Time to peak insulin level, relative bioavailability, and effect of site of deposition of nebulized insulin in patients with noninsulin-dependent diabetes mellitus

Seven fasting patients with noninsulin-dependent diabetes mellitus (NIDDM) inhaled 1.0 U/kg of body weight of nebulized regular pork insulin by mouth or were subcutaneously (sc) injected with 0.1 U/kg of body weight of insulin in the upper arm on two different occasions. The time to peak insulin level was compared for the two treatment modalities. Insulin bioavailability after inhalation was quantified relative to sc injected insulin. Deposition of a radiolabeled insulin surrogate aerosol (insulin diluent) in the larger central airways versus the peripheral airways, expressed as the inner-to- outer (I:O) ratio, and in the lung apex versus the lung base, expressed as the apex-to-basal (A:B) ratio, was quantified with gamma scintigraphy. Ratios were related to glucose responses after inhalation of insulin. Times to peak insulin level were similar for the two methods of treatment, averaging 43 ± 16 and 64 ± 40 minutes after inhalation and sc injection of insulin, respectively. The bioavailability of inhaled insulin averaged 14.7% ± 5.8% relative to sc injected insulin. This was significantly less than the average bioavailability of deposited drug (18.9% ± 5.3%) relative to sc injected insulin (P < 0.05). I:O and A:B ratios for the surrogate aerosol averaged 1.3 ± 0.4 and 0.7 ± 0.2, respectively. Linear regression analysis revealed that the maximum percentage of decrease in glucose after insulin inhalation was significantly related to the A:B ratio such that percentage decrease in glucose was greater in patients who demonstrated a lower A:B ratio (P = 0.003). Percentage decrease in glucose was not related to the I:O ratio. These results indicate that the bioavailability of nebulized insulin inhaled by mouth is approximately 20% when calculated in terms of drug deposited and suggest that increasing the distribution of insulin aerosol to the base of the lung enhances the glucose response in patients with NIDDM during the fasting state.