Discordance between patient-reported and objectively measured internal rotation after reverse shoulder arthroplasty

Background Patient satisfaction after reverse shoulder arthroplasty (RSA) partly relies on restoring functional internal rotation (IR). Although postoperative assessment of IR includes objective appraisal by the surgeon and subjective report from the patient, these evaluations may not vary together uniformly. We assessed the relationship between objective, surgeon-reported assessments of IR and subjective, patient-reported ability to perform IR-related activities of daily living (IRADLs).Methods Our institutional shoulder arthroplasty database was queried for patients undergoing primary RSA with a medialized-glenoid lateralized-humerus design between 2007-2019 and minimum 2-year follow-up. Patients who were wheelchair bound or had a preoperative diagnosis of infection, fracture, and tumor were excluded. Objective IR was measured to the highest vertebral level reached with the thumb. Subjective IR was reported based on patients' rating (normal, slightly difficult, very difficult, or unable) of their ability to perform 4 IRADLs (tuck in shirt with hand behind back, wash back or fasten bra, personal hygiene, and remove object from back pocket). Objective IR was assessed preoperatively and at latest follow-up and reported as median and interquartile ranges.Results A total of 443 patients were included (52% female) at a mean follow-up of 4.4 +/- 2.3 years. Objective IR improved pre- to postoperatively from L4-L5 (buttocks to L1-L3) to L1-L3 (L4-L5 to T8-T12) (P < .001). Preoperatively reported IRADLs of "very difficult" or "unable" significantly decreased postoperatively for all IRADLs (P <= .004) except those unable to perform personal hygiene (3.2% vs. 1.8%, P > .99). The proportions of patients who improved, maintained, and lost objective and subjective IR was similar between IRADLs; 14%-20% improved objective IR but lost or maintained subjective IR and 19%-21% lost or maintained the same objective IR but improved subjective IR depending on the specific IRADL assessed. When ability to perform IRADLs improved postoperatively, objective IR also increased (P < .001). In contrast, when subjective IRADLs worsened postoperatively, objective IR did not significantly worsen for 2 of 4 IRADLs assessed. When examining patients who reported no change in ability to perform IRADLs pre- vs. postoperatively, statistically significant increases in objective IR were found for 3 of 4 IRADLs assessed.Conclusions Objective improvement in IR parallels improvements in subjective functional gains uniformly. However, in patients with worse or equivalent IR, the ability to perform IRADLs postoperatively does not uniformly correlate with objective IR. When attempting to elucidate how surgeons can ensure patients will have sufficient IR after RSA, future investigations may need to use patient-reported ability to perform IRADLs as the primary outcome measure rather than objective measures of IR.