Sunday, May 15, 2016

3D image intensifier navigation of K-wire placement in the glenoid - statistical vs clinical significance

Improved accuracy of K-wire positioning into the glenoid vault by intraoperative 3D image intensifier-based navigation for the glenoid component in shoulder arthroplasty

These authors hypothesized that navigated k-wire placement is more accurate and shows a smaller deviation angle to the standard centerline compared to the classical “free hand technic” in thirty-four pairs of fresh female sheep scapulae

We include some details of the method:
First, the shoulders from the navigated group 1 were mounted onto the operating table in the upright position. A small incision wasmade in the area of the scapular spine and a carbon-clamp withthree tracking points was fixed. A preoperative scan was performed with Ziehm VisionFD Vario 3D. The middle of the scanner was placed at the central point of the glenoid. The scan performed consisted of 110 single images witha radius of 136◦within 110 seconds. The DICOM-format raw datawere consequently transferred from the 3D image converter to theVectorVision©navigation system. In the navigation system, the data record was used as a CT data record. After creating a 3D image, the data was verified by comparing the image and the anatomy. This control was performedat three defined points (proximal glenoid, central glenoid and the distal glenoid).The instruments were calibrated using the instrument calibration matrix. A 1.8-mm navigable drill sleeve was used. By using the navigation screen, the wire wasplaced as centrally as possible. Subsequently, a K-wire was positioned and shortened to fit into the glenoid. 
In the freehand group 2, the shoulders were stabilized in the upright positionon the operating table. In contrast to group 1, the central point of theglenoid vault was determined visually using electrocauterization. Subsequently, drilling was also performed with the same navigable 1.8 mm sleeve in a perpendicular direction to the glenoid face planeand marked with a K-wire.

 The relation to glenoid standard and alternative centerlines (CL) and the position within the glenoid vault were analyzed.

In groups 1 and 2 the angle between the K-wire and standard CL was 2.2° and 4.7°, respectively (P = 0.01). The angle between the K-wire and alternative CL was 14.4° for group 1 and 17.2° for group 2 (P = 0.02). More navigated K-wire positions were identified within a 5 mm corridor along the glenoid vault CL (52 vs. 39; P = 0.004).

Comment: This article represents yet another method for placing a K-wire in the scapula, ostensibly to guide the preparation of the glenoid for glenoid component placement.

We do not know the cost and time for implementing this approach, whether it would be applicable in the clinical situation or whether the improvement of three degrees would lead to better results for patients than other methods.

As is well know to the readers of this blog, we do not use a guide wire, intaoperative imaging, or patient specific instrumentation to guide the reaming of the glenoid. By contrast (as shown in "D" below), we position a nubbed reamer without a guide wire so that the minimal amount of bone is removed without an attempt to 'correct' glenoid version. Using this method and soft tissue balancing without or with an anteriorly eccentric humeral head component (see this link), we have yet to experience problems with posterior instability or glenoid component failure. In this nice figure from Dr Lippitt, one can see some other approaches and their potential downsides.