Meta-analysis: efficacy and safety of inhaled insulin therapy in adults with diabetes mellitus.

BACKGROUND: Injection insulin therapy is not readily accepted by patients and many health care providers; therefore, less invasive options for insulin therapy are desirable.
PURPOSE: To examine the efficacy, safety, and patient acceptability of inhaled insulin therapy in nonpregnant adults with diabetes mellitus. DATA SOURCES: English-language studies in MEDLINE, the Cochrane Clinical Trials Register (through June 2006), and U.S. Food and Drug Administration review documents of the first formulation of inhaled insulin approved for clinical use. STUDY SELECTION: Randomized, controlled trials of at least 12 weeks' duration that compared inhaled insulin with another active therapy and reported hemoglobin A1c levels in adults with type 1 or type 2 diabetes mellitus. DATA EXTRACTION: Two reviewers independently assessed trials for inclusion and extracted data. Differences were resolved by consensus. DATA SYNTHESIS: Sixteen open-label trials met the inclusion criteria (4023 patients; age range, 18 to 80 years). Among patients with type 1 or type 2 diabetes, there was a small decrease in hemoglobin A1c level from baseline that favored subcutaneous insulin over inhaled insulin (weighted mean difference, 0.08% [95% CI, 0.03% to 0.14%]), although there was no difference in the proportion of participants achieving hemoglobin A1c levels less than 7%. Inhaled insulin lowered hemoglobin A1c levels more (weighted mean difference favoring inhaled insulin, -1.45% [CI, -1.80% to - 1.10%]) compared with fixed doses of oral agents but much less when compared with oral agents titrated to glycemic efficacy (weighted mean difference favoring inhaled insulin, -0.20% [CI, -0.34% to - 0.07%]). Severe hypoglycemia was more likely to occur with inhaled insulin than with oral agents (risk ratio, 3.1 [CI, 1.0 to 9.1]), but there was no increased risk compared with subcutaneous insulin. There was an increased incidence of mild to moderate nonprogressive dry cough in patients treated with inhaled insulin (risk ratio, 3.5 [CI, 2.2 to 5.6]) and a mild decrease in certain pulmonary function testing variables, which did not progress over 2 years. Patients preferred inhaled insulin over subcutaneous insulin. LIMITATIONS: All trials were open label, which may introduce bias. Most of the trials were of 24 weeks' duration or less, limiting assessment of long-term safety.
CONCLUSIONS: Inhaled insulin offers an alternative noninvasive option for premeal insulin administration, with glycemic efficacy slightly less than subcutaneous regular insulin and increased patient acceptability. Until long-term safety data are available, inhaled insulin should be reserved for nonpregnant adults with diabetes who are opposed to injections and who would otherwise delay appropriate and timely therapy with insulin.