W.A.J. Crane, G. F. Wilgram and D. J. Ingle Department of Pathology, University of Sheffield and Department of Physiology, University of Chicago.

The concept that derailment of adrenal cortical function as the result of exposure to stress is a primary cause of many human diseases is principally supported by Selye (1956). The evidence in favour of the hypothesis is impressive and many biologists and physicians have felt compelled to accept the proposition. The hypothesis also has an appeal to some who have recognized it as logical and heuristic but who have not been persuaded to accept it without further testing. In our opinion critical examination of the evidence does not confirm the validity of the concept although all of the evidence can be interpreted as supporting the less bold conclusion that hormones of the adrenal cortex have a permissive or supporting r61e in certain metabolic responses to stress and the overt manifestation of certain diseases. Addison's disease, Cushing's syndrome, adrenal virilism, aldosteronism, and the related mixed syndromes are known diseases of the adrenal glands and will not be discussed here. We are concerned with the role of the adrenal cortex in the 'diseases of adaptation'.

There are four principal lines of supporting evidence which link the adrenal cortex with other diseases.

(1) Corticotrophin (ACTH), cortisone, Cortisol and certain synthetic derivatives of Cortisol can suppress the symptoms of a large number of human diseases, most of which are characterized as inflammatory diseases. The growth of some tumours can be retarded by these hormones. There is, however, no satisfactory evidence that the hormones cure these diseases or that these diseases represent any form of adrenocortical insufficiency, but it has been postulated that a deficiency or an imbalance in adrenal steroids can make the inflammatory diseases more likely to occur.

(2) Removal of the adrenal glands may be followed by amelioration of diabetes, hypertension, and certain cancers in experimental animals and in patients.

(3) Over-dosing of experimental animals with one or another corticosteroid can cause a variety of pathological changes, including steroid diabetes, hypertension, nephrosclerosis, nephritis, arteriosclerosis, polyarteritis nodosa and gastro-intestinal ulcers. Resistance to infectious agents is dramatically decreased in animals and patients with severe hyper-corticalism.

(4) Enucleation of the adrenal glands in the immature rat sensitized by unilateral nephrectomy and a high sodium load is followed by hypertension and renal and cardiovascular lesions during adrenal cortical regeneration (Skelton, 1959).

All these lines of evidence can be interpreted as supporting the concept advanced by Selye of the role of the adrenal cortices in the aetiology of the so-called adaptation diseases, especially when one considers that many stressful situations cause at least a temporary increase in the secretory activity of the adrenal cortices.

The first weakness of this concept is that most of the supporting evidence is derived from conditions which do not occur naturally; the second is the absence of any convincing evidence that naturally occurring stressors cause adaptation diseases under natural conditions. Even if such evidence were available, it would remain to be shown that the adrenal glands must be present in order for the disease to be produced. We believe that in many instances in both animals and man the relationship of the adrenal cortical hormones to the manifestation of certain metabolic and morphological responses can best be described as permissive or supporting rather than directly causative. The conditions under which the term permissive is applied are as follows: (1) The response to a stimulus occurs in the presence of the adrenal gland. (2) The response fails to become overt when the gland is removed and no replacement therapy is given. (3) The response is again elicited by the appropriate stimulus when a steady intake of hormone is substituted for the adrenal. The hypothesis that the response is mediated solely by an increase in the secretory activity of the adrenal is made untenable by this third condition in which the presence of constant small doses of adrenal cortical hormone, but not the gland itself, are found to permit or support the responses.

The purpose of this paper is to present a coordinated account of our own studies and ideas on the role of the adrenal glands in the aetiology of experimentally induced diseases. Details of many of the experiments, to which repeated references will be made, have already been published. In most experiments male rats of the Sprague-Dawley strain have been used. They were maintained on commercial diets until the beginning of experiments; then we used a fluid medium carbohydrate suitable for either force-feeding or pair-feeding (Table I). The experiments were conducted in an air-conditioned laboratory with the temperature 74° to 78°F and the humidity at approximately 50 per cent of saturation. Blood pressure measurements were made by a direct arterial cannulation technique at a constant level of light anaesthesia, and the tissues were examined microscopically using a variety of histological and histochemical methods.

Table I. Medium carbohydrate diet. A fluid diet suitable for tube-feeding is prepared by the addition of water to a total volume of 2,000 ml. When a salt diet is used the extra sodium chloride is expressed as a percentage of the dry diet.