A STUDY ON EXPERT READING OF PATCH TEST REACTIONS - INTERINDIVIDUAL ACCORDANCE

To diagnose allergic or irritant contact dermatitis, a clinically rele vant contact allergy has to be demonstrated or ruled out, respectively . Although patch testing has been used for 100 years, it remains the m ethod of choice for diagnosing contact allergy A disadvantage of patch testing is that reading is subjective, based on inspection and palpat ion of the test area, implying that the assessment is subject to the r eader's knowledge and experience. This study was carried out to invest igate the accordance in reading patch test reactions between 5 dermato logists. 4 groups, each with 10 patients, participated. Within each gr oup, all 10 were allergic to one and the same sensitizer; nickel, epox y resin, ethylenediamine, or Kathon CG. The sensitizers were tested in serial dilutions and applied randomly to the back. The tests were rea d independently by the dermatologists in a blinded fashion. A protocol was used where the dermatologists had to note the presence of the mor phological features erythema, infiltration, papules, vesicles, and bul lae. In this way, it was possible to allocate the various readings int o 4 classification systems, 3 European and one American, although the definition of the various classifications might differ slightly. Based on the readings and classifications, it was possible to calculate the degree of accordance within the 4 systems used. It was also possible to analyze the degree of accordance for the various morphological feat ures. Total accordance for the 5 reading dermatologists for positive a nd negative test reactions was noted in 36% and 46%, respectively. The 3 European systems require homogenous erythema and infiltration for a reaction to be classified as allergic, while the American system also requires the presence of papules. Therefore, a higher number of aller gic reactions (9%) was obtained with the European systems. Between the 4 classification systems, there were statistically significant differ ences, with the highest number of deviations for the system in which p apules had discriminatory significance. When looking at the various mo rphological features, papules were most difficult to assess, followed by vesicles. Preferably, a classification system should not depend on papules. In conclusion, there was good accordance among the dermatolog ists in reading patch test reactions, since the 5 dermatologists read 82% of all tests similarly.