Post-marketing surveillance of the quinolones: a view from Great Britain.

All available information on adverse drug reactions (ADRs) to quinolones comes from clinical trials or spontaneous reports to regulatory authorities. Clinical trials so far are too small and suffer from artificial selection of patients. The major value of spontaneous reports of ADRs is in detection of an adverse effect that is virtually restricted to a single member of a drug class; comparisons of several members of a drug class with regard to ADRs are generally unreliable if these reactions are shared by all members of the class (i.e., photosensitivity or CNS stimulation with quinolones). Prescription event monitoring and record linkage are techniques for large, controlled, prospective studies that would be more suitable for monitoring these adverse reactions to quinolones. The response of the pharmaceutical industry to requests for information on ADRs to quinolones is variable, and even when information is made available it is not of sufficient quality for meaningful assessment. All of the current data suggest that quinolones are safe, and marketing departments need to be convinced that a well-defined ADR profile is in their best interests.