Antimicrobial coated peripherally-inserted central venous catheters: a beneficial intervention?
Antimicrobial coated central venous catheters (CVCs) have demonstrated efficacy for prevention of central line-associated bloodstream infections (CLABSIs). Both antibiotic (minocycline/rifampicin) and antimicrobial (chlorhexidine/silver sulfadiazine) impregnated devices have been shown to decrease bacterial colonisation and CLABSI. Clinically, these may be used in the setting of increased infection rates despite utility of recommended prevention bundles. Peripherally-inserted CVCs (PICCs) are increasingly being used in place of non-tunnelled CVCs, but the potential benefit of antimicrobial coating of these devices has not been well-studied. In 2011, the FDA approved a chlorhexidine-impregnated PICC. Storey S et al. (Am J Infect Control 2016; 44:636-641) sought to determine the impact of this device upon risk for CLABSI and thromboembolism.
The study was conducted at a US centre over an 18-month period. Patients admitted to cardiovascular thoracic, medical intensive care and oncology units and who required a PICC device were eligible for inclusion, and were randomised to receive either a chlorhexidine-impregnated or non-impregnated PICC. Devices were either single or double lumen.
One hundred and eighty-seven patients were analysed (n=80 and 87 in chlorhexidine PICC and non-impregnated PICC groups, respectively). Studied devices were in situ for a median of 8 days in both groups. Three patients developed CLABSI – 2 in the chlorhexidine-impregnated PICC group, and one in the control group (p=0.067). Three patients developed venous thromboembolism – 1 in the chlorhexidine-impregnated group, and 2 in the control group (p>0.99). Moderate bleeding was more frequent in the chlorhexidine-impregnated device group, compared to the control group (32.5 vs. 1.1%, p<0.001).
Findings suggest that in high-risk patient populations, chlorhexidine-impregnated PICC devices are not associated with a reduction in CLABSI or venous thromboembolism. Further, the risk of post-insertion bleeding may be increased. The study is limited by small sample size, and may not have been sufficiently powered to detect a difference in outcomes. Larger multicentre trials, including patients with longer PICC dwell times, are required.
Risk of infection following colonisation with carbapenem-resistant Enterobacteriaceae
In recent years, carbapenem-resistant Enterobacteriaceae (CRE) have emerged internationally as significant pathogens accounting for healthcare-associated infections. Colonisation is considered to be a risk factor for subsequent infection. However, the attributable risk related to colonisation has not been quantified. Tischendorf J et al. (Am J Infect Control 2016; 44:539-543) undertook a systematic review to evaluate the relationship between colonisation with CRE and subsequent infection.
Literature published between 1991 and 2015 were searched for clinical trials or observational studies reporting both CRE colonisation and infection. Point-prevalence studies were excluded. For evaluable studies, patient populations and nature of screening methods were determined. The proportion of CRE-colonised patients who developed infection was summarised, as was the infection site and all-cause mortality.
Initial search identified 178 potentially relevant abstracts. Of these, 10 studies fulfilled inclusion criteria (total of 1,806 colonised patients). All were studies of adult inpatients, with 4 exclusively in an intensive care setting. Four studies were conducted in an outbreak setting. Seven studies reported carbapenem-resistant Klebsiella pneumoniae, 2 reported carbapenem-resistant Acinetobacter baumannii, and one study reported multiple CRE. Colonisation was defined as detection of CRE in the gastrointestinal tract, by rectal swab in the majority of instances. Among all colonised patients, 299 infections (rate of 16.5%) were reported, although rates of infection in individual studies varied widely (0-89%). Half of all described infections were pneumonia, followed in decreasing order by urinary tract infections, bloodstream infections, and skin/soft tissue infections. Overall
mortality rate was 10%.
Findings demonstrate that a considerable proportion of patients colonised with CRE go on to develop infection with CRE (16.5%). Limitations include the variation in screening practices and potential for reporting bias. Notably, time to onset of infection following colonisation was not evaluable, and studies varied considerably in duration of follow-up of patients.