Inhaled corticosteroids (ICS) are now very widely used in high doses in the management of COPD patients. In sharp contrast to the situation in asthma, ICS provide little or no benefit in COPD patients and may have long-term detrimental effects. High doses of ICS fail to reduce disease progression or mortality, even when combined with a long-acting β₂-agonist (LABA). Several trials have demonstrated that ICS reduce exacerbations by 20–25%, particularly in patients with more severe disease, but these studies are confounded by poor trial design and more appropriate analysis shows no benefit. Indeed, the benefit of combination inhalers seems to be largely due to the effect of the LABA, and long-acting bronchodilators – including tiotropium – provide similar benefits in reducing exacerbations. However, there may be some COPD patients, for example those with concomitant asthma, who benefit from ICS. Yet it has not been possible to identify any clinical factors that predict corticosteroid responsiveness in COPD patients in the large clinical trials. There is increasing evidence that high doses of ICS may have detrimental effects on bones and may increase the risk of pneumonia. ICS fail to suppress inflammation in COPD patients because there is a marked reduction in histone deacetylase-2, the nuclear enzyme that corticosteroids require to switch off activated inflammatory genes. In the future, alternative anti-inflammatory treatments will be needed for COPD or therapeutic strategies which reverse the molecular pathways that causes corticosteroid resistance.