (6 cases) placed 1.5 cm above the tibiotalar articular surface at a 30 angle (Fig 3).12,25-27 For chronic medial ankle instability, a skin incision measuring about 4 cm was made from the medial malleolus to the navicular bone, followed by incision of the crural fascia in the same direction as the skin incision to expose the

tibionavicular band. The end of the band was cut off at a point approximately 2 to 3 mm from the medial malleolus. A suture anchor was placed at a 45 angle, and the distal ligament was pulled to repair it. The surrounding tissue was repaired with No. 2-0 nonabsorbable sutures. The anterior crural fascia was pulled posteriorly and sutured, and the subcutaneous tissue was repaired. Rehabilitation The postoperative rehabilitation regimen included wearing a short leg cast for 4 weeks and then, if possible, substituting a removable ankle air brace, which permitted a greater range of motion and weight bearing. This regimen was also used for patients with syndesmotic instability who had undergone suture button fixation. However, patients with syndesmotic instability who were treated with screw fixation were permitted to bear weight at 8 weeks. After the cast had been removed, increased mobility of the joints was promoted by plantar flexion and dorsiflexion for the first week, followed by alphabet writing, 1-leg standing, and wobble-board exercises from week 2. Exercises to strengthen the peroneal tendon and ankle joint muscles were performed from week 3. Jogging and a return to daily activities were permitted after 10 weeks. Patients

1552

J. S. KIM ET AL.

with syndesmotic instability who had undergone screw fixation were allowed to jog at 12 weeks, after removal of the syndesmotic screw in all 6 patients.

Results

the mean AOFAS score had increased significantly from a preoperative score of 67.8 to 90.5, whereas the mean VAS score had decreased significantly from a preoperative score of 5.9 to 2.5 (P < .001). The mean clinical scores were significantly improved both in patients with satisfactory outcomes and in those with unsatisfactory outcomes (Table 2). However, the differences between preoperative AOFAS and VAS scores and those at last follow-up were significantly greater for patients with satisfactory outcomes than for patients with unsatisfactory outcomes (P < .001) (Table 3). The patients’ clinical characteristics are shown in Table 1. Notably, chronic medial ankle instability (n ¼ 14, 35%) and chronic syndesmotic instability (n ¼ 12, 30%) occurred significantly more often in patients with unsatisfactory outcomes (P < .001). The odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for associated pathologies (Table 4). A combination of chronic medial ankle instability and chronic syndesmotic instability was the worst prognostic factor (OR, 11.70; 95% CI, 2.60 to 52.70). The concomitant presence of chronic medial ankle instability was associated with a 4.03-fold higher risk (95% CI, 1.15 to 14.15) of unsatisfactory outcomes compared with patients with satisfactory outcomes; chronic syndesmotic instability, a 3.07-fold higher risk (95% CI, 0.97 to 9.71); and sinus tarsi syndrome, an 8.9-fold higher risk (95% CI, 1.06 to 75.41). Other associated lesions made no significant difference. We defined 41 patients (15.0%) as having MCAI because they had 2 or more types of ankle instability (Fig 4). The mean AOFAS score increased significantly from 65.4 before surgery to 82.4 at last follow-up (P < .001) in the MCAI group and from 68.2 to 89.9 in patients with isolated CAI (P < .001). The mean VAS score decreased significantly between baseline and follow-up in both groups: from 6.0 to 3.6 in the MCAI group and from 5.7 to 2.6 in patients with isolated CAI (Table 5).

Preoperative MRI detected os subfibulare in 63 of 279 ankles (23%), OLT in 33 (12%), anterior bony impingement in 9 (3%), sinus tarsi syndrome in 8 (3%), peroneal tendon injury in 6 (2%), and os subtibiale in 5 (2%). At the last postoperative follow-up,

CAI patients with both chronic medial ankle instability and chronic syndesmotic instability had an

Clinical Assessment Clinical outcomes were evaluated preoperatively and at last follow-up by the American Orthopaedic Foot & Ankle Society (AOFAS) ankle-hindfoot score,28 a visual analog scale (VAS) score, and a specialized ankle functional satisfactory scale score29 designed for ankle instability. The ankle functional satisfactory scale was used to measure patient satisfaction, with a total score of 25 to 30 being excellent; 20 to 24, good; and less than 20, poor. The participants were classified as having satisfactory outcomes (excellent and good) or unsatisfactory outcomes (poor). No significant differences were found between the groups in terms of age, height, weight, and body mass index, whereas there was a significant difference in the follow-up period (Table 1). Participants were also classified as having isolated CAI or MCAI. Relevant patient characteristics and concurrent pathologies are shown, according to outcome group, in Tables 1 and 2. Statistical Assessment We performed a paired-samples Student t test to compare clinical outcomes and Pearson c2 test and Fisher exact test to compare the characteristics of the 2 groups. Multivariate logistic regression analysis was used to assess the probability of each concomitant pathology being associated with outcome. Preoperative and final follow-up clinical scores for patients with isolated CAI and MCAI were compared by Student paired t test. P  .05 was considered significant. SPSS software for Windows (version 16.0; SPSS, Chicago, IL) was used for statistical analyses.

Discussion

Table 1. Patient Characteristics and Demographic Data No. of patients Age, yr Height, cm Weight, kg BMI, kg/m2 Duration of follow-up, mo

All Patients 276 25.7  6.0 172.5  7.7 72.6  11.3 25.0  3.3 26.7  4.6

Satisfactory Outcome 236 (85.5%) 25.9  6.0 172.5  8.0 72.2  11.6 25.41  3.4 25.9  4.9

NOTE. Data are presented as mean  standard deviation unless otherwise indicated. BMI, body mass index. *Paired Student t test. y Significant difference between satisfactory and unsatisfactory outcome groups.

Unsatisfactory Outcome Group 40 (14.5%) 24.7  5.7 171.5  6.9 68.5  9.2 23.8  2.6 28.0  7.3

P Value* .246 .588 .072 .064 .011y

1553

MULTIDIRECTIONAL CHRONIC ANKLE INSTABILITY Table 2. Associated Pathologies in Chronic Lateral Ankle Instability Patients and Clinical Results According to Outcome No. of patients Chronic medial ankle instability: medial ankle ligament repair with anchor suture Chronic syndesmotic instability: syndesmotic fixation with suture button or screw Os subfibulare Osteochondral defect Osteochondral defect size, mm Anterior bony impingement Sinus tarsi syndrome Peroneal tendon injury Os subtibiale D AOFAS D VAS Ankle functional satisfactory scale score

All Patients 276 33 (12%)

Satisfactory Outcome Group 236 (85.5%) 19 (8%)

25 (9%) 63 (23%) 33 (12%) 8.6  2.75 9 (3%) 8 (3%) 6 (2%) 5 (2%) 22.7  11.0 3.4  1.7 26.5  3.9

Unsatisfactory Outcome Group 40 (14.5%) 14 (35%)

P Value* .999

Concomitant Syndesmotic Instability and Medial Ankle Instability Are Risk Factors for Unsatisfactory Outcomes in Patients With Chronic Ankle Instability.

To determine whether concomitant chronic syndesmotic injury or chronic medial ankle instability is associated with unsatisfactory outcomes in patients...
930KB Sizes 1 Downloads 7 Views