Search

Needleless connectors (NCs) are medical devices which connect to the end of vascular access devices. Since their introduction in the 1990s, in response to the increasing incidence of needle stick injuries experienced by nurses, they have become standard equipment for accessing venous access devices. The central venous access device (CVAD) is a specialised vascular device placed into a large vein leading to the heart for the administration of intravenous fluids, medications, blood products and specialised treatments. While NCs significantly reduced the occurrence of needle stick injuries, they also contributed to a rise in catheter-associated bloodstream infections (CABSI). This is largely due to inadequate cleaning and decontamination of microorganisms from the external surface of the NC by nurses prior to use. Over 1,000 cases of CABSI are reported in Australia annually. A CABSI can have a significant impact on patient morbidity and mortality, contributing to extended hospitalisation and affecting long-term health and lifestyle. Current research into NC decontamination practices is generally of low quality and indicates that current approaches remain ineffective, particularly in clinically vulnerable patients and those requiring long-term CVAD use. Aim: The overarching aim of this project was to investigate the role of NC decontamination products in preventing CABSI and to assess the feasibility of conducting a superiority randomised controlled trial (RCT). Methods: This research was underpinned by the United Kingdom Medical Research Council Framework for Developing and Evaluating Complex Interventions and Casadevall and Pirofski’s Damage-Response Framework of Microbial Pathogenesis and Infectious Diseases. This research was conducted in two phases: an in vitro study and a pilot RCT. A systematic review and meta-analysis were undertaken to inform the two research phases. In Phase One, the in vitro study compared three decontamination products (70% alcohol wipe; 2% chlorhexidine in 70% alcohol wipe; and a 70% alcohol-impregnated cap) for comparative efficacy. Six hundred and forty-eight NCs were used in the experiments consisting of three different types of NC designs, plus a contaminated and uncontaminated control. The connectors were contaminated with microorganisms commonly associated with CABSI, specifically Staphylococcus aureus, Staphylococcus epidermidis, Pseudomonas aeruginosa and Candida albicans. Half of the NCs were pre-treated with human serum to mimic real-life use of the connector. In Phase Two, a pilot RCT was undertaken to assess the feasibility of a large, multi-site superiority trial comparing the same three decontamination products used in the in vitro phase. Recruitment for this pilot RCT occurred from 31 August 2017 to 8 March 2018. Participants were recruited from the surgical (general elective, emergency, vascular and orthopaedic), oncology (medical, surgical, haematological) and general medical units of RBWH. Elective surgical patients made up the largest cohort. Orthopaedic, medical oncology, and general medical patients were the least likely to be recruited. Overall, the average rate of recruitment across all admission types was 10 participants per month (range 1–17). Patients with a current bloodstream infection were excluded. The primary outcomes of feasibility were defined as: > 80% of patients screened were eligible; > 80% of eligible participants agreed to enrol; > 80% of participants in the study groups received their allocated treatment; < 5% of participants were lost to follow-up; there was < 5% missing outcome data. Based on these feasibility criteria, the hypothetical conclusions would be interpreted as: (i) all criteria will be met and trial is feasible without modification; (ii) one criterion not be met and trial is feasible with close monitoring; (iii) two or three criteria not be met and trial is feasible with modification; (iv) four or more criteria not be met and trial is not feasible. Secondary outcomes were rates of CABSI and microbial colonisation of the internal surface of NCs following removal of the CVAD. Results of the in vitro study: The 2% chlorhexidine in 70% alcohol wipe resulted in a greater reduction of microorganisms than either the 70% alcohol wipe (t(70) = 74.32, p < 0.01) or the alcohol-impregnated cap (t(70) = 28.25, p < 0.01). The alcohol-impregnated cap was more effective than the alcohol wipe (t(70) = 5.01, p < 0.01). In the presence of human serum, all three products were effective in reducing colony counts of microorganisms, with the difference between the 2% chlorhexidine in 70% alcohol wipe and the other two products remaining statistically significant. Results of the pilot randomised controlled trial: Two of the feasibility criteria were not met; therefore, it was decided that a superiority RCT would be feasible to proceed with modification. For the primary predefined feasibility outcomes, 152 patients were screened for eligibility, 53 were excluded; therefore, 99 patients were eligible for inclusion in the study. This proportion (72%) did not meet the feasibility criteria of 80% eligible for recruitment. Of the 99 patients eligible to participate, 91 patients were recruited; therefore, this criterion was met, with 92% of eligible patients recruited. Over the course of the study, 239 protocol checks were conducted on 74 of the 90 participants of which 78% were compliant. Hence, the criteria that > 80% of participants would receive their allocated treatment was not met. No participants were lost to follow-up and there was no missing outcome data, so both of these criteria were met. Five patients had blood cultures collected during the study; none were diagnosed as CABSI. Of the 12 NC’s assessed for microbial growth, all were contaminated. Discussion: In the in vitro study the 2% chlorhexidine in 70% alcohol wipe out-performed the other products for NC decontamination. The pilot RCT established feasibility of the larger superiority trial, with modifications. It is recommended that to address the two unmet criteria a superiority multisite RCT may be needed to increase the proportion of potentially eligible participants. Recruitment should focus on the patient admission types of elective, emergency and oncology surgery, and haematology. Outpatients related to these groups should also be included. Trial fidelity may be enhanced through the adoption of a minimum of second daily checks and implementation of strategies such as bedside information indicating allocated treatment and increased engagement with clinical staff. Although no CABSI was recorded during the pilot RCT, secondary outcomes suggest that microbial colonisation of the internal surface of the NC may occur despite current decontamination practices which, coupled with extended CVAD dwell time, may contribute to infection. Follow-up of CVAD dwell time would facilitate quantification of rates of CABSI per 1,000 catheter days. Overall, the results of this doctoral research suggest that the 2% chlorhexidine in 70% alcohol wipe may be the most effective approach to NC decontamination. However, further factors may impact its clinical efficacy to prevent CABSI, such as a 30-second duration of active decontamination, which may be difficult to achieve in practice. Neither are the results of in vitro research directly translatable to practice. This emphasises the need for a fully powered RCT to directly compare NC decontamination products and practice issues such as decontamination and drying times to determine the most effective approach. The pilot RCT detailed the protocol and design modifications required to optimise recruitment and trial fidelity to undertake such research. Conclusion: This doctoral research has demonstrated the possibility of effective NC decontamination for CVADs and the need for, and feasibility of, further research to provide high-level evidence on the most efficacious and cost-effective product for NC decontamination. The resulting reduction in the occurrence of CABSI would have a significant impact on patient outcomes, patient quality of life, and healthcare-related costs.

Additional file 1: Table S1. Costs associated with each study group: products and staff time and costs of responding to failure.