Inpatient surgical site infection after shoulder arthroplasty

Abstract

Background: Surgical site infection (SSI) after joint arthroplasty is associated with prolonged hospitalization, reoperation, inferior outcomes, and substantial resource utilization. As the number of shoulder replacements performed in the United States continues to rise, measuring the incidence of inpatient SSI after hemiarthroplasty (HSA) and total shoulder arthroplasty (TSA), and associated risk factors for infection is worthwhile.

Methods: Using the Nationwide Inpatient Sample (NIS), we reviewed 241,193 patients undergoing TSA or reverse TSA and 159,795 undergoing HSA between 2002 and 2011 and identified patients with an associated diagnosis of SSI during the admission. Demographic characteristics, preoperative diagnoses, further surgical procedures, associated comorbidities, and in-hospital events associated with SSI were sought in multivariable logistic regression analysis.

Results: An in-hospital SSI developed in 0.08% of patients undergoing TSA or reverse TSA and in 0.11% of patients undergoing HSA. Independent risk factors for inpatient SSI included TSA vs HSA (odds ratio [OR], 1.83), Medicaid insurance vs private insurance (OR, 3.93), diagnosis of fracture nonunion (OR, 5.76), avascular necrosis (OR 2.71), or proximal humeral fracture (OR, 2.62) vs primary osteoarthritis, comorbidities, in-hospital events (blood transfusion, pneumonia, and acute renal failure), and increased duration of hospital stay.

Conclusions: The small percentage of SSI that occurs during the initial inpatient stay after shoulder arthroplasty is related to diagnoses other than primary osteoarthritis in more infirm patients with low-income government insurance (Medicaid). Patients considering shoulder arthroplasty can use this information to help decide between the potential improvement in comfort and function of the shoulder and the potential for major adverse events such as infection.

Keywords: Shoulder arthroplasty; acute infection; perioperative infection.