Original Paper Fetal Diagn Ther 2015;37:206–210 DOI: 10.1159/000363654
Received: March 3, 2014 Accepted after revision: May 10, 2014 Published online: July 12, 2014
Is MSAFP Still a Useful Test for Detecting Open Neural Tube Defects and Ventral Wall Defects in the Era of First-Trimester and Early Second-Trimester Fetal Anatomical Ultrasounds? Ashley S. Roman a Simi Gupta b, c Nathan S. Fox a–c Daniel Saltzman a–c Chad K. Klauser a–c Andrei Rebarber a–c a Department of Obstetrics and Gynecology, New York University School of Medicine, b Carnegie Imaging for Women, PLLC, and c Department of Obstetrics, Gynecology, and Reproductive Science, Mount Sinai School of Medicine, New York, N.Y., USA
Abstract Introduction: To evaluate whether maternal serum α-fetoprotein (MSAFP) improves the detection rate for open neural tube defects (ONTDs) and ventral wall defects (VWD) in patients undergoing first-trimester and early second-trimester fetal anatomical survey. Material and Methods: A cohort of women undergoing screening between 2005 and 2012 was identified. All patients were offered an ultrasound at between 11 weeks and 13 weeks and 6 days of gestational age for nuchal translucency/fetal anatomy followed by an early second-trimester ultrasound at between 15 weeks and 17 weeks and 6 days of gestational age for fetal anatomy and MSAFP screening. All cases of ONTD and VWD were identified via query of billing and reporting software. Sensitivity and specificity for detection of ONTD/VWD were calculated, and groups were compared using the Fisher exact test, with p < 0.05 as significance. Results: A total of 23,790 women met the criteria for inclusion. Overall, 15 cases of ONTD and 17 cases of VWD were identified; 100% of cases were diag-
© 2014 S. Karger AG, Basel 1015–3837/14/0373–0206$39.50/0 E-Mail
[email protected] www.karger.com/fdt
nosed by ultrasound prior to 18 weeks’ gestation; none were diagnosed via MSAFP screening (p < 0.001). First-trimester and early second-trimester ultrasound had 100% sensitivity and 100% specificity for diagnosing ONTD/VWD. Discussion: Ultrasound for fetal anatomy during the first and early second trimester detected 100% of ONTD/VWD in our population. MSAFP is not useful as a screening tool for ONTD and VWD in the setting of this ultrasound screening protocol. © 2014 S. Karger AG, Basel
Introduction
Neural tube defects (NTDs) are the second most common major congenital anomalies after congenital cardiac defects [1]. In the USA in 2006, the combined incidence of spina bifida and anencephaly, which represent a large proportion of NTDs, was about 0.3 per 1,000 births [2]. Maternal serum α-fetoprotein (MSAFP) testing in the second trimester was introduced in the 1970s and 1980s as a screening test for open NTDs (ONTDs) [3, 4]. Using
This study was submitted for Special Issue on Spina Bifida.
Simi Gupta 70 East 90th Street New York, NY 10128 (USA) E-Mail sgupta @ mfmnyc.com
Downloaded by: NYU Medical Center Library 216.165.126.139 - 5/7/2015 1:42:28 PM
Key Words Maternal serum α-fetoprotein · Ultrasound · Neural tube defect · Ventral wall defect
Utility of MSAFP Screening after Early Anatomy Ultrasound
Fetal Diagn Ther 2015;37:206–210 DOI: 10.1159/000363654
Material and Methods
207
Downloaded by: NYU Medical Center Library 216.165.126.139 - 5/7/2015 1:42:28 PM
This study was a retrospective study of all patients with singleton gestations undergoing ultrasound in a single ultrasound referral practice center between May 2005 and August 2011. IRB approval was obtained. All patients were offered 3 ultrasound screening examinations for aneuploidy and fetal malformations as
follows: (1) nuchal translucency ultrasound at between 11 weeks and 13 weeks and 6 days of gestational age, at which time an evaluation of the fetal anatomy was performed, (2) early second-trimester anatomical ultrasound at between 15 weeks and 17 weeks and 6 days of gestational age, after which the second half of the modified sequential screen was sent (MSAFP, estriol, inhibin, free β-HCG), or MSAFP alone was sent if the patient did not desire aneuploidy screening or had a prior chorionic villus sampling and (3) detailed anatomical ultrasound at 20–22 weeks’ gestation. Our ultrasound units function with established protocols which describe expectations for fetal evaluation at each of these studies. All ultrasound examinations were performed using GE Voluson E8 or Medison Accuvix machines. All ultrasound examinations were performed by 1 of 18 sonographers and interpreted under the direct supervision of maternal-fetal medicine attendings. All sonographers were registered diagnostic medical sonographers certified for nuchal translucency and nasal bone assessment through the Nuchal Translucency Quality Review Program. All ultrasound examinations were performed using transabdominal imaging. Transvaginal ultrasound was used during the first trimester only if there was a suspicion for structural malformation based on transabdominal imaging or if the nuchal translucency could not be measured appropriately using transabdominal imaging. The firsttrimester evaluation of the fetal anatomy included sonographic assessment of fetal structures according to a pre-established unit protocol consistent with recent ISUOG (International Society of Ultrasound in Obstetrics and Gynecology) guidelines, including a transverse image of the head documenting the falx, choroid plexus, spine, and fetal cord insertion [13]. The second-trimester anatomical scan was performed according to published AIUM (American Institute of Ultrasound in Medicine) guidelines and included the lateral ventricles, cerebellum, cisterna magna, falx, cavum septum pellucidum, complete spine, and fetal cord insertion, with the addition of the corpus callosum at the 20- to 22-week scan [14]. Invasive diagnostic testing for aneuploidy with chorionic villus sampling or amniocentesis was recommended for any patient carrying a fetus with suspected omphalocele, myelomeningocele or encephalocele. Invasive testing was offered to patients with a suspected diagnosis of gastroschisis. All specimens for the modified sequential screen or MSAFP screening were sent to NTD Laboratories. Elevated MSAFP was defined as ≥2.0 MoM. First, we identified all patients who had an elevated MSAFP and queried their ultrasound reports to determine whether they were found to have a fetus with an ONTD or VWD. Next, all cases of ONTDs (spina bifida, encephalocele, exencephaly, and anencephaly) diagnosed by ultrasound in our unit during this time frame were identified using the ICD-9 code 655.03 and by searching our ultrasound reporting software (Sonultra) using the keywords ‘anencephaly’, ‘exencephaly’, ‘spina bifida’, ‘encephalocele’, and ‘myelomeningocele’. Cases of VWD were identified by a search of our ultrasound reporting software (Sonultra) using the key words ‘gastroschisis’ and ‘omphalocele’. This search included cases diagnosed at any point during the pregnancy, including at any of the 3 above-mentioned ultrasounds. Demographics, including age, race, parity and body mass index, were collected for each of the patients diagnosed with ONTD or VWD. The primary outcome was test sensitivity for detecting ONTD or VWD. The sensitivity of MSAFP performed within the described ultrasound screening protocol was compared with the sen-
a screen-positive cutoff of 2.5 multiples of the median (MoM), the screen positive rate is 5% or less with a sensitivity of 85% in singleton pregnancies [1]. A recent metaanalysis that included over 600,000 women demonstrated that MSAFP has a sensitivity of 75% and specificity of 97.7% for detecting ONTDs and a detection rate of 95% for anencephaly [5]. MSAFP was later found to be an effective screening test for fetuses with ventral wall defects (VWDs). Using a screen-positive cutoff of ≥2.0 MoM, MSAFP was found to have a sensitivity of 85–89% with a false positive rate of 4.5% for detecting all VWDs [6]. A more recent article showed that using MSAFP testing in the first trimester could also provide a 50% detection rate for ONTDs with a 10% false positive rate [7]. Other reports indicate that with improvements in ultrasound technology, ultrasound in the second trimester has a sensitivity of 97% or more and specificity of 100% in diagnosing ONTDs [8, 9]. Furthermore, in recent years, a number of experts have advocated first-trimester fetal anatomical scans at the time of the nuchal translucency ultrasound and/or early second-trimester (15–17 weeks) fetal anatomical ultrasound as a means to detect clinically significant fetal malformations at an earlier gestational age [10–12]. It has been reported that abnormalities of the abdominal wall and skeletal systems can be detected effectively by ultrasound performed between 11 and 14 weeks of gestation [12]. Despite these advances in ultrasound technology, the American College of Obstetricians and Gynecologists still advises that all women should be offered MSAFP screening for ONTDs unless they plan to undergo amniocentesis with determination of amniotic fluid AFP levels [1]. To date, no study has evaluated whether MSAFP retains its utility screening for ONTDs and VWDs in the setting of first-trimester and early second-trimester anatomical ultrasound examinations (on the basis of a PubMed search using keywords ‘MSAFP’ and ‘ultrasound’). The objective of our study was to determine whether MSAFP screening enhances the detection rate of ONTDs and VWDs in patients undergoing routine firsttrimester and early second-trimester anatomical ultrasounds.
an ONTD or VWD in our population (n = 36) Median maternal age, years Median parity Median BMI Prevalence of BMI ≥30 Prevalence of BMI ≥25 Conceived via IVF
33 (22–47) 0 (0–6) 22 (18–39) 4/36 (11.1%) 11/36 (30.6%) 5/36 (13.9%)
Median values (in parentheses are ranges). IVF = In vitro fertilization.
Table 2. Cases of ONTD or VWD and method of diagnosis in our
population between 2005 and 2012 Patient
Diagnosis
Method of detection (U/S or MSAFP)
Gestational age at diagnosis, weeks
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Encephalocele Encephalocele Exencephaly Exencephaly Exencephaly Exencephaly Exencephaly Exencephaly Exencephaly Exencephaly MMC – cervical MMC – lumbar-sacral MMC – lumbar-sacral MMC – sacral MMC – thoracic Gastroschisis Gastroschisis Gastroschisis Omphalocele Omphalocele Omphalocele Omphalocele Omphalocele Omphalocele Omphalocele Omphalocele Omphalocele Omphalocele Omphalocele Omphalocele Omphalocele, ectopia cordis Omphalocele/limb body wall complex
U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S U/S
12 12 4/7 11 1/7 11 5/7 12 12 12 1/7 12 4/7 12 6/7 13 5/7 11 1/7 16 3/7 17 16 3/7 12 13 3/7 13 4/7 13 5/7 11 11 6/7 12 12 12 2/7 12 2/7 12 3/7 12 4/7 12 6/7 12 6/7 13 3/7 16 3/7 13 6/7
U/S
12 1/7
32
Gestational age is expressed as weeks with days out of 7. Exencephaly: absence of the cranial vault and presence of brain tissue. U/S = Ultrasound; MMC = myelomeningocele.
208
Fetal Diagn Ther 2015;37:206–210 DOI: 10.1159/000363654
sitivity of first-trimester and early second-trimester ultrasound. Groups were compared using the Fisher exact test, with p < 0.05 defined as significance.
Results
During the study period, 23,790 first-trimester anatomy scans were performed (21,092 at the time of nuchal translucency ultrasound and 2,698 at the time of chorionic villus sampling) and 27,770 MSAFP specimens were sent. Of the 27,770 MSAFP tests sent, 341 (1.2%) returned with an elevated result (≥2.0 MoM). During the study period, 32 patients were carrying a fetus diagnosed with ONTD (n = 15) or VWD (n = 17). The incidence of ONTD in singleton pregnancies undergoing first-trimester and early second-trimester anatomy scans was 0.6 per 1,000 in our population. The incidence of VWD was 0.7 per 1,000. Patient demographics of women carrying a fetus diagnosed with ONTD or VWD are shown in table 1. In table 2, cases of ONTD and VWD are listed. All cases in our population were diagnosed antenatally by ultrasound at either the first-trimester or early second-trimester anatomical scan. The median gestational age at diagnosis was 12 weeks and 4 days (range 11–17 weeks). No cases were diagnosed via MSAFP screening, after ultrasound at or beyond 18 weeks’ gestation, or postnatally. Of the 15 cases of ONTD, 12 (63.2%) were diagnosed during the first trimester and the remaining cases were diagnosed in the early second trimester. All 3 cases that were missed during the first trimester were lumbar or sacral myelomeningoceles. Of the 17 VWD cases, 16 (94.1%) were diagnosed in the first trimester. The 1 case that was diagnosed during the second trimester had a first-trimester anatomical scan at which time the ventral wall was noted to be suboptimally visualized. The sensitivity and specificity of first-trimester or early second-trimester ultrasound for detecting ONTD/VWD were both 100% (p < 0.001). First-trimester ultrasound alone had a sensitivity of 80.0% (12/15), specificity of 100% (23,786/23,786), positive predictive value of 100% (12/12) and negative predictive value of 99.987% (23,787/23,790) for the diagnosis of ONTD (p < 0.001). For the diagnosis of encephalocele and myelomeningocele specifically, firsttrimester ultrasound alone had a sensitivity of 57% (4/7), specificity of 100% (23,786/23,786), positive predictive value of 100% (4/4) and negative predictive value of 99.983% (23,787/23,790; p < 0.001). For the diagnosis of VWD, firsttrimester ultrasound alone had a sensitivity of 94.1% Roman/Gupta/Fox/Saltzman/Klauser/ Rebarber
Downloaded by: NYU Medical Center Library 216.165.126.139 - 5/7/2015 1:42:28 PM
Table 1. Demographics of women carrying a singleton fetus with
For patients undergoing routine sonographic evaluation of the fetal anatomy during the first and early second trimester, MSAFP did not improve the detection rate for ONTDs or VWDs. In our population, MSAFP screening did not identify a single case of ONTD or VWD. All cases of these anomalies were detected by ultrasound prior to 18 weeks’ gestation. In fact, first-trimester ultrasound had a sensitivity of 87.5% for diagnosing ONTD or VWD. This study holds several important implications. First, if MSAFP is not useful for detecting these structural malformations in patients undergoing first-trimester and early second-trimester anatomical ultrasound, its use as a screening test for ONTDs and VWDs should be reconsidered. Centers that follow the type of ultrasound screening protocol described in this study may safely omit MSAFP screening from routine testing for ONTDs and VWDs. In addition, this study underscores the benefits of the first-trimester anatomy scan. Despite potential changes in aneuploidy screening algorithms since the introduction of noninvasive fetal testing of maternal serum for fetal aneuploidy, the first-trimester ultrasound at 11–14 weeks’ gestation remains a valuable time to evaluate not only the nuchal translucency but also the major fetal anatomical structures. During the time period of this study, evaluating the intracranial translucency for the detection of ONTDs was not utilized. The addition of this screening method may increase the sensitivity of first-trimester screening even further. Additionally, it should be noted that all of our ultrasounds were performed transabdominally; transvaginal ultrasound was only used to confirm a suspected fetal anomaly seen on transabdominal imaging. A number of previous studies on first-trimester detection of central nervous system and skeletal anomalies relied on
transvaginal ultrasound [15–17]. Our study demonstrates that major structural anomalies such as ONTD and VWD can be identified at this early gestational age with high sensitivity and specificity using transabdominal ultrasound as the primary imaging modality. Finally, 2D ultrasound was used initially for all anatomical ultrasounds, with 3D ultrasound as needed. While the benefit of 3D ultrasound has been shown in suspicious cases, this study did not evaluate routine 3D anatomical surveys and we cannot recommend this based on our study [18]. In our study, all ultrasound examinations were performed in a unit with established, written protocols under the direct supervision of maternal-fetal medicine attendings, therefore these findings may not be applicable to all practices. Additionally, studies have shown that body mass index (BMI) can affect the sensitivity of ultrasound for detecting fetal structural malformations. Our population with a median BMI of 22 may not be representative of other populations in which the incidence of obesity is higher. Another limitation of this study is its retrospective design. However, given the rarity of ONTDs and VWDs (0.6 per 1,000 in our population), this type of anomaly does not easily lend itself to be studied prospectively. A final limitation is the lack of documented birth data on every patient. Our practice delivered approximately 25% of the patients in this cohort, of whom we know there were no undiagnosed cases. The other 75% were delivered by locally referring physicians who inform us of any undiagnosed conditions, and no undiagnosed ONTDs or VWDs have been reported. Therefore, while the potential exists for an undiagnosed case by both ultrasound and MSAFP, this is probably very low. MSAFP is used in practice primarily as a screening test for ONTD and VWD. However, elevated MSAFP has also been associated with adverse obstetrical events such as intrauterine growth retardation and preeclampsia. Therefore, if MSAFP is omitted from routine practice, one may miss a proportion of patients at increased risk for these obstetrical outcomes. Whether the use of MSAFP specifically as a screening test for adverse obstetrical outcomes is effective in improving outcomes remains to be determined. In conclusion, first-trimester and early second-trimester ultrasound was highly sensitive and specific for the detection of ONTDs and VWDs in our population. MSAFP did not improve the detection rate of these abnormalities in the setting of this screening protocol. For centers with practice patterns and a patient population similar to our population, MSAFP for detecting these structural malformations can be safely eliminated from routine practice.
Utility of MSAFP Screening after Early Anatomy Ultrasound
Fetal Diagn Ther 2015;37:206–210 DOI: 10.1159/000363654
Discussion
209
Downloaded by: NYU Medical Center Library 216.165.126.139 - 5/7/2015 1:42:28 PM
(16/17), specificity of 100% (23,786/23,786), positive predictive value of 100% (16/16) and negative predictive value of 99.996% (23,789/23,290; p < 0.001). No cases of ONTD or VWD were diagnosed using MSAFP screening. Of the 341 patients with elevated MSAFP, none were diagnosed with ONTD or VWD. A review of the records of cases of ONTD/VWD revealed that none of the patients diagnosed with ONTD or VWD had serum sent for MSAFP screening as the diagnosis had already been made by ultrasound. In our population of fetuses with ONTD/VWD, ultrasound diagnosed 32/32 cases of ONTD/VWD (100%), and none were diagnosed via MSAFP screening (p < 0.001).
References
210
6 Palomaki GE, Hill LE, Knight GJ, Haddow JE, Carpenter M: Second-trimester maternal serum alpha-fetoprotein levels in pregnancies associated with gastroschisis and omphalocele. Obstet Gynecol 1988;71:906–909. 7 Bredaki FE, Poon LC, Birdir C, Escalante D, Nicolaides KH: First-trimester screening for neural tube defects using alpha-fetoprotein. Fetal Diagn Ther 2012;31:109–114. 8 Lennon CA, Gray DL: Sensitivity and specificity of ultrasound for the detection of neural tube and ventral wall defects in a high risk population. Obstet Gynecol 1999;94:562. 9 Morrow RJ, McNay MB, Whittle MJ: Ultrasound detection of neural tube defects in patients with elevated maternal serum alpha-fetoprotein. Obstet Gynecol 1991;78:1055. 10 Timor-Tritsch IE, Fuchs KM, Monteagudo A, D’Alton ME: Performing a fetal anatomy scan at the time of first-trimester screening. Obstet Gynecol 2009;113:402–407. 11 Souka AP, Pilalis A, Kavalakis Y, Kosmas Y, Antsaklis P, Antsaklis A: Assessment of fetal anatomy at the 11–14 week ultrasound examination. Ultrasound Obstet Gynecol 2004;24: 730–734. 12 Syngelaki A, Chelemen T, Dagklis T, Allan L, Nicolaides KH: Challenges in the diagnosis of fetal non-chromosomal abnormalities at 11– 13 weeks. Prenat Diagn 2011;31:90–102.
Fetal Diagn Ther 2015;37:206–210 DOI: 10.1159/000363654
13 Salomon LJ, Alfirevic Z, Bilardo CM, Chalouhi GE, Ghi T, Kagan KO, et al: ISUOG practice guidelines: performance of first-trimester fetal ultrasound scan. Ultrasound Obstet Gynecol 2013;41:102–113. 14 AIUM practice guideline for the performance of an antepartum obstetric ultrasound examination. J Ultrasound Med 2003;22:1111–1115. 15 Bronshtein M, Deret D, Deutsch M, Liberson A, Bar Chava I: Transvaginal sonographic detection of skeletal anomalies in the first and early second trimesters. Prenat Diagn 1993; 13:597–601. 16 Timor-Tritsch IE, Monteagudo A, Warren WB: Transvaginal ultrasonographic definition of the central nervous system in the first and early second trimesters. Am J Obstet Gynecol 1991;164:497–503. 17 Carvalho MH, Brizot ML, Lopes LM, Chiba CH, Miyadahira S, Zugaib M: Detection of fetal structural abnormalities at the 11–14 week ultrasound scan. Prenat Diagn 2002;22:1–4. 18 Buyukkurt S, Binokay F, Seydaoglu G, Gulec UK, Ozgunen FT, Evruke C, Demir C: Prenatal determination of the upper lesion level of spina bifida with three-dimensional ultrasound. Fetal Diagn Ther 2013;33:36–40.
Roman/Gupta/Fox/Saltzman/Klauser/ Rebarber
Downloaded by: NYU Medical Center Library 216.165.126.139 - 5/7/2015 1:42:28 PM
1 ACOG Committee on Practice Bulletins: ACOG practice bulletin. Clinical management guidelines for obstetrician-gynecologists. Number 44, July 2003 (replaces Committee Opinion Number 252, March 2001). Obstet Gynecol 2003;102:203–213. 2 Martin JA, Hamilton BE, Ventura SJ, et al: Births: Final Data for 2009. Natl Vital Stat Rep 2009. http://www.cdc.gov/nchs/data/nvsr/ nvsr60/nvsr60_01.pdf (accessed November 29, 2011). 3 Wald NJ, Cuckle H, Brock JH, Peto R, Polani PE, Woodford FR: Maternal serum-alpha-fetoprotein measurement in antenatal screening for anencephaly and spina bifida in early pregnancy. Report of UK collaborative study on alpha-fetoprotein in relation to neural-tube defects. Lancet 1977;1: 1323– 1332. 4 Milunsky A: Prenatal detection of neural tube defects. VI. Experience with 20,000 pregnancies. JAMA 1980;244:2731–2735. 5 Wang ZP, Li H, Hao LZ, Zhao ZT: The effectiveness of prenatal serum biomarker screening for neural tube defects in second trimester pregnant women: a meta-analysis. Prenat Diagn 2009;29:960.
