Patients with myxedema present an unusual opportunity for studying the effect of lowered metabolism on organ blood flow and for observing the effect of changes in peripheral circulation on cardiac output. Studies of cerebral metabolism in myxedema are of interest to the physician because changes in cerebral metabolism may be of importance in understanding the alterations in mental status which occur with this disease (1-3). was recorded with a Hamilton manometer. The nitrous oxide technique for measuring cerebral blood flow has been previously described (4), and the modification in use in this laboratory of collecting simultaneous, continuous samples from the artery and internal jugular bulb, rather than five separate arterial and venous samples, has been reported elsewhere, together with our normal values (7). In the cerebral metabolism study, blood oxygen content was measured by the spectrophotometric method of Hickam and Frayser (8); in the cardiac output study, it was measured by the manometric method of Van Slyke (9). Blood glucose content was measured by Nelson's modification of Somogyi's method (10); in most instances, each arterial cerebral venous glucose difference represents an average of four such arteriovenous differences drawn during the procedure. The calculation of cerebral oxygen consumption, cerebral glucose consumption, and cerebral vascular resistance has been previously described (4, 7). Cerebral venous oxygen tension was not measured directly, but computed from the per cent oxygen saturation of the cerebral venous blood using a standard oxyhemoglobin dissociation curve drawn for pH 7.4. Mean arterial pressures in the cerebral metabolism study were calculated from the formula MP = diastolic pressure plus pulse pressure the readings 3 being obtained at one minute intervals by the standard auscultation procedure with the arm held at heart level.
RESULTSCardiac output. The essential data are presented in detail in Table I. The patients with myxedema were compared with a group of normal subjects studied under the same laboratory circumstances (11). The mean decrease in cardiac index from nornal was 47 per cent, and the mean decrease in total oxygen consumption was 37 per cent. There was an excellent correlation between cardiac index and basal metabolic rate, the correlation coefficient (r) being 0.90. The apparent moderate increase in mixed arterial-venous oxygen difference proved not to be statistically significant. The two patients who were restudied after treatment showed sizable increases in both cardiac index and total oxygen consumption. 1139