Evaluation of lengthening in reverse shoulder arthroplasty comparing X-rays and computerised tomography

The correct amount of arm lengthening in reverse shoulder arthroplasty is crucial to provide joint stability and good results. Determination has been proposed according to radiographs. However, radiographic measurements are error prone in regards to positioning of the arm with regard to the radiographic beam. The purpose of this study was to evaluate the precision of radiographic measurements compared to CT scans of the upper limb following reverse shoulder arthroplasty.

Thirty patients undergoing onlay reverse shoulder arthroplasty with comparative radiographs and CT scans of both humeri were included in this study. Arm length, humeral length as well as the arm and humeral lengthening were evaluated on pre- and postoperative radiographs compared to postoperative CT scans following a previously validated protocol.

We found an excellent correlation for arm length and humeral length for radiographic and CT measurements (r > 0.90). The postoperative humeral and arm lengthening compared to the contralateral side was 0.1 (-1.2 to 1.2) cm and 2.8 (0.2 to 5.2) cm for the CT scans, and -0.6 (-4.1 to 2.0) cm and 1.9 (-2.3 to 5.0) cm for the radiographs. For arm lengthening, correlation coefficient was good (r = 0.7) even though radiographs indicated arm shortening in five cases whereas arm lengthening was observed on CT scans.

Measurements on radiographs and CT scans are comparable in most of the cases. However, we observed some important variations that question the reliability of radiographic measurements in up to 20 % of cases. Therefore, a CT scan appears to be necessary in the event of postoperative complications (e.g., instability, neurological problems).