Instability of the sternoclavicular (SC) joint is an uncommon shoulder pathology (accounting for 3% of shoulder girdle injuries) that can cause significant morbidity due to the proximity of important neurovascular structures.1–3 This injury is associated with pain, mechanical symptoms at the joint, tenderness to palpation, swelling, and recurrent episodes of instability despite appropriate management. Posterior dislocations, although more rare than anterior dislocations, have additional concerns of thoracic outlet syndrome, vascular compromise, and injury to the hilar structures.2 Open reduction and stabilization of the SC joint is indicated for patients with severe symptoms who have failed conservative and closed management.
The SC joint is a rigid structure composed of tendinous insertions of the sternocleidomastoid, subclavius, sternothyroid, and sternohyoid muscles as well as ligamentous attachments of the costoclavicular, posterior SC, anterior SC, and interclavicular ligaments and the intraarticular discoligamentous complex.4 Although there is some debate as to the primary stabilizing ligament of the SC joint,1,5 frank dislocations result in disruption of all ligamentous attachments in skeletally mature individuals.2 Therefore, despite acute management of instability through closed reduction procedures, many patients will continue to have recurrent symptoms of instability due to the lack of ligamentous stability.1
Figure-of-eight tendon reconstruction of the SC ligaments has been recommended in cases of recurrent dislocations of the SC joint or locked posterior dislocations.6 Reconstruction requires high technical skill due to the blind posterior dissection required to mobilize and reduce the medial clavicular head, and the procedure must be performed with a thoracic or vascular surgeon present in the event of injury to the great vessels, esophagus, or lung.2,7–9 Recent case series of various SC joint tendon reconstructions indicate clinical improvements in patient-reported outcome scores with limited recurrent instability.10–13
Although recent case series have shown promise for this surgical reconstruction, there is a paucity of literature to document the ability to return to recreational and work-related activities following surgery. The purpose of this study was to establish rates of return to sport and return to work such that physicians may be able to provide realistic expectations for patients considering surgery.
Materials and Methods
An institutional registry was queried for all cases of SC reconstruction performed between 2005 and 2016 to ensure minimum 2-year follow-up. Once institutional review board approval was obtained, operative reports were reviewed to verify consistency between surgical techniques. Patients younger than 18 years or those requiring revision SC reconstruction were excluded. Similar allograft techniques were performed by 2 senior surgeons (G.P.N., A.A.R.) based on a biomechanical study defining the most stable configuration.6 Indications for surgery included persistent symptoms of instability, usually associated with significant pain, that failed to respond to conservative treatment. None of the patients were surgically treated acutely following the initial dislocation event. Three shoulders had evidence of anterior dislocations. One shoulder was initially a posterior dislocation that underwent a closed reduction under anesthesia. Although the remainder of the shoulders (n=8) did not have either a clearly defined anterior or posterior dislocation observable on imaging (Figure 1), they demonstrated instability on physical examination. All patients had mechanical symptoms that included apprehension, crepitus, and frank movement of the medial clavicle that was exacerbated with forward flexion and abduction.
Axial computed tomography slices at the level superior to the sternoclavicular joint demonstrating the right medial clavicle (†; A) and at the level of the sternoclavicular joint demonstrating the manubrium of the sternum (‡; B). Compared with the left sternoclavicular joint (normal), the right clavicle is superiorly and medially dislocated relative to the sternum.
A standard and previously used return to sport/work survey was the primary mode of data collection for these patients. Each patient was contacted via phone and asked to complete this survey by trained research staff. Collected outcomes include listing of preoperative sport, postoperative sport, ability to perform push-ups, preoperative work, postoperative work, and time to full activity.
A horizontal incision inferior to and paralleling the inferior border of the medial clavicle was used to expose the SC joint. The capsule of the SC joint was incised from the superior surface of the medial clavicle directed inferomedially onto the sternum. This created 2 capsular flaps and exposed the medial clavicle and lateral sternum and the SC joint. Medial clavicle resection of 4 to 6 mm was performed in all cases with an oscillating saw. The intra-articular disk was debrided to a stable structure or removed. Two 4.5-mm holes were drilled symmetrically in both the sternum and the clavicle while using a retractor to protect posterior structures. A hamstring allograft was thawed and prepared and a No. 2 FiberWire (Arthrex Inc, Naples, Florida) was sutured to each end. With a tendon-holding stitch, a palmaris longus autograft was used in 2 cases. The graft was passed in a figure-of-eight fashion6 and secured by suturing the graft to itself while the assistant surgeon was manually maintaining reduction. The graft was sized to bring the edges of the clavicle and sternum together. Nonabsorbable sutures were used to further reinforce the graft and to secure the reconstruction prior to performing capsular closure of the SC joint.
Patients were admitted to the hospital for a planned 23-hour observation period to ensure that there were no symptoms related to respiration, digestion, or compression of the neurovascular structures. Patients remained in an abduction sling with no activities or lifting of the arm away from the body for 6 weeks postoperatively. Supervised physical therapy was initiated at this point with progression of passive to active range of motion. Resistance exercises and lifting activities were allowed after 3 months and continually progressed based on the patient's symptoms.
Statistical analysis was performed using RStudio software version 1.0.143 (R Foundation for Statistical Computing, Vienna, Austria). The Student's t test was used to compare improvement regarding continuous variables. Statistical significance was set at P<.05.
During the included years, 14 patients (16 shoulders) met the inclusion/exclusion criteria. Ten patients (71.4%; 12 shoulders) were available for follow-up assessment of their return to function. Mean age was 34.7±17.9 years. Mean body mass index was 27.0±6.9 kg/m2. Mean follow-up was 86.1±41.4 months (range, 26–145 months). All patients were right hand dominant. Seven surgeries were performed on the right SC joint, while 5 were performed on the left. One patient did not have any associated trauma with injury and received bilateral SC joint reconstructions. This patient was also the only one to receive gracilis autograft. The remaining patients had traumatic injuries. In total, 4 patients received gracilis grafts (2 autograft, 2 allograft) and 8 patients received semitendinosus allografts. The injury occurred a mean of 9.3±8.4 months prior to surgery. One patient had a posterior dislocation of bilateral SC joints. This was treated initially with closed reduction. The patient subsequently underwent bilateral medial clavicle resection prior to reconstructive surgery to either side. Another patient had a midshaft clavicle fracture preoperatively and underwent open reduction and internal fixation with a plate. The clavicle plate and screws were removed during the SC reconstruction procedure.
Mean preoperative visual analog scale and American Shoulder and Elbow Surgeons scores for the patients were 5.4±1.9 and 44.8±16.2, respectively. At final follow-up, visual analog scale and American Shoulder and Elbow Surgeons scores were 2.8±2.0 and 70.8±16.3, respectively. There were significant improvements in both visual analog scale and American Shoulder and Elbow Surgeons scores (P<.001, respectively). Six patients reported being very satisfied with their surgery, 1 reported being satisfied, 2 reported being fairly satisfied, and 1 reported being dissatisfied. All patients (10 of 10) reported that they would consent to the procedure again given their preoperative symptoms and postoperative improvement (Table 1). There were no postoperative complications within this cohort. None of the patients required return to the operating room for reasons related to their SC joint or were readmitted in the first 30 days.
Outcomes Following Sternoclavicular Reconstruction
A total of 8 patients were able to perform a push-up prior to injury. Only 4 (50%) of these patients achieved the ability to perform a push-up again following surgery. For these 4 patients, it was estimated that they could perform a mean of 25.8±27.1 push-ups before surgery and only 21.0±10.2 push-ups after surgery (P=.712). Nine patients reported participating in sports prior to injury. Following surgery, 4 of the 9 (44.4%) were able to return to sporting activity. Only 1 patient reported returning to a higher level of intensity than prior to injury. This patient also had bilateral instability that was of atraumatic origin since childhood. Patients participated in a mean of 3.9±1.8 sporting activities per week before injury and 1.8±2.1 at final follow-up (P=.022). One patient reported improved physical fitness, 1 reported the same level of fitness, and 8 reported a worse level of physical fitness compared with their preinjury level. Seven patients reported hindrance from sport due to the injured joint. Hand dominance was not a statistically significant variable regarding return to sport. Sports-specific outcomes are summarized in Figure 2.
Collected sports-specific outcomes preoperative and postoperative.
Seven patients had jobs prior to surgery. Their occupations are listed in Table 2. Two patients were insured by workers' compensation. At final follow-up, 3 of 7 patients (42.9%) were able to return to work at a mean of 6.3±10.1 months following surgery. Only the patient who was a sedentary worker was able to return to work at the same level; the other 2 had to reduce the intensity of their work. Regarding the 2 patients covered by workers' compensation who were unable to return to the same intensity of work, 1 was able to return to gainful employment, whereas the other went on permanent disability and was not working.
Occupations of Patients Undergoing Sternoclavicular Reconstruction for Instability of the Sternoclavicular Joint
Persistent SC joint instability is a challenging pathologic condition that can be treated with operative stabilization, but the return to work and sport is unpredictable. In the current study, a minority of patients were able to return to normal sport (44.4%), perform push-ups (50%), or return to their previous work (42.9%) following SC joint reconstruction for symptomatic instability. The 1 patient who returned to sport at a higher level was an unusual case of bilateral, atraumatic dislocations since childhood. The remaining 8 patients, who all experienced a traumatic injury to their SC joint and subsequent reconstruction, did not demonstrate a consistent return to function at the same level as before their injury. However, the majority of patients were satisfied (70% good to excellent satisfaction) with the results of their reconstructive procedure, and all of the patients reported that they would undergo the procedure again based on their current result. Despite the complexity and infrequency of SC joint reconstruction, the current group of patients did not experience an intraoperative or postoperative complication, and no patient had a revision procedure for the SC joint. Consistent improvements were seen in patient outcome scores (American Shoulder and Elbow Surgeons and visual analog scale). Based on the retrospective review of this case series, figure-of-eight reconstruction with a hamstring allograft can produce reasonable outcomes regarding relieving preoperative symptoms and improving function. However, the majority of patients will not achieve their preinjury level of function.
Prior studies of autograft and allograft reconstruction of the SC joint have reported improved clinical results regarding patient satisfaction and patient-reported outcomes, similar to those of the current study.10–13 Singer et al12 reported return to activity without physician-imposed restrictions in a cohort of 6 patients who were young and active (mean age, 22 years). However, there was no further analysis regarding which activities, time to return to activities, and level of function relative to preinjury status. Traditional outcome scores (eg, the American Shoulder and Elbow Surgeons score) are limited by the ceiling effect because evaluation is based on simple and universal tasks. The ceiling effect is defined as a limitation of an outcome measurement when the highest possible score is reached, as it may no longer accurately reflect the intended domain.14 Patients with greater functional demand will have optimal postoperative scores, although they are still restricted when compared with their preinjury level. Previous literature has been limited regarding accounting for the ceiling effect. From the current study, the authors found that, following surgical stabilization of the SC joint, patients were still restricted compared with preinjury function but demonstrated consistent improvements regarding clinical outcomes.
Instability of the SC joint is challenging to resolve with a soft tissue reconstruction because of the relative contributions of the bone and soft tissue anatomy. Anatomical study has shown that only 50% of the medial articular surface of the clavicle articulates with the sternum.2,15 The SC joint must therefore rely primarily on the soft tissue structures of the ligaments and joint capsule for stability.2,15 The figure-of-eight reconstruction has been previously determined to provide the greatest mechanical strength, although biomechanical testing in a cadaver model showed a modest 36.9% reduction in stiffness in the anterior direction and 3.8% reduction in stiffness in the posterior direction relative to the native ligaments.6 Therefore, it is likely that some residual laxity of the SC joint greater than the native SC joint occurs even with a technically secure tendon reconstruction. The inability of patients to tolerate and compensate for this increased laxity over their preinjury stability is likely to be the underlying cause of their inability to achieve preinjury levels of activity and work capacity. Still, patients consistently report positive clinical outcomes in terms of pain relief and overall function when compared with their presurgical status.
Differences in technique may also contribute to the variation in outcome. The current study was heterogeneous in its use of both allograft and autograft tendons for the reconstruction of the soft tissue injury. Bak and Fogh16 previously reported an experience with significant donor site morbidity (in 68% of patients) with the use of gracilis and palmaris longus autografts. However, other series have used autograft hamstring tendons with no reports of donor site morbidity.10,13,17,18 One series used 4.75-mm diameter polyetheretherketone interference screws as additional fixation for figure-of-eight reconstruction in an effort to stiffen the construct and improve the bone–tendon fixation strength.11 Alternatively, suture anchors, in addition to graft, have also been used, with positive clinical outcomes reported.16 Further biomechanical research is warranted to determine whether there is an added benefit to certain tendon grafts, the graft weave configuration, and the use of synthetic implants or augmentation devices such as suture tapes to improve the biomechanical characteristics of the reconstruction in an effort to more closely mimic or surpass the characteristics of the native joint.
Sternoclavicular instability is also seen with bone resection of the medial clavicle, with and without subsequent injury to the costoclavicular ligaments. Any bone resection removes the stabilizing effect of the articular geometry and will increase the laxity between the clavicle and the sternum. One patient within this series experienced bilateral SC instability that was treated with closed reduction of the posterior dislocation initially and medial clavicle resection to alleviate pain subsequently. Although the status of the patient's SC joint was unknown prior to the medial clavicle resection, it is likely that this procedure contributed to the patient's symptoms of instability based on previous reports of this as a potential adverse event.3,19 Recent biomechanical evidence has suggested that SC diskectomy alone does not result in significant reduction of forces on the SC joint.20 However, a 5-mm resection by either an oblique or a parallel (relative to the SC joint) technique resulted in reduction in joint forces equivalent to that of a 10-mm resection.20 Given that the average anatomical distance of the costoclavicular ligament has been reported as 12.6 mm from inferior articular cartilage of the medial clavicle, minimizing resection is warranted to prevent iatrogenic SC instability.4,21 Medial clavicle resection, SC joint instability due to ligamentous insufficiency, or recurrence of pathologic laxity after reconstruction may have profound effects on the ability to perform activities of sport or work. The inability of half of the patients to perform a push-up and the majority failing to return to sports and work at their preinjury level indicates the importance of the SC joint and the challenges of restoring stability after it has lost its integrity.
Limitations of this study included its retrospective design, the small number of patients, the variation in the graft used for reconstruction, and the inability to achieve follow-up for all patients. Although outcomes were collected prospectively, recall bias can occur owing to the difference in follow-up time between patients. Additionally, the infrequency of this injury leads to a limited sample, which restricts the conclusions that may be reached. Inclusion of additional patients may allow for further analysis regarding factors contributing to successful return to function. It was also assumed that patients have completed their recovery after a minimum of 2 years following the index surgery. Further follow-up may provide differing results if patients are able to return to function at longer intervals after surgery. Some heterogeneity existed within the patient population, as 2 patients had a history of prior clavicle surgery (resection and fracture fixation), 2 had bilateral injury, and 1 had atraumatic dislocations. These additional factors may have confounded the results, although this could not be verified within the limited sample. The purpose of this study was to better define the outcomes after this relatively uncommon procedure in relation to sport and work. The results offer useful information for counseling patients preoperatively and setting appropriate expectations for the outcome of the surgical procedure.
Figure-of-eight reconstruction of the SC joint using a hamstring graft through bone tunnels provides consistent clinical improvement and reasonable levels of patient satisfaction. However, the majority of the current patients were unable to return to preinjury levels of activity. Notably, only 50% of the patients could return to performing push-ups, 44.4% returned to sport, and 42.9% returned to work at the same level as before their injury. Patients must be counseled preoperatively regarding appropriate expectations for activity levels following SC joint reconstruction surgery.
- Groh GI, Wirth MA. Management of traumatic sternoclavicular joint injuries. J Am Acad Orthop Surg. 2011;19(1):1–7. doi:10.5435/00124635-201101000-00001 [CrossRef]
- Martetschläger F, Warth RJ, Millett PJ. Instability and degenerative arthritis of the sternoclavicular joint: a current concepts review. Am J Sports Med. 2014;42(4):999–1007. doi:10.1177/0363546513498990 [CrossRef]
- Panzica M, Zeichen J, Hankemeier S, Gaulke R, Krettek C, Jagodzinski M. Long-term outcome after joint reconstruction or medial resection arthroplasty for anterior SCJ instability. Arch Orthop Trauma Surg. 2010;130(5):657–665. doi:10.1007/s00402-009-0911-z [CrossRef]
- Lee JT, Campbell KJ, Michalski MP, et al. Surgical anatomy of the sternoclavicular joint: a qualitative and quantitative anatomical study. J Bone Joint Surg Am. 2014;96(19):e166. doi:10.2106/JBJS.M.01451 [CrossRef]
- Spencer EE, Kuhn JE, Huston LJ, Carpenter JE, Hughes RE. Ligamentous restraints to anterior and posterior translation of the sternoclavicular joint. J Shoulder Elbow Surg. 2002;11(1):43–47. doi:10.1067/mse.2002.119394 [CrossRef]
- Spencer EE Jr, Kuhn JE. Biomechanical analysis of reconstructions for sternoclavicular joint instability. J Bone Joint Surg Am. 2004;86(1):98–105. doi:10.2106/00004623-200401000-00015 [CrossRef]
- Clark RL, Milgram JW, Yawn DH. Fatal aortic perforation and cardiac tamponade due to a Kirschner wire migrating from the right sternoclavicular joint. South Med J. 1974;67(3):316–318. doi:10.1097/00007611-197403000-00017 [CrossRef]
- Gerlach D, Wemhöner SR, Ogbuihi S. Two cases of pericardial tamponade caused by migration of fracture wires from the sternoclavicular joint [in German]. Z Rechtsmed. 1984;93(1):53–60.
- Leonard JW, Gifford RW Jr, . Migration of a Kirschner wire from the clavicle into the pulmonary artery. Am J Cardiol. 1965;16(4):598–600. doi:10.1016/0002-9149(65)90040-8 [CrossRef]
- Petri M, Greenspoon JA, Horan MP, Martetschläger F, Warth RJ, Millett PJ. Clinical outcomes after autograft reconstruction for sternoclavicular joint instability. J Shoulder Elbow Surg. 2016;25(3):435–441. doi:10.1016/j.jse.2015.08.004 [CrossRef]
- Sabatini JB, Shung JR, Clay TB, Oladeji LO, Minnich DJ, Ponce BA. Outcomes of augmented allograft figure-of-eight sternoclavicular joint reconstruction. J Shoulder Elbow Surg. 2015;24(6):902–907. doi:10.1016/j.jse.2014.10.001 [CrossRef]
- Singer G, Ferlic P, Kraus T, Eberl R. Reconstruction of the sternoclavicular joint in active patients with the figure-of-eight technique using hamstrings. J Shoulder Elbow Surg. 2013;22(1):64–69. doi:10.1016/j.jse.2012.02.009 [CrossRef]
- Uri O, Barmpagiannis K, Higgs D, Falworth M, Alexander S, Lambert SM. Clinical outcome after reconstruction for sternoclavicular joint instability using a sternocleidomastoid tendon graft. J Bone Joint Surg Am. 2014;96(5):417–422. doi:10.2106/JBJS.M.00681 [CrossRef]
- Sciascia AD, Morris BJ, Jacobs CA, Edwards TB. Responsiveness and internal validity of common patient-reported outcome measures following total shoulder arthroplasty. Orthopedics. 2017;40(3):e513–e519. doi:10.3928/01477447-20170327-02 [CrossRef]
- Sellards R. Anatomy and biomechanics of the acromioclavicular joint. Oper Tech Sports Med. 2004;12(1):2–5. doi:10.1053/j.otsm.2004.04.006 [CrossRef]
- Bak K, Fogh K. Reconstruction of the chronic anterior unstable sternoclavicular joint using a tendon autograft: medium-term to long-term follow-up results. J Shoulder Elbow Surg. 2014;23(2):245–250. doi:10.1016/j.jse.2013.05.010 [CrossRef]
- Armstrong AL, Dias JJ. Reconstruction for instability of the sternoclavicular joint using the tendon of the sternocleidomastoid muscle. J Bone Joint Surg Br. 2008;90(5):610–613. doi:10.1302/0301-620X.90B5.20293 [CrossRef]
- Bae DS, Kocher MS, Waters PM, Micheli LM, Griffey M, Dichtel L. Chronic recurrent anterior sternoclavicular joint instability: results of surgical management. J Pediatr Orthop. 2006;26(1):71–74. doi:10.1097/01.bpo.0000187998.91837.b2 [CrossRef]
- Di Fabio S, Fusi P, Bonaspetti G, Pazzaglia UE. Bilateral spontaneous atraumatic anterior subluxation of the sternoclavicular joint. J Orthop Traumatol. 2004;5(2):110–112. doi:10.1007/PL00021720 [CrossRef]
- Katthagen JC, Marchetti DC, Dahl KD, Turnbull TL, Millett PJ. Biomechanical comparison of surgical techniques for resection arthroplasty of the sternoclavicular joint. Am J Sports Med. 2016;44(7):1832–1836. doi:10.1177/0363546516639302 [CrossRef]
- Carrera EF, Archetti Neto N, Carvalho RL, Souza MA, Santos JB, Faloppa F. Resection of the medial end of the clavicle: an anatomic study. J Shoulder Elbow Surg. 2007;16(1):112–114. doi:10.1016/j.jse.2006.04.010 [CrossRef]
Outcomes Following Sternoclavicular Reconstruction
|Shoulder outcome, mean±SD, change in score|
|Good/excellent satisfaction, No./total no.|
| Surgery||7/10 (70%)|
| Ability to play sports||3/10 (30%)|
| Physical fitness||3/10 (30%)|
| Cosmetic appearance||4/10 (40%)|
| Would have this procedure again||10/10 (100%)|
| Return to push-up, No./total no.||4/8 (50.0%)|
| Return to sport at any level, No./total no.||4/9 (44.4%)|
| Return to sport at same/better level, No./total no.||1/9 (11.1%)|
| Change in sporting activities, mean±SD, days per week||1.9±2.4a|
| Time to return, mean±SD, mo||14.0±9.2|
| Return to work at any duty, No./total no.||3/7 (42.9%)|
| Return to work at same/better intensity, No./total no.||1/7 (14.3%)|
| Time to return, mean±SD, mo||6.3±10.1|
Occupations of Patients Undergoing Sternoclavicular Reconstruction for Instability of the Sternoclavicular Joint
|Level of Dutya||Occupation|
|Sedentary (n=1)||Software developerb|
|Moderate (n=2)||Aviation technician; highway bridge crew member|
|Heavy (n=4)||Athletic trainerc; farmer/horse breeder; truck driverc; wallpaper hanger|