Arthroscopic-Assisted Coracoclavicular Ligament Reconstruction: Clinical Outcomes and Return to Activity at Mean Six-Year Follow-up

OBJECTIVES: Over the past decade, there have been advances in arthroscopic-assisted approaches for coracoclavicular (CC) reconstruction with less surgical morbidity and enhanced visualization while also allowing for the treatment of concomitant glenohumeral pathology. Previous studies reporting outcomes using both open and arthroscopic-assisted techniques are limited by short-term follow-up and small patient populations. It also remains unclear how maintenance of reduction and clinical outcomes correlate with one another. The purpose of our study was to report clinical and functional outcomes including return to pre-injury activity level following arthroscopic-assisted CC ligament reconstruction (AA-CCR) and to determine associations between return to pre-injury activity level, radiographic outcomes and patient-reported outcomes scores following AA-CCR. We hypothesized that patients undergoing AA-CCR would have a high rate of return to pre-injury activity level, clinical outcomes would not be associated with RLOR, and that the treatment of concomitant glenohumeral pathology would not adversely affect outcomes. METHODS: A retrospective review of prospectively collected data from an institutional registry of all AA-CCR performed from January 2007-January 2016 was performed. Exclusion criteria included revision CCR, open CCR, and patients with less than two-year follow-up. Demographics and patient characteristics including sex, age at index surgery, grade of AC joint injury, duration between injury and index surgery, concomitant glenohumeral pathologies and procedures performed, complications, and subsequent surgeries were recorded. Grade of AC joint injury was determined using the Rockwood classification, and patients indicated for surgery had at least a Type III injury. Time elapsed between injury and index surgery was recorded and classified as acute (0–30 days) or chronic (> 30 days). The arthroscopic-assisted portion of the CC reconstruction has been prevously described and is as follows: The base of the coracoid then exposed either through a subacromial or intraarticular approach. Passing sutures were then placed around the coracoid for later shuttling of the soft tissue graft (allograft semitendinosus/posterior tibialis/anterior tibialis or autograft semitendinosus, according to surgeon preference) and heavy suture, which was used for ancillary fixation. Postoperative radiographs were obtained at approximately two weeks and six months following surgery. The CC distance was measured at final radiographic follow-up and compared to the unaffected contralateral side on an anteroposterior (AP) radiograph. Radiographic loss of reduction (RLOR), was defined as at least a 25% increase in CC distance as measured from the superior cortex of the coracoid process and the undersurface of the clavicle using a radiographic ruler compared to the contralateral side. Clinical assessment at final follow-up included SANE score, and additionally, patients were asked which sports(s) and/or recreational activity(s) they participated in prior to injury. For each sport or recreational activity, they were then specifically asked: “Were you able to return to the same or higher level of (specific sport or activity) as prior to your injury?” Failure of AA-CCR was defined as any one of the following: 1.) Patient underwent revision AC joint stabilization surgery, 2.) Patient was unable to return to the same or higher level of sport(s) and/or recreational activity(s) as prior to injury, 3.) Patient had RLOR as defined above. For comparative analysis, patients were characterized as having one primary mode of treatment failure. Post-hoc analysis was performed considering that patients may have more than one mode of treatment failure. RESULTS: There were 88 patients (89.8% male) with a mean age of 39.6 years (range 18-65) and minimum 2-year follow-up (mean 6.1 years, range 2.1-10.3). Follow-up rate was 67.7%. Mean time from injury to surgery was 7.2±2.4 months, with 70% chronic injuries and 63.6% grade V. Concomitant arthroscopic procedures were performed in 48.9% of cases. Overall, mean SANE score was 86.3 ± 17.5. Treatment failure occurred in 17.1%, with 8.0% unable to return to activity, 5.7% with RLOR, and 3.4% undergoing revision surgery for failed AA-CCR. Each patient undergoing revision surgery had an identifable traumatic event. All patients with RLOR were able to return to pre-injury activity level. SANE score was lower among patients who were unable to return to activity compared to those with RLOR and compared to non-failures (p=0.0002) (Table 1). Post-hoc analysis considering multiple modes of treatment failure for individual patients demonstrated that SANE score was still significantly lower among those unable to return to pre-injury activity level compared to patients with RLOR and compared to patients considered non-failures (p=0.00003). Ninety three percent of patients who participated in weightlifting, 97% who participated in swimming, and 83% of those who participated in yoga were able to return to their respective activity at the same or higher level as pre-injury at final followup. For all other sporrts, all patients returned to their pre-injury activity level. There were no differences in revision surgery rates, return to activity, or SANE scores according to the specific surgical technique used, Rockwood grade, or if concomitant pathology was treated (Table 2-4). CONCLUSIONS: AA-CCR with free tendon grafts resulted in good clinical outcomes and a high rate of return to pre-injury activity level. RLOR did not correlate with return to pre-injury activity level. Concomitant pathology that required treatment did not adversely affect outcomes. Return to pre-injury activity level may be a more clinically relevant outcome measure than radiographic maintenance of AC joint reduction.