The effectiveness of rectal administration of insulin suppository on normal and diabetic subjects

The effectiveness of insulin administration by rectal suppository was examined in normal and non-insulin-dependent nonobese diabetic subjects. A 100-U insulin suppository (mean 1.8 U/kg) given to the diabetic subjects caused four times as great a fall in plasma glucose compared with the normal subjects given the same dose (mean 1.6 U/kg). The insulin response after suppository administration demonstrated a significantly positive correlation (r = 0.83, P less than 0.01) with the plasma glucose level before administration. Diabetic subjects given a 100-U insulin suppository (mean 1.7 U/kg) 15 min after meals three times daily showed a significant (P less than 0.05) improvement in postprandial hyperglycemia accompanied by a restoration of the normal circadian profile of plasma IRI and a reduction of urinary glucose from 26 +/- 5.9 to 2.0 +/- 1.0 g/day. No untoward reactions were observed. These data strongly imply a unique characteristic of the insulin suppository in spite of low bioavailability.