Dialysis dependence and modality impact complication rates after shoulder arthroplasty

The goals of the study were to determine (1) the incidence of dialysis patients undergoing shoulder arthroplasty (SA), (2) the association of dialysis modality with complications after SA, and (3) the association of dialysis dependence with death in patients undergoing SA and nonsurgical dialysis-dependent controls.

Using an insurance database, we identified dialysis-dependent patients undergoing SA and compared them with a matched control cohort without dialysis use. We performed an analysis comparing patients using peritoneal dialysis (PD), patients using hemodialysis (HD), matched non-dialysis-dependent controls, and matched PD and HD nonsurgical patients. Complications including in-hospital death, emergency department visits, hospital readmission, infection, and revision surgery were assessed.

The incidence of SA in dialysis patients has significantly increased. Compared with controls, dialysis-dependent patients had increased rates of in-hospital death (odds ratio [OR], 7.60; P < .0001), emergency department visits (OR, 4.16; P < .0001), hospital admission (OR, 1.63; P < .0001), and infection within 1 year (OR, 1.90; P = .009). Compared with patients receiving HD, PD patients had lower rates of death (OR, 0.40; P = .008), hospital readmission (OR, 0.43; P = .047), and revision surgery (OR, 0.23; P = .037). as well as a lower incidence of infection (OR, 0.30; P = .018). Compared with controls, PD patients had similar rates of most complications. Compared with nonsurgical dialysis-dependent patients, the SA procedure itself was not associated with an increased mortality rate.

Although dialysis represents a significant independent risk factor for complications after SA, this risk is highly influenced by the type of dialysis.