September 2024
THE RATIONAL USE OF BLOOD CULTURES IN ADULT PATIENTS
LABUPDATE no. 54
Blood cultures are considered the “gold standard” to diagnose bloodstream infections, a major cause of
death worldwide.1 Ninety percent or more of blood cultures do not grow any organisms, suggesting many of
these were likely not indicated.1 In addition, 30−50% of positive blood cultures recover contaminants, which
can result in unnecessary antibiotic use, prolonged hospital stays, additional testing, unnecessary removal of
vascular catheters, and increased health care costs.1 The rational use of blood cultures is thus important. To
ensure optimal blood culture utilisation, the following needs to be considered:
• When are primary blood cultures indicated?
• How many sets do I need to send in/request?
• Is the timing of sampling important?
• Is a follow-up blood culture indicated?
WHEN ARE PRIMARY BLOOD CULTURES INDICATED?
Blood cultures should be taken when there is a high likelihood of bacteraemia or for syndromes with
a moderate likelihood of bacteraemia when cultures from the primary source of infection are not
readily available or will be delayed. Blood cultures are also indicated when the patient is at high risk of
endovascular infection as is the case in patients with prosthetic heart valves, vascular grafts, an implantable
defibrillator/pacemaker, valvular heart disease or in injection drug users. Blood cultures should also be taken
when the results are anticipated to impact patient management, for example, in severe non-purulent
cellulitis in an immuno-compromised patient or when a resistant organism is suspected.
Clinical judgement should always be used when deciding whether blood cultures are indicated, especially
in the immune-compromised patient or patients with co-morbidities such as diabetes, and end-stage liver
and end-stage renal disease. Figure 1 summarises the clinical scenarios where blood cultures may be of
value in the non-neutropaenic patient.
Are blood cultures indicated in my patient?
BC likely to be of intermediate
diagnostic value/yield
Send two sets of pheripheral BC if:
At risk for endovascular infection
OR
Primary site of infection not readily
available
for culture before starting antibiotics
OR
BC results will influence management
• Acute pyelonephritis
• Cholangitis
• Non-vascular shunt infections
• Prosthetic vertebral OM
• Severe CAP
• Cellulitis in patient with co-morbidities
• VAP
BC will have high diagnostic value/
yield in the following cases:
Send two sets of peripheral BC
BC generally not recommended
in immuno-competent patient
without significant comrbidities
• Severe sepsis/septic shock
• Infective endocarditis/
endovascular infection
• CLABSI
• Discitis/native vertebral OM
• Non-traumatic native septic arthritis
• Epidural abscess
• Meningitis
• Ventiriculoatrial shunt infections
BC likely to be of low diagnostic
value/yield
• Isolated fever/leukocytosis
• Non-severe cellulitis
• Lower UTI: cystitis, prostatitis
• Non-severe CAP or HAP
• Post-operative fever within
48 hours of surgery
FIGURE 1: CLINICAL SCENARIOS WHERE BLOOD CULTURES MAY BE OF VALUE IN THE NON-NEUTROPAENIC PATIENT 2,3
* CLABSI – central line associated blood stream infection, OM – osteomyelitis, CAP – community-acquired pneumonia, HAP – hospital-acquired pneumonia, VAP – ventilator-
associated pneumonia, UTI – urinary tract infection, BC – blood culture
ampath.co.za
HOW MANY SETS OF BLOOD CULTURES DO I NEED TO SEND IN/ REQUEST?
The two most important factors that impact the performance of blood cultures are proper skin antisepsis
and the volume of blood cultured.4 Various studies have demonstrated that single blood cultures are
inadequate to detect yeast and bacteria in the blood.1 Single sets can miss as many as 10−40% of
pathogens.1
Another disadvantage of taking a single blood culture is the inability to distinguish culture contamination
from true bacteraemia.1 Current guidelines recommend that at least two sets of blood cultures should be
collected for an adult patient, with each set consisting of 10 mL of blood inoculated into an aerobic and
anaerobic blood culture bottle (thus 40 mL of blood collected in total). The yield of detected pathogens
increases with each additional set of blood cultures collected: 61.4% with one set, 78.2% with two sets, and
93.1% with three sets (Figure 2).4
FIGURE 2: THE YIELD OF PATHOGENS FROM BLOOD CULTURE IN RELATION TO THE VOLUME OF BLOOD CULTURED4
IS THE TIMING OF SAMPLING IMPORTANT?
With regard to the timing of blood cultures, the only important aspect is to take the blood cultures before
the administration of antibiotics. Neither spacing of the blood cultures over a 24-hour period nor the timing
of the blood cultures close to a fever spike has any influence on the positivity rate of blood cultures.1
IS A FOLLOW-UP BLOOD CULTURE INDICATED?
When considering when to repeat blood cultures, three factors need to be considered:
• The identified pathogen
• The source of infection
• The clinical response and source control
Repeat blood cultures are indicated in the following cases:1-3
• To document the clearance of Staphylococcus aureus bactereamia
• To document the clearance of Staphylococcus lugdunensis bacteraemia
• To document the clearance of candidaemia
• For any organism when there is suspicion of infective endocarditis or endovascular infection
• Where persistent bacteraemia is suspected
• To distinguish contamination from true bacteraemia.
61.4% yield of pathogens
78.2% yield of pathogens
93.1% yield of pathogens
KEY POINTS
• The volume of blood collected for blood culture is the single most important determinant of the ability to
detect pathogens.
• At least two to three sets of blood cultures should be collected – with each set consisting of 10 mL of blood
inoculated into an aerobic and anaerobic blood culture bottle (thus 40–60 mL of blood collected in total).
• Blood cultures should ideally be collected before the administration of antibiotics.
• Blood cultures should not be spaced over a 24-hour period.
• Blood cultures need not be collected during a fever spike.
REFERENCES
1. Fabre V., Carroll K.C., Cosgrove S.E. 2022. Blood culture utilization in the hospital setting: A call for
diagnostic stewardship. Journal of Clinical Microbiology 16;60(3):e01005-21.
2. Papavarnavas N.S., Brink A.J., Dlamini S., Wasserman S., Whitelaw A., Ntusi N.A., Mendelson M. 2022.
Practice update to optimise the performance and interpretation of blood cultures: 2022. South African
Medical Journal. 112(6):397−402.
3. Fabre V., Sharara S.L., Salinas A.B., Carroll K.C., Desai S., Cosgrove S.E. 2020. Does this patient need blood
cultures? A scoping review of indications for blood cultures in adult nonneutropenic inpatients. Clinical
Infectious Diseases. 71(5):1339−47.
4. Snyder J.W. 2015. Blood cultures: The importance of meeting pre-analytical requirements in reducing
contamination, optimizing sensitivity of detection, and clinical relevance. Clinical Microbiology
Newsletter. 37(7):53−7.