Examination of Blood Cultures

Depending on the number of bacteria in the bloodstream and other factors, the first appearance of growth may take place after as little as a few hours or only after a longer period of 1 or a few days or occasionally 1 or weeks and medium incubated at 37°C for up to 6 weeks for Brucella.
As it is desired to detect any positive culture as early as possible and yet not miss any late-growing culture the ideal procedure is to make repeated examinations of the culture bottle at different timer during its incubation from 4 h to 2 weeks longer. if. however, many requests for blood culture are received each day, the amount of labor required for making more than a few examinations on each bottle may be excessive . Moreover, every opening of a bottle for examination affords an extra opportunity for contamination, and overlapping examinations on successive batches of bottles on the same day increase the chance of confusion and mistakes.
- Trials of different schedules suggest that a minimal, reasonably effective procedure is the making of just two examinations on each bottle, the first after incubation at 37°C overnight (i.e. for 12-24 h), the second after 4-7 days. Only very few cultures negative at 4 days become positive on longer incubation. When a report is required urgently on a suspected septicemia, an additional examination may be made after only 4 h incubation.
Positive findings at that stage are available for earlier reporting to the clinician and are less likely to be due to the growth of scanty contaminants than those obtained after longer incubation. The negative cultures, which constitute the majority of all those received, have to be examined again at 18-24 h and 4-7 days. requiring a total of three examinations. The extra work of the early. First-day examination may be minimized by confining it to the making of subcultures and omitting the more time-consuming examination by Gram film.
It is recommended that on each subsequent occasion of examination, the blood culture should be examined both by a Gram film and by subculturing on two plates of blood agar, the one incubated in air plus 5-10% C02, the other anaerobically in nitrogen/hydrogen plus 5-10% C02. If the presence of Meningococcus or Haemophilus seems likely, e.g. in suspected septicemia meningitis, a heated-blood (choco¬late) agar plate may be substituted for the aerobic plain blood agar plate. Gram filming is particularly valuable when the first examination is made only after overnight incubation, i.e. after 12-24 h, for when bacteria are seen to be present in the film, a preliminary report may be phoned to the physician at once, without the need to wait another day to see growth of the bacteria in the subcultures.
The procedures of filming and subculture must be performed in such a way as to minimize the opportunities for contamination of the blood culture with bacteria from the air, fingers or equipment. The bacteriologist should gently mix the contents of the bottle without wetting the inside of the cap, then remove the cap, withdraw 0.5-1ml of the culture fluid into a sterile Pasteur pipette, and at once replace the cap. (Contami¬nation of the bottle is less likely to occur during one dip with a pipette than during three with an inoculating loop.) He should then let a large drop of the fluid fall from the pipette on to each subculture plate and finally spread a drop on a slide for Gram staining. Lastly, he should spread the drops on the plates over large "well" areas and streak them out on the remainder.
Gram film. The Gram film should be examined as soon as the subcultures have been set up, and positive findings reported at once to the physician so that the morphological type of the organism present may serve as a guide to his initial choice of antibiotic. The finding will also indicate to the bacteriologist what further subcultures and tests should be seeded from the positive bottle. Two extra plates should be seeded confluent and the appropriate antibiotic disks applied to them for sensitivity testing on aerobic and anaerobic incubation. Optochin and bacitracin disks may be used on the subculture plates to identify organisms resembling Pneumococci and Streptococci. A MacConkey plate may be seeded to assist in the identification of Gram-negative bacilli, a DNase test plate for staphylococci, an aesculin-bile plate for diphtheroid bacilli (to exclude Listeria, which is positive), and a Nagier plate for bacilli resem Wing Clostridium perfringens.
A possible cause of error must be borne in mind when reading the Gram film. Some media, particularly some batches of meat pieces incorporated in cooked meat medium, may be contaminated with saprophytic bacteria, which, although they have been killed in the sterilization procedure , may be sufficiently numerous to be seen in the film, if only one morphological type is seen, its presence may lead to the making of a false provisional report, but if a mixture of types is seen, their origin as contaminants will be suspected. The error will be recognized next day when the subcultures will be found to be negative.
Examination of subcultures. The plates should be inspected after incubation for 18-24 h, when any growth should be examined in a Gram film and by the setting up of appropriate identifying and sensitivity tests. Plates showing what is possibly very slight growth and plates seeded from film-positive bottles but still devoid of growth should be re-incubated for up to three more days before being discarded as negative.
Reports. Relevant positive findings should be reported by phone to the physician as soon as is practicable and, if requested. advice should be given on the choice of antibiotic therapy. A written report should be issued when the exam¬inations made after the first 12-24 h of incu¬bation have been completed. For positive cultures, the species and drug sensitivities should be reported, and for negative cultures, the absence of growth and a note that a further report will be issued if growth appears on longer incubation.
When Staphylococcus albus or a commensal type of Diphtheroid bacillus is found and no predisposition to opportunistic infection is known, the report should state that the organism is probably a contaminant from the patient’s skin and omit the drug sensitivity findings. If, however, the patient has suspected endocarditis, an artificial heart valve, an intravenous catheter or a cerebrospinal fluid shunt, or has had heart surgery, or is immunosuppressed or on cytotoxic drugs, the finding of the organism and its sensi¬tivities should be reported, but with the caution that it may possibly be a contaminant from the patient’s skin.
The isolation of a bacterium from the blood of a patient with a local infection such as wound sepsis or pneumonia is valuable firstly in indi¬cating the urgent need for antibacterial therapy, secondly in revealing the species of bacterium against which therapy should be directed. and thirdly in providing a culture for the performance of in-vitro drug sensitivity tests. Culture of a specimen from a local site of infection, e.g. a wound swab or sputum specimen, often yields a mixture of contaminating commensal bacteria and potential pathogens, the clinical significance of which is unclear, but the demonstration of one of these species in the blood indicates that it at least is an invader of clinical significance. If, for instance, a Pneumococcus is cultured from sputum from a patient with suspected pneumonia, it may have been derived from its site of commensal carriage in the throat and not from the lungs, which may instead be infected with an undetected virus or Mycoplasma. But if the Pneumococcus is cultured from the blood, it may be concluded that it is probably present as a pathogen in the lungs and that the patient certainly requires anti-pneumococcal therapy.
Contamination:
A difficulty in interpreting results arises from the liability of the specimen to be contaminated with organisms that are commensals or common contaminants on the skin. Staphylococci albus are the commonest contaminants, but Dphtberoid bacilli, Coliform bacilli, Anthracoid bacilli, Clostridium perfrin¬gens, candida and other organisms are some¬times present. and the finding of one of these organisms in a blood culture should be viewed with the strong suspicion that it is a contaminant from the skin. But in a patient with debility, immune deficiency, an intravascular line, a pros¬thetic implant or some other condition favoring an opportunistic infection, the saprophytic or commensal organism isolated may indeed have an important pathogenic role. Consideration must be given to the probability of its having such an opportunistic role in the circumstances of each particular case. To this end, culture should be attempted on several separately collected samples of blood, for the isolation of the same species of organism from repeated specimens suggests that it is probably present as a cause of infection and not merely as a con¬taminant. This probability is strengthened if it can be shown that the different isolates are iden¬tical with one another In their pattern of anti¬biotic sensitivities, biotype or phage-type.
Examination in the laboratory should be planned with a view to minimizing the opportunities for contamination of the specimen or its subcultures with airborne dust or from the bacteriologist’s skin or equipment. However it should be noted that despite insistence on aseptic precautions, most laboratories report finding contamination in 1—5% of the blood cultures they examine.
False-negative results. The media and methods used for blood culture should be such as to support growth of nutritionally exacting organisms, and both strict aerobes and strict anaerobes as well as facultative organisms. Even SO, a single negative culture should not be taken as proof that a patient does not have a blood infection. in bacteriaemic illnesses the number of bacteria in the blood may vary rapidly between many and few or none as they are introduced intermittently from a local infective focus and removed by the reticuloendothelial tissues. To maximize the chances of isolation when the bacteria may be scanty, a large volume of blood, e.g. 10—20 ml, from each collection should be cultured and a number of samples should be collected on separate occasions. Thus, in attempting to discover the organism causing a suspected endocarditis, at least six samples collected at intervals of several hours in the course of 3-6 days should be cultured.
Antibiotics in the blood: The blood sample should be diluted between 1 in 5 and 1 in 10 in the culture medium in order to reduce the concentration of natural antibacterial constitu¬ents to a sub-effective level. The dilution also reduces the concentration of any therapeutically administered antibiotic. But when an antibiotic is present in the undiluted blood at more than 10 times its minimum inhibitory concentration, t may still, after dilution, prevent the growth of viable bacteria in the specimen.
For this reason, every attempt should be made to ensure that samples of blood are collected before the start of antibiotic therapy. If an anti¬biotic has already been given to the patient, the collection of the blood sample should preferably be delayed until at least 24 h after a dose has last been given. Otherwise, it may be possible to add an antibiotic-inactivating agent to the culture bottle. Para-aminobenzoic acid used to be added routinely to blood culture media to neutralize any sulphonamide administered. If the patient has been treated with a β-Iactam antibiotic (a penicillin or cephalosporin), a broad-spectrum β-lactamase should be added to the culture bottle as soon as possible after introduction of the blood, e.g. 0.2ml of Wellcome penicillinase to a bottle containing 50 ml medium. Special blood culture media are commercially available that contain resins that are said to be capable of inactivating a variety of antibiotics (Marion).