Shared learning database

Type and Title of Submission


Supporting a 36 hour Neonatal Blood Culture status check by developing the availability of blood culture status in real time.


This example describes how Liverpool Women's Hospital in collaboration with Royal Liverpool University Hospital have:
- Developed real time availability of blood culture status for clinicians on the neonatal unit.
- Introduced a policy of discontinuing antibiotics in babies with negative blood cultures and no lab evidence of infection at 36 hours.
- Introduced measures to embed these into clinical practice.

Does the submission relate to the general implementation of all NICE guidance?


Does the submission relate to the implementation of a specific piece of NICE guidance?


Full title of NICE guidance:

CG149 - Antibiotics for early onset neonatal infection

Is the submission industry-sponsored in any way?


Description of submission

Aims and objectives

To reduce antibiotic pressure on the neonatal unit by reducing the number of doses given to babies without infection


Untreated bacterial infection in the neonatal period can progress rapidly to become life threatening. Clinical signs of bacterial infection in these patients are non-specific. Because of this, antibiotics are commonly given to babies with risk factors or potential signs of early infection, who are subsequently found to be uninfected. Local audit had shown that 80% of the antibiotic courses started on our own unit were not completed as the baby is determined to be uninfected. This leads to a significant antibiotic burden within the neonatal unit. Reducing the use of antibiotics in non-infected individuals has been identified as a potential strategy to reduce the emergence of multiresistant pathogens in the hospital environment.

Standard neonatal practice has been for antibiotics to be continued until a negative blood culture result has been obtained after 48 hours. This result is traditionally only available during laboratory 'office hours', leading to continuation of treatment for many babies until the following day. In addition, there is good evidence to show that antibiotics can be safely discontinued after a negative result at 36 hours.


The first step was to make the status (positive or negative) of each blood culture in progress available in real time to the clinicians on the neonatal unit so that antibiotics could be discontinued at any time of the day or night. The blood culture analyser (BacT/Alert Microbial Detection System; Organon Teknika Corporation, Durham, North Carolina, USA) evaluates each culture for a positive status every 10 minutes. This is monitored by a computer system. We established a terminal of the laboratory computer system on the neonatal unit that allows the clinicians to view the status of any blood culture in process at any time. We then implemented a policy for antibiotics to be stopped at 48 hours using this computer link.

The next step was to modify our policy further so that antibiotics should be stopped at 36 hours if the blood culture was negative. We have developed a fully electronic patient record in use on our unit (Badger 3 system, Clevermed, UK) and have built in an automatic prompt for a blood culture status check to appear at 36 hours after any blood culture is taken. We have also developed a "gentamicin pathway" which is a prescription document that prompts for an active decision to continue antibiotics after 36 hours before allowing prescription after this time.

Results and evaluation

We completed an audit cycle to evaluate the impact of the computerised system. From historical data collected during 1997 and 1998, we identified 451 negative blood cultures and found that 144 (38%) had antibiotics continued beyond 48 hours unintentionally. In a repeat audit, following the implementation of the computer system, we identified 179 negative blood cultures during 2000 and found that antibiotics were continued unintentionally beyond 48 hours in only 20 (11.2%) of cases. We measured the combined effect of the computer link and the 36 hour screen check policy by measuring the change in the total antibiotic exposure per baby from a period before the changes were made (1998) with a period after the changes were made (2004). The total number of antibiotic doses received per baby was measured from a random sample of 55 case notes from the 674 babies who had blood cultures taken in 1998. This was compared with the total number of antibiotic doses per baby measured from a random sample of 64 case notes from the 579 babies who had blood cultures in 2004. There were no significant differences in demographics or outcomes between the two groups. The median (range) number of antibiotic doses per baby in 1998 was 25 (0 to 231) versus 14 (0 to 233) in 2004 (p=0.021). From this we estimated a reduction in the total number of antibiotic doses administered on the unit from 27700 to 16900, an equivalent of 10800 per year, or 30 doses per day based on just over 1000 admissions per year. This represents a significant reduction workload for nurses administering these drugs an also reduces the number of medical interventions to which babies are exposed to.

Key learning points

The period of overtreatment for each baby is short when antibiotics are continued longer than planned in babies with negative blood cultures. But because such episodes are numerically much larger than episodes requiring a full course of antibiotics, such overtreatment leads to a significant preventable antibiotic burden for the neonatal unit. Using relatively simple technological solutions to make blood culture status information available in real time allows this to be significantly reduced.

Further reducing the antibiotic burden can be achieved by reducing the 'screen check' time from 48 to 36 hours.

Changing clinical practice requires a clear guideline supported by appropriate staff training. The reduction in workload and the clear improvement in patient experience that accompanies these interventions appear to support their acceptance by clinical staff. Further strategies, such as building decision support into electronic systems, or modifying antibiotic prescription sheets can provide extra support for these policies.

View the supporting material

Contact Details

Name:Charles William Yoxall/ Timothy J Neal
Job Title:Consultant Neonatologist/ Consultant Microbiologist
Organisation:Liverpool Womens NHS Foundation Trust/ Royal Liverpool and Broadgreen University Hospitals NHS Trust
Address:Crown Street/ Prescot Street
Postcode:L23 6UQ


NICE handles personal information provided to the Institute in accordance with the Data Protection Act 1998. Find further details in our data protection policy.

This page was last updated: 16 July 2012

Accessibility | Cymraeg | Freedom of information | Vision Impaired | Contact Us | Glossary | Data protection | Copyright | Disclaimer | Terms and conditions

Copyright 2014 National Institute for Health and Care Excellence. All rights reserved.

Selected, reliable information for health and social care in one place

Accessibility | Cymraeg | Freedom of information | Vision Impaired | Contact Us | Glossary | Data protection | Copyright | Disclaimer | Terms and conditions

Copyright 2014 National Institute for Health and Care Excellence. All rights reserved.