Objective
To assess maternal-fetal outcomes in pregnancies associated with persistently elevated second-trimester maternal serum alpha-fetoprotein.
Study Design
A retrospective cohort study in 658 patients with maternal serum alpha-fetoprotein ≥2.5 multiple of median, performed at routine Down syndrome screening. Maternal serum alpha-fetoprotein was assayed a second time in 341 of them. Outcomes were recorded in all cases.
Results
The group with unexplained maternal serum alpha-fetoprotein persistently ≥2.5 multiple of median was associated with more pregnancy complications 37 of 92 (40.2%) as fetal death, preeclampsia, intrauterine growth restriction, and congenital nephrotic syndrome, compared with the group with maternal serum alpha-fetoprotein that returned to a normal level 37 of 226 (16.4%) ( P < .001).
Conclusion
When maternal serum alpha-fetoprotein returns to a normal level on a second assay, the risk of adverse outcome significantly decreases, but these pregnancies are still at risk of complications and therefore need close surveillance. Repeat maternal serum alpha-fetoprotein assay allows identification of patients who should be offered amniocentesis to evaluate the risk of nephrotic syndrome and epidermolysis bullosa. Alpha-fetoprotein should be monitored in pregnancies associated with unexplained high maternal serum alpha-fetoprotein. A management strategy based on ultrasound examination, second maternal serum alpha-fetoprotein assay and amniocentesis is proposed to improve prenatal counseling and management of such pregnancies. However, a prospective study remains necessary to evaluate it.
Alpha-fetoprotein (AFP) is a glycoprotein produced during pregnancy by the fetal yolk sac and the fetal liver, which crosses the placental barrier, the fetal membrane and the decidua to reach the maternal circulation. It is also produced by the choroid plexus and is released into the cerebrospinal fluid. First used in screening for neural tube defects, maternal serum AFP assay was combined with human chorionic gonadotropin hormone (hCG) in second-trimester biochemical screening for chromosomal abnormalities. An increased level of maternal serum AFP (MSAFP) was subsequently recognized as being associated with an increased risk of pregnancy complications as intrauterine fetal death (IUFD), intrauterine growth restriction (IUGR), preterm birth, and preeclampsia. An elevation of MSAFP is also seen in several fetal anomalies such as ventral wall defects, and rare congenital disorders, including among others nephrotic syndrome and epidermolysis bullosa. In France, the national consensus threshold used is 2.5 multiples of the median (MoM) and concerns 1% of pregnancies. Besides morphologic anomalies visible on ultrasound examination, there are many cases for which high MSAFP is not directly explicable. This can be worrying for the patient and physician. To our knowledge there are no explicit guidelines for pregnancies with abnormal MSAFP elevation. Some clinicians assay MSAFP again later in pregnancy, but this approach has never been evaluated. Our aim was to study the relevance of a second MSAFP assay and to propose a management strategy based on it.
Materials and Methods
A retrospective cohort study was conducted over the period 2004-2008 in our biochemical laboratory, which is accredited for trisomy 21 screening. Second-trimester (14-18 weeks) maternal serum markers used in screening for Down syndrome in singleton pregnancies, including hCGb and AFP, were assayed (Autodelfia, dual kit; PerkinElmer, Turku, Finland). Maternal sera from an unselected population were routinely sent from 8 hospitals. Gestational age was determined by crown-rump length at first-trimester ultrasound examination (expressed in weeks and days). Patients with a high MSAFP (defined as AFP ≥2.5 multiple of median [MoM]) were offered a second AFP assay as part of the diagnostic work-up. This second assay was performed at least 18 days later (which represents approximately 3 times the half-life of AFP in maternal serum ). Reasons for not perform a second assay were as follows: (1) diagnosis of a severe fetal malformation at ultrasound examination or an IUFD before 18 days, and cases of vanishing twin or selective fetal reduction because of the artificial increase in AFP. (2) Physician or maternal failure to adhere to the procedure. Patient management consisted of standard ultrasound surveillance and, if necessary, amniocentesis for fetal karyotyping, amniotic fluid AFP assay (Thermo Fisher Scientific, Hennigsdorf, Germany), and cholinesterase electrophoresis (AChE). No particular antenatal testing protocol for vascular follow-up was undertaken and patients were managed in their center of origin by the physician responsible. Consent for amniocentesis was obtained from all patients. A second MSAFP assay was part of the routine diagnostic work-up. For these reasons this study was exempt from ethical review board approval. Patients who underwent a second MSAFP assay were classified into 2 groups: (1) group with MSAFP remaining ≥2.5 MoM and (2) group with MSAFP returning to a normal level. Clinical data, ultrasonographic reports, laboratory findings, outcome, and final diagnosis were recorded. Pregnancy complications were classified in 5 groups: severe fetal anomalies, IUFD, IUGR <3rd percentile, preeclampsia, and spontaneous premature birth. Severe fetal anomaly was defined as a severe morphologic anomaly or syndrome, requiring possible extensive hospital care or to which French law about medical termination of pregnancy may apply. Only spontaneous preterm deliveries before 34 weeks of gestation were taken into account.
Pregnancy outcome was documented in all cases. In accordance with French law, termination of pregnancy was possible at the parents’ request in cases of severe fetal anomaly and poor prognosis.
Results are presented as median, interquartile range (IQR), range, or proportions accordingly. The χ 2 test, or the Fisher exact test (numbers <5) were used for comparison of groups. A probability value of .05 or less was considered statistically significant. Statistical analyses were performed using R software, version 2.0.0 ( www.r-project.org , Foundation for Statistical Computing, Vienna, Austria).
Results
Between 2004 and 2008, 658 second-trimester maternal serum marker samples with an MSAFP ≥2.5 MoM were retrieved. Cases of vanishing twin and selective fetal reduction were discarded (n = 44). In 341 of the remaining 614 cases, a second MSAFP assay was performed ( Figure 1 ).
Overall population
In these 614 cases of elevated MSAFP, median gestational age at sampling was 16 +1 weeks (IQR, 15 +3 −17 +5 ). Median maternal age was 29 years (IQR, 25−34). Final diagnosis and pregnancy outcomes are presented in Table 1 . No complication was observed in 372 of 614 cases (60.6%). Complications were observed in 242 of 614 pregnancies (39.4%). We observed 65 (10.6%) severe fetal anomalies, 60 (9.8%) IUFDs (15-30 weeks, median, 17 weeks), 36 (5.8%) IUGRs, 24 (3.9%) preeclampsias, and 57 (9.3%) spontaneous premature deliveries. The 65 severe fetal anomalies are detailed in Table 2 . Termination of pregnancy was performed at the parents’ request in 45 of 65 cases (69.2%). The main fetal anomalies involved were fetal cutaneous defect such as neural tube defect (n = 25), gastroschisis (n = 4) and cervical teratoma (n = 1), fetal glomerular defects such as nephrotic syndrome (n = 5), and placental anomalies such as triploidy (n = 3). There were 22 cases of undefined multiple malformation syndromes. Of the 65 fetal anomalies, 60 morphologic anomalies were visible at ultrasound examination, and only the 5 cases of nephrotic syndrome were undetectable at ultrasound examination. These 5 cases of nephrotic syndrome were confirmed by postnatal genetic testing. Amniocentesis was performed in 3 of these 5 cases and amniotic fluid AFP > 7 MoM associated with normal AChE electrophoresis findings was highly suggestive of nephrotic syndrome. The 2 other cases of nephrotic syndrome were diagnosed at birth.
| Outcomes (n = 614) | n (%) |
|---|---|
| Pregnancies without complication | 372 (60.6) |
| Complicated pregnancies | 242 (39.4) |
| Severe fetal anomalies | 65 (10.6) |
| IUFD | 60 (9.8) |
| IUGR | 36 (5.8) |
| Preeclampsia | 24 (3.9) |
| Spontaneous preterm delivery <34 wks | 57 (9.3) |
| Anomalies | n |
|---|---|
| Neural tube defect | 25 |
| Multiple malformation syndrome | 22 |
| Nephrotic syndrome | 5 a |
| Gastroschisis | 4 |
| Cerebral anomalies | 3 |
| Triploidy | 3 |
| Tuberous sclerosis complex | 1 |
| Cervical teratoma | 1 |
| Bilateral renal dysplasia | 1 |
Population without a second AFP assay
A second MSAFP assay was not performed in 273 pregnancies ( Figure 1 ). A complication occurred in 145 of 273 of these pregnancies (53.1%). We observed 46 (16.8%) IUFDs, 38 (13.9%) severe fetal anomalies, 27 (9.9%) spontaneous preterm deliveries, 18 (6.6%) IUGRs, and 16 (5.9%) preeclampsia. In 21 cases termination of pregnancy because of severe fetal anomaly or severe IUGR was performed less than 18 days after MSAFP assay. In 31 of 39 cases, the IUFD occurred less than 18 days after MSAFP assay.
Population with a second AFP assay
A second MSAFP assay was performed in 341 pregnancies ( Figure 1 ). Median maternal age was 29 years (IQR, 25−33), with no significant difference between the group with remained elevated MSAFP (median, 29 years; IQR, 25−33) and the group with decreased MSAFP (median, 29 years; IQR, 25−33). The median interval between the first and second MSAFP assays was 35 days (IQR, 26−55). Pregnancy complications were observed in 97 of 341 (28.4%) in the group with a second MSAFP assay. In 23 of these 341 pregnancies, a diagnosis was possible by ultrasound examination because of morphologic abnormalities. In the remaining 318 pregnancies with MSAFP ≥2.5 MoM there was no clear explanation ( Figure 1 ). No statistical difference between the 2 groups was observed for the first MSAFP value (2.85 MoM when AFP returned to normal vs 3.11 MoM when AFP remained elevated).
When AFP returned to normal, pregnancies were complicated in 37 of 226 cases (16.4%), and when AFP remained high, pregnancies were complicated in 37 of 92 cases (40.2%) ( P < .001). Among these 37 complications, there were 4 nephrotic syndromes, of which 2 had been prenatally suspected because of amniotic fluid AFP > 7 MoM.
The relationship between pregnancy outcomes and MSAFP evolution after exclusion of cases with a clear ultrasound explanation is presented in Table 3 . Pregnancies with a MSAFP remaining ≥2.5 MoM had significantly higher rates of IUFD (21-31 weeks; median, 26 weeks) ( P = .005), IUGR ( P = .01), and preeclampsia ( P = .047) and would have allowed identification of cases of fetal nephrotic syndrome ( P = .006). The rate of spontaneous preterm delivery was high in the 2 groups (9%), but did not differ significantly between the groups ( P > .99).
| Variable | AFP remained high n = 92 | AFP returned to normal n = 226 | P value |
|---|---|---|---|
| Initial MSAFP (MoM), median (range) | 3.11 (2.5–48.4) | 2.85 (2.5–38.7) | |
| Repeat MSAFP (MoM), median (range) | 3 (2.50–43.7) | 1.6 (0.12–2.47) | |
| Outcomes | |||
| Pregnancies without complication | 55 (59.8%) | 189 (83.6%) | < .001 |
| Complicated pregnancies | 37 (40.2%) | 37 (16.4%) | < .001 |
| IUFD | 9 (9.8%) | 5 (2.1%) | .005 |
| IUGR | 10 (10.9%) | 8 (3.5%) | .01 |
| Preeclampsia | 5 (5.4%) | 3 (1.3%) | .047 |
| Spontaneous preterm delivery <34 wks | 9 (9.8%) | 21 (9.3%) | > .99 |
| Congenital nephrotic syndrome | 4 (4.3%) | 0 (0%) | .006 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
