Modified double-bundle coracoclavicular ligament reconstruction with a single coracoid tunnel and locking loop versus conventional single-bundle reconstruction for acute acromioclavicular joint dislocation: a cardinality-matched cohort study

We described a modified double-bundle coracoclavicular (CC) ligament reconstruction using a single coracoid tunnel and locking loop technique for high-grade (Rockwood type III–V) acute acromioclavicular (AC) joint dislocation. This study sought to compare clinical function and radiographic outcomes between this technique and conventional single-bundle CC ligament reconstruction. This retrospective cohort study enrolled patients with high-grade (Rockwood type III–V) acute AC joint dislocation treated at our institution between January 2018 and January 2024. A total of 92 consecutive patients were initially screened, and 68 patients were finally included in the present study according to the predefined inclusion and exclusion criteria. According to the different surgical methods of CC ligament reconstruction, the patients were divided into modified double-bundle CC ligament reconstruction with a single coracoid tunnel and locking loop technique group (Group A, n = 24) and conventional single-bundle reconstruction technique group (Group B, n = 44). Cardinality matching (CM) was used to control confounding factors. After CM matching, 22 patients were included in each group. The visual analogue scale (VAS), Constant–Murley score (CMS), shoulder range of motion (ROM), coracoclavicular distance (CCD) and clavicular tunnel width (CTW) were compared between the two groups, at 3 and 6 months postoperatively and at the final follow-up. The postoperative failure rate and complication rate were compared between the two groups. At 3 and 6 months postoperatively and at the final follow-up, both groups showed significant improvements in VAS, CMS, and shoulder ROM compared to the preoperative state (all Bonferroni-adjusted P < 0.001). At 3 months postoperatively, Group A had a lower VAS score than Group B, with an estimated adjusted between-group difference of −0.65 points (95% CI −0.98 to −0.33; Bonferroni-adjusted P = 0.001), and greater forward flexion, with an estimated adjusted between-group difference of 12.64° (95% CI 7.25–18.03; Bonferroni-adjusted P < 0.001). No statistically significant between-group differences in VAS or forward flexion were observed at the remaining follow-up time points, and no statistically significant between-group differences in CMS, abduction, external rotation, or internal rotation were observed at any follow-up time point (all Bonferroni-adjusted P > 0.05). Regarding radiographic outcomes, neither CCD nor CTW differed significantly between the groups at 1 week postoperatively. At 3 months postoperatively, CCD was lower in Group A than in Group B, with an estimated between-group difference of −0.28 mm (95% CI −0.46 to −0.10; Bonferroni-adjusted P = 0.018). At 6 months postoperatively and at the final follow-up, CTW was lower in Group A than in Group B, with estimated between-group differences of −0.49 mm (95% CI −0.81 to −0.16; Bonferroni-adjusted P = 0.023) and −0.47 mm (95% CI −0.77 to −0.18; Bonferroni-adjusted P = 0.013). Apart from these findings, no other statistically significant between-group differences in radiographic outcomes were observed (all Bonferroni-adjusted P > 0.05). In the original cohort, the surgical failure rate was 2.27% (1/44) in the conventional single-bundle reconstruction group, and no surgical failure occurred in the modified double-bundle reconstruction group. After CM to equilibrate the baseline, there were no surgical failures in the matched cohort (22 cases in each group). There was no fracture, infection, nerve injury and other related complications in the two groups. In patients with high-grade (Rockwood type III–V) acute AC joint dislocation, both modified double-bundle CC ligament reconstruction using a single coracoid tunnel and locking loop technique and conventional single-bundle reconstruction effectively restore shoulder function. At 3 months postoperatively, compared with conventional single-bundle reconstruction, the modified double-bundle technique yielded less postoperative pain, improved recovery of shoulder forward flexion, and lower CCD. At 6 months postoperatively and the final follow-up, the double-bundle group exhibited milder CTW enlargement. However, the two procedures show similar clinical outcomes in the late follow-up period. Level III, retrospective cohort study.