94 
Biological safety cabinets and other containment equipment are not 
required for work at the Pi level. It is recognized that under these 
conditions, exposures to aerosols may occur. However, Pi conditions 
are only allowed for experiments involving organisms that are known 
to possess no or minimal hazard to man and that exchange genetic 
information in nature. Prokaryotic and eukaryotic host organisms 
that will be used in the conduct of these experiments have traditionally 
been handled on the open bench, yet they have never been associated 
with any laboratory-acquired infection. Indeed, out of the 3,921 
laboratory-acquired infections reviewed by Pike (5), only 166 were 
caused by class I agents, of which 161 infections were superficial skin 
irritations caused by fungi. 
The Ft. Detrick experience discussed above provides an extremely 
conservative basis for assessing the capability of the safeguards required 
by the NIH Guidelines for recombinant DNA research. For example, it 
would be expected that the rate of laboratory-acquired infections indicated 
for Pi conditions would have been considerably lower if these conditions 
were only used for class 1 and class 2 organisms rather than for class 3 
and class 4 organisms. 
For example, enteric pathogens classified by the CDC as class 2 
agents are only a primary a hazard to the laboratory worker if large 
quantities of the organisms are ingested. The infectious dose--that is, 
the number of organisms ingested- -necessary to cause typhoid fever, 
cholera, and shigellosis in 2 5 to 50 percent of volunteers has been 
5 8 9 
reported to be 10 (1 00, 000) (7), 10 (1 00, 000, 000) (8), and 10 
(1,000, 000, 000) (9) respectively. These diseases are not known to 
be caused by exposures to infectious aerosols. On the other hand, 
most class 3 and class 4 agents can cause infection through inhalation 
of infectious aerosols. Of even greater importance, agents responsible 
for most documented laboratory-acquired infections have the potential 
to cause disease when fewer than 50 organisms are inhaled [Q fever 
(10), tularemia (11)]. Also, equivalent P4 safeguards were only employed 
when research was conducted that required the experimental generation 
of highly concentrated aerosols. Thus, the Ft. Detrick experience should 
be interpreted as a measure of the potential for aerosol exposure to 
research materials rather than as an indication of potential risks of 
conducting recombinant DNA research. Nevertheless, this experience 
demonstrates that P3- and P4-level safeguards can protect the laboratory 
worker from infection where research is being conducted with agents 
known to cause serious disease in man. 
