, Volume 49, Issue 5, pp 891-899

First online:

Efficacy and safety of preprandial human insulin inhalation powder versus injectable insulin in patients with type 1 diabetes

  • S. GargAffiliated withDepartment of Medicine and Paediatrics, Barbara Davis Center for Diabetes at the University of Colorado Health Sciences Center Email author 
  • , J. RosenstockAffiliated withDallas Diabetes and Endocrine Center
  • , B. L. SilvermanAffiliated withAlkermes
  • , B. SunAffiliated withLilly Research Laboratories, Lilly Corporate Center
  • , C. S. KonkoyAffiliated withLilly Research Laboratories, Lilly Corporate Center
  • , A. de la PeñaAffiliated withLilly Research Laboratories, Lilly Corporate Center
  • , D. B. MuchmoreAffiliated withLilly Research Laboratories, Lilly Corporate Center



The efficacy and safety of human insulin inhalation powder (HIIP) plus insulin glargine were compared to subcutaneously injected insulin (SC insulin) plus insulin glargine in patients with type 1 diabetes.


This was a randomised, open-label crossover study in which one group of patients received preprandial HIIP plus insulin glargine for 12 weeks, followed by the same duration with preprandial SC insulin (lispro or regular) plus insulin glargine. Another group of patients received the reverse treatment sequence. The trial was designed as a non-inferiority comparison of the two treatments for effect on HbA1c; blood glucose levels were also monitored. Safety assessments included adverse event reporting and hypoglycaemic events.


HbA1c at endpoint was 7.95±0.12% for the HIIP treatment and 8.06±0.12% for the SC insulin treatment; mean changes from baseline to endpoint were −0.08 and 0.00%, respectively, (p=NS). The upper limit of the 95% CI of mean difference in HbA1c between the two treatments was 0.02%, indicating that HIIP was not inferior relative to SC insulin, as measured against the pre-defined margin of 0.3%. Fasting blood glucose was significantly lower for HIIP treatment (8.09±0.33 mmol/l; n=117) than for SC insulin treatment (9.05±0.33 mmol/l; n=111) (p=0.01). Safety profiles were comparable between the two treatments. The rate of any hypoglycaemia (least-squares mean adjusted for 30 days±SE) was 8.9±0.7 and 8.2±0.8 for HIIP and SC insulin treatments, respectively, (p=0.29). The rate of nocturnal hypoglycaemia was greater for HIIP (4.2±0.4) than for SC insulin (2.7±0.4; p<0.001).


HIIP was similar in efficacy to SC insulin for glycaemic control in type 1 diabetes mellitus. The two treatments had comparable safety profiles.


Drug delivery HbA1c Hyperglycaemia Inhaled insulin Injection Pulmonary Type 1 diabetes mellitus