Background: Clinical trial registries operated by the pharmaceutical industry (PI) represent a potential data source for systematic reviews (SRs), as they often contain study results. SRs of drugs may therefore require a registry search to ensure an unbiased analysis. Objectives: a) To assess whether PI registries are regularly used as data sources in current SRs of drugs; b) to assess whether the non-consideration of registry data affects the conclusions of these SRs. Methods: Fifty current SRs (including Cochrane reviews) of drugs were identified in a systematic search of PubMed. We checked whether a search of PI registries had been conducted and whether registry data had been considered. If not, we performed a search in www.clinicalstudyresults.org and manufacturers’ registries for documents published within the search period of each SR. These were subdivided into two categories: 1) additional studies; and 2) additional data for studies already considered). The primary outcomes of the SRs will be re-analysed including the additional data. All analyses will be conducted by two reviewers independently of each other. Results: 44 out of 50 SRs (88%) did not search PI registries; four out of six SRs with a registry search were Cochrane reviews. For the 44 SRs without a registry search, our own search produced additional relevant data in nine cases (20%). In eight out of nine cases (89%), this concerned patented drugs. Twenty-nine of the 35 drugs (83%) for which no additional data were identified in PI registries were no longer patented. The effects of the registry data on the results of each SR will be examined and presented. Conclusions: PI registries are rarely used as data sources for SRs of drugs. Additional relevant data, particularly on patented drugs, can be retrieved from these registries. SRs of patented drugs that ignore PI registries may therefore be particularly biased.