Hormonal Changes During Pregnancy

Hormonal Changes During Pregnancy

Hormonal Changes During Pregnancy JOSEPH W. JAILER, PH.D., M.D.* DONALD LONGSON, M.D.** PREGNANCY is a limited condition which is characterized by pr...

1MB Sizes 3 Downloads 55 Views

Hormonal Changes During Pregnancy JOSEPH W. JAILER, PH.D., M.D.* DONALD LONGSON, M.D.**

PREGNANCY is a limited condition which is characterized by profound physiological and biochemical changes. Perhaps the most readily detectable changes occur in the secretions of the various hormones: gonadotropins, estrogens, progesterone, adrenal steroids and perhaps thyroxine. The placenta is an extremely active endocrine gland and is definitely responsible for the secretion of the chorionic gonadotropin, estrogens and progesterone and may possibly play some role in the secretion of some of the adrenal steroids. In addition there is evidence of increased secretion of the adrenal cortex and the thyroid gland. This short review summarizes some of the hormonal changes which occur during pregnancy. PROGESTERONE

Progesterone, the hormone of the corpus luteum and of the placenta, is not excreted in the urine in detectable amounts. However, certain metabolites appear in the urine after the administration of progesterone and are also known to be present during the luteal phase of the menstrual eyrIe and during pregnancy. These compounds, chiefly pregnanediol, pregnanolone, and allo-pregnanolone, occur as the water-soluble glucuronides and are known, collectively, as the "pregnanediol complex." Since the pregnanediol complex accounts for 10 to 20 per eent of administered progesterone, the total amount of such compounds in the urine is taken as an indirect measurement of endogenous progesterone production. Several methods exist for the estimation of urinary pregnanediol. The well known method of Venning l (1938) is a gravimetric procedure which estimates the pregnanediol and pregnanolone glucuronides as they exist in the urine. The accuracy of the method depends on the purity of the final extract and, hence, precision is lost when dealing with small amounts-a criticism which applies to all current methods. From the CoUege of Physicians and Surgeons, Columbia University, and the Presbyterian Hospital, New York, N. Y. * Associate Professor of Clinical Medicine and Associate Attending Physician. ** Dickenson Traveling Fellow, Columbia University, and Senior Registrar, Manchester Regional Hosp1:tal Board, Manchebter, England. 341

842

Joseph W. Jailer, Donald Longson

Another widely used method (Sommerville et al} 1948) employs acid hydrolysis to liberate the free steroids; the purified extract is then subjected to a sulfuric acid reaction and estimated colorimetrically. Since the 20-ketosteroids exhibit a very low chromogenicity with sulfuric acid, the method is relatively specific for the pregnanediol moiety of the complex. Numerous contaminants, however, give rise to color on the addition of sulfuric acid and adversely affect the specificity of the test at low values. Coyle et al. 3 (1955) have recently compared the results obtained on late pregnancy urine by four different methods; owing to the widely divergent values, they concluded that these methods are grossly inaccurate and should be discarded. With one exception, attempts to detect progesterone in plasma have been unsuccessful. Hooker and Forbes 4 (1949) improved a pre-existing assay for progestational activity. In this method the material under assay is introduced directly into the uterine cavity of the mouse and the end point consists of certain histological changes in the stromal nuclei of the endometrium. Amounts of progesterone as small as 2 times 10-4 /Lg. can be detected in this way. In a single sample of pregnancy plasma' the authors found progestational activity equivalents to 5.5 /Lg. progesterone per ml. Later (Forbes,5 1951), further observations during pregnancy demonstrated progestational activity equivalent to less than 2 /Lg. progesterone per ml., although Morris 6 (1952), using a polarographic technique of great sensitivity, was unable to detect progesterone in the plasma even in late pregnancy. Zander 7 (1954), using chemical methods involving paper chromatography, color reactions and infrared spectrophotometry, found 0.15 /Lg. per ml. during the second half of pregnancy. No fluctuations in this level were observed from day to day. Within five minutes after the administration of 200 mg. of progesterone intravenously the value was 1.48 /Lg. per ml. only to fall two hours later to 0.28 /Lg. per ml. In view of the current views concerning the intrahepatic metabolism of progesterone, it is of interest that Mossis 6 (1952) reported detectable amounts of progesterone in the peripheral blood of a pregnant woman who developed infective hepatitis. Much evidence points to the fact that progesterone disappears from the blood with great rapidity. It would be of great interest to know whether such disappearance is due entirely to reduction and conjugation, or whether there occurs storage of the free hormone in the tissues. Pregnanediol is not excreted during the follicular phase of the menstrual cycle. At ovulation, and coinciding with the rise in the basal body temperature, pregnanediol suddenly appears in the urine and reaches a peak value of under 10 mg. per day in the midluteal phase. There follows a fall in excretion and zero values are commonly observed at the onset of menstruation. The total excretion of pregnanediol during the menstrual cycle is of the order of 50 mg., which corresponds to the production of 250 to 500 mg. of progesterone. When pregnancy supervenes, the menstrual

Hormonal Changes During Pregnancy 140

. .... . . ...

120

100

% 80 ~

!3

i

60

40

20

40

80

120

160 DAYS

200

240

Fig. 176. Excretion of pregnanediol complex in pregnancy. (From Venning, E. H., Obst. & Gynec. Surv. 3: 661, 1948.)

pattern of pregnanediol excretion merges into a gradually rising curve, which reaches a peak at or near term (Fig. 176). During the first ten weeks or so, the values remain, broadly speaking, within the range found during normal menstrual cycles. Since persistence of the corpus luteum, or the occurrence of a corpus luteum cyst, could also account for such values, and because of the lack of specificity of the chemical methods, it has not been possible to establish a test for the early diagnosis of pregnancy on the basis of the pregnanediol output, although this has been attempted. Following the peak in the excretion of chorionic gonadotropin, which occurs at the end of the first trimester, the pregnanediol output increases more rapidly until a peak or a plateau is attained during the eighth month. This definite upward trend, occurring between the 70th and lOOth day, is usually attributed to the placenta assuming steroidogenic function. However, it seems that the placenta may produce progesterone at a much earlier stage of pregnancy; removal of the corpus luteum during the second month of gestation is compatible with the excretion of normal, or near normal, amounts of pregnanediol throughout pregnancy (Venning,S 1948) . The precise pattern of pregnanediol excretion during the last few weeks of pregnancy has been the subject of some debate. According to some investigations, a peak occurs during the eighth month and is followed by a gradual fall until term (de Watteville,9 1951). Venning S (1948) could find no distinctive pattern during this period. Similarly, Trolle1o (1955) found no clear trend and emphasized the enormous differences in pregnanediol excretion which occur from patient to patient and, indeed, in

Joseph W. Jailer, Donald Longson 140

\20

100

80

60

40

20

DAYS

30

25

2~

15

10

5

ANTE PARTUM

Fig. 177. Excretion of pregnanediol complex followed in individual patients during the last few weeks of pregnancy. Note especially the marked day-to-day fluctuations in output. (From Venning, E. H., Obst. & Gynec. Surv. 3: 661, 1948.)

the same patient from day to day. This is illustrated in Figures 176 and 177. These fluctuations do not appear related to urine volumes or to any obvious clinical feature of the pregnancy. However, it must be appreciated that such fluctuations may be a reflection of the lack of accuracy of the available chemical methods. Indeed, Coyle et alY (1956), reporting on the use of a more specific chromatographic method, were unable to confirm the existence of large day-to-day variations in output. All authors agree that the excretion rate falls to zero within three to six days after delivery. Despite earlier reports, no correlation has been demonstrated between pregnanediol output and placental weight (Trolle,lO 1955). The possible relationship of pregnanediol excretion to various pathological states in pregnancy has been extensively explored. Browne, Henry and Venning 12 (1939) were the first to call attention to low pregnanediol values in threatened abortion. Guterman13 (1945) and Koff and Tulsky 14 (1953) have suggested that, if repeated daily determinations of pregnanediol excretion show persistently low values (i.e., below 5 mg. per day) up to the twelfth week of pregnancy, the likelihood of miscarriage is very high. Others, however, have failed to confirm this finding (Venning,!" 1952). One of the factors which invalidates this approach is the great day-to-day variability of the excretion rates, which has been mentioned previously, and the considerable overlapping between pregnant and nonpregnant values in the earlier months of pregnancy. The alleged decrease in urinary pregnanediol value is the basis for the

Hormonal Changes During Pregnancy use of large amounts of progesterone or the longer acting progestins (17-0H-progesterone caproate) in threatened and habitual abortion. The results in the various clinics have not been consistent. The fetal salvage rate in some clinics has been no higher in the treated group as compared to a concurrently run control group. The role of progesterone in this condition has not been settled as yet. There is even more confusion concerning the relationship of pregnanediol excretion to toxemia of pregnancy. In 1938, Browne, Henry and Venning16 described an abnormally low pregnanediol excretion in most of their cases of toxemia, though no correlation could be found between the individual values and the severity of the disease. Smith and Smithl7 • 18 (1940 and 1954), on the basis of similar findings, expounded their well known theory concerning the etiology of toxemia. The hormonal changes, which they considered characteristic of this disease, consisted of a diminished output of pregnanediol and estrogens, together with an enhanced output of chorionic gonadotropin. Similar findings led de Watteville 9 (1951) to conclude that a prognosis could be made for the child on the basis of such data. He advocated termination of pregnancy when the values remained low despite conservative treatment. Much of the extensive work on this topic is far from conclusive. While it does seem that pregnanediol excretion falls in some patients with toxemia, the relationship of this phenomenon to fetal prognosis, or to the etiology and severity of the toxemia, is quite obscure. Many factors, apart from progesterone production, have to be considered in interpreting such results. For instance, defective conjugation of pregnanediol in the liver and failing renal function could produce identical results. ESTROGENS

Most of the available information concerning the excretion of estrogens during pregnancy is derived from the use of relatively crude methods of biological assay. Since the estrogens are excreted as relatively or totally inactive conjugates, it follows that these methods may seriously underestimate the amount of such substances originally present in the urine. Three estrogens have been isolated from human tissues and body fluids: estradiol 17{3, estrone and estriol. Estradiol (the most active compound) and estrone are interconvertible, the administration of one leading to the urinary excretion of the other. Estriol, however, is derived by an irreversible reaction from either of the other two compounds and, consequently, is probably a metabolite of the others. Estrogen synthesis is known to occur in the ovary, the adrenal cortex and the testes. Evidence of placental synthesis is provided by the experiments of Stewart 19 (1951), who demonstrated estrogenic stimulation of the uterus in spayed rabbits with intraocular implants of full-term human placenta. Chemical extraction procedures have demonstrated the presence of all three estrogens in the piacenta20 (Dicfalusy, 1953; Mitchell

Joseph W. Jailer, Donald Longson and Davies,21 1954). Estriol is the most abundant steroid with concentration of the order of 150-200 JLg/kg; estrone concentration is about half this value, while estradiol is present in only trace amounts. A similar preponderance of estriol had been noted in pregnancy urine (Smith and Smith,22 1941). Most of our information concerning the quantitative aspect of estrogen excretion during pregnancy has been obtained by means of biological assay procedures. All authors are agreed that the urinary values increase slowly in early pregnancy and that a sharp rise occurs from the sixth month onwards (Cohen et al.,23 1935; Venning,S 1948; Bradshaw and Jessop, 24 1953). The behavior of the urinary estrogens prior to the onset of labor is variable and unpredictable. In many cases a fall is observed during the two or three days preceding labor, but similar or greater variations in output may have occurred previously in the same patients without there having been any signs of labor. It is not possible, therefore, to draw any conclusions concerning the mechanism which initiates labor. The difficulties which beset the interpretation of the estrogen curves are very similar to those which have been described with urinary pregnanediol. Dorfman and van Wegenen26 (1941), using monkeys, showed that removal of both the ovaries and the fetus is ineffective in altering the excretion rate, indicating a major extragonadal source of these steroids, presumably the placenta. A similar bilateral ovariectomy in humans does not preclude the continuation of pregnancy. Many cases of late pregnancy toxemia are associated with low estrogen output. The significance of this feature in relation to prognosis is not yet clear. Improved analytical methods, such as that described by Brown26 (1955), may shed more light on the precise pattern of estrogen excretion in the course of normal and pathological pregnancies. CHORIONIC GONADOTROPIN

Chorionic gonadotropin, elaborated by cytotrophoblast, appears in the urine in clearly detectable amounts about two weeks after conception, i.e., 35 to 40 days after the first day of the last menstrual flow. The conventional tests for the diagnosis of pregnancy depend on the presence of chorionic gonadotropin in urine and are designed in such a way as to detect an arbitrary concentration clearly above those amounts which may be found in the urine from nonpregnant individuals. A rapid and spectacular rise in urinary chorionic gonadotropin is the earliest detectable hormonal change in pregnancy. Between the seventh and twelfth week the excretion reaches values between 30,000 and 400,000 LU. per day and the serum concentration may vary between 200 and 600 I.U./ml. At about the fifteenth week a rapid fall occurs and, thereafter, the urinary values remain fairly constant until term (5000 to 10,000 I.U.j24 hours). A second minor peak has occasionally been described shortly before parturition, but this is an inconstant feature. Since the production

Hormonal Changes During Pregnancy

347

of chorionic gonadotropin is a function of the trophoblast, it follows that interruption of pregnancy, whether at term or prematurely, leads to the disappearance of the hormone from the urine, i.e., the pregnancy test becomes negative. However, the primary cause of abortion may reside in other structures, i.e., the uterus, and trophoblast activity may continue for a while and the excretion of chorionic gonadotropin may be unaltered or moderately decreased. The pregnancy test, being primarily a qualitative test, fails to detect such changes. Zondek et alY (1948) have described a method for the rough titration of chorionic gonadotropin in urine which, it is claimed, is of considerable prognostic value in cases of threatened abortion. In the last trimester, fetal death is almost certain if the concentration falls to one-tenth the normal level. The importance of following trends rather than on relying on single assay is particularly stressed. The hormonal changes occurring during pregnancy were first ascertained very early in the history of modern endocrinology. However, so far as the estrogens-progesterone and the chorionic gonadotropin-are concerned, very little fundamental knowledge has been added during the past ten years while great strides have been taken in thyroid and adrenal physiology and chemistry. These latter advances have been due primarily to the development of sensitive and specific methods of estimating the thyroid hormones and adrenal steroids in urine and plasma. ADRENAL CORTICAL STEROIDS

Enlargement of the adrenal cortices occurs in certain animal species during pregnancy. It is not known whether a similar change occurs in the human glands. However, changes occur in certain commonly used indices of adrenal function during pregnancy; the most important of these will be discussed in this article. 17-}(etosteroids

Frequent estimations of the urinary 17-ketosteroids during normal pregnancy reveal progressively rising values, reaching a peak in the third trimester. However, the Zimmerman reaction which is commonly employed to measure these urinary metabolites is not absolutely specific for the ketone group at the 17th carbon atom. The ketone group present at the 20th position in certain metabolites of progesterone (such as pregnanolone) interferes with the reaction, leading to falsely high values. In actual fact, when the individual 17-ketosteroids are separated by chromatography and identified, an actual fall is observed in the output of certain 17-ketosteroids (androsterone and etiocholanolone) (Dobriner et al.,28 1948). The 17-ketosteroids oxygenated at the 11th position are excreted in normal amounts. The reactivity of the C-20 compounds in the Zimmerman test explains the high total 17-ketosteroid values seen

Joseph W. Jailer, Donald Longson whcn pregnancy supervenes in the patient with Addison's disease. 29 Migeon30 (1954) has demonstrated an appreciable fall in plasma dehydroisoandrosterone in maternal plasma during pregnancy.

C-21 Adrenal Steroids It is possible to demonstrate an increase in the glucocorticoid activity of urinary extracts during pregnancy. Venning 29 (1946) demonstrated the ability of such extracts to cause an increase in the deposition of glycogen in the liver of adrenalectomized mice, a property shared by adrenal steroids of the hydrocortisone series. The application of chemical tests for urinary corticoids (Tobian,31 1948) has amply confirmed these observations and there seems no reason to doubt that they indicate a trne increase in the output of adrenal type steroids. This, indeed, becomes a probability when one considers the data obtained from the study of plasma corticoids. A rise in plasma corticoids in the third trimester has been established beyond all reasonable doubt, using both the NelsonSamuel (Gemzell,32 1953) and the Porter-Silber (Christy et al.,33 1955; Bayliss,Z4 1955) methods. By chromatographic separation and measurement it can be demonstrated that these high values are due to the presence of approximately twice the normal amount of circulatory free hydrocortisone, i.e., 17 to 25 mg. per 100 cc. Physiological evidence of this increase in hydrocortisone in the plasma is suggested in such phenomena as the occurrence of a diabetic glucose tolerance curve in patients without pre-existing clinical diabetes mellitus, or the aggravation of true diabetes mellitus with advancing pregnancy. For a number of reasons the adrenal origin of the excess hydrocortisone cannot be taken for granted. The placenta is an active endocrine organ capable of the synthesis of very large amounts of steroid hormone, of which progesterone is, of course, the most important. Although there is as yet no evidence from either in vivo or in vitro studies that the placenta can synthesize compounds with a dihydroxyacetone side chain, such a possibility cannot be excluded. Observations during pregnancy in Addisonian patients are pertinent in this discussion. Jailer and Knowlton35 (1950) showed an increase in neutral reducing lipid excretion in such a case and the phenomenon has been confirmed using chemical techniques. The table on page 350 illustrates the plasma corticoid values in two patients with Addison's disease during their pregnancies. In nonpregnant individuals the normal range, as estimated by this technique, is 8 to 28 J..lg. per 100 cc. It will be seen that the values rose to the upper half of the normal range in these two patients, though, as will be mentioned later, the administration of ACTH failed to produce a further rise. It has not been possible to identify the substance responsible for these high values in Addisonian patients. Should hydrocortisone be present under these circumstances, it would, presumably, be derived from either the placenta or the fetus. In considering the claim that cer-

Hormonal Changes During Pregnancy PREGNANCY, THIRD TRIMESTER 115 110 105 100

95 C

~

~

"-

90 85 80

Ii

75

<;;

...,.e

70

,.

65

0 0


50

t?

0

45

IIt: 0

40

I

35

!.! u

is t:

«

:I!

30 25

UI

20

...l

15

«

"-

10 5 0

o

234

HOURS

Fig. 178. The hyperresponse of the adrenal to exogenous ACTH during pregnancy. The shaded area is the normal response.

tain Addisonian patients require less substitution therapy during pregnancy, it must be remembered that progesterone itself has been shown to exert some life-maintaining activity in adrenalectomized animals. Despite the foregoing evidence, there is a strong suggestion that in normal pregnancy most, if not all, the increment in hydrocortisone is of adrenal origin. Adrenal function may be readily investigated by means of the intravenous infusion of ACTH under standardized conditions. In one method, 25 mg. of ACTH in 500 ml. isotonic dextrose is administered over a four hour period, and the plasma corticoids, measured by the Porter-Silber method, rise from the normal resting range of 8 to 28 /Lg. per 100 cc. to a value between 35 to 55 /Lg. per 100 cc. When this procedure is applied to patients in the third trimester of pregnancy the increment in plasma corticoids is much larger, the level at four hours being between 70 and 111 /Lg. per 100 cc. (Fig. 178). These higher values can be shown by chromatography to reflect correctly the level of circulating hydrocortisone. The abnormally large response to ACTH could be due (1) to the presence of hydrocortisone synthesis in some site remote from the adrenal

Joseph W. Jailer, Donald Longson

350

(i.e., the placenta); (2) to a decreased rate of metabolism of hydrocortisone in pregnancy, or (3) to an increased sensitivity of the adrenal cortex to ACTH. The first of these possibilities is ruled out by data obtained in the pregnant Addisonian patients. The accompanying table shows the results Table 17-HYDROXy-CORTICOSTEROID VALUES IN Two PREGNANT ADDISONIAN PATIENTS PLASMA CORTJCOIDS (tLg./lOO Cc.)

R.L.

M.K. First pregnancy 3/30/54 4/5 6/7 7/6 Delivered 7/8/54 7/13/54 7/21 7/23 Second pregnancy 3/21/55

3/31 5/26

7/7 7/21 8/23 8/29 9/2 9/lO lO/4-Delivered 11/1

PLASMA CORTICOIDS (tLg./lOO Cc.) 4/29/55

6

6/8

18 23 19

6/29 10/11 11/1 11/9

3 20 26

8 13 20->ACTH-> 6

19 10 2

7 11 23 12 6 12 23->ACTH->14 14 12 14

o

of ACTH tests in two such individuals. Despite the fact that the resting values of Porter-Silber chromogen is elevated above the usual Addisonian level, ACTH did not elicit any further rise. Hence, the response seen in normal pregnancy does not depend on the presence of the placenta or fetus. There is evidence that there may be a slight delay in the disappearance rate of administered hydrocortisone during the third trimester of pregnancy, but this alone cannot account for the elevated plasma levels found during pregnancy. Apparently the third possibility is the mechanism responsible for this phenomenon. Aldosterone

The occurrence of edema during pregnancy and the well known relationship between sodium intake and toxemia of pregnancy have prompted several investigations into the role of the adrenocortical mineral corticoid. Gordon et aJ.36 (1954) found a slight increase in salt-retaining activity during normal pregnancy and a marked increase in toxemia. Recently,

Hormonal Changes During Pregnancy

351

Venning and Dyrenfurth37 (1956), employing more refined methods for the hydrolysis of conjugates, have demonstrated equally elevated values for total aldosterone excretion in both normal and toxemic pregnancies. Their results indicate, however, that patients with toxemia excrete a larger proportion of the total aldosterone output as the free (active) steroid. Corroborative evidence of aldosterone activity is found in the low Na/K ratio found in toxemia (Soiva and Parviainen,38 1955). However, in our clinic we could not detect aldosterone in the urine of an Addisonian patient with definite toxemia of pregnancy. It is difficult in view of this observation to see how aldosterone can play any etiological role in toxemia of pregnancy. REFERENCES 1. Venning, E. H.: Further Studies on Estimation of Small Amounts of Sodium Pregnanediol G1ucuronidate in Urine. J. BioI. Chern. 126: 595, 1938. 2. Sommerville, I. F., Gough, N. and Marrian, G. F.: Quantitative Determination of Small Amounts of Pregnanediol in Human Urine. J. Endocrinol. 5: 247, 1948. 3. Coyle, M. G., Mitchell, F. L., Russell, C. S. and Paine, C. G.: Errors in Determination of Urinary Pregnanediol. J. Obst. & Gynaec., Brit. Emp. 62: 291, 1955. 4. Hooker, C. W. and Forbes, T. R.: Transport of Progesterone in Blood. Endocrinology 44: 61, 1949. 5. Forbes, T. R.: Systemic Levels of Plasma Progesterone During Pregnancy in Women and Monkeys. Endocrinology 49: 218, 1951. 6. Morris, C. J. O. R.: Blood Progesterone in Pregnancy. Ciba Colloquia for Endocrinol., Vol. II, 1952, p. 359. 7. Zander, J.: Progesterone in Human Blood and Tissues. Nature 174: 404, 1954. 8. Venning, E. H.: Excretion of Various Hormone Metabolites in Normal Pregnancy. Obst. & Gynec. Surv. 3: 661, 1948. 9. Watteville, H. de: Pregnanediol Determination in the Clinic and in Research. J. Clin. Endocrinol & Metab. 11: 251, 1951. 10. Trolle, D.: Experimental and Clinical Investigation on the Pregnanediol Excretion in Human Urine. Acta Endocrinol. 19: 217, 1955. 11. Coyle, M. G., Mitchell, F. L. and Russell, C. S.: A Report on the Chromatographic Assay of Urinary Pregnanediol in Pregnancy. J. Obst. & Gynaec., Brit. Emp. 63: 560, 1956. 12. Browne, J. S. L., Henry, J. S. and Venning, E. H.: Significance of Endocrine Assays in Threatened and Habitual Abortion. Am. J. Obst. & Gynec. 38: 927, 1939. 13. Guterman, H. S.: Further Observations on Values of Pregnanediol Test for Pregnancy. J. Clin. Endocrinol. 5: 407, 1945. 14. Koff, A. K. and Tulsky, A. S.: Threatened Abortion. S. CLIN. NORTH AMERICA 33: 3,1953. 15. Venning, E. H.: Ciba Colloquia for Endocrinol., Vol. II, 1952. 16. Browne, J. S. L., Henry, J. S. and Venning, E. H.: Urinary Excretion of Prolan, Estrin and Pregnanediol in Normal Pregnancy and in Early and Late Pregnancy Toxemias. J. Clin. Invest. 17: 503, 1938. 17. Smith, G. V. S. and Smith, O. W.: Estrogen and Progesterone Metabolism in Pregnant Women. Am. J. Obst. & Gynec. 39: 405, 1940. 18. Smith, G. V. S. and Smith, O. W.: Internal Secretions and Toxemia of Late Pregnancy. Physiol. Rev. 28: 1, 1948. 19. Stewart, H. L.: Hormone Secretion by Human Placenta Grown in Eyes of Rabbits. Am. J. Obst. & Gynec. 61: 990, 1951. 20. Dicfalusy, E.: Chorionic Gonadotrophins and Estrogens in Human Placenta. Acta Endocrinol., Vol. 12 (Supplement), 1953.

352

Joseph W. Jailer, Donald Longson

21. Mitchell, F. L. and Davies, R. E.: Isolation and Estimation of Steroid Estrogens in Placental Tissue. Biochem. J. 56: 690, 1954. 22. Smith, G. V. S. and Smith, O. W.: Estrogen and Progestin Metabolism in Pregnancy. II. J. Clin. Endocrinol. 1: 470, 1941. 23. Cohen, S. L., Marrian, G. F. and Watson, M.: Excretion of Estrin During Pregnancy. Lancet 1: 674, 1935. 24. Bradshaw, T. E. T. and Jessop, W. J. E.: Urinary Excretion of Estrogens and Pregnanediol at End of Pregnancy, During Labour and During Early Puerperium. J. Endocrinol. 9: 427, 1953. 25. Dorfman, R. 1. and van Wegenen, G.: Sex Hormone Excretion of Adult Female and Pregnant Monkeys. Surg., Gynec. & Obst. 73: 545, 1941. 26. Brown, J. B.: A Chemical Method for Determination of Estriol, Estrone and Estradiol in Human Urine. Biochem. J. 60: 185, 1955. 27. Zondek, B., Sulman, F. and Black, R.: Hormone Test for Fetal Death in Disturbed Pregnancy. J.A.M.A. 136: 965, 1948. 28. Dobriner, K., Liberman, S., Rhoads, C. P. and Taylor, H. C., Jr.: Urinary Excretion of Ketosteroids in Pregnancy. Obst. & Gynec. Surv. 3: 677, 1948. 29. Venning, E. H.: Adrenal Function and Pregnancy. Endocrinology 39: 243,1946. 30. Migeon, C. J.: Dehydroisoandrosterone and Androsterone Levels in Maternal and Cord Plasma. A Symposium. Adrenal Function in Infants and Children. Syracuse, New York, 1954, pp. 96-101. 31. Tobian, L.: Cortical Steroid Excretion in Edema in Pregnancy Pre-eclampsia and Essential Hypertension. J. Clin. Endocrinol. 9: 319, 1948. 32. Gernzell, C. A.: Blood Levels of 17 Hydroxycorticosteroids in Normal Pregnancy. J. Clin. Endocrinol. 13: 898, 1953. 33. Christy, N. P., Wallace, E. Z. and Jailer, J. W.: Effect of Intravenously Administered ACTH on Plasma, 17, 21-Dihydroxy-20-Ketosteroids in Normal Individuals and in Patients with Disorders of the Adrenal Cortex. J. Clin. Invest. 34: 899, 1955. 34. Bayliss, R. 1. A., et al.: Plasma 17-Hydroxycorticosteroids in Pregnancy. Lancet 1: 62,1955. 35. Jailer, J. W. and Knowlton, A. 1.: Simulated Adrenocortical Activity During Pregnancy in an Addison,ian Patient. J. Clin. Invest. 29: 1430, 1950. 36. Gordon, E. S., Chart, J. J., Hagedorn, D. and Shipley, E. G.: Mechanism of Sodium Rf'tention in Preeclamptic Toxemia. Obst. & Gynec. Surv. 4: 39, 1954. 37. Venning, E. H. and Dyrenfurth, 1.: Aldosterone Excretion in Pregnancy. J. Clin. Endocrinol. & Metab. 16: 426, 1956. 38. Soiva, K. and Parviainen, S.: On the Sodium, Potassium and Chloride Concentrations of Salvia in Pregnancy and in Toxemia of Late Pregnancy. Acta Endocrinol. 20: 161, 1955. 630 W. 168th Street New York 32, N. Y.