There was no difference in episodes of severe hypoglycemia in patients using inhaled insulin compared to those using subcutaneous insulin. A higher proportion of patients using inhaled insulin reported at least one episode of severe hypoglycemia compared to those using oral agents (risk ratio, 3.06 [CI 1.03 to 9.07]; 9.4% versus 3.5%, respectively).
With respect to pulmonary safety, all trials selected patients without histories of pulmonary problems and with at least six months of nonsmoking status. Pulmonary safety was assessed by self-reported symptoms and by pulmonary function tests. The most common pulmonary symptom associated with inhaled insulin was nonproductive cough. This symptom was reported more frequently compared to patients using subcutaneous insulin or oral agents (risk ratio, 3.52 [CI 2.23 to 5.56]; 16.9% versus 5.0%). Cough was noted early in the treatment course and diminished in frequency over time. Patients receiving inhaled insulin had a greater decrease in FEV1 (forced expiratory volume in the first second) from baseline than the comparator group (weighted mean difference, -0.031 L [CI-0.043 L to -0.020 L]). This decrease progressed slowly over the first six months but stabilized in studies of up to two years’ duration.
Only four trials reported data on overall patient satisfaction for inhaled insulin versus subcutaneous insulin. All trials reported a statistically significant increase in overall patient satisfaction with inhaled insulin over subcutaneous insulin. Patients randomly assigned to inhaled insulin were more likely to continue taking inhaled insulin than to switch back to subcutaneous insulin.
This meta-analysis showed that inhaled insulin is comparable to subcutaneous insulin in lowering hemoglobin A1C levels in patients with type 1 or type 2 diabetes. The proportion of patients reaching a target hemoglobin A1C of less than 7% was much lower in the studies in this meta-analysis as compared to levels in trials of intensive subcutaneous insulin therapy.
It’s more difficult to compare inhaled insulin with oral hypoglycemic agents because most studies involving oral agents used fixed dosing with different types of oral agents. There was a three-fold risk of severe hypoglycemia in patients using inhaled insulin compared to those using oral hypoglycemic agents. This is probably due to overall improved glycemic control in the inhaled insulin group. Cough was more common in the inhaled insulin groups, and there were small decreases in FEV1, but these did not progress over two years. The potential for pulmonary toxicity with long-term administration has not been evaluated and deserves further study.