Stabilization of Acute High-Grade

Acromioclavicular Joint Separation

A Prospective Assessment of the Clavicular Hook Plate
Versus the Double Double-Button Suture Procedure
Thomas Stein,*yz MD, PhD, Daniel Müller,§k MD, Marc Blank,§ MD,
Yana Reinig,y§ MD, Tim Saier,{ MD, PhD, Reinhard Hoffmann,§ MD, PhD,
Frederic Welsch,y MD, and Uwe Schweigkofler,§ MD
Investigation performed at Berufsgenossenschaftliche Unfallklinik Frankfurt am Main,
Frankfurt am Main, Germany

Background: The stabilization strategy for acute high-grade acromioclavicular (AC) joint separations with AC-stabilizing clavic-
ular hook plate (cHP) or coracoclavicular (CC)–stabilizing double double-button suture (dDBS) is still under consideration.
Hypothesis: The CC-stabilizing dDBS is superior to the cHP according to an AC-specific radiologic assessment and score system.
Study Design: Cohort study; Level of evidence, 2.
Methods: Seventy-three consecutive patients with acute high-grade AC joint separation were prospectively followed in 2 treat-
ment groups (64.4% randomized, 35.6% patient-selected treatment): open reduction and cHP (cHP group) or arthroscopically
assisted dDBS (dDBS group) performed within 14 days of injury. Patients were prospectively analyzed by clinical scores (Taft,
Constant score [CS], numeric analog scale for pain) and AC-specific radiographs (AC distance, CC distance [CCD], relative
CCD [rCCD; 100 / AC distance 3 CCD]) at points of examination (preoperative and 6, 12, and 24 months). The minimal clinically
important differences (MCIDs) were assessed by the anchor-based method.
Results: Twenty-seven of 35 patients (mean age 6 SD: 37.7 6 9.7 years) after cHP implantation and 29 of 38 patients (34.2 6 9.7
years) after dDBS implantation were continuously followed until the 24-month follow-up. All patients showed significantly
increased scores after surgery as compared with preoperative status (all P \ .05). As compared with GI, GII had significantly bet-
ter outcomes at 24 months (Taft: cHP = 9.4 6 1.7 vs dDBS = 10.9 6 1.1, P \ .05, MCID = 2.9; CS: cHP = 90.2 6 7.8 vs dDBS =
95.3 6 4.4, P \ .02, MCID = 16.6) and at 24 months for Rockwood IV/V (Taft: cHP = 9.4 6 1.7 vs dDBS = 11.1 6 0.8, P \ .0005;
CS: cHP = 90.1 6 7.7 vs dDBS = 95.5 6 3.1, P \ .04). Clinically assessed horizontal instability persisted in 18.52% (GI) and 6.89%
(GII; P = .24). The rCCD showed equal loss of reduction at 24 months (GII = 130.7% [control = 111%] vs GI = 141.8% [control =
115%], MCID = 11.1%).
Conclusion: This prospective study showed significantly superior outcomes in all clinical scores between GII and GI. The sub-
analysis of the high-grade injury type (Rockwood IV/V) revealed that these patients showed significant benefits from the dDBS
procedure in the clinical assessments. The cHP procedure resulted in good to excellent clinical outcome data and displayed
an alternative procedure for patients needing less restrictive rehabilitation protocols.
Keywords: acute acromioclavicular separation; acromioclavicular joint (ACJ); arthroscopically assisted acromioclavicular joint
stabilization; double-CC-bundle acromioclavicular joint stabilization; relative CCD; MCID (minimal clinically important differences)

In the published literature, .160 acromioclavicular joint enables the adaption of the CC ligament remnants by
(ACJ)–stabilizing techniques are described and generally implant placements through the conoid and trapezoid
divided into acromioclavicular (AC)– and coracoclavicular insertion areas in single or double configurations. In con-
(CC)–stabilizing aspects.2 The CC stabilization strategy trast, the AC-stabilizing surgery techniques have a direct
CC-stabilizing aspect.2,3,6,29 Although it was shown that
both technique aspects enable anatomic healing of the
CC ligament complexes,10,13 a current meta-analysis
The American Journal of Sports Medicine
revealed superior outcomes after the CC-stabilizing techni-
1–10
DOI: 10.1177/0363546518788355 ques versus the AC-stabilizing procedures.1,19 Beitzel et al2
Ó 2018 The Author(s) hypothesized that anatomic CC-stabilizing device

1
2 Stein et al The American Journal of Sports Medicine

TABLE 1
Inclusion Criteria, Criteria for Crossover Treatment, and General Indication for Surgerya

Inclusion criteria Postoperative

Age at surgery: 18-45 y  No surgery at both shoulders (except implant removal)
Pre- and intraoperative  No macro reinjury at both shoulders
 Unilateral ACJ separation grade RW 3-5  Compliance to rehabilitation protocol
 Treatment in GI or GII 14 d after trauma Indication for surgery
 No rotator cuff injury . Ellman A1/Snyder A1  Acute ACJ separation type RW IV/V 14 d after trauma
 No preoperative osteoarthritic changes  Acute ACJ separation type RW III 14 d after trauma with combined
 No prior shoulder surgery vertical and horizontal instability and regular physical activity
 No bony injury Indication for crossover treatment
 Sport activity before injury . 13/week Concomitant lesions with indication for arthroscopic therapy (SLAP
lesions 1, Bankart lesions)

a
ACJ, acromioclavicular joint; GI, group I (clavicular hook plate); GII, group II (double double-button suture); SLAP, superior labrum
anterior and posterior; RW, Rockwood.

positioning would result in superior clinical outcomes prospectively analyzed (inclusion criteria and indication
when compared with nonanatomic device positionings. for surgery: Table 1; patient data: Table 2, Figure 1). All
Several biomechanical studies confirmed this hypothesis patients who consented to participate in the study protocol
and supported significant advantages of the double dou- were thoroughly informed of both specific surgical techni-
ble-button suture (dDBS) procedure versus clavicular ques, the current literature, and the existing rehabilitation
hook plate implantation.20,37 However, the current litera- programs (Appendix Table A1, available in the online ver-
ture lacks prospective studies comparing different stabiliz- sion of this article). Patients were then randomly assigned
ing techniques with short injury-to-surgery intervals and into 1 of 2 groups with freeware Research Randomizer (v
ACJ-specific radiologic and clinical outcome assessments. 2.0; http://www.randomizer.org). Patients who rejected
We aimed to prospectively assess, by radiologic and clini- the randomization and preferred 1 of the 2 treatment
cal assessment, 2 stabilizing procedures for acute high-grade groups were included as ‘‘intention to treat.’’ For both
ACJ separation. We hypothesized that an arthroscopically groups, the contralateral uninjured shoulder was used as
assisted double CC-stabilizing procedure would generate the radiologic control. Acute concomitant glenohumeral
superior clinical outcomes in the AC-specific Taft score anal- lesions were evaluated with a preoperative magnetic reso-
ysis when compared with the AC-stabilizing procedure. Addi- nance imaging (MRI). In case of acute concomitant lesions,
tionally, we hypothesized that persisting vertical and patients with assignment to the cHP group were trans-
horizontal AC instability was better restored by the double ferred to the dDBS group as ‘‘crossover treatment.’’ In
CC-stabilizing technique over the AC-stabilizing surgery. both groups, surgery was performed in an injury-to-sur-
gery interval 14 days. The following points of examina-
tion were established: P0, preoperative; P1, directly
METHODS postoperative (only radiologic assessment); P2, 6 months;
P3, 12 months; and P4, 24 months. Patients were allowed
Study Design and Cohort Selection
to miss only 1 follow-up (P2 or P3) if social or professional
Our institutional ethics committee approved all experi- reasons (eg, moving, temporary stay abroad) hindered
mental procedures, and all subjects provided informed con- them from appearing for the radiologic control assessment.
sent (data collection and clinical trial registry No. FF 42/
2011). Using a prospective study design, we determined Surgical Techniques
the radiologic and clinical outcomes after AC stabilization
with the clavicular hook plate (cHP group) versus arthro- Open Clavicular Hook Plate. Open reduction and fixa-
scopically assisted stabilization with the CC dDBS system tion were performed with the patient in the beach-chair
(dDBS group) of acute high-grade ACJ separations type position under general anesthesia. A mini-open sagittal
Rockwood III to V. During a 5-year period, all consecutive 6- to 8-cm incision was made between the lateral clavicle
patients with acute unilateral ACJ separations were and the ACJ space. After clavicular reduction, the

*Address correspondence to PD Dr med Dr rer nat Thomas Stein, MD, PhD, Department of Sporttraumatology, Knee, and Shoulder Surgery, Berufs-
genossenschaftliche Unfallklinik Frankfurt am Main, Friedberger Landstraße 430, 60389 Frankfurt am Main, Germany (email: dr.thomas.stein@me.com).
y
Department of Sporttraumatology, Knee, and Shoulder Surgery, Berufsgenossenschaftliche Unfallklinik Frankfurt am Main, Frankfurt am Main, Ger-
many.
z
Department of Sports Science, University of Bielefeld, Bielefeld, Germany.
§
Department of Trauma and Orthopedic Surgery, Berufsgenossenschaftliche Unfallklinik Frankfurt am Main, Frankfurt am Main, Germany.
k
Department of Trauma, Hand, and Reconstructive Surgery, University Hospital Frankfurt am Main, Frankfurt am Main, Germany.
{
Department of Reconstructive Joint Surgery, Berufsgenossenschaftliche Unfallklinik Murnau, Murnau am Staffelsee, Germany.
One or more of the authors has declared the following potential conflict of interest or source of funding: T.S. has received payment from Arthrex for
instructional courses.
AJSM Vol. XX, No. X, XXXX Stabilization of Acute High-Grade ACJ Separation 3

Enrollment Assessed for eligibility

(n = 112)

Excluded (n = 39)
Declined to participate (n = 23)
Other reasons:
- re-surgery (n = 5)
- no compliance (n = 8)
- injury-to-surgery-interval > 14d (n = 3)
Randomized (n = 47)
- Concomitant lesions (n = 1)
Patient-selected treatment (n = 25)

Allocaon
Allocated to cHP group (n = 35) Allocated to dDSB group (n = 37)
Received allocated intervention (n = 35) Received allocated intervention (n = 37)

Follow-up
Lost to follow-up (n = 8) Lost to follow-up (n = 8)
- missed one or more FU (n = 5) - missed one or more FU (n = 5)
- inability to contact patient (n = 3) - inability to contact patient (n = 3)

Analysis
Analysed (n = 27) Analysed (n = 29)
- Initial assignment: randomised (n = 19) - Initial assignment: randomised (n = 20)
- Initial assignment: Patient-selected treatment (n = 8) - Initial assignment: Patient-selected treatment (n = 9)
- Initial assignment: intention to treat (n = 10) - Initial assignment: intention to treat (n = 12)

Figure 1. CONSORT (Consolidated Standards of Reporting Trials) flow diagram of assignment to the cHP group and the dDBS
group. cHP, clavicular hook plate; dDBS, double double-button suture; FU, follow-up.

TABLE 2 the ACJ space under the acromion. After plate fixation by
Patient Characteristics for the cHP and dDBS Groupsa 3 bicortically placed screws, the torn CC ligaments were
repaired with absorbable sutures. The deltoidotrapezoid fas-
cHP dDBS P Value cia was plicated with absorbable sutures (Figure 2). All hook
Patients plates were removed by the third postoperative month after
RW III to V 35 of 52 37 of 60 .66 the index procedure according to standard protocol, as
(67.3) (61.7) described in previous studies.25,35
Follow-up 27 of 35 29 of 37 .93 Arthroscopically Assisted CC Stabilizing With dDBS
(77.1) (78.4) Systems. All surgery was performed with the patient in
Reasons for exclusion the beach-chair position and the ipsilateral arm in a pneu-
Revision-stabilization 5 3 .33 matic arm holder. In all cases, dDBS systems were used
No compliance 2 3 .09 (10-mm round clavicular titanium plate, 4-strand continu-
Missing 1 follow-up 8 9 .93
ous loop of nonresorbable suture material, 13 3 3.4–mm
Injury to surgery .14 d 3 .06
Sex, male:female 26:1 28:1 .71
oblong coracoidal titanium plate). The 3-portal technique
Age at injury, y 37.65 6 9.66 34.24 6 9.68 .2 for AC stabilization was performed with a 4-mm/30° angle
Injury to surgery, d 5.15 6 3.35 8.89 6 2.59 .00008b arthroscope. The surgical procedure was performed with
Side the 3-portal technique described by Scheibel et al.28 Based
Dominant 14 (51.85) 13 (44.83) on the anatomic study of Rios et al,23 the clavicular drill
Nondominant 13 (48.15) 16 (55.17) .6 holes were made at 25 mm and 45 mm from the lateral
Preoperative RW clavicle for the trapezoid and conoid, respectively. The del-
3bc 3 8 .34 totrapezial fascia was closed in a manner similar to that for
4:5 4:20 1:20 .76 the hook plate technique (Figure 3).
a For the postoperative rehabilitation protocol of the cHP
Values are presented as n (%) or mean 6 SD. cHP, clavicular
hook plate; dDBS, double double-button suture; RW, Rockwood.
and dDBS groups, see Appendix Table A1.
b
P \ .05.
c
Only with combined vertical and horizontal instability. Radiographic Assessment
clavicular hook plate (3.5-mm clavicular hook plate) was All patients underwent the preoperative radiologic assess-
positioned on the lateral clavicle. Analogous to previous ment, which included bilateral anteroposterior panorama
techniques,12 the acromial hook was placed posterior to stress views with 10-kg load, Y view, axillary lateral view,
4 Stein et al The American Journal of Sports Medicine

Figure 2. Implantation of the clavicular hook plate and trans- Figure 3. Implantation of the double double-button suture
osseous coracoclavicular ligament refixation with orientation systems with orientation to the anatomic insertion areas.
to the anatomic insertion areas. con, conoid ligament; tra, con, conoid ligament; tra, trapezoid ligament.
trapezoid ligament.

and MRI. The panorama view enables bilateral ACJ assess-

ment in the same view; therefore, the beam is directed at
the jugular with a 30° to 45° caudocranial tilt. The directly
postoperative radiograph controls (P1) were performed
without load, whereas 6-, 12-, and 24-month controls (P2-
P4) had, analogous to the P0 panorama stress view, a
10-kg load. The separations were graded according to Rock-
wood and Young.24 For radiograph analysis, the ‘‘coracoidal’’
parallel line was drawn horizontal to the ground and
through the superior coracoid cortex; the ‘‘clavicular’’ paral-
lel line was drawn through the inferior clavicular cortex
directly above the coracoid; and the ‘‘acromial’’ parallel
line was drawn through the inferior acromial cortex directly
at the center of the acromion. The absolute acromiocoracoid
distance (ACD) was defined as the distance between the acro- Figure 4. Analysis of the relative CCD = 100% / absolute
mial and coracoidal parallels, and the absolute CC distance ACD 3 absolute CCD. ACD, acromioclavicular distance;
(aCCD) as the distance between the clavicular and coracoidal CCD, coracoclavicular distance.
parallels. Principally, interindividual (patients’ height and
sex) and systemic measurement (varying beam tilting) errors records the clinical status for pain, range of motion, and
influence the absolute measurement values and the compro- strength. Additionally, the pain intensity status was mea-
mise assessments between different patients and different sured by a numeric analog scale (NAS). The horizontal
radiographs. The relative value assessments neutralize these instability was recorded as positive when the manual ante-
interindividual and systematic errors and allow dimension- rior-to-posterior translation of the lateral clavicle was
less measurements. The relative CC distance (rCCD) was increased as compared with the uninjured control shoul-
defined as the percentage ratio of the absolute distances of der. The AC-specific Taft score30 was applied. All clinical
the ACD and the aCCD (100% / ACD 3 aCCD) on the analyses were performed by the same orthopaedic nonop-
same panorama radiograph (Figure 4). The measurement erating scientist (D.M.). At all points of follow-up, the ques-
values (aCCD) and the reference (ACD) are similarly influ- tionnaire including the Taft score, the CS, and the NAS
enced by interindividual and systematic measurement and the radiologic assessment of the rCCD were applied.
errors, which were canceled out by the relative assessment Minimal clinically important differences (MCIDs) were
method. All radiologic analyses were performed by the assessed by the anchor-based method analogous to the
same orthopaedic nonoperating scientist (M.B.), with established protocol of Tashjian et al.31,32 Satisfaction
a Mac-based Osirix Imaging Software system (v 3.5). and improvement were graded by patients’ self-assessment
of the surgical treatment with a 6-item anchor question,
which was derived from established protocols.31-33
Clinical Shoulder Outcomes Measurement Tools
Clinical assessments for pre- and postoperative examina- Statistical Analyses
tions were standardized. At each follow-up, complete
examination of both shoulders was performed according An independent statistician evaluated the data. Mean and
to a standardized protocol. The function-specific Constant SD were calculated with Microsoft Excel for Mac (v 15.25)
score (CS)5 was applied, which subjectively and objectively and BIAS software (Biometric Analysis of Samples; v 8.4.2
AJSM Vol. XX, No. X, XXXX Stabilization of Acute High-Grade ACJ Separation 5

TABLE 3
Postoperative Assessment: cHP Group vs dDBS Groupa

P0 (Preoperative) P2 (6 mo) P3 (12 mo) P4 (24 mo) P Value

ACJ-specific Taft score

cHP 3.00 6 0.96 8.00 6 2.00 8.89 6 1.28 9.41 6 1.69 \.05b
dDBS 3.10 6 1.54 9.41 6 1.27 10.72 6 0.88 10.90 6 1.08 \.05b
P Value .16 \.05b \.05b \.05b
Function-specific Constant score
cHP 32.67 6 10.51 69.89 6 7.98 86.89 6 3.95 90.19 6 7.79 .0000001b
dDBS 27.45 6 6.31 85.79 6 8.54 93.34 6 5.51 95.31 6 4.42 .0000001b
P Value .07 .000001b .00001b .02b
NAS for pain
cHP 13.96 6 0.71 6.74 6 1.65 4.08 6 1.23 2.44 6 2.99 .000001b
dDBS 13.21 6 2.69 6.31 6 2.09 1.81 6 1.02 1.74 6 1.86 .00001b
P Value .22 .31 .000001b .67

a
ACJ, acromioclavicular joint; cHP, clavicular hook plate; dDBS, double double-button suture; NAS, numeric analog scale.
b
P \ .05.

for Microsoft). The nonparametric Wilcoxon-Mann-Whitney type 1; 4 supraspinatus lesions, Ellman grade A1-A2/Snyder
U test was used to detect statistically significant differences grade A1-A2), which were all treated with arthroscopic
in scoring systems. A P value \.05 was considered statisti- debridement. The mean points of follow-up were as follows:
cally significant. The sample size for the paired t test was P0, P1, P2 control (cHP 8.3 6 5.27 months; dDBS 5.2 6 5.34
performed with the software G*Power (v 3.1.9.3 for Mac), months), P3 control (cHP 12.54 6 2.41 months; dDBS 16.08
and the significance level (a) was 0.05 (2-tailed). For detect- 6 8.35 months), and P4 control (cHP 40.12 6 23.64 months;
ing differences for the primary outcome parameter (the Taft dDBS 30.75 6 1.41 months).
score) with an effect size (d) of 0.96 and 90% power, a sample
size of 48 shoulders would be adequate. MCIDs for each out-
come variable (Taft, CS, NAS, and rCCD) were calculated Clinical Outcomes
per Tashjian et al31,32 and Tubach et al.33 First, the change
scores for each variable were calculated from preoperative On preoperative radiologic assessment, both groups had
assessments to final postoperative assessments. Patients comparable distributions of different grades of ACJ insta-
were classified by the anchor question as having ‘‘change’’ bility according to Rockwood (Table 2). In this regard, the
(answers 2 and 3) or ‘‘no change’’ (answers 4 and 5). The preoperative clinical status in both groups was similarly
answer ‘‘excellent’’ (answer 1) and ‘‘bad’’ (answer 6) were decreased by the acute injury (Tables 3 and 4). All scores
not included per the protocols. The t tests were performed in both groups showed significant longitudinal improve-
to compare the means between the ‘‘change’’ and ‘‘no ments after surgical stabilization (P \ .05). The Taft score
change’’ status. The MCID is the mean difference between detected significantly superior data for the dDBS proce-
the changed and unchanged status. dure as compared with the cHP surgery at all points of
postoperative assessment. The clinically assessed horizon-
tal instability persisted in 18.52% (cHP) and 6.89% (dDBS;
P = .24). The CS detected for both groups consecutive post-
RESULTS operative improvements and excellent function outcome
status at the 24-month control. Both groups showed,
During the study period, 73 patients (cHP, n = 35; dDBS, n between different postoperative assessments, significantly
= 38) were included in the prospective assessment (Table 1, higher function with the dDBS procedure. The NAS assess-
Appendix Table A1, Figure 1). At the end of the follow-up ment revealed a significantly increased pain status for the
interval, 77.1% (cHP) and 76.3% (dDBS) fulfilled all postop- 12- and 24-month controls for the cHP group as compared
erative inclusion criteria and were included for scientific with the dDBS group. The separate MCID analysis
analysis with clinical and radiological assessments at all revealed that the statistically significant differences in
points of follow-up. The clavicular hook plates were removed all scores remain under the level of the MCIDs.
after 9.09 6 1.58 weeks; no clavicular buttons in the dDBS
group were removed within that time interval. There were
no significant differences in patient characteristics between Radiologic Outcomes
groups (Table 2). Diagnostic arthroscopy in the dDBS group
detected, per Pauly et al,22 additional chronic lesions or Table 5 shows radiologic outcome data from all points of
lesions with unclear traumatic correlations in 7 (24.1%) of assessment. Analysis of rCCD in both groups revealed sig-
29 patients (2 subscapularis lesions, Fox and Romeo type nificant improvements between P0 and all postoperative
1; 2 SLAP [superior labrum anterior and posterior] lesions, periods (P \ .05). Regarding the loss of reduction (LOR)18
6 Stein et al The American Journal of Sports Medicine

TABLE 4
Average Value of the Pre- and Postoperative Outcome Scores
for the Assessment of the MCID by Applying the Anchor Methoda

Taft Score Constant Score NASPain rCCD

Average value
Preoperative 3.05 6 1.29 29.96 6 8.91 13.57 6 2.02 234.55 6 24.54
Postoperative 10.18 6 1.59 92.84 6 6.73 2.08 6 2.47 136.05 6 27.27
Difference
Change 8.00 6 1.53 68.83 6 6.35 –12.12 6 1.67 –102.53 6 35.23
No change 5.13 6 1.25 52.20 6 9.01 –10.77 6 4.14 80.33 6 42.36
Difference: change – no change
MCID 2.87 16.63 –1.37 22.20
P value \.00001b \.00001b .71 .04b

a
MCID, minimal clinically important difference; NAS, numeric analog scale; rCCD, relative coracoclavicular distance.
b
P \ .05.

as compared with the postoperative status, the rCCD anal- With regard to the subdivision of ACJ-stabilizing proce-
ysis showed significant postoperative LOR within both dures,2 current meta-analysis1,19 and recent studies similarly
groups between P1 and P4 (cHP = 138.28%, P \ .05; showed that ‘‘nonanatomic’’ AC-stabilizing and single CC-sta-
dDBS = 126.55%, P \ .05). bilizing techniques generate inferior shoulder function with
inadequately high rates of remaining ACJ instabilities (Table
6).8,11,19,36 Neither open single CC-stabilizing procedures (1
double-button suture device vs Bosworth screw)8 nor open
Subanalysis According to the Grade of Injury
AC-stabilizing clavicular hook plate implantation (vs arthro-
The subanalysis according to the initial grade of ACJ dislo- scopically assisted 1 double-button suture device) enabled
cation revealed that all patients showed equal LOR in the sufficient ACJ stabilization outcomes.19 Compared with the
postoperative interval (Appendix Table A2). However, the double CC-stabilizing technique with dDBSs, the application
clinical score analysis revealed that patients with high- of a single double-button suture device showed increased
grade injuries (Rockwood IV/V) showed significant clinical rates of vertical (36% for 1 double-button suture device vs
benefits from the more extensive dDBS procedure as com- 13% for dDBS devices) and horizontal (26% vs 0%) instability
pared with the cHP treatment. For Rockwood III, there in the postoperative period.21 Jensen et al12 retrospectively
were no differences detectable (Appendix Tables A3 and showed, for the AC-stabilizing clavicular hook plate, similar
A4). clinical outcome data as compared with the double CC-stabi-
lizing technique with dDBS systems but applied no radiologic
CC distance assessments. The present study clarified that
the AC-stabilizing procedure resulted in inferior score out-
DISCUSSION come data for function, pain, and ACJ-specific score assess-
ment as compared with the double CC-stabilizing technique
The most striking finding of our study was that the arthro- (Table 3). At each point of postoperative assessment, the
scopically assisted CC-stabilizing dDBS technique created dDBS system showed superior score levels as compared
superior clinical outcomes as compared with the AC- with the cHP. Possible reasons for the clinical score differen-
stabilizing cHP procedure. Regarding our primary hypoth- ces include the LOR with persisting vertical and horizontal
esis, the clinical assessment arm revealed significantly instability, initially differing rehabilitation protocols, and
improved postoperative status for the dDBS procedure concomitant lesions. The indication for the surgical treat-
(Table 3, Appendix Table A3) without achieving MCIDs ment of acute ACJ separation grade 3 is controversial, but
(Table 4). With regard to our secondary hypothesis, both there exists consensus in the current literature for the surgi-
treatment groups revealed in the rCCD analysis a signifi- cal stabilization of acute ACJ injury grade 4/5.
cant LOR and significant differences as compared with The current Taft score subanalysis detected that the
the healthy control side. The radiologic assessments patients with a high-grade injury pattern showed signifi-
tended at all postoperative points of follow-up to smaller cant benefit from the more extensive dDBS procedure
rCCD of the dDBS group as compared with the cHP group, (Appendix Tables A2 and A3). The current literature lacks
but there were no significant differences detectable (Table 5, the analysis of MCID for treatments of ACJ separations.
Appendix Table A2). The secondary onset rate of osteoar- In particular, the patients with high-grade injuries (Rock-
thritic ACJ changes was significantly increased after the wood IV/V) showed significant superior clinical outcomes
cHP procedure versus the dDBS strategy. The subanalysis after the dDBS procedure as compared with the cHP treat-
of the high-grade instabilities (Rockwood IV/V) revealed ment (Appendix Tables A3 and A4). The MCIDs for the Taft,
that for these patients, the dDBS procedure enabled signifi- CS, and rCCD after surgical ACJ stabilization with an
cant benefits for the clinical score measurements. anchor-based method show that a 2.87-point change on
AJSM Vol. XX, No. X, XXXX Stabilization of Acute High-Grade ACJ Separation 7

TABLE 5
Analysis of the Absolute ACD, Absolute CCD, and Relative CCD for the cHP and dDBS Groupsa

cHP group dDSB group D Relative CCD (cHP-dDBS) P value

P0 (Preoperative)
Absolute ACD, mm 14.34 6 3.52 13.74 6 3.89 10.60 .23
Absolute CCD, mm 33.44 6 8.22 31.89 6 8.30 11.55 .28
Relative CCD, % 234.74 6 25.09 234.37 6 24.26 10.37 .68
P1 (Postoperative)
Absolute ACD, mm 15.56 6 3.17 14.71 6 4.48 10.86 .32
Absolute CCD, mm 16.51 6 3.88 15.44 6 5.08 11.07 .41
Relative CCD, % 105.59 6 8.55 104.14 6 8.55 11.45 .61
P2 (6 mo)
Absolute ACD, mm 14.44 6 4.60 13.20 6 4.24 11.23 .33
Absolute CCD, mm 17.87 6 6.33 15.41 6 4.97 12.46 .14
Relative CCD, % 124.14 6 24.51 119.82 6 25.07 14.32 .33
P3 (12 mo)
Absolute ACD, mm 13.82 6 3.55 14.26 6 5.53 -0.79 .63
Absolute CCD, mm 17.49 6 6.33 17.53 6 5.98 -0.05 .94
Relative CCD, % 127.41 6 39.40 124.42 6 26.98 12.99 .71
P4 (24 mo)
Absolute ACD, mm 14.21 6 5.42 13.23 6 3.67 10.97 .7
Absolute CCD, mm 19.38 6 6.42 16.92 6 4.16 12.46 .16
Relative CCD, % 141.81 6 31.20 130.69 6 22.25 111.13 .23
P Value
P0-P4 \.001 \.01
P1-P2 .01 .51
P2-P3 .89 .99
P3-P4 .54 .56
P1-P4 \.02 \.002

a
Data are reported as mean 6 SD, unless otherwise indicated. Relative CCD = 100 / absolute ACD 3 absolute CCD. Bold values are sta-
tistically significant (P \ .05). ACD, acromioclavicular distance; CCD, coracoclaviculare distance; cHP, clavicular hook plate; DDBS, double
double-button suture.

the Taft score, a 16.63-point change in the CS, and a 22.20% study, only 1 study group published radiograph-based
change in the rCCD assessment achieve clinically important ACJ-specific scores.17,28 Compared with cHP, the dDBS
improvements (Table 4). The present MCID analysis lets us system results in superior radiologic ACJ reduction (Table
conclude that the statistically significant differences in the 5), which appeared to be reflected by significantly
clinical score analysis between the two procedures were increased score outcomes (Table 3). Current longitudinal
not reflected by clinically apparent differences. With regard assessments26,28,34 detected, like the present data, that
to the MCID analysis, there existed no clinically meaningful the dDBS procedure had an atraumatic postoperative
advantage for the dDBS as compared with the cHP minor LOR up to the sixth postoperative month. For this
procedure. postoperative time interval, it can be hypothesized that
Analogous to the work of Pauly et al,22 in the present progressively repetitive stress loads after full weightbear-
study, the concomitant injuries were in 7% of the acute ing results in microtraumatic insufficiency of the dDBS
lesions and 28% of the chronic lesions or lesions with systems. In the present study, the AC-stabilizing cHP
unclear traumatic correlations. By performing the preoper- and the double CC-stabilizing technique with dDBS sys-
ative MRI in the early posttraumatic interval, the non- tems showed, in the entire postoperative interval, micro-
treated acute concomitant lesions can be excluded as an traumatic LOR (Table 5) without any significant
underlying reason for score impairments. Regarding the influence by grade of ACJ separation (Appendix Table
chronic and nontraumatic lesions, the data of Pauly et al A2). Regarding rCCD assessment for the LOR and the com-
and the present data let us suppose that both groups parison with the healthy control side, the cHP group
have similar percentages of chronic concomitant lesions. tended to have larger CC distance values as compared
Undetected chronic and nontraumatic intra-articular with the dDBS group. From the 6-month control, full
lesions in the cHP group can be discussed as an underlying weightbearing and full functional exposure for work and
reason for the present score difference. However, the score sport were permitted. The NAS measurement detected
differences were similarly reflected by the ACJ-specific a significant increased pain level for the cHP group after
Taft measurements. the 6-month controls, whereas the Taft score and the CS
The present literature analysis (Table 6) shows an recorded for each postoperative control significant showed
inconsistent outcome analysis. Aside from the present advantages for the dDBS group. Regarding the subdivision
8 Stein et al The American Journal of Sports Medicine

TABLE 6
Literature Overview (2010-2018) Focused on Clinical Trials Investigating Anatomic
and Nonanatomic ACJ-Stabilizing Proceduresa
Procedure Scores

Design; Treatment Injury to Follow-up, ACJ

Study Groups, n Nonanatomic Anatomic Surgery mo Specific Non-ACJ Specific Radiologic Assessments Findings

Darabos Randomized; 2 GI: Bosworth — < 2 wk Preoperative, — GI: CS (87.42), DASH (9.9), 1. Panorama view CCD vs Similar radiologic and
(2015)8 screw (n = 34); 6 mo Oxford Shoulder Score contralateral shoulder; clinical outcomes
GII: single TR (43.17); GII: CS (92.22), 2. MRI evaluation CC
(n = 34) DASH (6.46), Oxford ligament after 6 mo
Shoulder Score (44.59)
Horst Retrospective; 2 GI: K-wire (n = — 6 d (1-13) Preoperative, No clinical No clinical evaluation 1. True AP (ACD/CCD) 1. Good reconstruction
11
(2013) 11); GII: single 6 wk, evaluation preoperative/ alignment in both
TR (n = 16) 21 wk postoperative (6 of 21 wk groups; 2. increased
in the GI) vs the health ACD at the end in
side; 2. bilateral stress the TR group
view (10 kg) on trauma
day
Jensen Retrospective; 2 GI: hook plate (n GII: TR <3 wk GI: 48 mo; GI: modified GI: VAS (0.8), SST (11), 1. No radiograph; 2. 1. Good to excellent
12
(2013) = 30) (n = 26) GII: TAFT (10); CS (92.4); GII: VAS (0.4), sonographic CCD results in both
17 mo GII: modified SST (12), CS (94) unloaded and in stress groups without
TAFT (10) view (10 kg) vs the significant
healthy side differences; 2. no
significant
differences in CCDs
between groups; 3.
significant CCD
differences vs
healthy side in both
groups
—
Kraus Prospective; 1 GI: double TR V — GI: 24 mo; GI: TAFT (10.5), GI: SSV (96.2), CS (92.4); GII: 1. True AP of the affected 1. No significant
17
(2013) (n = 15); GII: GII: ACJI (75.9); SSV (93.9), CS (90.5) side, bilateral axillary differences in clinical
double TR II 24 mo GII: TAFT view, bilateral AP stress or radiologic results
(n = 13) (10.5), ACJI view (10 kg)
(84.5) posttraumatic; 2. CCD vs
the healthy side; 3.
bilateral Alexander view
Natera- Retrospective; 2 GI: hook plate — < 3 wk GI: 36 mo; — GI: physical SF-36 (53.7), 1. AP view (both shoulders); Patients with
Cisneros (n = 11); GII: GII: mental SF-36 (53.06), VAS 2. axillary view (only arthroscopically
19
(2016) singe TR 36 mo (1.45), DASH (4.79), CS injured shoulder); 3. no fixation with better
(n = 20) (91.36), global satisfaction evaluation of the CCD QoL
(8.0); GII: physical SF-36
(58.24) mental SF-36
(56.15), VAS (0.4), DASH
(2.98), CS (95.3), global
satisfaction (8.85)
Patzer Retrospective; 2 GI: Single TR GII: Double TR 6d Preoperative, — GI: CS, ASES, SST; GII: CS, 1. Panorama with Stress 1. Lower CC distance
21
(2013) (n = 14) (n = 15) 6 wk, ASES, SST (no values (10 kg) (preoperative with the double TR
3 mo, provided) and 3, 6, 12 mo vs the single TR; 2.
6 mo, postoperative); 2. no significant
12 mo Alexander view; 3. difference of CC
axillary view; 4. CCD distance and scores
vs healthy side

Salzmann Prospective; 1 — GI: Double TR < 3 wk Preoperative, — GI: VAS (0.25), CS (94.3), 1. Preoperative: axillary Satisfactory clinical
26
(2010) (n = 23) 6 mo, SST (12), SF-36 PCS view, scapular-Y view, results
12 mo, (56.2), SF-36 MCS (51.1) bilateral AP stress view;
24 mo 2. 6, 12, 24 mo: unilateral
stress view, axillary view;
3. CCD vs the healthy
side
Scheibel Prospective; 1 — 1. Double TR 7d 26.5 mo GI: TS (10.5), GI: SSV (95.1), CS (91.5) 1. Bilateral axillary view, AP 1. Good to excellent early
28
(2011) (n = 28) ACJI (79.9) stress view (10 kg); 2. CCD clinical results; 2.
vs the healthy side partial recurrent vertical
and horizontal AC joint
instability
Venjakob Prospective; 1 — GI: Double TR <3 wk GI: preoperative, — GI: VAS (0.3), CS (91.5), SST 1. Unilateral AP stress view; 2. Good to excellent early
34
(2013) (n = 23) 24 mo, (11.8), SF-36 PCS (57.9), unilateral axillary view; 3. clinical results after 58
58 mo SF-36 MCS (49.2) CCD vs data of Salzmann mo
26
et al
Vrgoc Retrospective (2 GI: K-wires + — GI: 6 d (0-21); GII: 12 mo — GI: VAS (1.27), CS (99), Radiologic analysis GI: shorter period for
36
(2015) centers); 2 FiberTape (n = 9 d (3-14) Oxford Shoulder Score complete recovery and a
10); GII: single (47.6); GII: VAS (1.33), significantly more cost-
TR (n = 6) CS (98), Oxford Shoulder effective outcome; GII:
Score (47.5) shorter operative
procedure, better
cosmetic result
Present Prospective; 2 GI: hook plate GII: dDBS GI: 6 d; GII: 9 d GI/GII: 6, 12, GI: TS (9.4); GI: CS (90), VASPAIN (2.4); ACD, CCD, relative dDBS: significantly
study (n = 27) (n = 29) 24 mo GII:TS GII: CS (95), CCD, panorama superior clinical
(2018) (10.9) VASPAIN (1.7) stress view outcome without
differences in radiologic
assessments and MCID

a
ACD, acromioclavicular distance; ACJ, acromioclavicular joint; ACJI, acromioclavicular joint injury; AP, anteroposterior; ASES, American Shoulder and Elbow Surgeons; CC, coracoclavicular; CCD,
coracoclavicular distance; CS, Constant-Murley score; DASH, Disabilities of the Arm, Shoulder and Hand; dDBS, double double-button suture; GI, group I (clavicular hook plate); GII, group II (dDBS);
MCID, minimal clinically important difference; MCS, mental component score; MRI, magnetic resonance imaging; PCS, physical component score; QoL, quality of life; TR, TightRope; TS, Taft score; SF-36,
Short Form–36; SST, Simple Shoulder Test ; SSV, Simple Shoulder Value; VAS, visual analog scale.
AJSM Vol. XX, No. X, XXXX Stabilization of Acute High-Grade ACJ Separation 9

of ACJ-stabilizing procedures,2 the present findings sug- instability were included (Table 1). Postoperative persisting
gest that the ‘‘anatomic’’ CC-stabilizing dDBS procedure horizontal and vertical residual ACJ instabilities were non-
enabled an increased functional status and a superior load- significant between groups (P = .231) and can be discussed
ing capacity status as compared with the AC-stabilizing as 1 underlying reason for the remaining clinical shoulder
cHP procedure. These data confirm current meta-analyses impairments in both groups. In situations of chronic residual
that revealed, for loop suspensory fixation, higher postop- AC instabilities, the medial scapula stabilization is
erative functional outcomes as compared with AC-stabiliz- decreased, and the patients show consecutive internal scap-
ing hook plate procedures, without differences in ula malrotation and scapulathoracic dysfunction. These
complication rates.1,19 Interestingly, the most deviant post- chronic glenohumeral malpositions increase general shoulder
operative LOR between the groups (D rCCD analysis in pain and additional functional limitations. These impair-
Table 5) occurred in the second half of the ments can even increase if the patients resume the full func-
follow-up period, when the postoperative rehabilitation tional shoulder exposure and loading capacity.14-16
protocol was exactly the same. Both groups had compara- Several limitations of the present study should be con-
ble rehabilitation protocols (Appendix Table A1), which dif- sidered. First, the rate of patients’ inclusion with intention
fered only in the first period and were similarly restrictive to treat was high, but all patients were thoroughly
when compared with the different rehabilitation protocols informed. Why patients chose the intention-to-treat arm
of the current literature (Table 6). From the fourth postoper- were, for the cHP group, the rehabilitation protocol and,
ative month, weightbearing and range of motion limitation for the dDBS group, the 1-step surgery. Second, horizontal
were similarly increased, and the presumptively healed CC stability was assessed by clinical estimation. The distribu-
and AC ligament structures are expected to stabilize the tion of Rockwood type III, IV, and V (Table 2) and the per-
ACJ. Regarding the influence of the interval between implan- sisting postoperative horizontal instability rates were
tation and removal of the hook plate, there exists no consen- equal between groups. Standardized radiologic stress
sus in the current literature. In several studies, the removal views were not established at the beginning of the present
of the implants was performed from the third postoperative study. Third, the acromioclavicular joint injury score was
month after the index procedure according to standard proto- not applied because the radiologic stress view of horizontal
col25,35; in a current meta-analysis, the fourth postoperative stability is part of this questionnaire. Fourth, concomitant
month was described.1 Analyzing the influence of the inter- lesions detected with different methods (preoperative MRI
val to hook plate removal, Jensen et al12 recorded, after and diagnostic arthroscopy), as well as the rehabilitation
hook plate removal in the fourth month, similar Taft score protocol at the preoperative education, have to be consid-
values as compared with the present cHP group with ered bias. The effect of the nondetected acute and chronic
removal in the third month (Table 3). In conclusion, it can concomitant lesions in the cHP group remain unclear
be supposed that even a more restrictive postoperative reha- because the initially diagnostic arthroscopy in the dDBS
bilitation protocol with an elongated interval for implant group is more sensitive than the native MRI assessment
removal would not affect the secondary LOR rates. in the cHP group. Standardized initially diagnostic
Standardized stress radiographs increase the value of the arthroscopies would minimize this limitation. Fifth, the
outcome assessment because stress views allow reproducible, patients’ phase of life with lots of social and professional
objective, and indirectly qualitative measurements of the lig- changes as well as the obligate radiologic assessment pro-
ament consolidation. In study protocols without radiologic tocol hindered equal intervals of follow-up.
assessments, postoperative LOR and secondary ACJ redislo-
cation remain undetected. Regarding the different and mar-
ginally varying techniques for the panoramic stress view
with 30° to 45° caudocranial tilting and the body height CONCLUSION
depending on differences of the aCCD, the direct comparison
This prospective study showed significantly superior out-
of aCCD includes systemic and interindividual errors. The
comes in all clinical scores for the double double-button
present evaluation of the rCCD neutralizes these systemic
suture technique as compared with the clavicular hook
and patient-related errors and tended to result in a smaller
plate technique. The subanalysis of the high-grade injuries
rCCD for the dDBS group without any significant difference
(Rockwood IV/V) revealed that these patients showed sig-
between treatment groups or in the analysis of the LOR. Hor-
nificant benefits from the dDBS procedure in the clinical
izontal stability is primarily generated by the superior AC
assessments. The clavicular hook plate procedure resulted
ligament and the trapezoid ligament.7,9 The persisting hori-
in good to excellent clinical outcome data too and displays
zontal instability rates after dDBS stabilization vary in the
an alternative procedure for patients with need for less
present literature, depending on the kind of assessment,
restrictive rehabilitation protocols.
between 17% and 43%4,27,34,37 (Table 6). The results of the
present clinically assessed persisting horizontal instabilities
confirm the findings of the literature: the cHP procedure
showed horizontal instability in 18.5%, as opposed to 6.9% ACKNOWLEDGMENT
for the dDBS group. The distribution of types of acute ACJ
separation according to Rockwood between groups (Table 2) The authors appreciate the assistance of Dr rer med Hanns
was not significantly different; furthermore, only patients Ackermann for analyzing the data and Dr Ing Tobias
with combined preoperative vertical and horizontal Bischof-Niemz for the assistance of the rCCD evaluation.
10 Stein et al The American Journal of Sports Medicine

REFERENCES 19. Natera-Cisneros L, Sarasquete-Reiriz J, Escola-Benet A, Rodriguez-

Miralles J. Acute high-grade acromioclavicular joint injuries treat-
1. Arirachakaran A, Boonard M, Piyapittayanun P, et al. Post-operative ment: arthroscopic non-rigid coracoclavicular fixation provides better
outcomes and complications of suspensory loop fixation device ver- quality of life outcomes than hook plate ORIF. Orthop Traumatol Surg
sus hook plate in acute unstable acromioclavicular joint dislocation: Res. 2016;102(1):31-39.
a systematic review and meta-analysis. J Orthop Traumatol. 20. Nuchtern JV, Sellenschloh K, Bishop N, et al. Biomechanical evalua-
2017;18(4):293-304. tion of 3 stabilization methods on acromioclavicular joint dislocations.
2. Beitzel K, Cote MP, Apostolakos J, et al. Current concepts in the Am J Sports Med. 2013;41(6):1387-1394.
treatment of acromioclavicular joint dislocations. Arthroscopy. 21. Patzer T, Clauss C, Kuhne CA, et al. Arthroscopically assisted reduc-
2013;29(2):387-397. tion of acute acromioclavicular joint separations: comparison of clin-
3. Beitzel K, Mazzocca AD, Bak K, et al. ISAKOS Upper Extremity Com- ical and radiological results of single versus double TightRope
mittee consensus statement on the need for diversification of the technique [in German]. Unfallchirurg. 2013;116(5):442-450.
Rockwood classification for acromioclavicular joint injuries. Arthros- 22. Pauly S, Kraus N, Greiner S, Scheibel M. Prevalence and pattern of
copy. 2014;30(2):271-278. glenohumeral injuries among acute high-grade acromioclavicular
4. Cohen G, Boyer P, Pujol N, Hamida Ferjani B, Massin P, Hardy P. joint instabilities. J Shoulder Elbow Surg. 2013;22(6):760-766.
Endoscopically assisted reconstruction of acute acromioclavicular 23. Rios CG, Arciero RA, Mazzocca AD. Anatomy of the clavicle and cor-
joint dislocation using a synthetic ligament: outcomes at 12 months. acoid process for reconstruction of the coracoclavicular ligaments.
Orthop Traumatol Surg Res. 2011;97(2):145-151. Am J Sports Med. 2007;35(5):811-817.
5. Constant CR, Murley AH. A clinical method of functional assessment 24. Rockwood CA, Young D. Disorders of the acromioclavicular joint. In:
of the shoulder. Clin Orthop Relat Res. 1987;214:160-164. Rockwood CJ, Matsen FA, eds. The Shoulder. Vol 2. Philadelphia,
6. Cook JB, Tokish JM. Surgical management of acromioclavicular dis- PA: WB Saunders; 1990:483-553.
locations. Clin Sports Med. 2014;33(4):721-737. 25. Rolf O, Hann von Weyhern A, Ewers A, Boehm TD, Gohlke F. Acromio-
7. Costic RS, Labriola JE, Rodosky MW, Debski RE. Biomechanical clavicular dislocation Rockwood III-V: results of early versus delayed
rationale for development of anatomical reconstructions of coraco- surgical treatment. Arch Orthop Trauma Surg. 2008;128(10):1153-1157.
clavicular ligaments after complete acromioclavicular joint disloca- 26. Salzmann GM, Walz L, Buchmann S, Glabgly P, Venjakob A, Imhoff
tions. Am J Sports Med. 2004;32(8):1929-1936. AB. Arthroscopically assisted 2-bundle anatomical reduction of acute
8. Darabos N, Vlahovic I, Gusic N, Darabos A, Bakota B, Miklic D. Is AC acromioclavicular joint separations. Am J Sports Med. 2010;
TightRope fixation better than Bosworth screw fixation for minimally 38(6):1179-1187.
invasive operative treatment of Rockwood III AC joint injury? Injury. 27. Salzmann GM, Walz L, Schoettle PB, Imhoff AB. Arthroscopic ana-
2015;46(suppl 6):S113-S118. tomical reconstruction of the acromioclavicular joint. Acta Orthop
9. Debski RE, Parsons IM, Woo SL, Fu FH. Effect of capsular injury on Belg. 2008;74(3):397-400.
acromioclavicular joint mechanics. J Bone Joint Surg Am. 28. Scheibel M, Droschel S, Gerhardt C, Kraus N. Arthroscopically assis-
2001;83(9):1344-1351. ted stabilization of acute high-grade acromioclavicular joint separa-
10. Di Francesco A, Zoccali C, Colafarina O, Pizzoferrato R, Flamini S. tions. Am J Sports Med. 2011;39(7):1507-1516.
The use of hook plate in type III and V acromio-clavicular Rockwood 29. Stucken C, Cohen SB. Management of acromioclavicular joint inju-
dislocations: clinical and radiological midterm results and MRI evalu- ries. Orthop Clin North Am. 2015;46(1):57-66.
ation in 42 patients. Injury. 2012;43(2):147-152. 30. Taft TN, Wilson FC, Oglesby JW. Dislocation of the acromioclavicular
11. Horst K, Dienstknecht T, Andruszkow H, Gradl G, Kobbe P, Pape HC. joint: an end-result study. J Bone Joint Surg Am. 1987;69(7):1045-1051.
Radiographic changes in the operative treatment of acute acromio- 31. Tashjian RZ, Deloach J, Green A, Porucznik CA, Powell AP. Minimal
clavicular joint dislocation—tight rope technique vs K-wire fixation. clinically important differences in ASES and simple shoulder test
Pol J Radiol. 2013;78(4):15-20. scores after nonoperative treatment of rotator cuff disease. J Bone
12. Jensen G, Katthagen JC, Alvarado L, Lill H, Voigt C. Arthroscopically Joint Surg Am. 2010;92(2):296-303.
assisted stabilization of chronic AC-joint instabilities in GraftRope 32. Tashjian RZ, Deloach J, Porucznik CA, Powell AP. Minimal clinically
technique with an additive horizontal tendon augmentation. Arch important differences (MCID) and patient acceptable symptomatic state
Orthop Trauma Surg. 2013;133(6):841-851. (PASS) for visual analog scales (VAS) measuring pain in patients treated
13. Jobmann S, Buckup J, Colcuc C, et al. Anatomic ligament consolidation for rotator cuff disease. J Shoulder Elbow Surg. 2009;18(6):927-932.
of the superior acromioclavicular ligament and the coracoclavicular lig- 33. Tubach F, Ravaud P, Baron G, et al. Evaluation of clinically relevant
ament complex after acute arthroscopically assisted double coracocla- changes in patient reported outcomes in knee and hip osteoarthritis:
vicular bundle stabilization [published online September 18, 2017]. Knee the minimal clinically important improvement. Ann Rheum Dis.
Surg Sports Traumatol Arthrosc. doi:10.1007/s00167-017-4717-1 2005;64(1):29-33.
14. Kibler WB. The role of the scapula in athletic shoulder function. Am J 34. Venjakob AJ, Salzmann GM, Gabel F, et al. Arthroscopically assisted
Sports Med. 1998;26(2):325-337. 2-bundle anatomic reduction of acute acromioclavicular joint separa-
15. Kibler WB, Dome D. Internal impingement: concurrent superior labral tions: 58-month findings. Am J Sports Med. 2013;41(3):615-621.
and rotator cuff injuries. Sports Med Arthrosc. 2012;20(1):30-33. 35. von Heideken J, Bostrom Windhamre H, Une-Larsson V, Ekelund A.
16. Kibler WB, Sciascia A. Current concepts: scapular dyskinesis. Br J Acute surgical treatment of acromioclavicular dislocation type V with
Sports Med. 2010;44(5):300-305. a hook plate: superiority to late reconstruction. J Shoulder Elbow
17. Kraus N, Haas NP, Scheibel M, Gerhardt C. Arthroscopically assisted Surg. 2013;22(1):9-17.
stabilization of acute high-grade acromioclavicular joint separations in 36. Vrgoc G, Japjec M, Jurina P, et al. Operative treatment of acute acro-
a coracoclavicular Double-TightRope technique: V-shaped versus par- mioclavicular dislocations Rockwood III and V—comparative study
allel drill hole orientation. Arch Orthop Trauma Surg. 2013;133(10):1431- between K-wires combined with FiberTape(Ò) vs TightRope Sys-
1440. tem(Ò). Injury. 2015;46(suppl 6):S107-S112.
18. Loriaut P, Casabianca L, Alkhaili J, et al. Arthroscopic treatment of 37. Walz L, Salzmann GM, Fabbro T, Eichhorn S, Imhoff AB. The ana-
acute acromioclavicular dislocations using a double button device: tomic reconstruction of acromioclavicular joint dislocations using 2
clinical and MRI results. Orthop Traumatol Surg Res. 2015;101(8): TightRope devices: a biomechanical study. Am J Sports Med.
895-901. 2008;36(12):2398-2406.

For reprints and permission queries, please visit SAGE’s Web site at http://www.sagepub.com/journalsPermissions.nav.