Minimum 10-year outcomes after arthroscopically assisted anatomic coracoclavicular ligament reconstruction for the treatment of type III and V acromioclavicular joint injuries

Abstract

Background

Acromioclavicular joint (ACJ) injuries can lead to severe pain and reduced shoulder function. Anatomic coracoclavicular ligament reconstruction (ACCR) using a free tendon graft has demonstrated favorable biomechanical properties and good short- to mid-term outcomes, but data on long-term outcomes are scarce. The purpose of the present study was to evaluate the long-term clinical and functional outcomes after ACCR with a focus on return to sport and revision surgery.

Methods

Patients who underwent ACCR using a free tendon allograft for the treatment of ACJ injuries type III or V between November 2006 and April 2011 by a single surgeon, with a minimum 10-year follow-up, were eligible for inclusion. Patient-reported outcome measures, including the American Shoulder and Elbow Surgeons score, short version of the Disabilities of the Arm, Shoulder and Hand questionnaire, Single Assessment Numeric Evaluation, patient satisfaction (1-10 scale with “10” indicating maximum satisfaction), rates of return to sport and revision surgery were evaluated minimum 10 years postoperatively. Pain (via visual analog scale) was assessed preoperatively and at final follow-up. Patients who underwent further surgery were excluded from analysis. Survivorship was reported as two scenarios: not undergoing revision ACJ stabilization and not undergoing any further surgery.

Results

Fourteen patients were evaluated after 13.0 (interquartile range, 12.0-15.0) years. Four patients (28.6%) underwent further surgery (revision ACJ stabilization [n = 2], exostosis removal [n = 2]) and were excluded from further analysis. Shoulder function was excellent at follow-up (American Shoulder and Elbow Surgeons score: 100 [96.6-100], Quick Disabilities of the Arm, Shoulder and Hand: 0 [0-5.1]), Single Assessment Numeric Evaluation: 99.0 [91.5-99.0]). Pain levels decreased significantly from preoperatively to follow-up (visual analog scale for pain: 2.0 [1.0-3.0] to 0 [0-0.5], P = .016). Satisfaction at follow-up was high (10 [5.0-10]). Nine patients reported their postoperative sporting level, of which eight returned to sport (88.9%). Four patients (44.4%) returned equal to or above the preoperative level and 4 patients (44.4%) returned below their preoperative level. One patient reported hypersensitivity and anterior shoulder pain postoperatively but reported low pain levels at follow-up. Survivorship was 85.7% for not undergoing revision ACJ stabilization and 71.4% for not undergoing any further surgery.

Conclusion

ACCR using a free tendon graft for the treatment of ACJ injuries type III or V resulted in good to excellent shoulder function, low pain levels, and high patient satisfaction at long-term follow-up. The rate of revision ACJ stabilization was 14.3%.

There exists a general consensus for nonoperative treatment of Rockwood type I and II injuries, initial nonoperative treatment of Rockwood type III injuries, and operative treatment in Rockwood type IV-VI injuries.^2,3,32 Nonetheless, early surgical stabilization of type III injuries may be considered in overhead athletes, workers, patients with a wish for surgical stabilization, and those with concomitant pathologies.^{10,12,32,34,36} Acromioclavicular joint (ACJ) reconstruction using a free tendon graft has been associated with good to excellent functional outcomes and low revision rates.²² Mazzocca et al¹⁸ proposed an anatomic coracoclavicular ligament reconstruction (ACCR) that utilizes a semitendinosus tendon autograft that is passed underneath the coracoid and secured in two clavicular bone tunnels using interference screws. Since then, ACCR has advanced to an arthroscopically assisted technique.²¹ Biomechanically, ACCR has been shown to better approximate the stability of the native ACJ than nonanatomic techniques, such as the modified Weaver–Dunn procedure.^11,16,19 Similarly, clinical studies have shown good to excellent outcomes, predominantly at short-^1,7,17,20,26 and mid-term²⁸ follow-up. Data reporting on long-term outcomes following ACCR in patients with Rockwood type III or V injuries are yet scarce.

The purpose of the present study was to report the clinical and functional outcomes and return to sport rates in patients who underwent arthroscopically assisted ACCR after a minimum follow-up of 10 years. It was hypothesized that the functional outcomes would be good to excellent and that return to sport rates would be high.

Methods

The present study was a retrospective review of prospectively collected data that was conducted with approval from the Vail Health Hospital institutional review board (reference: 2023-220-SPRI). Consecutive patients who underwent primary arthroscopically assisted ACCR using an allograft between November 2006 and April 2011 for the treatment of an ACJ injury Rockwood type III/V by the senior author (P.J.M.) were eligible for inclusion (minimum follow-up: 10 years).

Following thorough clinical and radiographic evaluation by the senior author (P.J.M.), patients were counseled regarding their treatment options. Acute surgical treatment was recommended for Rockwood type V injuries and for Rockwood type III injuries in patients with an active lifestyle or those with concomitant injuries. Chronic injuries (≥3 weeks)²⁹ were treated surgically if the patients had ACJ-related complaints despite a 4- to 6-week period of nonoperative treatment. Patients with concomitant fractures or a history of previous shoulder surgery were excluded. The outcomes of several patients included in the present study were previously assessed and reported at short-²⁰ and mid-term²⁸ follow-up.

Surgical technique

The arthroscopically assisted ACCR was performed with the patient in the beach-chair position. After diagnostic arthroscopy, a laterally based rotator interval working portal was established to address intra-articular pathologies and expose the inferior surface of the coracoid. The senior author then switched to the 70-degree arthroscope to adequately visualize the coracoid from the posterior portal. An accessory anterolateral portal was made approximately 5 cm distal to the anterolateral border of the acromion. This portal entered just superior to the subscapularis tendon. An 8.25-mm cannula was then placed through this portal to facilitate allograft passage.

Once the coracoid was exposed arthroscopically and accessory portals were established, a 2-cm incision was made 25 mm medial to the ACJ. The deltotrapezial fascia was incised perpendicular to the skin for later repair and imbrication of the ACJ capsule and its associated ligaments following coracoclavicular ligament reconstruction. The distal clavicle and ACJ were exposed. The distal clavicle was then drilled (either 2 drill holes for anatomic reconstruction of the conoid and trapezoid ligaments or 1 drill hole between the two ligament complexes) 6-7 mm in diameter to allow for graft passage. An allograft (typically tibialis anterior) was sized to 7 mm, whipstitched on both ends, and passed around the coracoid, with both limbs passed through the drill hole(s) in the distal clavicle. After tensioning and cycling the graft, the graft was fixed to the distal clavicle with a 5.5-mm PEEK interference screw to prevent slippage of the graft. The graft was also tied to itself with multiple figure-of-eight sutures using #2 permanent sutures (Ethibond; Ethicon, Raritan, NJ, USA). Throughout the study period, a 9-strand #1 polydioxanone suture cable was occasionally utilized (5 patients) and passed around the coracoid process to provide additional internal fixation. After copious irrigation, the ACJ capsule was meticulously closed with either #2 nonabsorbable or #1 absorbable sutures, see Figure 1. The deltotrapezial fascia was closed with #1 absorbable sutures, and the remainder of the incision was closed in layers.

Figure 1.

Anatomic coracoclavicular ligament reconstruction

Postoperative rehabilitation

After surgery, patients were placed into an abduction sling for 6 weeks. Supine passive range of motion exercises are allowed immediately after surgery. This was followed by upright passive range of motion beginning at 2 weeks. Active and active-assisted motion exercises were then started at 4 weeks. Strengthening exercises are initiated at 8 weeks. Return to full activities was typically accomplished by 4 months.

Patient characteristics and operative data

Chart review was performed to obtain data regarding patient characteristics, including age at the time of surgery, sex, laterality of the injury, injury mechanism, and operative data, as well as intra- and postoperative complications.

Outcome measurements

Patient-reported outcome measures, including the American Shoulder and Elbow Surgeons score, short version of the Disabilities of the Arm, Shoulder and Hand questionnaire, Single Assessment Numeric Evaluation, 12-Item Short Form Survey Physical Component Summary (PCS-12), Visual Analog Scale (VAS) for pain and subjective satisfaction (1-10 scale, with 10 indicating maximum satisfaction) were collected at long-term follow-up (minimum 10 years). In addition, PCS-12 and VAS for pain were collected preoperatively. Patients received optional return to sports questionnaires that evaluated the postoperative return to sporting activities, modifications in activity, and reasons for doing so both at short- and long-term follow-up. Lastly, rates of further ACJ-related surgery and other complications were recorded. Survivorship was reported as two scenarios: not undergoing revision ACJ stabilization and not undergoing any further surgery.

Statistical analysis

Data were analyzed using SPSS 28.0 (IBM Corp., Armonk, NY, USA). Categorical variables are reported as counts and corresponding percentages. Normal distribution of the collected continuous variables was assessed using the Shapiro–Wilk test. Normally distributed continuous variables are reported as mean ± standard deviation, whereas non–normally distributed continuous variables are reported as median (25%-75% interquartile range). For group comparisons of continuous variables, the t-test or Wilcoxon signed rank test was applied. Statistical significance was set at a P value of < .05. Due to the rarity of the injury, all eligible patients were considered, and a power analysis was not performed.

Results

The senior author performed arthroscopically assisted ACCR in 27 patients between November 2006 and April 2011, of which 18 were eligible for study inclusion, see Figure 2. Fourteen patients (77.8%) were available for follow-up. Four of these patients (28.6%) were excluded from all analyses due to undergoing revision surgery. Two patients (14.3%) underwent revision ACJ reconstruction 13 days and 13 months postoperatively, and exostosis removal due to tenderness over the distal clavicle 3 and 5 months postoperatively, respectively.

Figure 2.

Flowchart of patient inclusion. ACCR, anatomic coracoclavicular ligament reconstruction; ACJ, acromioclavicular joint.

The remaining 10 patients were followed up after 13.0 (12.0-15.0) years, for which preoperative VAS for pain and PCS-12 were available for 8 patients (80.0%). All patients were male and 48.3 ± 16.0 years old at the time of surgery, see Table I. The mechanism of injury was a fall in 9 patients (90.0%) and contact with an opponent during touch football in 1 patient (10.0%). Half of the patients underwent surgery following acute injury. Half of the patients suffered a Rockwood type III injury. All patients underwent concomitant partial claviculectomy. Superior labral anterior to posterior lesions were found in 7 patients (70.0%) and were treated with superior labral anterior to posterior débridement (n = 6; 60.0%) or subpectoral tenodesis of the long head of the biceps tendon (n = 1; 10.0%). Three patients (30.0%) underwent rotator cuff débridement for a small articular-sided partial-thickness supraspinatus tendon tear (n = 2; 20.0%) and an upper-border subscapularis tendon tear (n = 1; 10.0%).

Table I.

Patient characteristics.^∗

	Total study cohort
Number of patients (shoulder), n	10 (10)
Age at surgery, year	48.3 ± 16.0
Male sex, n (%)	10 (100%)
Laterality, right:left, n (%)	3:7 (70.0% left)
Follow-up, year	13.0 (12.0-15.0)

^∗Continuous data reported as mean ± standard deviation or median (interquartile range). Categorical data are reported as n (%).

Outcome assessment

Shoulder function at follow-up was excellent (American Shoulder and Elbow Surgeons score: 100 [96.6-100], Quick Disabilities of the Arm, Shoulder and Hand: 0 [0-5.1]), Single Assessment Numeric Evaluation: 99.0 [91.5-99.0]). Pain levels decreased significantly from preoperatively to follow-up (VAS for pain: 2.0 [1.0-3.0] to 0 [0-0.5], P = .016). Physical health did not change significantly (PCS-12: 42.8 [42.2-50.7] to 56.7 [36.0-57.9), P = .237).) Satisfaction at follow-up was high (10 [5.0-10]).

Nine patients reported their postoperative sporting level, of which 4 patients (44.4%) returned equal to or above the preoperative level and 4 patients (44.4%) returned below their preoperative level. One patient (11.1%) reported being unable to participate in any sport. Five patients (55.5%) modified their activities. Most frequently, lifestyle changes were reported as the reason for doing so (n = 4; 80.0%), but pain (n = 1; 20.0%), weakness (n = 1; 20.0%), fear of reinjury (n = 1; 20.0%), and fear of further surgery (n = 1; 20.0%) were also reported.

One patient reported hypersensitivity and anterior shoulder pain postoperatively but reported low pain levels (VAS for pain = 2) at follow-up. The complication rate was 35.7% (further surgery [n = 4], hypersensitivity [n = 1]). Survivorship was 85.7% for not undergoing revision ACJ stabilization and 71.4% for not undergoing any further surgery.

Discussion

The most important finding of the present study was that arthroscopically assisted ACCR with a free tendon allograft was associated with good to excellent shoulder function and high return to sport rates at long-term follow-up. Pain levels decreased significantly. Rates of further surgery, specifically revision ACJ reconstruction, were 28.6% and 14.3%, respectively.

Previous studies reported good to excellent short-^1,7,17,20,26 and mid-term^25,28 outcomes after open and/or arthroscopically assisted ACCR for the treatment of ACJ injuries. Studies on long-term outcomes are yet scarce.^4,24 It should, however, be noted that in patients who do not need to or do not wish to undergo surgery immediately, an initial trial of nonoperative treatment may also be feasible, as successful nonoperative treatment has been shown to lead to comparable results as ACCR for chronic injury.²⁵

Besides primary injuries, ACCR may also be performed in the revision setting following failed ACJ reconstruction.^4,6,8 Berthold et al⁴ reported good to excellent outcomes after ACCR for failed ACJ reconstruction at long-term follow-up. Their results were comparable to those of Kraus et al¹⁵ who performed an arthroscopically assisted acromio- and coracoclavicular reconstruction using a gracilis tendon autograft and suspensory fixation, and those of Tauber et al³¹ who utilized an arthroscopically assisted coracoclavicular ligament reconstruction with a semitendinosus tendon autograft and temporary coracoclavicular cerclage or Bosworth screw.

Few other studies evaluated mid- to long-term outcomes following ACJ stabilization using different techniques, including suspensory fixation,^14,23,35 hook plate fixation,¹⁴ and Ligament Augmentation Reconstruction System.^13,33 Ko et al¹⁴ reported no difference in patient-reported outcome measures between patients who underwent ACJ reconstruction using a suspensory fixation vs. hook plate fixation after a mean follow-up of 7.0 years. Patients who underwent suspensory fixation, however, had significantly better forward flexion at follow-up. Further, subacromial erosion was observed in 41.7% of patients following hook plate fixation, but this did not affect pain at follow-up. Good to excellent outcomes at mid-term follow-up following suspensory fixation were also reported by Venjakob et al³⁵ who utilized an arthroscopically assisted double-bundle reconstruction of the coracoclavicular ligaments for acute ACJ injuries, and Motta et al²³ who augmented an arthroscopically assisted single-tunnel suspensory fixation with an allograft in chronic injuries. Notably, ossifications in projection of the coracoclavicular ligaments were frequently observed in both studies.^23,35 This has been identified as a potentially prognostic factor for superior outcomes⁹ but could not be verified in the current study. Tiefenboeck et al³³ and Hunter et al¹³ reported mid- to long-term outcomes following ACJ stabilization using Ligament Augmentation Reconstruction System. Complications occurred in 10.6% and 20.6%, respectively.^13,33

Although functional outcomes were favorable in the current study, the rate of revision ACJ stabilization was high at 14.3%. Several modifications have been made to the surgical technique that was utilized in the present study as to reduce the risk for revision ACJ stabilization. These include looping the graft around the clavicle as to avoid drilling large holes, adding a knotless suspensory fixation system with suture tape that only necessitates small drill holes (2.4 mm), and, being knotless, decrease the risk of soft tissue irritation around the clavicle, which has been reported to occur following ACJ stabilization.³⁰

Some limitations should be noted when interpreting the findings of the present study. First, the number of included patients is low overall. Further, the study lacked a comparison group, such as patients that underwent nonoperative treatment, which has been shown to lead to outcomes comparable with ACJ reconstruction.^25,27,32 Finally, radiographs were not performed at follow-up and thus, several important factors such as, tunnel widening,⁵ loss of reduction,⁵ or the onset of progression of ACJ arthritis could not be evaluated.

Conclusion

Anatomic coracoclavicular ligament reconstruction using a free tendon graft for the treatment of ACJ injuries type III or V resulted in good to excellent shoulder function, low pain levels, and high patient satisfaction at long-term follow-up. The rate of revision ACJ stabilization was 14.3%.

Disclaimers:

Funding: No author received direct financial remunerations related to the subject of this article.

Conflicts of interest: Maximilian Hinz, MD; Ayham Jaber, MD; Christopher J. Hawryluk, MBS; Marilee P. Horan, MPH and Amelia Drumm, BA’s; positions are supported by the Steadman Philippon Research Institute, which is a 501(c) (3) nonprofit institution supported financially by private donations and corporate support. The Steadman Philippon Research Institute (SPRI) exercises special care to identify any financial interests or relationships related to research conducted here. During the past calendar year, SPRI has received grant funding or in-kind donations from Arthrex, Canon, DJO, Icarus Medical, Medtronic, Ossur, Smith&Nephew, SubioMed, and Stryker & Wright Medical. Peter J. Millett, MD, MSc receives royalties, consultant payments, and research support from Arthrex which is not related to the subject of this work. Dr. Millett also owns stock options with VuMedi. The other author, his immediate family, and any research foundation with which he is affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.