Sakel (1938a) drew attention to the difficulty of establishing satisfactory comas in a minority of patients attending for Deep Insulin therapy. This phenomenon has since been confirmed by other workers including Tillim (1938) whose patient received 500 units of insulin without the production of deep coma, by Hall (1940) who reported an instance in which 1,000 units of insulin was equally unsuccessful, by Reznikoff and Scott (1942) who described how neither 120 nor 1,000 units of insulin when injected intravenously produced any significant difference in hypoglycaemia in insulin resistant patients, and more recently by Fogarty (1953) whose case required 5,000 units of insulin for the production even of sopor. Other recorded examples of resistance to massive doses of insulin include those of Banting et al. (1938) and Tennent (1944).

Various explanations of this perverse response to insulin have been formulated, including that of Jones (1939) who proposed that resistance to insulin was both a problem of true insulin insensitivity and also of an anomalous response of the central nervous system to hypoglycaemia. Meduna et al. (1942) preferred to ascribe it to anti-insulin factors in the blood, but a more interesting interpretation derived from Freudenberg (1952) who suggested that if the effects of high insulin dosage employed in insulin coma treatment were regarded as a special instance of a stress response, then the fluctuations and differences in response could be equated in terms of the General Adaptation Syndrome of Selye. High doses were thus an index of an effective “alarm stage” characterized by a discharge from plentiful adrenocortical reserves.