Occult syndesmotic injury in chronic lateral ankle instability treated with Broström repair and suture-button stabilization

Abstract

BACKGROUND

Chronic lateral ankle instability may coexist with occult syndesmotic injury despite the absence of definite distal tibiofibular diastasis on static imaging. Because symptoms may overlap with those of isolated lateral ankle instability, this condition can remain unrecognized. This study hypothesized that selected patients with persistent anterior syndesmotic symptoms and stress-confirmed occult syndesmotic instability would demonstrate sustained improvement after combined Broström repair and suture-button stabilization.

AIM

To evaluate 5-year clinical outcomes of combined Broström repair and suture-button stabilization in this selected cohort.

METHODS

This uncontrolled retrospective case series at a tertiary hospital included 16 of 1487 patients treated from April 2018 to May 2021. Eligibility required chronic lateral ankle instability, focal syndesmotic symptoms, no static diastasis, failed conservative treatment, and qualitative confirmation by three intraoperative stress tests. No comparator or quantitative diagnostic threshold was used. All underwent combined Broström repair and suture-button stabilization. Repeated outcomes were analyzed using Friedman and Bonferroni-adjusted exact Wilcoxon tests.

RESULTS

The median injury-to-surgery interval was 21.0 months (interquartile range, 8.25-120.0 months). Preoperative visual analogue scale, American Orthopaedic Foot and Ankle Society, Foot and Ankle Ability Measure, and Karlsson scores were 3.00 ± 1.32, 69.69 ± 12.11, 69.67 ± 12.15, and 64.00 ± 11.86, respectively. At 5 years, scores were 0.44 ± 0.63, 92.06 ± 6.89, 89.21 ± 10.19, and 88.12 ± 8.69 (all adjusted P < 0.001; paired Cohen’s dz, -2.03, 2.06, 1.94, and 2.40). All patients returned to work at 2.44 ± 0.68 months. Seven patients (43.8%) underwent button removal at 1 year; none had implant-related symptoms at 5 years.

CONCLUSION

In this uncontrolled case series, combined treatment was associated with improvement, but the findings are exploratory and do not establish the independent benefit of suture-button stabilization.

Key Words: Syndesmosis; Occult syndesmotic injury; Suture-button; Chronic lateral ankle instability; Broström repair

Core Tip: This study highlights a selected subgroup of patients with chronic lateral ankle instability who had persistent anterior syndesmotic symptoms despite the absence of definite distal tibiofibular diastasis on static imaging. Occult syndesmotic instability was supported by qualitative stress testing under anesthesia. Combined Broström repair and suture-button stabilization was associated with sustained clinical improvement at 5 years. These hypothesis-generating findings support careful syndesmotic assessment and a selective approach to stabilization but do not establish an independent treatment effect.

INTRODUCTION

Chronic lateral ankle instability (CLAI), characterized by recurrent episodes of giving way, pain, and functional limitation following lateral ankle ligament injury, is a common sequela of ankle sprain[1]. Syndesmotic injuries, or high ankle sprains (HAS), account for approximately 10% of all ankle sprains and 25% of ankle sprains in athletic populations[2-4] and may occur concomitantly with lateral ligament injuries. The syndesmotic complex comprises the anterior inferior tibiofibular ligament (AITFL), interosseous ligament, interosseous membrane, and posterior inferior tibiofibular ligament and serves as an important stabilizing structure of the ankle joint[5-10]. HAS is commonly classified according to clinical and radiographic findings: Grade 1 injuries are considered stable, grade 2 injuries involve partial disruption with variable instability, and grade 3 injuries involve complete disruption[11-14]. However, chronic subtle syndesmotic lesions may be difficult to classify when static imaging demonstrates no definite diastasis despite persistent syndesmotic symptoms.

Unstable grade 2 and grade 3 syndesmotic injuries are commonly treated with flexible suture-button fixation or rigid screw fixation[15-17], whereas radiographically stable grade 1 injuries are generally managed nonoperatively with immobilization or protected weightbearing followed by progressive rehabilitation[12,18-21]. In this report, grade 1 denotes the absence of definite diastasis on static imaging, whereas occult instability denotes dynamic separation detected only during stress assessment; this radiographically stable but stress-positive phenotype differs from uncomplicated grade 1 HAS. A clinically important management gap remains for the small subgroup of patients with CLAI, persistent anterior syndesmotic pain, no definite diastasis on static imaging, and instability detectable only during stress assessment. Consensus statements and recent studies emphasize the diagnostic challenges of chronic syndesmotic injury, the limited sensitivity of static imaging for subtle or dynamic instability, and the potential role of surgical stabilization when instability is demonstrated by imaging, stress assessment, or arthroscopy[22-27]. Dynamic and arthroscopic studies further suggest that concomitant syndesmotic injury may influence ankle stability in the presence of lateral ligament disruption[28-30]. Nevertheless, evidence specifically addressing radiographically stable but stress-positive occult syndesmotic injury in patients with CLAI remains limited.

This retrospective case series aimed primarily to characterize this uncommon clinical phenotype and secondarily to describe 5-year outcomes after combined Broström repair and suture-button stabilization. The novel observation of interest was the selected radiographically stable but stress-positive phenotype; an independent treatment effect of suture-button stabilization was not tested. It was hypothesized that selected patients with CLAI, persistent anterior syndesmotic symptoms, no definite diastasis on static imaging, and positive intraoperative stress testing would demonstrate sustained postoperative improvement. This hypothesis applies specifically to a selected stress-positive subgroup and should not be interpreted to suggest that all grade 1 HAS injuries require stabilization. The study was reported in accordance with the Strengthening the Reporting of Cohort, Cross-sectional, and Case-control Studies in Surgery 2024 guideline[31].

MATERIALS AND METHODS

This retrospective case series included patients with CLAI and suspected occult syndesmotic injury who were treated between April 2018 and May 2021. CLAI was diagnosed on the basis of persistent instability symptoms or recurrent episodes of giving way, clinical findings consistent with lateral ligament insufficiency, and imaging or intraoperative evidence of lateral ligament injury. Preoperative suspicion of occult syndesmotic injury required focal pain or tenderness over the anterior syndesmosis (Figure 1), no definite distal tibiofibular diastasis on static imaging, and failure of conservative treatment. Clinical examination included the external rotation stress, squeeze, pronation stress, and anterior drawer tests. All 16 patients underwent non-weight-bearing radiographs and computed tomography (CT) at our institution (Figure 2). Magnetic resonance imaging (MRI) was available for all patients; eight MRI examinations were performed at our institution and eight at outside institutions. MRI was used to characterize ligament injury but not as a stand-alone test of mechanical instability. Radiographs and CT examinations followed the institution’s routine standardized imaging workflow. Detailed acquisition parameters, including exact foot positioning and limb rotation, were not consistently available for retrospective verification. CT images were assessed at a consistent distal tibiofibular level using multiplanar reconstructions; contralateral imaging was unavailable. Absence of definite diastasis meant no visually apparent distal tibiofibular widening or malalignment on static radiographs and CT. No prespecified quantitative measurement cutoff or contralateral comparison was applied.

Figure 1 Clinical photographs showing the sites of persistent pain or tenderness at the anterior syndesmosis (blue areas). A-C: The symptomatic area viewed from three directions.

Figure 2 Non-weight-bearing ankle radiographs showing no definite syndesmotic diastasis. A: Anteroposterior view; B: Lateral view.

Eligible patients had CLAI with persistent pain or tenderness over the anterior syndesmosis, radiographically stable grade 1 syndesmotic injury, at least 3 months of unsuccessful conservative treatment, and intraoperative confirmation of occult syndesmotic instability.

All included patients underwent a standardized framework of conservative treatment after injury. During the acute phase, the injured ankle was immobilized in a short-leg cast or rigid controlled ankle motion walking boot for 2-4 weeks. Weightbearing was restricted and protected with crutch assistance, with progression according to pain tolerance, and a short course of oral nonsteroidal anti-inflammatory drugs was prescribed for acute pain control. Thereafter, rehabilitation focused on progressive restoration of ankle range of motion, strengthening of the peroneal and other periankle muscles, and proprioceptive and neuromuscular training, including balance-board exercises. Conservative treatment was continued for at least 3 months. The precise progression of treatment was individualized according to symptoms and injury characteristics, including the presence of avulsion fragments.

Exclusion criteria comprised isolated lateral ligament injury without syndesmotic involvement; grade 2 or 3 syndesmotic injury; acute peri-ankle fracture requiring fixation; severe talar osteochondral injury; previous ankle surgery; neurological disease; systemic disease likely to impair healing or functional outcomes, including diabetes mellitus or chronic kidney disease; systemic inflammatory arthropathy; advanced ankle osteoarthritis; incomplete preoperative imaging or baseline data; and insufficient follow-up.

Of the 1487 patients screened, 1351 were excluded: 1050 had isolated lateral ligament injury without syndesmotic involvement, 160 had grade 2 or 3 syndesmotic injury, 90 had a peri-ankle fracture or severe talar osteochondral lesion, and 51 had previous ankle surgery, systemic inflammatory arthropathy, or advanced osteoarthritis. Among the remaining 136 patients with lateral ligament injury and grade 1 syndesmotic injury, 118 improved with conservative treatment and did not require surgery.

Eighteen patients remained eligible for operative treatment. One was excluded because of incomplete preoperative MRI or baseline clinical data, and one was excluded because follow-up was shorter than 24 months. The final cohort therefore comprised 16 patients. All 16 included patients completed the 1-, 3-, and 5-year assessments. None had undergone previous ankle surgery or had an osteochondral lesion, and no mild or severe articular cartilage lesions were identified during arthroscopy.

The study was approved by the Medical Ethics Committee of Peking University People’s Hospital on October 8, 2018, No. 2018PHB131-01. All participants provided written informed consent for retrospective use of their medical records and publication of clinical and intraoperative images. During surgery, patients were positioned supine under spinal anesthesia. All three intraoperative stress tests were performed using standard maneuvers[10,24].

For the external rotation stress test, the lower leg was stabilized and the dorsiflexed foot was manually externally rotated under fluoroscopy; visually apparent widening of the distal tibiofibular relationship compared with the unstressed position was considered supportive of instability. For the hook/Cotton test, a bone hook was placed around the distal fibula and lateral traction was applied with countertraction on the tibia; visible lateral translation or widening under fluoroscopy was considered supportive. During arthroscopy, the AITFL and distal tibiofibular interval were inspected and probed while external rotation and lateral fibular translation were applied; dynamic separation or loss of a congruent syndesmotic relationship was considered supportive. Exact applied forces and displacement were not recorded, and no predefined millimetric or quantitative arthroscopic threshold was used. The intraoperative stress assessment was considered positive when any one of the three maneuvers demonstrated a qualitative finding supportive of instability. All procedures were performed by one foot and ankle surgeon with more than 10 years of subspecialty experience using a uniform assessment and treatment protocol.

An endobutton (Smith & Nephew, Memphis, TN, United States) was placed on the anteromedial tibial cortex and connected by a suture passed through a tibiofibular tunnel to a second endobutton positioned on the lateral fibular cortex. The distal tibiofibular joint was not compressed with reduction forceps. The suture was tightened with the ankle in a neutral position under fluoroscopic guidance, and implant positions were confirmed fluoroscopically (Figure 3). Arthroscopic Broström repair was performed in patients without an os subfibulare, whereas open repair was performed in patients with an os subfibulare, an accessory ossicle or chronic avulsion fragment adjacent to the lateral malleolus that may require direct exposure during ligament repair[10,32].

Figure 3 Intraoperative steps of arthroscopic Broström repair. A-F: Sequential steps of the arthroscopic repair.

Patients subsequently followed a standardized postoperative rehabilitation framework. The ankle was maintained in a neutral position in a brace, and weightbearing was prohibited for 4 weeks. Toe and knee motion was initiated immediately, and ankle rehabilitation exercises commenced at 4 weeks[33]. Immediate postoperative radiographs were obtained on postoperative day 1. At the 4-week assessment, progression to partial weightbearing in a walking boot was permitted when implant position was satisfactory, the distal tibiofibular relationship remained stable, and no syndesmotic widening was evident. Weightbearing was then gradually advanced to full weightbearing according to clinical and radiographic assessment.

All 16 included patients were evaluated at 1 year, 3 years, and 5 years after surgery using physical examination, questionnaires, and radiographs. Complete Visual Analogue Scale (VAS), American Orthopaedic Foot and Ankle Society (AOFAS), Foot and Ankle Ability Measure (FAAM), Karlsson, and radiographic data were available at every time point; no imputation was performed. Immediate postoperative radiographs were obtained on postoperative day 1, and radiographic follow-up at 1 year, 3 years, and 5 years was used to assess syndesmotic alignment and implant-related complications. Outcomes included time to return to preinjury work, the VAS, AOFAS score, FAAM, and Karlsson score[34-37]. Time to return to preinjury work was defined as the interval from surgery to resumption of usual occupational duties. Because this was an exploratory retrospective case series, no single primary outcome was prespecified, and all reported clinical outcomes were considered exploratory. The patient constituted the independent unit of analysis. Baseline variables included sex, age, body mass index, injury mechanism, and injury-to-surgery interval; individual-level data are presented in Table 1. Physical examinations and questionnaires were completed by two experienced foot and ankle surgeons who were not the operating surgeon and were unaware of the intraoperative diagnostic findings and operative details; blinding to the follow-up time point was not feasible. Preoperative imaging demonstrated no definite syndesmotic diastasis in any patient, whereas occult instability was supported by at least one positive qualitative intraoperative stress maneuver under anesthesia. Individual stress-test grades and quantitative displacement measurements were not recorded. One illustrative patient sustained a sports-related injury in November 2019 and subsequently developed lateral ankle and anterior syndesmotic pain. Initial radiographs and CT demonstrated a lateral ankle avulsion fracture without syndesmotic diastasis (Figure 4). She was diagnosed with an anterior talofibular ligament tear and radiographically stable HAS and underwent conservative treatment. Persistent pain prompted reassessment in May 2020. Anterior syndesmotic tenderness was present; the external rotation and squeeze tests were negative, whereas the pronation stress and anterior drawer tests were positive. Repeat imaging again demonstrated no definite syndesmotic diastasis (Figure 5). Intraoperative stress testing under anesthesia was positive, and postoperative radiographs were obtained following combined surgery (Figure 6).

Figure 4 Baseline radiographs and computed tomography images obtained immediately after injury in November 2019 from the illustrative patient. A: Anteroposterior radiograph; B: Lateral radiograph; C: Three-dimensional computed tomography reconstruction; D: Axial computed tomography image.

Figure 5 Follow-up radiographs, computed tomography images, and magnetic resonance images obtained in May 2020, approximately 6 months after injury, from the same illustrative patient shown in Figure 4. A: Anteroposterior radiograph; B: Lateral radiograph; C: Three-dimensional computed tomography reconstruction; D: Axial computed tomography image; E and F: Coronal magnetic resonance images; G and H: Axial magnetic resonance images.

Figure 6 Postoperative radiographs after suture-button stabilization and Broström-Gould repair. A: Anteroposterior view; B: Lateral view.

Table 1 Individual characteristics, treatment details, and outcomes of the 16 patients.

Patient No.	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16
Age (years)	55	52	21	50	29	36	31	19	33	25	48	51	67	53	56	54
Side	R	L	R	R	L	R	L	L	R	R	L	R	R	L	L	R
Sex	F	F	M	F	F	F	M	F	F	F	M	F	F	F	F	F
Body mass index (kg/m2)	23.63	23.88	23.29	23.05	22.04	21.72	26.17	17.65	22.86	29.38	26.20	24.84	18.99	19.15	21.93	24.8
Injury mechanism	Walking downstairs	Walking on level ground	Sports	Sports	Walking downstairs	Sports	Sports	Sports	Walking downstairs	Sports	Walking downstairs	Sports	Sports	Fall from a height	Fall from a height	Fall from a height
Injury-to-surgery interval (months)	120	120	9	360	4	120	18	60	24	120	5	4	12	24	12	6
VAS, preoperative	3	3	3	1	6	4	4	2	2	2	3	4	2	4	1	4
VAS, 1 year postoperative	0	2	0	0	0	0	1	0	0	0	3	2	2	0	0	0
VAS, 3 years postoperative	0	2	0	0	1	0	0	0	0	0	2	2	1	0	0	1
VAS, 5 years postoperative	0	1	0	0	1	0	0	0	0	0	2	1	1	0	0	1
AOFAS, preoperative	75	58	80	83	44	64	60	77	79	84	70	69	61	54	87	70
AOFAS, 1 year postoperative	96	81	100	92	96	89	92	100	98	100	79	76	80	91	99	80
AOFAS, 3 years postoperative	96	84	100	92	92	90	92	100	98	100	80	78	82	91	99	84
AOFAS, 5 years postoperative	96	86	100	94	94	91	93	100	98	100	82	80	83	92	99	85
FAAM, preoperative	72.62	55.95	89.29	76.19	50.00	76.19	60.00	77.38	78.57	82.14	60.71	59.52	60.71	54.76	85.71	75.00
FAAM, 1 year postoperative	96.43	82.14	100.00	86.90	97.62	88.10	80.95	94.04	90.48	100.00	63.10	69.05	73.81	88.10	100.00	91.67
FAAM, 3 years postoperative	96.43	86.90	100.00	86.90	95.24	88.10	80.95	94.04	90.48	100.00	65.48	69.05	73.81	88.10	100.00	95.24
FAAM, 5 years postoperative	96.43	88.10	100.00	88.10	95.24	90.48	83.33	95.24	92.86	100.00	66.67	71.43	76.19	88.10	100.00	95.24
Karlsson, preoperative	62	52	80	75	39	64	55	72	74	79	60	59	57	52	80	64
Karlsson, 1 year postoperative	94	80	100	89	92	85	82	92	90	100	70	69	72	84	97	82
Karlsson, 3 years postoperative	94	82	100	90	89	87	82	95	92	100	72	70	75	85	97	84
Karlsson, 5 years postoperative	95	84	100	90	92	87	84	94	94	100	74	72	77	85	97	85
Residual focal syndesmotic pain	No	No	No	No	No	No	No	No	No	No	No	No	No	No	No	No
Button removal	No	Yes	No	Yes	No	Yes	No	Yes	Yes	No	Yes	No	No	Yes	No	No
Postoperative difficulty with squatting (yes/no)	No	No	No	No	No	Yes	No	Yes	No	No	No	No	No	Yes	No	No
Lateral ligament repair	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Lateral ligament repair approach	Arthroscopic	Arthroscopic	Arthroscopic	Arthroscopic	Arthroscopic	Arthroscopic	Arthroscopic	Open	Open	Open	Open	Open	Arthroscopic	Arthroscopic	Arthroscopic	Open
Return to preinjury work (months)	2.0	2.5	1.0	2.0	2.5	3.0	2.5	2.5	2.0	1.5	3.5	3.5	3.0	3.0	2.0	2.5
Satisfied with surgery	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes

Data are shown for each patient. Age is reported in years; body mass index in kg/m²; injury-to-surgery interval and return to preinjury work in months. AOFAS: American Orthopaedic Foot and Ankle Society; FAAM: Foot and Ankle Ability Measure; VAS: Visual Analogue Scale; R: Right; L: Left.

Statistical analyses were performed using SPSS version 26.0 (IBM Corp., Armonk, NY, United States) and Microsoft Excel 2016. Continuous variables are presented as the mean ± SD or median and interquartile range, as appropriate, whereas categorical variables are presented as counts and percentages. Repeated outcome measures were evaluated using the Friedman test. When the overall test result was significant, each postoperative time point was compared with baseline using a two-sided exact Wilcoxon signed-rank test. Zero differences were excluded from the signed-rank calculation. Three planned comparisons were performed for each outcome; raw P values were multiplied by three using the Bonferroni method and capped at 1.00. Table 2 presents the Friedman statistics, raw and adjusted P values, numbers analyzed, and paired Cohen’s dz effect sizes. Paired Cohen’s dz was calculated as the mean within-patient change divided by the standard deviation of the within-patient changes. No a priori power calculation was performed because all eligible patients treated during the retrospective study period were included. Accordingly, the analyses were considered exploratory and hypothesis-generating. Statistical significance was defined as an adjusted P < 0.05. This retrospective case series was not prospectively registered.

Table 2 Preoperative and postoperative Visual Analogue Scale, American Orthopaedic Foot and Ankle Society, Foot and Ankle Ability Measure, and Karlsson scores, Friedman statistics, exact P values, and paired effect sizes, mean ± SD.

Score	VAS	AOFAS	FAAM	Karlsson
Preoperative	3.00 ± 1.32	69.69 ± 12.11	69.67 ± 12.15	64.00 ± 11.86
Friedman χ2 (df = 3); P	37.59; < 0.001	40.49; < 0.001	40.83; < 0.001	41.35; < 0.001
1 year postoperative	0.62 ± 1.02	90.56 ± 8.64	87.65 ± 11.30	86.12 ± 9.96
Raw P; adjusted P; Cohen’s dz	0.000122; 0.000366; -1.46	0.000031; 0.000092; 1.77	0.000031; 0.000092; 1.66	0.000031; 0.000092; 2.07
3 years postoperative	0.56 ± 0.81	91.12 ± 7.54	88.17 ± 10.86	87.12 ± 9.34
Raw P; adjusted P; Cohen’s dz	0.000031; 0.000092; -1.85	0.000031; 0.000092; 2.00	0.000031; 0.000092; 1.76	0.000031; 0.000092; 2.38
5 years postoperative	0.44 ± 0.63	92.06 ± 6.89	89.21 ± 10.19	88.12 ± 8.69
Raw P; adjusted P; Cohen’s dz	0.000031; 0.000092; -2.03	0.000031; 0.000092; 2.06	0.000031; 0.000092; 1.94	0.000031; 0.000092; 2.40

n = 16 at every time point. Friedman χ² statistics have 3 degrees of freedom. Raw two-sided exact Wilcoxon signed-rank P values and Bonferroni-adjusted P values are reported for each postoperative-versus-preoperative comparison (three planned comparisons per outcome). Cohen’s dz is the mean within-patient change divided by the standard deviation of within-patient changes; negative Visual Analogue Scale values indicate pain reduction. Injury-to-surgery interval was 21.0 months (interquartile range, 8.25-120.0 months). AOFAS: American Orthopaedic Foot and Ankle Society; FAAM: Foot and Ankle Ability Measure; VAS: Visual Analogue Scale.

RESULTS

Of the 1487 patients screened, 16 met all eligibility and follow-up criteria (Figure 7). The cohort comprised three men and 13 women, with a mean age of 42.50 ± 14.60 years and a mean body mass index of 23.10 ± 2.97 kg/m². Nine ankles were right-sided and seven were left-sided. The mean injury-to-surgery interval was 63.62 ± 92.36 months, and the median interval was 21.0 months (interquartile range, 8.25-120.0 months), reflecting substantial clinical heterogeneity. Individual patient characteristics and outcomes are summarized in Table 1. Preoperative VAS, AOFAS, FAAM, and Karlsson scores were 3.00 ± 1.32, 69.69 ± 12.11, 69.67 ± 12.15, and 64.00 ± 11.86, respectively. At 1 year, the corresponding scores were 0.62 ± 1.02, 90.56 ± 8.64, 87.65 ± 11.30, and 86.12 ± 9.96; at 3 years, they were 0.56 ± 0.81, 91.12 ± 7.54, 88.17 ± 10.86, and 87.12 ± 9.34; and at 5 years, they were 0.44 ± 0.63, 92.06 ± 6.89, 89.21 ± 10.19, and 88.12 ± 8.69. Overall differences were significant for all outcomes (Friedman χ² = 37.59 for VAS[3], 40.49 for AOFAS, 40.83 for FAAM, and 41.35 for Karlsson; all P < 0.001). Each postoperative score differed from baseline after Bonferroni adjustment, with large standardized paired changes across all time points (absolute Cohen’s dz, 1.46-2.40). Exact raw and adjusted P values are reported in Table 2.

Figure 7  Flow diagram of patient screening, exclusion, and final inclusion.

All 16 patients contributed complete clinical scores and radiographs at the 1-, 3-, and 5-year assessments. Four patients (25.0%) sustained injuries while walking downstairs, eight (50.0%) during sports, one (6.3%) while walking on level ground, and three (18.8%) after a fall from height. No patient had definite syndesmotic diastasis on preoperative static imaging. Every patient underwent external rotation, hook/Cotton, and arthroscopic stress assessment under anesthesia, and at least one maneuver showed a qualitative finding supportive of occult instability in every patient; individual maneuver grades and quantitative displacement were unavailable. No recurrent syndesmotic widening or radiographic implant-related complication was identified on postoperative day 1 or at the 1-, 3-, or 5-year follow-up. No patient reported residual focal anterior syndesmotic pain, and all returned to preinjury work at a mean of 2.44 ± 0.68 months.

Seven patients (43.8%) underwent endobutton removal at 1 year. Among the three symptomatic patients, symptoms developed approximately 3 months postoperatively and persisted for more than 3 months. Initial nonoperative management consisted of reducing ankle activity, which provided partial symptom relief but was incompatible with the patients’ desired activity levels and daily functional demands; consequently, they elected implant removal. Symptoms improved immediately after removal in patient No. 14; patients No. 6 and No. 8 improved during subsequent rehabilitation. Four asymptomatic patients (No. 2, No. 4, No. 9, and No. 11) requested implant removal. Apart from these seven removals, no wound-healing problem, infection, neurologic complication, deep venous thrombosis, recurrent clinical instability, or other reoperation occurred. No patient had implant-related symptoms at 5 years (Tables 1 and 3).

Table 3 Adverse events and additional procedures, n (%).

Event	Patients	Details
Wound-healing complication	0/16 (0)	None
Surgical-site infection	0/16 (0)	None
Neurologic complication	0/16 (0)	None
Deep venous thrombosis	0/16 (0)	None
Recurrent clinical ankle instability	0/16 (0)	None
Suture-button removal	7/16 (43.8)	Three for persistent symptoms; four at patient request
Other reoperation	0/16 (0)	None
Implant-related symptoms at 5 years	0/16 (0)	None

No adverse event or additional procedure other than suture-button removal occurred during follow-up.

DISCUSSION

This study characterizes an uncommon clinical subgroup of patients with CLAI who had persistent anterior syndesmotic symptoms despite the absence of definite distal tibiofibular diastasis on preoperative static imaging. The findings should not be interpreted to suggest that all grade 1 HAS injuries require suture-button fixation. Rather, concomitant syndesmotic injury should be considered when focal syndesmotic symptoms persist and instability is supported by stress testing under anesthesia. Although combined Broström repair and suture-button stabilization was associated with sustained improvement, the absence of an isolated Broström control group precludes determining whether Broström repair alone would have produced comparable outcomes. The observed pre-post changes may also reflect rehabilitation, regression to the mean, and selection of patients with persistent symptoms after failed conservative treatment. Accordingly, the results generate awareness and hypotheses rather than direct a specific treatment strategy.

Syndesmotic injuries typically result from external rotation forces, whereas lateral ligament injuries more commonly result from inversion forces[1,8,10,16]. Additional mechanisms include talar eversion and excessive dorsiflexion[38]. Sports-related injury was the most common mechanism in this cohort. Diagnosing chronic subtle syndesmotic injury remains challenging because its symptoms overlap with those of lateral ankle instability and static imaging may not demonstrate definite abnormalities. Expert consensus supports an integrated diagnostic approach incorporating imaging, stress assessment, and arthroscopy[22-27]. Emerging dynamic imaging techniques may further improve detection: Dynamic ultrasound can quantify stress-dependent fibular motion[28], and a 2025 cadaveric diagnostic study demonstrated that three-dimensional weightbearing CT distance mapping could detect subtle widening under axial load, although clinical validation remains necessary[39]. In this cohort, standardized non-weight-bearing imaging demonstrated no definite diastasis, whereas all patients had persistent anterior syndesmotic symptoms and at least one positive qualitative stress maneuver under anesthesia. These patients were therefore considered radiographically stable but functionally stress-positive rather than representative of uncomplicated grade 1 HAS. Because the diagnosis depended on the operating surgeon’s qualitative judgment, the study cannot estimate the prevalence or reproducibility of this phenotype, and treatment selection may have been influenced by diagnostic selection bias. Because contralateral imaging was unavailable, individual anatomical variation could not be accounted for through bilateral comparison.

Stable syndesmotic injuries are generally managed nonoperatively, and isolated grade 1 HAS often has favorable outcomes[18,20,40-42]. The selected patients in this series differed in that they had CLAI, persistent anterior syndesmotic symptoms, and stress-confirmed occult instability. Flexible suture-button fixation can stabilize the syndesmosis while maintaining reduction[8,10,16,43,44], and reported surgical options for chronic syndesmotic instability also include arthroscopic debridement, screw fixation, and ligament repair or augmentation[25-27]. The distal tibiofibular syndesmosis undergoes physiological motion during loading and walking[45]. One possible but untested explanation for the persistent symptoms observed in this cohort is a subtle AITFL lesion or pathological distal tibiofibular micromotion. Similarly, the proposed ability of the suture-button construct to restrict pathological micromotion while preserving physiological motion represents a biomechanical interpretation rather than a mechanism demonstrated by the present data. Because all patients underwent both Broström repair and suture-button stabilization, the independent contribution or necessity of syndesmotic stabilization cannot be determined.

Seven of 16 patients (43.8%) underwent endobutton removal at 1 year, confirming that secondary implant removal was clinically relevant in this small cohort. The need for removal is consistent with prior reports recognizing implant-related symptoms and selective removal after suture-button fixation[46,47], although the observed proportion was higher than the pooled 6.0% rate reported in a meta-analysis of randomized trials[47]. Direct comparison is limited by the small, selected chronic cohort and the inclusion of four elective removals. Among the three symptomatic patients, discomfort developed approximately 3 months postoperatively and persisted for more than 3 months. Activity reduction provided partial relief but was incompatible with their desired activity levels and daily functional demands, leading them to elect implant removal at 1 year. Patient No. 14 experienced immediate improvement after removal, whereas patients No. 6 and No. 8 improved during subsequent rehabilitation. Four asymptomatic patients requested removal, and no patient had persistent implant-related symptoms at 5 years. Kim et al[46] reported symptom improvement without recurrent diastasis after suture-button removal at a mean of 12.2 months, consistent with the symptomatic improvement and maintenance of radiographic alignment observed here. These findings support counseling patients regarding implant-related discomfort and selective, on-demand removal but do not support routine implant removal.

This study has several limitations. First, the sample size was small, and no a priori power calculation was performed because all eligible patients treated during the retrospective study period were included. Accordingly, the findings are exploratory and hypothesis-generating rather than confirmatory, and their generalizability is limited. Second, the absence of an isolated Broström control group precludes establishing causal treatment effects or determining the independent contribution or necessity of suture-button stabilization. Third, no calibrated stress force, predefined quantitative fluoroscopic displacement cutoff, or universally accepted arthroscopic threshold was applied. Individual stress-test grades were not recorded, and standardized still images or videos of the dynamic stress maneuvers were unavailable. The any-one-positive rule may favor sensitivity at the expense of specificity when individual test reliability is not quantified. Diagnosis therefore depended in part on qualitative judgment by the operating surgeon, introducing potential diagnostic and selection bias despite the use of a uniform protocol by a single experienced surgeon.

Fourth, the injury-to-surgery interval varied widely, and the sample size was insufficient for reliable subgroup analysis. Heterogeneity in chronicity may have influenced the observed outcomes. Fifth, imaging followed the institution’s routine standardized workflow, but exact acquisition and positioning parameters could not be retrospectively verified; contralateral imaging was unavailable, and radiographic follow-up was assessed qualitatively rather than using prespecified quantitative measurements. Sixth, the open vs arthroscopic Broström approach varied according to the presence of an os subfibulare, and preoperative conservative treatment was individualized rather than identical. Seventh, seven patients underwent implant removal, which may have influenced subsequent symptom scores. Eighth, although outcome assessors were independent of the operating surgeon and unaware of operative details, they were not blinded to follow-up time point. Ninth, the AOFAS score has recognized psychometric limitations. Inclusion of the patient-reported FAAM and Karlsson score provided complementary functional assessment, but minimal clinically important difference thresholds were not prespecified. Future multicenter controlled studies should incorporate quantitative stress-testing criteria, bilateral dynamic imaging when feasible, standardized arthroscopic documentation, and validated patient-reported responder thresholds.

CONCLUSION

A rare subgroup of patients with CLAI may exhibit occult syndesmotic instability despite the absence of definite diastasis on preoperative static imaging. In this uncontrolled retrospective case series, combined Broström repair and suture-button stabilization after failed conservative treatment was associated with resolution of focal anterior syndesmotic pain, improvement in functional scores, and return to preinjury work. These exploratory, hypothesis-generating associations support careful syndesmotic assessment in selected patients but do not establish the independent efficacy of suture-button stabilization or justify its routine use in all grade 1 HAS injuries.