HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Goetzl, L. Articles by Wapner, R. J. Search for Related Content PubMed PubMed Citation Articles by Goetzl, L. Articles by Wapner, R. J. Related Collections Epidemiology/public health Genetics and teratology Prenatal Diagnosis

Pregnancy-Associated Plasma Protein A, Free ß-hCG, Nuchal Translucency, and Risk of Pregnancy Loss

From Baylor College of Medicine, Houston, Texas; NTD Laboratories, Huntington Station, New York; The Fetal Diagnostic Center, Evanston Hospital, Evanston, Illinois; Northwestern University Medical School, Chicago, Illinois; George Washington Biostatistics Center, Washington, DC; Cedars Sinai Medical Center, Los Angeles, California; Yale University, New Haven, Connecticut; and Drexel University College of Medicine, Philadelphia, Pennsylvania.

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION Appendix 1 REFERENCES

 
OBJECTIVE: To estimate the likelihood of clinical early and late pregnancy loss as a function of first-trimester maternal serum analytes and fetal nuchal translucency measurements.

METHODS: Study subjects were recruited for a National Institute of Child Health and Human Development–sponsored multicenter cohort study initially designed to study the detection of Down syndrome during the first trimester of pregnancy. The cohort consisted of women who had a live fetus between 10 and 14 weeks of gestation and had no significant vaginal bleeding. Women with prior fetal trisomy (T21/18) and those with structural or chromosomal abnormalities in the index pregnancy were excluded. First-trimester screening consisted of pregnancy-associated plasma protein A (PAPP-A), free ß-hCG, and nuchal translucency. Pregnancy loss rates in women with various levels of PAPP-A, free ß-hCG, or nuchal translucency (less than 1st, less than 5th, more than 95th, and more than 99th percentile) were compared with losses in women with normal values (5th to 95th percentile).

RESULTS: The mean gestational age at screening of 7,932 women meeting study criteria was 12.1 weeks. Loss rates were only 0.36% at less than 20 weeks after normal free ß-hCG, PAPP-A, and nuchal translucency. Conversely, low levels of PAPP-A and free ß-hCG as well as increased nuchal translucency were individually associated with increased early loss. These associations persisted after controlling for maternal age and race using logistic regression analysis.

CONCLUSION: Normal values of PAPP-A, free ß-hCG, and nuchal translucency are associated with a very low risk of pregnancy loss at less than 20 weeks.

LEVEL OF EVIDENCE: II-2

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION Appendix 1 REFERENCES

 
The cohort used for the current study was a subset of that recruited for the First Trimester Maternal Serum Biochemistry and Ultrasound Fetal Nuchal Translucency Screening Study (BUN), a National Institute of Child Health and Human Development multicenter study designed to evaluate the detection of Down syndrome using noninvasive methods during the first trimester of pregnancy. The study was reviewed and approved by the institutional review boards at the various institutions. Patients were enrolled between May 1998 and December 2000. Subjects recruited from 12 U.S. and Canadian centers had viable, singleton pregnancies between 10–4/7 and 13–6/7 weeks of gestation by crown rump length. Women were not enrolled if they had significant vaginal bleeding or known type 1 diabetes. The details of this study group and outcome ascertainment are explained elsewhere.² The initial study cohort consisted of 8,216 women; for the current analysis, an additional 284 women were excluded (155 with either structural or chromosomal abnormalities, 72 with trisomy 21 or 18 in the current pregnancy, 47 without delivery information, and 10 with terminations with no known abnormality). Maternal serum samples were obtained either before or on the same day as sonographic evaluation and before any invasive procedures for prenatal diagnosis.

The primary outcomes of the current analysis were rates of pregnancy loss at less than 20 weeks and 20 weeks or more of gestation. Pregnancy loss at 20 weeks of gestation or more included losses caused by cervical incompetence, delivery of a previable infant, intrauterine fetal demise, and stillbirth. Pregnancy loss at 20 weeks of gestation or more did not include liveborn infants experiencing subsequent neonatal demise. The cytogenetic normalcy of pregnancy losses was verified whenever possible. Baseline maternal variables included age, weight, race, and smoking status during the pregnancy. Serum analytes were analyzed based on multiples of the median (MoM) adjusted for gestational age and maternal weight. Nuchal translucency was analyzed based on MoM adjusted for gestational age. Rates of pregnancy loss were further stratified by levels of PAPP-A and free ß-hCG or by nuchal translucency measurement. Extreme values were defined as less than 5th, less than 1st, more than 95th, and more than 99th percentiles based on the distribution within the study cohort; the reference group was defined as women having normal values (5th to 95th percentile) for each factor.

Statistical comparisons of continuous variables were performed by using the Kruskal–Wallis test; rates were compared by using the Fisher exact test. A nominal P value of < .05 was considered statistically significant. A separate standard logistic regression using pregnancy loss as a dichotomous dependent variable was constructed for each abnormal analyte threshold to examine the association between abnormal first-trimester parameters and pregnancy loss after controlling for potential confounders. The potential confounding factors maternal age (in years) and race (African American, Hispanic, or Caucasian/other) were included in the logistic regression model. Maternal smoking status could not be included in the model because very few women smoked. Invasive diagnostic procedures (amniocentesis or chorionic villus sampling) were excluded because of high levels of missing data (n = 617). Maternal weight was not included in the models because maternal serum analytes are weight adjusted; thus, the presentation of unadjusted MoM would not be readily interpretable to clinicians. Secondary models, including maternal weight (pounds), were used to examine whether weight-adjusted MoM adequately addressed the effect of maternal weight on pregnancy loss at less than 20 weeks of gestation. Adjusted odds ratios with 95% confidence intervals are presented.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION Appendix 1 REFERENCES

 
Of the 8,216 patients initially described,² 7,932 met criteria for the study. The mean gestational age at screening of these 7,932 patients was 84.6 ± 6.2 days. Maternal and obstetric characteristics are presented in Table 1, stratified by pregnancy outcome. The overall rate of pregnancy loss was 0.53% at less than 20 weeks of gestation (n = 42) and 0.42% at 20 weeks of gestation or more (n = 33). Univariate comparisons suggest potential associations between early pregnancy loss and these maternal factors: older age, heavier weight, and maternal race. Although rates of smoking are higher in women who later experience a loss at 20 weeks of gestation or greater, this difference was not statistically significant. The rates of maternal smoking were too low to assess a relationship between smoking and early pregnancy loss. Neither early nor late pregnancy loss rates were higher in women who underwent invasive procedures for prenatal diagnosis.

Rates of pregnancy loss at less than 20 weeks of gestation as a function of serum analytes and nuchal translucency values are presented in Table 2. Levels less than the 5th percentile of either analyte, PAPP-A or free ß-hCG, were associated with a significantly increased rate of pregnancy loss at less than 20 weeks of gestation. The risk of pregnancy loss increased as depressed serum levels of either analyte became more extreme. In contrast, elevated levels of serum analytes were not associated with either an increased or decreased risk of early loss. Nuchal thickness greater than the 95th and 99th percentiles also was associated with a graded increase in the rate of early pregnancy loss. In contrast to results at less than 20 weeks of gestation, extreme values of either serum analytes or nuchal translucency (in either direction) were not associated with an increased risk of pregnancy loss at 20 weeks of gestation or more (Table 3). When PAPP-A, free ß-hCG, and nuchal translucency are all normal, loss rates were only 0.36% at less than 20 weeks of gestation and 0.48% at 20 weeks of gestation or more (Table 4).

Logistic regression analysis revealed that maternal age, race, and weight were all independent predictors of pregnancy loss at less than 20 weeks of gestation; however, only maternal age and race were included in the final regression models because serum analytes are already weight-adjusted; thus, the presentation of unadjusted MoM would not be readily interpretable. That maternal weight remained significant suggests that weight adjustments made to analyte MoM may not be sufficient for the optimal prediction of pregnancy loss at less than 20 weeks of gestation. Unadjusted odds ratios are presented for pregnancy loss at 20 weeks of gestation or greater because low rates of pregnancy loss at this gestation resulted in a lack of convergence. When all 3 first-trimester screening values are normal (5th to 95th percentile) the odds ratio for loss at less than 20 weeks of gestation is 0.4 (0.2–0.6) after adjusting for race and maternal age.

Maternal age was further examined to determine whether the increased pregnancy loss at less than 20 weeks of gestation was gradual or rose sharply at any point. We observed a noticeable increase in the rate of pregnancy loss at less than 20 weeks of gestation after a maternal age of more than 35 years (Fig. 1). In contrast, rates of pregnancy loss at 20 weeks of gestation or greater failed to increase with increasing maternal age category (Fig. 2). The effect of increasing maternal weight category on the rate of pregnancy loss at less than 20 weeks of gestation also was examined separately. Our analysis suggested that rates of loss may increase sharply when maternal weight exceeds approximately 185 pounds (Fig. 3). Maternal weight was not a predictor of pregnancy loss at 20 weeks of gestation or greater.

View larger version (21K): [in this window] [in a new window]   Fig. 1. Relationship between maternal age and pregnancy loss at less than 20 weeks of gestation. Goetzl. First-Trimester Markers and Pregnancy Loss. Obstet Gynecol 2004.

View larger version (20K): [in this window] [in a new window]   Fig. 2. Relationship between maternal age and pregnancy loss at 20 weeks of gestation or more. Goetzl. First-Trimester Markers and Pregnancy Loss. Obstet Gynecol 2004.

View larger version (23K): [in this window] [in a new window]   Fig. 3. Relationship between maternal weight and pregnancy loss before 20 weeks of gestation. Goetzl. First-Trimester Markers and Pregnancy Loss. Obstet Gynecol 2004.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION Appendix 1 REFERENCES

 
The message of greatest clinical utility is that when PAPP-A, free ß-hCG, and nuchal translucency are in the normal range, pregnancy loss rates are only 0.36% at less than 20 weeks of gestation and 0.48% at 20 weeks or more. By contrast, pregnancy loss at less than 20 weeks of gestation was increased with either low PAPP-A or free ß-hCG or elevated fetal nuchal translucency. The association becomes more marked with increasingly abnormal values. This association suggests that abnormally low values of PAPP-A or free ß-hCG may be early markers of abnormal implantation or other trophoblast dysfunction. However, even with these extreme values, the likelihood of pregnancy loss remains relatively low, not exceeding 4.1%. Elevated serum analytes and decreased nuchal translucency were not associated with an increased rate of pregnancy loss at less than 20 weeks of gestation.

This analysis of a large cohort clarifies previous uncertainties. Prior studies have reported that low first-trimester maternal serum PAPP-A or free ß-hCG may^8,9 or may not¹⁰ predict subsequent fetal loss in symptomatic patients with a viable fetus on sonographic evaluation. Ong et al¹¹ described a significantly lower median maternal serum PAPP-A between 10 and 14 weeks of gestation in women with subsequent pregnancy loss in a large prospective cohort. In contrast to our findings and those of others, their study did not show pregnancy loss to be associated with significantly lower ß-hCG levels. In addition, Ong et al¹¹ did not report the rate of subsequent pregnancy loss in women with abnormal serum analyte levels, limiting clinical application of these findings. Subsequently, Yaron et al^12,13 confirmed the association between low PAPP-A and free ß-hCG and early pregnancy loss in a cohort of 1,622 subjects. However, this analysis relied on specific multiple of the median cutoffs, determined by receiver operator curve, potentially limiting the application of their findings to other populations. Furthermore, the cohort contained only a single late intrauterine fetal demise, precluding the analysis of first-trimester parameters and fetal loss beyond 20 weeks of gestation that was possible in our data set. Smith et al¹⁴ reported an association between reduced first-trimester PAPP-A and stillbirth; however, data on ß-hCG and nuchal translucency were not available. Most recently, Wald et al¹ performed a nested case–control study of fetal loss from the Serum, Urine and Ultrasound Screening Study cohort. The sample size for this analysis was 1,452 of the original sample of 47,053 pregnancies ascertained in 24 United Kingdom and 1 Austrian center. Despite large sample size, results from Serum, Urine and Ultrasound Screening Study are difficult to interpret. One reason is that early loss and stillbirth are combined, precluding determination of the specific timing of the risk of pregnancy loss. Second, pregnancies with chromosomal or structural abnormalities were not excluded. Third, odds ratios for fetal loss were presented without 95% confidence intervals, hampering the interpretation of which markers are significantly associated with fetal loss.

The strengths of our study include its large sample size, our attention to confounders, the exclusion of patients with threatened miscarriage, and the exclusion of pregnancies with known chromosomal abnormalities. This has yielded an asymptomatic cohort with documented pregnancy viability that is ostensibly chromosomally normal. Our analysis is based on logically defined, a priori definitions for extreme values of first-trimester serum analytes and nuchal thickness. We present percentile-based results that can be extrapolated to any populations in which the normal distribution of serum analytes is known. In addition, we present multiple of the median cutoffs that are readily interpretable by clinicians. Finally, we present data on not only serum analytes but also nuchal thickness.

Despite the strength of our data set and the associations we demonstrate, limitations exist. First, we were not able to obtain detailed information on history of prior pregnancy loss, an additional risk factor that would have been important to include in our multivariate analysis. In women with a history of recurrent loss, the effect of normal serum analytes and nuchal translucency on risk of subsequent loss is not known, but some evidence suggests that lower serum total hCG may have some predictive value concerning prognosis.¹⁵ Also unavailable was information regarding maternal medical characteristics that could directly influence pregnancy outcome, such as diabetes mellitus or thrombophilia. Undiagnosed pregestational diabetes may explain at least part of the association between maternal obesity and increased rates of pregnancy loss at less than 20 weeks. Finally, karyotypic analysis of pregnancy losses was aggressively sought, but tissue could be obtained for analysis in only 62.7% of cases. Although unrecognized chromosomal abnormalities presumably account in part for our finding that advancing maternal age contributes to the risk of early pregnancy loss, losses with and without karotypic analysis were entirely comparable with regard to maternal age (data not shown).

Our results provide data that are useful for counseling pregnant patients undergoing first-trimester screening for Down syndrome. Those with normal values of serum analytes can be reassured that their risk of subsequent early pregnancy loss is exceedingly low (0.36% even at a mean maternal age of 34.4 ± 4.6 years), several-fold less that the 3% traditionally offered on the basis of a sonographically viable pregnancy.¹⁶ The lower rates of pregnancy loss in our cohort may be attributable not only on the basis of later gestation (mean 12.1 weeks), but the exclusion of patients with insulin-dependent diabetes. In patients having a single extreme value for any of the 3 screening values, we provide easily interpreted cutoffs to define the patient's increased risk for pregnancy loss, both in percentiles and MoM. Patients with elevated levels of serum analytes or small nuchal translucency can be counseled that their risk of subsequent pregnancy loss is not increased. Rates of pregnancy loss at less than 20 weeks of gestation do increase with extremely low levels of serum analytes or elevated nuchal translucency, yet absolute rates of pregnancy loss remain low, not exceeding 4.1%. Therefore, altered perinatal surveillance after a single extreme value may not necessarily be warranted.

	Appendix 1

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION Appendix 1 REFERENCES

 
The following individuals and centers participated in the multi-center study and comprise the BUN study group: J.L.S., Baylor College of Medicine, Houston, TX; D.K., T. Hallahan, NTD Laboratories, Huntington Station, NY: R.K.S., S. MacGregor, The Fetal Diagnostic Center, Evanston Hospital, Evanston, IL; E.P., R. Sabbagha, Northwestern University Medical School, Chicago, IL; L.D.P., R. Snijders, N. Greene, Cedars Sinai Medical Center, Los Angeles, CA; M.J.M., R. Bahado-Singh, Yale University, New Haven, CT; R.J.W., L. Jackson, Drexel University College of Medicine, Philadelphia, PA; E. Thom, J.M.Z., George Washington Biostatistics Center, Washington, DC; K. Filkins, UCLA Center for Health Sciences, Los Angeles, CA; A. Johnson, Wayne State University, Detroit, MI; W. A. Hogge, L. Hill, Magee Women's Hospital, Pittsburgh, PA; R. D. Wilson, A. Gagnon, British Columbia Women's Hospital, Vancouver, BC, Canada; P. Mohide, McMaster University Medical Centre, Hamilton, Ont., Canada; D. Hershey, Prenatal Diagnosis Center of Northern California, Sacramento, CA.

	Footnotes

 
Received December 9, 2003. Received in revised form March 3, 2004. Accepted April 4, 2004.

* For members of the BUN Study Group, see the Appendix.

Supported by R01 HD31991 and HD32109 from the National Institute of Child Health and Human Development.

Reprints are not available. Address correspondence to: Laura Goetzl, 6550 Fannin Street, Suite 901, Houston, TX 77030; e-mail: lgoetzl{at}bcm.tmc.edu.

10.1097/01.AOG.0000129969.78308.4f

	REFERENCES

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION Appendix 1 REFERENCES

 
1. Wald NJ, Rodeck C, Hackshaw AK, Walters J, Chitty L, Mackinson AM. First and second trimester antenatal screening for Down's syndrome: the results of the Serum, Urine and Ultrasound Screening Study (SURUSS). J Med Screen 2003;10:56–104.[Medline]

2. Wapner R, Thom E, Simpson JL, Pergament E, Silver R, Filkins K, et al. First-trimester screening for trisomies 21 and 18. N Engl J Med 2003;349:1405–13.[Abstract/Free Full Text]

3. Spencer K, Spencer CE, Power M, Dawson C, Nicolaides KH. Screening for chromosomal abnormalities in the first trimester using ultrasound and maternal serum biochemistry in a one-stop clinic: a review of three years prospective experience. BJOG 2003;110:281–6.[Medline]

4. Haddow JE, Palomaki GE, Knight GJ, Williams J, Miller WA. Johnson A. Screening of maternal serum for fetal Down's syndrome in the first trimester. N Engl J Med 1998;338:955–61.[Abstract/Free Full Text]

5. Crossley JA, Aitken DA, Cameron AD, McBride E, Connor JM. Combined ultrasound and biochemical screening for Down's syndrome in the first trimester: a Scottish multicentre study. BJOG 2002;109:667–76.[Medline]

6. Schuchter K, Hafner E, Stangl G, Metzenbauer M, Hofinger D, Philipp K. The first trimester "combined test" for the detection of Down syndrome pregnancies in 4939 unselected pregnancies. Prenat Diag 2002;22:211–5.[Medline]

7. von Kaisenberg CS, Gasiorek-Wiens A, Bielicki M, Bahlmann F, Meyberg H, Kossakiewicz A, et al, for the German Speaking Down Syndrome Screening Group. Screening for trisomy 21 by maternal age, fetal nuchal translucency and maternal serum biochemistry at 11–14 weeks: a German multicenter study. J Matern Fetal Neonatal Med 2002;12:89–94.[Medline]

8. Westergaard JG, Sinosich MJ, Bugge M, Madsen LT, Teisner B, Grudzinskas JG. Pregnancy-associated plasma protein A in the prediction of early pregnancy failure. Am J Obstet Gynecol 1983;145:67–9.[Medline]

9. Christodoulou CN, Zonas C, Loukaides T, Maniatis A, Giannikos L, Giannakopoulos C, et al. Low beta-hCG is associated with poor prognosis in association with an embryo with positive cardiac activity. Ultrasound Obstet Gynecol 1995;5:267–70.[Medline]

10. Ruge S, Pedersen JF, Sorensen S, Lange AP. Can pregnancy-associated plasma protein A (PAPP-A) predict the outcome of pregnancy in women with threatened abortion and confirmed fetal viability? Acta Obstet Gynecol Scand 1990;69:589–95.[Medline]

11. Ong CY, Liao AW, Spencer K, Munim S, Nicolaides KH. First trimester maternal serum free beta human chorionic gonadotrophin and pregnancy associated plasma protein A as predictors of pregnancy complications. BJOG 2000;107:1265–70.[Medline]

12. Yaron Y, Heifetz S, Ochshorn Y, Lehavi O, Orr-Urtreger A. Decreased first trimester PAPP-A is a predictor of adverse pregnancy outcome. Prenat Diag 2002;22:778–82.[Medline]

13. Yaron Y, Ochshorn Y, Heifetz S, Lehavi O, Sapir Y, Orr-Urtreger A. First trimester maternal serum free human chorionic gonadotropin as a predictor of adverse pregnancy outcome. Fetal Diag Ther 2002;17:352–6.[Medline]

14. Smith GC, Stenhouse EJ, Crossley JA, Aitken DA, Cameron AD, Connor JM. Early pregnancy levels of pregnancy-associated plasma protein A and the risk of intrauterine growth restriction, premature birth, preeclampsia, and stillbirth. J Clin Endocrin Metab 2002;87:1762–7.[Abstract/Free Full Text]

15. Li TC, Spring PG, Bygrave C, Laird SM, Serle E, Spuijbroek M, Adekanmi O. The value of biochemical and ultrasound measurements in predicting pregnancy outcome in women with a history of recurrent miscarriage. Hum Reprod 1998;13:3525–9.[Abstract/Free Full Text]

16. Simpson JL, Mills JL, Holmes LB, Ober CL, Aarons J, Jovanovic L, et al. Low fetal loss rates after ultrasound-proved viability in early pregnancy. JAMA 1987;258:2555–7.[Abstract]

This article has been cited by other articles:

				Normal Serum Analytes? Low Pregnancy Loss Rate Journal Watch Women's Health, August 24, 2004; 2004(824): 6 - 6. [Full Text]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Goetzl, L. Articles by Wapner, R. J. Search for Related Content PubMed PubMed Citation Articles by Goetzl, L. Articles by Wapner, R. J. Related Collections Epidemiology/public health Genetics and teratology Prenatal Diagnosis