terms of unwanted noise (from the perspective of humans) and
ultrasonic sound. Our instrumentation quickly confirmed that
the new lighting did not produce ultrasound in the frequencies between 20 kHz and 100 kHz which allowed the work to
proceed with the new LED lights.
The potential for ultrasonic emissions to harm laboratory
animals and/or compromise the studies they are involved
in, is quite well known. Our recent experience suggests this
knowledge is being used in an unsystematic way. In our first
case study the institution recognized that laboratory rats
were experiencing seizures during transport. Personnel were
aware of the potential of ultrasound to trigger seizures. They
called us in to determine if this was indeed the source of
their problem. In this case, ultrasonic detectors were already
operating in a laboratory where animal studies were on-going, and harm had already been done.
In our second example, the laboratory was more proactive.
They called us in to assess the potential harm that motion
detectors might cause. The detectors were already installed but
the facility was not yet occupied. This allowed them to replace
the motion detectors before animals were exposed.
In the third case study, the laboratory was aware that motion
detectors could be a potential source of ultrasonic sound in a
laboratory setting. In fact, some forms of lighting, computer
terminals, and ventilation systems have been shown to produce
ultrasound. This organization took an even more proactive step
of testing LED lights (for which data did not exist).
The progression of knowledge and pro-action is encouraging: from equipment installed and causing harm, to equipment
installed that has a known potential to cause harm, and to new
equipment for which data does not yet exist. With each step
the facilities are coming closer to providing an ultrasound free
environment for their animals from the very beginning.
While institutions are individually becoming more proactive con-
cerning ultrasound, unfortunately a more systematic approach
to keeping ultrasound out of vivaria has not been developed
industry-wide. The Association for the Assessment and Accredi-
tation of Laboratory Animal Care International (AAALAC Interna-
tional) offers a voluntary accreditation program. Currently there
are more than 900 organizations (primarily universities, hospi-
tals, and corporations) which have earned this accreditation.
Accreditation is based on the National Research Council's Guide
for the Care and Use of Laboratory Animals. The noise guidelines
state, “The location of equipment capable of generating sound
at ultrasonic frequencies is important as some species can hear
such high frequencies. Selecting equipment for rodent facilities
that does not generate noise in the ultrasonic range should be
considered.” 8 Unfortunately no further guidance is given.
One institution appears to have taken a leadership role in
terms of specifications for ultrasound, as well as for audible
noise and vibration in animal facilities. The MD Anderson
Cancer Center at the University of Texas, Houston, has studied
vibration, noise and ultrasound and published their results
in the “Noise, Vibration, and Ultrasonic Design Guide.” 9 This
report specifically states that “ultrasonic motion detectors
should not be used in animal housing facilities unless they
operate above 200 kHz; substantially above the hearing range
of animals.” The report also cautions against the use of ultra-
sonic cleaners unless they are situated “such that there is no
direct line of sight between the animals at any time and the
Ultrasonic cleaners can operate anywhere from 20 kHz to 400
kHz. While the design guide does not say so, it seems prudent
that ultrasonic cleaners in vivaria also be constrained to operate
above 200 kHz.
In the MD Anderson guide, there is a recommendation for
measurements of vibration, sound, and ultrasound during
the commissioning process. This suggests a process, similar
to that often used in cleanrooms, where specifications for
noise and vibration must be met before ownership is accepted. The guide specifically recommends measuring in at least
two animal rooms, as well as in corridors directly adjacent to
animal housing facilities. The report provides preliminary vivarium noise criteria for rodents which includes a NTE (not
to exceed) curve. The NTE curve varies, from approximately
Figure 1: Ultrasonic sound measured in proximity of two different types
of motion detectors
Figure 2: Ultrasonic sound measured in proximity of motion detectors