Continuous renal replacement therapy may increase the risk of catheter infection.

BACKGROUND AND OBJECTIVES

Little is known about the risks of catheter-related infections in patients undergoing intermittent hemodialysis (IHD) as compared with continuous renal replacement therapy (CRRT) techniques. We compared the two modalities among critically ill adults requiring acute renal replacement therapy (RRT).

DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: We used the multicenter Cathedia study cohort of 736 critically ill adults requiring RRT. Cox marginal structural models were used to compare time to catheter-tip colonization at removal (intent-to-treat, primary endpoint) among patients who started IHD (n = 470) versus CRRT (n = 266). On-treatment analysis was also conducted to take into account changes in prescription of RRT modality.

RESULTS

Hazard rate of catheter-tip colonization did not increase within the first 10 days of catheter use. Predictors of catheter-tip colonization were higher lactate levels and hypertension, while systemic antibiotics, antiseptics-impregnated catheters, and mechanical ventilation were associated with decreased risk. The incidence of catheter-tip colonization per 1000 catheter-days was 42.7 in the IHD group and 27.7 in the CRRT group (P < 0.01). This association was no longer significant after correction for channeling bias (weighted HR, 0.96; 95% CI: 0.77 to 1.20, P = 0.73). On-treatment analysis revealed an increased risk of primary endpoint during CRRT exposure as compared with IHD exposure (weighted HR, 0.71; 95% CI: 0.56 to 0.92, P < 0.009).

CONCLUSIONS

Our results do not support the use of CRRT when IHD could be an alternative to reduce the risk of catheter-related infection.