Epidemiology and Risk Factors for Hemodialysis Access–Associated Infections in Children: A Prospective Cohort Study From the SCOPE Collaborative

Abstract

Rationale & Objective: Infections cause significant morbidity and mortality for children receiving maintenance hemodialysis (HD). The Standardizing Care to Improve Outcomes in Pediatric End-Stage Kidney Disease (SCOPE) Collaborative is a quality-improvement initiative aimed at reducing dialysis-associated infections by implementing standardized care practices. This study describes patient-level risk factors for catheter-associated bloodstream infections (CA-BSIs) and examines the association between dialysis center–level compliance with standardized practices and risk of CA-BSI. Study Design: Prospective cohort study. Setting & Participants: Children enrolled in SCOPE between June 2013 and July 2019. Exposures: Data were collected on patient characteristics and center-level compliance with HD catheter care practices across the study period. Centers were categorized as consistent, dynamic (improved compliance over the study period), or inconsistent performers based on frequency of compliance audit submission and changes in compliance with HD care practices over time. Outcome: CA-BSIs. Analytical Approach: Generalized linear mixed models were used to evaluate (1) patient-level risk factors for CA-BSI and (2) associations between change in center-level compliance and CA-BSIs. Results: The cohort included 1,277 children from 35 pediatric dialysis centers; 1,018 (79.7%) had a catheter and 259 (20.3%) had an arteriovenous fistula or graft. Among children with a catheter, mupirocin use at the catheter exit site was associated with an increased rate of CA-BSIs (rate ratio [RR], 4.45; P = 0.004); the use of no antibiotic agent at the catheter exit site was a risk factor of borderline statistical significance (RR, 1.79; P = 0.05). Overall median compliance with HD catheter care practices was 87.5% (IQR, 77.3%-94.0%). Dynamic performing centers showed a significant decrease in CA-BSI rates over time (from 2.71 to 0.71 per 100 patient-months; RR, 0.98; P < 0.001), whereas no significant change in CA-BSI rates was detected among consistent or inconsistent performers. Limitations: Lack of data on adherence to HD care practices on the individual patient level. Conclusions: Improvement in compliance with standardized HD care practices over time may lead to a reduction in dialysis-associated infections.