LD50 is the dose of a chemical that is lethal to 50 percent of a test population. It is therefore a measure of a particular median response which, in this case, is death. The term is most frequently used to characterize the response of animals such as rats and mice in acute toxicity tests. The term is generally not used in connection with aquatic or inhalation toxicity tests. It is difficult, if not impossible, to determine the dosage of an animal in such tests; results are most commonly represented in terms of lethal concentrations (LC), which refer to the concentration of the substance in the air or water surrounding an animal.
In LD testing, dosages are generally administered by means of injection, food, water, or forced feeding. Injections are used when an animal is to receive only one or a few dosages. Greater numbers of injections would disturb the animal and perhaps generate some false-positive types of responses. Food or water may serve as a good medium for administering a chemical, but the amount of food or water wasted must be carefully noted. Developing a healthy diet for an animal which is compatible with the chemical to be tested can be as much art as science. The chemical may interact with the foods and become more or less toxic, or it may be objectionable to the animal due to taste or odor. Rats are often used in toxicity tests because they do not have the ability to vomit. The investigator therefore has the option of gavage, a way to force-feed rats with a stomach tube or other device when a chemical smells or tastes bad.
Toxicity and LD50 are inversely proportional, which means that high toxicity is indicated by a low LD50 and vice versa. LD50 is a particular type of effective dose (ED) for 50 percent of a population (ED50). The midpoint (or effect on half of the population) is generally used because some individuals in a population may be highly resistant to a particular toxicant, making the dosage at which all individuals respond a misleading data point. Effects other than death, such as headaches or dizziness, might be examined in some tests, so EDs would be reported instead of LDs. One might also wish to report the response of some other percent of the test population, such as the 20 percent response (LD20 or ED20) or 80 percent response (LD80 or ED80).
The LD is expressed in terms of the mass of test chemical per unit mass of the test animals. In this way, dose is normalized so that the results of tests can be analyzed consistently and perhaps extrapolated to predict the response of animals that are heavier or lighter. Extrapolation of such data is always questionable, especially when extrapolating from animal response to human response, but the system appears to be serving us well. However, it is important to note that sometimes better dose-response relations and extrapolations can be derived through normalizing dosages based on surface area or the weight of target organs.
[Gregory D. Boardman ]
Casarett, L. J., J. Doull, and C. D. Klaassen, eds. Casarett and Doull's Toxicology: The Basic Science of Poisons. 6th ed. New York: McGraw Hill, 2001.
Hodgson, E., R. B. Mailman, and J. E. Chambers. Dictionary of Toxicology. 2nd ed. New York: John Wiley and Sons, 1997.
Lu, F. C. Basic Toxicology: Fundamentals, Target Organs, and Risk Assessment. 3rd ed. Hebron, KY: Taylor & Francis, 1996.
Rand, G. M., ed. Fundamentals of Aquatic Toxicology: Effects, Environmental Fate, and Risk Assessment. 2nd ed. Hebron, KY: Taylor & Francis, 1995.