Changing sex ratio in diabetic ketoacidosis.

An annual audit of diabetic ketoacidosis and hyperosmolar non-ketotic state was made in one hospital from 1971 to 1988. There were 846 episodes of ketoacidosis and 126 episodes of hyperosmolar state. A relative fall occurred in the number of episodes of ketoacidosis compared with hyperosmolar state over this time (p less than 0.05), and there was a change of female:male ratio for episodes of ketoacidosis occurring in established diabetes from 2.79 to 1.59 (p less than 0.01). In contrast the female:male ratio remained unchanged (mean 1.16) for episodes of hyperosmolar state and remained less than 1.0 for all episodes of ketoacidosis in previously undiagnosed diabetes mellitus. Among patients who suffered recurrent ketoacidosis there was a reduction in the number of episodes occurring in female patients and an increase in the number of episodes occurring in male patients in each successive 6-year period with consequent change in female:male ratio for this subgroup from 7.33 to 4.75 to 1.12 (p less than 0.001).