Monday, February 8, 2016

What is the value of patient-specific instrument guidance in shoulder arthroplasty?

Patient-specific instrument guidance of glenoid component implantation reduces inclination variability in total and reverse shoulder arthroplasty.

The authors present 36  shoulder arthroplasties (12 TSAs, 24 RSAs) of which 18 procedures (6 TSAs, 12 RSAs) were executed using preoperative 3D planning and patient-specific (PSI) guides to position the central guide pin for glenoid component implantation. In 9 cases, the glenoid anatomy was severely distorted through wear or previous surgery. 

The inclination of the glenoid component was measured using the angle between the glenoid baseplate and the floor of the supraspinatus fossa (angle β) on postoperative radiographs.

For TSA, the average angle β was 74 ± 9 in the PSI group and 86 ± 12 in the non-PSI group. 
For RSA, the average angle β was 83 ± 7 in the PSI group and 90 ± 17 in the non-PSI group. 

The x-rays below show the subtle difference between a reverse glenoid inserted without PSI (left) and with PSI (right).

Extreme values of glenoid component inclination were more likely to occur in the non-PSI group than in the PSI group (P < .001 for TSA; P = .02 for RSA).

Comment: A few days ago we posted on an article stating that accurate measurement of glenoid inclination could not be made without 3D reconstruction of CT scans (link).

Here we have a 'case controlled' study, comparing glenoids placed with and without patient-specific instrument (PSI) guidance. However, the authors do not explain how patients were assigned to the two groups or whether the pathoanatomy was comparable. The clinical outcomes of the two groups are not compared. The increment in cost, preparation time and operative time for the PSI method is not provided. As a result, the value (benefit/cost) cannot be determined and we cannot be sure that the patient whose x-rays are shown on the right above will fare better than the one whose x-rays are shown on the left.

As the authors point out, there are many variables other than inclination (AP and superior/inferior orientation of the starting point, anterior and posterior inclination, bone quantity and quality, glenoid component geometry) to be considered in the selection and positioning of a glenoid component. They also point out that 
"(1) the manufacturing the PSI guide depends on an accurately defined surgical plan on a correct 3D model of the patient’s scapula. Segmenting an accurate 3D model  of the scapula, however, can be difficult because of cartilage loss, severe bone deformity, calcified labrum, and difficulty in segmenting glenoid bone from humeral bone.
(2) fitting the PSI guide on the glenoid and finding its stable and correct position depend on a meticulous exposure  of the anterosuperior glenoid rim with sufficient removal of the soft tissues.and 
(3)  the accuracy of the system is also dependent on the surgeon’s ability to ream to the proper depth in line with the central guide pin."

A final set of questions emerge in regard to such complex technologies: 
is it the recommendation that such a device should be used on all shoulder arthroplasties?
is the technology more appropriate for high volume or for less experienced surgeons?
what is the learning curve?
where does the money come from to cover the incremental costs?
if we learn to depend on such a system, what happens if it is not available?

We will have to ask and answer these questions for each new technology that is brought forward.