Article Text

Download PDFPDF

G540(P) An improvement project to reduce central line associated blood stream infection (clabsi) in newborn infants
  1. S Mohamed Cassim1,
  2. C Skiffington2,
  3. C Lucas3,
  4. D Anand1,2
  1. 1Neonatology, The Royal Hospital for Sick Children Yorkhill, Glasgow, UK
  2. 2Medicine, University of Glasgow, Glasgow, UK
  3. 3Microbiology, The Royal Hospital for Sick Children Yorkhill, Glasgow, UK


Context The work is being done in a level 3 neonatal unit providing intensive care for babies with surgical, cardiac, ENT and other complex problems requiring both neonatal and paediatric specialist input. All infants admitted between December 2010 and June 2014 with central lines (CL) and those that were sited during the admission period were included.

Problem There were concerns among staff regarding high rate of CLABSI and lack of standardised management of CL. This was resulting in increased morbidity and length of stay.

Assessment of problem and analysis of its causes The baseline CLABSI rate was quantified in the first 4 months by monitoring infection in all infants with CL. CLABSI was identified using pre-defined criteria. Champions were identified from all staff groups and a multidisciplinary team was formed. The practice of line insertion and maintenance; and published literature was reviewed to identify potential problems and areas for improvement. An improvement project charter was developed describing the aims and objectives, components of the care bundle and plan for measuring and monitoring of processes and outcomes.

Intervention The intervention was a care bundle comprising of 1) line trolley with standardised equipment 2) hand hygiene 3) standardised checklist of techniques for aseptic insertion and maintenance of CL 4) daily review of need for CL.

Study design The study was designed using improvement methodology. A base line rate of CLABSI per 1000 central line days (CLD) was monitored monthly during the first four months, followed by implementation of care bundle and monthly monitoring of CLABSI rates for rest of the study period.

Strategy for change Through teaching sessions and meetings, all staff were informed of the base line CLABSI rates and the proposal to initiate an improvement project. Structured education and training was carried out by the champions in the multidisciplinary group. Every month, a consultant neonatologist and clinical scientist in microbiology reviewed the infections for diagnosing CLABSI. Data is displayed in the unit to feedback to the staff. The initial timescale for change was 4 months.

Measurement of improvement The measurement for improvement was done through monitoring process and outcome measures. The process measures monitored were compliance with 1) hand hygiene 2) using insertion and maintenance checklists 3) changing of dressings on surgically inserted CL once a week 3) changing of hubs twice a week and 4) removing lines when not required. The data was analysed using Statistical Process Charts. Run charts were used for process measures and control charts were used for outcome measures.

Between December 2010 and June 2014, 350 children had 571 central lines during their stay. Of these, 315 were peripherally inserted central catheters (PICC), 161 were surgically inserted central catheters (SICC) and 95 umbilical lines. Together they amounted to 12,204 CLD. There were 53 CLABSI resulting in the overall rate of 4.3/1000 CLD. However, the base line rate between December 2010 and March 2011 was 8.3/1000 CLD and post intervention the rate has fallen to 3.7/1000 CLD (Figure 1).

Abstract G540(P) Figure 1

Rate of Central line associated blood stream infection per 1000 central line days

Effects of changes The CLABSI rate was reduced by 55%. It could be inferred that in a similar proportion of patients it may have reduced the morbidity and length of stay. It is suggested that each CLABSI increases the length of stay by 3 weeks and cost of care by £10 000.

Due to problems with person dependency, CLABSI rate increased in the second year but once this issue was addressed, improvement was regained.

Lessons learnt We learnt to avoid person dependency and build CL surveillance into routine daily work.

Message for others Through this project we demonstrated that improvement methodology has a potential role in ensuring reliable delivery of evidence based medicine and improving outcomes.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.