Clavicular tunnel position is associated with reduction loss after arthroscopic coracoclavicular ligament reconstruction: a computed tomography–based analysis Authors: Kenta Inagaki, Nobuyasu Ochiai, Yasuhito Sasaki, Seiji Ohtori, Yasuchika Aoki Published: 2027 ABSTRACT BackgroundPost-operative reduction loss remains a major concern after arthroscopic coracoclavicular ligament (CCL) reconstruction. Although bone tunnel position has been suggested as a contributing factor, most previous studies have relied on plain radiographs, which are limited in evaluating three-dimensional positioning. The purpose of this study was to evaluate clavicular and coracoid bone tunnel positions using computed tomography and to investigate their association with post-operative reduction loss.MethodsThis retrospective, single-center study included 22 patients who underwent arthroscopic CCL reconstruction for acute acromioclavicular joint dislocation or distal clavicle fracture with CCL rupture. Post-operative computed tomography with multiplanar reconstruction was used to assess the anteroposterior (AP) and mediolateral (ML) positions of the clavicular and coracoid tunnels. Reduction loss was defined as the percentage increase in coracoclavicular distance at final follow-up relative to the immediate post-operative value. Correlation analyses were performed between tunnel positions and reduction loss. Redisplacement was defined as ≥25% compared with the contralateral side, and clinical outcomes (American Shoulder and Elbow Surgeons, Constant score, and Subjective Shoulder Value) were compared between groups.ResultsThe ML position of the clavicular tunnel showed a moderate correlation with reduction loss, whereas the AP position of the clavicular tunnel and both ML and AP positions of the coracoid tunnel showed weak or no significant correlation. The overcorrection rate demonstrated a strong correlation with reduction loss, whereas the pre-operative displacement rate showed no significant association. Redisplacement occurred in 7 of 22 patients (31.8%) but was not associated with clinical outcomes.ConclusionThe ML position of clavicular tunnel plays an important role in post-operative stability following arthroscopic CCL reconstruction. In contrast, AP tunnel position and coracoid tunnel position appear to have less influence on reduction loss. Although redisplacement was not associated with clinical outcomes, its relatively high incidence (31.8%) remains a concern. Accordingly, preventing redisplacement remains important, and particular attention should be paid to the ML position of the clavicular tunnel during arthroscopic CCL reconstruction. Acromioclavicular (AC) joint dislocation and distal clavicle fractures associated with coracoclavicular ligament (CCL) rupture are common shoulder injuries. For these conditions, arthroscopic CCL reconstruction has become a widely accepted surgical technique because it provides stabilization of the AC joint or fracture site with minimal invasiveness and favorable short-term clinical outcomes.8,17 Despite advances in surgical techniques and fixation devices, post-operative reduction loss remains a major concern following CCL reconstruction. Post-operative redisplacement has been reported in up to 50% of cases in previous studies after AC joint reconstruction, potentially compromising both radiographic and clinical outcomes.3,10,13 Therefore, identifying factors associated with post-operative redisplacement is essential for improving surgical results. Several predictors of post-operative redisplacement have been reported. Clavicular tunnel placement located more medially than 25% of the total clavicular length from the lateral end has been associated with an increased risk of redisplacement.1,3 Furthermore, previous biomechanical studies have suggested that optimal tunnel positioning lies approximately within 17%-24% of the clavicular length from the lateral end, corresponding to the native attachment sites of CCL.15 In addition, improper tunnel placement in the coracoid process, particularly deviation from the central portion of the coracoid, has also been suggested as a risk factor.4,5 Furthermore, previous studies have reported that the degree of over-reduction and the magnitude of pre-operative displacement may influence post-operative redisplacement.9,14 However, most previous evaluations of tunnel position have relied on plain radiographs. These are limited in accurately assessing three-dimensional positioning, particularly in the anterior–posterior direction. To date, few studies have evaluated the relationship between tunnel positioning and post-operative redisplacement using computed tomography (CT), which enables precise multiplanar reconstruction (MPR) assessment of both clavicular and coracoid tunnels, including their anterior–posterior orientation. The purpose of this study was to evaluate the post-operative tunnel positions in the clavicle and coracoid process using CT following arthroscopic CCL reconstruction, and to investigate their association with the degree of reduction loss. In addition, we assessed whether the degree of over-reduction and the magnitude of pre-operative displacement were associated with redisplacement and whether redisplacement affected clinical outcomes. Materials and methods This retrospective, single-center observational study was conducted at our institution and approved by the institutional review board. Study information was publicly disclosed on the institutional website, and patients who opted out were excluded from the analysis, in accordance with the opt-out consent process. Patients with acute AC joint dislocation or distal clavicle fracture with CCL rupture who underwent arthroscopic CCL reconstruction within 3 weeks of injury were included. Patients with chronic injuries, prior shoulder surgery, or insufficient imaging data were excluded. A total of 22 patients were included in this study. The study population consisted of 15 patients with AC joint dislocation and 7 patients with distal clavicle fractures classified as Neer type II,12 in which the distal clavicular fragment measured ≤10 mm. In cases where the distal fragment measured ≤10 mm, plate fixation was deemed technically difficult due to insufficient bone stock. Both AC joint dislocation and distal clavicle fracture with CCL rupture result in loss of vertical stability of the clavicle due to CCL insufficiency. All patients were followed for a minimum of 12 months post-operatively. The cohort comprised 21 men and 1 woman, with a mean age of 44.4 years (range, 23 to 71 years). The mean follow-up period was 16.2 ± 6.1 months. Patient demographics were comparable between the AC joint dislocation group and the distal clavicle fracture group (Table I).Table IPatient demographics acromioclavicular joint dislocation and distal clavicle fracture.DemographicsAC joint dislocation (n = 15)Distal clavicle fracture (n = 7)P valueAge (mean ± SD, y)40.1 ± 17.553.6 ± 18.0.110Sex (male/female, n)15/06/1.318Right/left (n)10/52/5.172Follow-up (mean ± SD, m)15.6 ± 6.317.6 ± 6.1.500Time from injury to surgery (mean ± SD, d)12.9 ± 5.910.6 ± 4.1.138AC, acromioclavicular; SD, standard deviation.The mean age, follow-up, and time from injury to surgery were analyzed using Student t-test, and sex and side were analyzed using Fisher exact test. Table IPatient demographics acromioclavicular joint dislocation and distal clavicle fracture.DemographicsAC joint dislocation (n = 15)Distal clavicle fracture (n = 7)P valueAge (mean ± SD, y)40.1 ± 17.553.6 ± 18.0.110Sex (male/female, n)15/06/1.318Right/left (n)10/52/5.172Follow-up (mean ± SD, m)15.6 ± 6.317.6 ± 6.1.500Time from injury to surgery (mean ± SD, d)12.9 ± 5.910.6 ± 4.1.138AC, acromioclavicular; SD, standard deviation.The mean age, follow-up, and time from injury to surgery were analyzed using Student t-test, and sex and side were analyzed using Fisher exact test. Patient demographics acromioclavicular joint dislocation and distal clavicle fracture. AC, acromioclavicular; SD, standard deviation. The mean age, follow-up, and time from injury to surgery were analyzed using Student t-test, and sex and side were analyzed using Fisher exact test. Surgical technique All procedures were performed under general anesthesia with the patient in the beach chair position, using a combination of arthroscopy and fluoroscopic guidance. After reduction of the dislocation or fracture, temporary fixation was achieved using a 2.0-mm Kirschner wire inserted transacromially. Arthroscopic inspection of the glenohumeral joint was then performed; the rotator interval was debrided and the inferior surface of the coracoid process was exposed. Subsequently, subacromial arthroscopy was carried out to identify the inferior border and base of the coracoid process. A 2-cm skin incision was made directly over the clavicle, and a bone tunnel was created from the clavicle to the base of the coracoid process using a dedicated guide, with the tunnel position determined based on the relationship between the clavicle and coracoid process to achieve appropriate fitting of the guide. CCL reconstruction was performed using a single-tunnel technique. Fixation was achieved using a low-profile AC repair system (Arthrex, Naples, FL, USA), consisting of 2 independent tensionable TightRope sutures and a Dog Bone button. After provisional tightening of the synthetic ligament, the Kirschner wire was removed, followed by final tightening of the implant. The remaining portion of the synthetic ligament was then passed arthroscopically through tunnels created in the central and anterior portions of the acromion and secured to provide additional stabilization.16 Post-operative rehabilitation Post-operatively, the operated shoulder was immobilized in a sling for 3 weeks. From 1 week after surgery, passive range-of-motion exercises were initiated with unrestricted internal and external rotation, while forward elevation was limited to 90°. Active range-of-motion exercises were started at 3 weeks post-operatively under the same range-of-motion restrictions. From 6 weeks after surgery, forward elevation was allowed without restriction. Weight-bearing activities, heavy labor, and return to sports were permitted at 3 months post-operatively. Computed tomography evaluation of bone tunnel positions Post-operative CT was performed within 1 month after surgery to evaluate the positions of bone tunnels in the clavicle and coracoid process. MPR images were generated to align the coracoid process and clavicle along their anatomic axes with a picture archiving and communication system (Centricity, GE Healthcare, Chicago, IL, USA). The anteroposterior (AP) and mediolateral (ML) positions of each bone tunnel were assessed using standardized measurement methods. Axial CT images were reconstructed, and an image in which the entire length of the clavicle and the most cranial aspect of the tunnel were visualized in the same plane was selected. The total length of the clavicle and the distance from the lateral end of the clavicle to the center of the tunnel were measured on this image (Fig. 1, A).Figure 1(a) Reconstructed axial CT image. The solid white arrow indicates the total length of the clavicle, and the dotted arrow indicates the distance from the lateral edge of the clavicle to the Center of the bone tunnel. (b) Reconstructed axial CT image. The solid black arrow indicates the anteroposterior width of the clavicle, and the dotted arrow indicates the distance from the posterior cortex of the clavicle to the Center of the bone tunnel. CT, computed tomography. (a) Reconstructed axial CT image. The solid white arrow indicates the total length of the clavicle, and the dotted arrow indicates the distance from the lateral edge of the clavicle to the Center of the bone tunnel. (b) Reconstructed axial CT image. The solid black arrow indicates the anteroposterior width of the clavicle, and the dotted arrow indicates the distance from the posterior cortex of the clavicle to the Center of the bone tunnel. CT, computed tomography. The ML position of the clavicular tunnel was calculated as follows: (Distance from the lateral end of the clavicle to the center of the tunnel [mm]/total clavicular length [mm]) × 100 (%). Previous anatomic studies have shown that the clavicular insertions of the CCL are located at approximately 17–24% of the clavicular length from the lateral end.15 Therefore, 20% of the clavicular length was used as the reference value, and the absolute deviation from this value was calculated for analysis. In addition, axial CT images perpendicular to the clavicular bone tunnel were reconstructed, and the AP position of the bone tunnel was calculated as follows: (Distance from the posterior cortex of the clavicle to the center of the tunnel [mm]/AP width of the clavicle [mm]) × 100 (%) (Fig. 1, B). On coronal CT images, an image perpendicular to the coracoid tunnel was reconstructed, and the ML position of the tunnel was calculated as follows: (Distance from the lateral border of the coracoid process to the center of the tunnel [mm]/coracoid width [mm]) × 100 (%) (Fig. 2, A).Figure 2(a) Reconstructed coronal CT image. The solid white arrow indicates the width of the coracoid process, and the dotted arrow indicates the distance from the lateral edge of the coracoid process to the Center of the bone tunnel. (b) Reconstructed sagittal CT image. The solid white arrow indicates the distance from the anterior margin to the base of the coracoid process, and the dotted arrow indicates the distance from the base of the coracoid process to the Center of the bone tunnel. CT, computed tomography. (a) Reconstructed coronal CT image. The solid white arrow indicates the width of the coracoid process, and the dotted arrow indicates the distance from the lateral edge of the coracoid process to the Center of the bone tunnel. (b) Reconstructed sagittal CT image. The solid white arrow indicates the distance from the anterior margin to the base of the coracoid process, and the dotted arrow indicates the distance from the base of the coracoid process to the Center of the bone tunnel. CT, computed tomography. Based on previous studies identifying 50% as the optimal location, the absolute deviation from this 50% reference value was calculated and used for analysis.19 In addition, on sagittal CT images, an image in which the tip of the coracoid process, the most caudal aspect of the tunnel, and the base of the coracoid process were visualized in the same plane was reconstructed. The AP position of the coracoid tunnel was calculated as follows. (Distance from the base of the coracoid process to the center of the tunnel [mm]/distance from the tip to the base of the coracoid process [mm]) × 100 (%) (Fig. 2, B). Radiographic evaluation of reduction loss AC joint AP plain radiographs were obtained in the standing position pre-operatively, immediately post-operatively, and at the final follow-up. The degree of post-operative reduction loss was assessed by comparing the coracoclavicular (CC) distance measured immediately after surgery and at the final follow-up using X-ray (Fig. 3). The rate of reduction loss was calculated as follows:Figure 3Anteroposterior plain radiograph. The solid white arrow indicates the coracoclavicular distance. Anteroposterior plain radiograph. The solid white arrow indicates the coracoclavicular distance. ((Final follow-up CC distance [mm] − immediate post-operative CC distance [mm])/immediate post-operative CC distance [mm]) × 100 (%). Correlation analyses were performed between the tunnel position parameters and the rate of reduction loss. Additionally, the relationships between overcorrection rate, pre-operative displacement rate, and rate of reduction loss were analyzed using the following formulas: The overcorrection rate: ((contralateral CC distance [mm] − immediate post-operative CC distance [mm])/contralateral CC distance [mm]) × 100 (%). The pre-operative displacement rate: (pre-operative CC distance [mm]/contralateral CC distance [mm]) × 100 (%). Group classification Patients were categorized into 2 groups based on post-operative redisplacement at final follow-up. Redisplacement of ≥25% compared with the contralateral side was defined as significant redisplacement.11 Clinical outcomes, including the American Shoulder and Elbow Surgeons (ASES) score, the Constant score, and the Subjective Shoulder Value (SSV), were compared between patients with and without redisplacement. Statistical analysis Statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R. Correlations between continuous variables were assessed using Pearson correlation coefficient. Differences in continuous variables between the 2 groups were analyzed using the Mann–Whitney U test. A P value of <.05 was considered statistically significant. Results The correlation between tunnel position and the rate of reduction loss is shown in Table II, and the corresponding scatter plots are shown in Figure 4. The ML position of the clavicular tunnel showed a moderate correlation (r = 0.503). The AP position of the coracoid tunnel showed weak correlations, whereas the AP position of the clavicular tunnel and ML position of the coracoid tunnel showed no correlation.Table IICorrelation between bone tunnel position and the rate of reduction loss.Bone tunnelr95% CIP valueClavicle ML0.5030.104-0.763.017 AP−0.053−0.464 to 0.377.816Coracoid ML0.100−0.336 to 0.501.658 AP0.222−0.22 to 0.589.320CI, confidence interval; ML, mediolateral; AP, anteroposterior.Figure 4(a) Scatter plot showing the correlation between the ML position of the clavicular tunnel and the rate of reduction loss. (b) Scatter plot showing the correlation between the AP position of the clavicular tunnel and the rate of reduction loss. (c) Scatter plot showing the correlation between the ML position of the coracoid tunnel and the rate of reduction loss. (d) Scatter plot showing the correlation between the AP position of the coracoid tunnel and the rate of reduction loss. ML, mediolateral; AP, anteroposterior. Table IICorrelation between bone tunnel position and the rate of reduction loss.Bone tunnelr95% CIP valueClavicle ML0.5030.104-0.763.017 AP−0.053−0.464 to 0.377.816Coracoid ML0.100−0.336 to 0.501.658 AP0.222−0.22 to 0.589.320CI, confidence interval; ML, mediolateral; AP, anteroposterior. Correlation between bone tunnel position and the rate of reduction loss. CI, confidence interval; ML, mediolateral; AP, anteroposterior. (a) Scatter plot showing the correlation between the ML position of the clavicular tunnel and the rate of reduction loss. (b) Scatter plot showing the correlation between the AP position of the clavicular tunnel and the rate of reduction loss. (c) Scatter plot showing the correlation between the ML position of the coracoid tunnel and the rate of reduction loss. (d) Scatter plot showing the correlation between the AP position of the coracoid tunnel and the rate of reduction loss. ML, mediolateral; AP, anteroposterior. The correlation between the overcorrection rate, pre-operative displacement rate, and rate of reduction loss are shown in Table III. The overcorrection rate showed a strong correlation (r = 0.711), whereas the pre-operative displacement rate showed no correlation. The mean overcorrection rate was 21.8% in the AC joint dislocation group and 13.7% in the distal clavicle fracture group, with no significant difference between the groups (P = .180).Table IIICorrelation between the overcorrection rate, pre-operative displacement rate, and the rate of reduction loss.Radiographic parametersr95% CIP valueOvercorrection rate0.7110.414-0.871<.001Pre-operative displacement rate0.124−0.314 to 0.519.582CI, confidence interval. Table IIICorrelation between the overcorrection rate, pre-operative displacement rate, and the rate of reduction loss.Radiographic parametersr95% CIP valueOvercorrection rate0.7110.414-0.871<.001Pre-operative displacement rate0.124−0.314 to 0.519.582CI, confidence interval. Correlation between the overcorrection rate, pre-operative displacement rate, and the rate of reduction loss. CI, confidence interval. Redisplacement was observed in 7 of 22 cases (31.8%), comprising 5 cases of AC joint dislocation and 2 cases of distal clavicle fracture. There were no significant differences in ASES score, Constant score, or SSV between patients with and without redisplacement (Table IV).Table IVComparison between with and without redisplacement.OutcomesRedisplacement (n = 7)Without redisplacement (n = 15)P valueASES score (mean ± SD)91.7 ± 13.196.3 ± 3.9.534Constant score (mean ± SD)93.8 ± 9.793.4 ± 5.6.524SSV (mean ± SD)89.0 ± 8.991.0 ± 8.0.682SD, standard deviation; ASES, American Shoulder and Elbow Surgeons; SSV, Subjective Shoulder Value. Table IVComparison between with and without redisplacement.OutcomesRedisplacement (n = 7)Without redisplacement (n = 15)P valueASES score (mean ± SD)91.7 ± 13.196.3 ± 3.9.534Constant score (mean ± SD)93.8 ± 9.793.4 ± 5.6.524SSV (mean ± SD)89.0 ± 8.991.0 ± 8.0.682SD, standard deviation; ASES, American Shoulder and Elbow Surgeons; SSV, Subjective Shoulder Value. Comparison between with and without redisplacement. SD, standard deviation; ASES, American Shoulder and Elbow Surgeons; SSV, Subjective Shoulder Value. Discussion To the best of our knowledge, this is the first study to use CT-based MPR to evaluate tunnel position in both the clavicle and coracoid and to investigate its relationship with reduction loss. The ML position of the clavicular tunnel showed a moderate correlation and was the strongest among the evaluated parameters. This finding is consistent with previous reports suggesting that non-anatomic placement of the clavicular tunnel, particularly medial malposition, increases the risk of reduction loss.1,15 In this study, the optimal ML position of the clavicular tunnel was defined as 20% from the lateral edge of the clavicle, based on previous reports,1,3,15 and analysis was performed using the absolute distance from this reference point. Our results demonstrated that greater deviation of the clavicular tunnel from the 20% position was associated with greater reduction loss. These findings suggest that careful attention to the ML position of the clavicular tunnel is important to minimize reduction loss following arthroscopic CCL reconstruction. Regarding the ML position of the coracoid tunnel, deviation from the central portion of the coracoid process, which is considered the optimal position19 was not associated with a significant increase in the reduction loss. This may be because coracoid tunnel placement is more anatomically and technically constrained, resulting in less variability than clavicular tunnel position. However, excessive lateral or medial placement of the coracoid tunnel may increase the risk of cutout. Therefore, careful attention should still be paid to appropriate tunnel position. Although AP position could be accurately assessed using CT, no significant association with reduction loss was observed, suggesting that ML positioning may play a more critical role in post-operative stability. Thus, with regard to the AP position, it may be reasonable to follow conventional anatomic landmarks, such as the center of the clavicle and the base of the coracoid process, when creating tunnels.6 The overcorrection rate showed a strong correlation with the reduction loss. However, previous study has reported that approximately 50% overcorrection relative to the contralateral side can result in a final anatomic position with favorable clinical outcomes.7 Thus, it should be noted that overcorrection does not necessarily lead to poor outcomes. In the same report,7 no cases remained in an overcorrected state at final follow-up, even when overcorrection was intentionally performed. Our findings are consistent with these observations, suggesting that even when overcorrection is applied, reduction loss may occur over time, ultimately resulting in an appropriate anatomic position. In the present study, the pre-operative displacement rate showed no association with the reduction loss. Olivos-Meza et al13 reported no significant difference in the rate of redisplacement between Rockwood type III and type V injuries, which is consistent with our findings. Redisplacement after arthroscopic CCL reconstruction has been defined variably in the literature, and its relationship with clinical outcomes remains controversial.2,4,11 In the present study, redisplacement was defined relatively strictly as a ≥25% increase compared with the contralateral side, and clinical outcomes were compared accordingly. Redisplacement was observed in 7 of 22 cases (31.8%), and no significant differences were found in ASES score, Constant score, or SSV between patients with and without redisplacement. However, clinical outcomes were favorable in both groups, suggesting a potential ceiling effect, which should be taken into consideration when interpreting these results. Several surgical techniques have been reported for CCL reconstruction, including mini-open procedures and techniques using double clavicular tunnels.2,1819 These techniques may have different advantages in terms of surgical exposure and fixation configuration. Further comparative studies are warranted to evaluate the differences among these techniques. This study has several limitations. First, the relatively small sample size may have limited the statistical power of the study. In addition, the lack of adjustment for multiple comparisons may have increased the risk of type I error. Therefore, the results of the group comparison should be interpreted with caution. Second, both AC joint dislocation and distal clavicle fracture were included. However, distal clavicle fractures were limited to cases with concomitant CCL rupture and were therefore considered to have a comparable pathophysiology. Nevertheless, the healing process of the fracture may have influenced the stability of the reconstructed CCL and cannot be completely excluded. Third, this was a retrospective study, which may have introduced selection bias and limited the ability to control for potential confounding variables. Fourth, the tension applied to the TightRope was not objectively quantified, which may have limited the reproducibility of the surgical procedure. Finally, post-operative immobilization was limited to 3 weeks in our protocol. Although this duration is within the range reported in previous studies, the potential effect of a longer immobilization period on maintenance of reduction cannot be excluded. Conclusion The ML position of clavicular tunnel plays an important role in post-operative stability following arthroscopic CCL reconstruction. In contrast, AP tunnel position and coracoid tunnel position appear to have less influence on reduction loss. Although redisplacement was not associated with clinical outcomes, its relatively high incidence (31.8%) remains a concern. Accordingly, preventing redisplacement remains important, and particular attention should be paid to the ML position of the clavicular tunnel during arthroscopic CCL reconstruction. Disclaimers: Funding: This study was supported by the Goudou Shigen Academic Activity Support Fund. Conflicts of interest: The authors, their immediate family, and any research foundation with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.