U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Cover of Evidence reviews for maternal and neonatal risk factors for early-onset neonatal infection

Evidence reviews for maternal and neonatal risk factors for early-onset neonatal infection

Neonatal infection: antibiotics for prevention and treatment

Evidence review D

NICE Guideline, No. 195

London: National Institute for Health and Care Excellence (NICE); .
ISBN-13: 978-1-4731-4080-6

Risk factors for early-onset neonatal infection

1.1. Review question

What is the accuracy of clinical prediction models for early-onset neonatal infection and what is their effectiveness in guiding management in the baby?

1.1.1. Introduction

Neonatal infection is a significant cause of mortality and morbidity in newborn babies. It can lead to life-threatening sepsis, which accounts for 10% of all neonatal deaths. Early-onset neonatal infection is typically defined as infection that occurs within 72 hours of birth.

Predicting which babies are most at risk of early-onset neonatal infection is important to help determine who should receive antibiotic treatment. A culture taken from blood or cerebrospinal fluid which tests positive for the organisms associated with infection is the gold-standard for identifying which babies should receive treatment. However, waiting for the results of these tests may delay treatment. A tool which can predict which babies are most at risk of early-onset neonatal infection is therefore important to help identify those who will benefit from early treatment whilst reducing the number of babies who receive unnecessary treatment. This will also reduce other associated risks, such as the development of antimicrobial resistance. The aim of this review is therefore to evaluate existing clinical prediction models for early-onset neonatal infection and determine their effectiveness in guiding management of the baby.

1.1.2. Summary of the protocol

The review was divided into 2 parts. Part A aimed to identify studies assessing the accuracy of clinical prediction models in identifying babies with early onset infection. Part B aimed to identify ‘test and treat’ randomised controlled trials that assessed the effectiveness of clinical prediction models in guiding management.

Part A.

Table

Part A.

Part B.

Table

Part B.

1.1.3. Methods and process

This evidence review was developed using the methods and process described in Developing NICE guidelines: the manual. Methods specific to this review question are described in the review protocol in Appendix A. For full details of the methods used in this review, see the methods document.

Declarations of interest were recorded according to NICE’s 2018 conflicts of interest policy.

Prospective and retrospective observational cohort or cross-sectional studies (part A) and test and treat randomised controlled trials (part B) were considered in addition to systematic reviews of these study types. The review protocols specified that, where possible, subgroup analyses would be conducted for gestational age of the baby (preterm vs term). However, this was not possible as most studies included both preterm and term babies, and the results were not separated by gestational age. Results were stratified by population where the models were evaluated on different cohorts, such as mothers with chorioamnionitis.

1.1.4. Prognostic evidence

1.1.4.1. Included studies

A joint search was carried out to identify studies specified for this evidence review, and a similar evidence review for studies assessing clinical prediction models for late-onset infection (for details, see evidence review E – Risk factors for late onset). This returned a total of 1,252 results, of which 68 were identified as potential included studies. Full text articles were ordered and reviewed against the inclusion criteria, of which 7 met the inclusion criteria for the review. Two additional studies were included from a systematic review making 9 included references in total.

The search was re-run in July 2020 to identify any studies which had been published since the date of the original search. This combined search for early- and late-onset prediction models returned a total of 244 results of which 14 were identified as possible included studies. After full text review, 13 were excluded. The one included study examined the use of a clinical prediction model for early-onset infection. In total there were therefore 10 studies which met the inclusion criteria for this review (5 prospective cohort studies, 5 retrospective cohort studies).

The majority of the evidence (9 studies) investigated the use of the Kaiser Permanente neonatal sepsis calculator, including one study that compared the prognostic accuracy of the calculator against the recommendations from the 2012 version of this guideline. One study examined the use of a different model, based on various demographic and clinical factors. Studies reported the information needed to calculate prognostic outcomes for sensitivity, specificity and likelihood ratios. Only one study reported model fit statistics (c-statistic), and none reported hazard or odds ratios. No studies matched the protocol for Part B of the review (RCTs for different risk predictor tools).

1.1.4.2. Excluded studies

See Appendix J for excluded studies and reasons for exclusion.

1.1.5. Summary of studies included in the prognostic evidence

Table 2. Summary of included clinical studies.

Table 2

Summary of included clinical studies.

See appendix D for full evidence tables.

1.1.6. Summary of the prognostic evidence

1.1.6.1. Model summaries
Kaiser Permanente neonatal sepsis model

The Kaiser Permanente neonatal sepsis model was developed in the USA and is designed to predict the risk of early-onset neonatal infection for any baby born at or after 34 weeks’ gestational age. The model was developed from data that can be obtained from a patient’s electronic medical record and requires a clinician to enter information on the local incidence of early-onset infection, gestational age of the baby, highest maternal antepartum temperature, duration of rupture of membranes, maternal group B streptococcal (GBS) status and the type and duration of intrapartum antibiotics given to the mother. This information is used to calculate a baby’s risk of infection at birth. The clinician then determines whether the baby is well appearing, equivocal or has clinical illness and this information is used to provide guidance on how the baby should be treated. The calculator produces three recommendations depending on the baby’s risk of infection; if a baby is at low risk then ‘no culture or antibiotics’ are recommended, if they are at moderate risk of infection then ‘blood culture’ is recommended alongside vitals every 4 hours for 24 hours and if the baby is at high risk then the recommendation is for ‘empiric antibiotics’.

Popowski 2011 model

The model reported by Popowski was developed in France and designed for babies born at or after 34 weeks’ gestational age whose mothers had prelabour rupture of membranes. The model was developed from information from serum samples and vaginal swabs at admission and data that could be obtained from a patient’s electronic medical record. The final model includes white blood cell count and C-reactive protein levels. The study reported information on the algorithms used for the model but there is no evidence of a web-based tool or software that can be used directly by a clinician.

1.1.6.2. Summary of clinical findings included in the evidence review
Sensitivity, specificity and likelihood ratios.

Table

Sensitivity, specificity and likelihood ratios.

C-statistics.

Table

C-statistics.

See appendix F for full GRADE tables.

1.1.7. Economic evidence

1.1.7.1. Included studies

A single search was performed to identify published economic evaluations of relevance to any of the questions in this guideline update (see Appendix B). This search retrieved 4,398 studies. Based on title and abstract screening, all of the studies could confidently be excluded for this question, as none of them were found to be relevant.

The search was re-run in July 2020 to identify any studies which had been published since the date of the original search. This returned a total of 577 results. Based on title and abstract screening, all the studies could confidently be excluded for this question. Thus, the review for this question does not include any study from the existing literature.

1.1.8. Economic model

This question was not prioritised for original economic analysis

2.1. Review question

Which maternal and fetal risk factors for early-onset neonatal infection should be used to guide management in the baby?

2.1.1. Introduction

Neonatal infection is a significant cause of mortality and morbidity in newborn babies. It can lead to life-threatening sepsis, which accounts for 10% of all neonatal deaths. For the purpose of this guideline, early-onset neonatal infection is defined as infection which occurs in babies up to 72 hours of age (corrected for gestational age).

Predicting which babies are most at risk of early-onset neonatal infection is important to help determine who should receive antibiotic treatment. A culture taken from blood or cerebrospinal fluid which tests positive for the organisms associated with infection is the gold-standard for identifying which babies should receive treatment. However, waiting for the results of these tests may delay treatment. Identifying the factors which can put a baby at high risk of neonatal infection is therefore important to help determine whether a baby should be given antibiotics while waiting for the results of a blood culture to confirm infection. The aim of this review is therefore to evaluate potential risk factors in the mother and the fetus and determine how well they can guide management of the baby.

2.1.2. Summary of the protocol

Population
  • Unborn or newborn babies under 72 hours
  • Pregnant women
Risk factors
  • Behavioural and hygienic factors (for example adherence to infection control measures by medical professionals and parents/carers)
  • Gestational age
  • Intrapartum antibiotic prophylaxis (including the time before birth that it is received)
  • Intrapartum fever higher than 38°C, or confirmed or suspected chorioamnionitis
  • Invasive group B streptococcal (GBS) infection in a previous baby
  • Maternal carriage of Methicillin-resistant Staphylococcus aureus (MRSA)
  • Maternal GBS colonisation, bacteriuria, detection (including vaginal or rectal swab) or infection in the current pregnancy
  • Maternal GBS colonisation, bacteriuria, detection (including vaginal or rectal swab) or infection in a previous pregnancy (including where the baby was well)
  • Maternal obesity
  • Maternal perineal infections
  • Maternal suspected bacterial infection in the puerperium period
  • Parenteral antibiotic treatment given to the woman for confirmed or suspected invasive bacterial infection (such as septicaemia) at any time during labour, or in the 24-hour periods before and after the birth [This does not refer to intrapartum antibiotic prophylaxis]
  • Preterm prelabour rupture of membranes
  • Suspected or confirmed infection in another baby in the case of a multiple pregnancy
  • Suspected or confirmed rupture of membranes for more than 18 hours in a preterm birth
Reference standard
  • culture-proven infection from a sample taken within 72 hours of birth where available or within the study-defined period for early onset neonatal infection
  • antibiotics for suspected bloodstream infection (in neonate) given within 72 hours of birth where available or within the study-defined period for early onset neonatal infection
OutcomesOutcomes for predictive accuracy studies:
  • Sensitivity
  • Specificity
  • Positive and negative predictive values
  • Positive and negative likelihood ratios
If association studies are included due to a lack of predictive accuracy data:
  • Adjusted Risk ratios, Odds ratios, hazard ratios

2.1.3. Methods and process

This evidence review was developed using the methods and process described in Developing NICE guidelines: the manual. Methods specific to this review question are described in the review protocol in Appendix A. For full details of the methods used in this review, see the methods document.

Declarations of interest were recorded according to NICE’s 2018 conflicts of interest policy.

Predictive accuracy studies were considered in addition to systematic reviews. For outcomes where no predictive accuracy studies were available, multivariate cohort studies that reported adjusted measures of association were included. The review protocol specified that, where possible, subgroup analyses would be conducted for gestational age of the baby (preterm vs term) and multiple births. Some evidence was available for single and multiple births, and where studies have included babies of different gestational ages this has been highlighted in the results.

Some studies reported outcomes that matched the protocol but were only reported as part of univariate analysis and not included in multivariate analysis. These outcomes are stated in the clinical evidence tables (Appendix D) but not reported in the analysis as they did not meet the criteria for multivariate analyses.

2.1.4. Prognostic evidence

2.1.4.1. Included studies

A joint search was carried out to identify studies specified for this review question, and a similar review question for studies assessing risk factors and signs and symptoms in the baby for early-onset infection (for details, see section 3.1 of this evidence review). This returned a total of 1,825 results, of which 55 were identified as potential included studies. Full text articles were ordered and reviewed against the inclusion criteria, of which 11 met the inclusion criteria for this review.

The search was re-run in July 2020 to identify any studies which had been published since the date of the original search. This combined search for late-onset maternal and fetal risk factors and risk factors in the baby returned a total of 143 results, of which 3 were identified as possible included studies. After full text review, all 3 were excluded. In total there were therefore 11 studies which met the inclusion criteria for this review (3 prospective cohort studies, 8 retrospective cohort studies). No studies reported predictive accuracy data and so prognostic association data was considered instead.

Most of the multivariate cohort studies identified reported on the association between early-onset neonatal infection and chorioamnionitis (6 studies) or singleton births (3 studies). One study reported on the association between maternal obesity and early-onset infection and another examined the effects of intrapartum fever.

2.1.4.2. Excluded studies

See Appendix J for excluded studies and reasons for exclusion.

2.1.5. Summary of studies included in the prognostic evidence

StudyStudy type and follow-up timePopulationPredictive factors

Dempsey 2005

(n=392)

  • Retrospective cohort
  • 72 hour follow-up
  • All singleton neonates delivered at <30 weeks gestational age
  • Chorioamnionitis

Dior 2016

(n=46,560)

  • Retrospective cohort
  • 72 hour follow-up
  • Women in labour who had a singleton live birth
  • <50 years old
  • Had a term pregnancy (≥ 37 weeks’ gestation)
  • Baby with a birth weight <5000 g
  • Spent >1 hour in the delivery room
  • Intrapartum fever

Garcia-Munoz 2014a

(n=451)

  • Prospective cohort
  • 72 hour follow up
  • Born in maternity unit or admitted to Neonatal Intensive Care Unit in the first 28 days of life
  • Birth weight <1500 g
  • or <30 weeks’ gestational age
  • Chorioamnionitis

Garcia-Munoz 2014b

(n=8330)

  • Retrospective cohort
  • Duration of follow-up not reported
  • Birth weight <1500 g
  • <32 weeks’ gestational age
  • Admitted to a neonatal unit
  • Chorioamnionitis

Hakansson 2008

(n=344,127)

  • Retrospective cohort
  • Follow-up for rfirst 27 days of life
  • Gestational age >22 weeks
  • Vaginal birth or emergency caesarean section
  • Maternal BMI
    (weight grouped by BMI. Obesity classified as BMI ≥30.0)

Klinger 2009 *

(n=15,839)

  • Retrospective cohort
  • Duration of follow up unclear
  • Infants whose data was collected by the Israel Neonatal Network on very low birth weight newborn infants (BW <1500 g)
  • Chorioamnionitis
  • Maternal fever
  • Single/multiple birth

Mularoni 2014

(n=14,719)

  • Prospective cohort
  • 72 hour follow-up
  • Babies weighing 401 - 1500 g
  • Babies with a positive blood culture and clinical signs of sepsis during the first 72 hours of life
  • Twin/singleton births

Ofman 2016 *

(n=2192)

  • Retrospective cohort
  • 72 hour follow-up
  • Moderately preterm infants
  • Chorioamnionitis

Ronnestad 2005

(n=462)

  • Prospective cohort
  • Follow-up for first week of life
  • Birth weight <1000 g
  • Chorioamnionitis
  • Gestational age

Soraisham 2009

(n=3094)

  • Retrospective cohort
  • Follow-up for first 48 hours after birth
  • All singleton infants with birth gestational age <33 weeks
  • No congenital anomalies
  • Chorioamnionitis
*

Also included in review question on risk factors in the baby for neonatal infection

See appendix D for full evidence tables.

2.1.6. Summary of the prognostic evidence for predicting the development of early-onset neonatal infection

Risk factorNo. studiesSample sizeEffect size (95% CI)Quality
Chorioamnionitis
Histological chorioamnionitis (babies <30 weeks’ gestational age)1 (Dempsey 2005)392

Adjusted OR 6.9

(2.2, 20.0)

Moderate
Clinical chorioamnionitis in very low birth weight babies1 (Garcia-Munoz 2014a)451

Adjusted RR 6.13

(1.67, 22.58)

Moderate
1 (Garcia-Munoz 2014b)8330

Adjusted OR 3.10

(2.31–4.17)

Moderate
Clinical chorioamnionitis in preterm babies1 (Soraisham 2009)3094

Adjusted OR 5.54

(2.87–10.69)

Moderate
Clinical chorioamnionitis in moderately preterm babies1 (Ofman 2016)2192

Adjusted OR 4.1

(2.83–5.30)

Low
Clinical chorioamnionitis in extremely preterm babies1 (Ronnestad 2005)451

Adjusted OR 10.5

(3.3–33.4)

Moderate
Intrapartum fever
Low febrile fever (38.0–38.9◦C)1 (Dior 2016)43,560

Adjusted OR 7.44

(3.29, 16.85)

Moderate
High febrile fever (>39◦C)1 (Dior 2016)43,560

Adjusted OR 16.08

(2.15, 120.3)

Moderate
Maternal obesity
Overweight mothers (BMI 25–29.9)1 (Hakansson 2008)344,127

Adjusted OR 1.3

(0.9, 2.0)

Low
Obese mothers (BMI 30.0)1 (Hakansson 2008)344,127

Adjusted OR 1.8

(1.1, 3.0)

Moderate
Single vs multiple births
Very low birth weight babies1 (Mularoni 2014)14,719

Adjusted OR 1.4

(1.1, 1.8)

Moderate
1 (Klinger 2009)15,839

Adjusted OR 1.4

(1.1, 1.8)

Moderate
Babies born ≥22 weeks’ gestational age1 (Hakansson 2006)319

Adjusted OR 1.1

(0.6, 2.0)

Moderate

See appendix F for full GRADE tables.

2.1.7. Economic evidence

2.1.7.1. Included studies

A single search was performed to identify published economic evaluations of relevance to any of the questions in this guideline update (see appendix B). This search retrieved 4,398 studies. Based on title and abstract screening, all of the studies could confidently be excluded for this question, as none of them were found to be relevant.

The search was re-run in July 2020 to identify any studies which had been published since the date of the original search. This returned a total of 577 results. Based on title and abstract screening, all the studies could confidently be excluded for this question. Thus, the review for this question does not include any study from the existing literature.

2.1.8. Economic model

No economic modelling was undertaken for this review because of a lack of economic evidence and because the committee agreed that other topics were higher priorities for economic evaluation.

3.1. Review question

Which risk factors in the baby (including symptoms and signs) should raise suspicion of early-onset neonatal infection?

3.1.1. Introduction

Neonatal infection is a significant cause of mortality and morbidity in newborn babies. It can lead to life-threatening sepsis, which accounts for 10% of all neonatal deaths. For the purpose of this guideline, early-onset neonatal infection is defined as infection which occurs within 72 hours of birth (corrected for gestational age).

Predicting which babies are most at risk of early-onset neonatal infection is important to help determine who should receive antibiotic treatment. A culture taken from blood or cerebrospinal fluid which tests positive for the organisms associated with infection is the gold-standard for identifying which babies should receive treatment. However, waiting for the results of these tests may delay treatment. Identifying the factors which can put a baby at high risk of neonatal infection is therefore important to help determine whether a baby should be given antibiotics while waiting for the results of a blood culture to confirm infection. These factors can either be diagnostic, such as the signs and symptoms that babies commonly display when they have an infection, or prognostic, such as factors that are commonly associated with babies subsequently developing an infection. The aim of this review is therefore to evaluate potential risk factors as well as signs and symptoms in the baby to determine how they can guide management of the baby.

3.1.2. Summary of the protocol

Population
  • Newborn babies under 72 hours, or study definition for ‘early onset’ infection
Risk factorsSigns and symptoms (diagnostic)
  • Abnormal heart rate (bradycardia or tachycardia)
  • Altered behaviour or responsiveness
  • Altered glucose homeostasis (hypoglycaemia or hyperglycaemia)
  • Altered muscle tone (for example, floppiness)
  • Apnoea
  • Feed intolerance, including vomiting, excessive gastric aspirates and abdominal distension
  • Feeding difficulties (for example, feed refusal)
  • Jaundice
  • Local signs of infection (for example, affecting the skin or eye)
  • Metabolic acidosis (base deficit of 10 mmol/litre or greater)
  • Need for cardio-pulmonary resuscitation
  • Need for mechanical ventilation
  • Oliguria
  • Reduced oxygen saturation level
  • Seizures
  • Signs of neonatal encephalopathy
  • Signs of respiratory distress
  • Signs of shock
  • Temperature abnormality (lower than 36°C or higher than 38°C) unexplained by environmental factors
  • Unexplained excessive bleeding, thrombocytopenia, or abnormal coagulopathy (International Normalised Ratio greater than 2.0)
Risk factors (prognostic)
  • Gestational age
  • Colonisation with Group B streptococcus (GBS) or Methicillin-resistant Staphylococcus aureus (MRSA) in the baby
  • Persistent fetal circulation (persistent pulmonary hypertension)
Reference standard
  • culture-proven infection from a sample taken within 72 hours of birth where available or within the study-defined period for early onset neonatal infection
  • antibiotics for suspected bloodstream infection (in neonate) given within 72 hours of birth where available or within the study-defined period for early onset neonatal infection
OutcomesOutcomes for diagnostic/predictive accuracy studies:
  • Sensitivity
  • Specificity
  • Positive and negative predictive values
  • Positive and negative likelihood ratios
If association studies are included due to a lack of predictive accuracy data:
  • Adjusted Risk ratios, Odds ratios, hazard ratios

3.1.3. Methods and process

This evidence review was developed using the methods and process described in Developing NICE guidelines: the manual. Methods specific to this review question are described in the review protocol in Appendix A. For full details of the methods used in this review, see the methods document.

Declarations of interest were recorded according to NICE’s 2018 conflicts of interest policy.

Predictive accuracy studies were considered in addition to systematic reviews. For outcomes where no predictive accuracy studies were available, multivariate cohort studies that reported adjusted measures of association were included. The review protocol specified that, where possible, subgroup analyses would be conducted for gestational age of the baby (preterm vs term), multiple births and babies who have been admitted to hospital from home. Data was available for gestational age, but no data was reported for babies admitted to hospital from home. Some evidence was available for multiple births, and this was reported as part of the maternal risk factors review (for details, see section 2.1). Evidence was separated by those which reported prognostic (risk factors for infection) and diagnostic (signs and symptoms of infection) factors.

3.1.4. Prognostic and diagnostic evidence

3.1.4.1. Included studies

A joint search was carried out to identify studies specified for this review question, and a similar review question for studies assessing maternal and fetal risk factors for early-onset infection (for details, see section 2.1). This returned a total of 1,825 results, of which 55 were identified as potential included studies. Full text articles were ordered and reviewed against the inclusion criteria, of which 4 met the inclusion criteria for this review. No studies reported diagnostic or predictive accuracy data.

The search was re-run in July 2020 to identify any studies which had been published since the date of the original search. This combined search for late-onset maternal and fetal risk factors, and risk factors in the baby returned a total of 143 results, of which 3 were identified as possible included studies. After full text review, all 3 were excluded. In total there were therefore 4 studies which met the inclusion criteria for this review (1 prospective cohort study, 3 retrospective cohort studies). No studies reported predictive accuracy data and so prognostic and diagnostic association data was considered instead.

Of the 4 multivariate cohort studies identified, most were prognostic and reported on the association between early-onset neonatal infection and gestational age (3 studies). One diagnostic study reported on the association between respiratory distress syndrome and early-onset infection.

3.1.4.2. Excluded studies

See Appendix J for excluded studies and reasons for exclusion.

3.1.5. Summary of studies included in the prognostic and diagnostic evidence

StudyStudy type and follow-up timePopulationPredictive factors

Hakansson 2006

(n=319)

  • Retrospective cohort (prognostic)
  • Follow-up for first 27 days of life
  • Gestational age >21 weeks
  • Gestational age
  • Twin/singleton births (subgroup analysis)

Klinger 2009 *

(n=15,839)

  • Retrospective cohort (prognostic)
  • Duration of follow up unclear
  • Infants whose data was collected by the Israel Neonatal Network on very low birth weight newborn infants (BW <1500 g)
  • Single/multiple birth
  • Gestational age

Ofman 2016 *

(n=2192)

  • Retrospective cohort (diagnostic)
  • 72 hour follow-up
  • Moderately preterm infants
  • Respiratory distress syndrome

Ronnestad 2005 *

(n=462)

  • Prospective cohort (prognostic)
  • Follow-up for first week of life
  • Birth weight <1000 g
  • Gestational age
*

Also included in review question on maternal and fetal risk factors for neonatal infection

See appendix D for full evidence tables.

3.1.6. Summary of the prognostic evidence for predicting the development of early-onset neonatal infection

3.1.6.1. Risk factors
Risk factorNo. studiesSample sizeEffect size (95% CI)Quality
Gestational age
Very early onset infection in extremely premature babies1 (Ronnestad 2005)462

Adjusted OR1 1.1

(0.4, 3.6)

Low
Early onset infection in extremely premature babies1 (Ronnestad 2005)462

Adjusted OR1 3.0

(0.6, 14.9)

Low
Very low birthweight babies (1-week increase)1 (Klinger 2009)15,839

Adjusted OR1 0.98

(0.94, 1.03)

Low
Babies born ≥22 weeks’ gestational age (<28 weeks vs 40 weeks)1 (Hakansson 2006)319

Adjusted OR1 22.1

(8.5, 57.4)

Moderate
Babies born ≥22 weeks’ gestational age (28–31 weeks vs 40 weeks)1 (Hakansson 2006)319

Adjusted OR1 34.1

(18.6, 62.7)

Moderate
Babies born ≥22 weeks’ gestational age (32–34 weeks vs 40 weeks)1 (Hakansson 2006)319

Adjusted OR1 11.2

(6.0, 21.0)

Moderate
Babies born ≥22 weeks’ gestational age (35–36 weeks vs 40 weeks)1 (Hakansson 2006)319

Adjusted OR1 4.7

(2.5, 8.9)

Moderate
Babies born ≥22 weeks’ gestational age (37 weeks vs 40 weeks)1 (Hakansson 2006)319

Adjusted OR1 3.5

(1.8, 6.5)

Moderate
Babies born ≥22 weeks’ gestational age (≥42 weeks vs 40 weeks)1 (Hakansson 2006)319

Adjusted OR1 1.9

(0.9, 3.7)

Low
3.1.6.2. Signs and symptoms
ComparisonNo. studiesSample sizeEffect size (95% CI)Quality
Respiratory distress syndrome in moderately preterm babies1 (Ofman 2016)2192

Adjusted OR 2.05

(1.62–3.14)

Low

See appendix F for full GRADE tables.

3.1.7. Economic evidence

3.1.7.1. Included studies

A single search was performed to identify published economic evaluations of relevance to any of the questions in this guideline update (see appendix B). This search retrieved 4,398 studies. Based on title and abstract screening, all of the studies could confidently be excluded for this question, as none of them were found to be relevant.

The search was re-run in July 2020 to identify any studies which had been published since the date of the original search. This returned a total of 577 results. Based on title and abstract screening, all the studies could confidently be excluded for this question. Thus, the review for this question does not include any study from the existing literature.

3.1.8. Economic model

This question was not prioritised for original economic analysis.

4.1. The committee’s discussion and interpretation of the evidence

4.1.1. The outcomes that matter most

The committee discussed the potential effects of true positive, true negative, false positive and false negative outcomes from tools designed to predict a baby’s risk of early-onset infection. A model that correctly identifies all those with infection (true positives) would result in antibiotics being prescribed to all those who need treatment, reducing the serious harms associated with neonatal infection. If a model correctly identifies all those without infection (true negatives) then it will avoid over-prescribing of antibiotics. This is a particular issue when evaluating neonatal infection as it can be difficult to diagnose and can therefore result in many babies being prescribed antibiotics to avoid any infections being missed and left untreated.

If a model does not accurately predict true positives and true negatives, then there are a number of potential harms. False positive results will result in babies being given antibiotics unnecessarily, and either admission to hospital or prolonged admission in hospital will lead to separation of the mother and baby, potentially causing anxiety and distress to the family. The mother may also have to remain in hospital for longer than she otherwise would. This has an impact on the family as well as the increasing costs of a longer hospital stay. False positives can also contribute to the development of antibiotic resistance. However, a false negative result is the biggest concern for parents and clinicians as there can be serious consequences if neonatal infection is left untreated. The most serious consequence is death of the baby, but delayed treatment can also have long-term health consequences, such as neuro-disability, which can have both emotional and financial impacts on the family as well as downstream treatment costs for the healthcare system. The committee therefore discussed how negative likelihood ratios should be prioritised over positive likelihood ratios as they believed that it was important that negative test results were accurate, and that neonatal infection was not incorrectly ruled out. As a result, the committee used a combination of likelihood ratios, sensitivity and specificity to examine the effectiveness of each clinical prediction model. Some studies also reported c-statistics. The committee agreed that this outcome was less useful as it weighs false negatives and false positives as equally important, which the committee agreed was not appropriate.

The committee highlighted that while babies tend to be discharged between 6–12 hours after birth if they appear healthy, there are often no follow-up observations in the community during the first 72 hours of life. It is therefore important for any prognostic model or risk factor tool to avoid false negative results, thereby ensuring that any babies at high risk of infection remain in the hospital for treatment and monitoring.

The committee were also interested in the diagnostic accuracy of risk factors and signs and symptoms for early-onset infection. No studies reported sensitivity, specificity, predictive values or likelihood ratios for these outcomes. Instead, studies reported adjusted odds ratios, and risk ratios. These do not provide an indication of how well each risk factor or symptom can correctly identify a baby with or without early-onset neonatal infection. However, the committee were interested in these outcomes as they indicate which factors have the strongest association with infection. This helped them to decide which were the most important risk factors and signs to include in the risk factor and clinical indicators boxes in the recommendations. As the committee only wanted to consider the factors which were most relevant to early-onset infection, studies were only included if they had used a multivariate analysis which had adjusted for other potential risk factors for infection.

4.1.2. The quality of the evidence

4.1.2.1. Clinical prediction models

The outcomes from the evidence for prognostic models ranged from high to very low quality, with most outcomes either low or very-low quality. The quality of outcomes were commonly downgraded for imprecision in the results and for including studies at moderate risk of bias. All of the studies were directly applicable to the research question. Most of the evidence examined the use of the Kaiser Permanente neonatal sepsis calculator, with only one study reporting the details of a different prognostic model. One other study also compared the use of the calculator to the NICE recommendations. The committee therefore decided that only the neonatal sepsis calculator had sufficient evidence to be considered as an alternative to the NICE risk factors and clinical indicators.

Four studies examined the use of the Kaiser Permanente neonatal sepsis calculator in all babies over 34-, 35- or 36-weeks’ gestational age. Five studies examined the use of the calculator specifically for babies who were born to mothers with chorioamnionitis. The committee discussed whether the results from the chorioamnionitis studies could be applied to all babies but, as the results did not differ greatly from the three studies that included all babies, the committee thought that the results could be generalised. Six of the studies in the review were based in the USA where the calculator was developed. Only one study (Goel 2020) examined the use of the calculator in the UK. The committee discussed whether differences in the demographics of the population, in addition to differences in group B streptococcal (GBS) screening policies, could mean that the outcomes of the calculator may differ between the UK and the USA. GBS status is one of the factors in the neonatal sepsis calculator and the committee were unsure whether the unknown status in the UK would make assumptions about the mother that might change the outcome of the neonatal sepsis calculator. However, GBS status only contributes approximately 2% to the predictive weight of the model. This, in addition to the similarities in results between the UK study and the USA validation studies, meant that the committee did not feel it should recommend against the use of the neonatal sepsis calculator based on GBS status.

An important aspect of the neonatal sepsis calculator is the need for clinicians to enter a background rate for the incidence of early-onset neonatal infection. Further research is needed in the UK to identify what rate should be applied to reflect the local prevalence of neonatal infection, and whether this should be a single rate that is used across the UK or if it should vary by hospital. The committee did not feel that this should be a reason to recommend against the use of the neonatal sepsis calculator but potential differences in the incidence rate of infection between the UK and the USA, and the differences in GBS screening policies, meant that the studies that reported on the use of the calculator in the USA were downgraded as being partially applicable to the research question. The committee decided that the rate of infection should be considered as part of future research into the use of the neonatal sepsis calculator in the UK. This will help to ensure the accuracy of the outcomes of the calculator when used in NHS practice. For clinicians who are currently using the calculator, the committee decided to recommend that they should use either the national or local rate of infection, whichever is higher. This will help to increase the sensitivity of the calculator, thereby reducing the risk of a baby who has infection not being recommended for antibiotic treatment.

The imprecision of some of the results from the neonatal sepsis calculator studies was discussed, particularly when comparing the overall results of the calculator to the NICE guidelines. The neonatal sepsis calculator identified a slightly lower number of true positives to the NICE guidelines, and the likelihood ratios indicated that a positive outcome with the neonatal sepsis calculator would indicate a slightly higher chance of a baby having infection than a positive test from the NICE guidelines. However, the wide confidence intervals suggested variation in the results. This was particularly evident for the negative likelihood ratio for the NICE guidelines, which suggested that a negative outcome would indicate anything between a large decrease in the probability of a baby having infection to a slight increase in their chance of having infection. This made it difficult for the committee to decide on recommendations based on likelihood ratios alone. This imprecision in the results may be due to the very low incidence of culture-confirmed early-onset neonatal infection in the studies, with many studies reporting only one baby with a confirmed infection. Given that both the NICE recommendations and the neonatal sepsis calculator were associated with a wide degree of imprecision, the committee did not think this should be a reason against recommending the neonatal sepsis calculator. Instead, they decided to recommend that the risk factors specified in boxes 1 and 2 is used to identify babies at risk of early-onset infection, but also stated that if the calculator is being used in clinical practice it should be used as part of a clinical audit. This will help give a clearer understanding of its effectiveness. Examples of the outcomes that should be collected in an audit (number of babies assessed using the calculator and number of true positive, false positive and false negative results when using the calculator) were included in the recommendation to ensure that the most relevant information on effectiveness is available. These outcomes were chosen as the ones needed to form a 2 x 2 table to examine the effectiveness of a prognostic tool. This information will be useful in future updates of this guideline to decide whether a stronger recommendation can be made in favour of the use of the Kaiser Permanente neonatal sepsis calculator.

Given the issues discussed above, the committee agreed that more evidence is needed before it could recommend the Kaiser Permanente neonatal sepsis calculator as the sole option for predicting risk of early-onset neonatal infection in the UK. More research is particularly important for babies born before 34 weeks’ gestational age, as the neonatal sepsis calculator is only designed for babies born at a gestational age of 34 weeks or above. As such, the committee could not recommend the use of the calculator for this group of babies and this was stated as part of the calculator recommendations. Instead, a research recommendation relating to prognostic models was made which did not specify gestational age (Appendix K). This should help to ensure that the most effective prognostic tool can be determined for all babies, and not just those covered by the calculator.

4.1.2.2. Maternal and neonatal risk factors and clinical indicators

The evidence for risk factors and signs and symptoms ranged from high to low quality and most studies were directly applicable to the research question. Each of the studies reported the use of a multivariate model but there was a wide range in the factors that the models were adjusted for, and most studies did not explain why those particular factors were chosen. The committee agreed that this could affect the validity of the data and so the quality of the outcomes from these studies were downgraded for risk of bias. However, the factors that were highlighted as potential risks for infection were consistent with their clinical experience. As such, they decided that they could still be identified as risk factors in the recommendations.

There was also variation in the populations that were included in each study, with some basing the inclusion criteria on gestational age or birthweight while other studies included all babies born in a particular setting. Consequently, the results were presented by individual study outcomes rather than pooled effect estimates. An additional issue was that many studies only reported the significant results from the models. This means that while one study may have reported an association between a particular factor and neonatal infection, it is unclear how many other studies also investigated that risk factor but found non-significant results. Where studies only reported significant results, they were therefore downgraded for risk of bias. The committee decided that risk factors would only be included as part of the recommendations if the evidence corresponded with their clinical experience.

Only one study examined the effects of maternal obesity on the risk of early-onset neonatal infection. The committee questioned the applicability of this research as mothers who were included in the analysis were grouped by the World Health Organisation definition of obesity (BMI of 30.0 and above). It was highlighted that in clinical practice maternal obesity is now often defined as women with a BMI of 35 and above. Due to these differences in classification, this study was graded as partially applicable to the research question and the quality of the outcomes were downgraded. With such limited and partially applicable evidence, the committee decided that more relevant research was needed before recommendations could be made on maternal obesity and neonatal infection. A research recommendation was therefore made to reflect this (Appendix K).

The committee discussed the criteria for chorioamnionitis that was used in the research. Many studies used the Gibbs criteria or similar, and this is known to have relatively low sensitivity. It was raised that these criteria do not reflect the complexities of diagnosing chorioamnionitis in clinical practice where clinicians tend to look for more subtle signs to enable earlier diagnosis and treatment. These differences in definition may change the association between chorioamnionitis and neonatal infection. However, the committee was confident that its clinical experience supported the findings that chorioamnionitis is a risk factor for neonatal infection. They therefore agreed that these studies should remain applicable to the research question and that chorioamnionitis should remain part of the maternal risk factors table.

The studies which examined individual risk factors examined a wider range of populations, including babies born before 34 weeks’ gestational age. As a result, the risk factor tables and their accompanying recommendations can be used for babies of any gestational age, including the population that are not covered by the neonatal sepsis calculator.

4.1.3. Benefits and harms

4.1.3.1. Clinical prediction models

The main concern of the committee in relation to any prognostic model or management tool for neonatal infection was the trade-off between the potential issues associated with over-treatment versus the risks from lack of treatment where a baby does have infection. Although there are harms associated with unnecessary antibiotic treatment, such as the potential for nephrotoxicity when a baby is given gentamicin, and increased length of stay in hospital, the committee decided that these were smaller than the risk of a baby not receiving treatment when they do have infection. It is therefore important that any clinical prediction model or framework can maximise the number of babies that are correctly identified as needing treatment, while minimising the number who are given antibiotics unnecessarily.

Most of the data on clinical prediction models was for the Kaiser Permanente neonatal sepsis calculator. The committee agreed that overall pooled results from 9 studies showed that the tool had good specificity, but that there was substantial uncertainty about the sensitivity, with wide confidence intervals in the results. The positive likelihood ratio was above the clinical decision threshold, suggesting that a positive test result from the neonatal sepsis calculator indicated a large to very large increase in the probability of a baby having infection. However, the negative likelihood ratio was on the clinical decision threshold, and so the committee thought that more research was needed before the calculator could be considered as the sole option for predicting a baby’s risk of infection. One UK study compared the predictive accuracy of the Kaiser Permanente neonatal sepsis calculator with the recommendations in the 2012 guideline on neonatal infection, and the committee placed particular weight on this study when making recommendations. This study showed that both tools had similar sensitivity, but that the specificity of the neonatal sepsis calculator was higher, suggesting that using the neonatal sepsis calculator may result in fewer babies being treated with antibiotics who do not have infection. However, data on sensitivity were very uncertain because of the low number of cases of culture confirmed early-onset neonatal infection. This was particularly the case for the framework outlined in the 2012 NICE guideline which had data from only a single study with just 6 cases of confirmed neonatal infection. Given the uncertainties in the evidence, the committee decided that a recommendation for a framework based on individual factors was appropriate, with an option for obstetric or paediatric centres to consider the Kaiser Permanente neonatal sepsis calculator as part of an audit which assesses and manages the risk of neonatal infection. The committee thought that using a framework based on individual risk factors (described in more detail in the section below) was likely to be a conservative approach which would result in more antibiotics being prescribed that the neonatal sepsis calculator, but might also identify more true cases of infection. However, as the evidence did not show one option to be clearly better than the other, and as the neonatal sepsis calculator is already used in some centres in the UK, the committee decided that the recommendation to use the calculator in the context of an audit was appropriate, as long as this was for babies with a gestational age of 34 weeks and above. In situations where the Kaiser Permanente neonatal sepsis calculator is used, the committee decided that clinicians should use the recommendations within the tool to decide whether to treat with antibiotics or monitor further, as the evidence included as part of this review is based on these categorisations.

An additional aspect of the recommendation for use of the Kaiser Permanente calculator is that is should only be used for babies who are being cared for in a neonatal unit (neonatal intensive care units, local neonatal units and special care units), transitional care or postnatal ward. The committee did not think the calculator should be recommended for use in the emergency department or other settings, as babies who are brought in from home are likely to already be showing signs of being unwell and therefore need more immediate treatment than babies who are being assessed for risk of infection in a neonatal unit. In these cases, waiting to consult the calculator could instead delay treatment.

Given the limited evidence on clinical prediction models in the UK (including the Kaiser Permanente neonatal sepsis calculator as well as the framework set out by NICE), the committee decided that a research recommendation was needed for the use of prognostic tools for early-onset neonatal infection specifically in the UK. This recommendation was for any prognostic model, meaning that other models can be designed and evaluated, which may be particularly important for specific populations, such as babies born before 34 weeks gestational age (Appendix K).

4.1.3.2. Individual risk factors and clinical indicators

The committee decided that current evidence was not sufficient to make recommendations based only on clinical prediction models, so they also reviewed the evidence on individual risk factors and signs and symptoms of early-onset neonatal infection. The committee agreed that the structure of the recommendations outlined in the 2012 version of this guideline was still appropriate, with tables of risk factors and clinical indicators, some of which were designated as ‘red flag’ indicators. Red flag indicators were selected based on committee experience and are those thought to be the most high risk factors that require immediate treatment. Non-red flag indicators are those that can have causes other than neonatal infection and therefore do not always signal the need for immediate treatment. They decided that the recommendations in the 2012 version of this guideline on when to start antibiotic treatment or carry out further monitoring based on the number of indicators and red flag indicators met were still appropriate, and so made recommendations that were very similar to those made previously.

The committee thought it was important to retain the separate tables for risk factors and for clinical indicators that were used in the 2012 version of the guideline. The risk factors list (Box 1) gives an indication of the factors that a clinician should be aware of before the birth and the clinical indicators list (Box 2) highlights the signs and symptoms to look for in the baby after birth. Separating risk factors and clinical indicators should make it clearer for clinicians when they are trying to make important decisions about whether a baby should be treated for neonatal infection. Although the committee decided to keep the format of the recommendations the same as the previous version of the guideline, they made some changes to the risk factors and clinical indicators based on the updated evidence and their knowledge and experience. These changes are outlined in the sections below.

4.1.3.3. Maternal risk factors

When presented with the evidence for maternal risk factors, the committee decided to use a modified version of the risk factors table used in the 2012 version of the guideline (Box 1), as the main factors identified as risks for neonatal infection in the evidence review (intrapartum fever and chorioamnionitis) were already included in the table. Intrapartum fever and chorioamnionitis were a single risk factor in the 2012 guideline but, based on their clinical experience, the committee decided to separate these into two risk factors, as fever can also indicate other bacterial infections that are not chorioamnionitis. However, it specified that fever should only be considered a risk factor when there is suspected bacterial infection. This will avoid women who have a high temperature for other reasons, such as the side-effects of an epidural, receiving antibiotics unnecessarily. The committee also discussed how a woman can have chorioamnionitis without having fever, and this should be considered a risk factor for infection. There was discussion about whether this separation of fever and chorioamnionitis into separate risk factors would result in an increase in the number of women prescribed antibiotics. However, as women with both chorioamnionitis and fever would generally be given antibiotics in practice, the committee did not think this would result have a big impact on antibiotic prescription or resistance. The committee also updated the terminology for chorioamnionitis from suspected or confirmed chorioamnionitis to clinical chorioamnionitis, as a diagnosis during the intrapartum period is usually based on clinical signs rather than a histological diagnosis.

The committee decided to remove parenteral antibiotic treatment from the list of risk factors. This decision was made based on changes in obstetric practice since the previous guideline update, meaning that the threshold for diagnosing a mother with septicaemia is now lower. This means that many babies are now receiving antibiotics based on this red flag risk factor alone. This is considered to be an issue for the overprescribing of antibiotics.

The committee combined the risk factors ‘Invasive group B streptococcal infection in a previous baby’ and ‘maternal group B streptococcal colonisation, bacteriuria or infection in the current pregnancy’ into a single item in the list because they considered that these items relate to the same risk factor. Having a previous baby with invasive group B streptococcal infection does not increase the risk of neonatal infection further if a mother is known to have group B streptococcal colonisation, bacteriuria or infection in her current pregnancy, and so this should not be considered an additional risk factor.

The committee revised the table to make the sections on rupture of membranes at term and preterm clearer, and to reflect current NICE guidance in this area. Confirmed pre-labour rupture of membranes at term for more than 24 hours was included to correspond with the recommendations in the intrapartum care guideline (NICE clinical guideline 190, recommendation 1.11.6). However, the committee decided that the time period for confirmed rupture of membranes for more than 18 hours in a preterm birth should be retained (as in the 2012 version of the guideline) because there was no new evidence available in relation to this risk factor. Confirmed prelabour rupture of membranes was removed from the table because the committee felt that it is now covered by other risk factors in the table (preterm birth and confirmed rupture of membranes in a preterm or term birth). Consequently, they agreed that babies born to these mothers would still receive treatment when using the updated version of the NICE guidelines.

One other factor (single compared to multiple births) was identified in the evidence review. However, the committee did not think that it would be useful to highlight single births as a risk factor for infection, as this could result in most babies being identified as at higher risk of infection. It decided that there were other factors in the review that were more important to identify as potential risks for infection.

4.1.3.4. Neonatal risk factors and clinical indicators

The committee also reviewed the risk factors and clinical indicators in the baby for early-onset infection. They decided to recommend a modified version of the risk factors (Box 1) and clinical indicators (Box 2) tables used in the 2012 version of the guideline as a starting point to modify based on new evidence as appropriate.

One of the most common risk factors identified in the evidence review was low gestational age. The committee discussed whether this should be an additional risk factor but agreed that this is already covered by the preterm birth risk factor. It was highlighted that while there are a number of risk factors in the baby, many of these are as a result of low gestational age. The evidence suggested that babies born at less than 32 weeks’ gestational age are at greater risk of developing infection than those born at 32 weeks or greater. The committee discussed whether this should be added as additional information to the preterm birth factor, but it was deemed unnecessary as it would still count as one risk factor in the table. It was highlighted that babies who are slightly preterm and on a postnatal ward might be more at risk of an infection being missed because they might not be considered a high-risk group. The inclusion of babies born before 37 weeks’ gestation as a risk factor should therefore highlight to clinicians that these babies should be monitored for other signs of infection.

There was very little evidence on clinical indicators for early-onset infection and so the committee decided to base their recommendations on the table of clinical indicators in the 2012 version of the NICE guideline on neonatal infection. They made changes to this table based on their knowledge and experience and to make the table applicable to current practice (Box 2).

The committee decided to remove respiratory distress starting 4 hours after birth from the table of risk factors because they agreed that they would not want clinicians to wait 4 hours for treatment on the basis of this recommendation. Instead, they chose to retain ‘signs of respiratory distress’ as a factor in the table as this would include the group of babies who have symptoms beyond 4 hours after birth and should ensure that any babies who have infection will still receive the necessary treatment. The 2012 recommendations also had two recommendations for need for mechanical ventilation; one for preterm babies which was not a red flag risk factor and one for term babies which was a red flag. The committee agreed that mechanical ventilation is a risk factor for infection regardless of prematurity and so they decided to merge these into one recommendation which did not refer to whether a baby was born pre-term or at term for simplicity. They decided that this should be a red flag indicator.

For signs and symptoms in the baby, the committee decided to remove oliguria and local signs of infection from the recommendations table. Oliguria persisting beyond 24 hours after birth was removed because there is no clear definition of this risk factor, and the 24-hour time point is beyond the time when babies typically present with early-onset infection. The committee also considered oliguria to be a poor indicator of early-onset sepsis. The committee felt that local signs of infection should be removed because such infections are very common in newborn babies and including ‘local infection’ in the table may result in overprescribing antibiotics. Many local infections also require different management pathways to sepsis, such as oral or topical antibiotics.

4.1.3.5. Management of babies at increased risk of infection

The committee made recommendations based on their knowledge and experience that were consistent with the recommendations from the 2012 version of this guideline and current best practice. These recommendations were designed to direct clinicians to other, evidence-based, sections of the guideline where they could receive guidance when deciding whether a baby should be given antibiotic treatment, what antibiotics should be given, or what to do when discharging the baby if there are no further concerns.

The committee discussed how the recommendations for the use of the NICE risk factors and clinical indicators, or the neonatal sepsis calculator may result in some babies beginning to receive treatment when they do not have an infection. However, this risk is mitigated by clear recommendations in other sections of the guideline (recommendations 1.9.1–1.9.4) on when clinicians should stop giving antibiotics. This will minimise the time that a baby receives treatment if a negative blood culture result is returned. Recommendations on duration of antibiotics for early-onset infection were from the previous version of the guideline, and were thought to be appropriate in ensuring that babies receive necessary and timely treatment as well as making sure that treatment is stopped as soon as it is safe to do so. As such, the committee did not think that these needed to be changed.

4.1.4. Cost effectiveness and resource use

The committee were mindful that, as well as having potentially catastrophic consequences for the neonate, any infection that is missed can generate very substantial costs for the health and care system. They noted the clinical evidence, including one UK-based study suggesting that the neonatal sepsis calculator results in a similar number of false negatives as the NICE guidelines. However, the same study suggests that the calculator would lead to a substantial reduction in false-positive diagnoses. This could be important in reducing the number of babies who receive unnecessary treatment for infection, which in turn results in decreasing hospital stays. Thus, the committee made a ‘consider’ recommendation to use either one of the tools to predict newborn babies’ risk of sepsis, which is unlikely to be associated with increased NHS resource use.

4.1.5. Other factors the committee took into account

The committee discussed the recommendations from the previous guideline and agreed that the guidance to perform an immediate clinical assessment, review the maternal and neonatal history and carry out a physical examination was important. This will ensure that a clinician has all the information they need to assess the baby’s risk of infection and decide whether blood tests and treatment are needed. The committee did not think that this advice had changed since the previous update of the guideline and so this was included as part of the recommendations.

The committee highlighted the changes in obstetric practice since the previous update of the neonatal infection guideline in 2012. The awareness of maternal sepsis has changed, and fear of missing a diagnosis of sepsis has led to more women now being given antibiotics during labour. Currently, use of parenteral antibiotics during, before or after labour for confirmed or suspected bacterial infection (such as septicaemia) is one of the red flag indicators for babies being at high risk of early-onset neonatal infection. The change in perception of maternal sepsis (based on SEPSIS 6) has therefore led to a rise in the number of babies being given potentially unnecessary antibiotic treatment.

The committee also discussed the results of two additional studies that compared the suggested management of the NICE risk factors tables and the Kaiser Permanente neonatal sepsis calculator but did not meet the inclusion criteria for the review (Appendix D3). One (Pettinger 2019) was a systematic review and meta-analysis paper which retrospectively compared the management of the two tools, and the other (Morris 2020), was a cohort study which retrospectively compared the suggested management of the two tools at sites in England and Wales. The study designs did not meet the inclusion criteria for the review, because the sensitivity and specificity of the neonatal sepsis calculator were not assessed, and the Morris study only included babies with culture-confirmed infection. However, it was decided that it was important to consider the results given the direct comparison between the two tools included in the recommendations for this review. The committee felt confident that the Pettinger results reflected issues that had already been considered within the review and did not think that the findings should change any of the recommendations. The Morris paper highlighted instances for a number of babies with culture-confirmed infection where, at 4 hours, antibiotic treatment was recommended by the NICE recommendations but not by the Kaiser neonatal sepsis calculator. The lower sensitivity of the Kaiser neonatal sepsis calculator supported the results of the review, and as this was based specifically in the UK, the committee decided that the findings were therefore relevant to take into consideration when deciding on recommendations. This further supported its decision that the Kaiser Permanente neonatal sepsis calculator could be used, but only where a clinical audit was taking place to help better determine its effectiveness.

The committee also considered equality issues. It noted that risk factors for neonatal infection vary according to ethnicity and the age of the mother. It was also particularly aware of evidence that group B streptococcus (GBS) colonisation was higher in women of Black African family origin, as was the likelihood of having a baby who is preterm (Puthussery et al. 2019). They noted that the likelihood of having a baby who is preterm also increased with maternal age (Fuchs et al 2018). Having a preterm birth and GBS colonisation are included as risk factors in box 1 and so will be used to assess the risk of early-onset infection and determine antibiotic treatment.

4.2. Recommendations supported by this evidence review

This evidence review supports recommendations 1.3.1–1.3.9 and the research recommendations on clinical prediction models for early-onset infection and the risk of early-onset infection with maternal obesity.

4.3. References – included studies

4.3.1. Clinical prediction models

  • Carola, David, Vasconcellos, Mansi, Sloane, Amy et al (2018) Utility of Early-Onset Sepsis Risk Calculator for Neonates Born to Mothers with Chorioamnionitis.. The Journal of pediatrics 195: 48–52e1 [PubMed: 29275925]
  • Dhudasia MB; Mukhopadhyay S; Puopolo KM (2018) Implementation of the Sepsis Risk Calculator at an Academic Birth Hospital.. Hospital pediatrics 8(5): 243–250 [PubMed: 29666161]
  • Goel N., Shrestha S., Smith R. et al (2020) Screening for early onset neonatal sepsis: NICE guidance-based practice versus projected application of the Kaiser Permanente sepsis risk calculator in the UK population. Archives of disease in childhood. Fetal and neonatal edition 105: 118–122 [PubMed: 31296696]
  • Hershkovich-Shporen, C., Ujirauli, N., Oren, S. et al (2019) Not all newborns born to mothers with clinical chorioamnionitis need to be treated. Journal of Maternal-Fetal and Neonatal Medicine [PubMed: 31409159]
  • Joshi NS, Gupta A, Allan JM et al (2019) Management of Chorioamnionitis-Exposed Infants in the Newborn Nursery Using a Clinical Examination-Based Approach.. Hospital pediatrics 9(4): 227–233 [PubMed: 30833294]
  • Money, N, Newman, J, Demissie, S et al (2017) Anti-microbial stewardship: antibiotic use in well-appearing term neonates born to mothers with chorioamnionitis.. Journal of perinatology : official journal of the California Perinatal Association 37(12): 1304–1309 [PubMed: 28981079]
  • Popowski, Thomas, Goffinet, Francois, Maillard, Francoise et al (2011) Maternal markers for detecting early-onset neonatal infection and chorioamnionitis in cases of premature rupture of membranes at or after 34 weeks of gestation: a two-center prospective study.. BMC pregnancy and childbirth 11: 26 [PMC free article: PMC3088535] [PubMed: 21470433]
  • Shakib, Julie, Buchi, Karen, Smith, Elizabeth et al (2015) Management of newborns born to mothers with chorioamnionitis: is it time for a kinder, gentler approach?.. Academic pediatrics 15(3): 340–4 [PubMed: 25906702]
  • Sloane A.J., Coleman C., Carola D.L. et al (2019) Use of a Modified Early-Onset Sepsis Risk Calculator for Neonates Exposed to Chorioamnionitis. Journal of Pediatrics [PubMed: 31208783]
  • Strunk T., Buchiboyina A., Sharp M. et al (2018) Implementation of the Neonatal Sepsis Calculator in an Australian Tertiary Perinatal Centre. Neonatology 113(4): 379–382 [PubMed: 29514161]

4.3.2. Maternal and neonatal risk factors

  • Dempsey, E, Chen, M-F, Kokottis, T et al (2005) Outcome of neonates less than 30 weeks gestation with histologic chorioamnionitis. American journal of perinatology 22(3): 155–9 [PubMed: 15838750]
  • Dior, Uri P, Kogan, Liron, Eventov-Friedman, Smadar et al (2016) Very High Intrapartum Fever in Term Pregnancies and Adverse Obstetric and Neonatal Outcomes. Neonatology 109(1): 62–8 [PubMed: 26536344]
  • Garcia-Munoz Rodrigo, Fermin; Galan Henriquez, Gloria M; Ospina, Cristina Gomez (2014) Morbidity and mortality among very-low-birth-weight infants born to mothers with clinical chorioamnionitis. Pediatrics and neonatology 55(5): 381–6 [PubMed: 24745649]
  • Garcia-Munoz Rodrigo, Fermin, Galan Henriquez, Gloria, Figueras Aloy, Josep et al (2014) Outcomes of very-low-birth-weight infants exposed to maternal clinical chorioamnionitis: a multicentre study. Neonatology 106(3): 229–34 [PubMed: 25011418]
  • Hakansson, S and Kallen, K (2006) Impact and risk factors for early-onset group B streptococcal morbidity: analysis of a national, population-based cohort in Sweden 1997–2001. BJOG : an international journal of obstetrics and gynaecology 113(12): 1452–8 [PubMed: 17083655]
  • Hakansson, Stellan and Kallen, Karin (2008) High maternal body mass index increases the risk of neonatal early onset group B streptococcal disease. Acta paediatrica (Oslo, Norway : 1992) 97(10): 1386–9 [PubMed: 18647277]
  • Klinger, Gil, Levy, Itzhak, Sirota, Lea et al (2009) Epidemiology and risk factors for early onset sepsis among very-low-birthweight infants. American journal of obstetrics and gynecology 201(1): 38e1–6 [PubMed: 19380122]
  • Mularoni, Alessandra, Madrid, Marisela, Azpeitia, Agueda et al (2014) The role of coagulase-negative staphylococci in early onset sepsis in a large European cohort of very low birth weight infants. The Pediatric infectious disease journal 33(5): e121–5 [PubMed: 24168984]
  • Ofman, Gaston; Vasco, Natalia; Cantey, Joseph B (2016) Risk of Early-Onset Sepsis following Preterm, Prolonged Rupture of Membranes with or without Chorioamnionitis. American journal of perinatology 33(4): 339–42 [PubMed: 26469992]
  • Ronnestad, Arild, Abrahamsen, Tore G, Medbo, Sverre et al (2005) Septicemia in the first week of life in a Norwegian national cohort of extremely premature infants. Pediatrics 115(3): e262–8 [PubMed: 15687417]
  • Soraisham, Amuchou S, Singhal, Nalini, McMillan, Douglas D et al (2009) A multicenter study on the clinical outcome of chorioamnionitis in preterm infants. American journal of obstetrics and gynecology 200(4): 372e1–6 [PubMed: 19217596]

4.3.2. Other citations

  • Fuchs, F., Monet, B., Ducruet, T., Chaillet, N. and Audibert, F., 2018. Effect of maternal age on the risk of preterm birth: A large cohort study. PloS one, 13(1), p.e0191002. [PMC free article: PMC5791955] [PubMed: 29385154]
  • Puthussery, S., Li, L., Tseng, P.C., Kilby, L., Kapadia, J., Puthusserry, T. and Thind, A., 2019. Ethnic variations in risk of preterm birth in an ethnically dense socially disadvantaged area in the UK: a retrospective cross-sectional study. BMJ open, 9(3), p.e023570. [PMC free article: PMC6429724] [PubMed: 30852531]

Appendices

Appendix A. Review protocols

Review protocols for review protocols on the accuracy and effectives of clinical prediction models (A.1, Part A – Prognostic accuracy studies and A.2, Part B – Test and treat RCTs), individual maternal risk factors (A.3) and individual neonatal risk factors (A.4) are all included in the appendices below.

A.1. Review protocol for clinical prediction models - part A (prognostic accuracy studies) (PDF, 286K)

A.2. Review protocol for clinical prediction models - part B (test and treat RCTs) (PDF, 291K)

A.3. Review protocol for maternal risk factors (PDF, 289K)

A.4. Review protocol for neonatal risk factors (PDF, 280K)

Appendix B. Literature search strategies

Literature search strategies for prognostic accuracy models and maternal and neonatal risk factors are all included in the appendices below.

B.1. Clinical search: Clinical prediction models

The search was conducted on 14th August 2019. Given the broad range of publication types included in the review protocol, no in-house publication type filters were used. The following databases were searched: Medline, Medline In Process, Medline E-pub Ahead of print, Embase, (all via the Ovid platform), Cochrane Database of Systematic Reviews, CENTRAL (all via the Wiley platform), and the DARE database (via the CRD platform).

Medline. Medline In Process, Medline E-pub
  1. exp Infant, Newborn/
  2. Term Birth/
  3. Infant Care/
  4. Perinatal Care/
  5. Intensive Care Units, Neonatal/
  6. Intensive Care, Neonatal/
  7. Infant Health/
  8. (newborn* or new born* or neonat* or neo-nat* or perinat* or peri-nat*).tw.
  9. ((premature or pre-mature* or preterm or pre-term) adj4 (child* or infant* or baby* or babies* or offspring)).tw.
  10. or/1-9
  11. exp Bacterial Infections/
  12. ((bacter* or strep* or staph* or GNB) adj4 (infect* or diseas* or contaminat* or mening* or pneumon* or nosocomial*)).tw.
  13. exp Sepsis/
  14. (sepsis or septic?emia* or py?emia* or pyho?emia*).tw.
  15. (septic* adj4 shock*).tw.
  16. or/11-15
  17. exp Streptococcus/
  18. exp Staphylococcus/
  19. (streptococc* or staphylococc*).tw.
  20. (GBS or MRSA or NRCS-A or MSSA).tw.
  21. (met?icillin-resistant adj3 aureus).tw.
  22. exp Escherichia coli/
  23. ((Escheric* or E) adj2 coli).tw.
  24. exp Listeria/
  25. listeria*.tw.
  26. exp Klebsiella/
  27. klebsiella*.tw.
  28. exp Pseudomonas/
  29. (pseudomonas or chryseomonas or flavimonas).tw.
  30. Enterobacteriaceae/
  31. (enterobact* or sodalis or paracolobactrum or ewingella or leclercia).tw.
  32. ((enteric or coliform) adj2 bac*).tw.
  33. exp Neisseria/
  34. neisseria*.tw.
  35. exp Haemophilus influenzae/
  36. ((h?emophil* or H or bacter* or bacill* or mycobacter* or coccobac*) adj2 (influenz* or pfeiffer* or meningitidis)).tw.
  37. exp Serratia/
  38. serratia*.tw.
  39. exp Cronobacter/
  40. (cronobact* or sakazaki* or malonatic*).tw.
  41. exp Acinetobacter/
  42. (acinetobact* or herellea* or mima or baumanni* or genomosp* or calcoacetic*).tw.
  43. exp Fusobacterium/
  44. (fusobact* or sphaerophor* or necrophorum or nucleatum).tw.
  45. exp Enterococcus/
  46. enterococc*.tw.
  47. or/17-46
  48. 16 or 47
  49. 10 and 48
  50. ((newborn* or new born* or neonat* or neo-nat* or perinat* or peri-nat*) adj4 infect*).tw.
  51. ((premature or pre-mature* or preterm or pre-term) adj4 (child* or infant* or baby* or babies* or offspring) adj4 infect*).tw.
  52. 50 or 51
  53. 49 or 52
  54. (bacter?emia* or bacill?emia*).tw.
  55. (blood* adj4 (infect* or contamin* or invas* or invad*)).tw.
  56. 54 or 55
  57. 10 and 56
  58. 53 or 57
  59. Risk Assessment/mt [Methods]
  60. ((risk* or predict* or probab* or prognos* or quantitativ*) adj2 (model* or tool* or algorithm* or rul*)).tw.
  61. (diagnos* adj2 (model* or algorithm*)).tw.
  62. ((sepsis* or septic* or Bayes* or EOS or LOS) adj4 calculator*).tw.
  63. (NEOSC or EOSCAL* or SRC).tw.
  64. (Kaiser adj2 Permanente).tw.
  65. (Kaiser adj10 calculator*).tw.
  66. ((sepsis or septic*) adj4 risk* adj4 scor*).tw.
  67. SRS.tw.
  68. ((sepsis* or septic*) adj4 (metascore* or meta-score*)).tw.
  69. Diagnosis, Computer-Assisted/
  70. Algorithms/
  71. ((computer* or vital signs* or math* or manag* or clinic* or medic* or stratif* or prevent* or therap*) adj4 algorithm*).tw.
  72. RALIS.tw.
  73. (computer* adj4 (analys* or template*)).tw.
  74. Decision Support Techniques/
  75. (decision* adj4 (aid* or analys* or support* or assist*)).tw.
  76. CDSS*.tw.
  77. or/59-76
  78. 58 and 77
  79. Animals/ not Humans/
  80. 78 not 79
  81. limit 80 to english language
Embase
  1. newborn/
  2. term birth/
  3. infant care/
  4. perinatal care/
  5. neonatal intensive care unit/
  6. newborn intensive care/
  7. child health/
  8. (newborn* or new born* or neonat* or neo-nat* or perinat* or peri-nat*).tw.
  9. ((premature or pre-mature* or preterm or pre-term) adj4 (child* or infant* or baby* or babies* or offspring)).tw.
  10. or/1-9
  11. exp bacterial infection/
  12. ((bacter* or strep* or staph* or GNB) adj4 (infect* or diseas* or contaminat* or mening* or pneumon* or nosocomial*)).tw.
  13. exp sepsis/
  14. (sepsis or septic?emia* or py?emia* or pyho?emia*).tw.
  15. (septic* adj4 shock*).tw.
  16. or/11-15
  17. exp Streptococcus/
  18. exp Staphylococcus/
  19. (streptococc* or staphylococc*).tw.
  20. (GBS or MRSA or NRCS-A or MSSA).tw.
  21. (met?icillin-resistant adj3 aureus).tw.
  22. exp Escherichia coli/
  23. ((Escheric* or E) adj2 coli).tw.
  24. exp Listeria/
  25. listeria*.tw.
  26. exp Klebsiella/
  27. klebsiella*.tw.
  28. exp Pseudomonas/
  29. (pseudomonas or chryseomonas or flavimonas).tw.
  30. Enterobacteriaceae/
  31. (enterobact* or sodalis or paracolobactrum or ewingella or leclercia).tw.
  32. ((enteric or coliform) adj2 bac*).tw.
  33. exp Neisseria/
  34. neisseria*.tw.
  35. exp Haemophilus influenzae/
  36. ((h?emophil* or H or bacter* or bacill* or mycobacter* or coccobac*) adj2 (influenz* or pfeiffer* or meningitidis)).tw.
  37. exp Serratia/
  38. serratia*.tw.
  39. exp cronobacter/
  40. (cronobact* or sakazaki* or malonatic*).tw.
  41. exp Acinetobacter/
  42. (acinetobact* or herellea* or mima or baumanni* or genomosp* or calcoacetic*).tw.
  43. exp Fusobacterium/
  44. (fusobact* or sphaerophor* or necrophorum or nucleatum).tw.
  45. exp Enterococcus/
  46. enterococc*.tw.
  47. or/17-46
  48. 16 or 47
  49. 10 and 48
  50. ((newborn* or new born* or neonat* or neo-nat* or perinat* or peri-nat*) adj4 infect*).tw.
  51. ((premature or pre-mature* or preterm or pre-term) adj4 (child* or infant* or baby* or babies* or offspring) adj4 infect*).tw.
  52. 50 or 51
  53. 49 or 52
  54. (bacter?emia* or bacill?emia*).tw.
  55. (blood* adj4 (infect* or contamin* or invas* or invad*)).tw.
  56. 54 or 55
  57. 10 and 56
  58. 53 or 57
  59. *risk assessment/
  60. ((risk* or predict* or probab* or prognos* or quantitativ*) adj2 (model* or tool* or algorithm* or rul*)).tw.
  61. (diagnos* adj2 (model* or algorithm*)).tw.
  62. ((sepsis* or septic* or Bayes* or EOS or LOS) adj4 calculator*).tw.
  63. (NEOSC or EOSCAL* or SRC).tw.
  64. (Kaiser adj2 Permanente).tw.
  65. (Kaiser adj10 calculator*).tw.
  66. ((sepsis or septic*) adj4 risk* adj4 scor*).tw.
  67. SRS.tw.
  68. ((sepsis* or septic*) adj4 (metascore* or meta-score*)).tw.
  69. computer assisted diagnosis/
  70. algorithm/
  71. ((computer* or vital signs* or math* or manag* or clinic* or medic* or stratif* or prevent* or therap*) adj4 algorithm*).tw.
  72. RALIS.tw.
  73. (computer* adj4 (analys* or template*)).tw.
  74. exp decision support system/
  75. (decision* adj4 (aid* or analys* or support* or assist*)).tw.
  76. CDSS*.tw.
  77. or/59-76
  78. 58 and 77
  79. nonhuman/ not human/
  80. 78 not 79
  81. limit 80 to english language
  82. limit 81 to (conference abstract or conference paper or “conference review”)
  83. 81 not 82
Cochrane Database of Systematic Reviews, CENTRAL
#1.

MeSH descriptor: [Infant, Newborn] explode all trees

#2.

MeSH descriptor: [Term Birth] this term only

#3.

MeSH descriptor: [Infant Care] this term only

#4.

MeSH descriptor: [Perinatal Care] this term only

#5.

MeSH descriptor: [Intensive Care Units, Neonatal] this term only

#6.

MeSH descriptor: [Intensive Care, Neonatal] this term only

#7.

MeSH descriptor: [Infant Health] this term only

#8.

((newborn* or new born* or new-born or neonat* or neo-nat* or perinat* or peri-nat*)):ti,ab,kw

#9.

((premature or pre-mature* or preterm or pre-term) near/4 (child* or infant* or baby* or babies* or offspring)):ti,ab,kw

#10.

{or #1-#9}

#11.

MeSH descriptor: [Bacterial Infections] explode all trees

#12.

((bacter* or strep* or staph* or GNB) near/4 (infect* or diseas* or contaminat* or mening* or pneumon* or nosocomial*)):ti,ab,kw

#13.

MeSH descriptor: [Sepsis] explode all trees

#14.

((sepsis or septic?emia* or py?emia* or pyho?emia*)):ti,ab,kw

#15.

((septic* near/4 shock*)):ti,ab,kw

#16.

{or #11-#15}

#17.

MeSH descriptor: [Streptococcus] explode all trees

#18.

MeSH descriptor: [Staphylococcus] explode all trees

#19.

((streptococc* or staphylococc*)):ti,ab,kw

#20.

((GBS or MRSA or NRCS-A or MSSA)):ti,ab,kw

#21.

((met?icillin-resistant near/3 aureus)):ti,ab,kw

#22.

MeSH descriptor: [Escherichia coli] explode all trees

#23.

((Escheric* or E) near/2 (coli)):ti,ab,kw

#24.

MeSH descriptor: [Listeria] explode all trees

#25.

(Listeria*):ti,ab,kw

#26.

MeSH descriptor: [Klebsiella] explode all trees

#27.

(klebsiella*):ti,ab,kw

#28.

MeSH descriptor: [Pseudomonas] explode all trees

#29.

((pseudomonas or chryseomonas or flavimonas)):ti,ab,kw

#30.

MeSH descriptor: [Enterobacteriaceae] this term only

#31.

((enterobact* or sodalis or paracolobactrum or ewingella or leclercia)):ti,ab,kw

#32.

((enteric or coliform) near/2 (bac*)):ti,ab,kw

#33.

MeSH descriptor: [Neisseria] explode all trees

#34.

(neisseria*):ti,ab,kw

#35.

MeSH descriptor: [Haemophilus influenzae] explode all trees

#36.

((h?emophil* or H or bacter* or bacill* or mycobacter* or coccobac*) near/2 (influenz* or pfeiffer* or meningitidis)):ti,ab,kw

#37.

MeSH descriptor: [Serratia] explode all trees

#38.

(serratia*):ti,ab,kw

#39.

MeSH descriptor: [Cronobacter] explode all trees

#40.

((cronobact* or sakazaki* or malonatic*)):ti,ab,kw

#41.

MeSH descriptor: [Acinetobacter] explode all trees

#42.

((acinetobact* or herellea* or mima or baumanni* or genomosp* or calcoacetic*)):ti,ab,kw

#43.

MeSH descriptor: [Fusobacterium] explode all trees

#44.

((fusobact* or sphaerophor* or necrophorum or nucleatum)):ti,ab,kw

#45.

MeSH descriptor: [Enterococcus] explode all trees

#46.

(enterococc*):ti,ab,kw

#47.

{or #17-#46}

#48.

#16 or #47

#49.

#10 and #48

#50.

((newborn* or new born* or new-born* or neonat* or neo-nat* or perinat* or peri-nat*) near/4 (infect*)):ti,ab,kw

#51.

((premature or pre-mature* or preterm or pre-term) near/4 (child* or infant* or baby* or babies* or offspring) near/4 (infect*)):ti,ab,kw

#52.

#50 or #51

#53.

#49 or #52

#54.

((bacter?emia* or bacill?emia*)):ti,ab,kw

#55.

((blood*) near/4 (infect* or contamin* or invas* or invad*)):ti,ab,kw

#56.

#54 or #55

#57.

#10 and #56

#58.

#53 or #57

#59.

MeSH descriptor: [Risk Assessment] this term only and with qualifier(s): [methods - MT]

#60.

((risk* or predict* or probab* or prognos* or quantitativ*) near/2 (model* or tool* or algorithm* or rul*)):ti,ab,kw

#61.

((diagnos*) near/2 (model* or algorithm*)):ti,ab,kw

#62.

((sepsis* or septic* or Bayes* or EOS or LOS) near/4 (calculator*)):ti,ab,kw

#63.

((NEOSC or EOSCAL* or SRC)):ti,ab,kw

#64.

((Kaiser) near/2 (Permanente)):ti,ab,kw

#65.

((Kaiser) near/10 (calculator*)):ti,ab,kw

#66.

((sepsis or septic*) near/4 (risk*) near/4 (scor*)):ti,ab,kw

#67.

(SRS):ti,ab,kw

#68.

((sepsis* or septic*) near/4 (metascore* or meta-score*)):ti,ab,kw

#69.

MeSH descriptor: [Diagnosis, Computer-Assisted] this term only

#70.

MeSH descriptor: [Algorithms] this term only

#71.

((computer* or vital signs* or math* or manag* or clinic* or medic* or stratif* or prevent* or therap*) near/4 (algorithm*)):ti,ab,kw

#72.

(RALIS):ti,ab,kw

#73.

((computer*) near/4 (analys* or template*)):ti,ab,kw

#74.

MeSH descriptor: [Decision Support Techniques] this term only

#75.

((decision*) near/4 (aid* or analys* or support* or assist*)):ti,ab,kw

#76.

(CDSS*):ti,ab,kw

#77.

{or #59-#76}

#78.

#58 and #77

#79.

(conference):pt

#80.

((clinicaltrials or trialsearch)):so

#81.

#79 or #80

#82.

#78 not #81

DARE
  1. MeSH DESCRIPTOR Infant, Newborn EXPLODE ALL TREES
  2. MeSH DESCRIPTOR Term Birth
  3. MeSH DESCRIPTOR Infant Care
  4. MeSH DESCRIPTOR Perinatal Care
  5. MeSH DESCRIPTOR Intensive Care Units, Neonatal
  6. MeSH DESCRIPTOR Intensive Care, Neonatal
  7. MeSH DESCRIPTOR Infant Health
  8. (((newborn* or new born* or neonat* or neo-nat* or perinat* or peri-nat*)))
  9. (((premature or pre-mature* or preterm or pre-term) NEAR4 (child* or infant* or baby* or babies* or offspring)))
  10. #1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9
  11. MeSH DESCRIPTOR Bacterial Infections EXPLODE ALL TREE
  12. (((bacter* or strep* or staph* or GNB) NEAR4 (infect* or diseas* or contaminat* or mening* or pneumon* or nosocomial*)))
  13. MeSH DESCRIPTOR Sepsis EXPLODE ALL TREES
  14. (((sepsis or septic?emia* or py?emia* or pyho?emia*)))
  15. (((septic* NEAR4 shock*)))
  16. #11 OR #12 OR #13 OR #14 OR #15
  17. MeSH DESCRIPTOR Streptococcus EXPLODE ALL TREES
  18. MeSH DESCRIPTOR Staphylococcus EXPLODE ALL TREES
  19. (((streptococc* or staphylococc*)))
  20. (((GBS or MRSA or NRCS-A or MSSA)))
  21. (((met?icillin-resistant NEAR3 aureus)))
  22. MeSH DESCRIPTOR Escherichia coli EXPLODE ALL TREES
  23. (((Escheric* or E) NEAR2 (coli)))
  24. MeSH DESCRIPTOR Listeria EXPLODE ALL TREES
  25. ((listeria*))
  26. MeSH DESCRIPTOR Klebsiella EXPLODE ALL TREES
  27. ((klebsiella*))
  28. MeSH DESCRIPTOR Pseudomonas EXPLODE ALL TREES
  29. ((pseudomonas or chryseomonas or flavimonas))
  30. MeSH DESCRIPTOR Enterobacteriaceae EXPLODE ALL TREES
  31. (enterobact* or sodalis or paracolobactrum or ewingella or leclercia)
  32. ((enteric or coliform) NEAR2 (bac*))
  33. MeSH DESCRIPTOR Neisseria EXPLODE ALL TREES
  34. (neisseria*)
  35. MeSH DESCRIPTOR Haemophilus influenzae EXPLODE ALL TREES
  36. ((h?emophil* or H or bacter* or bacill* or mycobacter* or coccobac*) NEAR2 (influenz* or pfeiffer* or meningitidis))
  37. MeSH DESCRIPTOR Serratia EXPLODE ALL TREES
  38. (serratia*)
  39. MeSH DESCRIPTOR Cronobacter EXPLODE ALL TREES
  40. (cronobact* or sakazaki* or malonatic*)
  41. MeSH DESCRIPTOR Acinetobacter EXPLODE ALL TREES
  42. ((acinetobact* or herellea* or mima or baumanni* or genomosp* or calcoacetic*))
  43. MeSH DESCRIPTOR Fusobacterium EXPLODE ALL TREES
  44. ((fusobact* or sphaerophor* or necrophorum or nucleatum))
  45. MeSH DESCRIPTOR Enterococcus EXPLODE ALL TREES
  46. (enterococc*)
  47. #17 OR #18 OR #19 OR #20 OR #21 OR #22 OR #23 OR #24 OR #25 OR #26 OR #27 OR #28 OR #29 OR #30 OR #31 OR #32 OR #33 OR #34 OR #35 OR #36 OR #37 OR #38 OR #39 OR #40 OR #41 OR #42 OR #43 OR #44 OR #45 OR #46
  48. #16 OR #47
  49. #10 AND #48
  50. ((newborn* or new born* or neonat* or neo-nat* or perinat* or peri-nat*) NEAR4 (infect*))
  51. ((premature or pre-mature* or preterm or pre-term) NEAR4 (child* or infant* or baby* or babies* or offspring) NEAR4 (infect*))
  52. #50 OR #51
  53. #49 OR #52
  54. ((bacter?emia* or bacill?emia*))
  55. ((blood*) NEAR4 (infect* or contamin* or invas* or invad*))
  56. #54 OR #55
  57. #10 AND #56
  58. #53 OR #57
  59. MeSH DESCRIPTOR Risk Assessment WITH QUALIFIER MT
  60. ((risk* or predict* or probab* or prognos* or quantitativ*) NEAR2 (model* or tool* or algorithm* or rul*))
  61. ((diagnos*) NEAR2 (model* or algorithm*))
  62. ((sepsis* or septic* or Bayes* or EOS or LOS) NEAR4 (calculator*))
  63. (NEOSC or EOSCAL* or SRC)
  64. ((Kaiser) NEAR2 (Permanente))
  65. ((Kaiser) NEAR10 (calculator*))
  66. ((sepsis or septic*) NEAR4 (risk*) NEAR4 (scor*))
  67. (SRS)
  68. ((sepsis* or septic*) NEAR4 (metascore* or meta-score*))
  69. MeSH DESCRIPTOR Diagnosis, Computer-Assisted
  70. MeSH DESCRIPTOR Algorithms
  71. ((computer* or vital signs* or math* or manag* or clinic* or medic* or stratif* or prevent* or therap*) NEAR4 (algorithm*))
  72. (RALIS)
  73. ((computer*) NEAR4 (analys* or template*))
  74. MeSH DESCRIPTOR Decision Support Techniques
  75. ((decision*) NEAR4 (aid* or analys* or support* or assist*))
  76. (CDSS)
  77. #59 OR #60 OR #61 OR #62 OR #63 OR #64 OR #65 OR #66 OR #67 OR #68 OR #69 OR #70 OR #71 OR #72 OR #73 OR #74 OR #75 OR #76
  78. #58 AND #77
  79. * IN DARE
  80. #78 AND #79

B.2. Clinical search: Maternal and neonatal risk factors

The search was conducted on 28th November 2019. A single search strategy was developed for review questions 1.1 and 1.2. The following databases were searched: Medline, Medline In Process, Medline E-pub Ahead of print, Embase, (all via the Ovid platform), Cochrane Database of Systematic Reviews, (via the Wiley platform), and the DARE database (via the CRD platform).

The search focused on unique risk factors for review questions 1.1 and 1.2 not previously considered for review questions 5.1 and 5.2.

Population and risk factor terms

The search terms used to identify information on population and risk factors are reproduced below for all databases. The population and risk factor terms were combined as ‘And’ to identify papers that discussed both.

Medline, Medline in Process & Medline E-pub Ahead of Print
  1. exp Infant, Newborn/
  2. Term Birth/
  3. Infant Care/
  4. Perinatal Care/
  5. Intensive Care Units, Neonatal/
  6. Intensive Care, Neonatal/
  7. Infant Health/
  8. (newborn* or new born* or neonat* or neo-nat* or perinat* or peri-nat*).tw.
  9. ((premature* or pre-mature* or preterm* or pre-term*) adj4 (child* or infant* or baby* or babies* or offspring)).tw.
  10. or/1-9
  11. exp Bacterial Infections/
  12. ((bacter* or strep* or staph* or GNB) adj4 (infect* or diseas* or contaminat* or mening* or pneumon* or nosocomial*)).tw.
  13. exp Sepsis/
  14. (sepsis or septic?emia* or py?emia* or pyho?emia*).tw.
  15. (septic* adj4 shock*).tw. (
  16. (bacter?emia* or bacill?emia*).tw.
  17. (blood* adj4 (infect* or contamin* or invas* or invad*)).tw.
  18. or/11-17
  19. exp Streptococcus/
  20. exp Staphylococcus/
  21. (streptococc* or staphylococc*).tw.
  22. (GBS or MRSA or NRCS-A or MSSA).tw.
  23. (met?icillin-resistant adj3 aureus).tw.
  24. exp Escherichia coli/
  25. ((Escheric* or E) adj2 coli).tw.
  26. exp Listeria/
  27. listeria*.tw.
  28. exp Klebsiella/
  29. klebsiella*.tw.
  30. exp Pseudomonas/
  31. (pseudomonas or chryseomonas or flavimonas).tw.
  32. Enterobacteriaceae/
  33. (enterobact* or sodalis or paracolobactrum or ewingella or leclercia).tw.
  34. ((enteric or coliform) adj2 bac*).tw.
  35. exp Neisseria/
  36. neisseria*.tw.
  37. exp Haemophilus influenzae/
  38. ((h?emophil* or H or bacter* or bacill* or mycobacter* or coccobac*) adj2 (influenz* or pfeiffer* or meningitidis)).tw.
  39. exp Serratia/
  40. serratia*.tw.
  41. exp Cronobacter/
  42. (cronobact* or sakazaki* or malonatic*).tw.
  43. exp Acinetobacter/
  44. (acinetobact* or herellea* or mima or baumanni* or genomosp* or calcoacetic*).tw.
  45. exp Fusobacterium/
  46. (fusobact* or sphaerophor* or necrophorum or nucleatum).tw.
  47. exp Enterococcus/
  48. enterococc*.tw.
  49. or/19-48
  50. 18 or 49
  51. 10 and 50
  52. ((newborn* or new born* or neonat* or neo-nat* or perinat* or peri-nat*) adj4 infect*).tw.
  53. ((premature* or pre-mature* or preterm* or pre-term*) adj4 (child* or infant* or baby* or babies* or offspring) adj4 infect*).tw.
  54. 52 or 53
  55. 51 or 54
  56. exp Fetal Membranes, Premature Rupture/
  57. ((preterm* or pre-term* or premature* or pre-mature* or prelabor* or pre-labor* or pre) adj4 (ruptur* or membrane* or disrupt* or erupt* or sever or severed or tear* or breach*)).tw.
  58. (prom or proms or pprom*).tw.
  59. Gestational Age/
  60. ((gestat* or f?etal* or f?etus*) adj4 (age* or aging* or matur*)).tw.
  61. Fever/di, dg [Diagnosis, Diagnostic Imaging]
  62. ((intrapartum* or intra-partum* or labo?r* or deliver* or childbirth* or child-birth* or congenit* or conatal*) adj4 (fever* or deliriu* or pyrexia* or hyperthermia*)).tw.
  63. Chorioamnionitis/
  64. (chorioamnionit* or amnioniti* or funisiti*).tw.
  65. (parenteral* adj4 (antibacter* or anti-bacter* or antibiotic* or anti-biotic* or antimycobact* or anti-mycobact* or bacteriocid*)).tw.
  66. Antibiotic Prophylaxis/
  67. ((antibacter* or anti-bacter* or antibiotic* or anti-biotic* or antimycobact* or antimycobact* or bacteriocid*) adj4 (prophyla* or premedic* or pre-medic* or prevent*)).tw.
  68. Brain Diseases/
  69. ((brain* or intracranial* or intra-cranial* or encephalon*) adj4 (diseas* or disorder* or defect* or illness* or inflam* or syndrom*)).tw.
  70. encephalopath*.tw.
  71. Persistent Fetal Circulation Syndrome/
  72. ((persist* or misalign* or mis-align*) adj4 (f?etal* or f?etus* or pulmonar*) adj4 (circulat* or hypertens* or vein*)).tw.
  73. (PPHN or PFC or ACD MPV or ACDMPV).tw.
  74. or/56-73
  75. 55 and 74
  76. Animals/ not Humans/
  77. 75 not 76
  78. limit 77 to english language
Embase
  1. newborn/
  2. term birth/
  3. infant care/
  4. perinatal care/
  5. neonatal intensive care unit/
  6. newborn intensive care/
  7. child health/
  8. (newborn* or new born* or neonat* or neo-nat* or perinat* or peri-nat*).tw.
  9. ((premature* or pre-mature* or preterm* or pre-term*) adj4 (child* or infant* or baby* or babies* or offspring)).tw.
  10. or/1-9
  11. exp bacterial infection/
  12. ((bacter* or strep* or staph* or GNB) adj4 (infect* or diseas* or contaminat* or mening* or pneumon* or nosocomial*)).tw.
  13. exp sepsis/
  14. (sepsis or septic?emia* or py?emia* or pyho?emia*).tw.
  15. (septic* adj4 shock*).tw.
  16. (bacter?emia* or bacill?emia*).tw.
  17. (blood* adj4 (infect* or contamin* or invas* or invad*)).tw.
  18. or/11-17
  19. exp Streptococcus/
  20. exp Staphylococcus/
  21. (streptococc* or staphylococc*).tw.
  22. (GBS or MRSA or NRCS-A or MSSA).tw.
  23. (met?icillin-resistant adj3 aureus).tw.
  24. exp Escherichia coli/
  25. ((Escheric* or E) adj2 coli).tw.
  26. exp Listeria/
  27. listeria*.tw.
  28. exp Klebsiella/
  29. klebsiella*.tw.
  30. exp Pseudomonas/
  31. (pseudomonas or chryseomonas or flavimonas).tw.
  32. Enterobacteriaceae/
  33. (enterobact* or sodalis or paracolobactrum or ewingella or leclercia).tw.
  34. ((enteric or coliform) adj2 bac*).tw.
  35. exp Neisseria/
  36. neisseria*.tw.
  37. exp Haemophilus influenzae/
  38. ((h?emophil* or H or bacter* or bacill* or mycobacter* or coccobac*) adj2 (influenz* or pfeiffer* or meningitidis)).tw.
  39. exp Serratia/
  40. serratia*.tw.
  41. exp cronobacter/
  42. (cronobact* or sakazaki* or malonatic*).tw.
  43. exp Acinetobacter/
  44. (acinetobact* or herellea* or mima or baumanni* or genomosp* or calcoacetic*).tw.
  45. exp Fusobacterium/
  46. (fusobact* or sphaerophor* or necrophorum or nucleatum).tw.
  47. exp Enterococcus/
  48. enterococc*.tw.
  49. or/19-48
  50. 18 or 49
  51. 10 and 50
  52. ((newborn* or new born* or neonat* or neo-nat* or perinat* or peri-nat*) adj4 infect*).tw.
  53. ((premature* or pre-mature* or preterm* or pre-term*) adj4 (child* or infant* or baby* or babies* or offspring) adj4 infect*).tw.
  54. 52 or 53
  55. 51 or 54
  56. exp premature fetus membrane rupture/
  57. ((preterm* or pre-term* or premature* or pre-mature* or prelabor* or pre-labor* or pre) adj4 (ruptur* or membrane* or disrupt* or erupt* or sever or severed or tear* or breach*)).tw.
  58. (prom or proms or pprom*).tw.
  59. gestational age/
  60. ((gestat* or f?etal* or f?etus*) adj4 (age* or aging* or matur*)).tw.
  61. fever/di [Diagnosis]
  62. ((intrapartum* or intra-partum* or labo?r* or deliver* or childbirth* or child-birth* or congenit* or conatal*) adj4 (fever* or deliriu* or pyrexia* or hyperthermia*)).tw.
  63. exp chorioamnionitis/
  64. (chorioamnionit* or amnioniti* or funisiti*).tw.
  65. (parenteral* adj4 (antibacter* or anti-bacter* or antibiotic* or anti-biotic* or antimycobact* or anti-mycobact* or bacteriocid*)).tw.
  66. antibiotic prophylaxis/
  67. ((antibacter* or anti-bacter* or antibiotic* or anti-biotic* or antimycobact* or antimycobact* or bacteriocid*) adj4 (prophyla* or premedic* or pre-medic* or prevent*)).tw.
  68. brain disease/
  69. ((brain* or intracranial* or intra-cranial* or encephalon*) adj4 (diseas* or disorder* or defect* or illness* or inflam* or syndrom*)).tw.
  70. encephalopath*.tw.
  71. persistent pulmonary hypertension/
  72. ((persist* or misalign* or mis-align*) adj4 (f?etal* or f?etus* or pulmonar*) adj4 (circulat* or hypertens* or vein*)).tw.
  73. (PPHN or PFC or ACD MPV or ACDMPV).tw.
  74. or/56-73
  75. 55 and 74
  76. nonhuman/ not human/
  77. 75 not 76
  78. limit 77 to english language
CDSR
#1.

MeSH descriptor: [Infant, Newborn] explode all trees

#2.

MeSH descriptor: [Term Birth] this term only

#3.

MeSH descriptor: [Infant Care] this term only

#4.

MeSH descriptor: [Perinatal Care] this term only

#5.

MeSH descriptor: [Intensive Care Units, Neonatal] this term only

#6.

MeSH descriptor: [Intensive Care, Neonatal] this term only

#7.

MeSH descriptor: [Infant Health] this term only

#8.

((newborn* or new born* or neonat* or neo-nat* or perinat* or peri-nat*)):ti,ab,kw

#9.

((premature* or pre-mature* or preterm* or pre-term*) near/4 (child* or infant* or baby* or babies* or offspring)):ti,ab,kw

#10.

{or #1-#9}

#11.

MeSH descriptor: [Bacterial Infections] explode all trees

#12.

((bacter* or strep* or staph* or GNB) near/4 (infect* or diseas* or contaminat* or mening* or pneumon* or nosocomial*)):ti,ab,kw

#13.

MeSH descriptor: [Sepsis] explode all trees

#14.

(sepsis or septic?emia* or py?emia* or pyho?emia*):ti,ab,kw

#15.

(septic* near/4 shock*):ti,ab,kw

#16.

(bacter?emia* or bacill?emia*):ti,ab,kw

#17.

((blood*) near/4 (infect* or contamin* or invas* or invad*)):ti,ab,kw

#18.

{or #11-#17}

#19.

MeSH descriptor: [Streptococcus] explode all trees

#20.

MeSH descriptor: [Staphylococcus] explode all trees

#21.

(streptococc* or staphylococc*):ti,ab,kw

#22.

(GBS or MRSA or NRCS-A or MSSA):ti,ab,kw

#23.

(met?icillin-resistant near/3 aureus):ti,ab,kw

#24.

MeSH descriptor: [Escherichia coli] explode all trees

#25.

((Escheric* or E) near/2 (coli)):ti,ab,kw

#26.

MeSH descriptor: [Listeria] explode all trees

#27.

(listeria*):ti,ab,kw

#28.

MeSH descriptor: [Klebsiella] explode all trees

#29.

(klebsiella*):ti,ab,kw

#30.

MeSH descriptor: [Pseudomonas] explode all trees

#31.

(pseudomonas or chryseomonas or flavimonas):ti,ab,kw

#32.

MeSH descriptor: [Enterobacteriaceae] explode all trees

#33.

(enterobact* or sodalis or paracolobactrum or ewingella or leclercia):ti,ab,kw

#34.

((enteric or coliform) near/2 (bac*)):ti,ab,kw

#35.

MeSH descriptor: [Neisseria] explode all trees

#36.

(neisseria*):ti,ab,kw

#37.

MeSH descriptor: [Haemophilus influenzae] explode all trees

#38.

((h?emophil* or H or bacter* or bacill* or mycobacter* or coccobac*) near/2 (influenz* or pfeiffer* or meningitidis)):ti,ab,kw

#39.

MeSH descriptor: [Serratia] explode all trees

#40.

(serratia*):ti,ab,kw

#41.

MeSH descriptor: [Cronobacter] explode all trees

#42.

(cronobact* or sakazaki* or malonatic*):ti,ab,kw

#43.

MeSH descriptor: [Acinetobacter] explode all trees

#44.

(acinetobact* or herellea* or mima or baumanni* or genomosp* or calcoacetic*):ti,ab,kw

#45.

MeSH descriptor: [Fusobacterium] explode all trees

#46.

(fusobact* or sphaerophor* or necrophorum or nucleatum):ti,ab,kw 296

#47.

MeSH descriptor: [Enterococcus] explode all trees

#48.

(enterococc*):ti,ab,kw

#49.

{or #19-#48}

#50.

#18 or #49

#51.

#10 and #50

#52.

((newborn* or new born* or neonat* or neo-nat* or perinat* or peri-nat*) near/4 (infect*)):ti,ab,kw

#53.

((premature* or pre-mature* or “preterm*” or “pre-term*”) near/4 (child* or infant* or baby* or babies* or offspring) near/4 (infect*)):ti,ab,kw

#54.

#52 or #53

#55.

#51 or #54

#56.

MeSH descriptor: [Fetal Membranes, Premature Rupture] explode all trees

#57.

((preterm* or pre-term* or premature* or pre-mature* or prelabor* or pre-labor* or pre) near/4 (ruptur* or membrane* or disrupt* or erupt* or sever or severed or tear* or breach*)):ti,ab,kw

#58.

(prom or proms or pprom*):ti,ab,kw

#59.

MeSH descriptor: [Gestational Age] this term only

#60.

((gestat* or f?etal* or f?etus*) near/4 (age* or aging* or matur*)):ti,ab,kw

#61.

MeSH descriptor: [Fever] this term only and with qualifier(s): [diagnostic imaging - DG, diagnosis - DI]

#62.

((intrapartum* or intra-partum* or labo?r* or deliver* or childbirth* or child-birth* or congenit* or conatal*) near/4 (fever* or deliriu* or pyrexia* or hyperthermia*)):ti,ab,kw

#63.

MeSH descriptor: [Chorioamnionitis] this term only

#64.

(chorioamnionit* or amnioniti* or funisiti*):ti,ab,kw

#65.

((parenteral*) near/4 (antibacter* or anti-bacter* or antibiotic* or anti-biotic* or antimycobact* or anti-mycobact* or bacteriocid*)):ti,ab,kw

#66.

MeSH descriptor: [Antibiotic Prophylaxis] this term only

#67.

((antibacter* or anti-bacter* or antibiotic* or anti-biotic* or antimycobact* or antimycobact* or bacteriocid*) near/4 (prophyla* or premedic* or pre-medic* or prevent*)):ti,ab,kw

#68.

MeSH descriptor: [Brain Diseases] this term only

#69.

((brain* or intracranial* or intra-cranial* or encephalon*) near/4 (diseas* or disorder* or defect* or illness* or inflam* or syndrom*)):ti,ab,kw

#70.

(encephalopath*):ti,ab,kw

#71.

MeSH descriptor: [Persistent Fetal Circulation Syndrome] this term only

#72.

((persist* or misalign* or mis-align*) near/4 (f?etal* or f?etus* or pulmonar*) near/4 (circulat* or hypertens* or vein*)):ti,ab,kw

#73.

(PPHN or PFC or ACD MPV or ACDMPV):ti,ab,kw

#74.

{or #56-#73}

#75.

#55 and #74

DARE
  1. MeSH DESCRIPTOR Infant, Newborn EXPLODE ALL TREES
  2. MeSH DESCRIPTOR Term Birth
  3. MeSH DESCRIPTOR Infant Care
  4. MeSH DESCRIPTOR Perinatal Care
  5. MeSH DESCRIPTOR Intensive Care Units, Neonatal
  6. MeSH DESCRIPTOR Intensive Care, Neonatal
  7. MeSH DESCRIPTOR Infant Health
  8. (newborn* or new born* or neonat* or neo-nat* or perinat* or peri-nat*)
  9. ((premature or pre-mature* or preterm or pre-term) NEAR4 (child* or infant* or baby* or babies* or offspring))
  10. (#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9)
  11. MeSH DESCRIPTOR Bacterial Infections EXPLODE ALL TREES
  12. ((bacter* or strep* or staph* or GNB) NEAR4 (infect* or diseas* or contaminat* or mening* or pneumon* or nosocomial*))
  13. MeSH DESCRIPTOR Sepsis EXPLODE ALL TREES
  14. (sepsis or septic?emia* or py?emia* or pyho?emia*)
  15. ((septic* NEAR4 shock*))
  16. (bacter?emia* or bacill?emia*)
  17. ((blood*) NEAR4 (infect* or contamin* or invas* or invad*))
  18. (#11 OR #12 OR #13 OR #14 OR #15 OR #16 OR #17)
  19. MeSH DESCRIPTOR Streptococcus EXPLODE ALL TREES
  20. MeSH DESCRIPTOR Staphylococcus EXPLODE ALL TREES
  21. (streptococc* or staphylococc*)
  22. (GBS or MRSA or NRCS-A or MSSA)
  23. ((met?icillin-resistant NEAR3 aureus))
  24. MeSH DESCRIPTOR Escherichia coli EXPLODE ALL TREES
  25. ((Escheric* or E) NEAR2 (coli))
  26. MeSH DESCRIPTOR listeria EXPLODE ALL TREES
  27. (listeria*)
  28. MeSH DESCRIPTOR Klebsiella EXPLODE ALL TREES
  29. (klebsiella*)
  30. MeSH DESCRIPTOR Pseudomonas EXPLODE ALL TREES
  31. (pseudomonas or chryseomonas or flavimonas)
  32. MeSH DESCRIPTOR Enterobacteriaceae EXPLODE ALL TREES
  33. (enterobact* or sodalis or paracolobactrum or ewingella or leclercia)
  34. ((enteric or coliform) NEAR2 (bac*))
  35. MeSH DESCRIPTOR Neisseria EXPLODE ALL TREES
  36. (neisseria*)
  37. MeSH DESCRIPTOR Haemophilus influenzae EXPLODE ALL TREES
  38. ((h?emophil* or H or bacter* or bacill* or mycobacter* or coccobac*) NEAR2 (influenz* or pfeiffer* or meningitidis))
  39. MeSH DESCRIPTOR Serratia EXPLODE ALL TREES
  40. (serratia*)
  41. MeSH DESCRIPTOR Cronobacter EXPLODE ALL TREES
  42. (cronobact* or sakazaki* or malonatic*)
  43. MeSH DESCRIPTOR Acinetobacter EXPLODE ALL TREES
  44. (acinetobact* or herellea* or mima or baumanni* or genomosp* or calcoacetic*)
  45. MeSH DESCRIPTOR Fusobacterium EXPLODE ALL TREES
  46. (fusobact* or sphaerophor* or necrophorum or nucleatum)
  47. MeSH DESCRIPTOR Enterococcus EXPLODE ALL TREES
  48. (enterococc*)
  49. (#19 OR #20 OR #21 OR #22 OR #23 OR #24 OR #25 OR #26 OR #27 OR #28 OR #29 OR #30 OR #31 OR #32 OR #33 OR #34 OR #35 OR #36 OR #37 OR #38 OR #39 OR #40 OR #41 OR #42 OR #43 OR #44 OR #45 OR #46 OR #47 OR #48)
  50. (#18 OR #49)
  51. (#10 AND #50)
  52. ((newborn* or new born* or neonat* or neo-nat* or perinat* or peri-nat*) NEAR4 (infect*))
  53. ((premature or pre-mature* or preterm or pre-term) NEAR4 (child* or infant* or baby* or babies* or offspring) NEAR4 (infect*))
  54. (#52 OR #53)
  55. (#51 OR #54)
  56. MeSH DESCRIPTOR Fetal Membranes, Premature Rupture EXPLODE ALL TREES
  57. ((preterm* or pre-term* or premature* or pre-mature* or prelabor* or pre-labor* or pre) NEAR4 (ruptur* or membrane* or disrupt* or erupt* or sever or severed or tear* or breach*))
  58. (prom or proms or pprom*)
  59. MeSH DESCRIPTOR Gestational Age
  60. ((gestat* or f?etal* or f?etus*) NEAR4 (age* or aging* or matur*))
  61. MeSH DESCRIPTOR Fever WITH QUALIFIER DI
  62. MeSH DESCRIPTOR Fever WITH QUALIFIER DG
  63. ((intrapartum* or intra-partum* or labo?r* or deliver* or childbirth* or child-birth* or congenit* or conatal*) NEAR4 (fever* or deliriu* or pyrexia* or hyperthermia*))
  64. MeSH DESCRIPTOR Chorioamnionitis
  65. (chorioamnionit* or amnioniti* or funisiti*)
  66. ((parenteral*) NEAR4 (antibacter* or anti-bacter* or antibiotic* or anti-biotic* or antimycobact* or anti-mycobact* or bacteriocid*))
  67. MeSH DESCRIPTOR Antibiotic Prophylaxis
  68. ((antibacter* or anti-bacter* or antibiotic* or anti-biotic* or antimycobact* or antimycobact* or bacteriocid*) NEAR4 (prophyla* or premedic* or pre-medic* or prevent*))
  69. MeSH DESCRIPTOR Brain Diseases
  70. ((brain* or intracranial* or intra-cranial* or encephalon*) NEAR4 (diseas* or disorder* or defect* or illness* or inflam* or syndrom*))
  71. (encephalopath*)
  72. MeSH DESCRIPTOR Persistent Fetal Circulation Syndrome
  73. ((persist* or misalign* or mis-align*) NEAR4 (f?etal* or f?etus* or pulmonar*) NEAR4 (circulat* or hypertens* or vein*))
  74. (PPHN or PFC or ACD MPV or ACDMPV)
  75. #56 OR #57 OR #58 OR #59 OR #60 OR #61 OR #62 OR #63 OR #64 OR #65 OR #66 OR #67 OR #68 OR #69 OR #70 OR #71 OR #72 OR #73 OR #74
  76. #55 AND #75
  77. * IN DARE
  78. #76 AND #77
Search Filters

The following search filters were combined as ‘And’ with the population and risk factor terms for the Medline databases and Embase. CDSR and DARE are systematic review databases so did not require the addition of a filter.

The Medline versions of the filters are reproduced below. Embase has validated translations of these that were used in the search.

Systematic Review
  1. MEDLINE or pubmed).tw.
  2. systematic review.tw.
  3. systematic review.pt.
  4. meta-analysis.pt.
  5. intervention$.ti.
  6. or/1-5
Observational studies

The in-house observational studies filter was adapted to focus on cross-sectional studies this was then supplemented with the McMaster diagnostic and prognostic filters.

  1. Cohort Studies/
  2. Prospective Studies/
  3. Retrospective Studies/
  4. Cross-Sectional Studies/
  5. cohort:.mp.
  6. predictor:.tw.
  7. cross sectional.tw.
  8. prospective*.tw.
  9. retrospective*.tw.
  10. sensitiv:.mp.
  11. predictive value:.mp.
  12. accurac:.tw.
  13. prognosis.sh.
  14. diagnosed.tw.
  15. death.tw.
  16. exp models, statistical/
  17. or/1-16
Risk terms

Following combination of population, risk factor and filter terms (if an appropriate database) the number of results were still considered too high. Additional risk terms were combined as ‘And’ with the other sections of the search strategy to reduce numbers.

The Medline risk terms are listed below. These were translated across all databases used in the search:

  1. exp Risk/
  2. exp Risk Management/
  3. Pregnancy, High Risk/
  4. risk*.tw.
  5. exp Health Status Indicators/
  6. ((health* or illness* or wellness* or wellbeing* or well-being*) adj4 (indicat* or index* or indices* or apprais* or barometer* or gaug* or mark* or warn* or ratio or ratios)).tw.
  7. (sever* adj4 illness*).tw.
  8. exp “Signs and Symptoms”/
  9. ((symptom* or sign or signs or manifest* or phenomenon*) adj8 (infect* or diseas* or contaminat* or mening* or pneumon* or nosocomial*)).tw.
  10. or/1-9

Virus terms

The following terms were combined as ‘Not’ with the other sections of the search strategy to remove any papers focused on viral illness.

The Medline virus terms are listed below. These were translated across all databases used in the search:

  1. exp Virus Diseases/
  2. exp Viruses/
  3. (virus* or viral* or retrovir* or arbovir* or lentivir* or deltaretrovir* or adenovir*).tw.
  4. HIV*.tw.
  5. (cytomegalovir* or CMV*).tw.
  6. herpes*.tw.
  7. (papillomavir* or HPV*).tw.
  8. ((hepatitis* or hepatitid*) adj2 (A or B or C or D or E)).tw.
  9. (parechovir* or echovir*).tw.
  10. (yellow* adj2 fever*).tw.
  11. rhinovir*.tw.
  12. (coronavir* or deltacoronavir*).tw.
  13. rotavir*.tw.
  14. (enterovir* or coxsackie*).tw.
  15. exp Malaria/
  16. (malaria* or paludism*).tw.
  17. exp Syphilis/
  18. (syphili* or neurosyphili* or neuro-syphili*).tw.
  19. or/1-18

Risk factors included in the search for review questions 5.1 and 5.2

The risk factors searched for as part of review questions 5.1 and 5.2 were considered for this evidence review and results from that search considered as part of the analysis. As a result of this they were removed from this search. This was to ensure there was no duplication of effort and there was a unique set of results for this search.

The following risk factors were combined as ‘Not’ with the other sections of the search strategy. The Medline risk factor terms are listed below. These were translated across all databases used in the search:

  1. ((previous or preceding or earlier or prior or antecedent) adj4 (child* or infant* or baby* or babies* or offspring or delivery or deliveries)).tw.
  2. ((later or next or succeeding) adj4 (child* or infant* or baby* or babies* or offspring or delivery or deliveries)).tw.
  3. (Infectious Disease Transmission, Vertical/ or Carrier State/) and (Streptococcal Infections/ or Methicillin-Resistant Staphylococcus aureus/)
  4. ((vertical* or maternal* or mother* or mum* or mom* or parental* or fetomaternal* or fetomaternal* or woman* or women* or pregnan* or gestat* or parturition* or birth* or childbirth* or labo?r*) adj4 (GBS* or group B* or MRSA* or met?icillin-resist*) adj4 (transmission* or transmit* or transfer* or infect* or diseas* or contaminat* or coloni?ation* or contagio* or bacteriuria* or carrier* or carriage or heterozygo*)).tw.
  5. exp Pregnancy, Multiple/
  6. ((multiple or twin or triplet or quadruplet or quintuplet or superfetation) adj4 (pregnan* or gestat* or parturition* or birth* or childbirth* or labo?r* or delivery or deliveries)).tw.
  7. Wound Infection/
  8. (wound* adj4 (infect* or diseas* or contaminat* or coloni?ation* or contagio* or sepsis)).tw.
  9. Postpartum Period/
  10. (postpartum or post-partum or puerperium or puerperal).tw.
  11. ((perineal or perineum) adj4 (infect* or diseas* or contaminat* or coloni?ation* or contagio*)).tw.
  12. exp Obesity/
  13. ((obesity or obese or overweight or over-weight) adj8 risk*).tw.
  14. exp Hygiene/
  15. exp Sanitation/
  16. (hygien* or saniti?e* or sanitation* or sanitary*).tw.
  17. exp Maternal Behavior/
  18. ((behavio?r* or attitud*) adj4 (factor* or aspect* or consider* or circumstanc* or component* or influenc* or feature*)).tw.
  19. Illness Behavior/
  20. ((alter* or chang* or illness*) adj4 (behavio?r* or respons* or feedback*) adj8 risk*).tw.
  21. Muscle Hypotonia/
  22. (flop* or flaccid* or hypoton* or hypomyotoni*).tw.
  23. ((alter* or chang* or poor* or decreas* or atonic* or feeble or insuffic* or meag* or weak* or unsatisfact* or imperfect* or impair* or declin* or reduc* or diminish*) adj4 musc*).tw.
  24. Feeding Behavior/
  25. ((feed* or bottle* or breast*) adj4 (behavio?r* or difficult* or refus* or intoleran* or declin* or ignor* or withdraw* or problem*)).tw.
  26. exp Vomiting/
  27. (vomit* or emesis*).tw.
  28. ((gastric* or nasogastric* or naso-gastric*) adj4 (aspirat* or suction*)).tw.
  29. (abdom?n* adj4 disten*).tw.
  30. Arrhythmias, Cardiac/ or Atrial Fibrillation/ or Atrial Flutter/ or Cardiac Complexes, Premature/ or Parasystole/ or Ventricular Fibrillation/ or Ventricular Flutter/
  31. (arr?ythmia* or dysrhythmia*).tw.
  32. ((abnormal* or anomal* or atypical* or irregular* or uncommon* or unexpect*) adj4 (heart* or cardiac* or vascular*) adj2 (rate* or pace* or measure* or rhythm* or beat*)).tw.
  33. Bradycardia/ or Tachycardia/
  34. (bradycardia* or bradyarrhythmia* or tachycardia* or tachyarrhythmia*).tw.
  35. Respiratory Distress Syndrome, Newborn/
  36. ((respirat* or breath*) adj4 (distres* or troubl* or discomfort*)).tw.
  37. exp Hypoxia/
  38. (hypoxia* or hypoxic* or hypoxem* or anoxia* or anoxem*).tw.
  39. (oxygen* adj4 (deficien* or reduc* or suturat* or concentrat* or measur*)).tw.
  40. exp Cyanosis/
  41. exp Oximetry/
  42. (cyanos?s* or cyanotic* or oximet*).tw.
  43. exp Jaundice, Neonatal/
  44. (jaundice* or icterus*).tw.
  45. exp Apnea/
  46. apn?ea*.tw.
  47. Seizures/
  48. ((seizure* or convuls* or paroxysm*) adj8 risk*).tw.
  49. exp Cardiopulmonary Resuscitation/
  50. (((cardiopulmon* or cardio-pulmon* or mouth-to-mouth*) adj4 resuscitat*) or CPR).tw.
  51. exp Respiration, Artificial/
  52. ((artificial* or mechanic* or automat* or machine* or control*) adj4 (respirat* or ventilat* or breath* or oxygenat*)).tw.
  53. exp Body Temperature/
  54. ((body* or organ* or skin* or high* or low* or excess* or reduc*) adj4 temperat*).tw.
  55. ((“36*” or “38*”) adj2 (C or celsius)).tw.
  56. ((“96*” or “100*”) adj2 (F or fahrenheit)).tw.
  57. exp Shock/
  58. (shock not (septic or sepsis)).tw.
  59. (circulat* adj4 (collaps* or fail*)).tw.
  60. ((pale* or cold* or clammy or chill* or blanch*) adj4 skin*).tw.
  61. Sweat/ or Sweating/
  62. (sweat* or perspir*).tw.
  63. ((rapid* or shallow* or accelarat* or hollow* or flat*) adj4 (breath* or respirat*)).tw.
  64. (weakness* or fragilit*).tw.
  65. Dizziness/
  66. (dizz* or orthostas* or lighthead* or light-head*).tw.
  67. Thirst/
  68. thirst*.tw.
  69. Yawning/
  70. (yawn* or sigh or sighs).tw.
  71. exp Hemorrhage/
  72. (bleed* or h?emorrhag*).tw.
  73. (blood* adj4 (loss or effus* or excess*)).tw.
  74. exp Thrombocytopenia/
  75. (thrombocytop?enia* or thrombop?enia*).tw.
  76. Blood Coagulation/
  77. ((coagulat* or clot or clott*) adj8 risk*).tw.
  78. Oliguria/
  79. oliguria*.tw.
  80. ((decreas* or diminish* or dwindl* or reduc* or wane) adj4 urin*).tw.
  81. Homeostasis/
  82. (homeostas* or homeostat* or autoregulat* or auto-regulat*).tw.
  83. exp Hypoglycemia/
  84. exp Hyperglycemia/
  85. (hypoglyc?emi* or hyperglyc?emi*).tw.
  86. ((low* or high*) adj4 blood* adj4 (sugar* or glucose*)).tw.
  87. exp Acidosis/
  88. acidos?s*.tw.
  89. ((local* or region* or limit*) adj4 (infect* or contamin* or invas*)).tw.
  90. ((previous* or preced* or earlier or prior* or anteceden* or histor* or past*) adj4 (surg* or operat*)).tw.
  91. exp Catheters/ or Catheterization/ or Catheterization, Central Venous/ or exp Catheterization, Peripheral/
  92. ((catheter* or cannula*) adj4 (present* or presence* or exist* or attend* or current*)).tw.
  93. ((indwell* or in-dwell*) adj4 (devic* or apparat* or applianc* or equip* or gadget* or machine* or mechanism*)).tw.
  94. (prematur* adj8 risk*).tw.
  95. ((newborn* or new born* or neonat* or neo-nat* or perinat* or peri-nat*) adj4 (admiss* or admit*)).tw.
  96. ((previous* or preced* or earlier or prior* or anteceden* or histor* or past*) adj4 (GBS* or group B*) adj4 (infect* or contamin* or invas*)).tw.
  97. ((newborn* or new born* or neonat* or neo-nat* or perinat* or peri-nat* or infant* or baby* or babies*) adj4 (GBS* or group B* or MRSA* or met?icillin-resist*) adj4 (contaminat* or coloni?ation* or contagio*)).tw.
  98. or/1-97

B.3. Economic search: Health Economics literature search strategy

Download PDF (360K)

Appendix C. Prognostic and diagnostic evidence study selection

C.1. Clinical prediction models (PDF, 162K)

C.2. Maternal and neonatal risk factors (PDF, 136K)

Appendix D. Prognostic and diagnostic evidence

D.1. Clinical prediction models (PDF, 533K)

D.2. Maternal and neonatal risk factors (PDF, 465K)

D.3. Additional studies (PDF, 246K)

Appendix E. Forest plots

Download PDF (586K)

Appendix F. GRADE tables

F.1. Clinical prediction models (PDF, 318K)

F.2. Maternal risk factors (PDF, 200K)

F.3. Neonatal risk factors (PDF, 200K)

Appendix G. Economic evidence study selection

Download PDF (134K)

Appendix H. Economic evidence tables

No economic evidence is available as none of the studies in the economic search results were found to be relevant.

Appendix I. Health economic model

This question was not prioritised for original economic analysis.

Appendix J. Excluded studies

J.1. Clinical prediction models (PDF, 326K)

J.2. Maternal and neonatal risk factors (PDF, 262K)

Appendix K. Research recommendations – full details

K.1.1. Research recommendation

What is the accuracy of clinical prediction models for early-onset neonatal infection in the UK and what is their effectiveness in guiding management in the baby?

K.1.2. Why this is important

Nine observational studies were identified evaluating the accuracy of clinical prediction models for early-onset neonatal infection. These primarily evaluated the use of the Kaiser Permanente neonatal sepsis calculator. However, most of the evidence has validated the use of this tool in the USA, with only one study examining its use in the UK. In addition, the neonatal sepsis calculator is designed for use with babies at or over 34 weeks’ gestational age. There is currently no evidence for the use of clinical prediction models for babies born at a gestational age of less than 34 weeks.

Further research is needed using a robust study design such as prospective cohort studies, parallel RCTs or cluster RCTs to either examine the effectiveness of existing clinical prediction models for early-onset neonatal infection, or to develop new clinical prediction models designed for use in UK clinical practice. Research in this area is essential to help develop accurate methods of identifying newborn babies most at risk of developing earlyonset neonatal infection whilst avoiding over-prescribing of antibiotics.

K.1.3. Rationale for research recommendation

Download PDF (198K)

K.1.4. Modified PICO table (Part A – prognostic accuracy)

Download PDF (206K)

K.1.5. Modified PICO table (Part B – clinical effectiveness)

Download PDF (154K)

K.2.1. Research recommendation

What is the risk of early-onset neonatal infection with maternal obesity and how does this change with increasing BMI?

K.2.2. Why this is important

One retrospective cohort study (Hakansson 2008) was identified evaluating the association between maternal obesity and early-onset neonatal infection. This defined maternal obesity using the World Health Organisation definition based on BMI (underweight <18.5, normal 18.5–24.9, overweight 25–29.9 and obese ≥30). However, this definition does not reflect the definition that is used in current practice, where mothers with a BMI greater than 35 are now classified as obese. The differences in BMI classifications between this study and clinical practice mean that there is currently no evidence that can be directly applied to clinical practice in the UK. In addition, with only one study currently available it is difficult to assess any associations between BMI and the risk of neonatal infection.

Further research is needed using a robust study design such as prospective cohort studies, to determine the association between maternal obesity and the risk of babies developing early-onset neonatal infection. Research in this area is important to help provide clinicians with a more detailed understanding of the risk factors associated with early-onset infection, thereby helping to identify which babies are most at risk of infection within the first 72 hours of life. Such research is relevant to the NHS due the potential resource impact. Infections are both costly to treat and may result in severe adverse health outcomes. As such, research that helps clinicians identify which babies are most at risk, is likely to result in cost-savings at the population level and also improve health outcomes.

K.2.3. Rationale for research recommendation

Download PDF (202K)

K.2.4. Modified PICO table

Download PDF (185K)

Final

Evidence reviews underpinning recommendations 1.3.1-1.3.9 and research recommendations in the NICE guideline

These evidence reviews were developed by NICE Guideline Updates Team

Disclaimer: The recommendations in this guideline represent the view of NICE, arrived at after careful consideration of the evidence available. When exercising their judgement, professionals are expected to take this guideline fully into account, alongside the individual needs, preferences and values of their patients or service users. The recommendations in this guideline are not mandatory and the guideline does not override the responsibility of healthcare professionals to make decisions appropriate to the circumstances of the individual patient, in consultation with the patient and/or their carer or guardian.

Local commissioners and/or providers have a responsibility to enable the guideline to be applied when individual health professionals and their patients or service users wish to use it. They should do so in the context of local and national priorities for funding and developing services, and in light of their duties to have due regard to the need to eliminate unlawful discrimination, to advance equality of opportunity and to reduce health inequalities. Nothing in this guideline should be interpreted in a way that would be inconsistent with compliance with those duties.

NICE guidelines cover health and care in England. Decisions on how they apply in other UK countries are made by ministers in the Welsh Government, Scottish Government, and Northern Ireland Executive. All NICE guidance is subject to regular review and may be updated or withdrawn.

Copyright © NICE 2021.
Bookshelf ID: NBK571216PMID: 34133104

Views

  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (3.1M)

Other titles in this collection

Supplemental NICE documents

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...