Serum arachidonic acid levels in normal and preeclamptic pregnancies

Cesarean section in twin gestations

Volume 148 Number I

4. Pritchard, J. A., and MacDonald, P. A.: Williams' Obstetrics, ed. I6, New York, I980, Appleton-Century-Crofts, p. 662. 5. Cetruio, C. L., Ingardia, C. J., and Sbarra, A. J.: Management of multiple gestation, Clin. Obstet. Gynecol. 23: 533, I980. 6. Keith, L., and Hughey, M. J.: Twin gestation, in Gerbie, A. B., and Sciarra, J. J ., editors: Gynecology and Obstetrics, ed. 2, New York, I98I, vol. 2, Harper & Row, Publishers, Inc., p. 8. 7. Danforth, D. N.: Obstetrics and Gynecology, ed. 4, Philadelphia, 1982, Harper & Row, Publishers, Inc., pp. 733-735. 8. Ho, K. S., and Wu, P. Y. K.: Perinatal factors and neonatal morbidity in twin pregnancy, AM. J. OBSTET. GYNECOL. 122:979, I975. 9. Ware, H. H.: The second twin, AM. J. 0BSTET. GYNECOL. 110:865, 1971. 10. Chervenak, F. A., Johnson, R. E., Berkowitz, R. L., et al.: Intrapartum external version of the second twin, Obstet. Gynecol. 62:I60, I983.

II. Goldenberg, R. L., and Nelson, K. G.: The premature breech, AM. J. 0BSTET. GYNECOL. 127:240, I977. I2. Duenhoelter, J. H., Wells, C. E., and Reisch, J. S.: A paired controlled study of vaginal and abdominal delivery of the low birth weight breech fetus, Obstet. Gynecol. 54:3IO, I979. I3. Drage, J. S., Kennedy, C., and Berendes, H.: The Apgar score as an index of infant morbidity, Dev. Med. Child Neurol. 8:141, 1966. I4. Acker, D., Lieberman, M., Holbrook, H., et al.: Delivery of the second twin, Obstet. Gynecol. 59:710, I982. 15. Collea, J. V., Rabin, S. C., Weghorst, G. R., et al.: The randomized management of term frank breech presentation: Vaginal delivery vs. cesarean section, AM. J. 0BSTET. GYNECOL. 131:186, 1978. I6. Shepard, M.J., Richard, V. A., Berkowitz, R. L., eta!.: An evaluation of two equations for predicting fetal weight by ultrasound, AM.J. 0BSTET. GYNECOL. 142:47, I982.

Serum arachidonic acid levels in normal and preeclamptic pregnancies PaulL. Ogburn, Jr., M.D., Preston P. Williams, M.D., Susan B. Johnson, B.S., and Ralph T. Holman, Ph.D. Minneapolis and Austin, Minnesota Fifteen serum samples from 11 women with preeclampsia and 19 samples from 10 normal third-trimester pregnancies were analyzed for total nonesterified fatty acids and total nonesterified arachidonic acid. The percentages of arachidonic acid in nonesterified fatty acids, in phospholipids, in triglycerides, and in cholesterol esters were also measured in each sample. The same analyses were done on serum from six samples of cord blood from each group. Cord blood sera from preeclamptic and normal pregnancies had much less total nonesterified fatty acids than the corresponding maternal sera but had much higher percentages of arachidonic acid in each of the major lipid categories. The maternal phospholipids and cholesterol esters had higher proportions of arachidonic acid in preeclampsia than in normal pregnancy. Samples from placentas of preeclamptic pregnancies had significantly lower proportions of arachidonic acid in the nonesterified fatty acids and triglycerides than normal placentas. These findings suggest a decreased availability of arachidonic acid in the fetal circulation which may result in decreased production of prostacyclin in preeclampsia. (AM. J. OssTET. GYNECOL. 148:5, 1984.)

Recently, there has been expanding interest in the role of arachidonic acid and its metabolites in the physFrom the Division if Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, University if Minnesota, and the Harmel Institute. Supported in part by Grant AM04524 National Institutes if Health, by United States Public Health Service Grant HL08214, Prog;ram Projects Branch, Extramural Prog;rams, National Heart and Lung Institute, by a g;rant from the Graduate School, University if Minnesota, and by the H ormel Foundation. Received for publication August 26, 1983. Revised September 15, 1983. Accepted September 16, 1983. Reprint requests: Paul L. Ogburn, Jr., M.D., Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Box 395, Mayo Memorial Building, University of Minnesota Hospitals, Minneapolis, Minnesota 55455.

iology of pregnancy and labor. Since the discovery that arachidonic acid is the natural precursor of prostaglandin F 2 a and E2 , extensive research has been done suggesting that the onset of labor is initiated by the enzymatic release of arachidonic acid from placental tissues.!. 2 Other investigators have suggested that another arachidonic acid metabolite, prostacyclin (prostaglandin 12 ), is elevated in normal pregnancy, leading to vasodilation and smooth muscle relaxationY· 4 The production of prostaglandin 12 is decreased in the maternal and fetal systems in preeclampsia. The increase in the ratio of thromboxane A2 to prostaglandin 12 has been suggested as a mechanism for many of the pathologic changes seen in preeclampsia. 4 - 6 Alteration in fetal prostaglandin 12 activity in preeclampsia has

5

6

Ogburn et al.

January 1, 1984 Am.]. Obstet. Gynecol.

Table I. Arachidonic acid patterns in maternal serum Measurement

Total NEFA (mg/dl) Total NEAA (t..t.g/dl) % NEFA AA % PLAA %TGAA % CEAA

Preeclampsia

31.9 391.9 1.35 10.39 0.98 6.80

± ± ± ± ± ±

9.7 115.8 0.17 0.58 0.12 0.54

NorrruJl pregnancy

16.8 274.8 1.56 8.98 l.l3 5.13

± ± ± ± ± ±

3.4 80.4 0.18 0.36 0.14 0.35

Significance

NS NS NS p < 0.05 NS p < 0.02

Arachidonic acid patterns in maternal serum are given as the mean value± SEM. Total NEFA =Total nonesterified fatty acids. Total NEAA = Total nonesterified arachidonic acid. % NEFA AA = Percent of total nonesterified fatty acids made up of arachidonic acid. % TG AA = Percent triglycerides made up of arachidonic acid. % CE AA = Percent of cholesterol esters made up of arachidonic acid. NS = Not significant.

Table II. Arachidonic acid patterns in cord serum Measurement

Total NEFA (mg/dl) Total NEAA (t..t.gldl) % NEFA AA % PLAA %TGAA % CEAA

Preeclampsia (n = 6)

6.96 176.7 2.58 18.41 2.55 12.1

± ± ± ± ± ±

0.56 19.9 0.30 0.41 0.26 1.88

Normal pregnancy (n = 6)

6.24 262.8 5.04 17.42 4.95 15.28

± ± ± ± ± ±

2.41 84.1 0.67 0.71 0.5 1.89

Significance

NS NS p < 0.01 NS p < 0.005 NS

Arachidonic acid patterns in cord serum are given as mean ± SEM. Total NEFA = Total nonesterified fatty acids. Total NEAA =Total nonesterified arachidonic acid. % NEFA AA =Percent of total nonesterified fatty acids made up by arachidonic acid. % PL AA = Percent of phospholipids made up of arachidonic acid. % TG AA = Percent of trigylcerides made up of arachidonic acid. % CE AA = Percent of cholesterol esters made up of arachidonic acid. been related to placental insufficiency, and attempts to treat preeclampsia with prostaglandin 12 administration have been made. 7 Our previous work has shown the patterns of esterified and nonesterified arachidonic acid in cord blood to be quite different from those of maternal blood in normal pregnancy. 8 The present study was undertaken to determine whether these arachidonic acid patterns are altered in preeclampsia. Methods

Fifteen blood samples were drawn from 11 volunteers with preeclampsia in the third trimester of pregnancy. All patients had blood pressures of at least 150/100 mm Hg on at least two occasions with either significant proteinuria or severe edema (or both). All samples of blood were drawn prior to the administration of oxytocin for induction of labor. Four of the patients with preeclampsia were multiparous, and five patients had preterm gestations. All patients were white and one was carrying twins. Magnesium sulfate administration had begun in four patients. The range of ages of these patients was 18 to 34 years, and gestational age ranged from 30 to 44 weeks. Nineteen control blood samples were drawn from 10 volunteers with normal pregnancies in the third trimester or in early labor. Samples of cord blood were obtained from the umbilical veins of the placentas of six preeclamptic and

six control pregnancies. The control cord blood samples came from pregnancies of gestational ages ranging from 40 to 41 \t2 weeks. Two of the control patients were multiparous; three cesarean sections were performed, and oxytocin had been used in two. Control infants weighed from 2,840 to 4,500 gm. The preeclamptic cord blood samples came from pregnancies ranging from 31 to 44 weeks' gestational age and all patients were primiparous. One cesarean section was performed. Oxytocin was used in five cases, and magnesium sulfate was used in three cases. Infants born to preeclamptic mothers ranged from 760 to 3,260 gm in weight. The serum samples were frozen until analyses could be performed. The samples were thawed, and a measured amount of heptadecanoic acid was added as an internal standard for quantification. The lipids were extracted from samples with chloroform/methanol (111, v/v). The extract was dried under a stream of nitrogen, and the lipids were redissolved in 0.1 ml of chloroform/methanol (211, v/v). Thin-layer chromatography was performed on silicic acid-impregnated paper (Gelman Instrument Co., Ann Arbor, Michigan). The developing solvent for thin-layer chromatography was 30° to 60° C petroleum ether/diethylether/acetic acid (90/10/1). The thin-layer chromatography papers were developed with 0.1% dichlorofluorescein solution, and the bands of lipids were visible under ul-

Serum arachidonic acid levels

Volume 148 Number I

traviolet light. The phospholipids, triglycerides, nonesterified fatty acids, and cholesterol esters appeared as distinct bands which were then cut apart and put into Teflon-lined screw-capped glass tubes. These were esterified with boron trifluoride-methanol at 75° C for 1 hour. After esterification, the lipids were extracted with petroleum ether (30° to 60° C) three times to ensure quantitative transfer. The samples were taken to dryness under nitrogen at room temperature and redissolved in a small amount of petroleum ether for gas-liquid chromatographic analysis. Gas-liquid chromatography of the fatty acid methyl esters was carried out on a Packard Model428 gas chromatograph (Packard Instrument Co., Inc., Downers Grove, Illinois) equipped with a flame ionization detector and an aluminum column, 12 feet by Ys inch, packed with 10% Silar 10C on 100-120 Gas Chrom Q. The temperature was programmed from 110° to 230° C at 4 o per minute with a final hold of 8 minutes. This procedure allowed separation of methyl esters of fatty acids from eight carbons to 22 carbons. Peak areas were measured by electronic digital integrator (Model CRS-104, Infotronics, Houston, Texas). Quantification of fatty acids was accomplished by injection of a known amount of methyl heptadecanoate in comparison with the internal standard of each sample. Identification of fatty acid methyl esters was made by comparison with known methyl ester standards. Results were tested for significance by the two-tailed Student t test, with p < 0.05 providing evidence for significant differences. Results

The characteristics of the control patients have been previously published. 8 Table I compares lipid patterns in preeclamptic and normal maternal sera. The mean total nonesterified fatty acid and mean total nonesterified arachidonic acid levels were higher in the preeclamptic group, but the difference did not reach statistical significance. Analysis of the esterified fatty acids shows that, in preeclampsia, the phospholipids and cholesterol esters have a significantly higher proportion of arachidonic acid compared to that in normal pregnancy. Table II compares lipid patterns in preeclamptic and normal cord sera. Samples from preeclamptic placentas had significantly lower proportions of arachidonic acid in the nonesterified fatty acid and triglyceride components compared to those of normal placentas. Comparison of cord blood values to corresponding maternal serum values showed significant differences except for the total nonesterified arachidonic acid (p < 0.005). The cord blood samples had quantitatively less total nonesterified fatty acids in both groups. Cord sera had a greater percentage of arachidonic acid in

Maternal Blood

Phospholipid AA

7

Cord Blood

I

INCREASE

0~

I

Nonesterified AA

I

Cholesterol Ester AA ~---~~--~L_:T_::ri~gl~y~ce~r~id~e~A:_:A_j BLOCKAGE

Fig. l. Net effect of preeclampsia on maternal and fetal serum arachidonic acid.

nonesterified fatty acids, in phospholipids, in triglycerides, and in cholesterol esters than corresponding maternal sera. Total nonesterified arachidonic acid was not significantly different in maternal or cord sera of either the preeclamptic group or the normal pregnancy group in spite of large mean differences. Comment

Although extensive work has been done on the phospholipids and nonesterified arachidonic acid in the placenta, membranes, and amniotic fluid (as well as work documenting arachidonic acid metabolites), very little documentation of blood levels of arachidonic acid has previously been published. 2 - 8 Esterified and nonesterified arachidonic acid is relatively abundant in the bloodstream compared to its very active metabolites (the prostaglandins, thromboxane A 2 , prostaglandin I 2 , and others). It is precisely in the bloodstream where the arachidonic acid metabolites, prostaglandin I 2 and thromboxane A2 , may have their most notable effects. Prostaglandin I 2 is held responsible for vasodilation in normal pregnancy, while thromboxane A2 activity, unchecked by counterbalancing prostaglandin I 2 , is held responsible for the vasoconstriction (and occasionally platelet activation) seen in preeclampsia. 2 - 7 Our previous publications in this area have proposed theories of circulating arachidonic acid conversion to thromboxane A 2 as part of the pathogenesis of preeclampsia. 9 • 10 The results presented above show significant differences in arachidonic acid patterns and maternal and cord sera when normal third-trimester and preeclamptic pregnancies are compared. There is a higher percentage of arachidonic acid in phospholipids and cholesterol esters in preeclamptic women than in control subjects. In cord blood there is relatively more nonesterified and triglyceride arachidonic acid in the control pregnancies than in the preeclamptic pregnancies. The effect in preeclampsia would seem to be a net shift of arachidonic acid from the nonesterified (and the companion triglyceride) components in the fetus to the phospholipids and cholesterol esters in the maternal circulation (see Fig. 1). This shift of arachidonic acid from the fetal to the maternal compartment may also be seen by examination of the maternal: cord non esterified arachidonic acid ratio. In the control pregnancies, this ra-

8 Ogburn et al.

January I , 1984 Am. ]. Obstet. Gynecol.

Lipolysis

Lipoxygenase

Cyclooxygenase

HPETE HETE --+---, SRS-A Leucotrienes

Inhibits

Edema Capillary Leak Vasospasm

Bronchospasm Vasodilation Smooth Muscle Rela·xation

Platelet Aggregation Smooth Muscle Stimulation

Fig. 2. Theoretical model for arachidonic acid metabolism for normal and preeclamptic pregnancies.

tio is (274.8 J.Lg/dl: 262.8 J.Lg/dl) = 1.05. The ratio is much higher in preeclamptic pregnancies (391.9 J.Lg/ dl: 176.7 J.Lg/dl) = 2.22. Because the fetal circulation favors the production of prostaglandin I 2 , decreased availability of arachidonic acid for the fetus may result in a decrease in the total prostaglandin I 2 produced in the maternal-fetal unit in preeclampsia. Another aspect of the pathogenesis of preeclampsia may be the specific inhibition of prostaglandin I 2 production in the maternal and fetal circulations. The substances which are the most likely candidates for this prostaglandin I 2 inhibition are the lipoxygenase products of arachidonic acid, including hydroperoxyeicosatetraenoic acid and hydroxyeicosatetraenoic acid. 10 - 14 Not only does hydroperoxyeicosatetraenoic acid inhibit prostaglandin I 2 production, but preliminary studies have shown placental production of hydroperoxyeicosatetraenoic acid and hydroxyeicosatetraenoic acid to be higher in preeclamptic pregnanciesY Lipoxygenase products of arachidonic acid (including hydroxyeicosatetraenoic acid) make up a family of compounds known as the leukotrienes, which can cause contraction of airway smooth muscle, increased vascular permeability, and chemotaxis. 14 Fig. 2 assimilates the above information into a theory which relates arachidonic acid metabolism to normal and preeclamptic pregnancies. In the normal pregnancy, arachidonic acid is converted by cyclooxygenase to prostaglandin H 2 which, in turn, converts to prostaglandin I 2 (causing smooth muscle relaxation and vasodilation). In labor, a greater portion of prostaglandin H 2 is converted to prostaglandins F 2 a and E2 [which causes uterine contractions and lipolysis (which increase arachidonic acid levels)]. 15 In preeclampsia, we have seen that relatively less nonesterified arachidonic acid is available to the fetal circulation than in normal pregnancy. With the propensity of fetal vascular tissue to produce prostaglandin I 2 , this decrease in arachi-

donie acid in the fetus would, in turn, decrease the total prostaglandin I 2 in the maternal-fetal unit. In addition, the increased placental production of lipoxygenase products in preeclampsia would inhibit the normal rate of prostaglandin I 2 production. "1 With decreased prostaglandin I 2 production, thromboxane A 2 effects (vasospasm, platelet aggregation, and smooth muscle stimulation) would predominate and uterine contractions could be more easily stimulated. The increased lipoxygenase activity postulated for preeclampsia could explain edema and the occasional pulmonary edema as well as proteinuria (secondary to capillary leakage) seen with this condition. Four of the preeclamptic patients included in this study were multiparous, and five preeclamptic. patients were preterm; a homogeneous group may have been more desirable. However, the control group included multiparous and primigravid patients, and no significant differences in arachidonic acid patterns were seen at various gestations from 30 weeks to early labor at term. 8 Moreover, the same general mechanisms described above could apply to preterm and multiparous preeclamptic gestations as well as to the primigravid patients at term. Another source of potential variability for this study is the fact that arachidonic acid in nonesterified fatty acids levels varies with dietary intake and starvation. Again, we have shown no significant difference in nonesterified fatty acids or arachidonic acid patterns in the control patients whether fasting or nonfasting prior to labor or in early labor. 8 A theory relating arachidonic acid serum patterns in preeclamptic patients and cord blood to the pathogenesis of the disorder has been outlined above. It is hoped that future research will either refine or refute this theory. REFERENCES 1. Bergstrom, S., Danielsson, H., and Samuelsson, B.: The enzymatic formation of prostaglandin E 2 from arachi-

Serum arachidonic acid levels

Volume 148 Number I

2. 3. 4. 5. 6. 7. 8. 9.

donie acid. 32. Prostaglandins and related factors, Biochem. Biophys. Acta 90:207, I964. MacDonald, P. C., Schultz, M., Duenhoelter, J. H., et al.: Initiation of human parturition. I. Mechanisms of action of arachidonic acid, Obstet. Gynecol. 44:629, I974. Lewis, P. J., Boylan, P., Friedman, L.A., et al.: Prostacyclin in pregnancy, Br. Med. J. 280:I58I, I980. Remuzzi, G., Zoja, C., Marchesi, D., et al.: Plasmatic regulation of vascular prostacyclin in pregnancy, Br. Med. J. 282:5I2, I982. Remuzzi, G., Marchesi, D., Zoja, C., et al.: Reduced umbilical and placental vascular prostacyclin in severe preeclampsia, Prostaglandins 20:I05, I980. Bussolino, F., Benedetto, C., Massobrio, M., et al.: Maternal vascular prostacyclin activity in preeclampsia, Lancet 2:702, I980. Lewis, P.J., Shepherd, G. L., Ritter,]., et al.: Prostacyclin and preeclampsia, Lancet 1:599, I98l. Ogburn, P. L., Jr., Johnson, S. B., Williams, P. P., et al.: Levels of free fatty acids and arachidonic acid in pregnancy and labor, J. Lab. Clin. Med. 95:943, I980. Ogburn, P. L., Jr., Brenner, W. E., Dingfelder, J. R., et al.: Arachidonic acid's role in the pathology and physiology of pregnancy and labor, Prog Lipid Res. 20:243, I98I.

IO. Ogburn, P. L., Jr., and Brenner, W. E.: The physiologic actions and effects of prostaglandins, Kalamazoo, I98I, The Upjohn Co. II. Bunting, S., Gryglewski, R. J ., Mancada, S., et al.: Arterial walls generate from prostaglandin endoperoxides a substance (prostaglandin X) which relaxes strips of mesenteric and coeliac arteries and inhibits platelet aggregation, Prostaglandins 12:897, I976. I2. El Tahir, K. E. H., and Williams, K. I.: Trapped blood elements within the decidua of the rat pregnant uterus generate a lipoxygenase product(s) which inhibit myometrial prostacyclin synthesis, Br. J. Pharmacal. 73:695, 1981.

I3. Ogburn, P. L., Jr., Maynard, S., Williams, P. P., et al.: Arachidonic acid metabolism and preeclampsia, in Xth World Congress of Gynecology and Obstetrics Abstracts, San Francisco, California, October I7-22, I982, p. 82. (Abst.) I4. Samuelsson, B.: Leukotrienes: A novel group of compounds including SRS-A, Prog. Lipid Res. 20:23, I98l. I5. Ogburn, P. L., Jr., and Brenner, W. E.: Circulating free arachidonic acid responses to prostaglandin F2a treatment in pregnancy, Prog. Lipid Res. 20:247, I98l.

Treating genital condyloma during pregnancy with the carbon dioxide laser Alex Ferenczy, M.D. Montreal, Quebec, Canada The therapeutic effectiveness of the carbon dioxide laser was evaluated in 43 pregnant women with extensive urogenital and anal condylomas. All patients received one treatment and were followed up for an average of 9 months after delivery. Laser failures (persistent disease) and recurrences (new disease) were stratified according to the location of the lesions and the gestational age of the patients. The overall failure rate was 5%; 6% of the women with multiple sites involved had persistent disease but none of the lesions confined to the vulva persisted after laser treatment. The recurrence rate was 14%; 33% and 17% of the patients treated during the first and second trimesters, respectively, had recurrent disease irrespective of the areas involved. Recurrences were not observed in women treated during the third trimester of pregnancy. Laser vaporization of genital condylomas was not associated with perioperative or postoperative bleeding or infections. Laser therapy is an attractive means of treating urogenital and anal condylomas during pregnancy and is most effective near term. (AM. J. OBSTET. GYNECOL. 148:9, 1984.)

Genital warts or condyloma acuminatum is caused by the sexually transmitted human papillomavirus (HPV),

From the Departments r-if' Pathology and Obstetrics and Gynecology, The Sir Mortimer B. Davis Jewish General Hospital and McGill University. This work was supported by the SD-Gyn-Path-4 and Ob/Gyn-6 Developmental Research Funds af The Sir Mortimer B. Davis jewish General Hospital. Received for publication May 12, 1983. Accepted August 30, 1983. Reprint requests: Alex Ferenczy, M.D., Department of Pathology, The Sir Mortimer B. Davis Jewish General Hospital, Department af Pathology, 3755 Cote Ste-Catherine Road, Montreal, Quebec, Canada H3T JE2.

type 6 and type II. 1 In addition to their carcinoma precursor potentials, 2 • " during pregnancy they may rapidly enlarge and obliterate the birth canal, thereby precluding vaginal delivery. Also, transmission of HPV from warts to the newborn infant is possible when condylomas are present along the lower female genital tract. Several studies found that a large proportion (40%) of infants and children developing laryngeal warts were delivered of mothers harboring genital condyloma. 4 - 6 The fact that both laryngeal papillomas and genital condylomas contain the same type of HPV (type II) lends credence to the concept that HPV may be transmitted from the mother to the newborn m9