Article Text
PDF
Statistics from Altmetric.com
In April 2021, the National Institute for Health and Care Excellence (NICE) published new guidelines for neonatal infection,1 replacing the 2012 guidance. In order to reduce antibiotic usage in healthy infants, NICE modified risk factors and allowed use of Kaiser Permanente Sepsis Risk Calculator (KP-SRC)2 as an alternative with prospective audit. KP-SRC provides recommendations for antibiotics, culture with observation or observation only. Previous studies,3 including from our group,4 demonstrated significant reductions (up to 84%) in antibiotic usage on virtual application of KP-SRC compared with NICE 2012. No studies examining the impact of NICE 2021, or its effect in conjunction with KP-SRC, have yet been published.
We performed a retrospective application of NICE 2021 using data collected from 11 neonatal units (NNUs) in the West Midlands, UK. We included all infants born ≥34 weeks’ gestation between 1 January 2020 and 29 February 2020, not directly admitted to NNU, started on antibiotics following NICE 2012 guidelines. KP-SRC was also retrospectively applied (early-onset Sepsis (EOS) incidence rate 2/1000; regional rates vary from 0.7 to 1.3/1000).
Data from 626 infants were collected and 572 were included for analysis (figure 1). NICE 2012 recommended antibiotics in 95.5% (of the 626); therefore, 4.5% may have had unnecessary treatment. NICE 2021 reduced antibiotic usage by 39.5% compared with appropriate application of NICE 2012. When KP-SRC was additionally applied, there was an overall reduction from NICE 2012 of 82%. If those recommended cultures by KP-SRC also received antibiotics, as recommended by regional guidelines, the overall reduction from NICE 2012 would be 63.5%.
Retrospective virtual application of the National Institute for Health and Care Excellence (NICE) guidelines for neonatal infection 2021 compared with Kaiser Permanente Sepsis Risk Calculator (KP-SRC) and previous NICE guidelines 2012.
At baseline, 7.3% of live births >34 weeks in postnatal settings were recommended antibiotics using NICE 2012 (table 1). NICE 2021 may reduce this to 4.4% and KP-SRC may reduce it to 2.7%, treating those recommended both culture and antibiotics.
Estimated percentages of live births >34 weeks receiving antibiotics in postnatal settings, following virtual application of NICE 2021 guidelines and KP-SRC
Three infants had a positive blood culture (0.5%), all were recommended antibiotics using both NICE 2012 and 2021. One had group B Streptococcus bacteraemia (C reactive protein (CRP) 28), KP-SRC recommended antibiotics. Two infants had Escherichia coli bacteraemia; both were feverish. One with CRP 88, KP-SRC recommended culture and one with CRP 37, KP-SRC recommended observations. As symptomatic, it is likely that these infants would receive antibiotics under whichever system used. All cerebrospinal fluid cultures were negative. No infants received mechanical ventilation or inotropes, and there were no deaths.
Of the 26 (4.5%) infants with CRP >60, NICE 2021 recommended antibiotics in 18 and KP-SRC recommended antibiotics in 9, culture in 11 and observations in the rest (6).
We conclude that NICE 2021 may reduce antibiotic exposure in infants on the postnatal ward by up to 39.5%. Subsequent implementation of KP-SRC to those infants could reduce antibiotic exposure by up to 63.5%–82%. In clinical practice, some infants on KP-SRC observations may later become symptomatic needing treatment and therefore the reduction may be slightly less. A prospective audit with KP-SRC as recommended by NICE would be beneficial in reviewing the safety and efficacy of KP-SRC.
Ethics statements
Patient consent for publication
Acknowledgments
We would like to acknowledge the input of all the units in the West Midlands operational network who took part in the data collection and all of our PRAM collaborators. Specific mention goes to PRAM lead Helen Mcdermott for her support and guidance and to Megha Jagga for her input in the beginning of the project.
Footnotes
Twitter @drlauramac, @94naseem
Correction notice This paper has been corrected since it was first published. In Table 1, column 3 it should read '2012' instead of '2021'.
Collaborators PRAM Collaborators: Early project advice and support: Helen McDermott Local leads: Dr Hannah Vawda, Ms Sarah Kirk, Dr Moataz Badawy, Dr Pooja Siddhi. Local data collection and analysis: Megha Jagga, Oluwaseyi Alake, Canada Cherry, Benjamin Miguras, Raunak Jain, Nicola McMullan, Rawia Eltahir, Dr Sebastian Brown. Local data collection: Dr Mshail Mahmoud, Dr Vishanna Balbirsingh, Dr Catherine Lomax, Dr Emmeline Parfitt-Roche, Dr Andrew Pearce, Dr Lindsay Mulligan, Dr Afza Sadiq. Initial pilot work: Dr Sonia Goyal. Local Consultant Leads: Dr Kate Palmer, Dr Chandan Gupta, Dr Mohammed Buhary, Dr Michael Plunkett, Dr Paul Watson, Dr Robert Negrine, Dr Penny Broggio, Dr Raghu Krishnamurthy, Dr Cathryn Seagrave.
Contributors This study is a collaboration lead by Paediatric Research Across the Midlands (PRAM), the West Midlands trainee-led regional research group. LM (1), SS (2), TvH (2), NM (2), AKE (2,3), PS (1), Paediatric Research Across the Midlands (PRAM) Network: (1) Birmingham Heartlands Hospital, (2) Birmingham Women’s and Children’s NHS Foundation Trust, (3) Institute of Metabolism and Systems Research, University of Birmingham. Supervision, project design, manuscript review: PS, AKE. Data analysis and manuscript: LM, SS, TvM, MN.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; internally peer reviewed.