Infections and mortality in ICU patients undergoing continuous renal replacement therapy: a retrospective cohort study

Abstract

Background: Critically ill patients receiving continuous renal replacement therapy (CRRT) are at increased risk for multidrug-resistant infections and infection-related mortality. Altered pharmacokinetics in CRRT may contribute to inadequate antimicrobial exposure and therapeutic failure. However, limited data exist on infection burden and resistance patterns specific to this population.

Methods: We conducted a retrospective cohort study of ICU patients receiving continuous venovenous hemodialysis (CVVHD) at a tertiary academic center between May 2016 and April 2020. Patients were included if they received CRRT for ≥ 48 h, had at least one positive microbial culture, and received at least one antimicrobial of interest. Data were collected on infection sources, pathogens, resistance patterns, and mortality.

Results: Among 661 CRRT recipients, 394 (59.6%) had at least one positive culture. The most common infection sites were respiratory (69.0%), skin and soft tissue (53.8%), and intra-abdominal (38.8%). Intra-abdominal and bloodstream infections had the highest mortality (63.7% and 57.7%, respectively). Vancomycin-resistant E. faecium (83.3%), cefepime-resistant A. baumannii (100%), and P. aeruginosa with high β-lactam resistance were prominent. These resistance profiles diverged from general ICU trends.

Conclusion: ICU patients receiving CRRT experience high rates of multidrug-resistant infections and associated mortality. Tailored dosing strategies, including dual empiric coverage in select cases, and CRRT-specific antimicrobial stewardship are essential to improve outcomes in this high-risk population.

Keywords: Antimicrobial resistance; Continuous renal replacement therapy; Critical care; Infection-related mortality.

Fig. 1

CONSORT diagram for cohort selection. Patients were included if they received continuous renal replacement therapy (CRRT) and at least one of cefepime, meropenem, piperacillin-tazobactam, or vancomycin. Patients with CRRT duration <48 hours were excluded. Of the remaining cohort, microbiologic culture results were assessed, and the presence of antimicrobial resistance was determined

Fig. 2

Proportion of positive and negative culture results by source in critically Ill patients. The bar chart displays the distribution of positive and negative culture samples from various sources in critically ill patients. Respiratory samples had the highest positivity rate (69.0%), followed by skin and soft tissue infections (SSTI, 53.6%), intraabdominal sources (39.0%), and urinary samples (28.1%). Blood cultures demonstrated a relatively low positivity rate (10.1%), while central nervous system (CNS) cultures had the lowest positivity rate (4.5%). Device cultures showed a balanced distribution with 50.0% positive and 50.0% negative samples, though the total sample size was small (n = 8). Device infections were exclusively pacemakers and other cardiac device leads. These findings highlight the variability in culture positivity depending on the source, with respiratory and SSTI sources having the highest rates of culture-confirmed infections

Fig. 3

Distribution of microbial isolates across infection sites in critically ill patients undergoing CRRT. (A) Overall distribution of the most frequently isolated pathogens across all positive culture sites (n = 394). (B) Bloodstream infections, where coagulase-negative staphylococci (22.0%) were the most common, followed by S. aureus (13.4%) and Staphylococcus epidermidis (10%). (C) Respiratory cultures, with P. aeruginosa (20.5%) and S. aureus (20.5%) as the predominant pathogens, followed by K. pneumoniae (12.8%). (D) Intra-abdominal infections, where P. aeruginosa (20.5%), E. faecium (16.4%), and E. coli (13.7%) were most frequently identified. (E) Skin and soft tissue infections (SSTI), where E. faecium (15.3%), E. faecalis (14.0%), and S. aureus (14.0%) were the leading pathogens, with P. aeruginosa (12.7%) also present. (F) Urinary cultures, in which E. coli (30.7%) was most frequently isolated, followed by K. pneumoniae (14.8%) and E. faecalis (14.8%). Values represent the percentage of positive cultures for each pathogen within the respective infection site

Fig. 4

Heatmap of antibiotic resistance patterns for Gram-positive organisms isolated from ICU patients undergoing CRRT. The heatmap displays the percentage of resistance for E. faecalis, E.s faecium, and S. aureus across various antibiotics. Colors indicate resistance levels, with green (low resistance, 0–25%), yellow (moderate resistance, 25–50%), and red (high resistance, ≥ 75%), as shown in the accompanying scale bar. E. faecium exhibits higher resistance rates, particularly to ampicillin and vancomycin, while S. aureus shows notable resistance to cephalosporins. Percentages and sample sizes for each antibiotic-pathogen combination are displayed within the cells

Fig. 5

Heatmap of antibiotic resistance patterns for Gram-negative organisms isolated from ICU patients undergoing continuous renal replacement therapy (CRRT). The heatmap displays the percentage of resistance for A. baumannii, E. coli, Enterobacter species, K. pneumoniae, P. aeruginosa, P. mirabilis, and S. maltophilia across various antibiotics. Colors indicate resistance levels, with green (low resistance, 0–25%), yellow (moderate resistance, 25–50%), and red (high resistance, ≥ 75%), as shown in the accompanying scale bar. A. baumannii and K. pneumoniae demonstrate high resistance to beta-lactams and carbapenems, while E. coli and P. aeruginosa show moderate resistance to fluoroquinolones and cephalosporins. Percentages and sample sizes for each antibiotic-pathogen combination are displayed within the cells