Introduction

Acromioclavicular (AC) dislocation is common in young active patients. The injury occurs especially in contact sports or after a fall while cycling or winter sports [8, 14, 15, 19, 37, 51]. Approximately 9 % of shoulder girdle injuries involve the AC joint [34]. Depending on the type and the intensity of the sports activity being practised, the percentage increases up to 40 % [23]. Classified by Rockwood et al. [42] into six grades, there is consensus about initial conservative treatment of Type I and II injuries, and though lacking evidence, in high-grade dislocations (Type IV–VI) a primarily surgical treatment is considered in most of the cases [2]. The treatment of Type III joint separations is still under discussion [25]. A variety of surgical techniques has been described, but none of them is considered to be a gold standard. In this context, the anatomic AC joint stabilisation with a synthetic non-absorbable suture system in mini-open or arthroscopic technique is a well observed and increasingly used treatment option for Type III to VI lesions [17, 22, 28, 30, 36, 43, 46, 50]. Short- and midterm follow-up presented good to excellent clinical and radiographic results leaving open issues with respect to the optimal operation technique, the placement of the drilling hole or the required number of sutures [26, 27, 47, 52, 53].

Despite the high frequency of AC joint injury and its relation to an active patient cohort, no study has considered the sports activity after AC joint stabilisation in detail. Especially for surgically treated high-grade AC joint separations (Type V), there is a lack of evidence about the sports activity after surgery. To the authors’ best knowledge, this is the first study to comprehensively analyse sports activity after AC joint stabilisation.

The purpose of this study was to determine the rates of return to sports, the time it took the patients to return to sports, level of activity following stabilisation of high-grade (Type V) acromioclavicular joint dislocation and to investigate the relevance of overhead sports on activity following AC injury. It was hypothesised that overhead athletes exhibit more limitations regarding return to sports compared to non-overhead athletes after AC joint stabilisation.

Materials and methods

Between 2011 and 2014, a consecutive series of 68 patients with a Rockwood type V [42] injury, irrespective of pre-injury sports activity, was enrolled retrospectively. Initial radiographs included true anteroposterior/axillary view of the affected shoulder and bilateral anteroposterior stress view (10 kg). In all cases, an acromioclavicular joint stabilisation in a single TightRope technique was performed at a level 1 trauma centre.

The following inclusion criteria were set: AC dislocation Type V according to Rockwood [42]; the injury occurred within the last 4 weeks; no concomitant injury of upper extremity; no shoulder problems concerning the injured or contralateral side before injury; and age ≥18.

A questionnaire regarding sports activity, the Constant–Murley Score (CMS) for self-assessment [6, 10] and the Disability of the Arm, Shoulder and Hand (DASH) questionnaire [21] were sent to all 68 patients. Thirteen patients (19.1 %) did not reach follow-up: Six refused to participate and seven could not be contacted.

In the end, 55 patients, including 47 males (85.5 %) and eight females (14.5 %), completed the questionnaires (follow-up: 80.9 %). They were followed for a median period of 24 months (range, 18–45 months).

Surgical procedure and post-operative care

All patients were operated on a standard procedure using the mini-incision single TightRope technique as described by Beris et al. [5] (Fig. 1). All surgeries were performed by one of two shoulder specialists (SSF, TG).

Fig. 1
figure 1

Anteroposterior view after single TightRope stabilisation

Full size image

The post-operative rehabilitation protocol consisted of immobilisation of the arm in a shoulder abduction orthosis (Medi SAK®; Medi, Bayreuth, Germany) at 30° for 4 weeks. Pain-free, passive range of motion up to 90° abduction and flexion and 40° internal/external rotation (IR/ER) was started on the first day post-surgery. Active motion restricted at 90° abduction and flexion and 40° IR/ER rotation was allowed after four weeks. Active range of motion on all ranges was allowed after 6 weeks, but overhead motion was especially trained under supervision of a physiotherapist at the start. Strength training for rotator cuff, deltoid and scapula stabilisers was gradually increased beginning at 6 weeks post-operative. Three months after surgery carrying weight and non-contact sports were allowed. After 6 months, the patients were allowed to return to sports without restriction.

Data acquisition

For all included participants, electronic patient records were screened for injury–surgery time interval, surgery time, hospitalisation time and documented revisions.

The participants received a questionnaire regarding general demographic data, pre-injury sports activity, post-injury sports activity and time to return to sports (see “Appendix in ESM”).

The type of sport was graded with regard to its strain on the shoulder according to Allain et al. (GI = “non-impact sports”; GII = “high-impact sports”; GIII = “overhead sports with hitting movements”; GIV = “overhead sports with hitting movements and sudden stops”) [1, 40]. Patients were grouped according to their pre-injury sports activity according to Allain et al. Group 1 included patients practising sports without overhead components (GI and G II according to Allain). Group 2 consisted of patients participating frequently in overhead sports (G III and G IV). A stratification between recreational and competitive or elite athletes was not conducted. None of the patients was a professional athlete. Patients who were not participating in sports before injury (non-athlete) were considered in regard to clinical outcome as well, but were not taken into account at the analysis of post-operative sports activity.

According to this assessment, 17 patients (30.9 %) were referred to group 1 (non-overhead sports) and 26 patients (47.3 %) were referred to group 2 (overhead sports). 12 patients (21.8 %) did no sports before injury (non-athlete) (Fig. 2).

Fig. 2
figure 2

Flow diagram; Group distribution

Full size image

The functional outcome was evaluated using an adopted Constant–Murley Score (CMS) for self-assessment, which was inaugurated and validated by Böhm et al. (pts) [6, 10], the age and gender adjusted CMS (%) [54] and the DASH questionnaire [21].

This study was approved by the local ethics committee of the board of Medical Profession of Rhineland-Palatinate in Mainz (No. 837.009.15/9777).

Statistical analysis

Statistical analysis was performed, advised by a biometrician, using SPSS. Descriptive results are given as median and minimum/maximum for continuous variables, frequencies and percentage for categorical data. Nominal/categorical data were compared using the Chi-square test or Wilcoxon signed-rank test. For metric data, the Mann–Whitney U test was used. The level of significance was defined as p = 0.05. Evaluation of the data was retrospective; an a priori sample size estimation has not been conducted. A post hoc sample size calculation regarding the hypothesis computed, that a difference of 40 %, as found at this study, can be detected with a sample size of N = 19 per group with a power of 80 % at the level of significance of 0.05.

Results

Basic demographic data, concomitant injuries and surgery-related information for all 55 patients are given in Table 1. There were no significant discrepancies concerning descriptive data (Table 1) between group 1 and 2. A sports-related injury was reported in 28 cases (50.9 %), most frequently due to a bicycle accident (N = 16, 19.1 %).

Table 1 Demographic data
Full size table

Clinical outcome

For all 55 patients, the median CMS was 85.0 pts (range, 44.0–100.0 pts) at final follow-up. The CMS of the contralateral side was 99.3 pts in median (range, 70.0–100.0 pts), which was significantly higher than the involved shoulder (p < 0.001). The median DASH Score was 5.0 (range, 0.0–43.9). Comparisons of both groups revealed no significant differences between overhead and non-overhead athletes concerning CMS or DASH (Table 2).

Table 2 Clinical Outcome
Full size table

Sports activity before injury

Forty-three patients (78.2 %) performed regular sports activity before injury. The median of sports frequency was three times a week (Fig. 3). According to Allain, 26 patients (60.5 %, group 2) participated in overhead sports and 17 patients (39.5 %, group 1) practised non-overhead sports (Table 4; Fig. 4).

Fig. 3
figure 3

Comparison of pre- and post-injury frequency of sports (N = 43); (p = 0.004)

Full size image
Fig. 4
figure 4

Comparison of pre- and post-injury sports level according to Allain (N = 43); (p = 0.051)

Full size image

Sports activity after surgery

At final follow-up, 41 of 43 (95.3 %) of the patients, who were participating in sports before injury, returned to sports. In detail, two patients (4.7 %) had to stop their sports activity due to injury-related shoulder problems. Overall, the sports frequency was significantly lower at follow-up (p = 0.004, Fig. 3). The level of sports according to Allain showed a trend towards lower shoulder demanding activity (n.s., Fig. 4).

Sixteen patients (37.2 %) had to change their sports activity after the operation (Table 3).

Table 3 Return to sports and changes of sports activity
Full size table

Athletes participating in overhead sports (group 2) needed to change sports activity significantly more often (53.8 vs. 11.8 %; p = 0.011). Mainly, overhead athletes had to reduce intensity (N = 13, 50 %) and frequency (N = 5, 19.2 %). Interestingly, seven patients (26.9 %) participating in overhead sports before injury had to reduce their sports level to low demanding sports, while none of the non-overhead athletes (group 1) lowered their level (p = 0.029) (Table 4).

Table 4 Detailed type of sports and post-injury changes
Full size table

The median time to return completely to their pre-injury sports activity was 9.5 months (range, 3.0–18.0 months). Overhead athletes needed a significantly longer time to completely return to sports (9.5 vs. 4.5 months; p = 0.009).

Discussion

The most important finding of the present study was that despite high general return rates to sports (95.3 %) there are considerable restrictions following a surgical AC joint stabilisation. For overhead athletes, in particular, we noticed a significant alteration of sports activity after surgery. More than half of them reported that they had to change their sports activity. In detail, 26.9 % of the overhead athletes had to choose another less demanding type of sport and 50 % reported that they reduced the intensity of sports. In contrast, only two of the non-overhead athletes (11.8 %) had to change their sports activity in general after surgery.

This supports the notion that overhead activity is more demanding for the AC joint [31, 33]. In fact, all of the overhead athletes, who needed to change the type of sports, practised non-shoulder demanding sports afterwards (Table 4). In addition, a reduction in sports frequency was only found in group 2. Despite the high return to sports rates, more than half of the patients reported soreness of the operated AC joint during sports activity. These complaints were independent from the type and intensity of sports.

Data concerning sports activity after AC joint injury are lacking, despite the fact that AC joint injury has been identified as a typical sports-related injury [12, 15, 38]. According to the literature, 42.9 % of AC joint injuries are sports related [8]. This is in agreement with our findings (50 % sports-associated traumas). Here, bicycle accidents were found to be responsible for most cases. However, it must be noted that we did not distinguished between road traffic and sports activity associated accidents.

There are several epidemiologic studies regarding professional athletes, which analyse incidence and mechanisms of AC joint injury in detail. These studies provide only sporadic information on sports activity after AC joint injury [13, 24, 32, 38]. In these studies, the time to return to sports varied between 9.8 and 22.1 days on average [24, 32, 38], differing considerably from our data. Further analyses of return to sports rates and sports activity after AC joint injuries were not provided. Moreover, the studies dealt with a very restricted cohort (solely professional football players) with various grades of injury (Rockwood I–IV). Most often, the injuries were rated as Rockwood I–III injuries and were treated conservatively.

In conclusion, the existing data about sports activity after a conservative treatment of AC joint injury report a remarkably shorter rehabilitation period and less impairment of sports activity [24, 32, 38]. In this context, the ongoing controversial debate about the benefit of surgical treatment should be considered. There is a general consensus about primary non-operative treatment of Rockwood I and II lesions [4]. Several studies promote a shorter rehabilitation period—an immobilisation of the shoulder is only recommended for short-term pain reduction [4]—and fewer complications in conservative treatment [49]. The aforementioned studies which are dealing with professional football players seem to confirm this argument.

In contrast, data about optimal treatment of Type III injury remain uncertain. An older survey among American team orthopaedics revealed that 69 % would treat the injury non-operatively [35]. Recently, Balke et al. reported a considerable preference among German surgeons for surgical treatment of Type III lesions (73 %). For Type V lesions, 99 % of surgeons prefer a surgical treatment [2].

Especially for active patient cohorts, surgical treatment is recommend [4, 18]. The data from this study call this approach into question. As elaborated above, in this series in particular the more active overhead athletes complained about impairment of sports activity post-operatively.

A comparison with conservative treated patients with Type V injury could clarify this point. Unfortunately, a conservative treated control group is lacking in our study and in the existing literature as well. It can be assumed that at least the time to return to sports would be shorter in favour of conservative treated patients. However, due to the heterogenic cohort and inconsistent evaluation of sports activity in comparable studies a substantiated statement cannot be made at this point.

Data about return to sports capacity after AC joint surgery are heterogeneous. Rangger et al. reported a reduction in sports activity in 28 % of patients after conservative or surgical treatment of AC joint injuries [41]. Gstettner et al. demonstrated that four of 19 patients (21 %) who were treated with a hook plate were not able to return to their former level of sports [18]. Bannister et al. compared conservative versus surgical treated patients and found a significant longer time to return to sports after surgical treatment (16 vs. 7 weeks) [3]. In contrast, Cardone et al. reported that the time to return to pre-injury sports level was in favour of the surgical group (18.8 vs. 26.2 weeks) [7]. These studies are limited by outdated surgical techniques and heterogeneous types of injury severity, and they also lack detailed information regarding post-operative sports activity.

To date, minimal invasive anatomic coracoclavicular reconstructions have been established as the gold standard. After TightRope stabilisation, 100 % return to sports rates are reported, without specification of type of sports or impairments after surgery [11, 16]. Time to return to sports (120 days) reported by De Carli et al. were comparable to our findings [9]. However, the influence of the type of sport was not considered. Loriaut et al. reported about an average time of 17 weeks before sports were resumed after TightRope stabilisation of distal clavicle fractures. Fourteen of 17 patients reached the same level as before trauma. A differentiation of sports activity was lacking [29].

To the best of our knowledge, our study is the first to provide detailed information on sports activity and impairments after AC stabilisation and to consider the factors that might affect the return to sports rates.

Overall, the clinical results of our study are in agreement with others reporting good to excellent functional results after TightRope stabilisation [9, 22, 45, 46]. A median CMS of 92.9 % and a DASH Score of 5.0 reports indicate satisfactory results.

Interestingly, the present study demonstrates higher revision rates (18 %) compared to published data. Scheibel et al. [46] reported no revisions in 28 cases, the highest incidence of revision rates (12.5 %) was found by Murena et al. [36]. In the present study, five patients (9.1 %) developed wound-healing impairments. This might be related to the mini-open technique. According to the literature, wound-healing problems seem to be less frequent after arthroscopic stabilisation [22, 39, 46]. Conversely, Rosslenbroich et al. reported only one infection out of 83 patients operated on with mini-open technique [43]. All documented wound-healing impairments affected the clavicular incision. We identified a prominent knot of the TightRope, superior to the clavicular button as a potential cause of mechanical irritation or even wicking. Modifying our technique, by submerging the knot beneath the deltoid muscle using an additional vicryl suture, the wound-healing impairments seemed to decrease although there is no statistical evidence to support this. Any further investigation should clarify this point. The rate of revisions due to recurrent instability (7.3 %) was comparable to existing data about single or double TightRope stabilisation [36, 43, 45, 46]. The TightRope technique used in this study only addressed the horizontal instability. Recently, Saier et al. [44] promoted an additional acromioclavicular cerclage for horizontal stability. This might have provided a lower rate of re-instability or even better clinical outcome. Patients were included to this study if the surgical procedure was done within 4 weeks after injury. Several authors recommend an early operation when using artificial augmentation of CC ligaments. They hypothesised that the soft tissue which forms around the device is required for long-term stability [20]. Derived from that, the authors suggested a surgery within the first 3 weeks. To the best of our knowledge, there is no study providing evidence-based data concerning the optimal time to operate. In clinical routine, we are performing a AC stabilisation in TightRope technique within the first 4 weeks after injury and have not found major impairments in patients operated on in the third or fourth week. In this series, a subanalysis revealed no significant differences in clinical outcome between early (within 2 weeks) and late (after 2 weeks) surgical stabilisation. Despite these current findings, we tend to stabilise the AC joint as soon as possible after injury. The outcome of long-term follow-up is important in this regard.

This study was limited by its retrospective design and the relatively short median follow-up of 24 months. The main purpose of this study was to analyse the return to sports. Most of the patients return to sports within the first 18 months. Information about the time to return to sports or sports activity before injury could have been lost due to oblivion in the case of a longer follow-up period. Even so, the retrospective survey of pre-injury sports activity, in particular, might have caused a recall bias. In addition to our main objective, we evaluated the global functional outcome using the Constant–Murley Score. In consideration of the clarity of the study, we decided to omit an additional AC joint-specific evaluation tool as introduced by Scheibel et al. [46] or Taft et al. [48]. Scheibel et al. developed the ACJI because they observed an overestimation of the functional outcome after AC joint injuries by other shoulder scoring systems. This point might explain the high values of CMS, despite the remarkable limitation of sports activity. Applying such an AC joint-specific score could have provided a more differentiated analysis of AC joint function.

It should be noted that there were no professional athletes in this cohort. Here, higher demands on the stabilised joint on the one hand, but more comprehensive medical support on the other hand might lead to different results. The fact that patients with a surgical treatment within 4 weeks after injury were included might have lead to adverse results. All patients were operated on by one of two specialised surgeons, which provides a high standardisation of the surgical procedure, but might imply a performance bias. Beside the surgical technique, the rehabilitation protocol was standardised.

Another strength of this study is, besides its large number of patients, the homogeneous cohort. Only patients with a Rockwood Type V injury were enrolled.

Enhancing new knowledge about limitations of return to sports after AC stabilisation, especially for overhead athletes, can give the findings impact with respect to therapeutic decision-making after injury and help in prognosis and assessment of the rehabilitation progress. Patients can be informed about realistic chances and risks regarding sports activity after a surgical AC stabilisation.

Conclusion

After minimal invasive stabilisation of the AC joint using a single TightRope technique, the majority of patients return to sports within the first year. Besides this quite long rehabilitation period, a notable risk of surgical associated complication should be considered. In particular, athletes participating in overhead activity should be informed about the frequent necessity to reduce the intensity of the sporting activity or to even change the type of sport practised. On the basis of our data, the current assumption that sports activity is a relative indication to prefer surgical treatment should be further reviewed, particularly in high-grade AC instability. Further studies, including a conservative control group and other surgical techniques, need to be implemented.