Menu Search Donate
Guideline identity image

Microbiological testing in PICU

This document is only valid for the day on which it is accessed. Please read our disclaimer.

This is a guideline for the taking of microbiological samples in PICU to diagnose or exclude infection.

The diagnosis of infection requires:

  • Ruling out non-infectious causes of fever
  • Localizing the site of infection
  • Taking the correct specimens
  • Interpreting microbiological findings - can be more difficult in PICU due to high colonisation rates, prior antibiotic use, unusual organisms, polymicrobial results and knowing the significance of normally low virulence organisms in immunosuppressed patients.

History, risk factor assessment, and clinical examination will often reveal potential sources of infection and guide further investigations.

Daily Review

It is important to review daily all indwelling catheters to both assess the need for them, duration of insertion and to check for evidence of infection around insertion sites.

Risk of infection increases with length of insertion and there should be a low threshold to remove/replace lines or catheters that have been in for longer than several days if there is a suspicion of infection related to the device.

Check the nature of tracheal secretions, urine, and any drain discharge. Check the skin integrity plus all surgical wounds and drains.

In some patients the use of nonspecific markers of infection may be useful:

  • WBC Indices - Total WBC count and even WBC differential may not be helpful in differentiating infection from inflammation. More sensitive is the WBC left shift index. This is the ratio of immature neutrophil forms to total neutrophils and is normally < 0.15 ( <0.2 in a neonate). This can be elevated postoperatively.
  • C-Reactive Protein (CRP) - Accurately reflects inflammation but not necessarily infection. Cut-off values for viral versus bacterial infection are not clear. Serial values may be useful for tracking clinical progress. Some evidence that this is useful in neonatal sepsis.
  • Procalcitonin - Similar limitations to CRP but rises and falls quicker. Multiple studies in adult ICU have shown procalcitonin to be useful in guiding antibiotic treatment especially when done serially. This test is not currently available at ADHB but is at MMH as a send away.

Microbiological Sampling

Blood Cultures

Always take blood cultures in suspected sepsis before commencing antibiotics. Take as soon as possible after fever develops to maximise chance of positive result. Obtain 2 sets of blood cultures - A paediatric set is 1 aerobic bottle, an adult set is 2 bottles (1 aerobic and 1 anaerobic). Either take the 2 sets at the same time from 2 different sites or take each set at a different time. Ideally take from a clean peripheral site.

  • If endocarditis suspected obtain 2 sets separated by at least an hour (ideally 3 sets).
  • Prolongation of cultures is necessary for fungi, brucellosis and endocarditis and the laboratory should be notified if these are suspected.

If a central line is in situ take a blood culture from the line PLUS a peripheral culture. 30-50% of patients presenting with severe sepsis will have positive blood cultures. Most patients with bacterial pneumonia will not have a positive blood culture.

Repeat cultures:

  • With each NEW suspected episode of bacteremia.
  • If initial blood cultures are negative but infection is still strongly suspected. Additional specialist blood culture media may be required if mycobacteria are strongly suspected. Anaerobic cultures can be used but only in specific clinical scenarios. Discuss with ID.
  • If the patient is still septic after 48-72 hours of appropriate treatment for a previous positive blood culture.
  • If a blood culture is positive but the clinical significance is doubtful then repeat the culture prior to starting or altering antibiotics.

Good technique is vital in taking blood cultures to limit contamination:

  • Wash hands and wear gloves (non sterile).
  • Skin prep with chlorhexidene 2% + alcohol 70%, swab concentrically starting at the centre and allow to dry.
  • If drawing blood from a line, cleanse the connector with skin prep as above.
  • Swab stopper of culture bottle.
  • DO NOT change needles prior to placing sample in culture bottle.
  • For children < 30kg take 0.25ml/kg (minimum 1ml) to a maximum of 5ml in paediatric blood culture bottle.
  • For adults and children > 30kg take 20ml and place 10ml in each adult blood culture bottle.


Three possible pathogens in patients with new onset diarrhoea are rotavirus, Clostridium difficile, and norovirus.


  • Predominantly a disease of infants and young children.
  • In immunocompetent patients lasts average of 5 days.
  • Can cause chronic infection in immunosuppressed.
  • Test once only within 30 days.
  • Do not test adults.
  • Do not check for clearance - cease isolation when diarrhoea settles.
  • Discuss with infection control nurse

Clostridium difficile Toxin

  • Test once only within a 14 day period.
  • Patients require isolation.
  • Do not test for clearance. Patients with well formed stool do not need isolating.
  • Testing involves detection of C. difficile and detection of toxin - need both to have clinical disease.
  • Standard treatment is oral metronidazole 10-14 days.
  • High rate of relapse following treatment - retreat without retesting.
  • Discuss with ID team.


  • Highly contagious and responsible for sporadic outbreaks of diarrhoea in hospitals.
  • Can only detect with PCR.
  • Discuss with ID and Virology BEFORE sending sample for testing.

Respiratory Tract Infection

Nasopharyngeal Aspirate

Respiratory Panel PCR Mon-Fri before 10am should ensure same day result. Includes the following:

  • Influenza A and B
  • Adenovirus
  • RSV
  • Parainfluenza 1, 2, 3

Sample can also be processed for other organisms on request - including Bordetella pertussis PCR (performed twice weekly), enterovirus in neonates , or cornavirus, bocavirus, rhinovirus and metapneumovirus. Not suitable for microscopy and culture.

Tracheal aspirate

  • Derived from suctioning ETT.
  • Not suitable for Respiratory Viruses.
  • Routine is microscopy and culture primarily for bacterial pathogens.
  • Routine test reports Appearance, Gram Stain, and Culture results
  • If many epithelial cells specimen likely from oropharynx and not cultured.
  • Gram Stain reported semiquantitatively as occasional, moderate, or large numbers of organisms.
  • Specialised tests = Legionella culture, Acid fast bacilli culture, Pneumocystis jiroveci stain, and fungal culture.

Do not repeat unless patient thought to have developed a new respiratory tract infection.

Bronchoalveolar Lavage (BAL)

  • Performed blindly in PICU or under bronchoscopic vision in OR.
  • In PICU BAL avoids problems of colonisation due to ETT.
  • Special kits have been pre-prepared in PICU for blind BAL.
  • Use 1ml/kg sterile saline to a maximum of 10ml.
  • Usually done in immunosuppressed or chronically infected patients.
  • As well as all tests for tracheal aspirate also sent for Cryptococcus culture, TB stain, and fungal stain.
  • Sample sent to virology for PCR of Respiratory Viruses (10 panel multiplex as above). Atypical respiratory pathogens such as Legionella and pertussis can also be tested by PCR on BAL on request.
  • Additionally parvovirus, tuberculosis stain (+/- PCR) can be requested.

Urinary antigen tests

are available for legionella and Streptococcus pneumoniae. The legionella test may be useful in severe community acquired pneumonia. There is no role for urinary pneumococcal antigen test in children due to the high rate of false positives.

Pleural fluid

As well as routine microscopy and culture, pneumococcal antigen test on pleural fluid can be requested as this is very specific when used on sterile fluids (CSF and pleural fluid).

Urinary Tract Infection

Bag urine sampling is a screening test and these samples should not be sent for culture. Dipstick for nitrates and leucocytes and if either positive obtain clean sample. Catheter urine samples should be obtained in suspected UTI and sent for microscopy and culture not just dipstick.

  • Pure growth >108, WBC > 100, and single organism equate to UTI.
  • Equivocal results should be repeated.
  • Do not repeat unless concern over clinical symptoms.

CNS Infection

CSF analysis

If there is concern about the intrathecal pressure, measure the opening pressure with a manometer.

Collect 3 tubes. Always send for:

  • Cell Count and WBC differential
  • Glucose
  • Protein
  • Gram Stain
  • Culture - bacterial +/- fungal

In cases of suspected infection ALWAYS send for PCR panel. PCR panel includes:

  • HSV types 1 and 2
  • VZV
  • Enterovirus
  • Parechovirus
  • S. pneumoniae
  • N. meningitidis

For other tests for infectious causes of meningoencephalitis please refer to the Starship Encephalitis guideline.

Did you find this information helpful?

Document Control

  • Date last published: 08 August 2012
  • Document type: Clinical Guideline
  • Services responsible: Paediatric Intensive Care Unit
  • Editor: John Beca
  • Review frequency: 2 years