10. Direct Antibiotic Susceptibility Testing

10. Direct antibiotic susceptibility testing

The temptation to perform a susceptibility test directly on the specimen isolated from a patient is great as it has the potential to save at least 24 hours in the provision of the result. It is argued that this may be of critical value in the management of the patient. However, this is only the case if the result can be guaranteed to be accurate and this is not always the case. A premature inaccurate result may do more harm than good and in the vast majority of cases it is of little advantage to hasten the result because the patient can be maintained on a broad antibiotic cover until an accurate assessment is made. One area where we are often tempted to provide direct susceptibility testing is in blood cultures. However, in an extensive retrospective analysis Mahrer showed that while there is a reasonable agreement between direct and formal susceptibility testing there were a number of serious discrepancies between the two examinations1. Consequently, we do not recommend direct CDS susceptibility testing of blood cultures; the only acceptable direct susceptibility testing is on urine and this is described below.

10.1 Urine

Direct antibiotic susceptibility testing of urine facilitates the early reporting of antimicrobial susceptibilities. The usefulness of early reporting was established previously2.

In a recent study, the volume of the urine sample used for direct susceptibility testing was adjusted based upon the sample bacterial concentration, to achieve a suspension equivalent to the standard CDS suspension (1010/L)2. Bacterial concentrations were estimated semi-quantitatively by manual phase contrast microscopy using Kova® slides. When the bacterial concentration was estimated at ≥ 9 x 1010/L, 25 μl of the urine sample was added to 2.5 ml of sterile saline yielding a bacterial suspension of ≥ 109/L. A 250 μl inoculum was used for all lower bacterial concentrations giving a bacterial suspension of < 1010/L. In this study an acceptable lawn of bacterial growth was achieved for 95% of urine samples having bacterial concentration ranging from < 109/L to > 1011/L, thereby reducing the number of susceptibilities that had to be repeated.

Bacterial concentrations reported by automated urine analysers may not be accurate enough to permit reliable adjustment of the inoculum. We recommend that laboratories using these analysers use a 250 μl inoculum for all samples.

When manual microscopic morphology suggests enterococci, streptococci or staphylococci, a larger inoculum is required to compensate for the lower viable count of these organisms. Add 50 μl of the urine sample for bacterial concentrations of ≥ 9 x 1010/L and 250 μl for all lower concentrations.

After incubation, the lawn of growth should be confluent, i.e., similar to that obtained with the standard CDS inoculum. If the lawn is semi‑confluent or worse, too heavy or mixed, the result should be withheld and susceptibility testing repeated from a plate culture of the organism using a conventional prepared CDS inoculum (section 2.2.2, page 16).


Table 11.1. Urine sample inocula for direct susceptibility testing.

Bacterial Morphology Quantitation* Urine Sample Inoculum
Rods < 9 x 1010/L 250 μl
≥ 9 x 1010/L 25 μl
Cocci < 9 x 1010/L 500 μl
≥ 9 x 1010/L 50 μl
Automated Urine Analysers 250 μl

*  A sample bacterial concentration of 9 x 1010/L is equivalent to 1000 bacteria per small square of the Kova® chamber.

†     Add the inoculum to 2.5 ml sterile saline and use this suspension to flood the susceptibility plate.


1       Mahrer, S., Reinbott, P. & Bell, S.M. 2008. Blood Cultures – A Retrospective evaluation of Direct Antibiotic Susceptibility testing using Positive Broth cultures detected by the BacT/Alert system. Australian Society for Microbiology. Annual Conference, Adelaide, Australia. (A pdf of the poster is available on the CDS website)

2       Mukerjee, C. & Reiss‑Levy, E. 1994. Evaluation of direct disc diffusion susceptibility testing for bacteriuria using diluted urine against standard CDS method. Pathology. 26, 201‑7.

3       Fisher, G., Simos, A. & Bell, S.M. 2005. Direct sensitivity testing of urine samples based on the microscopic quantitation and morphology of organisms. Australian Society for Microbiology. Annual Conference, Canberra, Australia.