Premium
Negative predictive value of drug skin tests in investigating cutaneous adverse drug reactions
Author(s) -
Waton J.,
Tréchot P.,
LossAyav C.,
Schmutz J.L.,
Barbaud A.
Publication year - 2009
Publication title -
british journal of dermatology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.304
H-Index - 179
eISSN - 1365-2133
pISSN - 0007-0963
DOI - 10.1111/j.1365-2133.2008.08975.x
Subject(s) - predictive value , value (mathematics) , medicine , dermatology , library science , computer science , machine learning
Summary Background Drug skin tests are useful in aetiological analyses of cutaneous adverse drug reactions to determine if the drug can be rechallenged, or to avoid a cross‐reaction with a substitute drug. Objectives To evaluate the negative predictive value of drug skin tests. Methods We retrospectively analysed the files of patients referred for drug reactions. We have enrolled those having strictly determined drug reactions with clinical features, delayed onset after drug intake, drug causality assessment, and negative drug skin tests followed by drug administration. Oral provocation tests or substitution tests with a drug of the same class as that suspected of causing the drug reactions were performed. Results From 1957 files analysed, 200 patients were included. After 403 patch tests, 403 prick tests and 304 intradermal tests, which were all negative, 260 oral provocation tests and 143 substitution tests were done; 307 different drugs were rechallenged. There were 42 positive drug re‐administrations in 27 oral provocation tests and 15 substitution tests. The negative predictive value of our drug skin tests was 89·6%. The negative predictive value for beta‐lactams was 87% for oral provocation tests and 96% for substitution tests, and for corticosteroids it was 100% and 74%, respectively. Conclusions Negative drug skin tests do not eliminate the responsibility of a drug in drug reactions, and must be followed by drug re‐administration under hospital surveillance.