Optimizing Reproducibility for Clinical Studies Involving Patch Testing and Application of Topical Preparations

Abstract

The reproducibility and consistency of patch-test techniques can be problematic, making controlled clinical trials of patch testing difficult. To measure the accuracy and reproducibility of applying controlled quantities of petrolatum onto Finn Chambers. Four dermatology nurses applied a total of 240 samples of white petrolatum, using three syringe sizes and types. Three different amounts of white petrolatum (0.02 mL, 0.05 mL, and the "usual" amount) were expressed onto previously weighed Finn Chambers, using three different syringe sizes (5 mL, 1 mL, and 0.5 mL), five times each on two separate days. The average weights of 0.05 mL of petrolatum expressed with each type of syringe (5 mL: 0.04252 g; 1 mL: 0.04084 g; and 0.5 mL: 0.04139 g) were not significantly different from each other in pairwise comparisons (p > .36) or from the "gold standard" expected value (0.04138 g, p > .72). The average weights of 0.02 mL of petrolatum expressed with two types of syringes (5 mL: 0.02138 g; 0.5 mL: 0.01778 g) were significantly different from each other (p = .0012), but neither differed significantly from the expected value (0.01655 g, p > .08). The variance due to the effect modifications of nurse, day, and interaction of nurse and day was statistically significant for measurements made with the 5 mL syringe but not for measurements made with the 1 mL or 0.5 mL syringe. The average amounts of petrolatum extruded from smaller syringes (0.5 mL and 1 mL) were less variable and more reproducible than those extruded from a 5 mL syringe.