- Email: [email protected]
The effect of indomethacin and prostacyclin agonists on blood pressure in a rat model of preeclampsia
The effect of indomethacin and prostacyclin agonists on blood pressure in a rat model of preeclampsia Marya G. Zlatnik, MD,a Irina Buhimschi, MD,a Kristof Chwalisz, MD, PhD,b Q. Ping Liao, MD,a George R. Saade, MD,a and Robert E. Garfield, PhDa Galveston, Texas, and Berlin, Germany OBJECTIVE: This study was designed to determine the effects of cyclooxygenase inhibition and prostacyclin agonists on the hypertension induced by nitric oxide synthase blockade in a previously characterized rat model of preeclampsia. STUDY DESIGN: A condition similar to preeclampsia was induced by infusing pregnant rats with the nitric oxide synthase inhibitor NG-nitro-L-arginine methyl ester through subcutaneously implanted osmotic minipumps. Blood pressure was measured with the tail cuff method. In the first experiment the rats received either vehicle alone (control group), NG-nitro-L-arginine methyl ester (50 mg/d), indomethacin (0.1 mg/d), or NG-nitro-L-arginine methyl ester plus indomethacin beginning on day 17 of pregnancy. In the second experiment the rats received vehicle alone (control group), NG-nitro-L-arginine methyl ester (50 mg/d), or NG-nitroL-arginine methyl ester plus iloprost (31 µg/d). In a third experiment cicaprost (15 µg/d) was substituted for iloprost. RESULTS: Except for an increase on the day after insertion of the pump indomethacin had no significant effect on the hypertension induced by NG-nitro-L-arginine methyl ester. Both prostacyclin agonists (iloprost and cicaprost), however, attenuated the rise in blood pressure usually seen after NG-nitro-L-arginine methyl ester administration. CONCLUSIONS: Nonselective inhibition of the cyclooxygenase enzymatic system does not influence the hypertension seen in the rat preeclampsia model induced by chronic nitric oxide deficiency. The hypertension in this model can be partially reversed with prostacyclin analogs. (Am J Obstet Gynecol 1999;180:1191-5.)
Key words: Cyclooxygenase, nitric oxide, preeclampsia, rat model, prostacyclin.
Preeclampsia, characterized by hypertension, edema, and proteinuria, is among the most common causes of maternal morbidity and mortality. The pathogenesis of preeclampsia remains unclear. It is believed that vascular changes are important not only in the symptoms but also in the causation of preeclampsia. Normal pregnancy is characterized by a decrease in vascular tone, a decrease in systemic vascular resistance, and a refractoriness to pressor agents. These maternal vascular adaptations are disturbed in preeclampsia. Several factors have been implicated in these abnormalities, including disturbances in prostaglandin function, endothelial dysfunction, and increased oxidant stress. The endothelial prostacyclin/ thromboxane production ratio is also altered in preeclampsia.1 In contrast to normal pregnancy the bal-
From the Department of Obstetrics and Gynecology, The University of Texas Medical Branch,a and the Research Laboratories of Schering AG.b Received for publication August 25, 1998; revised December 28, 1998; accepted January 27, 1999. Reprint requests: Marya G. Zlatnik, MD, Department of Obstetrics and Gynecology, The University of Texas Medical Branch, 301 University Blvd, Galveston, TX 77555-0587. Copyright © 1999 by Mosby, Inc. 0002-9378/99 $8.00 + 0 6/1/97401
ance is tilted in favor of thromboxane, a potent vasoconstrictor. In addition, inhibition of cyclooxygenase enzyme with aspirin has been shown to restore the refractoriness to angiotensin in some patients2 and to decrease the production of lipid peroxide and thromboxane A2 in the placentas of patients with preeclampsia.3 There is evidence to suggest that the normal physiologic vascular adaptations to pregnancy (increased blood volume, increased cardiac output, and decreased vascular resistance) are accompanied by an increase in endogenous nitric oxide4, 5 and an increase in the release of nitric oxide in response to known stimuli.6 Several studies have demonstrated that in pregnant rats treated with nitric oxide inhibitors, such as NG-nitro-L-arginine methyl ester (L-NAME) or NG-nitro-L-arginine, findings similar to those of preeclampsia develop; these include hypertension, fetal growth restriction, and proteinuria.4, 7, 8 Recent studies have demonstrated an interaction between the nitric oxide–generating pathway and the cyclooxygenase system. Both nitric oxide synthase and cyclooxygenase exist in inducible and constitutive isoforms. Nitric oxide has been shown to interact directly with the cyclooxygenase enzymes to modulate their activity.9 Moreover, low concentrations of nitric oxide increase the 1191
1192 Zlatnik et al
May 1999 Am J Obstet Gynecol
Fig 1. Effects of indomethacin on systolic blood pressure of pregnant rats in presence or absence of L-NAME. Data presented as mean ± SE. Circle, Control; square, L-NAME; triangle, L-NAME plus indomethacin; inverted triangle, indomethacin. Each day’s means were compared; those with different superscripts were significantly different from each other (P < .05).
expression of inducible cyclooxygenase in vitro, whereas high concentrations inhibit this expression.10 It has also been reported that nitric oxide modulates the release of prostacyclin from endothelial cells.11 The objective of our studies was to determine the effect of nonspecific inhibition of prostaglandin production and the effect of selective prostacyclin agonists on the hypertension seen in a rat model of preeclampsia. We hypothesized that inhibition of cyclooxygenase (by indomethacin) or administration of prostacyclin agonists would dampen the hypertension induced by chronic inhibition of nitric oxide synthase with L-NAME. Material and methods Animals. Adult nulliparous pregnant rats (300-325 g body weight) were purchased from Harlan Sprague Dawley, Inc (Indianapolis, Ind), and were received in the animal facilities on day 16 of pregnancy (with day 1 being the day of the positive sperm smear result). All animals were given free access to food and water. All procedures were approved by the Animal Care and Use Committee of the University of Texas Medical Branch. Induction of preeclampsia-like condition and treatment groups. All infusions were started on day 17 of pregnancy. In the first experiment the rats received vehicle alone (control group), L-NAME (50 mg/d), indomethacin (0.1 mg/d), or L-NAME plus indomethacin. In the second experiment the rats received vehicle alone (control group), L-NAME (50 mg/d), or L-NAME plus iloprost (31 µg/d). In the third experiment cicaprost (15 µg/d) was substituted for iloprost. The specific nitric oxide synthase inhibitor L-NAME (Sigma, St Louis, MO) was dissolved in sterile saline solution. Osmotic minipumps (model 2ML2 with a pumping rate of 5 mL/h; ALZA Corporation, Palo
Alto, Calif) were filled with vehicle with or without LNAME and placed subcutaneously during halothane anesthesia (Halocarbon Products Corporation, NJ). Indomethacin was injected subcutaneously twice daily, with a total dose of 0.1 mg/d. Iloprost and cicaprost were administered with osmotic minipumps. Each rat was used for only a single experiment. Blood pressure measurement. Starting on day 17 of pregnancy systolic blood pressure was measured daily or every other day with a pneumatic tail cuff device (NARCO Biosystems, Inc, Austin, Tex) in animals that had been warmed in a metal chamber maintained at approximately 30°C. Blood pressure values obtained from 3 consecutive measurements were averaged and recorded as the pressure for a given rat at each point. Measurements of systolic blood pressure of all animals were carried out until delivery and for 2 days post partum. All animals delivered on day 22 of gestation, the expected day of delivery. Statistical analysis. Mean ± SEM values were calculated for blood pressure levels on each day of pregnancy and post partum. Statistical analysis was performed with 1-way analysis of variance followed by multiple comparison with the Fisher least significant difference criteria. P < .05 was considered significant. Results Except for an increase on the day after insertion of the pump, indomethacin had no significant effect on the hypertension induced by L-NAME (Fig 1). Iloprost (31 µg/d) partially attenuated the rise in blood pressure expected after L-NAME administration, although not at every time point measured (Fig 2). Cicaprost (15 µg/d) significantly attenuated the rise in blood pressure ex-
Volume 180, Number 5 Am J Obstet Gynecol
Zlatnik et al 1193
Fig 2. Effects of iloprost on systolic blood pressure of pregnant rats in presence or absence of L-NAME. Data presented as mean ± SE. Open bar, Control; solid bar, L-NAME; hatched bar, L-NAME plus iloprost. Each day’s means were compared; those with different superscripts were significantly different from each other (P < .05).
Fig 3. Effects of cicaprost on systolic blood pressure of pregnant rats in presence or absence of L-NAME. Data presented as mean ± SE. Open bar, Control; solid bar, L-NAME; hatched bar, L-NAME plus cicaprost. Each day’s means were compared; those with different superscripts were significantly different from each other (P < .05).
pected after L-NAME administration, reducing the blood pressure to control levels or below on days 17 through 22 and to near control levels on postpartum days 1 through 4 (Fig 3). The effect of cicaprost on blood pressure was greater than that of iloprost on all days measured.
oxide synthase inhibition in the pregnant rat mimics the clinical findings seen in women with preeclampsia, including hypertension, fetal growth restriction, proteinuria, intravascular volume constriction, renal abnormalities, and thrombocytopenia.12, 13 Findings in this animal model are also consistent with other disturbances reported in human preeclampsia, such as an increase in serum levels of endothelin 1 and increased responsiveness to angiotensin.14, 15 One difference noted between the rat model and clinical preeclampsia is that thus far the rats have not been noted to progress to an eclamptic state. Although our findings suggest that prostacyclin analogs may be helpful in the management of preeclampsia, we only evaluated blood pressure and did not determine whether the other manifestations of preeclampsia (thrombocytopenia, proteinuria, and tissue ischemia) were reversed. It is therefore possible that the hypertension is related to a deficiency in prostacyclin but that the other manifestations of preeclampsia are mediated by other factors. Another possibility is that the cyclooxygenase enzymatic system is already down-regulated as a result of the hypertension and that the hypertension is not caused by prostacyclin deficiency but can be reversed by a prostacyclin analog acting as a nonspecific vasodilator. It is therefore possible that prostacyclin is not directly involved in the hypertension seen in this model. Although our data support the concept that cyclooxygenase products (namely prostacyclin) are involved in the pathogenesis of preeclampsia, the lack of effect from
Comment Prostacyclin analogs dampened the rise in blood pressure in this animal model of preeclampsia. This supports the hypothesis that some of the cyclooxygenase products (namely prostacyclin) may be involved in the causation of hypertension induced in the pregnant rat after nitric oxide synthase blockade. We did not find, however, any effect on blood pressure from nonselective inhibition of the cyclooxygenase system in the rat preeclampsia model induced by nitric oxide deficiency. Our findings do not exclude a role for a relative deficiency or excess of individual cyclooxygenase products. Because indomethacin may have inhibited production of both prostacyclin and thromboxane A2, our results do not refute the hypothesis that the hypertension seen in this animal model of preeclampsia is caused by a relative deficiency of prostacyclin with respect to thromboxane. Our findings support the hypothesis suggested by others that an imbalance between thromboxane and prostacyclin production or function plays an important role in the development of preeclampsia. One limitation of this study was the use of an animal model. Research in preeclampsia has been hindered by the lack of a naturally occurring equivalent condition in nonhuman animals. The condition induced by nitric
1194 Zlatnik et al
nonspecific inhibition of cyclooxygenase products with indomethacin is somewhat puzzling. This lack of effect is probably explained by the nonspecific nature of indomethacin—it is expected to inhibit both the thromboxane and prostacyclin products of the cyclooxygenase enzymatic system. It may be that in our experiment the balance of thromboxane A2/prostacyclin ratio was not altered and that is why no effect was seen. In addition, the dose used may not have been appropriate for selective inhibition of thromboxane rather than prostacyclin. In fact, the blood pressure was actually higher in the LNAME plus indomethacin group than in the L-NAME only group on the first day of treatment, perhaps indicating that indomethacin initially inhibited prostacyclin production more than it did thromboxane production. As in the case of aspirin, the relative effects of indomethacin on thromboxane and prostacyclin production may be dose dependent. A number of cyclooxygenase inhibitors have been studied clinically for the prevention of preeclampsia. Aspirin is a nonspecific inhibitor of the cyclooxygenase enzymatic system. It has been studied extensively as a possible medication to prevent preeclampsia. Aspirin has been shown to influence the imbalance of prostacyclin and thromboxane; however, results of clinical trials have been mixed. Several small studies16 and a meta-analysis17 have shown a beneficial effect, especially in certain highrisk subpopulations. Most recently, 2 large randomized placebo-controlled trials (ECPPA18 and CLASP19 trials) showed no benefit from aspirin in the prevention of preeclampsia. It has been argued that an incorrect dosage of aspirin was used in these trials and that the balance of cyclooxygenase products was therefore not altered.20 Other investigators have studied the effects of various prostacyclin analogs and thromboxane synthase inhibitors in various animal models and on a smaller scale in pregnant women with (or at risk for) preeclampsia. Results of these studies have generally been favorable. Furuhashi et al21 studied the effects of OKY-046, a thromboxane synthase inhibitor, in normotensive and spontaneously hypertensive pregnant rats. They found that OKY-046 decreased the blood pressure in the hypertensive rats but not in the healthy rats. Placental blood flow remained unchanged in both groups of rats. Keith et al22 used a thromboxane synthase inhibitor (U-63,557A) in an ovine pregnancy toxemia syndrome model. In these sheep with starvation- or stress-induced toxemia thromboxane synthase inhibitors reversed the hypertension and the proteinuria. In addition, platelet dysfunction and fetal heart rate decelerations resolved with treatment. Another thromboxane synthase inhibitor, dazoxiben, was also studied in a small group of pregnant women with severe preeclampsia.23 Although not all women had responses to the medication, those who did
May 1999 Am J Obstet Gynecol
so had significant pregnancy prolongation and improvement in blood pressure, proteinuria, and other symptoms. Of the 11 patients studied, 7 were considered to have had clinically significant improvement. Keith et al24 also evaluated ridogrel, a thromboxane synthase inhibitor and endoperoxide receptor agonist, in pregnant rhesus monkeys. They measured serum thromboxane and prostacyclin levels after intravenous ridogrel administration to evaluate its effects on the prostacyclin/thromboxane ratio. The compound was well tolerated and it increased prostacyclin and decreased thromboxane levels at both doses evaluated, thereby making the prostacyclin/thromboxane ratio more favorable. Seki et al25 studied the use of ozagrel, a thromboxane A2 synthase inhibitor, in 9 women with severe preeclampsia or a history of severe preeclampsia. They reported that in their small uncontrolled trial ozagrel decreased the blood pressure and perhaps prevented development of preeclampsia, with no effect on fetal growth or the incidence of fetal distress. In a randomized study Di Iorio et al26 evaluated the use of picotamide (a selective inhibitor of thromboxane A2 synthase and thromboxane A2 platelet receptors) in late term pregnancy in women with diagnosed pregnancy-induced hypertension. They found significantly less platelet aggregation in response to collagen and a trend toward increased gestational age, birth weight, and Apgar scores in the treated group with respect to control subjects. Moodley and Gouws27 studied the use of intravenous epoprostenol (prostacyclin) in women with severe hypertension in pregnancy. Epoprostenol was found to be as efficacious as hydralazine in lowering blood pressure in the short term in their randomized clinical trial. We found that prostacyclin analogs dampen the rise in blood pressure seen in a rat model of preeclampsia. In our system the effect on blood pressure of cicaprost was greater than that of iloprost. Nonspecific inhibition of the cyclooxygenase system did not influence the hypertension in the rat model. These data support the hypotheses that a relative prostacyclin deficiency is involved in the causation of the hypertension induced in the rat model by nitric oxide synthase blockade and that the imbalance between the cyclooxygenase products prostacyclin and thromboxane may play an important role in the pathogenesis of preeclampsia. Furthermore, this study suggests that possible preventive or therapeutic regimens for preeclampsia might involve alteration of the balance of cyclooxygenase products. Further clinical studies of prostacyclin agonists and thromboxane antagonists are needed. REFERENCES
1. Walsh SW. Preeclampsia: an imbalance in placental prostacyclin and thromboxane production. Am J Obstet Gynecol 1985;152:335-40. 2. Brown CE, Gant NF, Cox K, Spitz B, Rosenfeld CR, Magness RR. Low-dose aspirin. II. Relationship of angiotensin II pressor re-
Zlatnik et al 1195
Volume 180, Number 5 Am J Obstet Gynecol
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
sponses, circulating eicosanoids, and pregnancy outcome. Am J Obstet Gynecol 1990;163:1853-61. Wang Y, Walsh SW. Preeclampsia: Aspirin inhibits both lipid peroxides and thromboxane in preeclamptic placentas. Free Radical Biol Med 1995;18:585-91. Molnar M, Hertelendy F. N-omega-nitro-L-arginine, an inhibitor of nitric oxide synthesis, increases blood pressure in rats and reverses the pregnancy-induced refractoriness to vasopressor agents. Am J Obstet Gynecol 1992;166:1560-7. Okutomi T, Nomoto K, Nakamura K, Goto F. Nitric oxide metabolite in pregnant women before and after delivery. Acta Obstet Gynecol Scand 1997;76:222-6. Weiner CP, Martinez E, Zhu LK, Ghodsi A, Chestnut D. In vitro release of endothelium-derived relaxing factor by acetylcholine is increased during the guinea pig pregnancy. Am J Obstet Gynecol 1989;161:1599-605. Yallampalli C, Garfield RE. Inhibition of nitric oxide synthesis in rats during pregnancy produces signs similar to those of preeclampsia. Am J Obstet Gynecol 1993;169:1316-20. Baylis C, Mitruka B, Deng A. Chronic blockade of nitric oxide synthesis in the rat produces systemic hypertension and glomerular damage. J Clin Invest 1992;90:278-81. Salvemini D, Misko TP, Masferrer JL, Seibert K, Currie MG, Needleman P. Nitric oxide activates cyclooxygenase enzymes. Proc Natl Acad Sci U S A 1993;90:7240-4. Swierkosz TA, Mitchell JA, Tomlinson A, Botting RM, Warner TD, Vane JR. Relationship between different isoforms of nitric oxide synthase and cyclooxygenase in various cell types. Polish J Pharm 1994;46:587-92. Soma M, Izumi Y, Watanabe Y, Kanmatsuse K. A nitric oxide synthesis inhibitor decreased prostaglandin production in rat mesenteric vasculature. Prostaglandins 1996;51:225-32. Molnar M, Suto T, Toth T, Hertelendy F. Prolonged blockade of nitric oxide synthesis in gravid rats produces sustained hypertension, proteinuria, thrombocytopenia, and intrauterine growth retardation. Am J Obstet Gynecol 1994;170:1458-66. Novak J, Reckelhoff J, Bumgamer L, Cockrell K, Kassab S, Granger JP. Reduced sensitivity of the renal circulation to angiotensin II in pregnant rats. Hypertension 1997;30:580-4. Edwards DL, Arora CP, Bui DT, Castro LC. Long-term nitric oxide blockade in the pregnant rat: effects on blood pressure and plasma levels of endothelin-1. Am J Obstet Gynecol 1996;175:484-8. Ahokas RA, Sibai BM. Endothelium-derived relaxing factor inhibition augments the vascular angiotensin II reactivity in the pregnant rat hind limb. Am J Obstet Gynecol 1992;167:1053-8. Surean C. Prevention of perinatal consequences of pre-eclamp-
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
sia with low-dose aspirin: results of the EPREDA trial. Eur J Obstet Gynecol Reprod Biol 1991;41:71-3. Imperiale TF, Petrulis AS. A meta-analysis of low-dose aspirin for the prevention of pregnancy-induced hypertensive disease. JAMA 1991;266:260-4. Anonymous. ECPPA: randomised trial of low dose aspirin for the prevention of maternal and fetal complications in high risk pregnant women. ECPPA (Estudo Colaborativo para Prevenção da Pré-eclampsia com Aspirina) Collaborative Group. Br J Obstet Gynaecol 1996;103:39-47. Anonymous. CLASP: a randomised trial of low-dose aspirin for the prevention and treatment of pre-eclampsia among 9364 pregnant women. CLASP (Collaborative Low-dose Aspirin Study in Pregnancy). Collaborative Group. Lancet 1994;343:619-29. Goffinet F, Bréart G, Uzan S. ECPPA: randomised trial of low dose aspirin for the prevention of maternal and fetal complications in high risk pregnant women [letter; comment]. Br J Obstet Gynaecol 1996;103:719-20. Furuhashi N, Tsujiei M, Kimura H, Nagae H, Yajima A. Effects of thromboxane synthetase inhibitor (OKY-046) on placental blood flow, placental weight and fetal weight in normotensive and spontaneously hypertensive rats. Gynecol Obstet Invest 1993;35:23-6. Keith JC Jr, Spitz B, Van Assche FA. Thromboxane synthetase inhibition as a new therapy for preeclampsia: animal and human studies minireview. Prostaglandins 1993;45:3-13. Van Assche FA, Spitz B. Thromboxane synthetase inhibition in pregnancy-induced hypertension. Am J Obstet Gynecol 1988;1159:1015-6. Keith JC Jr, Goodrich JA, Endo Y, Taub D, Mehlman P, Ward G. Effects of ridogrel on the prostacyclin-thromboxane ratio in nulliparous third trimester-pregnant rhesus monkeys. Prostaglandins 1993;45:547-55. Seki H, Kuromaki K, Takeda S, Kinoshita K, Satoh K. The possibility of clinical application of the thromboxane A2 synthase inhibitor, ozagrel, for the treatment and prevention of preeclampsia: a preliminary report. J Obstet Gynaecol 1995;4:357-65. Di Iorio R, Marinoni E, Horvath S, Pulcinelli FM, Manzari G, Martinico E, et al. Selective inhibition of thromboxane A2 synthetase in the treatment of pregnancy-induced hypertension. In: Cosmi EV, DiRenzo CG, editors. Hypertension in pregnancy. Bologna, Italy: Monduzzi Editore; 1991. p. 289-93. Moodley J, Gouws E. A comparative study of the use of epoprostenol and dihydralazine in severe hypertension in pregnancy. Br J Obstet Gynaecol 1992;99:727-30.
Key words: Cyclooxygenase, nitric oxide, preeclampsia, rat model, prostacyclin.
Preeclampsia, characterized by hypertension, edema, and proteinuria, is among the most common causes of maternal morbidity and mortality. The pathogenesis of preeclampsia remains unclear. It is believed that vascular changes are important not only in the symptoms but also in the causation of preeclampsia. Normal pregnancy is characterized by a decrease in vascular tone, a decrease in systemic vascular resistance, and a refractoriness to pressor agents. These maternal vascular adaptations are disturbed in preeclampsia. Several factors have been implicated in these abnormalities, including disturbances in prostaglandin function, endothelial dysfunction, and increased oxidant stress. The endothelial prostacyclin/ thromboxane production ratio is also altered in preeclampsia.1 In contrast to normal pregnancy the bal-
From the Department of Obstetrics and Gynecology, The University of Texas Medical Branch,a and the Research Laboratories of Schering AG.b Received for publication August 25, 1998; revised December 28, 1998; accepted January 27, 1999. Reprint requests: Marya G. Zlatnik, MD, Department of Obstetrics and Gynecology, The University of Texas Medical Branch, 301 University Blvd, Galveston, TX 77555-0587. Copyright © 1999 by Mosby, Inc. 0002-9378/99 $8.00 + 0 6/1/97401
ance is tilted in favor of thromboxane, a potent vasoconstrictor. In addition, inhibition of cyclooxygenase enzyme with aspirin has been shown to restore the refractoriness to angiotensin in some patients2 and to decrease the production of lipid peroxide and thromboxane A2 in the placentas of patients with preeclampsia.3 There is evidence to suggest that the normal physiologic vascular adaptations to pregnancy (increased blood volume, increased cardiac output, and decreased vascular resistance) are accompanied by an increase in endogenous nitric oxide4, 5 and an increase in the release of nitric oxide in response to known stimuli.6 Several studies have demonstrated that in pregnant rats treated with nitric oxide inhibitors, such as NG-nitro-L-arginine methyl ester (L-NAME) or NG-nitro-L-arginine, findings similar to those of preeclampsia develop; these include hypertension, fetal growth restriction, and proteinuria.4, 7, 8 Recent studies have demonstrated an interaction between the nitric oxide–generating pathway and the cyclooxygenase system. Both nitric oxide synthase and cyclooxygenase exist in inducible and constitutive isoforms. Nitric oxide has been shown to interact directly with the cyclooxygenase enzymes to modulate their activity.9 Moreover, low concentrations of nitric oxide increase the 1191
1192 Zlatnik et al
May 1999 Am J Obstet Gynecol
Fig 1. Effects of indomethacin on systolic blood pressure of pregnant rats in presence or absence of L-NAME. Data presented as mean ± SE. Circle, Control; square, L-NAME; triangle, L-NAME plus indomethacin; inverted triangle, indomethacin. Each day’s means were compared; those with different superscripts were significantly different from each other (P < .05).
expression of inducible cyclooxygenase in vitro, whereas high concentrations inhibit this expression.10 It has also been reported that nitric oxide modulates the release of prostacyclin from endothelial cells.11 The objective of our studies was to determine the effect of nonspecific inhibition of prostaglandin production and the effect of selective prostacyclin agonists on the hypertension seen in a rat model of preeclampsia. We hypothesized that inhibition of cyclooxygenase (by indomethacin) or administration of prostacyclin agonists would dampen the hypertension induced by chronic inhibition of nitric oxide synthase with L-NAME. Material and methods Animals. Adult nulliparous pregnant rats (300-325 g body weight) were purchased from Harlan Sprague Dawley, Inc (Indianapolis, Ind), and were received in the animal facilities on day 16 of pregnancy (with day 1 being the day of the positive sperm smear result). All animals were given free access to food and water. All procedures were approved by the Animal Care and Use Committee of the University of Texas Medical Branch. Induction of preeclampsia-like condition and treatment groups. All infusions were started on day 17 of pregnancy. In the first experiment the rats received vehicle alone (control group), L-NAME (50 mg/d), indomethacin (0.1 mg/d), or L-NAME plus indomethacin. In the second experiment the rats received vehicle alone (control group), L-NAME (50 mg/d), or L-NAME plus iloprost (31 µg/d). In the third experiment cicaprost (15 µg/d) was substituted for iloprost. The specific nitric oxide synthase inhibitor L-NAME (Sigma, St Louis, MO) was dissolved in sterile saline solution. Osmotic minipumps (model 2ML2 with a pumping rate of 5 mL/h; ALZA Corporation, Palo
Alto, Calif) were filled with vehicle with or without LNAME and placed subcutaneously during halothane anesthesia (Halocarbon Products Corporation, NJ). Indomethacin was injected subcutaneously twice daily, with a total dose of 0.1 mg/d. Iloprost and cicaprost were administered with osmotic minipumps. Each rat was used for only a single experiment. Blood pressure measurement. Starting on day 17 of pregnancy systolic blood pressure was measured daily or every other day with a pneumatic tail cuff device (NARCO Biosystems, Inc, Austin, Tex) in animals that had been warmed in a metal chamber maintained at approximately 30°C. Blood pressure values obtained from 3 consecutive measurements were averaged and recorded as the pressure for a given rat at each point. Measurements of systolic blood pressure of all animals were carried out until delivery and for 2 days post partum. All animals delivered on day 22 of gestation, the expected day of delivery. Statistical analysis. Mean ± SEM values were calculated for blood pressure levels on each day of pregnancy and post partum. Statistical analysis was performed with 1-way analysis of variance followed by multiple comparison with the Fisher least significant difference criteria. P < .05 was considered significant. Results Except for an increase on the day after insertion of the pump, indomethacin had no significant effect on the hypertension induced by L-NAME (Fig 1). Iloprost (31 µg/d) partially attenuated the rise in blood pressure expected after L-NAME administration, although not at every time point measured (Fig 2). Cicaprost (15 µg/d) significantly attenuated the rise in blood pressure ex-
Volume 180, Number 5 Am J Obstet Gynecol
Zlatnik et al 1193
Fig 2. Effects of iloprost on systolic blood pressure of pregnant rats in presence or absence of L-NAME. Data presented as mean ± SE. Open bar, Control; solid bar, L-NAME; hatched bar, L-NAME plus iloprost. Each day’s means were compared; those with different superscripts were significantly different from each other (P < .05).
Fig 3. Effects of cicaprost on systolic blood pressure of pregnant rats in presence or absence of L-NAME. Data presented as mean ± SE. Open bar, Control; solid bar, L-NAME; hatched bar, L-NAME plus cicaprost. Each day’s means were compared; those with different superscripts were significantly different from each other (P < .05).
pected after L-NAME administration, reducing the blood pressure to control levels or below on days 17 through 22 and to near control levels on postpartum days 1 through 4 (Fig 3). The effect of cicaprost on blood pressure was greater than that of iloprost on all days measured.
oxide synthase inhibition in the pregnant rat mimics the clinical findings seen in women with preeclampsia, including hypertension, fetal growth restriction, proteinuria, intravascular volume constriction, renal abnormalities, and thrombocytopenia.12, 13 Findings in this animal model are also consistent with other disturbances reported in human preeclampsia, such as an increase in serum levels of endothelin 1 and increased responsiveness to angiotensin.14, 15 One difference noted between the rat model and clinical preeclampsia is that thus far the rats have not been noted to progress to an eclamptic state. Although our findings suggest that prostacyclin analogs may be helpful in the management of preeclampsia, we only evaluated blood pressure and did not determine whether the other manifestations of preeclampsia (thrombocytopenia, proteinuria, and tissue ischemia) were reversed. It is therefore possible that the hypertension is related to a deficiency in prostacyclin but that the other manifestations of preeclampsia are mediated by other factors. Another possibility is that the cyclooxygenase enzymatic system is already down-regulated as a result of the hypertension and that the hypertension is not caused by prostacyclin deficiency but can be reversed by a prostacyclin analog acting as a nonspecific vasodilator. It is therefore possible that prostacyclin is not directly involved in the hypertension seen in this model. Although our data support the concept that cyclooxygenase products (namely prostacyclin) are involved in the pathogenesis of preeclampsia, the lack of effect from
Comment Prostacyclin analogs dampened the rise in blood pressure in this animal model of preeclampsia. This supports the hypothesis that some of the cyclooxygenase products (namely prostacyclin) may be involved in the causation of hypertension induced in the pregnant rat after nitric oxide synthase blockade. We did not find, however, any effect on blood pressure from nonselective inhibition of the cyclooxygenase system in the rat preeclampsia model induced by nitric oxide deficiency. Our findings do not exclude a role for a relative deficiency or excess of individual cyclooxygenase products. Because indomethacin may have inhibited production of both prostacyclin and thromboxane A2, our results do not refute the hypothesis that the hypertension seen in this animal model of preeclampsia is caused by a relative deficiency of prostacyclin with respect to thromboxane. Our findings support the hypothesis suggested by others that an imbalance between thromboxane and prostacyclin production or function plays an important role in the development of preeclampsia. One limitation of this study was the use of an animal model. Research in preeclampsia has been hindered by the lack of a naturally occurring equivalent condition in nonhuman animals. The condition induced by nitric
1194 Zlatnik et al
nonspecific inhibition of cyclooxygenase products with indomethacin is somewhat puzzling. This lack of effect is probably explained by the nonspecific nature of indomethacin—it is expected to inhibit both the thromboxane and prostacyclin products of the cyclooxygenase enzymatic system. It may be that in our experiment the balance of thromboxane A2/prostacyclin ratio was not altered and that is why no effect was seen. In addition, the dose used may not have been appropriate for selective inhibition of thromboxane rather than prostacyclin. In fact, the blood pressure was actually higher in the LNAME plus indomethacin group than in the L-NAME only group on the first day of treatment, perhaps indicating that indomethacin initially inhibited prostacyclin production more than it did thromboxane production. As in the case of aspirin, the relative effects of indomethacin on thromboxane and prostacyclin production may be dose dependent. A number of cyclooxygenase inhibitors have been studied clinically for the prevention of preeclampsia. Aspirin is a nonspecific inhibitor of the cyclooxygenase enzymatic system. It has been studied extensively as a possible medication to prevent preeclampsia. Aspirin has been shown to influence the imbalance of prostacyclin and thromboxane; however, results of clinical trials have been mixed. Several small studies16 and a meta-analysis17 have shown a beneficial effect, especially in certain highrisk subpopulations. Most recently, 2 large randomized placebo-controlled trials (ECPPA18 and CLASP19 trials) showed no benefit from aspirin in the prevention of preeclampsia. It has been argued that an incorrect dosage of aspirin was used in these trials and that the balance of cyclooxygenase products was therefore not altered.20 Other investigators have studied the effects of various prostacyclin analogs and thromboxane synthase inhibitors in various animal models and on a smaller scale in pregnant women with (or at risk for) preeclampsia. Results of these studies have generally been favorable. Furuhashi et al21 studied the effects of OKY-046, a thromboxane synthase inhibitor, in normotensive and spontaneously hypertensive pregnant rats. They found that OKY-046 decreased the blood pressure in the hypertensive rats but not in the healthy rats. Placental blood flow remained unchanged in both groups of rats. Keith et al22 used a thromboxane synthase inhibitor (U-63,557A) in an ovine pregnancy toxemia syndrome model. In these sheep with starvation- or stress-induced toxemia thromboxane synthase inhibitors reversed the hypertension and the proteinuria. In addition, platelet dysfunction and fetal heart rate decelerations resolved with treatment. Another thromboxane synthase inhibitor, dazoxiben, was also studied in a small group of pregnant women with severe preeclampsia.23 Although not all women had responses to the medication, those who did
May 1999 Am J Obstet Gynecol
so had significant pregnancy prolongation and improvement in blood pressure, proteinuria, and other symptoms. Of the 11 patients studied, 7 were considered to have had clinically significant improvement. Keith et al24 also evaluated ridogrel, a thromboxane synthase inhibitor and endoperoxide receptor agonist, in pregnant rhesus monkeys. They measured serum thromboxane and prostacyclin levels after intravenous ridogrel administration to evaluate its effects on the prostacyclin/thromboxane ratio. The compound was well tolerated and it increased prostacyclin and decreased thromboxane levels at both doses evaluated, thereby making the prostacyclin/thromboxane ratio more favorable. Seki et al25 studied the use of ozagrel, a thromboxane A2 synthase inhibitor, in 9 women with severe preeclampsia or a history of severe preeclampsia. They reported that in their small uncontrolled trial ozagrel decreased the blood pressure and perhaps prevented development of preeclampsia, with no effect on fetal growth or the incidence of fetal distress. In a randomized study Di Iorio et al26 evaluated the use of picotamide (a selective inhibitor of thromboxane A2 synthase and thromboxane A2 platelet receptors) in late term pregnancy in women with diagnosed pregnancy-induced hypertension. They found significantly less platelet aggregation in response to collagen and a trend toward increased gestational age, birth weight, and Apgar scores in the treated group with respect to control subjects. Moodley and Gouws27 studied the use of intravenous epoprostenol (prostacyclin) in women with severe hypertension in pregnancy. Epoprostenol was found to be as efficacious as hydralazine in lowering blood pressure in the short term in their randomized clinical trial. We found that prostacyclin analogs dampen the rise in blood pressure seen in a rat model of preeclampsia. In our system the effect on blood pressure of cicaprost was greater than that of iloprost. Nonspecific inhibition of the cyclooxygenase system did not influence the hypertension in the rat model. These data support the hypotheses that a relative prostacyclin deficiency is involved in the causation of the hypertension induced in the rat model by nitric oxide synthase blockade and that the imbalance between the cyclooxygenase products prostacyclin and thromboxane may play an important role in the pathogenesis of preeclampsia. Furthermore, this study suggests that possible preventive or therapeutic regimens for preeclampsia might involve alteration of the balance of cyclooxygenase products. Further clinical studies of prostacyclin agonists and thromboxane antagonists are needed. REFERENCES
1. Walsh SW. Preeclampsia: an imbalance in placental prostacyclin and thromboxane production. Am J Obstet Gynecol 1985;152:335-40. 2. Brown CE, Gant NF, Cox K, Spitz B, Rosenfeld CR, Magness RR. Low-dose aspirin. II. Relationship of angiotensin II pressor re-
Zlatnik et al 1195
Volume 180, Number 5 Am J Obstet Gynecol
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
sponses, circulating eicosanoids, and pregnancy outcome. Am J Obstet Gynecol 1990;163:1853-61. Wang Y, Walsh SW. Preeclampsia: Aspirin inhibits both lipid peroxides and thromboxane in preeclamptic placentas. Free Radical Biol Med 1995;18:585-91. Molnar M, Hertelendy F. N-omega-nitro-L-arginine, an inhibitor of nitric oxide synthesis, increases blood pressure in rats and reverses the pregnancy-induced refractoriness to vasopressor agents. Am J Obstet Gynecol 1992;166:1560-7. Okutomi T, Nomoto K, Nakamura K, Goto F. Nitric oxide metabolite in pregnant women before and after delivery. Acta Obstet Gynecol Scand 1997;76:222-6. Weiner CP, Martinez E, Zhu LK, Ghodsi A, Chestnut D. In vitro release of endothelium-derived relaxing factor by acetylcholine is increased during the guinea pig pregnancy. Am J Obstet Gynecol 1989;161:1599-605. Yallampalli C, Garfield RE. Inhibition of nitric oxide synthesis in rats during pregnancy produces signs similar to those of preeclampsia. Am J Obstet Gynecol 1993;169:1316-20. Baylis C, Mitruka B, Deng A. Chronic blockade of nitric oxide synthesis in the rat produces systemic hypertension and glomerular damage. J Clin Invest 1992;90:278-81. Salvemini D, Misko TP, Masferrer JL, Seibert K, Currie MG, Needleman P. Nitric oxide activates cyclooxygenase enzymes. Proc Natl Acad Sci U S A 1993;90:7240-4. Swierkosz TA, Mitchell JA, Tomlinson A, Botting RM, Warner TD, Vane JR. Relationship between different isoforms of nitric oxide synthase and cyclooxygenase in various cell types. Polish J Pharm 1994;46:587-92. Soma M, Izumi Y, Watanabe Y, Kanmatsuse K. A nitric oxide synthesis inhibitor decreased prostaglandin production in rat mesenteric vasculature. Prostaglandins 1996;51:225-32. Molnar M, Suto T, Toth T, Hertelendy F. Prolonged blockade of nitric oxide synthesis in gravid rats produces sustained hypertension, proteinuria, thrombocytopenia, and intrauterine growth retardation. Am J Obstet Gynecol 1994;170:1458-66. Novak J, Reckelhoff J, Bumgamer L, Cockrell K, Kassab S, Granger JP. Reduced sensitivity of the renal circulation to angiotensin II in pregnant rats. Hypertension 1997;30:580-4. Edwards DL, Arora CP, Bui DT, Castro LC. Long-term nitric oxide blockade in the pregnant rat: effects on blood pressure and plasma levels of endothelin-1. Am J Obstet Gynecol 1996;175:484-8. Ahokas RA, Sibai BM. Endothelium-derived relaxing factor inhibition augments the vascular angiotensin II reactivity in the pregnant rat hind limb. Am J Obstet Gynecol 1992;167:1053-8. Surean C. Prevention of perinatal consequences of pre-eclamp-
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
sia with low-dose aspirin: results of the EPREDA trial. Eur J Obstet Gynecol Reprod Biol 1991;41:71-3. Imperiale TF, Petrulis AS. A meta-analysis of low-dose aspirin for the prevention of pregnancy-induced hypertensive disease. JAMA 1991;266:260-4. Anonymous. ECPPA: randomised trial of low dose aspirin for the prevention of maternal and fetal complications in high risk pregnant women. ECPPA (Estudo Colaborativo para Prevenção da Pré-eclampsia com Aspirina) Collaborative Group. Br J Obstet Gynaecol 1996;103:39-47. Anonymous. CLASP: a randomised trial of low-dose aspirin for the prevention and treatment of pre-eclampsia among 9364 pregnant women. CLASP (Collaborative Low-dose Aspirin Study in Pregnancy). Collaborative Group. Lancet 1994;343:619-29. Goffinet F, Bréart G, Uzan S. ECPPA: randomised trial of low dose aspirin for the prevention of maternal and fetal complications in high risk pregnant women [letter; comment]. Br J Obstet Gynaecol 1996;103:719-20. Furuhashi N, Tsujiei M, Kimura H, Nagae H, Yajima A. Effects of thromboxane synthetase inhibitor (OKY-046) on placental blood flow, placental weight and fetal weight in normotensive and spontaneously hypertensive rats. Gynecol Obstet Invest 1993;35:23-6. Keith JC Jr, Spitz B, Van Assche FA. Thromboxane synthetase inhibition as a new therapy for preeclampsia: animal and human studies minireview. Prostaglandins 1993;45:3-13. Van Assche FA, Spitz B. Thromboxane synthetase inhibition in pregnancy-induced hypertension. Am J Obstet Gynecol 1988;1159:1015-6. Keith JC Jr, Goodrich JA, Endo Y, Taub D, Mehlman P, Ward G. Effects of ridogrel on the prostacyclin-thromboxane ratio in nulliparous third trimester-pregnant rhesus monkeys. Prostaglandins 1993;45:547-55. Seki H, Kuromaki K, Takeda S, Kinoshita K, Satoh K. The possibility of clinical application of the thromboxane A2 synthase inhibitor, ozagrel, for the treatment and prevention of preeclampsia: a preliminary report. J Obstet Gynaecol 1995;4:357-65. Di Iorio R, Marinoni E, Horvath S, Pulcinelli FM, Manzari G, Martinico E, et al. Selective inhibition of thromboxane A2 synthetase in the treatment of pregnancy-induced hypertension. In: Cosmi EV, DiRenzo CG, editors. Hypertension in pregnancy. Bologna, Italy: Monduzzi Editore; 1991. p. 289-93. Moodley J, Gouws E. A comparative study of the use of epoprostenol and dihydralazine in severe hypertension in pregnancy. Br J Obstet Gynaecol 1992;99:727-30.