Article Text

Download PDFPDF

The hidden harms of Matching Michigan
  1. Thomas J Heron,
  2. Christopher M Faraday,
  3. Paul Clarke
  1. Neonatal Intensive Care Unit, Norfolk & Norwich University Hospitals NHS Foundation Trust, Norwich, UK
  1. Correspondence to Dr P Clarke, Neonatal Intensive Care Unit, Norfolk & Norwich University Hospitals NHS Foundation Trust, Norwich NR4 7UY, UK; paul.clarke{at}

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.

Matching Michigan (MM) was a 2-year interventional programme run by the National Health Service (NHS) National Patient Safety Agency during 2009–2011. It was a safety initiative which aimed to reduce central line-related bloodstream infections in intensive care patients through introduction of care bundles for the insertion and maintenance of central venous catheters (CVCs). Successful reductions in infections from CVCs were realised for adult and paediatric patients.1 Neonatal intensive care units (NICUs) were also encouraged to join the MM programme, and 18 confirmed their intention to participate. Neonatal MM data were never reported by the National Patient Safety Agency.1 Consequently the true impact of MM in the NICU is unclear; data are presently limited to a report from a single centre.2 ,3 It is important that neonatal MM data be formally reported because more evidence is needed regarding the efficacy of catheter-care bundles in NICUs.4 We are pleased to learn that these analyses are currently being undertaken by the Neonatal Data Analysis Unit ( (personal communication, Professor N Modi, 2 May 2013).

Skin antisepsis using 2% chlorhexidine gluconate (CHG) in 70% isopropyl alcohol was one of the key technical interventions in the MM catheter-care bundle.1 ,5 The recommended antiseptic in the neonatal care bundle for participating NICUs was ambiguous because 2% CHG in 70% alcohol and 0.5% CHG in 70% alcohol were separately listed among the MM technical interventions.5 Choice of antiseptics for neonates remains contentious because comparative efficacy data are lacking and alcohol-based CHG agents and 2% aqueous CHG have caused chemical skin burns.6 ,7 In 2007, only 7/50 (14%) UK NICUs used alcohol-based antiseptics and none used 2% aqueous CHG.8

We aimed to determine current prevalence of catheter-care bundles and antiseptic use in the wake of the MM programme. During March–April 2013 we telephone-surveyed all 57 UK tertiary-level NICUs. We asked a senior staff member whether their unit had a catheter-care bundle in place for percutaneous CVC (pCVC) insertions, which topical antiseptic they used for skin disinfection prior to pCVC insertion and whether they had seen chemical skin burns resulting from antiseptic use.

We obtained responses from all 57 (100%) NICUs. Forty (70%) had a MM-type care bundle in place. Seven different cutaneous antiseptics were in use. Thirty of 57 (53%) were now using alcohol-based antiseptics (n=22, 2% CHG in 70% alcohol; n=7, 0.5% CHG in 70% alcohol; n=1 both preparations) and three used 2% aqueous CHG; most (27/33; 82%) were using these ‘stronger’ agents as part of a catheter-care bundle. Twenty-four of 30 NICUs used alcohol-based CHG agents for pCVC insertions in all infants irrespective of gestational age, while six used an aqueous or lower concentration alcohol CHG agent for very premature neonates. Seven (14%) NICUs reported limb skin chemical burns, of which three had subsequently ceased using the 2% CHG in 70% alcohol for very preterm infants.

Our study shows that most tertiary NICUs have now adopted catheter-care bundles. We also show that the widespread introduction of catheter-care bundles into UK NICUs has led to a significant increase in use of alcohol-based CHG antiseptics for neonatal skin preparation, and that chemical burns are indeed still occurring in neonates.6 The lack of safety and efficacy data for antiseptic agents in neonates is of great concern. More potent antiseptics may possibly reduce CVC-associated infections but at the expense of a hidden cost in terms of their own associated morbidity and mortality.6 In the absence of randomised controlled trials to examine safety and efficacy in the neonatal population, babies will potentially continue to be put at risk from introduction of antiseptic agents based on safety and efficacy data that were derived solely from adults and older children.


We thank all practice development nurses, senior nurses and doctors who contributed data, and Professor Neena Modi for helpful comments.