The possibility that estrogens might under certain circumstances influence water and electrolyte metabolism is suggested by several observations. For instance, estrogens are related chemically to the strongly sodium-retaining steroid desoxycorticosterone (1); the recurrent ankle edema observed in some women before menstruation was considered by Fluhmann (2) to be associated with an increase in blood estrogen levels; and the edema frequently found in late pregnancy occurs at a time when estrogen production is at its height (3).These considerations led several workers to attempt to establish a direct relationship between water and salt retention on the one hand, and estrogenic activity on the other. Thorn and Engel (4) found sodium, chloride and water retention after a single injection of estrogen into male dogs. Later, the same group of workers (5), in a study of premenstrual fluid retention, found that decreased urinary excretion of salt and water was in two patients associated with an increased output of estrogenic material in the urine. Knowlton, Kenyon, Sandiford, Lotwin, and Fricker (6) found diminished urinary excretion of sodium following the daily administration of estradiol in one normal woman, but Sharpey-Schafer and Schrire (7), using twice the amount of estradiol in ten women, found no alteration in urinary volume. As part of an investigation into the possible role of estrogens in the fluid retention of liver disease it became necessary to establish the effect of continued estrogen administration on water and electrolyte metabolism in the normal. Water and electrolyte balance studies were therefore carried out in eleven normal subjects, and the effect of daily administration of estrogen observed.
PROCEDUREEleven subjects without evidence of cardiac, hepatic, renal or other disease likely to affect water and electrolyte metabolism, were selected from the wards of The London Hospital. Five subjects were under treatment for peptic ulcer, during which time no medication other than magnesium trisilicate or aluminum hydroxide was given. Two subjects had sciatic nerve pain, one osteoarthritis of the hip, one torn medial meniscus of the knee, one hallux valgus, and in one a tuberculous pleural effusion had subsided three weeks previously.Each subject underwent a water and electrolyte balance study, lasting approximately two weeks and divided into two periods-the first a "control period," and the second an "estrogen period," during which daily injections of estrogen were given. The subjects were given a normal diet, gauged in amount and composition to individual appetite, and remaining constant throughout the study in respect of Cl, Na, K, PO, water and calorie content. Diet composition was calculated from appropriate tables (8). Sample dietary analyses were not carried out. The balance diet was commenced 2 to 3 days before any observations were made. Since this diet corresponded closely in composition to previous intake in each case, no longer period for metabolic adjustment was considered necessary. In ...