Is There an Association Between Postoperative Internal Rotation and Patient-reported Outcomes After Total Shoulder Arthroplasty?

Background: Deficient internal rotation after shoulder arthroplasty can inhibit specific essential activities of daily living that require behind-the-back arm positioning. Although postoperative internal rotation deficits occur, their impact on outcomes of total shoulder arthroplasty (TSA) is not well established. Previous authors have validated the Single Assessment Numeric Evaluation (SANE) as a patient-reported assessment of acceptable outcomes of TSA.

Questions/purposes: (1) Is there an association between postoperative internal rotation and acceptable outcomes following TSA as assessed by SANE? (2) Is there a threshold for internal rotation after TSA beyond which increasing internal rotation no longer improves odds of acceptable outcomes?

Methods: A single institution's longitudinally maintained shoulder arthroplasty registry was used to identify patients undergoing primary anatomic or reverse TSA (RTSA). The registry provides postoperative patient-reported outcomes, including SANE scores. Postoperatively, patients complete a previously validated ROM self-assessment to quantify their current abduction, forward elevation, external rotation in adduction and abduction, and internal rotation in adduction. Data on patient age, preoperative Patient-Reported Outcomes Measurement Information System mental health scores, gender, surgery performed, and hand dominance were also obtained. In all, 784 patient-reported surveys were available. Thirty-four percent (268 of 784) of the surveys were collected at 1 year, 52% (410 of 784) at 2 years, 11% (87 of 784) at 5 years, and 2% (19 of 784) at 10 years. More than 50% percent (446 of 784) of patients underwent RTSA, 48% were men, and the mean ± SD age was 68 ± 8 years at the time of surgery. A logistic multivariate analysis was used to assess the association of internal rotation with an acceptable outcome (defined as a SANE score of > 75%). A receiver operating characteristic curve was used to assess an internal rotation threshold associated with an acceptable SANE score.

Results: After accounting for age, gender, hand dominance, pain level, and surgical procedure, patients with internal rotation below the upper back had lower odds of achieving a SANE score of > 75% (p < 0.05). The threshold for SANE scores > 75% was identified to be internal rotation to the midback and higher, resulting in an area under the curve of 0.71 (95% CI 0.67 to 0.75; p < 0.001) with sensitivity of 57% (95% CI 0.56 to 0.58) and specificity of 75% (95% CI 0.73 to 0.77).

Conclusion: After shoulder arthroplasty, shoulder normalcy was associated with postoperative internal rotation. Although our study has not proven a causal relationship between limited internal rotation and poorer SANE scores, our clinical experience combined with these findings suggests that limited internal rotation indeed is likely clinically important in this context, and so addressing postoperative internal rotation deficits, especially for RTSA, may improve the quality of shoulder arthroplasty. Further research is needed to understand the modifiable factors that prevent sufficient internal rotation following shoulder arthroplasty.

Level of evidence: Level III, therapeutic study.