Outcomes of shoulder arthroplasty in diabetic patients as assessed by peri-operative A1C

Although diabetes mellitus (DM) has an adverse effect on complication rates in orthopaedic surgery, neither the effect of DM nor the association between haemoglobin A1C (HbA1C) and outcomes after shoulder arthroplasty (SA) has been studied.

A retrospective review of 406 SAs (70 HAs, 188 total shoulder arthroplasties [TSAs], 148 reverse total shoulder arthroplasties [RSAs]) with HbA1Cs within 90 days of surgery was conducted. The average age was 70 years (range 27-97) and 55% were female. The average peri-operative HbA1C was 6.4% (range 4.7-9.8%), with 104 (26%) having a HbA1C of 7.0% or greater. Kaplan-Meier curves were constructed to determine complication, re-operation, and revision rates at two, five and ten years post-operatively.

At mean follow-up of four years, 58 (16.5%) SAs were associated with post-operative complications including six (1.5%) infections. Kaplan-Meier two and five year survivorship free of any complication was 87.3 and 82.6% and of infection was 95.8 and 90.9%, respectively. Thirty-two (7.9%) SAs required reoperation with 26 (6.4%) of these being revisions. Survivorship at two and five years post-operatively was 97.1 and 92.7% for re-operation and 98.6 and 98.2% for revision, respectively. No increased risk of complications, re-operation, revision, or infection was seen with increased HbA1C analyzed as a continuous variable (hazard ratio = 0.97-1.11, 95% CI = 0.28-3.94, p = 0.5882-0.9445) or as a dichotomous variable with a cutoff of 7.0% (hazard ratio = 1.02-1.47, 95% CI = 0.20-7.48, p = 0.3253-0.9544).

A collaborative and comprehensive approach to the pre-operative medical evaluation of patients with DM is critical, as is future investigation into alternative methods associated with outcomes after shoulder arthroplasty in patients with DM.

Level IV.