Symptoms and signs of serious illness

Citation/ELMethodResults
Baraff163

study type:
Systematic review and meta-analysis

El: 2+
Aim:
They aimed to determine the prevalence of meningitis, bacteraemia and all SBIs in the febrile infants < 3 months according to commonly used clinical and lab factors. Moreover, to identify the nature and aetiology of SBIs in this age group to determine the outpatient management.
Method:
They searched English language literature using MEDLINE from 1972 to May 1991. They only included original studies concerning febrile infants < 3 months. SBI was defined as sepsis, meningitis, bacteraemia, pneumonia, UTI, bacterial enteritis, septic arthritis and osteomyelitis.
They used hierarchical Bayesian meta-analysis to combine data from individual publications.
The mean risk of bacteraemia of the individual studies ranged from 0–3.2%, the mean of the probability distribution of the combined studies was 1.4% and the upper limits of the 95% CI was 2.7%. The results also showed that the classification of Rochester criteria results in two populations at significantly different risk of bacteraemia.

Table : Hierarchical Bayesian meta-analysis: probability of bacterial infections in infants ≤ months of age as a function of clinical and lab findings*
% of patients
RochesterLow risk**Non-toxicToxicHigh risk
SBI1.4 (0.4–2.7)2.6 (1.5–4.0)8.6(3.7–15.6)17.3 (8.0–30.0)24.3 (18.2–31.4)
Bacteraemia1.1 (0.2–2.6)1.3 (0.8–2.1)2.0 (0.8–3.8)10.7 (6.7–15.7)12.8 (7.3–19.9)
Meningitis0.5 (0.0–1.0)0.6 (0.3–1.0)1.0 (0.2–2.4)3.9 (1.7–7.1)3.9 (1.7–7.0)
*: numbers in parentheses, 95% CI of the probability distribution.
** low risk were defined as previously healthy, non-toxic appearance, no focal bacterial infection on physical exam and negative lab screening. If the authors defined the low risk differently, they re-classified infants to meet the criteria whenever possible.

There was no overlap of the 95% credible sets of the low and high risk groups for the infectious groups. The relative risk of the mean risks of each of the infections between the high and low risk groups is SBI 9.3, bacteraemia 9.8, and meningitis 6.5.
Hewson93

study type:
prospective cohort study
EL:2+
Country:
Australia
Aim:
To perform a multicentre follow-up study to determine if previously identified markers of serious illness in early infancy were robust and statistically reliable.
Setting, inclusion/exclusion:
This study was conducted from July 1991 to June 1992. This was a study on the clinical marks of serious illness in young infants aged 1-to 26 weeks presenting to the Emergency Departments of Royal Children’s Hospital and two general Melbourne metropolitan Hospitals for 12 months.
Rectal temperature was used in this study. Type of thermometer is not specified. The predictive values of temp. < 36.4 °C, > 38.0 °C and > 38.9 °C were explored. Exclusion criteria were not reported
Clinical markers:
13.

Drowsiness

  1. occasional
  2. frequent
  3. on examination
  4. any ( history or on exam)
14.

Decreased activity

15.

a. difficult breathing

b. moderate – severe chest wall recession

16.

a. pale on history

b. pallor on exam

17.

a. feeding 2/3–1/2

b. feeding < 1/2

18.

Urine output

19.

Vomits: > 5/24 hr

20.

Convulsion

21.

Bile-stained vomiting

22.

Respiratory grunt

23.

Lump > 2 cm

24.

Temp. (RT, type of thermometer not reported)

  1. 38.1–38.9 °C
  2. > 38.9 or < 36.4 °C
  3. > 38.1 or < 36.4 °C
Definition of serious illness:
Either having a serious investigation result (i.e. positive pathological bacterial culture from blood, urine, CSF, faeces, or a chest-x ray reported as showing consolidation in a febrile patient ) or by requiring significant treatment in hospital as supervised by independent staff (i.e. NG or IV fluid, parental antibiotics, O2 > 30% or surgery).
From 3806 assessments (mean age: 77 days. 62.4% were < 13 weeks) there were 312 infants assessed as being seriously ill (8.2%).
Table :The diagnostic values of the markers of serious illness for all infants from 0–26 weeks.
No.PPV (%)NPV (%)Relative riskSensitivity (%)Specificity (%)
Drowsiness
  1. occasional
  2. frequent
  3. on examination any ( history or on exam)

219
32
26
262

27.4
59.4
57.7
32.1

93.0
92.2
92.1
93.6

3.91
7.62
7.30
5.02

19.2
6.1
4.8
26.9

95.4
99.6
99.7
94.9
Decreased activity3745.992.25.885.499.4
  1. difficult breathing
  2. moderate – severe chest wall recession
484
84
10.7
40.5
92.2
92.5
1.37
5.4
16.7
10.9
87.6
98.6
  1. pale on history
  2. pallor on exam
134
63
32.1
49.2
92.7
92.5
4.40
6.56
13.8
9.9
97.4
99.1
  1. feeding 2/3–1/2
  2. feeding < 1/2
647
195
14.5
30.8
93.1
93.0
2.07
4.40
30.1
19.2
84.2
96.1
Urine output:< 4 wet nappies98
196
31.6
16.8
92.3
92.4
4.10
2.21
9.9
10.6
98.1
95.3
Convulsion3327.390.82.973.599.0
Bile-stained vomiting1747.190.85.123.199.6
Respiratory grunt4619.690.72.113.598.5
Lump > 2 cm18041.792.65.6431.995.8
Temp.
  1. 38.1–38.9 °C
  2. gt; 38.9 or < 36.4 °C
  3. > 38.1 or < 36.4 °C

252
101
353

29.0
41.6
32.6

92.2
91.7
93.0

3.62
5.13
4.71

17.5
10.1
27.6

95.8
98.6
94.4
Table :The cumulative diagnostic values of the markers of serious illness*.
Cumulative Sensitivity (%)Specificity (%)PPV (%)NPV (%)Relative risk
Drowsiness26.994.432.193.65.02
Pale on history or exam36.992.630.794.34.58
Difficult breathing50.097.719.194.83.67
Temp. > 38.1 or < 36.4 °C62.276.818.995.54.2
Lump82.573.522.197.79.61
Feeding < 1/283.971.821.397.89.68
> 5 vomits/24 hr87.368.520.198.211.2
< 4 wet nappies/24 hr87.968.220.198.311.8
  • excluding infants with inguinal hernia.
Data collection was not blind, randomised and didn’t report the measurements of reference standard before and after intervention. Control Group: not reported. No details of follow-up although this study was claimed as multicentre follow-up study. The sensitivity, specificity, positive predictive value and negative predictive value were used for statistical analysis but 95% CI did not report. The risk of bias on this study was likely to affect the result although the study related to infant with fever.
Nademi121

Study type
Prospective cohort study

EL:2+
Country:
UK.
Aim:
To assess the causes of fever and identify clinical and laboratory features suggesting serious disease in U.K.
Setting, inclusion/exclusion:
This study was conducted in August and October 1999
All patients presenting fever to the paediatric assessment units at Newcastle General Hospital. Children presenting to hospital with temperatures ≥38 °C were included and patients with a temp. < 38 °C were excluded.
Definition of serious illness: sepsis, meningitis, toxic shock syndrome, brain abscess, pneumonia, UTI, ischiorectal abscess, appendicitis.
Twenty two (16%) had already received antibiotics (usually amoxicillin) within last 24 h, including 8 serious illness.
Axillary temperature was measured routinely in children < 3yr; tympanic temperature in children > 3yr. Type of thermometer not specified.
One hundred and forty one children between 8 days and 16 years of age (mean age 3.3 years) 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.

Table :Comparison of sensitivity, specificity, PPV and NPV of all variables with 95% CI to detect serious illness (n = 41)
Sensitivity %Specificity %PPV %NPV %Relative risk
T> 39 °C.14 (3–25)82 (74–89)25 (7–42)70 (61–78)0.83
T> 39.5 °C.7 (0–15)93 (87–98)30 (1–58)71 (63–78)1.03
Poor feeding78 (65–90)43 (33–52)36 (25–45)83 (72–92)2.12
Vomiting59 (43–73)60 (50–69)38 (25–49)78 (68–87)1.73
Restlessness76 (62–88)43 (33–52)35 (25–45)81 (70–91)1.84
Petechial rash29 (15–43)98 (95–100086 (67–100)77 (69–84)3.74
WBC
> 15 00010 (0.6–18)95 (90–99044 (11–76)72 (64–79)2.44
> 20 00029 (15–43)93 (87–98)63 (41–84)76 (68–83)2.63
Weber98

Study type:
prospective cohort study
EL: 2+
Country:
Ethiopia, the Gambia. Papua New Guinea and the Philippines.
Aim:
To identify simple procedures for identifying infants with infection that need referral for treatment are therefore of major public health importance.
Setting, inclusion/exclusion:
At hospitals or outpatient clinics where large number of sick infants were seen from April 1978 to March 1979.
Rectal temperature for children < 5; oral temperature for > 5 years. Type of thermometer not reported.
At each study site, infants < 91 days of age seen consecutively for acute care with chief complaints indicating possible infection were eligible. This report only analyse the age group 0–59 days. Entry criteria were intended to include a wide spectrum of illness severity and to ensure that virtually all infants with serious infection would be included. Children with congenital heart disease and hypoxemia were excluded.
All infants underwent a standardized history and physical exam to assess the degree of signs and symptoms. All had and pulse oximetry. Infants with pre- specified symptoms associated with bacterial infection had lab evaluation that included blood culture, WBC, CXR (n = 1809). Specific criteria were used to identify infants for lumbar puncture (n = 401).
Definition of sepsis:
The growth of an unknown pathogen in cultures of blood.
Ranking of disease severity:
 Level 0: No abnormality
 Level 1: Mild hypoxemia (90%≤SaO2< 95%) or radiologic pneumonia.
 Level 2: Severe hypoxemia (SaO2< 90%) or bacteraemia or meningitis.
Death was separately analysed.
They recruited 3303 infants < 2 mo.
Level 0: No abnormality, n = 2585 (78.3%); level 1: Mild hypoxemia (90%≤SaO2< 95%) or radiologic pneumonia; n = 346 (10.5%); and level 2: Severe hypoxemia (SaO2< 90%) or bacteraemia or meningitis: n = 372 (11.3%); and 194 (5.9%) died. There were 120 cases of sepsis, 34 of meningitis and 259 of hypoxemia.

Table : Independently significant predictors of Ordinal Outcome 1 or 2 vs 0 in the three groups of general status, respiratory signs and meningitis signs, for the age group 0–6 days.
Signs or symptomPrevalence (%)
General status
 • Feeding ability reduced17*
 • No spontaneous movement11*
 • Temp. > 38 °C19*
 • Drowsy7
 • History of feeding problem16
 • Hx of change in activity21
 • Agitated4
 • Digital capillary refill11*
Respiratory signs
 • Lower chest wall indrawing14*
 • Res rate > 623*
 • Grunting2*
 • Cyanosis4*
Meningitis signs
 • Hx of convulsion4*
 • Bulging fontanel2
*: these signs comprise a restricted group that were considered for a more specific diagnostic algorithm, ( see next table)

Table :Sensitivity, specificity and negative likelihood ratio of different combination rules for predicting severe illness by ordinal outcome scale (0 vs 1+2)
0–59 days0–6 days7–59 days
Any sign of previous tableSn 87LR+1.89Sn 95LR+1.28Sn 85LR+1.98
Sp 54LR− 0.24Sp 26LR− 0.19Sp 57LR− 0.26
Any one sign from list of 9 marked in the previous tableSn 83LR+2.18Sn 92LR+1.31Sn 82LR+2.28
Sp 62LR− 0.27Sp 30LR− 0.27Sp 64LR− 0.28
Any sign omitting resp rate (n = 13)Sn 80LR+2.05Sn 94LR+1.32Sn 78LR+2.11
Sp 61LR− 0.33Sp 29LR− 0.21Sp 63LR− 0.35
Feeding ability: reduced or lower chest indrawing or history of convulsion (n = 3, most predictive signs only)Sn 60LR+3.53Sn 80LR+1.60Sn 56LR+3.73
Sp 83LR− 0.48Sp 50LR− 0.40Sp 85LR− 0.52
Any 1 sign from general status + I from other groupSn 51LR+3.92Sn 65LR+1.76Sn 48LR+4.00
Sp 87LR− 0.56Sp 63LR− 0.56Sp 88LR− 0.59
Any 2 signsSn 69LR+3.00Sn 87LR+1.47Sn 66LR+3.14
Sp 77LR− 0.40Sp 41LR− 0.32Sp 79LR− 0.43
Any 1 sign if wt < 3kg or any 2 signs if wt> 3kgSn 72LR+2.88Sn 91LR+1.36Sn 68LR+3.09
Sp 75LR− 0.37Sp 33LR− 0.27Sp 78LR− 0.41
Fever (temp.> 38 °C) and any other signSn 25LR+2.78Sn 21LR+1.31Sn 26LR+3.25
Sp 91LR− 0.82Sp 84LR− 0.94Sp 92LR− 0.80
*: Sn: sensitivity, Sp: specificity, LR+: positive likelihood ration; LR−: negative likelihood ratio.

Table :Association of clinical signs with sepsis, meningitis, hypoxemia and death. OR adjusted for place of study, weight and age.
SepsisMeningitis
Prevalence (%)OR95% CIOR95% CI
Hx of cough75--------
Hx of fast breathing35--------
Hx of change in level of activity213.62.5–5.16.43.1–13.5
Hx of change of crying381.91.4–2.72.11.1–4.1
Hx of convulsion44.22.6–7.012.26.2–23.9
Hx of feeding problem153.92.6–5.76.03.0–12.4
Lower chest wall indrawing141.51.0–2.2----
Nasal flaring41.60.8–2.9----
Grunting32.81.5–5.13.71.3–10.1
Crepitations171.30.9–2.0----
Wheeze110.60.3–1.2----
Drowsy/unconscious73.02.0–4.74.62.2–9.6
Agitated52.41.5–4.03.81.7–8.4
Lethargy162.31.6–3.32.41.2–4.7
Feeding ability reduced155.13.4–7.78.13.7–17.9
No spontaneous movement103.02.0–4.63.61.7–7.5
Consolability: continues to cry/fuss42.91.8–4.83.41.3–8.6
Central cyanosis32.41.3–4.32.00.6–6.5
Dehydration71.10.7–1.9----
Digital capillary refill 2+s112.21.5–3.31.70.8–3.4
Umbilical discharge41.10.5–2.3----
Bulging fontanel210.05.6–18.021.410.0–45.8
Resp rate < 40191.20.8–1.91.30.6–3.0
Resp rate ≥60232.21.5–3.12.01.0–4.1
Temp. < 35.523.71.8–7.34.20.8–22.5
Temp.≥ 38173.62.6–5.111.85.7–24.6
Hypoxemia82.31.5–3.71.70.7–4.2
Invasive bacterial infection4--------
Meningitis1--------
HypoxemiaDeath
Prevalence (%)OR95% CIOR95% CI
Hx of cough751.51.1–2.0----
Hx of fast breathing353.62.7–4.7----
Hx of change in level of activity213.22.5–4.23.72.7–5.1
Hx of change of crying381.71.3–2.11.00.8–1.4
Hx of convulsion41.50.9–2.65.33.4–8.3
Hx of feeding problem152.92.2–3.94.63.3–6.4
Lower chest wall indrawing146.44.9–8.42.82.0–3.9
Nasal flaring46.84.5–10.13.82.5–5.9
Grunting34.52.8–7.35.13.1–8.3
Crepitations179.57.1–12.71.91.3–2.8
Wheeze112.21.5–3.10.90.6–1.5
Drowsy/unconscious76.14.4–8.48.05.7–11.2
Agitated53.12.0–4.71.30.8–2.2
Lethargy163.82.9–5.04.53.3–6.1
Feeding ability reduced157.95.8–10.78.96.1–13.0
No spontaneous movement105.33.9–7.17.75.6–10.7
Consolability: continues to cry/fuss44.02.5–6.24.73.0–7.3
Central cyanosis315.09.9–22.65.73.6–8.8
Dehydration7----1.81.2–2.6
Digital capillary refill 2+s112.71.9–3.73.42.4–4.6
Umbilical discharge4----1.70.9–3.0
Bulging fontanel2----5.52.9–10.4
Resp rate < 40191.10.7–1.71.71.2–2.5
Resp rate ≥60234.53.3–6.22.31.6–3.3
Temp. < 35.523.21.9–5.43.11.8–5.3
Temp. ≥ 38172.41.7–3.22.31.7–3.2
Hypoxemia8----4.53.0–6.7
Invasive bacterial infection4----5.23.3–8.2
Meningitis1----11.05.1–23.5
Table :Association of clinical signs with the age group 0–6 days. OR adjusted for the place of study.
Age group 0–6 days
Outcome: level 1 or 2 (cf.0)Outcome: level 2 (cf.0 or 1)
Prevalence (%)OR95% CIOR95% CI
Hx of cough181.90.8–4.30.90.3–2.5
Hx of fast breathing382.51.5–4.31.91.1–3.5
Hx of change in level of activity311.40.8–2.41.60.8–3.0
Hx of change of crying301.30.7–2.11.60.9–2.9
Hx of convulsion91.00.4–2.41.00.4, 2.5
Hx of feeding problem481.91.1–3.43.61.7, 7.6
Hx of diarrhoea110.40.2–0.90.30.1, 1.0
Lower chest wall indrawing201.91.0–3.62.41.2–4.7
Nasal flaring121.60.7–3.42.10.9–4.8
Grunting91.90.8–4.51.60.6–3.9
Crepitations67.22.0–26.33.31.1–9.3
Wheeze50.60.2–1.90.80.2–3.1
Drowsy/unconscious213.72.0–6.93.41.8–6.5
Agitated71.20.5–3.31.50.5–4.3
Lethargy401.50.9–2.52.11.2–3.9
Feeding ability reduced575.02.5–9.94.62.0–10.7
No spontaneous movement371.81.1–3.12.41.3–4.3
Consolability: continues to cry/fuss121.80.7–4.31.50.7–3.7
Central cyanosis93.51.4–8.44.01.7–9.3
Dehydration101.20.5–2.71.60.7–3.7
Digital capillary refill 2+s232.91.6–5.21.70.9–3.2
Skin rash90.30.1–1.70.50.0–4.3
Umbilical discharge171.40.7–2.81.10.5–2.6
Bulging fontanel31.50.4–6.31.60.4–6.9
Eye discharge101.70.7–4.21.70.5–5.2
Jaundice450.70.4–1.20.80.4–1.4
Resp rate < 40211.80.9–3.53.41.5–7.7
Resp rate ≥60371.81.0–3.32.21.1–4.6
Temp. < 35.5152.00.9–4.22.10.9–4.8
Temp. ≥ 38221.00.5–1.91.10.5–2.2
Table :Association of clinical signs with the age group 7–60 days. OR adjusted for the place of study and weight.
Age group 7–60 days
Outcome: level 1 or 2 (cf.0)Outcome: level 2 (cf.0 or 1)
Prevalence (%)OR95% CIOR95% CI
Hx of cough761.10.9–1.40.70.6–0.9
Hx of fast breathing342.62.2–3.22.52.0–3.3
Hx of change in level of activity203.62.9–4.55.03.7–6.6
Hx of change of crying371.31.1–1.61.41.1–1.9
Hx of convulsion44.02.7–6.04.93.1–7.6
Hx of feeding problem122.92.3–3.73.92.9–5.2
Hx of diarrhoea170.70.6–1.00.80.6–1.1
Lower chest wall indrawing135.64.4–7.03.92.9–5.1
Nasal flaring46.94.5–10.84.52.9–6.9
Grunting28.14.4–15.15.73.2–10.2
Crepitations167.35.8–9.24.73.6–6.2
Wheeze92.31.7–3.11.30.9–1.9
Drowsy/unconscious65.84.1–8.17.04.9–9.9
Agitated42.91.9–4.32.91.8–4.6
Lethargy153.12.4–3.94.03.0–5.2
Feeding ability reduced136.65.1–8.79.46.9–12.8
No spontaneous movement95.34.0–7.06.44.7–8.7
Consolability: continues to cry/fuss44.22.7–6.75.23.2–8.3
Central cyanosis310.86.5–17.812.27.6–19.5
Dehydration61.30.9–1.81.51.0–2.2
Digital capillary refill 2+s102.51.9–3.33.32.4–4.6
Skin rash90.80.6–1.10.90.6–1.4
Umbilical discharge51.00.6–1.51.10.6–2.0
Bulging fontanel14.32.3–8.25.32.7–10.5
Eye discharge---------------------------------------------
Jaundice---------------------------------------------
Resp rate < 40180.90.7–1.21.10.8–1.6
Resp rate ≥60223.83.0–4.63.82.9–5.0
Temp. < 35.522.41.2–4.73.41.7–6.8
Temp. ≥ 38152.72.2–3.43.4-2.6–4.5
Ronfani99

Study type:
prospective cohort study

EL: 2+
Country:
Brazil
Aim:
To estimate sensitivity, specificity, and predictive value of different signs of severe bacterial infection (SBI) in neonates upon presentation to an emergency and neonatology department
Setting, inclusion/exclusion
All neonates (< 28 days) presenting at hospital and admitted to the emergency and neonatology department of Instituto Materno Infantil de Pernambuco from1 March 1995 to 29 Feb 1996 infants with ‘birth-related problems’ were excluded. Number not reported.
Data on age, sex, type of delivery, birthweight, gestational age, weight and length at admission, type of feeding collected at admission

Signs reported by mother/carer:
  • Difficult breathing
  • Diarrhoea
  • Cough
  • Vomiting
  • Duration of all the above
Signs reported by doctor: Lab:
  • Complete blood count
  • Blood culture
  • Chest x-ray, CSF microscopy and culture, and urine culture only when CNS infections and UTI were suspected
Designation of infection status by doctor at discharge (reference standard):
  • SBI, included sepsis, meningitis, severe diarrhoea, lower respiratory tract infection, UTI, severe omphalitis
  • Probable SBI
  • Other disease
They recruited 83 (42 male, 39 female) in total. SBI = 41 (49.4%); probable SBI = 9 (10.8%); other disease = 33 (39.8%)

Most common diagnosis:
Among SBI:
  • pneumonia, n = 22
  • sepsis, n = 10
  • meningitis, n = 4
  • conjunctivitis, n = 4
Among other diseases:
  • jaundice, n = 9
  • mild diarrhoea, n = 6
  • convulsions, n = 4
Signs most frequently reported by mother/carer:
  • Difficult breathing, 32%
  • Diarrhoea, 26%
  • Fever, 19%
  • Cough, 19%
  • Vomiting, 19%
  • Jaundice, 16%
  • Cyanosis, 14%
  • Not feeding well, 11%
Signs most frequently observed by doctor:
  • Severe chest indrawing, 46%
  • Fast breathing (60+ breaths/minute), 40%
  • Jaundice, 29%
  • ‘Not looking well’, 25%
  • pallor, 23%
  • hypotonia, 22%
  • cyanosis, 19%
  • dehydration, 18%
Sensitivity, specificity and predictive values of best performing signs for SBI
PPV (%)*Sensitivity (%)Specificity (%)
By mothers
 Difficult breathing784282
Fever10033100
 Diarrhoea733282
 Cough882894
 Vomiting752488
By doctors
 S. chest indrawing765873
 Fast breathing795278
 Not looking well954097
*No negative predictive value was reported.

Fever and ‘not looking well’ were the only two signs independently associated with SBI:
Fever RR = 6.47, 95% CI 2.07 to 20.23, P< 0.001
Not looking well RR = 7.17, 95% CI 2.44 to 21.02, P< 0.001

Best sensitivity (74%) found with signs in parallel:
Doctor observed severe chest indrawing or fast breathing or ‘not looking well’ (specificity 67%, PPV 77%)

6 deaths: 4 from SBI group (2 sepsis, 1 pneumonia, 1 meningitis), and 2 from ‘other disease’ group (1 severe rhesus isoimmune haemolytic disease, 1 adrenogenital syndrome)

Table :Sensitivity, specificity and predictive values of best performing signs for pneumonia
PPV (%)*Sensitivity (%)Specificity (%)
By mothers
 Difficult breathing637784
 Cough886497
Fever564389
By doctors
 S. chest indrawing457766
 Fast breathing395967
 Not looking well292775
*No negative predictive value was reported.
Hiew97

Study type
prospective cohort study EL:2-
Country:
Singapore
Aim:
To identify the clinical features and haematological indices of bacterial infection amongst young infants and to determine retrospectively the findings significantly associated with positive bacterial cultures.
Setting, inclusion/exclusion:
July 1989–Febuary 1991, infants ≤3 mo with suspected bacterial infection and admitted to the Paediatric Department, Tan Tock Seng Hospital. Patents already on antibiotics before evaluation were excluded.
Evaluations were:
  • General features
  • Cardiovascular system
  • Respiratory system
  • Central nervous system
  • Gastrointestinal system
  • Skin
Lab test:
  • Total white blood cell count
  • Absolute neutrophil count
  • Platelet count
  • Immature to total neutrophil ratio (I/T ratio)
  • Nitroblue Tetrazolium test (NBT)
  • CXR
  • Blood culturex2
  • Urine culturex2
  • CSF FEME and culture (only with suspected meningitis)
  • Skin/umbilical cord culture
Designation of infection status: Proven bacterial infection: infants with positive bacterial cultures of a pathogenic organism from the blood, CSF, urine, sputum, pustules or umbilicus.
The recruited 100 infants with mean age of 46wk (SD:3.06)., 60 male & 40 female. The most common clinical features among the 100 infants are fever (n = 85), lethargy (n = 44), hepatomegaly (n = 39), poor feeding (n = 35), irritability (n = 30), splenomegaly (n = 23), skin mottling (n = 17), diarrhoea (n = 15), respiratory distress (n = 12), hypotonia (n = 12).

Table :Most common clinical features of bacterial infections in young infants. Positive/Total evaluations 30/100 (30%).
FeatureInfected (n)Non-infected (n)PPV (%)Sensitivity (%)P value
Respiratory distress755823< 0.01
Cyanosis655520< 0.05
Grunting574217Ns
Slenomegaly9143930Ns
Hepatomegaly15243850Ns
Fits473613Ns
Mottled skin6113520Ns
Hypotonia483313Ns
Diarrhoea5103317Ns
Fever28573393Ns
Lethargy13313043Ns
Poor feeding10252933Ns
Irritability7232323Ns
Vomiting210177Ns
Table :Haematological findings in young infants with bacterial infections Positive/Total evaluations 30/100 (30%).
Total +ve tests+ve tests & +ve culture*+ve tests & − ve culture*PPV (%)Sensitivity (%)Specificity (%)NPV (%)P value
Abnormal WBC218133826.781.472Ns
Absolute neutrophil counts55163929534468Ns
Abnormal platelet counts514203.39469Ns
Raised I/T ratio1541126.7138469.4Ns
Raised NBT134930.813.387.170.1Ns
Raised CRP66254137.983.341.485.3< 0.01
Raised ESR54213338.97052.980.4< 0.05
*: duplication in the report (both were reported as +ve tests & −ve culture), use the provided numbers to deduce the correct column title.

Table : Results of combination of some haematological tests
Total +ve tests+ve tests & +ve culturePPV (%)Sensitivity (%)Specificity (%)NPV (%)P value
CRP& ESR431842606478< 0.05
CRP&WBC16850278984< 0.05
CRP& neutrophil counts391436476474Ns
ESR& WBCl counts15853279074< 0.05
ESR& neutrophil counts331339437175ns
WBC & neutrophil counts19737238372Ns
No report on the number of withdrawals, exclusions and drop outs. PPV is reported as positive predictive accuracy (if clinical feature is present or test abnormal, what is the probability of infection being present? )

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.

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