Until recently, people with diabetes had to inject their insulin. Now they can inhale it instead. A meta-analysis of 16 trials concluded recently that inhaled insulin works almost as well as subcutaneous insulin for patients with either type 1 or type 2 diabetes. More specifically, subcutaneous insulin reduced concentrations of haemoglobin A1c slightly more than inhaled insulin (weighted mean difference 0.08%, 95% CI 0.03% to 0.14%), but similar proportions in both groups reached the treatment target of less than 7%. Inhaled insulin worked significantly better than oral hypoglycaemic agents, but it was more likely to cause severe episodes of hypoglycaemia (9.4% v 3.5% of patients reported at least one episode; risk ratio 3.06, 1.03 to 9.07). Patients with both types of diabetes preferred inhaled insulin to subcutaneous insulin despite an increase in the incidence of cough and a slight but significant reduction in lung function.
These results are promising, but they don't support a wholesale switch to insulin inhalers, say the authors. The trials were mostly short, open label, and funded by the manufacturers of inhaled insulin. They couldn't rule out bias, and long term side effects are unknown. For now, inhaled insulin should be offered only to patients who face damaging delays in their treatment because they won't or can't inject. Inhaled insulin is contraindicated in smokers.
References
- Ann Intern Mede 2006;145:665-75 [DOI] [PubMed] [Google Scholar]

