Preoperative CT planning of screw length in arthroscopic Latarjet

The Latarjet procedure has shown its efficiency for the treatment of anterior shoulder dislocation. The success of this technique depends on the correct positioning and fusion of the bone block. The length of the screws that fix the bone block can be a problem. They can increase the risk of non-union if too short or be the cause of nerve lesion or soft tissue discomfort if too long. Suprascapular nerve injuries have been reported during shoulder stabilisation surgery up to 6 % of the case. Bone block non-union depending on the series is found around 20 % of the cases. The purpose of this study was to evaluate the efficiency of this CT preoperative planning to predict optimal screws length. The clinical importance of this study lies in the observation that it is the first study to evaluate the efficiency of CT planning to predict screw length.

Inclusion criteria were patients with chronic anterior instability of the shoulder with an ISIS superior to 4. Exclusion criteria were patients with multidirectional instability or any previous surgery on this shoulder. Thirty patients were included prospectively, 11 of them went threw a CT planning, before their arthroscopic Latarjet. Optimal length of both screws was calculated, adding the size of the coracoid at 5 and 15 mm from the tip to the glenoid. Thirty-two-mm screws were used for patients without planning. On a post-operative CT scan with 3D reconstruction, the distance between the screw tip and the posterior cortex was measured. A one-sample Wilcoxon test was used to compare the distance from the tip of the screw to an acceptable positioning of ±2 mm from the posterior cortex.

In the group without planning, screw 1 tended to differ from the acceptable positioning: mean 3.44 mm ± 3.13, med 2.9 mm, q1; q3 [0.6; 4.75] p = 0.1118, and screw 2 differed significantly from the acceptable position: mean 4.83 mm ± 4.11, med 3.7 mm, q1; q3 [1.7; 5.45] p = 0.0045. In the group with planning, position of screw 1 or 2 showed no significant difference from the acceptable position: mean 2.45 mm ± 2.07 med 1.8 mm, q1; q3 [1; 3.3] p = 1; mean 2.75 mm ± 2.32 med 2.3 mm, q1; q3 [1.25; 3.8] p = 0.5631.

Unplanned Latarjet can lead to inaccurate screw length especially in the lower screw and can increase the risk of non-union and nerve damage. The clinical relevance of this article is that CT planning of screw length before surgery showed good results on post-operative CT.