Preoperative Sepsis Is Associated With Increased 90-Day and 2-Year Complications Following Total Shoulder Arthroplasty: A Time-Stratified and Organism-Specific Analysis.

PURPOSE: To evaluate the association between preoperative sepsis and 90-day and 2-year complications following total shoulder arthroplasty, with particular attention to the timing and causative organism of sepsis in relation to postoperative outcomes.

METHODS: A retrospective cohort study was conducted using the PearlDiver M170 national administrative claims database. Patients undergoing shoulder arthroplasty with at least four years of data were included. Those with sepsis within two years prior to surgery were propensity score matched 1:3 to controls based on age, sex, year of surgery, Elixhauser Comorbidity Index, and comorbidities. Patients were stratified by timing of sepsis (<3 months, 3-6 months, 6-12 months, 12-18 months, 18-24 months preoperative) and by organism (gram-positive, gram-negative, Staphylococcal, Streptococcal, Escherichia coli). Complications were compared at 90 days and 2 years postoperatively. P values were Bonferroni-adjusted within each outcome period and subgroup.

RESULTS: The cohort included 14,786 patients (3,742 with preoperative sepsis; 11,044 controls). At 90 days, preoperative sepsis was strongly associated with postoperative sepsis (OR 3.89, 95% CI 2.75-5.50), and moderately associated with pneumonia (OR 2.08, 95% CI 1.70-2.54), acute kidney injury (OR 1.57, 95% CI 1.30-1.90), deep vein thrombosis (OR 1.69, 95% CI 1.17-2.46), urinary tract infection (OR 1.76, 95% CI 1.53-2.03), readmission (OR 1.49, 95% CI 1.28-1.75), and wound dehiscence (OR 1.92, 95% CI 1.12-3.28). At two years, preoperative sepsis was moderately associated with periprosthetic joint infection (OR 2.04, 95% CI 1.61-2.59), and weakly associated with implant loosening (OR 1.29, 95% CI 1.03-1.62) and postoperative stiffness (OR 1.28, 95% CI 1.07-1.52). The highest odds of complications occurred when sepsis was documented within three months before surgery and declined as the interval increased. Gram-positive sepsis was strongly associated with periprosthetic joint infection (OR 3.27, 95% CI 1.87-5.70), whereas gram-negative sepsis was strongly associated with wound dehiscence (OR 5.17, 95% CI 2.01-13.32) and moderately associated with acute kidney injury (OR 2.28, 95% CI 1.45-3.60), urinary tract infection (OR 2.79, 95% CI 1.99-3.90). Staphylococcal sepsis was strongly associated with periprosthetic joint infection at two years (OR 5.98, 95% CI 3.39-10.56).

CONCLUSION: Preoperative sepsis within two years of TSA was associated with increased odds of 90-day and 2-year complications, with the highest odds when surgery occurred within three months of sepsis. Prior sepsis and its temporal proximity may be relevant to preoperative risk assessment and patient counseling.

LEVEL OF EVIDENCE: Level III, Retrospective Cohort Comparison Using Large Database, Prognosis Study.