Comparative Study of Surgical Treatment of Acromioclavicular Luxation Komparativna Studija Hirurškog Lijeþenja Akromioklavikularne Luksacije

Background/Aim. Acromioclavicular (AC) luxations most often affect athletes. The published results regarding the treatment of AC joint luxations vary. Each method has its advantages and disadvantages, so there is still no consensus on the best method of treatment. The aim of this study was to review the results of a number of surgical approaches to stabilization of AC joint recorded over the span of five years. Methods. This study was based on the data acquired from the analysis of 28 patients with AC luxation surgically treated in the Clinical Center of Monte-negro. One group of 16 patients underwent the traditional AO method (with transfixation of AC joint with Kirschner wire and Zuggurtung tension bands) or the Bosworth method (using the coracoclaviculartransfixation screw – Zugg-Bosw group). The second group of 12 patients underwent a newer techinque with the Hook plate (Hook plate group). Results. All the patients had AC luxation of higher degree, stage IV–VI acording to the Rockwood scale. The average age of the two groups was very similar, with 28 being the average age of the Zugg-Bosw group, and 25 of the Hook plate group. Most patients were males


Conclusion.
The Hook plate method achieved somewhat better results, which indicate that this method is one of the ways to ensure a strong, stable fixation of the AC joint without transfixation.At the same time, this method does not inhibit the ligament healing and allows an early mobilisation.

Introduction
The first studies on acromioclavicular (AC) luxation repair were by Hippocrates, Galen, and Paul of Aegina.They recommended conservative management with compressive bandages to keep the clavicle in a normal position.The first surgical repair of an acute AC dislocation is credited to sir Samuel Cooper who, back in 1861, used a loop of silver wire to approximate the clavicle and acromion process 1 .Subsequently, numerous other techniques were reported, including suture repair of the AC ligaments and coracoclavicular (CC) ligament, tendon graft for reconstruction, and fixation with nails, screws or wires.Even today, however, there is no consensus on the best resolution of this problem 2,3 .
More often than not, the injury occurs when a direct force is applied to the upper part of the acromion, when, during the fall, the arm is in adduction.Less comonly, the injury occures when a force is applied indirectly like, for instance, when a person falls on a streched-out arm 4,5 .
AC luxations mostly affect athletes, especially those who engage in contact sports (football, rugby, judo, hokey) 6 .Young athletes (in their teens or twenities) are particularily prone to this type of injury.Also, men are five to ten times more likely to be affected than women.These injuries are very common, and cause up to 40% of all shoulder injuries and up to 3% of all sports injuries 7 .
AC joint luxations can be classified in several ways.Cadenat 8 differentiated incomplete injuries, in which only the capsular AC ligaments were torn, from injuries that involved disruption of both the AC and CC ligamentous structures.Allman 9 and Tossy et al. 10 later recognized 3 different types of AC separation based on the similar criteria.
Rockwood 11 has expanded this classification by including 3 additional variants.Type IV injuries are defined by posterior displacement of the clavicle relative to the acromion with buttonholing through the trapezius muscle.In type V injuries, the clavicle is widely displaced superiorly relative to the acromion as a result of disruption of muscle attachments.The rare type VI injuries are characterized by inferior displacement of the distal clavicle below the acromial process or the coracoid process.This classification is dominant today 11,12 .
In general, it is commonly accepted that lower degree AC luxations (I-III degree) are treated conservativelly, while higher degree AC luxations (IV-VI degree) are treated surgically.There is, however, lack of concensuse on the treatment of III and IV degree of AC luxation.Minority of authors 13,14 argue for a conservative approach, while the majority contends that surgery is the more approprate approch.
There is a number of different surgical approaches to affixiate the AC joint, such as the use of Kirschner wires, cerclage wires, transfixation screws, different types of plates, together with the use of sutures, ligament transpositions, or various ransplants (fascia lata, hamstring tendons, etc.) 15,16 .The newest technques include artroscopic fixations, the use of which requires endobutton, special types of hard seams, anchor with or without tendom grafts.Also, the authors often favorize the modified Weaver-Dunn method 17,18 .The published results of all these approaches vary, every method has its strengths and weaknesses, so consensus on the best approach remains unclear 19 .
The goal of this study was to compare both early and later results of the surgical approaches to luxation of the AC joint, as they are used at the Traumatology Clinic at the Clinical Center of Montenegro (CCM).The focus will be on the approach using Kirschner wire -the "Zuggurtung" and Bosworth method, and on the approach using the Hook plates.The later approach has been utilized at the CCM since 2005.

Methods
This study analysed the results of surgical treatment of 28 patients treated from January 2005 to June 2010 in the CCM Traumatology Department.These patients were divided in two groups based on the method of the surgical approach to AC joint luxation.The first group was made of 16 patients treated with, up to that point, widely accepted surgical methods: the AO method 20 , which encompasses the use of two Kirschner wires and the Zuggurtung tension band, and the Bosworth method, which requires the use of coracoclavicular transfixation screw 21,22 .This group we marked as the Zugg-Bosw group (Figures 1a, b).
The second group was made of 12 patients who were treated with the newer method which utlizes the Hook plate, and the group was named accordingly 23 (Figure 2).
Presurgical procedure was standard for all patients and included detailed anamnesis, clinical examination of the AC joint (with all patients suffering a loss of normal shoulder contours, with a sticking-out clavicle, pain and an inability to function normaly) and radiographic diagnostics (axillary, radiographic, radiographic in AP position without pressure and radiographic with a 5 kg pressure).
Indication for surgical treatment was based on the degree of the joint instability, which in turn was based on Rockwood's classification of different degrees of the AC joint luxation.The patients with IV-VI degree underwent surgery.

Fig. 2 -Clavicle Hook plate
The average duration of the postsurgical follow-up review was 11 months (a span of 6-36 months).To achieve the most objective analysis of the clincal results of these patinets we used the following parameters: pain, activities, range of motion, power (PARP) -where we used the Constant Score Scale (CSS) 24,25 .
Naturally, we also understood that a final result would be incomplete without a subjective evaluation by patients.We asked our patients about their opinion on the achieved results like, for instance, whether they are able to work in the same manner as before the injury, whether any pain was left in the shoulder, whether they felt an equal mobility and strenght in the shoulder as before the injury, and whether they experinced any other complications connected to their treatment at the CCM.The subjective evaluation was done according to the Likert scale 26 .
The comparisons between the groups were carried out using one way ANOVA, with Bonferroni post-hoc testing for multiple comparisons.A repeated measure ANOVA model was fit for each response using SAS Proc Mixed software (SAS Institute, Inc, Cary, NC), and the Bonferroni test was again employed to control for multiple comparisons.The t values and degrees of freedom were reported for all linear regression ANOVAs.Differences were considered significant at values of p 0.05.All results were presented as mean ± SD.

Results
The results are based on the five years of a follow-up examination of the two groups of pateints treated in the CCM, the Zugg-Bosw group (n = 16) and Hook plate group (n = 12).The average age in the first group was 28 (18-52 years), and in the second 25 (17-50 years).Statistically, no significance existed between the two groups with respect to age.In addition, with respect to gender, there were much more male than female patients (82% vs 12%, respectively).The Zugg-Bosw group had 13 male and 3 female patients (81% and 19%, respectively), and the Hook plate group had 10 male and 2 female patients (83% and 17%, respectively).
With respect to the manner of the injury of the AC joint, athletic activity proved to be the main cause in both groups.The Zugg-Bosw group numbered 12 (75%) and the Hook plate group numbered 10 (83%) such patients.Statistically, there was no significant difference between the two groups with respect to the manner of injury (p = 0,61; t = 0,52; df = 27).A small number of patients were injured due to traffic accidents (Zugg-Bosw, n = 2; Hook plate, n = 1).Accidental falls also had a minor presence (Zugg-Bosw, n = 2; Hook plate, n = 1).The analysis showed no significant difference in the presence of either of these two causes in either of the two groups (p = 0.74; n = -0.34;df = 27).The causes of the AC joint injuries are presented in Table 1.Most patients did not wear any imobilizators after the surgery.In the Zugg-Bosw group, 2 patients wore Desault's bandage for 3 and 4 weeks, respectively, and 2 patients wore arm slings for 2 weeks each.In the Hook plate group, only 3 patients wore imobiliztors, that is, arm slings, and again for 2 weeks each.The results show that patients in the Zugg-Bosw group spent on average more time in immobilization when compared to patients from the Hook plate group (0.75 + 0.34 vs 0.41 + 0.29, respectively).However, there was no statisticaly significant difference between these two groups (Figure 3).During surgeries, we used the appropriate osteosynthetic material.The next surgical manuever involved the removal of that material.Our results showed that the average period from surgery to removal of osteosynthetic material was 4.6 months in the Zugg-Bosw group, and 4.5 months in the Hook plate group.There was no statisticaly significant difference in the lenght of time from surgery to the removal of osteosynthetic material between these two groups (p = 0.82; t = -0.22;df = 27) (Figure 4).Postsurgical complications were different in kind, time of apperance and degree.The patients from the Zugg-Bosw group had a higher number of later complications (8 vs 16) than was the case with the Hook plate group patients (2 vs 12; p = 0.19; t = -1.34;df = 27).There were also more deformities in the first group (2 vs 1; (p = 0.74; t = -0.34;df = 27).In addition, there were more calcifications found in the Zugg-Bosw group (2/16) than in the Hook plate group (1/12; p = 0.74; t = -0.34;df = 27).However, statistical analysis revealed no significant difference in the occurence of these complications.Notably, the Zugg-Bosw group had 2 post-surgical infections and 3 looseing of alenthesis (p = 0.12; t = -1.6,df = 27).Such complications were absent the Hook plate group (Table 2).
Evaluation of patient's subjective satisfation showed a very significant difference between two goups.Using the Likert scale from 1 (bad) to 5 (excellent) to grade their postsurgical states, the patients from the Zugg-Bosw group evaluated their state as good (3.0 ± 0.39), while patients from the Hook plate group gave a much better grade (4.4 ± 0.19; p = 0.007, t = 2.95, df = 27) (Table 3).An important objective indicator of the postsurgical state -Pain, Activity, Range of Motion, Power -was graded by using the CSS.The maximum number of points for Pain was 15, for Activity 20, for Range of Motion 40 and for Power 25 (Table 3).There was statistically no significant difference between the two groups (p = 0.078, t = 1.8, df = 27).Patients from the Hook plate group had a better median score (90 ± 0.18) with respect to the patients from the Zugg-Bosw group (85 ± 0.40).
The latter two parameteres show that the Hook plate method gave somewhat better results for our patients who suffered from the AC joint luxation than did the classical Zugg-Bosw method which was almost an exclusive approach in the preceding years.

Discussion
In the last five years (2005-2010), there were 28 patients treated in the CCM Traumatology Clinic for luxation of the AC joint.Indications for surgical treatment were clinical examination and radiography of the shoulder which revealed IV-VI degree of injury based on the Rockwood's scale.
Many methods of fixating the AC joint have been described, but there are certain dilemmas as to which implants are to be used.The AO technique with two Kirschner wiresand a wire binder of the Zuggurtung type often causes migration of the Kirschner wire, while transclavicular screw may break the clavicle, which warrants their early removal, which in turn brings more dislocation and instability to the AC joint, ultimatelly ending in pain and disfunctionality.
The Hook plates have been on the market for a relativelly short time.Its design, where the plate is above the clavicle and the hook below the acromion, provides a very good stability of the AC joint.However, it also provides a danger of a subacromial impingement.
As there is no unanimous view about the best approach to AC joint luxation, in our clinic we used both the traditional methods: Zugg-Bosw method and the newer Hook plate method.
With respect to gender, male patients were much more numerous than female patients (82% vs 12%, respectively).In the Zugg-Bosw group, gender distribution was 13 male and 3 female patients (81% and 19%, respectively).In the Hook plate group, there were 10 male and 2 female patients (83% and 17%, respectively).This is in accordance with the statistical data presented by other authors.Rockwood contends that AC joint injuries are seen especially in competitive athletes, and occur most frequently in the second decade of life.Males are more commonly affected than females, with a male-to-female ratio of approximately 5:1 27 .
Most of our patients injured themselves during athletic activities (22 out of 28, 78%), some received injury due to traffic accidents (3 out of 28, or 11%) and some received injury due to accidental falling (3 out of 28, 11%).The analysis revealed no statistically significant difference in the occurrence of these causes of injury (p = 0.74; n = -0.34;df = 27).
Most of our patients did not wore any imobilizators during the post-surgical tretment.The subjective opinion of the surgeon about the stability of the osteosynthesis was the key factor with regards to placement of the imobilizators (Desault or arm slings).Cirstoiu et al. 28 and Zarzycki et al. 29 contend that post-surgical imobilizators should be in lace for up to 4 weeks, especially in the case of percutaneous fixations.
The average period from surgery to removal of osteosynthetic material was 4.6 months in the Zugg-Bosw group, and 4.5 months in the second Hook plate group.There was no statisticaly significant difference in the lenght of time from surgery to the removal of osteosynthetic material between these two groups.Koukakis et al. 30 recommend extraction of the osteosynthetic material within 3 months because of the possibility of subacromial impingement.
With respect to postsurgical complications, the patinets from the Zugg-Bosw group had a larger number of complications than did the patients from the Hook plate group.Zugg-Bosw group had 2 exterior infections which were treated with regular dressings and oral antibiotics.The osteosynthetic material became loose (2 screws used in the Bosworth method and 1 Kirschner wire), so one patient experinced deformity in the AC joint, limited movement and pain.Two patients in this group had a deformity-limited movement.Calcification between clavicle and processue coracoideus was identical, but there were no functional or esthetic irregularities.There was one case of subacromial impingment.In the Hook plate group, there was one case of deformity-limited movement, and one case of calcification and subacromial impingment, without functional irregularities.Still, the difference was not statistically significant.This is in accordance with Winstein et al 31 .
The evaluation of the patients' subjective satisfation showed very significant difference between the two goups.Using the Likert scale from 1 (bad) to 5 (excellent) to grade their post-surgical states, the patients from the Zugg-Bosw group graded their state as good while patients from the Hook plate group gave a much better grade.In the available literature, we did not find that other authors graded the subjective satisfaction in this way.
Using the CSS, we received an objective indicator of the post-surgical state of our patients from the both groups.The PARMP results show that there was no statistically significant difference, but the patients from the Hook plate group received a better median score than the patients from the Zugg-Bosw group.These results are in accordance with those published by Ladermann et al. 32 .
The newest data reveals that posttraumatic artritis occurs more frequently when transarticular fixations of the AC joint are used 33 .Therefore, the majority of authors argues for a temporary fixation between the coracoideus and the clavicle, and not via the acromioclavicular joint.Otherwise, migration of the wires is likely to happen and will probably cause certain complications, as described by Lindsey et al. 34 .

Conclusion
Because of the small number of patients examined, the results of this study must be taken with some reserve.Still, we can conclude that Hook plate allows stabile and strong fixation of AC luxation, and at the same time does not inhibit healing of the ligaments and allows for early mobility.This method provides good short term results with a small number of complications.This study shows that further research on both short and long term results are needed to bring a clearer understanding of the more advanced techniques.
a) Two Kirschner wires and a Zuggurtung tension band; b) Coracoclavicular transfixation screw, Bosworth method

Fig. 3 -
Fig. 3 -Postoperative immobilization; Bar graph showing the average time (in weeks) that the patients spent in postoperative immobilization after Zugg-Bosw (white column, left) and Hook plate (dashed column, right) types of surgery.Note that patients in the Zugg-Bosw group spent a slightly longer time in immobilization than in the Hook plate groups, but the difference was non-significant.The results were presented as averages with standard error bars (SE)

Fig. 4 -
Fig. 4 -Removal of osteosynthetic material; Bar graph showing the average time (in months) for the removal of osteosynthetic material after Zugg-Bosw (white column, left) and Hook plate (dashed column, right) types of surgery.Note that there was no statistical difference in length of removal of osteosynthetic material between these two groups of patients.The results were presented as averages with standard error bars (SE)

Table 3 Evaluation of patient satisfaction and pain, activity, range of motion and power (PARMP)
Values are expressed as ± SD