Summary

AAGR using DTA has a short learning curve regardless of fixation type, with button fixation resulting in a shorter operative time

Abstract

Background

Arthroscopic anatomic glenoid reconstruction (AAGR) using distal tibia allograft (DTA) has been previously described as a technique used to treat recurrent anterior shoulder instability with bone loss. The original technique used screw fixation and more recently, a technique has been described using double round buttons. The purpose of this study was to compare the learning curve and graft positioning of screw fixation with button fixation for AAGR using DTA.

Methods

We performed a retrospective review on consecutive patients with prospectively collected data who underwent AAGR with button fixation from 2018 to 2020. The button fixation patients were compared to the first set of patients who received AAGR with screw fixation. The learning curve data was previously published for the screw technique for the first 27 patients who received this procedure. All patients received AAGR from the senior author for anterior shoulder instability with bone loss and had a short-term post-operative computed tomography (CT) scan. Exclusion criteria included patients with rotator cuff pathology, multidirectional instability and glenoid fractures. There were 43 patients (27 screw and 16 button) who met the inclusion/exclusion criteria. We collected the surgical times of each procedure and assessed the vertical and horizontal positioning of the allograft using post-operative CT scans. We also assessed differences in alpha angle between the two groups. Button and screw patients were placed in three clusters to determine whether there were any differences in surgical time or graft positioning over the course of the learning curve.

Results

The average operative time for the button and screw fixation groups were 1.35±0.22 and 1.55±0.32, respectively. The button fixation groups showed a significant reduction in operative time compared to the screw fixation group (p=0.029, a=0.05). The equality of variance test demonstrated that none of the clusters had a significant difference with respect to surgical time between both groups. The graft positioning for both groups was on the lower third (button: n=14; screw: n=24) and middle third (button: n=2; screw: n=3) with no significant difference between the groups (p=0.891, a=0.05). Our analysis of horizontal position shows that button fixation results in good positioning with respect to medial-lateral step formation, with slightly better positioning being achieved using screw fixation (p=0.033, a=0.05). Compared to the screw fixation group, the button fixation group had better alpha angles (p<0.001, a=0.05), likely resulting from the use of a surgical guide for placement of fixation.

Conclusion

AAGR using DTA has a short learning curve regardless of fixation type, with button fixation resulting in a shorter operative time. The surgical times for both fixation types are comparable to existing arthroscopic stabilization techniques such as the arthroscopic Bankart. Both button and screw fixation result in good post-operative graft positioning. Future studies investigating patient-reported outcomes, post-operative recurrence of instability, and longer-term radiographic outcomes are necessary to determine whether screw and button fixation have the same post-operative outcomes.