Over the past few years significant steps forward have been made towards the development of insulin formulations suitable for inhalation via several delivery systems. This innovative route of insulin delivery offers the potential of administering pre-meal insulin in a non-invasive way to patients currently receiving multiple daily injections. This article describes the pharmacodynamic and pharmacokinetic profiles and the efficacy and safety data of inhaled insulin preparations. Particular emphasis is placed on Exubera, which currently has the largest pool of efficacy and safety data. In patients with type 1 diabetes 24-week trials have demonstrated that inhaled insulin was equally efficacious to short acting insulin (2-4 daily injections). Results of trials conducted in type 2 diabetes showed inhaled insulin efficacy too, and a potential role for inhaled insulin in patients failing oral medications. Safety data have shown that the most common reported adverse event is mild cough, which appears to be decreasing in frequency during the course of therapy. Higher antibody titers have been observed in patients treated with inhaled insulin compared to subjects treated with subcutaneous insulin. However, the titers do not present any association with clinical correlates. The safety area in need of higher scrutiny is naturally the area of pulmonary function tests (PFTs). A brief synopsis of PFTs is followed by the review of PFTs data following short and long term treatment with inhaled insulin. Two-year data in type 2 diabetes showed a significant change in Forced Expiratory Volume in 1 second (FEV1) after 6 months of treatment, but not after 9, 12,18,24 months of treatment and 6 and 12 months of wash-out. In type 1 diabetes a significant change in Diffusing Lung Capacity of Carbon Monoxide (DLCO) was observed after 24 weeks of inhaled insulin therapy. Long-term data in type 1 diabetes are only available as part of a pooled sample composed of patients with both type 1 and type 2 diabetes. In these patients, who had received inhaled insulin for at least 4 years, annualized changes of FEV1 and DLCO were similar in the group treated with inhaled insulin compared to the group treated with sub-cutaneous insulin.