|
METHODS
Observers
Individuals from the University
of Chicago and three outside campuses participated in this study.
The conication principles, analysis methods, and practical application
were taught to each participant during his or her visit to the
University of Chicago. Between 25 and 50 trials of sonication
were necessary for the observers to become proficient at the production
of the small and stable microbubbles. After the instruction and
practice sessions, each participant performed three trials that
were analyzed by the laser counter and served as the basis of
this article on reproducibility. All studies were performed in
one laboratory at the University of Chicago.
Sonicator
A Heat-Systems (model 220, Plainsview,
New York) with a 1/2 –inch titanium-tipped horn was used as the
continuous source of power for the generation of the microbubbles.
The power control was set at 7, and the horn was routinely tuned
in Renografin-76 solution before each study.
Manufactured latex particles (Duke Scientific, Palo Alto,
Califronia) were used for calibration. Serial dilutions of sonicated
albumin microspheres of known concentrations based on Coulter
counter analysis were used to determine the laser particle concentration
measurements.
Light
Microscopic Analysis
A stereoscopic light microscope (model CHA, Olympus Corp.,
Lake Success, New York) with a net eyepiece grid graduated in
microns was used to visually assess the measurements performed
on the sonication trials.
Sonication
Technique
The half-inch titanium horn from a sonicator was placed approximately
0.5 cm beneath the surface of the 8 ml of Renografin-76 contained
in a plastic syringe. With the tip of the syringe held firmly
by hand, the energy was turned on. After 10 seconds the syringe
was lowered enough to permit the tip of the sonicator to briefly
contact the surface of the liquid, thus permitting a period of
surface agitation. Once the surface agitation occurred, the tip
of the sonicator was lowered beneath the surface of the liquid
for 10 seconds. The surface agitation process was briefly performed
a second time at 20 seconds, followed by the lowering of the horn
beneath the surface for the duration of the sonication time. Ultimately,
after a total sonication time of 30 seconds, the 8 ml solution
of Renografin-76 appeared translucent. A dense white coloration
should not be present in the final solution (Figure 1). This white,
opaque appearance implies the presence of uncontrolled surface
agitation with the development of relatively large and unstable
microbubbles.
Laser
Sampling Technique
Before each test background counts of particles contained
in Renografin-76 were performed. Three separate determinations
were recorded by the laser counter. With a predetermined threshold
correlation chart provided by the manufacturer, the background
counts were considered acceptable if the absolute counts did not
exceed 200 counts/ml³. The Soectrex Fourier analysis was not limited
by the threshold values, thus the frequency analysis (histogram
plots) included the distribution of all background counts.
Immediately after the sonication process was completed, 1
ml of the sonicated solution was
Discussion
The ability to quantify blood flow by contrast ultrasonography
requires that the ultrasound tracer substances (microbubbles)
be reproducible and quantifiable in production and analysis. Before
the development of the sonication technique, microbubbles for
use in contrast ultrasound imaging were created by manual or hand
agitation. The microbubbles produced by hand agitation proved
to be unstable and difficult to reproduce,
20 and the relatively
large diameters prohibited passage through the capillary vasculature,
9, 10 thus presenting
a potential threat to safety. In addition, studies have shown
that the sonicated microbubble transit times correspond to actual
blood flow as measured by electromagnetic flow meters at control
and altered physiologic states. 14-16
width (microns)
Figure 4 Frequency
histogram for expected microbubble distribution characteristics
with Poisson model.
Although earlier reports have
been published on the sonication method for use in producing contrast
ultrasound agents, ¹ ² no analysis of the interobserver reproducibility
of the sonication has been attempted.
This article, for the first time,
analyzes the actual production and subsequent analysis method
used in preparing sonicated agents for clinical imaging. All four
participants independently volunteered for the sonication reproducibility
studies. After a 2- to 4- hour instructional period, automated
laser analysis of the microbubble
|
Laser
Analysis
A scanning laser particle counter
(Spectrex Corporation, Redwood City, California) was used to determine
the in vitro diameters and concentrations of the microbubbles.
As the laser beam passed through the solution that contained the
microbubbles, a predesigned “sensitive zone” at the center of
the container served as the sampling site for the examination.
When the laser beam struck a bubble, there was near-angle scatter
of deflected light. The magnitude of the laser deflection was
directly proportional to the diameter of the bubble. Following
a 25-second counting period, the concentration of the bubbles
per cubic centimeter was displayed on an electronic readout.
In conjunction with the laser analysis of the absolute particle
counts, a Fourier transform of the laser pulse amplitudes measured
the diameters of the bubbles within the sensitive zone. Afer a
30-second sampling period, a frequency histogram for the sample
was generated. The absolute numbers of the bubbles found in each
size channel can be calculated by multiplying the percentages
listed in the frequency histogram by the absolute counts determined
from the laser scanner.
Withdrawn from the center of the syringe and mixed in a beaker
containing 120 ml of Renografin-76. The contents of the beaker
were then scanned with the laser counter to determine the in situ
diameters and concentration of the sonicated microbubbles. The
laser counter provided the absolute bubble concentrations within
the 1 ml³ region of analysis, and the Fourier analysis listed
the microbubble diameter frequency of occurrence within a sampled
time period. An example of the frequency histogram is shown in
Figure 2. Background particulate counts obtained before the analysis
were subtracted to obtain the actual concentration of the sonicated
microbubbles. In effect, the background particulate contamination
was small (less than 10%) relative to the large numbers of microbubbles
measured during the analysis. To minimize coincidence counts with
the laser counter, the microbubble concentrations were kept between
700 and 1500, requiring a 1:120 dilution, of sonicated Renografin-76.
Particle Size
Figure 2 Laser analysis of microbubbles created from Renografin-76
is displayed. Ordinate is size of microbubbles in microns, and
abscissa is frequency of occurrence.
Calibration
Studies
Particle size diameters were assessed with manufactured solid
latex spheres. The size distribution was described as 5.0 ± 0.38
μm. Coulter counter
determinations of serial dilutions of the albumin concentrations
(spheres per milliliter) were used to check the concentrations
recorded from the laser counter.
formation was accomplished. At
the end of the sessions, each participant was confident in the
production and analysis of the microbubbles created from the sonication
technique.
As a measure of quality control,
a light microscope was used for visual comparison of bubble sizes
as recorded by the laser counter. Although the optics of the two
measurement systems differ, the relative diameters could be easily
assessed when visualized in the field of the light microscope.
From our earlier studies ² it was observed that light microscope
measurements overestimate the diameters of the microbubbles partly
because of the compressive effects of the cover slip and the light
diffraction created from the refringent surface rim of the bubbles.
The relative size distribution of the microbubbles, however, whether
measured by laser or light microscopy, were similar in a variety
of solutions. ¹ ²
The 12 microbubble size distributions
obtained were similar. These distributions were modeled to quantify
any differences between them, and it was concluded that any differences
were negligible, since similarly shaped distributions with means
of 8 to 10 µm would be clinically applicable. Thus the method
is reproducible.
This study describes the results
of producing and analyzing the sonication method for generating
microbubbles. Once instructed in the methods of sonication, the
four participants demonstrated the reproducibility of the sonication
technique for the production of microbubbles. This technique promises
to provide a rapid, economical ultrasound contrast agent that
can be widely applied for use in tissue perfusion imaging. 11,17,18
|
