Injury patterns of the acromioclavicular ligament complex in acute acromioclavicular joint dislocations: a cross-sectional, fundamental study

In operative treatment of acute acromioclavicular joint (ACJ) dislocations, increasing attention is paid to reconstruction of the superior acromioclavicular ligament complex (ACLC) [2, 10, 12, 25, 26]. The superior ACLC functions as the major horizontal stabilizer of the ACJ [3, 5, 8, 16]. Insufficient superior ACLC healing may contribute to persistent horizontal ACJ instability, which is reported in up to 50 % of cases following both arthroscopic and open reconstruction of acute ACJ dislocations [13, 17, 24, 33]. Horizontal ACJ instability represents a clinically relevant issue, since several studies showed an association with inferior functional outcome [13, 19, 30]. Therefore, modern techniques of acute ACJ reconstruction employ additional synthetic acromioclavicular stabilization consisting of absorbable or non-absorbable high-strength suture cerclages [10, 12, 30]. Biomechanical cadaver studies have proven, that such combined coraco- and acromioclavicular stabilization enables restoration of physiological ACJ stability [2, 26]. However, time-dependent loosening of synthetic acromioclavicular stabilization results in recurrent instability [26]. Hence, definite horizontal ACJ stability will depend on the biomechanical quality of anatomical ACLC healing. Efforts to anatomically repair ACLC injuries might therefore restore horizontal instability in acute ACJ reconstruction [10].

Histological studies showed the acromioclavicular ligaments as a component of the joint capsule being confluent with the articular disc. If present, the articular disc may vary in size and appearance [6, 27]. The deltoid-trapezoidal (DT) fascia inserts on the superior lateral clavicle and superior ACLC providing additional biomechanical reinforcement [31]. Given the close anatomical and biomechanical relationship of the acromioclavicular ligaments, joint capsule, articular disc, and DT fascia insertions, these structures should be regarded as one anatomical unit, which we refer acromioclavicular ligament complex (ACLC). According to a current macroscopic anatomical study of Nakazawa et al. [20], the acromioclavicular ligament consists of a well-developed superoposterior (SP) bundle and a weaker anteroinferior (AI) bundle. In addition to detailed comprehension of biomechanics and anatomy, optimal operative repair also requires profound knowledge of principal characteristics of acute ACLC injuries. However, hardly any clinical findings exist on morphology and influencing cofactors of acute ACLC injuries.

Based on long-term clinical observations, we assumed that distinct and recurrent ACLC tear patterns would exist in acute ACJ dislocations. Another hypothesis was, that specific ACLC tear types would be associated with anatomical characteristics and injuries of related structures (articular disc, DT fascia, bony ACJ morphology), and the underlying type of ACJ dislocation.

Purposes of this prospective, cross-sectional study were 1.) to identify different ACLC tear patterns with development of a differentiated classification system, 2.) to assess prevalences of different ACLC tear patterns, and 3.) to identify influencing cofactors (articular disc, DT fascia, type of ACJ dislocation and bony ACJ morphology).

The study was approved by the Ethics Committee of the University of Freiburg (Vote-Nr.: 490/13) and written informed consent was obtained from all participating patients. From 1/2014 to 7/2016, the study consecutively included a total of 65 patients (6 female, 59 male) with complete, acute-traumatic ACJ dislocations type 4 or 5 according to the Rockwood classification [23] treated operatively by means of mini-open ACJ reduction and hook plate stabilization. Mean age was 38.2 years (range, 19–57 years) at the time of surgery. Inclusion criteria were age >18 years, isolated acute-traumatic ACJ dislocation type 4 or 5 and full legal competence. Exclusion criteria were age >60 years, incomplete ACJ dislocations (Rockwood-3A/B), subacute-chronic ACJ dislocations with an interval >3 weeks from trauma to surgery, radiological signs of ACJ degeneration/arthrosis and prior history of ACJ injury or surgery. Indication for surgery was not influenced by the patient’s decision for study participation. During the study period, 74 patients were eligible for study participation corresponding to an inclusion rate of 88 %. Non-inclusion of 9 patients was caused by refusal of study participation (n = 3) and incomplete documentation (n = 6).

All patients showed complete acute-traumatic ACLC disruptions. Each ACLC tear could be clearly assigned to one of four distinct and recurrent tear patterns. Based on these intraoperative observations, we propose a morphological classification of acute ACLC tears.

This study is the first to systematically investigate and describe ACLC tear patterns in acute ACJ dislocations, and hereby provides fundamental knowledge of characteristics of acute ACLC injuries. Using a novel morphological classification system, acute ACLC disruptions could be assigned to clavicular-sided (AC-1), oblique (AC-2), midportion (AC-3), and acromial-sided (AC-4) tears. Prevalences of specific ACLC tear types (AC-1-4) depended on the size of the articular disc. In accordance with Salter et al. [27], the majority of our patients (57 %) had a meniscoid (partial) articular disc. A proportion of 37 % showed no articular disc or only remnants of an articular disc. Histological studies have confirmed that the articular disc is confluent with the superior capsular-ligamentous structures forming an anatomical-biomechanical unit [6, 27]. Hereby, the articular disc seems to function as a biomechanical reinforcement of the superior ACLC. Recently, Nakazawa et al. [20] identified a strong superoposterior (SP) bundle and a more delicate anteroinferior (AI) bundle of the acromioclavicular ligament. The articular disc predominantly attaches to the supero-lateral (acromial) SP bundle shifting its biomechanical point of weakness to the clavicular insertion zone [6, 20].

These principal anatomical-biomechanical aspects strongly support our observations on acute ACLC tear morphotypes and their correlation to the size of articular disc manifestation. In acute-traumatic ACJ dislocation, presence of a pronounced articular disc predominantly induced a clavicular-sided (AC-1) tear due to increased stress shielding of the supero-lateral ACLC. ACLC detachment typically occurred subperiosteally, since bony insertion zones are biomechanically augmented by fibrocartilaginous entheses. Clinically, such subperiosteal clavicular-sided ACLC avulsion causes a pathognomonic “peeled-like” appearance, which we refer to as the “banana sign”. In case of a remnant or absent articular discs, the central and lateral portions of the superior ACLC became vulnerable to rupture. Accordingly, oblique and acromial-sided tears were associated with smaller articular disc sizes. Midportion tears were rare and only occurred in case of complete absence of the articular disc. Thus, the central superior ACLC in between fibrocartilaginous insertion zones represented the “weakest link of the chain”. Interestingly, bony ACJ morphology did not influence ACLC tear patterns. In addition to the size of articular disc manifestation, biomechanics of trauma is likely to substantially influence ACLC tear patterns, particularly with regard to the direction of translation (force vector) of the acromion related to the clavicle. In this context, principal biomechanical studies should follow.

To date, hardly any knowledge exists about the morphology of acute ACLC tears, and to our best knowledge no previous study focused on this topic. Rockwood et al. [23] described different degrees of acute ACLC injuries from sprain to complete rupture but did not analyze the morphology of ACLC tears. Hessmann et al. [9] performed AC ligament repair and augmentation in acute type 3, 4 and 5 injuries. They usually found the ACLC to be “avulsed from the lateral clavicle” without describing other tear types. In a biomechanical cadaver study, Freedman et al. [7] observed midsubstance ACLC tears resulting from superior translation of the clavicle. Present study entails important implications for acute ACJ reconstruction and operative ACLC repair. Specific types of ACLC tears were not related to different types of ACJ dislocation. However, a Rockwood-4 injury represented a prognostic risk factor for articular disc dislocation/rupture and ACLC incarceration. Fourteen (21.5 %) of patients (all Rockwood-4 and 10.5 % of Rockwood-5 dislocations) required mini-open surgery to obtain anatomical ACLC and ACJ reduction. In these cases, ACLC incarceration caused mechanical obstruction. Attempts of closed reduction probably would have caused non-anatomical ACJ and ACLC reduction, potentially resulting in insufficient ACLC healing and persistent horizontal ACJ instability. According to our experience, the articular disc can be preserved in majority of acute patients. We performed partial discectomy in only six (9.2 %) patients with unstable articular disc dislocation tears to obtain a stable residual articular disc. Remaining unstable tears might have a negative impact on outcome. With regard to the DT fascia, avulsion “in continuity” represented a consistent mode of failure. Insertional avulsion and thinning of the DT fascia were found in both Rockwood− 4 and −5 dislocations, whereas additional tears usually occurred in type 4 dislocations only. These longitudinal or transverse tears were rather small (<3 cm) and could be easily repaired. Perforation of the lateral clavicle throughout the DT fascia was previously considered as the pathoanatomical correlate of Rockwood-4 injuries. However, clavicular perforation was found in only one case, and therefore has to be regarded as an exceptional pattern of injury.

Current pathoanatomical findings substantially expand understanding of acute ACJ dislocations and significantly influence their operative treatment. This study showed, that complex ACLC injuries with incarceration rather than injury of the DT fascia prevent anatomical ACJ reduction. From our point of view, mini-open ACLC and ACJ reduction should be performed in Rockwood-4 injuries with true static posterior ACJ dislocation. Moreover, a relevant proportion of about 10 % of Rockwood-5 dislocations may also require mini-open reduction. At the beginning of the procedure, surgeons should carefully verify feasibility of anatomical ACJ reduction prior to application of closed techniques of ACJ reduction and stabilization. Any fluoroscopic evidence of joint space enlargement suggests presence of ACLC incarceration as an obstacle to reduction. Thus, both the surgeon and the patient should be prepared for conversion to a mini-open procedure. In this context, preoperative MRI could prove of value for assessment ACLC tear patterns and identification of patients benefiting from mini-open surgery due to otherwise limited healing potential. Current ACJ-specific MRI techniques allow detailed assessment of ligamentous structures including the articular disc [11, 21, 29]. In our own series, we were able to distinguish distinct ACLC tear types and to diagnose articular disc dislocations in selective cases by means of ACJ-specific MRI (Fig. 12). However, future studies should evaluate the role of preoperative MRI for diagnosis of acute ACLC injuries.

Several studies have shown significant association of horizontal instability with inferior clinical outcome following arthroscopically-assisted and open acute ACJ reconstruction [13, 19, 30]. Scheibel et al. [30] reported inferior Taft (9.2 versus 11.4. points) and ACJI scores (92.3 versus 63.3 points) compared to patients without residual horizontal instability following double TightRope ACJ reconstruction. Jensen et al. [13] found considerable impairment of not only ACJ-specific but also glenohumeral function (normalized Constant score [15]: 76 % versus 93 %) for both double TightRope and hook plate stabilization. Thus, insufficient ACLC healing causing persistent horizontal ACJ instability represents a clinically relevant issue. Beginning from the 1980s, ACLC suturing was an important component of acute ACJ stabilization using tension band wiring, screws or Balser (hook) plates [14, 18, 22]. However, none of the techniques succeeded substantially due to hardware complications, iatrogenic fractures, persistent ACJ instability and ACJ arthrosis/synostosis resulting from non-physiological ACJ kinematics. Biomechanical studies demonstrated that only combined, non-rigid coraco- and acromioclavicular stabilization may restore physiological ACJ stability [2, 26]. Saier et al. [26] employed an x-shaped acromioclavicular cerclage consisting of high strength suture tapes. Nonetheless, final horizontal ACJ stability will be determined by biomechanical quality of biologic ACLC healing due to the likelihood of time-dependent loosening/failure of synthetic stabilization material. Optimal biomechanical ACLC healing requires both anatomical ACJ/ACLC reduction and physiological ACJ stabilization [10]. Present study showed, that mini-open operative ACLC treatment was required in 21.5 % of acute ACJ dislocations to achieve anatomical reduction. This finding might at least partly explain the high rates of horizontal ACJ instability following closed techniques of acute ACJ reconstruction [13, 17, 24, 28, 30, 33]. Additional operative ACLC repair might support anatomical ACLC healing, if sufficient biomechanical augmentation is provided during the phase of healing [10]. In this context, we propose type-specific operative techniques for anatomical ACLC repair (Fig. 11).

Limiting, the superior surgical approach allowed accurate inspection and assessment of anterior, superior and posterior except inferior ACLC portions. Therefore, inferior ACLC tear morphology might vary from superior ACLC tear types. However, this limitation does not substantially affect significance of present study, since the superior ACLC represents the major restraint to posterior dislocation, and therefore clinical efforts focus on superior ACLC repair [2, 5, 10, 12, 16]. Analysis of clinical and radiological outcome was not aim of this cross-sectional study but should be subject of future comparative clinical trials.

There exist distinct and recurrent morphotypes of ACLC tears in acute ACJ dislocations. Clavicular-sided (AC-1) tears were observed in 72 %, oblique (AC-2) tears in 19 %, midportion (AC-3) tears in 4 % and acromial-sided (AC-4) tears in 5 %. Morphology of acute ACLC tears depended on anatomical conditions of the acromioclavicular ligament complex (ACLC), particularly on the size of the articular disc. AC-1 tears usually occur in presence of a pronounced articular disc, whereas AC-2-AC-4 tears are associated with a remnant or absent articular disc. Acute ACLC with dislocation and incarceration caused mechanical obstruction to anatomical ACJ reduction in 21.5 % of cases including all Rockwood-4 dislocations. In these cases, mini-open ACLC reduction was required to achieve anatomical ACLC and ACJ reduction. With regard to the DT fascia, avulsion “in continuity” represented the consistent mode of failure. Type-specific operative strategies enabled anatomical operative ACLC repair of all observed tear patterns. Prospective clinical studies should evaluate clinical implications of operative ACLC and whether combined coraco-, acromioclavicular stabilization and anatomical ACLC repair is able to diminish rates of horizontal ACJ instability in acute ACJ reconstruction.