Evidence table of likely bacterial causes of serious bacterial infections in the UK after 1992


Study type:
Prospective cohort study
EL: 2+
All children presenting to hospital with temperatures ≥38 °C see in two hospitals in New Castle between August to October 1999.One hundred and forty one children between 8 days and 16 years of age were studied, 64% male, 55% aged under 2 years. Serious disease was present in 41 (29%) with 31 (22%) microbiologically or radiologically proven and the other 10 given a diagnosis of sepsis cause including three patients with clinical signs of meningococcal disease but without any positive culture.

35/41 (86%) of patients with serious bacterial infections had temperatures between 38 and 39 °C and 3 (7%) had temperature between 38–39 °C. Ninety six percent were casualty or GP referrals and 4% were tertiary referrals. Twenty nine percent (41/141) had serious disease but microbiologically or radiologically proven in only 22% (31/141); pneumonia (nine), meningitis (seven), sepsis (five), urinary tract infection (five), brain abscess (two), toxic shock syndrome (one), appendicitis (one), ischiorectal abscess (one). Forty two percent (5/12) of microbiologically proven meningitis and sepsis and 36% (8/22) of all meningitis and sepsis were meningococcal. 71% had non-serious diseases. In cases of serious disease the temperature was >39 °C in 15% (sensitivity: 14%, specificity: 82%, PPV: 25%). Poor feeding and restlessness predicted serious disease with a sensitivity of 78% and 76%, respectively. Full blood count (FBC) was taken in 50% of patients on admission; in 44% of serious and 24% of non-serious diseases WBC was between 5000 and 15 000/mm3 and WBC ≥15 000/mm3 was seen in 39% of serious diseases (sensitivity:10%, specificity: 95%, PPV: 44%).

BT > 39 °C had sensitivity of 14% (3–25%), specificity 82% (74–89%), PPV 25% (7–42%) and NPV (70 61–78%).

Definition of serious infections: sepsis, meningitis, toxic shock syndrome, brain abscess, pneumonia, UTI, ischiorectal abscess, appendicitis.

Table : Lab investigations
No. of patientsNo of positive resultsDetails
Blood culture596N. meningitidis (3)
S. pneumoniae (2)
H. influenzae (1)
Lumbar puncture177S. pneumoniae (2)
S. epidermidis (1)
H. influenzae (1)
N. meningitidis GB (1)
Pos CSF profile (1)
Enterovirus (PCR) (1)
Urine microscopy9910
Throat swab2912HSV1 (4)
βH streptococci GA (2)
RSV (2)
S. aureus (1)
S. pneumoniae (1)
H. influenzae (1)
Pseudomonas (1)
Blood serology121Anti-streptolysin O=500
PCR113N. meningitidis (2)
Enterovirus (1)

Study type:
Prospective cohort study
EL: 2+
This is an observational prospective study done at the Accident and Emergency Department of the Royal Hospital for Sick Children, Edinburgh where approximately 30,000 patients under the age of fourteen are seen annually. The study period was chosen to coincide with the peak incidence of infectious diseases. Between 10 Dec. 1997 and 22 Feb. 1998, data were collected prospectively for all children presenting with an infectious illness. These were identified by the diagnosis made by the attending doctor, or else defined as symptoms suggestive of an infectious process, either specific e.g. cough or general such as fever or being unwell in the absence of a history of trauma or toxic ingestion. An experienced nurse recorded vital signs, including axillary temperature, for all patients. Doctors attending those children recorded the provisional diagnosis on a study pro forma, on which detailed clinical and laboratory data were then collected.A total of 5021 patients, including surgical patients, were seen during the nine week period with 1547 (31%) patients presenting with features of an infectious illness. Hereafter, the discussion concerns only this latter group. The median age was 17 months (range 0.1–224). There were 804 (52%) males and 743 (48%) females. Forty six per cent of the patients were self referred and 45% were referred by their general practitioners. The remainder were from other sources, such as school nurses. The mean temperature for all the patients was 37.5°C (SD 1.1, range 34.8–40.6) with 42% of the patients being febrile on presentation (axillary temperature ≥37.5°C) and 22% with a temperature ≥38.5°C.

Hospitalisation. The admission rate for the studied group was 41% (635/1547) Admitted children were younger and had a significantly higher temperature than these who were not admitted (both p=0.0005). More children were admitted when referred by their GPs than if they self referred (p=0.0005). In only 133 (21%) of cases was the reason for admission stated. The height of temperature was indicated as a reason for admission in 20% patients. In 11% of admissions parental anxiety or request were mentioned among the reasons for admission and a further 3% were for other social reasons. Uncertain diagnosis accounted for 10% of admissions, whereas in the majority of patients the severity of illness was identified as the reason for admission. The median duration of hospital stay was 2.7 days, with a range of 1 to 31 days.

Need for hospital treatment. Only 44% of the admitted patients had some form of treatment necessitating hospitalisation. The source of referral did not indicate the need for hospital treatment (χ2 0.484 df1, p=0.487), and there was no difference in length of hospital stay between GP and self-referred patients (p=0.547 MW test). Among many clinical and laboratory parameters evaluated, only the illness severity score predicted the need for hospital treatment.

Type of infection. The temperature was significantly higher in bacterial infections (p = 0.006). Less than 20% of patients with non-bacterial infections had a temperature greater than 38.5°C. However, the positive predictive value of fever of this magnitude in predicting bacterial infection was only 40%. At a cut-off value of 10x109/1, the absolute neutrophil count had a sensitivity and specificity of 54% and 72% respectively, for predicting bacterial infection (with 47% and 77% positive and negative predictive values respectively). The corresponding values for the white cell count were almost the same as those of the neutrophil count. Only the absolute neutrophil count and the possibility score retained significance in a regression model for the prediction of bacterial infection.

Blood culture. Blood was obtained for culture from 275 (43%) of the 635 patients admitted. Patients who were investigated with a blood culture were older than those who were not (median age 34.1m and 29.6m respectively, p=0.001). At higher temperatures, more patients were investigated with blood culture and more had bacteraemic illness (χ2 20.4, df2. p=0.0001, Kruskal-Wallis test). Seven (2.5%) blood cultures were positive for pathogenic organisms (SE:0.0095, CI:0.007 –0.044),. It is noteworthy that none of the three patients with clinically diagnosed and treated meningococcal septicaemia had a positive blood culture. A throat swab from one of these patients grew Neisseria meningitidis. The absence of Haemophilus influenzae type b was also evident.

Antecedent antibiotic treatment did not influence culture results (χ2 0.517, df=l, p=0.680, Fisher exact test). However, the small number of patients in the positive group limits the strength of this comparison. Bacteraemic patients had significantly higher temperatures (p=0.0372) and higher neutrophil (p=0.0056) and total white cell counts (p=0.0135). The sensitivity and specificity of temperature >38.5°C in predicting a positive blood culture were 71% and 63% respectively (PPV 5%, NPV 99%). The figures for neutrophilia as defined above were comparable to those of high fever. On logistic regression, only temperature remained a sole independent predictor of bacteraemia. It was notable that only a very small number (10%) of patients evaluated by a blood culture had either CRP or ESR done.

Table. Details of the positive blood cultures
Haemophilus influenzae biotype V noncapsulated1
Group B beta haemolytic streptococcus1
Streptococcus pneumoniae group 62
Streptococcus pneumoniae group 231
Streptococcus pneumoniae group 14a1
E. coli (coliforms)1
Total 7

Study type:
Prospective cohort study
EL: 2+
6-month prospective study of paediatric accident and emergency and general practice consultations with a diagnosis of community-acquired pneumonia (CAP).
The study population was from the catchments area for Royal London Hospital, East London from 30/09/2001–30/03/2002.
Any child younger than 5 years with symptoms and signs indicating CAP was eligible regardless of risk factors. VAP was defined as a respiratory illness with fever >38.5OC and tachypnoea (respiratory rate > 40/min in children 1–5 yr; >50/min for 1–11 mo and >60/min < 1 mo) with or without cough, plus evidence of consolidation from clinical exam or chest radiography. Children with fever who were not tachypnoeic but had clinical evidence of consolidation on chest radiogram were included.
Exclusion: young children with obvious bronchiolitis were excluded.
Nasopharyngeal aspirates for viral immunofluorescence and PCR studies and blood cultures for bacterial studies were taken from 51 (age ranged from 2 weeks to 4.8 years, median age of 1.3 years; 63% girl) children with symptoms, signs and chest radiographic features that satisfied a diagnosis of pneumonia.
45 patients (88%) were recruited from the hospital emergency department and 6 from GP. 42 (82%) were hospitalised.
An etiologic agent was isolated from 25 patients (49%). A viral cause was identified in 22 patients (43%), and influenza A virus and respiratory syncytial virus (RSV) were detected in 16 and 18% of all cases, respectively. Moreover, they found 1 case (2%) with enterovirus, 3 cases (6%) with parainfluenza, 3 with adenovirus (6%).
Only four patients (8%) had a positive bacterial blood culture; three had Streptococcus pneumoniae and one had Neisseria meningitidis W135. Mycoplasma pneumoniae was detected in 2 children, and mixed infections were detected in 5 (10%). The use of viral PCR increased the detection rate of influenza A virus by 100%.

Study type:
Prospective cohort study
EL: 2+
This is a multicentre prospective study including 21 hospitals in the south and west of England and South Wales between November 1993 to April 1995. 124 (83% of eligible population) children between the ages of 4 weeks and 16 years with newly diagnosed bacterial meningitis. The age ranged 0.1–15.6 years (median 2.1 yr) with no fatality in this series.Ninety two children (74%) had meningococcal and 18 (15%) had pneumococcal meningitis. Fifty two of these children had Neisseria meningitidis isolated by microscopy or culture of cerebrospinal fluid. Twenty six patients had cerebrospinal fluid pleocytosis plus positive meningococcal blood cultures or petechiae, and 14 children who did not undergo lumbar puncture had meningism and evidence of meningococcal disease. Streptococcus pneumoniae was isolated from the cerebrospinal fluid of 18 patients (15%). There was one case each of meningitis due to Haemophilus influenzae type b, Listeria monocytogenes, and group B streptococcus. In the remaining 11 cases (8%), all of whom had a cerebrospinal fluid neutrophil pleocytosis, the pathogen was unknown. Thirty four children had received parenteral penicillin before admission. This included 24 of the 26 patients with cerebrospinal fluid pleocytosis and signs of meningococcal disease.

Study type:
Prospective cohort
EL: 2+
During 1998, each case of meningococcal disease reported by all Medical Officers for Environment Health in Wales, the information resource included GP, hospital clinicians and microbiologists. Patients were identified from statutory notifications, reports from data provided by Manchester Public Health Lab (PHL). Meningococcal Reference Lab. Patients with clinical features of meningitis but whose diagnosis was not confirmed by blood or CSF culture, were considered to have meningococcal disease when a purpuric rush and an abnormal CSF were reported.In Wales, in 1998, 119 (63 male) patients with meningococcal were identified.
They included 10 without an organism cultured from blood or CSF and without evidence of purpuric rash. In five of those 10, diplococci were seen in the CSF; in the other five, the CSF was abnormal. The crude incidence, in 1998, in Wales was 4.2 cases/100 000. The age specific incidence was 83/100 000 in infants, 35/100 000 in 1–4 year-old, 5/100 000 in 5–14 year-old and 1/100 000 in adults. The peak incidence was between January and April, with a smaller peak towards the end of the year. The fatality rate was 3% (1/31) in infants, 10% (5/51) in 1–4 year-old, 18% (3/17) in 5–14 year-old and 20%(4/20) in older teenagers and adults.

Among 105 of the 111 meningococcal strains identified at Manchester PHL, there were 77 group B strains (74% sulphonamide sensitive), 27 group C strain (63% sulphonamide sensitive) and 1 group Y strain (sulphonamide sensitive). There were 26 different serotypes, the most common being B2bnt (n:23), B15P1.16 (n:12), Bnt nt (n:11), Bnt P1.15 (n:11) and Cnt nt (n:9).

Clinical features:
Signs and symptoms reported in cases of meningococcal disease
<1 yr (n=25)1–4 yr (n=39)5–14 yr (n=13)
Fever20 (80%)35 (90%)11 (85%)
Vomiting14 (56%)31 (79%)11 (85%)
Fever & vomiting13 (52%)28 (72%)8 (62%)
Refusal of feeds16 (64%)15 (38%)-
Loss of appetite-18 (46%)5 (38%)
Listless13 (52%)27 (69%)5 (38%)
Floppy5 (20%)12 (31%)1 (8%)
Pallor8 (32%)12(31%)6(46%)
Photophobia3 (12%)3 (8%)2(15%)
Neck stiffness5(20%)19 (49%)5(38%)

purpuric rash


other rash

19 (76%)
13 (52%)
35 (90%)
Fever, vomiting and rash9(36%)24(62%)5(38%)
Neck stiffness and rash3(12%)17(44%)5(38%)
Neck stiffness or rash21(84%)37(95%)12(92%)
Neck stiffness, rash and headache1 (4%)5(13%)4(31%)
Neck stiffness, vomiting and headache4(16%)12(31%)5(38%)

From: Evidence tables

Cover of Feverish Illness in Children
Feverish Illness in Children: Assessment and Initial Management in Children Younger than 5 Years.
NICE Clinical Guidelines, No. 47.
National Collaborating Centre for Women’s and Children’s Health (UK).
London (UK): RCOG Press; 2007 May.
Copyright © 2007, National Collaborating Centre for Women’s and Children’s Health.

No part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK [www.cla.co.uk]. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.