gms | German Medical Science

Objectives/Interrogation: Total elbow arthroplasty (TEA) is commonly performed in patients with rheumatoid arthritis(RA). Convertible systems provide the capability to choose between linked and unlinked designs intra-operatively using the same stems. There is a paucity of research comparing linked and unlinked designs in the rheumatoid population. We compared outcomes of linked and unlinked TEA arthroplasty using a single convertible implant in patients with rheumatoid patients.

Methods: All patients with RA who underwent total elbow arthroplasty at a single center and had a minimum of 2 years follow-up were reviewed. Exclusion criteria included patients with less than 2 years follow-up. Patient demographic information, outcome scores, functional assessment and radiographic parameters were evaluated.

Results and Conclusions: Eighty-two patients were evaluated. Mean age at index procedure was 60.9±10.5yrs with a mean follow-up of 5.6±4years. Twenty-seven unlinked and 55 linked TEAs were performed. Mean demographics were similar between linked and unlinked cohorts with the exception of follow-up (8±4years for unlinked & 5±3years for linked, p=0.001). No difference in range of motion or elbow strength compared to the contralateral elbow was noted between linked and unlinked cohorts other than pronation strength which was significantly better in the linked cohort (74±8% vs. 100±8%, p=0.03).

The mean patient reported outcome scores were MEPI (83±16), PREE (15±18), QuickDASH (34±20). There were no differences in outcome scores, incidence of radiographic lucency's or rates of reoperation (17% vs. 24%, p=0.4) complications (32% vs. 31%, p=0.4) and revision (13% vs. 17%, p=0.3) between the linked and unlinked cohorts (p>0.05). Ten patients developed post-operative infections (7 deep and 3 superficial) including 5 patients requiring 2-stage revision and 2 who were placed on suppressive antibiotics after I&D with implant retention. Superficial wound dehiscences were successfully treated without surgery. Four unlinked prosthesis were converted to a linked design for instability. 6 patients had ulnar nerve dysfunction postoperatively with 3 patients sustaining either partial or complete nerve transection.

TEA using a convertible implant provides good patient reported outcomes at mid-term follow-up in patients with RA. The lack of differences between these two designs suggests surgeons may elect to use either linked or unlinked designs as dictated by clinical and intra-operative findings.