Quantitative studies of cutaneous hypersensitivity: the prevalence of epicutaneous flare reactions to allergenic pollen extracts.

The flare reactions produced by epicutaneous tests with 68 undiluted allergenic pollen extracts were measured in 550 allergic patients. Skin test reactions greater than or equal to 2, greater than or equal to 5, greater than or equal to 10, greater than or equal to 20, and greater than or equal to 30 mm in diameter, respectively, were detected in approximately 67%, 22%, 10%, 3%, and 1% of the 34,700 skin tests. With the Kolmogorov-Smirnov difference test, the cumulative frequency of reaction diameters and loge-transformed diameters of all reactions and reactions to individual allergenic extracts differed significantly (p less than or equal to 0.01) from a normal distribution. The ability to identify specific differences between reactions to closely related pollen extracts was evaluated. Specific reactions could be reliably identified with greater than or equal to 10 mm diameter flares. This arbitrary conservative threshold was used to estimate the relative prevalence of positive reactions to each allergenic extract. Seven allergenic extracts elicited the first quartile of all positive reactions. Thirteen, 18, and 30 allergenic extracts, respectively, were needed to elicit the second, third, and fourth quartiles of all positive reactions. Reactions to amphiphilous, as well as anemophilous, pollens were detected. Skin test reactions to grasses were more prevalent than reactions to weeds and trees. The most informative allergenic extracts for the detection of patients who exhibited a positive reaction to any extract were from red fescue-grass pollens, mesquite, short ragweed, red clover, and timothy-grass pollens.