Zeeman, Gerda G. MD; Alexander, James M. MD; McIntire, Donald D. PhD; Byrd, William PhD; Leveno, Kenneth J. MD
Department of Obstetrics and Gynecology, The University of Texas Southwestern Medical Center, Dallas, Texas.
Address reprint requests to: Gerda G. Zeeman, MD, Department of Obstetrics and Gynaecologie, Isala Klinieken, Locatie Sophia, Dokter van Heesweg 2, Postbus 10400, 8000 GK Zwolle, The Netherlands; E-mail: firstname.lastname@example.org.
Received June 3, 2002. Received in revised form August 23, 2002. Accepted August 29, 2002.
OBJECTIVE: To determine if maternal serum inhibin-A can be used as a marker for subsequent development of superimposed preeclampsia in women with chronic hypertension.
METHODS: Serum for measurement of inhibin-A was obtained at monthly intervals in women with chronic hypertension requiring antihypertensive medications. Superimposed preeclampsia, the primary outcome of interest, was diagnosed when hypertensive women developed proteinuria (at least 300 mg per 24-hour urine specimen). Serum inhibin-A was considered abnormally elevated when the value exceeded the mean plus two standard deviations of the log for chronically hypertensive women who did not develop preeclampsia.
RESULTS: A total of 61 women were enrolled in this study, and 21 (34%) developed superimposed preeclampsia. Inhibin-A levels increased with advancing gestational age. Ten women had abnormally increased inhibin-A levels; eight (80%) developed superimposed preeclampsia, compared with 13 of 51 (26%) women with normal inhibin-A levels (P < .001). Sensitivity and specificity were 38% and 95%, respectively, whereas the positive and negative predictive values were 80% and 75%, respectively.
CONCLUSION: Although inhibin-A was abnormally increased an average of 3 weeks before the clinical onset of superimposed preeclampsia, the sensitivity of the test as a screen was too limited to be clinically useful.
A variety of peptides produced by the placenta are of interest as possible markers for the prediction of preeclampsia. Examples include corticotropin-releasing hormone,1 chorionic gonadotropin,2 and inhibin-A.3 The latter seems to be particularly promising in the search for early pregnancy markers related to the development of preeclampsia.3 For example, Muttukrishna and colleagues3 measured these placental peptides in 1496 healthy nulliparous women and found inhibin-A and activin-A to be elevated before the onset of preeclampsia. They concluded that such testing may have a clinical application in identifying women at risk for developing preeclampsia and, in particular, those at risk for early onset preeclampsia. At least four other groups of investigators have concluded that inhibin-A screening may be useful for early prediction of preeclampsia,4–7 whereas one has challenged this.8 We have previously demonstrated that third trimester maternal serum inhibin-A levels of chronically hypertensive women without superimposed preeclampsia are comparable to those of healthy pregnant women.9 According to our search of the literature via MEDLINE (1966–2001; English language; key words “inhibin” and “pregnancy”), prediction studies, particularly those including women with underlying essential hypertension, have not been performed to date.
This study was performed to assess whether maternal serum inhibin-A levels were predictive for the development of preeclampsia in a cohort of women at especially high risk for this pregnancy complication. Specifically, we studied women with chronic hypertension requiring antihypertensive medications throughout pregnancy because approximately a third of such women develop superimposed preeclampsia. A test that could reliably predict or rule out the later occurrence of superimposed preeclampsia would allow future intervention studies aiming to prevent preeclampsia.
MATERIALS AND METHODS
On August 1, 1999, a specialized prenatal clinic was initiated at Dallas, Texas' Parkland Hospital for women with chronic hypertension. This clinic was staffed by faculty and fellows in the Division of Maternal-Fetal Medicine, The University of Texas Southwestern Medical School. Women enrolled in this clinic were those using antihypertensive medications before the index pregnancy and those who required initiation of such therapy for hypertension before 20 weeks' gestation. The latter group of women were those with blood pressures of greater than 150/100 mm Hg during at least two successive prenatal visits. All blood pressures were measured using standard aneuroid sphygmomanometers, and Korotkoff 5 was used to determine the diastolic pressure. This study was approved by the Southwestern Medical School's Institutional Review Board, and written informed consent was obtained from all participating subjects.
Women with chronic hypertension were seen biweekly. Initial evaluation included collection of a 24-hour urine specimen for total protein excretion. Obstetric ultrasound was routinely employed for assessment of gestational age and fetal surveillance during the third trimester. Women developing hypertension exceeding their early pregnancy values or new-onset proteinuria of at least 1+ dipstick (30 mg/dL) were admitted to the High-Risk Pregnancy Unit. Delivery was effected for severe preeclampsia or fetal jeopardy manifest as suboptimal growth or nonreassuring antepartum tests of fetal well-being. Superimposed preeclampsia was diagnosed in the presence of worsening hypertension plus new-onset proteinuria of at least 300 mg per 24 hours.
Serum for measurement of inhibin-A levels was obtained at monthly intervals commencing as early as 16 weeks' gestation, but no later than 27 weeks. Inhibin-A was measured using an enzyme-linked immunosorbent assay kit (Serotec, Oxford, United Kingdom). The inhibin-A detection limit was less than 3.9 pg/mL. Standard serum control samples were run with each assay. The inter- and intraplate coefficients of variation were less than 10%. There was less than 1% cross-reactivity with inhibin-B. An upper limit threshold for abnormally increased inhibin-A levels was determined for three different gestational age categories and defined as the mean value plus two standard deviations (SDs) of the log for 16–20 weeks', 20–24 weeks', and 24–28 weeks' gestation, respectively. These thresholds were developed using specimens from women with chronic hypertension who did not develop superimposed preeclampsia.
Statistical analysis was performed using χ2 and Student t tests. All tests were two sided, and P values less than .05 were considered significant. Exact confidence intervals for estimated proportions are reported. Analysis was done using SAS 8.2 (SAS Institute Inc., Cary, NC).
A total of 61 women with chronic hypertension who had inhibin-A levels measured at least twice during the pregnancy were identified. The last measurement was obtained an average of 3.2 weeks before the diagnosis of preeclampsia in those women developing this complication. Superimposed preeclampsia developed in 21 women (34%); demographic characteristics for these women are compared with those for women without preeclampsia in Table 1. Approximately half of the women were of Hispanic origin, and the other half were black. A third of the women enrolled were nulliparous. There were no significant differences. Shown in Table 2 are selected pregnancy outcomes in women who developed superimposed preeclampsia versus those who did not develop this complication. As expected, women with superimposed preeclampsia had shorter gestations, increased incidence of fetal growth restriction, and more stillborn fetuses. Placental abruption occurred in 10% of women with superimposed preeclampsia, compared with 3% in those without (P = .228). Two of the stillborn fetuses occurred in women who refused hospitalization for severe superimposed preeclampsia, and one was due to placental abruption at 31 weeks' gestation.
Shown in Table 3 are inhibin-A levels throughout pregnancy in 61 women with chronic hypertension. Inhibin-A levels significantly increased as gestation advanced, from 88 pg/mL at 16–20 weeks' gestation to 906 pg/mL at term (P < .001). Inhibin-A results are shown as means (SD) and means plus two SD, respectively. Shown in Figure 1 are the maximum inhibin-A levels (mean ± two SD), according to weeks' gestation in the 40 women who did not develop superimposed preeclampsia versus the 21 women who did. Shown in Figure 2 is the distribution of women with superimposed preeclampsia in relation to abnormal or normal inhibin-A levels. The sensitivity and specificity for inhibin-A levels exceeding the mean plus two SD before 28 weeks' gestation for the prediction of superimposed preeclampsia were 38% and 95%, respectively, whereas the positive and negative predictive values were 80% and 75%, respectively (Table 4).
This study was designed to determine if maternal serum inhibin-A can be used as a marker for the subsequent development of superimposed preeclampsia in women with chronic hypertension. A marker with the potential to predict the later development of superimposed preeclampsia could be used for identification of a high-risk population for enrollment in future intervention studies. The sensitivity of inhibin-A for prediction of superimposed preeclampsia was poor (38%); however, the specificity was good (95%). We have tried to put these results into a meaningful clinical context, given the fact that the a priori risk of superimposed preeclampsia, based solely on preexisting chronic hypertension, was 34%. It appears that an abnormal inhibin-A level is more meaningful than history of preexisting hypertension alone in the prediction of superimposed preeclampsia because eight of ten women (80%) with abnormal levels developed preeclampsia. However, normal inhibin-A levels are less meaningful for the prediction of the absence of superimposed preeclampsia because 26% with normal levels also developed preeclampsia. It is important to emphasize that the use of a test such as inhibin-A for prediction of disease outcome is greatly influenced by the a priori risk of the study population. Put another way, a test used for prediction of a disease is more likely to be meaningful when the disease predicted occurs with high frequency in the study population.
To date, six reports3–8 have been published on the utility of maternal serum inhibin-A concentration for the prediction of preeclampsia. In five reports,3–7 inhibin-A levels were considered to be predictive (sensitivities 23–47%) for the development of preeclampsia in women at low risk for this disease (1.2–5.5% incidence of preeclampsia). None of these reports specifically addressed women with chronic hypertension. The general theme of these other reports is similar to our interpretation of our results; although inhibin-A levels are statistically related to the subsequent development of preeclampsia, they are not likely to be sufficient for screening in clinical medicine.
So what does the clinician do with our findings regarding inhibin-A levels in women with chronic hypertension? Considering that the inhibin-A levels were measured several antepartum weeks before the clinical onset of superimposed preeclampsia, the clinician who receives an abnormal inhibin-A result faces the dilemma of trying to prevent a disease that is currently not considered preventable. Conversely, the clinician who obtains a normal inhibin-A level cannot be reassured that preeclampsia will not supervene because almost one in four women with normal results also developed superimposed preeclampsia. In summary, we think that inhibin-A levels are not currently of great utility in clinical management. This does not, however, preempt the use of maternal serum inhibin-A as a marker for the development of preeclampsia in experimental intervention protocols.
1. Wolfe CDA, Patel SP, Linton EA, Campbell EA, Anderson J, Dornhorst A, et al. Plasma corticotrophin-releasing factor (CRF) in abnormal pregnancy. Br J Obstet Gynaecol 1988; 95:1003–6.
2. Muller F, Savey L, Le Fiblec B, Bussieres L, Ndayizamba G, Colau JC, et al. Maternal serum human gonadotropin level at fifteen weeks is a predictor for preeclampsia. Am J Obstet Gynecol 1996;175:37–40.
3. Muttukrishna S, North RA, Morris J, Schellenberg JC, Taylor RS, Asselin J, et al. Serum inhibin A and activin A are elevated prior to the onset of preeclampsia. Hum Reprod 2000;15:1640–5.
4. Cuckle H, Sehmi I, Jones R. Maternal serum inhibin A can predict preeclampsia. Br J Obstet Gynaecol 1998;105:1101–3.
5. Sebire NJ, Roberts L, Noble P, Wallace E, Nicolaides KH. Raised maternal serum inhibin A concentration at 10 to 14 weeks of gestation is associated with preeclampsia. Br J Obstet Gynaecol 2000;107:795–7.
6. Lambert-Messerlian GM, Silver HM, Petraglia F, Luisi S, Pezzani I, Maybruck WM, et al. Second trimester levels of maternal serum human chorionic gonadotropin and inhibin A as predictors of preeclampsia in the third trimester of pregnancy. J Soc Gynecol Investig 2000;7:170–4.
7. Aquilina J, Barnett A, Thompson O, Harrington K. Second trimester maternal serum inhibin A concentration as an early marker for preeclampsia. Am J Obstet Gynecol 1999; 181:131–6.
8. Grobman WA, Wang EY. Serum levels of activin A and inhibin A and the subsequent development of preeclampsia. Obstet Gynecol 2000;96:390–4.
9. Zeeman GG, Alexander JM, McIntire DD, Leveno KJ. Inhibin-A levels and severity of hypertensive disorders due to pregnancy. Obstet Gynecol 2002;100:140–5.