Abstract
Background:
Chronic ankle instability with generalized joint hypermobility (GJH) is considered a contraindication for the modified Broström procedure. The most widely accepted definition of GJH is a Beighton score of ≥4 on a 9-point scale. However, it is not clear whether this criterion can be applied to determine the GJH that would lead to a poor outcome after a modified Broström procedure. Some of the previous studies that report unfavorable outcomes do not specify the tests or cutoff scores used to determine the GJH, and, in fact, some of the patients with GJH in these studies had good outcomes.
Hypothesis:
The modified Broström procedure results in satisfactory outcomes in patients who have chronic ankle instability with GJH if the contralateral uninjured ankle shows a normal varus talar tilt and anterior talar translation during stress tests.
Study Design:
Case series; Level of evidence, 4.
Methods:
Modified Broström procedure was performed in 32 patients with chronic ankle instability with GJH if the contralateral uninjured ankle showed a normal varus talar tilt and anterior talar translation on stress tests. The mean patient age at surgery was 21.7 years, and the mean follow-up duration was 27.4 months.
Results:
The Karlsson-Peterson ankle score significantly improved from a mean ± SD of 63.6 ± 7.1 preoperatively to 90.4 ± 6.7 at the final postoperative follow-up (P < .001). Sixteen patients were very satisfied with the results, 10 patients were satisfied, 3 patients rated their satisfaction as fair, and 1 patient was dissatisfied with the results. Nine patients sustained ankle sprains after the surgery, 6 of which were mild sprains. Although 3 of these 9 patients had a mechanically unstable ankle on stress radiographs, they were satisfied with the postoperative results. None of the patients required a reoperation.
Conclusion:
GJH was not a contraindication for the modified Broström procedure if the contralateral uninjured ankle showed a normal varus talar tilt and a normal anterior talar translation on stress tests. Further studies are needed to better define GJH affecting the ankle.
Keywords
The modified Broström procedure, an anatomic repair of the anterior talofibular ligament (ATFL) and calcaneofibular ligament (CFL), is widely used in the surgical treatment of chronic ankle instability.2,10,14,15,20 However, direct repair of the ligaments is not recommended when the local ligamentous tissue is severely attenuated or when the patients have generalized joint hypermobility (GJH).1,3,4,7,12,31 Poor outcomes have been reported after the modified Broström procedure for chronic ankle instability with GJH.14,26 In such cases, lateral ankle ligament reconstruction using an allograft or autograft tendon is recommended under the assumption that the repaired ligaments would eventually stretch out in patients with GJH.4,7 However, this assumption has not been proven or thoroughly evaluated. This is probably true in patients with GJH secondary to connective tissue disorders such as Marfan syndrome or classic Ehlers-Danlos syndrome, as these patients have been found to have a marked reduction of elastin fiber content, abnormal collagen fibrils, or collagen V mutations, which explains the inherent connective tissue extensibility.8,19,22 However, GJH includes benign joint hypermobility syndrome, which is a clinical diagnosis with an unmapped causative gene and unknown biological markers. Several studies have reported that the biomechanical properties of a tendon in patients with benign joint hypermobility syndrome are similar to those of a tendon in normal healthy controls, whereas these properties are decreased in the tendons of those with classic Ehlers-Danlos syndrome. 22 GJH also includes mild joint hypermobility without any symptoms or problems except increased joint range of motion, and due to this, some investigators do not regard GJH as a discrete pathologic entity.13,27,28,35 Many different factors, such as patient age or sex, can contribute to increased joint range of motion; therefore, it cannot be postulated that all patients with GJH, as determined by physical tests, have connective tissue abnormalities that will cause the repaired ligaments to eventually stretch out.11,21,30,32
GJH is measured most commonly by use of the Beighton and Horan Joint Mobility Index (Beighton score), which evaluates joint mobility at several sites on a 9-point scale and defines hypermobility as a score of ≥4 (Table 1).9,27,28 However, controversy remains regarding the tests or cutoff score used to determine GJH.13,23,27 The Beighton score has been recommended by rheumatologists to determine GJH, but the applicability of the index for determining the degree of GJH that would lead to a poor outcome after the modified Broström procedure was not assessed.9,13,27 In a recent study of 49 patients with chronic ankle instability treated with modified Broström procedure, 3 cases failed, and all 3 failures had a Beighton score of ≥4; the authors reported that their findings were consistent with those of other studies, which had found that GJH resulted in poor outcomes. 26 However, 23 patients in that study had a Beighton score of ≥4, and among those 23 patients, 20 (89%) patients had a successful outcome after the modified Broström procedure. 26 A Beighton score of ≥4 may not be sufficient to determine whether the GJH would result in a poor outcome after a modified Broström procedure.
The Beighton and Horan Joint Mobility Index for Generalized Joint Hypermobility
We hypothesized that when the contralateral uninjured ankle shows a normal varus talar tilt and anterior talar translation during stress tests in patients with GJH as identified with a Beighton score of ≥4, GJH may have a smaller effect on the ankle ligaments, and the modified Broström procedure in these cases may have satisfactory outcomes.
Methods
This study involved 32 patients (32 ankles) with chronic ankle instability and GJH who underwent the modified Broström procedure from 2010 to 2013. Institutional review board approval was obtained. The functional and radiographic status was evaluated preoperatively, 6 months postoperatively, 12 months postoperatively, and at the final postoperative follow-up visit. The patients provided informed consent, and they were invited for a final follow-up office visit for a final evaluation. Six patients were not reachable; telephone numbers were incorrect for 2 of them, and 4 patients declined because they had moved to a distant location. We obtained the outcomes of these latter 4 patients through a telephone survey with respect to the final functional assessment and subjective satisfaction. The study included 29 men and 3 women. The mean patient age at surgery was 21.7 years (range, 20-37 years). The mean follow-up duration was 27.4 months (range, 17-51 months). Before surgery, clinical and radiographic examinations were performed to determine the presence of mechanical instability and concomitant pathologic abnormalities.
Patient Selection
The indications for the modified Broström procedure were unrelieved chronic ankle instability of at least 6 months duration and 3 months of failed rehabilitation. Chronic ankle instability with GJH and a normal varus talar tilt and anterior talar translation of the contralateral ankle during stress tests were indications for the modified Broström procedure. GJH was defined as a Beighton score of ≥4 on a 9-point scale.9,26,27 We determined the Beighton score by giving points for positive findings during examinations of the right and left passive dorsiflexion of the little finger beyond 90° (2 points), right and left passive apposition of the thumb to the flexor aspect of the forearm (2 points), right and left hyperextension of the knee beyond 10° (2 points), right and left hyperextension of the elbow beyond 10° (2 points), and the ability to place both hands flat on the floor with flexion at the waist and straight knees (1 point) (Table 1). We considered the varus stress test and anterior drawer test as normal for the contralateral ankle when both the physical stress examination and the instrumented stress radiographs were normal. Manual physical stress examination was considered normal when a hard stop could be appreciated; when the stop was soft with dimpling of the skin over the anterolateral ankle, it was considered pathologic. Stress radiographs were performed by use of the Telos device (Telos GmbH) with 15 daN of force applied, and results were considered pathologic with a varus talar tilt angle of greater than 10° and anterior talar translation greater than 6 mm. Contraindications for the modified Broström procedure were patient weight greater than 100 kg, failed Broström surgical cases, or presence of a large ossicle within the lateral ligaments, removal of which resulted in a lack of remnant ligamentous tissue. Intraoperative findings of much attenuated or deficient lateral ligaments were also considered contraindications for this procedure. Before surgery, magnetic resonance imaging was performed to evaluate any associated pathologic abnormalities. Hindfoot alignment view radiographs were taken to check for heel alignment. Any deformity was corrected before ligament repair. Nevertheless, all patients showed normal heel alignment before the operation in this study.
Operative Procedure
When ankle arthroscopy was indicated, it was performed before the Broström procedure. Ankle arthroscopy was not performed routinely for diagnostic purposes but was performed to address specific conditions, such as an osteochondral lesion of the talus, the os trigonum syndrome, and bony impingement. However, diagnostic ankle arthroscopy was indicated for deep ankle pain or pain that could not be approached through the Broström incision. After ankle arthroscopy, a 4-cm curvilinear incision was made from the fibular attachment of the ATFL to the distal tip of the lateral malleolus under tourniquet control. The ATFL and CFL were examined. When ossicles were present within the ligaments, the ligaments were gently peeled off, and the ossicles were removed. After the ligamentous tissue was confirmed to be suitable for repair, 2 anchors were inserted at the fibular attachment site of the ATFL and CFL to repair both the ligaments. Extensor retinaculum was augmented to the repaired ligaments.
Postoperatively, the ankle was immobilized in a splint with no weightbearing for 2 weeks, at which time the sutures were removed. A short leg cast was applied in a neutral position for 2 weeks, and tolerable partial weightbearing was allowed in a short leg cast. Thus, the immobilization period lasted 4 weeks. Active range of motion and partial weightbearing were allowed in a soft ankle orthosis. Full weightbearing was gradually allowed when tolerated. Soft ankle orthoses were worn for another 4 weeks after the immobilization period. Balance training and proprioceptive exercises were encouraged. A full range of sport activities was allowed 6 months after the surgery. Before starting sport activities, if the patients experienced ankle instability and a feeling of giving way, they were advised to delay their participation in sport activities and were encouraged to concentrate on balance training and peroneal strengthening exercises. Patients who achieved mechanical stability and did not experience any feeling of giving way in their ankles but were concerned about injuring their ankle again were assured that gradual participation in sport activities was possible.
Clinical Assessment
The Karlsson-Peterson ankle score was assessed preoperatively, 6 months postoperatively, 12 months postoperatively, and at the final follow-up examination. 16 The 100-point, 8-category Karlsson-Peterson ankle score scoring system combines subjective ankle instability (25 points), pain (20 points), swelling (10 points), stiffness (5 points), need for support (5 points), and symptoms associated with stair climbing (10 points), running (10 points), and work activities (15 points). 16 Functional assessments at final follow-up were graded according to the criteria described by Li et al 18 (excellent, 95-100; acceptable, 80-95; unacceptable, <80). Patients were asked to rate their overall satisfaction with the surgical results as very satisfied, satisfied, fair, or dissatisfied. Radiographically, the anterior talar translation and talar tilt angle were measured on stress radiographs preoperatively and at final follow-up.
Statistical Analysis
Statistical analysis was performed using SPSS version 12.0 (SPSS Inc) software. Data normality was assessed using the Kolmogorov-Smirnov test. The paired t test and Wilcoxon signed rank test were used to compare preoperative and postoperative values (Karlsson-Peterson ankle score and radiographic measurements). Bivariate associations between Beighton score and clinical outcomes were examined using the Spearman correlation analysis. Statistical significance was set at P < .05.
Results
The Karlsson-Peterson ankle score significantly improved by a mean (±SD) of 26.7 ± 7.9 points (95% CI, 22.1-29.7; P < .001), from 63.6 ± 7.1 preoperatively to 90.4 ± 6.7 at final postoperative follow-up. According to the criteria described by Li et al, 18 11 patients were graded as excellent, 18 patients as acceptable, and 1 patient as unacceptable. The mean talar tilt angle significantly improved, from 12.2° ± 3.5° preoperatively to 6.1° ± 2.2° at final follow-up (P < .05), and mean anterior talar translation also significantly improved, from 8.1 ± 1.5 mm preoperatively to 5.3 ± 0.8 mm at final follow-up (P < .05). We stratified the clinical outcomes according to Beighton score (Table 2). There was no correlation between the Beighton score and the Karlsson-Peterson ankle score at the last follow-up (Spearman correlation coefficient, –0.11; P = .591). However, 1 patient with a Beighton score of 8 and 1 patient with a score of 9 had lower Karlsson-Peterson ankle scores (82 and 85, respectively) compared with mean scores at the last follow-up.
Clinical Outcomes Stratified According to Beighton Score a
Spearman correlation coefficient, –0.11 (P = .591) between Beighton score and Karlsson-Peterson ankle score at last follow-up.
Twelve patients underwent ankle arthroscopy at the time of the modified Broström procedure. One patient had osteochondral lesion on the medial talus, which was treated operatively with microfracture bone marrow stimulation; 1 patient had os trigonum syndrome causing posterior impingement, which was treated with posterior ankle arthroscopy; 2 patients had anterior bony impingement, which was treated with ankle arthroscopy and miniarthrotomy for resection of the bony spurs. Eight patients underwent diagnostic ankle arthroscopy for their ankle pain, 8 patients had synovitis, and 4 patients had anterolateral soft tissue impingement addressed with ankle arthroscopy (Table 3).
Additional Procedures Combined With the Modified Broström Procedure
Sixteen patients were very satisfied with the results, 10 patients were satisfied, 3 patients rated their satisfaction as fair, and 1 patient was dissatisfied with the results. One patient who was dissatisfied with the result had a mechanically stable ankle but complained of pain 13 months after the surgery; however, the patient did not want additional surgery such as ankle arthroscopy. Nine patients sustained an ankle sprain after the surgery. Six of these injuries were mild sprains that did not require medication or immobilization. Three of these patients had a mechanically unstable ankle on stress radiographs but were satisfied with the result. None of the patients required a reoperation.
Discussion
The most important finding of the present study was that the modified Broström procedure may be successful in patients with chronic ankle instability and with GJH, as determined by a Beighton score of ≥4, if the contralateral uninjured ankle showed a normal varus talar tilt and anterior talar translation on stress tests.
Karlsson et al 14 reported unsatisfactory results after transection and imbrication of the lateral ankle ligament in patients with chronic ankle instability and GJH. Their study included 16 patients with GJH, and among these, 9 had fair or poor functional results; the authors concluded that GJH was a contraindication for this procedure. The results of this study have been commonly cited as a reference for the modified Broström procedure resulting in poor clinical outcomes in patients with GJH and for the necessity of lateral ligament reconstruction with an allograft or autograft tendon. However, among 9 patients with GJH who had fair or poor functional results, 4 patients also had long-standing ligamentous insufficiency (≥10 years), and most of the patients with GJH had undergone repair of only the ATFL and not the CFL; GJH therefore may not have been the only factor responsible for the poor outcome in these patients. Furthermore, among 16 patients with GJH, 7 had good results. Most of the other studies that report unfavorable outcomes do not specify the tests or cutoff scores used to determine the GJH. In fact, some patients with GJH in these studies had good results after the modified Broström procedure, leading to an assumption that appropriately selected cases with GJH could have good results after ligament repair.
Many authors report that GJH is not joint specific and that not all patients with GJH will show specific joint laxity.25,29 Sauers et al 29 found no moderate or stronger correlation between GJH and shoulder laxity. In a study of 57 athletes, a weak correlation was observed among the test variables for GJH and instrumented ankle laxity measured using an ankle arthrometer. 25 Not all patients with GJH as determined by a Beighton score of ≥4 may show increased ankle laxity. However, although not all patients with GJH will show increased ankle laxity on the uninjured normal ankle, its presence on the normal side may indicate ligamentous hyperextensibility of the ankle ligaments, and this ligamentous hyperextensibility can cause the repaired ligaments to stretch out eventually after the modified Broström procedure. We believe that excluding these patients with an increased varus talar tilt and anterior translation of the contralateral uninjured ankle led to good outcomes after the modified Broström procedure in the current study.
Several other reasons could account for the good results obtained in the current study after the modified Broström procedure for chronic ankle instability with GJH. First, we excluded patients who were expected to achieve a poor outcome by ligament repair, such as those with severely attenuated or deficient lateral ligaments that were found intraoperatively. 17 Excluding these patients may have led to the good results. Second, we combined ankle arthroscopy to address concomitant pathological abnormalities during the surgery, which could have also contributed to the satisfactory results. 6
We also encountered 18 patients with chronic ankle instability and GJH who had an increased varus talar tilt and anterior talar translation on the contralateral side on stress tests; we performed lateral ligament reconstruction using an allograft or autograft tendon. The Karlsson-Peterson ankle score significantly improved, from 61.3 ± 6.4 preoperatively to 89.4 ± 4.6 after the first postoperative year (P < .001). Among these patients, only 4 patients were without a history of ankle sprain or without a feeling of instability in the contralateral ankle; the remaining 14 patients had a history of ankle sprain or complained of ankle instability on the contralateral side. It was not clear whether inherent ligamentous hyperextensibility or a history of ligament injury had led to increased ankle laxity on the contralateral side. Nonetheless, we believe that reconstruction with an allograft or autograft tendon is required in cases that carry the possibility of inherent ligamentous hyperextensibility that may cause the ligaments to stretch out after the modified Broström procedure.4,7 However, compared with the modified Broström procedure, reconstruction with an allograft or autograft tendon is rather invasive as it requires bone tunnels to reconstruct the lateral ligaments. Therefore, if good results can be expected with both procedures, we believe that surgeons will prefer the modified Broström procedure, which is simple and less invasive. For those with GJH but normal stress tests on the contralateral ankle, the modified Broström procedure was successful during a short follow-up. However, 9 ankle sprains occurred after the surgery, 3 of which involved mechanically unstable ankles. Further study on more specific indications for this procedure may result in more cases with successful outcomes. Although the Beighton and Karlsson-Peterson ankle scores showed no correlation at the last follow-up in our study, 1 patient with a Beighton score of 8 and 1 patient with a score of 9 had lower Karlsson-Peterson ankle scores (82 and 85, respectively) compared with the mean scores. A study that includes more patients with higher Beighton scores is required to find out whether there is a higher cutoff point to determine unfavorable outcomes after this procedure.
The current study has several limitations. The primary limitation of this case series is the short follow-up period with no comparison group. It is possible that the repaired ligaments might have stretched out in a longer follow-up. However, many studies reporting poor results in patients with GJH have encountered unsatisfactory results in a short- to medium-term follow-up period, which leads to the assumption that when the repaired ligaments are stretched out due to their inherent hyperextensibility, this may occur early during short- to medium-term follow-up. Even though a direct comparison cannot be made with these studies, our outcomes with narrower indications for the modified Broström procedure in patients with GJH resulted in satisfactory results. However, it remains unclear whether these satisfactory results would continue over a long follow-up period, and further study is required. Another limitation of this study is that it has not been proven whether positive stress tests on the contralateral normal ankle are diagnostic of inherent ligamentous hyperextensibility. However, we believe that patients with positive stress tests on the contralateral normal ankle are more likely to have inherent ligamentous hyperextensibility than patients with normal, negative stress tests. Future studies on biomarkers to detect ligamentous tissue abnormality may help to determine whether the ligaments would stretch out after the modified Broström procedure.5,24,33,34 GJH is known to be age and sex specific, and young women tend to have more ligament laxity compared with older men. 11 Since many of our patients were young male soldiers, our results may not represent all patients with GJH. A randomized controlled trial on the modified Broström procedure versus tendon graft lateral ligament reconstruction may determine which procedure is better for patients with chronic ankle instability with GJH.
Conclusion
GJH was not a contraindication for the modified Broström procedure when the contralateral uninjured ankle showed a normal varus talar tilt and a normal anterior talar translation on stress tests. Further studies are needed to better define GJH affecting the ankle.
Footnotes
The authors declared that they have no conflicts of interest in the authorship and publication of this contribution.