Keywords distal clavicle resection - rotator cuff repair - acromioclavicular joint osteoarthritis
- shoulder
Introduction
Acromioclavicular joint (ACJ) osteoarthritis is common, with a radiological prevalence
in the general population that can reach 90% in patients over 30 years old.[1 ] Disorders of ACJ are an usual cause of shoulder pain in primary care, affecting
patients of all ages and levels of activity and are often described in patients with
rotator cuff (RC) tears.[2 ] Common treatments in painful ACJ osteoarthritis are oral analgesics, such as acetaminophen,
oral nonsteroidal anti-inflammatory drugs, topical analgesics, lidocaine patches,
and cortisone articular injections. In patients that do not respond to these pharmacological
treatments, a surgical option is considered and arthroscopic resection of the distal
clavicle (DCR) is the most commonly performed technique.[3 ]
Different studies have correlated ACJ osteoarthritis to the evolution of RC tears.[4 ]
[5 ] Some authors have described that these anatomical alterations, protruding downward
below the level of the adjacent acromion, could be a cause of subacromial impingement
with a focal pressure on RC.[6 ]
[7 ]
[8 ]
[9 ]
[10 ] Also, if the pathogenesis of RC lesions is multifactorial, the biomechanical consequences
of osteoarthritic changes in ACJ are considered a relevant factor determining RC tears.[6 ]
[7 ]
Theoretically, the inferior bony spurs in osteoarthritis of ACJ could cause not only
a primary RC lesions but also a delayed tendon healing after RC repair (RCR).[11 ] Considering that a cuff re-tear could evolve into a massive RC tear and require
more demanding surgeries, many authors have proposed the execution of DCR during the
surgical procedure of RC tear repair, both in symptomatic and asymptomatic ACJ osteoarthritis.[12 ]
[13 ] Clinical consequences of DCR in these patients are actually discussed among orthopaedic
surgeons, without definitive indications from medical literature.[4 ]
[14 ]
[15 ]
The aim of this scoping review is to determine the clinical consequences of the association
of DCR to cuff tears repair in patients with ACJ osteoarthritis.
Methods
A scoping review of the literature was performed using the Pubmed/MEDLINE and Embase
databases to identify all studies reporting ACJ osteoarthritis in patients undergoing
RCR. Authors followed the Preferred Reporting Items for Systematic reviews and Meta-Analyses
extension for Scoping Reviews checklist for the scoping review. The following search
string were used: ((rotator cuff tear[Title/Abstract]) OR (rotator cuff[Title/Abstract])
OR (rotator cuff repair[Title/Abstract])) AND ((acromioclavicular joint arthritis[Title/Abstract])
OR (ac joint arthritis[Title/Abstract]) OR (ac joint[Title/Abstract]) OR (acromioclavicular
joint[Title/Abstract])) Medical Subject Headings terms were used for “rotator cuff,”
and “acromioclavicular joint.” The search was performed on March 1, 2017, and was
restricted to English language literature. A new literature search has been performed
on May 5, 2020 with similar keywords and restrictions.
All randomized controlled trials and prospective cohort studies were included, as
well as retrospective comparative trials and therapeutic case series. The main inclusion
criteria were studies that evaluated patients with a full or partial thickness RC
tear treated with arthroscopic repair and symptomatic or asymptomatic ACJ degenerative
arthritis. Reviews, meta-analyses, expert opinions, and editorial pieces were excluded.
Other exclusion criteria included animal studies, in vitro studies, and biomechanical
studies on human cadaver specimens.
All abstracts from the searches were reviewed by two independent authors, who applied
the study inclusion and exclusion criteria. In case of disagreement between reviewers
(R.C., F.R.), consensus was obtained through discussion and, in case of persistent
disagreement, a third reviewer (C.S.) was consulted and the study was included until
full-text review could be performed. All full-text articles were then evaluated, and
manually cross-referenced to ensure that all potential studies were included. Five
studies were included in the review.
The level of evidence of each article was assessed using the 2003 Journal of Bone and Joint Surgery definitions for orthopaedic publications.[16 ]
Information regarding author, data, and journal of publication, study design and level
of evidence, patient demographics, modality for ACJ osteoarthritis diagnosis, treatment
intervention, follow-up duration, preoperative and postoperative validated outcome
measures, failure rates, and evidence of tendon healing, tear size, were extracted.
Where possible, the compiled data from individual studies with the same outcome measures
were pooled together. Mean weighted values were calculated for these outcomes and
minimal clinically important difference has been used as a benchmark of the difference
in the clinical postoperative performance. The differences among patients' inclusion
criteria in the evaluated studies suggest caution regarding definitive clinical considerations
on this aspect. Demographic data were compiled to assess weighted mean ages across
groups.
Results
The flowchart describing the selection of included papers is reported [Fig. 1 ]. After application of inclusion and exclusion criteria, two retrospective studies
and three randomized controlled trial were identified.[11 ]
[17 ]
[18 ]
[19 ]
[20 ] Clinical studies reported results of 443 shoulders, with an average age of 60.48
years and a mean follow-up of 31.7 months ([Table 1 ]). The evaluated outcomes included the Constant score,[21 ] the American Shoulder and Elbow Surgeons (ASES)[22 ] score, and the visual analog scale (VAS) that was reported, at least, in 4 of the
5 studies analyzed. All included studies compared isolated RCR with RCR associated
to DCR. Many differences were found among preoperative patients' selection, RC tear
classifications, associated surgical procedures, and postoperative follow-up. This
relevant aspect made difficult to compare clinical outcomes or pooling data of different
cohort with a clear statistical relevance. For example, preoperative RC tears were
classified in three studies, with different scales. Razmjou et al classified cuff
lesions as “large or massive” (33% of patients) or “small or moderate” (67% of patients).[20 ] Oh et al classified RC lesions as small (< 1 cm, 6 patients), medium (1–3 cm, 17
patients), large (3–5 cm, 2 patients), massive (> 5 cm, 11 patients), or partial (3
patients).[19 ] Kim et al reported the mean tear size, comparing the two groups (27.2 ± 3.7 mm in
group 1 and 24.7 ± 3.1 mm in group 2).[18 ] The two others studies did not report information about preoperative RC tears size.[17 ]
[18 ] The analyzed studies used different ACJ osteoarthritis classifications: the Collin's
system by Razmjou et al, the Stein magnetic resonance imaging (MRI) classification
by Kim et al (only patients with a grade 4 were included), and the Cook palpatory
tenderness scale by Park et al.[11 ]
[18 ]
[20 ] Only two studies, Oh et al and Kim et al, had a comparable population of asymptomatic
ACJ osteoarthritis patients, randomized in two groups, the first of DCR and RCR and
the second only RCR.[18 ]
[19 ] Acromioplasty was performed in 95% of patients with DCR in Razmjou et al study and
63% of the control group.[20 ] Kim et al performed a double row fixation in all RCRs, while all the other surgeons
performed a single row repair.[18 ] A lidocaine test was completed only in the study of Park et al, while other papers
only included clinical assessment of pain in this district.[11 ] Postoperative radiological outcomes and re-tear rate were evaluated by Oh et al
and Park et al: Oh et al analyzed radiological outcomes using ultrasonography, computed
tomography arthrography, MRI, and weighted stress radiography; Park et al performed
a MRI 6 months postoperatively and at final follow-up.[18 ] Results on tendon healing are reported by Oh et al and Park et al. If Oh et al described
the percentage of repaired cuffs healed at follow-up (77% with DCR; 74% without DCR),
Park et al reported the percentage of re-tear at follow-up, comparing patients with
and without DCR (9.5% vs. 15.3%). No statistical differences were found between the
groups (p = 0.95 in the first study, p = 0.678 in the second study), with comparable outcomes on tendon healing and reoperation
rate at final follow-up in both the reports.[18 ]
[19 ] Notably, the group of Oh et al patients were asymptomatic for AC pathology before
surgery while in the population of Park et al patients reported specific symptoms
related to ACJ.[18 ]
[19 ]
Table 1
Studies included in the review
Authors
Year
Study design
Level of evidence
No. shoulders
Sex no.
(male/female)
Age (y)
mean ± SD
Final follow-up (mo)
mean ± SD
Razmjou et al (Knee Surg Sports Traumatol Arthrosc)
2015
Retrospective outcome study
II
184
112/72
62 ± 9
24
Park et al (Am J Sports Med)
2015
Randomized controlled trial
I
58 (47 analyzed)
9/38
62.6 ± 9.31
44.45
61.7 ± 6.12
Oh et al (Am J Sports Med)
2014
Randomized controlled trial
I
78
44/34
63.6 ± 6.5
29.2
64.0 ± 7.5
Błasiak et al (Pol Orthop Traumatol)
2013
Retrospective outcome study
III
40
43/17
53.10 ± 13.60
54.4 ± 6.4
Kim et al (Arch Orthop Trauma Surg)
2011
Randomized controlled trial
II
83
40/43
59.8 ± 5.2
31.7
55.2 ± 6.1
Abbreviation: SD, standard deviation.
Note: Year of publication, study design, level of evidence, and patients information
are described.
Fig. 1 Flowchart of the literature review.
Considering the minimal clinically important difference for Constant score (10.4 points),
the overall analysis of the reported data showed a relevant increase in the mean Constant
score after surgery for both groups (RCR: + 29.7 points; RCR + DCR: + 39.2 points).[23 ] On the contrary, the difference in the postoperative results between the two groups
did not appear significant (Δ: 6.54 points). As reported in [Table 2 ], similar results have been found for the VAS and the ASES scores.[24 ]
[25 ]
Table 2
Clinical results in RCR and RCR + DCR groups
Increase in post-operative results
Difference in post-operative results between the two group
Minimal clinically important difference
RCR
RCR + DRC
Constant
29.7
39.2
6.5
10.4
VAS
–5.3
–5.9
–0.4
1.4
ASES
36.4
38.5
1.7
from 12 to 17
Abbreviations: ASES, American Shoulder and Elbow Surgeons Score; DCR, distal clavicle
resection; RCR, rotator cuff repair; VAS, Visual Analogue Scale.
Some differences were found regarding clinical scores used for measuring clinical
outcomes, and the most utilized were Constant, VAS, and ASES scores ([Table 3 ]). These scores were used differently among the groups, and this aspect could be
considered another bias that makes it difficult to compare outcomes of different studies.
Table 3
Clinical outcomes evaluated in the included studies
Authors
Year
Study groups
No.
subjects (%)
Outcomes
Razmjou et al (Knee Surg Sports Traumatol Arthrosc)
2015
DCR + RCR
144 (78)
ASES, Constant scores, strength
RCR
40 (22)
Park et al (Am J Sports Med)
2015
DCR + RCR
26 (21 analyzed)
Clinical outcomes: VAS, ASES, Constant scores, ROM
Radiological outcomes: MRI
RCR
32 (26 analyzed)
Oh et al (Am J Sports Med)
2014
DCR + RCR
39
Clinical outcomes: VAS, ASES, Constant scores, ROM
Radiological outcomes: Ultrasonography, computed tomography arthrography, MRI, weighted
stress radiography
RCR
39
Błasiak et al (Pol Orthop Traumatol)
2013
DCR + RCR
20
VAS, Constant scores
RCR
20
Kim et al (Arch Orthop Trauma Surg)
2011
DCR + RCR
31
VAS, UCLA, ASES scores, tenderness on ACJ, and cross-body adduction test
RCR
52
Abbreviations: ACJ, acromioclavicular joint; ASES, American Shoulder and Elbow Surgeons
Score; DCR, distal clavicle resection; MRI, magnetic resonance imaging; RCR, rotator
cuff repair; ROM, range of motion; SD, standard deviation; UCLA, The University of
California at Los Angeles Shoulder Score; VAS, visual analog scale.
Discussion
The aim of this study was to perform a systematic review on the management of ACJ
osteoarthritis in patients undergoing arthroscopic surgery for RC tears. Results of
this literature review underline a lack of evidence-based recommendations about the
management of ACJ in these patients, with low quality studies and a weak level of
recommendations. Current literature does not clarify the influence of ACJ osteoarthritis
on evolution of RC tears and the risk of evolution of the symptoms in patients with
asymptomatic shoulder in patients undergoing arthroscopic surgery. Osteoarthritis
of the ACJ is a common source of shoulder pain and a proper diagnosis requires a thorough
physical exam, plain-film radiograph, and, for a more accurate diagnosis, a local
anesthetic injection.[26 ] Nevertheless, clinical relevance of radiographic signs of osteoarthritis is not
clear, with some patients with well-defined degenerative imaging that do not report
any pain in this district. The clinical evolution of this condition is discussed,
without definitive conclusions, in a recent study regarding asymptomatic ACJ osteoarthritis
diagnosed with MRI.[27 ] Patients included in this study were followed for 7 years to evaluate the progression
of this pathology; the authors observed that asymptomatic acromioclavicular osteoarthritis
remained asymptomatic in 83% of cases, 7% turned better, and 10% turned worse.[27 ] The relevance of subacromial decompression on evolution of ACJ osteoarthritis is
also an interesting aspect. Some authors examined the differences in the reoperation
rate on the ACJ after arthroscopic subacromial decompression with and without concomitant
ACJ surgery. Authors concluded that violation of the ACJ during the initial surgery
by coplaning or DCR did not alter the reoperation rate due to ACJ symptoms.[28 ]
Focusing on the results of this review, in one study, clinical symptoms were not clearly
reported.[20 ] Oh et al found comparable results in nonsymptomatic patients undergoing lateral
clavicle resection plus RCR and patients treated with only RCR.[19 ] Similar results, but in patients with symptomatic ACJ osteoarthritis previously
examined with a lidocaine test, were found by Park et al in a prospective randomized
study.[11 ] Razmjou et al report that mild ACJ osteoarthritis is an independent negative predictor
of outcome at 2 years. This consideration supports the execution of DCR in patients
with mild ACJ osteoarthritis.[20 ]
This study presents two relevant limits: symptoms related to ACJ are not clearly described
in part of the included studies and different classifications of RC tears are used
in the included manuscripts. Moreover, the studies' design was different, with some
authors comparing treatment of symptomatic and nonsymptomatic AC joint osteoarthritis.
Błasiak et al described good results in patients undergoing DCR compared with a control
group, but the study has a small number of patients and group population is not the
same.[17 ] Another bias in this study, which should be considered, is the routine execution
of acromioplasty in the study published by Oh et al.[19 ]
In general, the surgical approach to ACJ osteoarthritis, when a RC tear repair is
performed, is not well studied in recent international literature. Studies have many
biases that limit data managements, making it difficult to extrapolate their application
to clinical practice. Considering the quality of the analyzed papers, Park et al and
Oh et al performed prospective randomized studies with high levels of evidence, ensuring
the best quality on this topic. They reported no difference between operative and
nonoperative management of ACJ osteophytes in RC tears.[18 ]
[19 ] Consistently with those results, this literature review draws similar conclusion
and shows no clinical differences in the two approaches. Surgical procedures such
as DCR should be performed carefully in this cohort of patients. Nevertheless, more
prospective randomized studies are needed to reach a consensus about the correct surgical
approach to DCR in patients with signs of ACJ osteoarthritis and RC tears.
