Herpes zoster complicated by unilateral pleural effusion

546 EJD, vol. 28, n◦ 4, July-August 2018 on patch testing. Though the number of diagnosed irritant reactions was noted to be equal in a comparison of the Finn Chambers and TRUE Test, this was for irritant reactions that were recognized and not for those inadvertently misread [7]. In many comparative studies, the TRUE Test underscores comparator systems for total number of allergic positives, leading to the assumption that the TRUE Test can give more false negatives than other methods [4, 8]. However, this may be an over-simplification. The TRUE Test may be a better system for detecting some relevant allergens (e.g. nickel and neomycin) whilst other systems are superior for others (e.g. fragrance mix I and thiuram mix) [8, 9]. Reproducibility in terms of obtaining the same result on successive testing is a useful measure of the reliability of a patch test system to show true allergic positives. The demonstration of a dose-response relationship using serial dilutions is the ideal discriminator for this, but it is impractical in clinical work. The TRUE Test shows a higher degree of reproducibility compared to other patch test methods. Non-reproducibility was 18% for the TRUE Test, compared to 38% for the Finn Chambers [4]. However, simultaneous right-versus-left testing using the TRUE Test revealed a discrepancy of 5% for the TRUE Test and 4.2% for the Epiquick Test [6, 10]. These predictable variabilities (known unknowns) exist for any system and have to be taken into account in methodological comparison. They may explain some of the disparity observed by Kwon et al. [1]. A demonstration of relevance is the gold standard for the applicative interpretation of results in the patch test clinic. Kwon et al. did not incorporate relevance into their analysis, concentrating on gross results, but others have examined it in comparative studies and found conflicting results [1]. Systems tended to have different strengths for different allergens. For example, more relevant allergic positives to methylisothiazolinone and nickel were found using the TRUE Test than with Finn Chambers, though the latter exceeded the former for fragrance mix I, wool alcohols, and Myroxylon pereriae, and findings were conflicted for para-phenylenediamine and neomycin [5, 9]. The message from Kwon et al. is that all patch test systems have their advantages and disadvantages and none at present is perfect [1]. Whether one methodology is preferred over another depends on local expertise, tradition, and availability of trained technical and nursing staff. There will be instances when the TRUE Test is the most appropriate method for assessing the possibility of contact allergy, as it has been well tested and validated.