Immediate postoperative radiographs after shoulder arthroplasty are often poor quality and do not alter care

It is technically difficult to obtain high-quality, postoperative shoulder radiographs immediately after surgery. Further, poor-quality radiographs may be unlikely to change clinical practice or improve patient outcomes. We therefore questioned the value of routine postoperative radiographs after shoulder arthroplasty.

We asked whether (1) postanesthesia care unit (PACU) radiographs can reasonably serve as a baseline for future studies; and (2) routine PACU radiographs change clinical care. (3) We also determined the charges associated with routine PACU radiographs and formal radiographic interpretation of these images.

We retrospectively compared the radiographs of 283 patients who had shoulder arthroplasties (Group 1) who underwent PACU radiographs with those of 241 patients (Group 2) who had their first postoperative radiographs at a later date. Radiographs were compared for quality, ability to serve as a baseline, and their influence on clinical course. Orthopaedic evaluation of each radiograph and the radiographic report were compared and charges were analyzed.

All images in Group 1 were single-view radiographs (88% internal rotation), most were underpenetrated (71%); no images changed postoperative management or were considered adequate to serve as a baseline. Group 2 radiographs were multiview radiographs, and 83% were deemed adequate to serve as baseline radiographs. Radiographic interpretation of immediate postoperative radiographs did not change the clinical course or treatment. The charges billed from radiographic evaluation in this study were $64,524 for Group 1.

Routine PACU radiographs, in the absence of a specific indication, may result in poor-quality images. Elimination of these radiographs and radiographic interpretation after shoulder arthroplasty may reduce charges without changing clinical care.

Level III, diagnostic study. See Guidelines for Authors for a complete description of levels of evidence.