The consequences of matrix effects in mass spectrometry analysis are a major issue of concern to analytical chemists. The identification of any ion suppressing (or enhancing) agents caused by sample matrix, solvent or LC-MS system components should be quantified and measures should be taken to eliminate or reduce the problem. Taking account of ion suppression should form part of the optimisation and validation of any quantitative LC-MS method. For example the US Food and Drug Administration has included the evaluation of matrix effects in its "Guidance for Industry on Bioanalytical Method Validation" (F.D.A. Department of Health and Human Services, Guidance for industry on bioanalytical method validation, Fed. Regist. 66 (100) 2001). If ion suppression is not assessed and corrected in an analytical method, the sensitivity of the LC-MS method can be seriously undermined, and it is possible that the target analyte may be undetected even when using very sensitive instrumentation. Sample analysis may be further complicated in cases where there are large sample-to-sample matrix variations (e.g. blood samples from different people can sometimes vary in certain matrix components, shellfish tissue samples sourced from different regions where different phytoplankton food sources are present, etc) and therefore exhibit varying ion-suppression effects. Although it is widely agreed that there is no generic method to overcome ion suppression, the purpose of this review is to: provide an overview of how ion suppression occurs, outline the methodologies used to assess and quantify the impact of ion suppression, discuss the various corrective actions that have been used to eliminate ion suppression in sample analysis, that is to say the deployment of techniques that eliminate or reduce the components in the sample matrix that cause ion suppression. This review article aims to collect together the latest information on the causes of ion suppression in LC-MS analysis and to consider the efficacy of common approaches to eliminate or reduce the problem using relevant examples published in the literature.