Traditional bacteriology systems for antibiotic susceptability testing (AST) generally fall into
two categories; those like the Kirby-Bauer system seen here which looks at the growth of a
confluent layer of organisms after overnight incubation, or more "rapid" systems
which look at metabolic or redox changes produced by the organisms over the
course of about six to eight hours. Both methods are subject to artifacts, in
some part because the any heterogeneous organisms in the population may be
swamped by their neighbors' metabolic activity.
Over the past few decades there have been a number of attempts to
rapidly determine the resistance of a microbe by looking at
changes to single organisms when exposed to an antibiotic. This technique works
well for a few antibiotics but is not generally applicable.
Our technology was developed to provide a microbial
ID/AST, and specifically, an ID and a MIC for a number of antibiotcs on a
simple, inexpensive platform. In operation, the technologist loads a
suspension of organisms from a primary isolate into a small cassette, which
is then placed into the instrument. The cassette is periodically examined by the
reader until all of the results have been collected, at which time the results
are presented. In (typically) two hours, a report is printed giving the
microbial ID and the exact MIC for any desired number of antibiotics.
Accompanying this report is an analysis of the required antibiotic dosage and total costs
of each treatment option for the particular patient.
The features and benefits of the technology are:
In contrast to all
previous methods, our technology employs advances in image capture and
recognition algorithms to digitally characterize the development of individual colony forming units (CFUs) over a few early
generations. A suspension of organisms in a nutrient broth is placed into the sample
cassette, which holds a thin-film hydrogel substrate where there is at least
one control area where the organisms can freely grow and one or more areas
containing growth-altering agents, such as antibiotics. The broth is absorbed
into the hydrogel, which plates a layer of organisms on the surface. The system
is relatively tolerant of variations in organism concentration and can
work directly with urine samples or CSF. The active areas of the cassette are
imaged, and any optical discontinuities are digitally mapped as shown above.
These include dust, defects in the substrate and the CFUs. Also present on the
surface of the substrate are small, randomly dispersed optical reference
particles of approximately 5 microns, which can be seen as the larger dark
object in the first picture. These allow focusing of the image (since the CFUs
themselves may be almost invisible), and they allow accurate frame registration
in the later images.
At intervals of from five to 30 minutes, the active areas are re-imaged
and reanalyzed as shown here after 90 minutes of incubation. Areas which
increase in size are interpreted as CFUs, and other areas are considered
artifact. The reference particles are used to provide exact localization of the
microcolony without the need for precisely re-positioning the cassette. Data is
gathered at least until growth is confirmed in the control area. The rate of
growth in the control area is compared with that in the area containing
antibiotics to determine susceptability. The calculation of relative colony growth versus antibiotic concentration provides a level of
information unavailable by any other technique. A video clip showing growth of micro-colonies and their digitization over the course of three hours can be seen here.
An exact MIC is obtained for all antibiotics. An inert optical marker is admixed
with each antibiotic so that there is a known ratio of marker for each unit of
antibiotic. In separate areas of the substrate, the mixture of each antibiotic
and marker is deposited on the hydrogel (by screen printing, jet printing,
etc) so as to form an approximate a gradient from zero to some point over
the presumed MIC. When the suspension of organisms hydrates the substrate, local
diffusion completes the gradient. During the measurement process, the signal
from the marker is taken along with the microcolony growth information, which
allows the calculation of the antibiotic concentration at any given point on the
gradient. Thus, microcolony growth versus antibiotic concentration can be plotted to
determine the MIC. Note that because the antibiotic concentration is calculated
during the measurement process, the initial deposition of the antibiotic mixture
does not have to be particularly accurate, which simplifies the manufacturing
process. A video of the growth of organisms on a control area and an area containing an antibiotic can be seen here.
This technique can be used with any growth-altering agent to provide information on growth rate versus agent concentration. This may be important for drug discovery or mutagen analysis.
On a separate section of the substrate are deposited a series of biochemical,
immunologic and nutrient markers which are also imaged by the reader, measuring
changes due to growth, metabolism, enzymatic cleavage of a substrate or
attachment of an immunologic marker. The combination of markers plus the
antibiotic sensitivity pattern identifies the organism.