Optimizing selection of sites for tissue blocks from arteries for microscopic evaluation

A probability model was developed to enable an investigator to estimate the probability of obtaining at least one tissue block for microscopic study from a gross lesion and at least one block from an area free of gross lesions in a sample of n sites in a gross arterial specimen. The model aids the investigator in choosing the location and the number of sites. Three situations were examined using the thoracic aortas of 118 baboons having experimentally induced atherosclerosis. In two situations, the sites were selected randomly and, in the third, purposively. The probability of obtaining the desired tissue characteristics in the purposive situation is greater than in the random situation in 118 baboon thoracic aortas when a low number of sites is utilized, but the probabilities in the two situations become closer as the number of sites increases and the increase in probability levels off with relatively few sites. However, purposive selection reduces the statistical inference.