Surgical decision making in shoulder arthroplasty

The surgeon makes critical decisions for each patient, striving for the optimal shoulder comfort and function. These decisions include those involving prosthesis selection and positioning.

As pointed out in preoperative planning for anatomic total shoulder and ream and run arthroplasty, aspects of planning can be accomplished using standardized plain films and the available PACs tools. However, image-based preoperative planning (whether based on plain films or on 3D CT scans) does not predict important characteristics of the shoulder that are only evident intraoperatively after surgical exposure and osteophyte resection.

Let’s consider two examples from this week’s surgical schedule.

Case 1- a 66 year old active man with a successful ream and run for right shoulder. arthritis who presented for a left sided ream and run because of refractory functional limitations. The shoulder was not especially stiff preoperatively (140 degrees of active elevation).

Preoperative x-rays showed osteoarthritis with mild-moderate posterior decentering seen on the axillary truth view.

Preoperative image-based planning showed the ideal head cut and suggested a concentric humeral head implant with 56 mm diameter and a thickness of 24 mm.

At surgery, however, after glenoid reaming the suggested head size overstuffed his soft tissues such that range of motion was excessively limited. Trialing with a 50 mm diameter 20 mm thick concentric head component provided the desired mobility, but this implant was posteriorly unstable when the arm was lifted into flexion.  An anteriorly eccentric 50 20 humeral trial provided both the desired mobility and stability. His postoperative x-rays are shown below.

The learning point here is that preoperative imaging could not have predicted the correct implant – that decision had to rest on intraoperative trialing.

The patient kindly gave us permission to show this video of his assisted motion three days after his ream and run.

Case 2: A 76 year old man, 5 ft 3 in with cuff tear arthropathy, pseudoparalysis and osteoporosis – all placing him at increased risk for instability and acromial/spine stress fractures.  His preoperative Grashey view is shown below.

Because of his pseudo paralysis, he was not a candidate for a CTA hemiarthroplasty (see Cuff tear arthropathy-current considerations); he elected to proceed with a reverse total shoulder. The surgical goals were to optimize stability without excessive lateralization or distalization and with glenosphere inferior tilt to match the orientation of the line (yellow) representing the base of the suprascapularis fossa.

Achieving this goal is a challenge: (1) the orientation of this reference line is not discernible at surgery and (3) the inferior prominence of glenoid would confound the use of a guide to orient the drill for the central screw. 

In this case a surgically visible landmark was the plane of the central glenoid (black line). A line (red) perpendicular to this plane provides a reference of zero degrees of inferior tilt.

The angle between the red and yellow lines indicates the desired orientation of the drill to achieve the desired inferior tilt of the baseplate.

The superior-inferior position of the baseplate was selected so that after reaming the inferior glenoid would be covered by the glenosphere.

The final reconstruction is shown below using a 36 mm glenosphere (selected intraoperatively because the 32 mm glenosphere did not provide sufficient stability to vigorous testing), a semiconstrained humeral liner both for stability and to minimize humeral lateralization, and a long smooth impaction grafted stem to avoid varus position of the humeral component (see Dislocation of the reverse total shoulder).

In these two cases the combination of preoperative image-based planning and intraoperative trialing led to the final choice of implant components and positioning.  

Recognizing that there is more than one way to skin a cat(bird).

it will be interesting to know the readers’ thoughts on alternate approaches.