Justine Pages1, Pascal Hazera2, Bruno Mégarbane3, Damien du Cheyron4,5, Marie Thuong6, Jean-Jacques Dutheil1, Xavier Valette4, François Fournel1, Leonard A Mermel7, Jean-Paul Mira8, Cédric Daubin4, Jean-Jacques Parienti9,10,11. 1. Department of Biostatistics and Clinical Research, Caen University Hospital, Caen, France. 2. Department of Intensive Care Medicine, General Hospital, Saint Lô, France. 3. Department of Medial and Toxicologic Intensive Care, Lariboisière University Hospital, Paris, France. 4. Department of Intensive Care Medicine, Caen University Hospital, Caen, France. 5. EA4655 Risque Microbiens, Caen Normandie Université, Caen, France. 6. Department of Intensive Care Medicine, General Hospital, Pontoise, France. 7. Warren Alpert Medical School of Brown University, Providence, RI, USA. 8. Department of Medial Intensive Care, Cochin University Hospital, Paris, France. 9. Department of Biostatistics and Clinical Research, Caen University Hospital, Caen, France. parienti-jj@chu-caen.fr. 10. EA4655 Risque Microbiens, Caen Normandie Université, Caen, France. parienti-jj@chu-caen.fr. 11. Department of Infectious Diseases, Caen University Hospital, Caen, France. parienti-jj@chu-caen.fr.
Abstract
PURPOSE: Compare the effectiveness of different cutaneous antiseptics in reducing risk of catheter-related infection in intensive care unit (ICU) patients. METHODS: We compared the risk of central venous catheter-related infection according to four-step (scrub, rinse, dry, and disinfect) alcoholic 5 % povidone-iodine (PVI-a, n = 1521), one-step (disinfect) alcoholic 2 % chlorhexidine (2 % CHX-a, n = 1116), four-step alcoholic <1 % chlorhexidine (<1 % CHX-a, n = 357), and four-step aqueous 10 % povidone-iodine (PVI, n = 368) antiseptics used for cutaneous disinfection and catheter care during the 3SITES multicenter randomized controlled trial. Within this cohort, we performed a quasi-experimental study (i.e., before-after) involving the four ICUs which switched from PVI-a to 2 % CHX-a. We used propensity score matching (PSM, n = 776) and inverse probability weighting treatment (IPWT, n = 1592). The end point was the incidence of catheter-related infection (CRI) defined as catheter-related bloodstream infection (CRBSI) or a positive catheter tip culture plus clinical sepsis on catheter removal. RESULTS: In the cohort analysis and compared with PVI-a, the incidence of CRI was lower with 2 % CHX-a [adjusted hazard ratio (aHR), 0.51; 95 % confidence interval (CI) (0.28-0.96), p = 0.037] and similar with <1 % CHX-a [aHR, 0.73; (0.36-1.48), p = 0.37] and PVI [aHR, 1.50; 95 % CI (0.85-2.64), p = 0.16] after controlling for potential confounders. In the quasi-experimental study and compared with PVI-a, the incidence of catheter-related infection was again lower with 2 % CHX-a after PSM [HR, 0.35; 95 % CI (0.15, 0.84), p = 0.02] and in the IPWT analysis [HR, 0.31; 95 % CI (0.14, 0.70), p = 0.005]. The incidence of CRBSI or adverse event was not significantly different between antiseptics in all analyses. CONCLUSIONS: In comparison with PVI-a, the use of 2 % CHX-a for cutaneous disinfection of the central venous catheter insertion site and maintenance catheter care was associated with a reduced risk of catheter infection, while the benefit of <1 % CHX-a was uncertain. CLINICAL TRIALS IDENTIFIER: NCT01479153.
RCT Entities:
PURPOSE: Compare the effectiveness of different cutaneous antiseptics in reducing risk of catheter-related infection in intensive care unit (ICU) patients. METHODS: We compared the risk of central venous catheter-related infection according to four-step (scrub, rinse, dry, and disinfect) alcoholic 5 % povidone-iodine (PVI-a, n = 1521), one-step (disinfect) alcoholic 2 % chlorhexidine (2 % CHX-a, n = 1116), four-step alcoholic <1 % chlorhexidine (<1 % CHX-a, n = 357), and four-step aqueous 10 % povidone-iodine (PVI, n = 368) antiseptics used for cutaneous disinfection and catheter care during the 3SITES multicenter randomized controlled trial. Within this cohort, we performed a quasi-experimental study (i.e., before-after) involving the four ICUs which switched from PVI-a to 2 % CHX-a. We used propensity score matching (PSM, n = 776) and inverse probability weighting treatment (IPWT, n = 1592). The end point was the incidence of catheter-related infection (CRI) defined as catheter-related bloodstream infection (CRBSI) or a positive catheter tip culture plus clinical sepsis on catheter removal. RESULTS: In the cohort analysis and compared with PVI-a, the incidence of CRI was lower with 2 % CHX-a [adjusted hazard ratio (aHR), 0.51; 95 % confidence interval (CI) (0.28-0.96), p = 0.037] and similar with <1 % CHX-a [aHR, 0.73; (0.36-1.48), p = 0.37] and PVI [aHR, 1.50; 95 % CI (0.85-2.64), p = 0.16] after controlling for potential confounders. In the quasi-experimental study and compared with PVI-a, the incidence of catheter-related infection was again lower with 2 % CHX-a after PSM [HR, 0.35; 95 % CI (0.15, 0.84), p = 0.02] and in the IPWT analysis [HR, 0.31; 95 % CI (0.14, 0.70), p = 0.005]. The incidence of CRBSI or adverse event was not significantly different between antiseptics in all analyses. CONCLUSIONS: In comparison with PVI-a, the use of 2 % CHX-a for cutaneous disinfection of the central venous catheter insertion site and maintenance catheter care was associated with a reduced risk of catheter infection, while the benefit of <1 % CHX-a was uncertain. CLINICAL TRIALS IDENTIFIER: NCT01479153.
Entities:
Keywords:
Central venous catheter; Chlorhexidine; Cutaneous disinfection; Nosocomial infection; Povidone–iodine
Authors: Naomi P O'Grady; Mary Alexander; Lillian A Burns; E Patchen Dellinger; Jeffrey Garland; Stephen O Heard; Pamela A Lipsett; Henry Masur; Leonard A Mermel; Michele L Pearson; Issam I Raad; Adrienne G Randolph; Mark E Rupp; Sanjay Saint Journal: Am J Infect Control Date: 2011-05 Impact factor: 2.918
Authors: Issam Raad; Hend A Hanna; Badie Alakech; Ioannis Chatzinikolaou; Marcella M Johnson; Jeffrey Tarrand Journal: Ann Intern Med Date: 2004-01-06 Impact factor: 25.391
Authors: K Cooper; G Frampton; P Harris; J Jones; T Cooper; N Graves; J Cleland; J Shepherd; A Clegg; B H Cuthbertson Journal: J Hosp Infect Date: 2013-09-21 Impact factor: 3.926
Authors: Niccolò Buetti; Jonas Marschall; Marci Drees; Mohamad G Fakih; Lynn Hadaway; Lisa L Maragakis; Elizabeth Monsees; Shannon Novosad; Naomi P O'Grady; Mark E Rupp; Joshua Wolf; Deborah Yokoe; Leonard A Mermel Journal: Infect Control Hosp Epidemiol Date: 2022-04-19 Impact factor: 6.520