Elevated serum human chorionic gonadotropin as evidence of secretory response in severe preeclampsia

Hsu et al.

Volume 170, Number 4 Am j Obstet Gynecol

14. Patrick J, Campbell K, Carmichael L, Natale R, Richardson B. Patterns of human fetal breathing during the last 10 weeks of pregnancy. Obstet Gynecol 1980; 56: 24-30. 15. Pillai M, James D. Development of human fetal behavior: a review. Fetal Diagn Ther 1990; 5: 15-32.

16. Fox HE, Inglis J, Steinbrecher M. Fetal breathing move1. Relationship to ments in uncomplicated pregnancies. age. Am J OBSFET GYNECOL 1979; 134: 544-6. gestational

17. Vintzileos AM, Campbell WA, Nochimson Dj, Weinbaum Pj. Fetal breathing as a predictor of infection in premature rupture of the membranes. Obstet Gynecol 1986; 67: 813-7. RL, Glass C, Boehm detection of membranes: rupture premature Am J OBSFEF GYNECOL 199 1; 165: 1099-104.

18. Roussis

P, Rosemond

FH.

Preterm of infection.

S, et al. Fetal body and 19. Goldstein 1, Romero R, Merrill infecbreathing of intraamniotic movements as predictors Am j tion in preterm rupture of membranes. premature OBSTF. 'r GYNECOI. 1988; 159: 363-8. R, Hobbins J. Fetal 20. Roberts AB, Goldstein 1, Romero breathing rupture of movements after preterin premature Am j OBSTET GYNECOL 199 1; 164: 821-5. membranes.

21. Sibai BM, Lipshitz J, Anderson GD, Dilts PV. Reassessment of intravenous MgSO4 therapy in preeclampsiaeclampsia. Obstet Gynecol 1981; 57: 199-202. 22. Elliott JP. Magnesium sulfate as a tocolytic agent, Contemp Obstet Gynecol 1985; 25: 49-61. 23. Hankins GDV, Hammond TL, Yeomans ER. Amniotic cavity accumulation of magnesium with prolonged magnesium sulfate tocolysis. j Reprod Med 1991; 36: 446-9.

Elevated serum human chorionic gonadotropin as evidence in of secretory response severe preeclampsia Chaur-Dong Hsu, MD, MPH, ' Daniel W. Chan, PhD, ' Brian Iriye, MD, ' Timothy R. B. Johnson, MD, ' Shih-Fen Hong, MPH, * and John T. Repke, MD' Baltimore, Ma?yland OBJECTIVE: Because preeciampsia is a trophoblastic disorder and human chorionic gonadotropin is secreted from trophoblast, we sought to determine whether measurement of serum human chorionic gonadotropin might reflect a different trophoblastic secretory response of preeclampsia. STUDY DESIGN: Twenty patients with mild preeclampsia and 12 with severe preeclampsia were matched with 32 healthy normotensive women in the third trimester with singleton pregnancies. Serum total human chorionic gonadotropin and total human chorionic gonadotropin-0 were measured by a two-site immunoenzymometric assay, and total hCG-txwas determined by a double-antibody radioimmunoassay. Wilcoxon signed-rank and Mann-Whitney rank-sum tests were used for statistical analysis. RESULTS: Serum total human chorionic gonadotropin, total human chorionic gonadotropin-a, and total human chorionic gonadotropin-P levels were significantly higher in severely preeclamptic women (p < 0.05), but not in those with mild preeclampsia, compared with those in their matched controls. CONCLUSION: Elevated serum human chorionic gonadotropin levels in severely preeclamptic women might reflect a significantly pathologic change and secretory reaction of the placenta. (Am J OBSTET 1994; 170: 1135-8. ) GYNECOL

Key words: Human chorionic gonadotropin, preeclampsia, trophoblast Preeclampsia.is one of the most common and least understood disorders of pregnancy, affecting 5% to 7% of pregnancies. Associatedwith hypertension, proteinFrom the Division of Maternal-Fetal Medicine, Department of Gynecology and ObstetnCS'a and the Division of Chncial Chemistr% DePartment ofPatholoob Johns Hopkins University School ofMedicine. Receivedfor publication November 1,1993; acceptedNovember 1, 1993. Reprint requests: Chaur-Dong Hsu, MD, MPH, Department of Gynecologyand Obstetrics,Yale University, P. O. Box 208063, New Haven, CT 06520. Copyright C 1994 by Mosby--YearBook, Inc. 0002-9378194 $3.00 +0 611152565

uria, and edema, it is a disorder that has been studied extensively,and yet the causeremains largely unknown. Becausepreeclampsiais likely a trophoblastic disorder, ' investigating pathologic changesand secretory reaction of the placenta may be essentialfor understanding this disease. Human chorionic gonadotropin (hCG), a glycoprotein hormone, is produced by normal and neoplastic trophoblastic tissues.' Twin pregnancies' and molar hCG levels higher of and are pregnancies' produce higher incidence of preeclampsia than a with associated Serum bCG singleton pregnancies. are uncomplicated 1135

April 1994 Am j Obstet Gynecol

1136 Hsuetal.

has been reported to be increased in patients with preeclampsia.' ' Serum levels of hCG-Pwere also shown to be increased in preeclamptic women.' Therefore we sought to determine whether the level of serum hCG was correlated with the severity of preeclampsia and might reflect a different trophoblastic secretory response of this disease. Materlal and methods This investigation was designed as a cross-sectional study. It was carried out at the Johns Hopkins Hospital with the approval of Institutional Review Board. Sixtyfour pregnant women were studied. Thirty-two pregnant women with preeclampsia were matched with 32 normotensive, healthy pregnant women with singleton pregnanciesin the third trimester. None of them had a history of chronic hypertension. Subjectswere matched for gestational age, parity, maternal age, and race. Among the 32 subjects20 had mild preeclampsiaand 12 had severe preeclampsia. Sixty-three percent of preeclamptic patients were nulliparous. Mild preeclampsiawas defined as (1) an absolute blood pressure 30 mm Hg systolic 2- 140/90mm Hg or an increaseof Z-_ or 15 mm Hg diastolic over baseline values in the first 20 weeks of pregnancy, (2) proteinuria (dipstick with 2-30 mg/dI or 2-1+ in a catheterized specimen), and (3) hyperuricernia (2t5.5 mg/dl). Additionally, patients returned to normal blood pressure and resolution of proteinuria within 12 weeks of delivery. The criteria for severe preeclampsia were (1) blood pressure ;--160 mm Hg systolic or 2-110 mm Hg diastolic and (2) proteinuria >5 gm in 24 hours. In addition, any patient with oliguria (s400 ml in 24 hours), cerebral or visual disturbances, epigastric pain, pulmonary edema, or abnormal platelet count and liver function profile was included in severe preeclampsia. Two patients were included in the severepreeclamptic group with only trace protein but who met all other severe preeclamptic criteria and had abnormalities of the platelet count and liver function profile. After informed consent was obtained from the women, venous blood samples were obtained. Blood samples were centrifuged at 20OOgfor 15 minutes at V C. Serawere collected and stored at - 70* C until the assaywas performed. Serum levels of total hCG and total hCG-P were measured by a two-site immunoenzymometric assay that was performed entirely in AIA-1200 system with AIA-PACK (Tosoh, San Francisco). The assay of total hCG has a high specificity and minimal crossreactivity with follicle-stimulating hormone of 0.060/c,luteinizing hormone of 0.70%, and thyroid-stimulating hormone of 0.06%. The intraassay and interassay coefficients of variance averaged 4.2% and 4.7%, respectively. The assayof total hCG-P has a high specificity to the

O-subunit and a minimal cross reactivity with folliclestimulating hormone of 0.16%, luteinizing hormone of 0%, and thyroid-stimulating hormone of 0%. The intraassayand interassaycoefficients of variance averaged 3.5% and 3.8%, respectively. Serum levels of total hCG-a were determined by a double antibody radioimmunoassay with chloramineT-iodinated a-subunit as tracer, a specific rabbit antia-subunit antibody (Weintraub/National Institutes of Health, Hazleton, VA.), and a goat antirabbit gamma globulin precipitating antibody.' The intraassay and interassay coefficients of variance averaged 6.2% and 13.3%. Statistical differences between matched pairs were evaluated by Wilcoxon signed-rank test, Wilcoxon Mann-Whitney rank-sum test, conventional correlation 0.05 Ap < value of and coefficient variance. coefficient, was used to determine statistical significance. All data are expressed as means t SE. Results Subject characteristicsand clinical laboratory data of patients with preeclampsia and their matched controls differin Table I. There significant were no are shown encesin maternal age, gestational age, parity, race, and hematocrit concentration between patients with preeclampsia and their matched controls. Patients with severe or mild preeclampsia had significantly higher systolic pressure, diastolic pressure, mean arterial pressure, urine protein, serum creatinine levels, and serum uric acid levels than did their matched controls. However, neither creatinine nor hernatocrit was corTelated positively with serum levelsof total hCG, total hCG-a or total hCG-0 in preeclamptic patients and their matched controls. Table 11showsthat patients with severepreeclampsia had significantly higher levels of serum total hCG, total hCG-otand total hCG-0 than did those in their matched controls (p < 0.05). However, there were no significant differences in serum levels of total hCG, total hCG-a, and total hCG-0 betweenpatients with mild preeclampsia and those of their matched controls (Table IIII).

Comment Although elevated serum levels of total hCG` ' and hCG-R' in patients with preeclampsia have been reported, less information is available on the correlation between serum hCG level and severity of preeclampsia. Our results clearly demonstrated that serum hCG levels were significantly elevated in patients with severe preeclampsia, but not with mild preeclampsia, compared with their matched controls. Similarly, our recent study showed that soluble thrombomodulin antigen, a cell surface glycoprotein in placenta, was also significantly higher in patients with severe preeclampsia than in their matched controls.' This might reflect a significant

Hsu et al. 1137

Volume 170, Number 4 Am j Obstet Gynecol

Table 1. Subject data Matched controls (n

Severe preeclampsia (n = 12) Maternalage(yr) Gestationalage(wk) Bloodpressure(mm Hg) Meanarterial Diastolic Systolic Proteinuria(dipstick, mg/dI) HematocritM Serumcreatinine(mg/dl) Serumuric acid(mg/dl)

Mild preeclampsia (n = 20)

Matched controls (n = 20)

24.0 :t2.6 34.4 t 1.3

25.3 :t2.6 35.4 ± 1.2

22.3 -i: 2.1 38.0 -t 1.1

24.0 t L8 38.5 ± 1.2

117.6 2.3* 96.9 2.3* 159.1 4.1* 3+ (338)*

74.9 t 2.6 59.5 ± 2.5 105.8 t 3.1 0 (:5 30)

112.6 2.6* 92.9 3.1* 151.9 3.0* 2+ (103)*

78.3 ± 2.5 62.7 3.0 ýt 109.7:t2.4 0 (!5 30)

36.1 ± 2.0 0.79 ± 0.04 4.5 :t- 0.4

36.2 ± 1.1 0.99 :t0.04* 7.0 ± 0.5*

35.2 ± 1.2 0.93 ± 0.04* 6.4 t: 0.3*

34.6 1.5 ýt 0.77 ± 0.02 4.3 t 0.4

< 0.05.

Table 11. Serum levels of total hCG, total hCG-a, and total hCG-P in severelypreeclamptic groups and matched controls (n = 12) Matched Severe (n = 12) preeclampsia contrals 46,538± 15,001

Total hCG(mIU/ml)

1,585± 456

Total hCG-a(ng/ml)

55,823± 18,051

Total hCG-P(mIU/ml)

11,203± 2,346 p<0.05 p<0.05

550 ± 64 14,307± 2,967

p<0.05

Table III. Serum levels of total hCG, total hCG-ot,and total hCG-P in mild preeclamptic groups and matched controls I Total hCG (mIU/ml) Total hCG-cL (ng/ml) Total hCG-ß (mIU/ml)

Mild preeclampsia (n = 20)

Matchedcontrols(n = 20)

16,098 ± 3,067

15,438 :L3,060

685

95 «-±:

20,664 f: 3,961

NS

609 i: 59 NS 19,769 :t3,862 NS

NS, Not significant.

pathologic change and secretory response of placenta in severepreeclampsia. The normal placenta differentiates during pregnancy with the cytotrophoblast dominant in early gestation and the syncytiotrophoblast dominant in late preg" is It well known that the cytotrophoblast is an nancy. undifferentiated stem cell, and the syncytiotrophoblast is a differentiated trophoblast transformed from cytotrophoblast. " Although the mechanism of regulation of gestational hCG remains largely unknown, it is generally accepted that hCG, hCG-a, and hCG-P are only " by secreted syncytiotrophoblast. However, it has been shown that a hCG-a can be secreted from cytotrophoblast."- " Also, hCG-a messengerribonucleic acid can be localized from cytotrophoblast with in situ hybridization. " Hoshina et al." summarized that cytotrophoblastscontain mostly hCG-(xand a small amount of total hCG and syncytiotrophoblasts contain all total hCG,

hCG-P, and hCG-a. Recently Maruo et al. " demonin be detected hCG-P the cytotrophothat could strated blast in very early pregnancy. In preeclampsia placental pathologic examination in focal the syncytiotrophoblast necrosis cellular reveals and increased mitotic activity with cellular proliferation in the cytotrophoblast." In addition, the proliferating is in rapidly transpreeclampsia severe cytotrophoblast formed into syncytiotrophoblast within 72 hours." Thus the balance betweencellular loss and replacement favor Elevated replacement. may of syncytiotrophoblast in hCG-a levels total patients with severe of serum preeclampsia. may be secondary to production ftom expanding cytotrophoblast and newly transformed syncytiotrophoblast. In addition to the elevation of serum total hCG-(x levels, our data also revealed that total hCG and total hCG-P were significantly higher in patients with severe

1138 Hsu et al.

preeclampsia than their matched controls. As previously mentioned, both total hCG and hCG-P are predominantly produced by syncytiotrophoblast." Additionally, they can also be detected in the differentiating cytotrophoblast." Thus an increased serum total hCG and total hCG-P in patients with severe preeclampsia may be related to increases of both differentiating cytotrophoblast and newly transformed syncytiotrophoblast. Although an effect of severe preeclampsia on intravascular volume might he considered as a confounder, the hematocrit concentrations of preeclamptic patients in this study were not significantly different from those of their controls. Furthermore, neither hematocrit nor creatinine was correlated positively with serum hCG level, and it is not likely that the elevated serum level of hCG was related to a confounding effect of hemoconcentration or renal impairment in patients with severe preeclampsia. In summary,serum levels of total hCG, total hCG-a, and total hCG-P are only significantly higher in patients with severe preeclampsia. Whether this abnormal trophoblastic secretory reactions in severe preeclamptic patients may reflect more severepathologic changesof the placenta or a different disease entity from mild preeclampsiaawaitsfurther investigation. Also, whether these three hCG levels can clinically serve as useful markers, measurements of free ix- and P-subunits of hCG with newer specific monoclonal antibodies and a large sample size to provide better specificity of assays and meaningful predictive values are currently under investigation. In additional, serial measurements of hCG for disease progress by means of a longitudinal study are also in progress. We thank Dr. Bruce D. Weintraub from the National Institutes of Health, who contributed the total hCG-(x assaysfor this study. REFERENCES

1. Redman CWG. Plateletsand the beginning of pre-eclampsia. N Engl j Med 1990;323:478-80. 2. Vaitukaitis j L, Ebersole ER. Evidencefor altered synthesis

Am]

April 1994 Obstet Gynecol

of human chorionic gonadotropin in gestational trophoblastic tumors. I Clin Endocrinol Metab 1976;42: 1048-55. 3, Long PA, Oat JN. Preeclampsiain twin pregnancy: severity and pathogenesis.Aust NZJ Obstet Gynaecol 1987; 27:1-5. 4. Curry SL, Hammond CB, Tyrey L, CreasmanWT Parker RT. Hydatidiform mole: diagnosis, management, and long-term follow-up of 347 patients. Obstet Gynecol 1975; 45: 1-8. 5. Smith GVS, Smith OW. Excessivegonadostimulatory horlate in toxemia of amounts oestrin of mone and subnormal pregnancy. Am J Physiol 1934;107;128-45. 6. Smith GVS, Smith OW. The anterior pituitary-like hormone in late pregnancy toxemia: a summary of results since 1932. Am J OBSTETGYNECOL1939; 36: 618-24.

7. Said ME, Campell DM, Azzam ME, MacGillivray 1. Betahuman chorionic gonadotropin levelsbefore and after the development of pre-eclampsia. Br J Obstet Gynaecol 1984,91:772-5. 8. Rosen SW,Weintraub BD. Ectopic production of the isolated alpha subunit of the glycoprotein hormone. N Engl J Med 1974;290: 1441-7. 9. Hsu CD, Iriye B, Johnson TRB, Witter FR, Hong SF, Chan DW. Elevated circulating thrombomodulin in severe preeclampsia. Am J OBSTETGYNECOL1993; 169: 148-9.

10. Enders AC. Formation of syncytium from cytotrophoblast in the human placenta. Obstet Gynecol 1965;25:378-86. 11. Kliman HJ, Feimann MA, StraussjF 111.Differentiation of human cytotrophoblast into syncytiotrophoblast in culture. Trophoblast Res 1987;2:407-21. 12. Fox H, Kharkongor FN. Immunofluorescent localization of chorionic gonadotropin in the placenta and in tissue cultures of human trophoblast. J Pathol 1970;101:277-82. 13. Hoshina M, Ashitaka Y, Tojo S. Inummohistochernical interaction on antisera to hCG and its subunits with chorionic tissue of early gestation. Endocrinol Jpn 1979; 26: 175-84. 14. GaspardJU, Hustin J, Reuter AM, Lambotte R. Franchimont P. Immunofluorescent localization of placental lactogen, chorionic gonadotropin and its alpha and beta subunits in organ cultures of human placenta. Placenta 1980;1:135-44. 15. Hoshina M, Boothby M, Boime 1. Cytological localization of chorionic gonadotropin a and placental lactogen mRNAs during development of the human placenta. J Cell Biol 1982;93: 190-8. 16. Maruo T, Ladines-Llave CA, Matsuo H, Manalo AS, Mochizuki M. A novel change in cytologic localization of human chorionic gondaotropin and human placental lactogen in first-trimester placenta in the course of gestation. Am J OBs-rETGYNECOL1992; 167: 217-22.

17. Jones CjP, Fox H. An ultrastructural and ultrahistochemical study of the human placenta in maternal pre-eclampsia. Placenta 1980;1:61-76.