Anatomic and non-anatomic reconstruction improves post-operative outcomes in chronic acromio-clavicular instability: a systematic review.

ABSTRACT

PURPOSE: To systematize the surgical outcomes of anatomic and non-anatomic reconstruction in patients with chronic acromio-clavicular joint (ACJ) instability and determine which technique is superior. METHODS: This review was conducted according to the PRISMA guidelines. PubMed and Cochrane Library databases were searched up to April 30th, 2018 for original articles that assessed the outcomes of one or more surgical techniques of anatomic and non-anatomic reconstruction in patients with chronic ACJ instability. The Methodological Index for Non-randomized Studies (MINORS) was used to assess study quality. Pre-to-post meta-analysis was performed for both anatomic and non-anatomic reconstructions. RESULTS: Twenty-eight studies were included comprising 799 patients (mean age of 36.6 years old and 84% males) with a mean follow-up of 34.6 months (range 13 to 74). Surgical techniques included anatomic (tendinous or synthetic grafts/constructs) and non-anatomic reconstruction (Weaver-Dunn or Modified Weaver-Dunn, conjoined tendon transfer, or temporary hook plate). There were significant pre-to-post improvements on the constant score with an average improvement ranging from 11.1 to 50.7 (p < 0.01). Average failure rate was 7.6% (7.5% for anatomic and 8.5% for non-anatomic reconstruction). Non-comparative studies had a mean MINORS score of 9 points (out of 16) and comparative studies 17 (out of 24) with excellent interrater agreement (k = 0.910). CONCLUSION: Both anatomic and non-anatomic ACJ reconstructions provide significant post-operative improvements, but definitive conclusions on optimal technique remain elusive. Notwithstanding, comparative studies support the use of anatomic ACJ reconstruction which should be preferably used. However, until superiority is demonstrated by robust studies, surgeons should supplement their decision-making with experience and patient preference. LEVEL OF EVIDENCE IV.