Identifying adverse drug reactions associated with drug-drug interactions: data mining of a spontaneous reporting database in Italy

Background: Drug-drug interactions (DDIs) are an important cause of adverse drug reactions (ADRs). Many studies have recently considered this issue, but most of them focus only on potential interactions and are often related to the hospital setting. A spontaneous reporting database could be a valuable resource for detection of ADRs associated with DDIs; however, data in the literature are limited. Objective: To detect those patients treated with potentially interacting drugs and the cases where reported adverse reactions are a possible consequence of DDIs, using an Italian spontaneous reporting database. Methods: The data were obtained from a database containing all reports of suspected ADRs from five Italian regions (January 1990 to December 2007) that are the main contributors to the Italian spontaneous reporting system. All reports containing at least two drugs, reported as being suspected of causing the ADR or as concomitant medication, were selected and a list of drug pairs was drawn up. We performed a search to verify which drug pairs are considered a potential DDI, using the Internet version of the DRUGDEX(R) system. For each report containing a potential DDI, we verified whether the description of the adverse reaction corresponded to the interaction effect. Results: The database contained 45 315 reports, of which 17 700 (39.1%) had at least two reported drugs. We identified 5345 (30.2%) reports with potential DDIs, and in 1159 (21.7%) of these reports a related ADR was reported. The percentage of reports with potential DDIs increased in relation to the number of concomitantly administered drugs, ranging from 9.8% for two drugs to 88.3% for eight or more drugs. The percentages of serious or fatal reports of ADRs associated with a DDI were significantly higher than other reports analysed. The mean age, percentage of male patients and the mean number of drugs were also significantly higher in reports with DDIs than in other reports. In 235 of 1159 reports (20.3%), both interacting drugs were recognized as suspect by the reporter. This percentage varies in relation to the drugs involved, ranging from 2% to about 65%. The most frequently reported interaction was digoxin and diuretics, but no fatal ADRs were reported with this combination. The combination of anticoagulant and antiplatelet agents was responsible for the greatest number of serious reactions and deaths. Conclusions: This study validates that spontaneous reporting, despite its limitations, can be an important resource for detecting ADRs associated with the concomitant use of interacting drugs. Moreover, our data confirm that DDIs could be a real problem in clinical practice, showing that more than one in five patients exposed to a potential DDI experienced a related ADR