- Email: [email protected]
Maternal prostacyclin, thromboxane, and placental blood flow
Maternal prostacyclin, blood flow 0.
thromboxane,
and placental
YLIKORKALA
P. JOUPPILA P. KIRKINEN L. VIINIKKA Helsinki, Finland The vasoactive prostanoids-prostacyclin (PGI,) and thromboxane A, (TX&)-and their metabolites-6-keto-prostaglandin F,, (CLketo-PGF,,) and thromboxane B2 (TxB2), respectively-have been implicated as regulators of uteroplacental blood flow in animals. To study their roles in placental blood flow in human beings, plasma or serum samples were collected for the measurement of 6-keto-PGF,, and TxB, from 42 women during late pregnancy on the same occasion, when placental intervillous blood flow (IVBF) was determined with 13sXe isotope method. The concentrations of 6-keto-PGF,, in plasma or those of TxB, in plasma or serum were not related to the IVBF. The intravenous infusion of ritodrine. for 1 hour up to the dose of 200 &min increased (p < 0.05) the plasma 6keto-PGF,, concentrations, but caused no changes in the TxB, levels or IVBF in seven women with premature uterine contractions. We conclude that if PGI, and TxA, participate in the control of placental blood flow, their changes are located in the fetoplacental unit and, thus, are not reflected by the levels of their metabolites in maternal circulation, and ritodrine may stimulate PGI, synthesis in human beings. (AM. J. OBSTET. GYNECOL. 14.5730, 1983.)
PROSTACYCLIN (PG12), the major active metabolite of arachidonic acid in the vascular wall, is a powerful vasodilator and inhibitor of platelet aggregation, whereas thromboxane AZ (TxA,), mainly synthesized in the platelets, has counteracting vasoconstrictive and proaggregative activities.’ A balance between PGI, and TxA, evidently regulates the vascular tone and platelet aggregation in the systemic circulation.’ There is abundant evidence from animal models that the PGI*/ TxA, balance also controls the uterofetoplacental blood flo~.~-~ That this might be the case in human beings as well is suggested by the fact that several pregnancy-associated human tissues, such as the cervix,
From the Departments of Obstetrics and Gynecology, University of Helsinki and University of Oulu, and the Department of Clinical Chemistry, University of Oulu. This study was supported Science.
by the Finnish
Received for publication
March
Revised August Accepted October
Academy
of
29, 1982.
30, 1982. 6, 1982.
Reprint requests: O&vi YlikorkaLa, M.D., Department Obstetrics and Gynecology, University of Helsinki, Haartmuninkatu 2, 00290 Helsinki 29, Finland. 730
oj
myometrium, placenta, decidua, and fetal membranes, and the umbilical cord vessels produce PG12 and/or TxA2 in vitro, as reviewed recently.’ To clarify the role of PGI, and TxAz in the control of the placental blood flow in human subjects, we studied the relationships between PGI,/TxA, productions and the placental blood flow in women with complicated late pregnancy.
Subjects and methods Subjects. The volunteers of this study were selected on the basis of intervillous blood flow (IVBF) from a larger patient series in which IVBF was followed systematically throughout pregnancy* so that 18 women with high IVBF (>140 mi/min/lOO ml of intervillous volume) and 24 women with low IVBF (< 100 ml/mini 100 ml of intervillous volume) were obtained. These groups were comparable in other respects (Table I). The pregnancy was complicated with pre-eclampsia (blood pressure > 140/90 mm Hg and/or proteinuria) in 22 women, with intrauterine fetal growth retardation (
P., et al.: To be published.
Volume
145
Prostacyclin, thromboxane, and placental blood flow ,
Number 6
the women with high and low IVBF. No participant had taken drugs known to interfere with the synthesis of prostaglandin during pregnancy. The study plan was approved by the local committee on ethics, and an informed consent was obtained from every volunteer. Measurement of intervillous blood flow. The IVBF was determined by means of the isotope ‘33Xe, as described and validated elsewhere.’ The result is expressed as blood flow in milliliters per minute per 100 ml of intervillous volume. Measurement of PGIg and TxA2. Because of the instability of PGI, and TxA,, their productions were evaluated bv the measurements of their stable metabolites, 6-ieto-prostaglandin F,, (6-keto-PGF1,) and thromboxane B, (TxB,), respectively. To obtain plasma, blood was taken into ice-cold heparinized vacuum tubes containing indomethacin (10 pmole/L final concentration), centrifuged immediately, and stored frozen (-25” C) until radioimmunoassayed for 6-ketoPGF I
Results 6-Keto-PGF,,. Plasma 6-keto-PGF,, ranged from 25 to 228 pg/ml without any correlation with the simultaneous IVBF in the whole series or in the subgroups studied (Table II). The ritodrine infusion was accompanied by increased plasma 6-keto-PGF,, concentrations, but by no changes in IVBF (Table III). TxB,. No relation was seen between plasma or serum concentrations of TxBz and IVBF in the whole series or in any subgroup (Table II). Ritodrine infusion did not change TxB, (Table III).
Comment Placental blood flow increases in parallel with the increase in placental weight and can be reduced considerably in some but not in all pregnancies complicated with preeclampsia.8s ll. I2 The factors responsible for
731
Table I. Clinical characteristics (mean + SE:) of study populations I (IVBF > 140 mUmin/ 100 ml of intervillous volume) and II (IVBF < 100 ml/ min/ 100 ml of intervillous volume)
No. of women Age (yr) Gestational age (wk) IVBF
croup I
GroupII
18
24 27.0 i 1.3
26.4 -t 1.3 37.7 k 0.7 216.4 t “0.3
--..--__
Yti. 1 5 0.6 73.2 -c 3.5
Table II. IVBF and the concentrations of 6-keto-PGF,, and ‘T’xB* in plasma or serum in women with high (group I) and low I\‘Bf; (group II)
IVBF (ml/min/100 ml of intervillous volume?) Plasma 6-keto-PCF,, (pg/ml) Plasma TxB2 ( pgiml) Serum ‘TxBI (rig/ml)
70.2 Ir 6.2 163.3 + 30.0 197.7 it 29.2
ti8.0 z!I 5.9 160.8 rt Pti.5 164.0 2 21.3
*No differences are seen in the concCntrati<)ns of prostanoids between the groups. tAl1 values are mean t SE. Table III. IVBF and the concentr‘ations 01‘ 6-ketoPGF1, and TxBf in plasma or serum before and after 1 hour’s infusion of ritodrine (up to ‘LOO pglmin) in seven women Before
IVBF (ml/min/lOO ml of intervillous volume*) Plasma 6-keto-PGF,, (pg/ml) Plasma TxB2 (pg/ml) Serum TxB? (rig/ml)
‘4fier
93.4 + 29.3
94.0 2 30.7
51.9 5 6.2 153.9 ? 38.3 298.2 * 32.2
ti8.0 If: 5.9t 134.3 t 8.3 “74.1 t 37.6
*All values are mean f SE. tP < 0.05 in comparison between “before” levels.
and “after”
these flow changes are unknown,‘” but much interest has been focused on the potently vasoactive prostanoids, PGI, and TxA2.‘-‘, I” .4n intraventricular injection of PGI, reduced the uterine blood fiow in pregnant ewes, but this was interpreted to be due to the reduction of systemic blood pressure.* When PGI, was injected into the uterine artery, the uterine blood flow increased in nonpregna& ’ and pregnant sheep,” and the injection of tranylcypromine, a PGIp synthesis inhibitor, by the same route resulted in ~asoccrlistriction in pregnant ewes.6 No data exist on the relationships between the uteroplacental blood Now and PGI, in human subjects, although a preliminary trial points
732
Ylikorkala
et al.
March Am. J. Obstet.
to the possible beneficial effects of the intravenous infusion of PGIz in pre-eclamptic w0men.l” These dataze7. 13, I4 suggest, but do not prove, a cause-andeffect relation between PGI, and uteroplacental vasculature. Our data show that the IVBF is unrelated to the maternal PG12, as assessed by the 6-keto-PGF,, levels in peripheral plasma. However, they do not exclude the possibility that local changes of PGI, production regulate the placental blood flow in human beings, because the representativeness of the maternal plasma 6-ketoPGF,, measurement as an index of fetoplacental PGIp release is not known. Specific inhibitors of PGI, synthetase should be available also for human studies, before the final role of PGI, in the regulation of placental blood flow is determined. Moreover, our data show that 6-keto-PGF,, itself in maternal plasma is not related to placental blood flow in human beings, although the injection of 6-ketoPGF,, into the uterine artery resulted in slight reduction in the uterine blood flow in pregnant sheep.j The plasma TxBz concentrations are supposed to reflect the circulating vasoconstrictory TxA,, although
15, I%% Gynecol.
the effect of the sampling on plasma TxBz cannot be ruled out. The amounts of TxBz in serum during spontaneous clotting illustrate the capacity of the platelets to generate TxA,. Both the plasma and serum TxBz levels are increased during pregnancy.‘j Although a direct infusion of TxBz into the uterine artery diminished the uterine blood flow in pregnant sheep,j the TxB, concentrations in maternal plasma or serum did not correlate with the simultaneous IVBF in human subjects in the present study. Naturally, it is possible that the TxB, concentrations in peripheral plasma do not detect the fetoplacental TxA,/TxB2 synthesis. In the present work we demonstrate for the first time that a beta-mimetic drug, ritodrine, stimulates PGT, generation. The tocolytic effect of ritodrine is, however, hardly mediated through PGI, increase, since the infusion of PGI, caused no changes in the uterine contractility of the pregnant women.” It is not known whether the PGI, rise is a primary effect of ritodrine or a secondary change to the pronounced maternal hemodynamic or hormonal and metabolic alternations during ritodrine administration.‘”
REFERENCES 1.
2.
3.
4.
5.
6.
Moncada, S., and Vane, J. R.: Arachidonic acid metabolites and the interactions between platelets and blood vessel walls, N. Engl. J. Med. 300: 1142, 1979. Rankin, J. H. G., Phernetton, T. M., Anderson, D. F., and Berssenbrugge, A. D.: Effect of prostaglandin I2 on ovine placental vasculature, J. Dev. Physiol. 1:151, 1979. Resnik, R., and Brink, G. W.: Uterine vascular response to prostacyclin in nonpregnant sheep, AM. J. OBSTET. GYNECOL. 137:267, 1980. Clark, K. E., Austin, J. E., and Stys, S. J.: Effect of bisenoic prostaglandins on the uterine vasculature of the nonpregnant sfieep, Prostaglandins 22:333, 1981. Clark. K. E.. Austin. 1. E.. and Seeds. A. E.: Effect of bisenoic prostaglandini and arachidonic acid on the uterine vasculature of pregnant sheep, AM. J. OBSTET. GYNECOL. 142:261, 1982. Clark, K. E., and Harrington, D. J.: Effect of the prostacyclin synthetase inhibitor tranylcypromine on uterine blood flow in pregnancy, Prostaglandins 23:227, 1982. Tuvemo, T.: Role of prostaglandins, prostacyclin, and thromboxanes in the control of the umbilical placental circulation, Semin. Perinatol. 4:91, 1980. Renkonen, A., Luotola, H., Pitkinen, M., Kuikka, J., and Pyiirili, T.: Measurement of intervillous and myometrial blood flow by an intravenous 133Xe method, Br. J. Obstet. Gynaecol. 83:723, 1976. Ylikorkala, O., Kaila, J., and Viinikka, L.: F’rostacyclin
and thromboxane in diabetes, Br. Med. J. 283:1148, 1981. 10. Viinikka, L., and Ylikorkala, 0.: Measurement of thromboxane B, in human plasma or serum using radioimmunoassay, Prostaglandins 20:759, 1980. 11. KPlr, K., Jouppila, P., Kuikka, J., Luotola, H., Toivanen, J., and Rekonen, A.: Intervillous blood flow in normal and complicated late pregnancy measured by means of an intravenous 133Xe method, Acta Obstet. Gynecol. Stand. 59:7, 1980. 12. Bruce, N. W., and Abdul-Karim, R. W.: Mechanism controlling maternal placental circulation, Clin. Obstet. Gynecol. 17:135, 1974. 13. Editorial: Pregnancy and the arachidonic-acid cascade, Lancet 1:997, 1982. 14. Fidler, J., Bennet, M. J., deSwiet, M., Ellis, C., and Lewis, P. J.: Treatment of pregnancy hypertension with prostacyclin, Lancet 2:31, 1980. 15. Ylikorkala, O., and Viinikka, L.: Thromboxane AZ in pregnancy and puerperium, Br. Med. J. 281: 1601, 1980. 16. Ylikorkala, O., Kauppila, A., Tuimala, R., Haapalahti, J., Karppanen, H., and Viinikka, L.: Effects of intravenous isoxsuprine and ritodrine with and without concomitant dexamethasone, on fetoplacental and pituitary hormones and cyclic adenosine monophosphate during late pregnancy, AM. J. OBSTET. GYNECOL. 130:302, 1978.
thromboxane,
and placental
YLIKORKALA
P. JOUPPILA P. KIRKINEN L. VIINIKKA Helsinki, Finland The vasoactive prostanoids-prostacyclin (PGI,) and thromboxane A, (TX&)-and their metabolites-6-keto-prostaglandin F,, (CLketo-PGF,,) and thromboxane B2 (TxB2), respectively-have been implicated as regulators of uteroplacental blood flow in animals. To study their roles in placental blood flow in human beings, plasma or serum samples were collected for the measurement of 6-keto-PGF,, and TxB, from 42 women during late pregnancy on the same occasion, when placental intervillous blood flow (IVBF) was determined with 13sXe isotope method. The concentrations of 6-keto-PGF,, in plasma or those of TxB, in plasma or serum were not related to the IVBF. The intravenous infusion of ritodrine. for 1 hour up to the dose of 200 &min increased (p < 0.05) the plasma 6keto-PGF,, concentrations, but caused no changes in the TxB, levels or IVBF in seven women with premature uterine contractions. We conclude that if PGI, and TxA, participate in the control of placental blood flow, their changes are located in the fetoplacental unit and, thus, are not reflected by the levels of their metabolites in maternal circulation, and ritodrine may stimulate PGI, synthesis in human beings. (AM. J. OBSTET. GYNECOL. 14.5730, 1983.)
PROSTACYCLIN (PG12), the major active metabolite of arachidonic acid in the vascular wall, is a powerful vasodilator and inhibitor of platelet aggregation, whereas thromboxane AZ (TxA,), mainly synthesized in the platelets, has counteracting vasoconstrictive and proaggregative activities.’ A balance between PGI, and TxA, evidently regulates the vascular tone and platelet aggregation in the systemic circulation.’ There is abundant evidence from animal models that the PGI*/ TxA, balance also controls the uterofetoplacental blood flo~.~-~ That this might be the case in human beings as well is suggested by the fact that several pregnancy-associated human tissues, such as the cervix,
From the Departments of Obstetrics and Gynecology, University of Helsinki and University of Oulu, and the Department of Clinical Chemistry, University of Oulu. This study was supported Science.
by the Finnish
Received for publication
March
Revised August Accepted October
Academy
of
29, 1982.
30, 1982. 6, 1982.
Reprint requests: O&vi YlikorkaLa, M.D., Department Obstetrics and Gynecology, University of Helsinki, Haartmuninkatu 2, 00290 Helsinki 29, Finland. 730
oj
myometrium, placenta, decidua, and fetal membranes, and the umbilical cord vessels produce PG12 and/or TxA2 in vitro, as reviewed recently.’ To clarify the role of PGI, and TxAz in the control of the placental blood flow in human subjects, we studied the relationships between PGI,/TxA, productions and the placental blood flow in women with complicated late pregnancy.
Subjects and methods Subjects. The volunteers of this study were selected on the basis of intervillous blood flow (IVBF) from a larger patient series in which IVBF was followed systematically throughout pregnancy* so that 18 women with high IVBF (>140 mi/min/lOO ml of intervillous volume) and 24 women with low IVBF (< 100 ml/mini 100 ml of intervillous volume) were obtained. These groups were comparable in other respects (Table I). The pregnancy was complicated with pre-eclampsia (blood pressure > 140/90 mm Hg and/or proteinuria) in 22 women, with intrauterine fetal growth retardation (
P., et al.: To be published.
Volume
145
Prostacyclin, thromboxane, and placental blood flow ,
Number 6
the women with high and low IVBF. No participant had taken drugs known to interfere with the synthesis of prostaglandin during pregnancy. The study plan was approved by the local committee on ethics, and an informed consent was obtained from every volunteer. Measurement of intervillous blood flow. The IVBF was determined by means of the isotope ‘33Xe, as described and validated elsewhere.’ The result is expressed as blood flow in milliliters per minute per 100 ml of intervillous volume. Measurement of PGIg and TxA2. Because of the instability of PGI, and TxA,, their productions were evaluated bv the measurements of their stable metabolites, 6-ieto-prostaglandin F,, (6-keto-PGF1,) and thromboxane B, (TxB,), respectively. To obtain plasma, blood was taken into ice-cold heparinized vacuum tubes containing indomethacin (10 pmole/L final concentration), centrifuged immediately, and stored frozen (-25” C) until radioimmunoassayed for 6-ketoPGF I
Results 6-Keto-PGF,,. Plasma 6-keto-PGF,, ranged from 25 to 228 pg/ml without any correlation with the simultaneous IVBF in the whole series or in the subgroups studied (Table II). The ritodrine infusion was accompanied by increased plasma 6-keto-PGF,, concentrations, but by no changes in IVBF (Table III). TxB,. No relation was seen between plasma or serum concentrations of TxBz and IVBF in the whole series or in any subgroup (Table II). Ritodrine infusion did not change TxB, (Table III).
Comment Placental blood flow increases in parallel with the increase in placental weight and can be reduced considerably in some but not in all pregnancies complicated with preeclampsia.8s ll. I2 The factors responsible for
731
Table I. Clinical characteristics (mean + SE:) of study populations I (IVBF > 140 mUmin/ 100 ml of intervillous volume) and II (IVBF < 100 ml/ min/ 100 ml of intervillous volume)
No. of women Age (yr) Gestational age (wk) IVBF
croup I
GroupII
18
24 27.0 i 1.3
26.4 -t 1.3 37.7 k 0.7 216.4 t “0.3
--..--__
Yti. 1 5 0.6 73.2 -c 3.5
Table II. IVBF and the concentrations of 6-keto-PGF,, and ‘T’xB* in plasma or serum in women with high (group I) and low I\‘Bf; (group II)
IVBF (ml/min/100 ml of intervillous volume?) Plasma 6-keto-PCF,, (pg/ml) Plasma TxB2 ( pgiml) Serum ‘TxBI (rig/ml)
70.2 Ir 6.2 163.3 + 30.0 197.7 it 29.2
ti8.0 z!I 5.9 160.8 rt Pti.5 164.0 2 21.3
*No differences are seen in the concCntrati<)ns of prostanoids between the groups. tAl1 values are mean t SE. Table III. IVBF and the concentr‘ations 01‘ 6-ketoPGF1, and TxBf in plasma or serum before and after 1 hour’s infusion of ritodrine (up to ‘LOO pglmin) in seven women Before
IVBF (ml/min/lOO ml of intervillous volume*) Plasma 6-keto-PGF,, (pg/ml) Plasma TxB2 (pg/ml) Serum TxB? (rig/ml)
‘4fier
93.4 + 29.3
94.0 2 30.7
51.9 5 6.2 153.9 ? 38.3 298.2 * 32.2
ti8.0 If: 5.9t 134.3 t 8.3 “74.1 t 37.6
*All values are mean f SE. tP < 0.05 in comparison between “before” levels.
and “after”
these flow changes are unknown,‘” but much interest has been focused on the potently vasoactive prostanoids, PGI, and TxA2.‘-‘, I” .4n intraventricular injection of PGI, reduced the uterine blood fiow in pregnant ewes, but this was interpreted to be due to the reduction of systemic blood pressure.* When PGI, was injected into the uterine artery, the uterine blood flow increased in nonpregna& ’ and pregnant sheep,” and the injection of tranylcypromine, a PGIp synthesis inhibitor, by the same route resulted in ~asoccrlistriction in pregnant ewes.6 No data exist on the relationships between the uteroplacental blood Now and PGI, in human subjects, although a preliminary trial points
732
Ylikorkala
et al.
March Am. J. Obstet.
to the possible beneficial effects of the intravenous infusion of PGIz in pre-eclamptic w0men.l” These dataze7. 13, I4 suggest, but do not prove, a cause-andeffect relation between PGI, and uteroplacental vasculature. Our data show that the IVBF is unrelated to the maternal PG12, as assessed by the 6-keto-PGF,, levels in peripheral plasma. However, they do not exclude the possibility that local changes of PGI, production regulate the placental blood flow in human beings, because the representativeness of the maternal plasma 6-ketoPGF,, measurement as an index of fetoplacental PGIp release is not known. Specific inhibitors of PGI, synthetase should be available also for human studies, before the final role of PGI, in the regulation of placental blood flow is determined. Moreover, our data show that 6-keto-PGF,, itself in maternal plasma is not related to placental blood flow in human beings, although the injection of 6-ketoPGF,, into the uterine artery resulted in slight reduction in the uterine blood flow in pregnant sheep.j The plasma TxBz concentrations are supposed to reflect the circulating vasoconstrictory TxA,, although
15, I%% Gynecol.
the effect of the sampling on plasma TxBz cannot be ruled out. The amounts of TxBz in serum during spontaneous clotting illustrate the capacity of the platelets to generate TxA,. Both the plasma and serum TxBz levels are increased during pregnancy.‘j Although a direct infusion of TxBz into the uterine artery diminished the uterine blood flow in pregnant sheep,j the TxB, concentrations in maternal plasma or serum did not correlate with the simultaneous IVBF in human subjects in the present study. Naturally, it is possible that the TxB, concentrations in peripheral plasma do not detect the fetoplacental TxA,/TxB2 synthesis. In the present work we demonstrate for the first time that a beta-mimetic drug, ritodrine, stimulates PGT, generation. The tocolytic effect of ritodrine is, however, hardly mediated through PGI, increase, since the infusion of PGI, caused no changes in the uterine contractility of the pregnant women.” It is not known whether the PGI, rise is a primary effect of ritodrine or a secondary change to the pronounced maternal hemodynamic or hormonal and metabolic alternations during ritodrine administration.‘”
REFERENCES 1.
2.
3.
4.
5.
6.
Moncada, S., and Vane, J. R.: Arachidonic acid metabolites and the interactions between platelets and blood vessel walls, N. Engl. J. Med. 300: 1142, 1979. Rankin, J. H. G., Phernetton, T. M., Anderson, D. F., and Berssenbrugge, A. D.: Effect of prostaglandin I2 on ovine placental vasculature, J. Dev. Physiol. 1:151, 1979. Resnik, R., and Brink, G. W.: Uterine vascular response to prostacyclin in nonpregnant sheep, AM. J. OBSTET. GYNECOL. 137:267, 1980. Clark, K. E., Austin, J. E., and Stys, S. J.: Effect of bisenoic prostaglandins on the uterine vasculature of the nonpregnant sfieep, Prostaglandins 22:333, 1981. Clark. K. E.. Austin. 1. E.. and Seeds. A. E.: Effect of bisenoic prostaglandini and arachidonic acid on the uterine vasculature of pregnant sheep, AM. J. OBSTET. GYNECOL. 142:261, 1982. Clark, K. E., and Harrington, D. J.: Effect of the prostacyclin synthetase inhibitor tranylcypromine on uterine blood flow in pregnancy, Prostaglandins 23:227, 1982. Tuvemo, T.: Role of prostaglandins, prostacyclin, and thromboxanes in the control of the umbilical placental circulation, Semin. Perinatol. 4:91, 1980. Renkonen, A., Luotola, H., Pitkinen, M., Kuikka, J., and Pyiirili, T.: Measurement of intervillous and myometrial blood flow by an intravenous 133Xe method, Br. J. Obstet. Gynaecol. 83:723, 1976. Ylikorkala, O., Kaila, J., and Viinikka, L.: F’rostacyclin
and thromboxane in diabetes, Br. Med. J. 283:1148, 1981. 10. Viinikka, L., and Ylikorkala, 0.: Measurement of thromboxane B, in human plasma or serum using radioimmunoassay, Prostaglandins 20:759, 1980. 11. KPlr, K., Jouppila, P., Kuikka, J., Luotola, H., Toivanen, J., and Rekonen, A.: Intervillous blood flow in normal and complicated late pregnancy measured by means of an intravenous 133Xe method, Acta Obstet. Gynecol. Stand. 59:7, 1980. 12. Bruce, N. W., and Abdul-Karim, R. W.: Mechanism controlling maternal placental circulation, Clin. Obstet. Gynecol. 17:135, 1974. 13. Editorial: Pregnancy and the arachidonic-acid cascade, Lancet 1:997, 1982. 14. Fidler, J., Bennet, M. J., deSwiet, M., Ellis, C., and Lewis, P. J.: Treatment of pregnancy hypertension with prostacyclin, Lancet 2:31, 1980. 15. Ylikorkala, O., and Viinikka, L.: Thromboxane AZ in pregnancy and puerperium, Br. Med. J. 281: 1601, 1980. 16. Ylikorkala, O., Kauppila, A., Tuimala, R., Haapalahti, J., Karppanen, H., and Viinikka, L.: Effects of intravenous isoxsuprine and ritodrine with and without concomitant dexamethasone, on fetoplacental and pituitary hormones and cyclic adenosine monophosphate during late pregnancy, AM. J. OBSTET. GYNECOL. 130:302, 1978.