Antibiotic Prophylaxis is Often Unnecessary for Screening Colonoscopies Following Total Knee Arthroplasty

Abstract

BackgroundIncidence of bacteremia following screening colonoscopy is low, but risk of hematogenous spread causing prosthetic joint infection (PJI) may exist in total knee arthroplasty (TKA) patients. In multivariate analyses, we examined PJI risk among 3 TKA cohorts: (1) colonoscopy recipients given antibiotic prophylaxis; (2) colonoscopy recipients not given antibiotic prophylaxis; and (3) no colonoscopy. We assessed 90-day to 1-year (A) PJI risk and (B) risk factors for postcolonoscopy PJI. MethodsWe queried a national, all-payer database for primary TKA recipients from 2010 to 2020. Patients who had colonoscopies and who did (n = 2,558) or did not have antibiotic prophylaxis (n = 20,000) were identified. These were compared those who did not undergo colonoscopy (n = 20,000). The 20,000 patients were randomly selected to mitigate type-1 errors. Multivariate regressions compared PJI risk factors, such as alcohol abuse (AA), rheumatoid arthritis (RA), and diabetes. ResultsBoth colonoscopy cohorts had no increased PJI risk compared to noncolonoscopy (odds ratio (OR) ≤ 2.20, P ≥ .064). Alcohol abuse, diabetes, and RA were found to be risk factors further enhancing likelihood of PJI for TKA patients not receiving antibiotics undergoing colonoscopies (OR ≥ 1.35, P ≤ .044). ConclusionOverall, antibiotic prophylaxis does not decrease PJI risk following colonoscopy TKA recipients. After adjusting for known risk factors, both colonoscopy cohorts demonstrated similar PJI risks compared to the noncolonoscopy cohort. However, AA, diabetes, and RA were associated with further increased PJI risk for TKA patients undergoing colonoscopies compared to those who did not. Therefore, if undergoing colonoscopy after TKA, our findings suggest that most patients do not need to have antibiotics except for these high-risk patients.