Nội dung

Obstet Gynecol Clin N Am 31 (2004) xi – xiii Preface Ultrasound in obstetrics Lynn L. Simpson, MD Guest Editor Ultrasound has become an integral component of obstetric care, with the vast majority of patients having at least one ultrasound examination during pregnancy. From the determination of early pregnancy and gestational age to the evaluation of fetal growth and well being, ultrasound is a valuable diagnostic tool for the practicing obstetrician. Recent advances in obstetric ultrasonography have increased its importance in managing pregnancies at risk for aneuploidy, structural anomalies, preterm delivery, and blood flow abnormalities. Compiled of contributions from leading experts across the country, this issue of Obstetrics and Gynecology Clinics of North America demonstrates the expanding role of ultrasound in the field of obstetrics. In the United States, ultrasound has been incorporated into prenatal screening programs aimed at identifying fetal chromosomal abnormalities. From their important work on the FASTER Trial (First and Second Trimester Evaluation of Risk), a multicenter prospective study comparing first and second trimester methods of screening for fetal aneuploidy, Karlla Brigatti and Dr. Malone provide a thorough review of first trimester screening including the ultrasonographic evaluation of nuchal translucency. The genetic sonogram, comprised of an evaluation of various sonographic markers during the second trimester, has been used to provide an individualized risk assessment for patients. An expert in both Maternal Fetal Medicine and Genetics, Dr. Stewart presents the potential benefits and obvious limitations of ultrasound in the detection of various fetal chromosomal abnormalities. 0889-8545/04/$ – see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ogc.2004.02.001 xii L.L. Simpson / Obstet Gynecol Clin N Am 31 (2004) xi–xiii In addition to decreasing the likelihood of fetal aneuploidy, patients want reassurance that their infants will be born without major structural abnormalities. Dr. Goldberg, who has devoted his career to prenatal diagnosis, provides an excellent overview of the routine screening ultrasound examination and the expected detection rates for fetal anomalies. My chapter on screening for congenital heart disease follows with the conclusion that the evaluation of multiple cardiac views at the time of routine prenatal ultrasound has the highest probability of detecting heart defects prior to birth. In contrast to the prenatal detection of major fetal malformations, there are many ultrasonographic findings that may or may not represent true pathology. Drs. Rochon and Eddleman present a detailed review of the most controversial ultrasound findings and provide a useful evidence-based approach to their management. Diagnostic and therapeutic interventions are often necessary for patients at risk for aneuploidy or when an ultrasonographic abnormality is identified. Experienced clinicians, Drs. Ralston and Craigo provide a comprehensive review of the various ultrasound-guided procedures that are in use today for fetal diagnosis and therapy. Although the fetus is often the focus during obstetric ultrasound examination, an evaluation of the cervix may be of importance in some patients. Drs. Doyle and Monga present an excellent discussion on the utility of ultrasound in women with prior second trimester pregnancy loss, previous preterm delivery, and multiple gestation. They provide logical guidelines for the ultrasonographic assessment of cervical length in patients at risk for preterm birth, emphasizing that the transvaginal approach is the optimal way to evaluate the cervix during pregnancy. In addition to an evaluation of cervical length, obstetric ultrasound plays an important role in multiple gestations. Drs. Egan and Borgida provide an extensive review of the use of ultrasound in twins, from diagnosis to delivery, demonstrating its favorable impact on the management of these highrisk pregnancies. Ultrasound evaluations in the third trimester involve assessments of fetal growth and well-being. An expert in ultrasonography, Dr. Lerner presents an overview of fetal growth and the accuracy of ultrasound to detect abnormalities such as intrauterine growth restriction and macrosomia. In addition to fetal growth, obstetric ultrasound permits an evaluation of the intrauterine environment. In a well-illustrated review, Dr. Marino discusses the use of ultrasound to evaluate the amniotic fluid volume, fetal membranes, umbilical cord, and placenta. This issue of Obstetrics and Gynecology Clinics of North America is concluded with a comprehensive presentation on fetal Doppler velocimetry. All leaders in the field, Drs. Mari, Detti, Cheng, and Bahado-Singh present the major applications of Doppler velocimetry in obstetrics. Although Doppler velocimetry is a relatively new technique, it has become an integral component of fetal testing and represents a significant advance in the field of obstetric ultrasound. I would like to extend my sincere thanks to the authors who contributed to this issue on ‘‘Ultrasound in Obstetrics’’. It provides a thorough update L.L. Simpson / Obstet Gynecol Clin N Am 31 (2004) xi–xiii xiii on recent advances in the field and it is my hope that the contents will be useful to practitioners providing care to pregnant women. Lynn L. Simpson, MD Guest Editor Associate Professor of Obstetrics and Gynecology Director of Labor and Delivery Division of Maternal Fetal Medicine Columbia Presbyterian Medical Center 622 West 168th Street, PH-16 New York, NY 10021, USA E-mail address: ls731@columbia.edu Obstet Gynecol Clin N Am 31 (2004) 1 – 20 First-trimester screening for aneuploidy Karlla W. Brigatti, MS, Fergal D. Malone, MD Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Columbia University College of Physicians and Surgeons, Columbia Presbyterian Medical Center, 622 West 168th Street, PH16, New York, NY 10032, USA Prenatal screening for Down syndrome and other aneuploidies has expanded substantially over the past 20 years. Initially only women of advanced maternal age ( 35 years old at delivery) or those with a previously affected pregnancy were offered the option of invasive prenatal diagnosis using amniocentesis or chorionic villus sampling (CVS). Subsequently, prenatal diagnosis of aneuploidy became possible for those in the general obstetric population identified at increased risk for Down syndrome by second-trimester multiple marker serum screening or abnormal second-trimester sonographic markers, or soft signs, for Down syndrome. At present, the most efficient multiple marker screening test in the second trimester is known as the ‘‘quad’’ screen, a biochemical marker panel comprised of alpha-fetoprotein (AFP), human chorionic gonadotropin (hCG), unconjugated estriol, and inhibin-A [1]. This combination approach yields sensitivities for Down syndrome of 67% to 76% for a 5% false-positive rate, depending on whether menstrual or sonographic dating are used [2]. This common method of screening has several limitations. The earliest it can reliably be performed is 15 weeks gestation, limiting the choice of definitive diagnosis of aneuploidy to amniocentesis and pushing prenatal diagnosis into the latter second trimester. Furthermore, over 25% of Down syndrome cases are not detected with this screening approach, and the 5% false-positive rate ensures that as many as 60 amniocentesis procedures need to be performed for every single case of Down syndrome detected [3]. Given the pregnancy loss rate of 1 in 200 associated with amniocentesis, about one normal fetus is lost for every three fetuses with Down syndrome detected. Clearly, the current approach of second-trimester screening is not ideal. A great deal of interest has been directed toward shifting prenatal screening for Down syndrome and other aneuploidies to the first trimester using the sonographic measurement of the fetal nuchal translucency (NT) alone and in com- E-mail address: fdm9@columbia.edu (F.D. Malone). 0889-8545/04/$ – see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/S0889-8545(03)00119-0 2 K.W. Brigatti, F.D. Malone / Obstet Gynecol Clin N Am 31 (2004) 1–20 bination with other sonographic and serum markers. This article focuses on the current data and status of first-trimester screening for Down syndrome and addresses the issues of implementation before it can be endorsed for widespread use in everyday clinical practice. Fetal nuchal translucency Nuchal translucency refers to the normal subcutaneous fluid-filled space between the back of the fetal neck and the overlying skin. In most cases, this area can be measured accurately and reproducibly on ultrasound between 10 and 14 weeks’ gestation. It is commonly believed that the larger the NT measurement, the greater it’s association with Down syndrome, other aneuploidy, major structural malformations, and adverse pregnancy outcome (Fig. 1) [4,5]. The etiology of increased NT may be variable, but it is commonly believed to be caused by fluid accumulation in the nuchal region because of aortic isthmic narrowing or other fetal cardiovascular defects [4], abnormalities in the extracellular matrix, or abnormal or delayed development of the lymphatic system [6]. Nuchal translucency screening for Down syndrome Earlier studies of NT-based screening were generally performed on small numbers of subjects and retrospective in nature, drawn from select high-risk populations. They demonstrated substantial variation in Down syndrome detection rates ranging from 46% to 62%, likely caused by differing criteria and skill Fig. 1. Ultrasound image of a fetus with Down syndrome at 12 weeks gestation with an increased nuchal translucency of 3.7 mm. K.W. Brigatti, F.D. Malone / Obstet Gynecol Clin N Am 31 (2004) 1–20 3 levels at measuring NT, differences in success of obtaining measurements, variation in gestational ages included in screening, and varying definitions of normal versus abnormal NT cutoffs [3]. These studies using high-risk women could not effectively extrapolate their results to the role of NT screening in the general population, because it overestimates the true performance of the test. Results of studies in the general obstetric population in a routine clinical setting have been mixed, with a range of detection rates for Down syndrome between 29% and 100%. Table 1 includes 30 published studies on the performance of NT-based screening for Down syndrome in the general population between 1966 and April 2003 [7– 36]. Studies were included in this table if patients were reported as being unselected or from the general population, but excluded if they described less than five cases of Down syndrome or retrospective case:control series [37 – 40]. In total these studies include 316,311 patients screened by NT measurement in the first trimester. A total of 1177 fetuses with Down syndrome were ascertained in this population, for a prevalence of 3.7 per 1000 pregnancies. In 11 of the 30 studies included in Table 1, the prevalence of Down syndrome was 5 per 1000 or greater, suggesting that these studies were not representative of the general obstetric population [7,13 – 15,19,20,25,26,30, 34,35]. Using data from all 30 studies, NT screening had an overall sensitivity for Down syndrome of 77% with a 6% false-positive rate. The odds of a positive screen result being a true positive for Down syndrome were approximately 5%. The data from these studies suggest that an abnormal NT measurement is 13 times more likely to be present in cases of Down syndrome, compared with when the fetus does not have this condition. Conversely, a normal NT measurement is about one quarter as likely in unaffected cases. It should be noted that these likelihood ratios may be overestimated because of the lack of accounting for the intrauterine lethality of Down syndrome in most of these studies; as many as 40% of fetuses alive at the time of first-trimester screening result in spontaneous intrauterine demise [41]. Underascertainment of Down syndrome is a significant limitation of studies in which a fetal or neonatal karyotype is not obtained on all patients. Because Down syndrome pregnancies are more likely to result in fetal demise, a significant portion of early pregnancy losses may have Down syndrome. In one review of the topic, the mean Down syndrome detection rate for studies subject to ascertainment bias was 77%, whereas it was only 55% in studies not subject to it [42]. Only 9 of the 30 studies listed in Table 1 described efforts to maximize the ascertainment of Down syndrome cases in stillbirth or early pregnancy losses [8 –10,16,17,23,28,33,36]. Ultimately, under ascertainment of Down syndrome cases can only be minimized by study methodologies that use extensive pregnancy follow-up, and eliminated altogether with complete karyotypic information on all pregnancies that were subjected to screening. This has been a criticism of the largest study to date on NT-based screening in the general population, conducted by the Fetal Medicine Foundation in London on 96,127 unselected patients at 22 centers between 10 and 14 weeks gestation. That series reported a Down syndrome detection rate of 82% for an 8% false- 4 Down syndrome Study Kornman et al [7] Taipale et al [8] Hafner et al [9] Economides et al [10] Theodoropoulos et al [11] Snijders et al [12] Pajkrt et al [13] De Biasio et al [14] Quispe et al [15] Whitlow et al [16] Schwarzler et al [17] Thilaganathan et al [18] Krantz et al [19] O’Callaghan et al [20] Niemimaa et al [21] Schuchter et al [22] Audibert et al [23] Michailidis et al [24] Number of fetuses 537 6939 4233 2256 3550 96,127 1473 1467 424 6443 4523 9802 5809 1000 1602 9342 4130 7447 Prevalence* Sensitivity (%) FPR % PPV % LR (+) LR ( ) 13 0.9 1.7 3.5 3.1 3.4 6.1 8.9 16.5 3.6 2.7 2.1 5.7 8 3.1 2 2.9 3.1 2/7 (29) 4/6 (67) 3/7 (43) 5/8 (63) 10/11 (91) 268/326 (82) 6/9 (67) 8/13 (62) 7/7 (100) 13/23 (57) 10/12 (83) 16/21 (76) 24/33 (73) 6/8 (75) 3/5 (60) 11/19 (58) 9/12 (75) 19/23 (83) 6.4 0.8 1.7 1 2.6 8 1.8 6.7 1.7 0.3 4.9 4.7 5 6.2 11.6 2.3 4.9 4.5 5.6 6.7 4.1 17.9 9.9 3.4 18.2 7.5 50 37.1 4.3 3.3 7.6 8.8 1.6 5 4.3 5.4 5 83 25 63 35 10 37 9 59 188 17 16 15 12 5 25 15 18 0.8 0.3 0.6 0.4 0.1 0.2 0.3 0.4 — 0.4 0.2 0.3 0.3 0.3 0.5 0.4 0.3 0.2 K.W. Brigatti, F.D. Malone / Obstet Gynecol Clin N Am 31 (2004) 1–20 Table 1 Studies of nuchal translucency ultrasound in an unselected prenatal population TOTAL 21,959 10,157 2557 6841 4939 1152 6234 17,229 16,237 14,383 7536 39,983 9.6 6.3 3.9 2.5 2.8 12.2 3.4 2.6 2.2 5.7 5 2.1 316,311 3.7 174/210 (83) 58/64 (91) 7/10 (70) 17/17 (100) 8/14 (57) 9/14 (64) 13/21 (62) 20/37 (54) 24/35 (69) 64/82 (79) 38/38 (100) 54/85 (63) 8.9 9.6 6.5 4.3 4.9 4.2 2.8 5 5 5 5 5 8.2 5.7 4 5.5 3.2 15.8 7 2.3 2.9 8.3 9.4 2.6 9 9 11 23 12 15 22 11 14 16 20 13 0.2 0.1 0.3 — 0.5 0.4 0.4 0.5 0.3 0.2 — 0.4 910/1,177 (77.3) (95% CI: 75 – 80) 5.9 (5.8 – 6) 4.7 (4.5 – 4.8) 13.1 (12.7 – 13.5) 0.24 (0.22 – 0.27) Pooled 95% confidence intervals given in parentheses at bottom of table. Abbreviations: FPR, Falsepositive rate; LR (+), likelihood ratio for Down syndrome given positive result; LR ( ), likelihood ratio for Down syndrome given negative result; MoM, multiples of median; PPV, positive predictive value. * Prevalence of Down syndrome per 1000 ascertained pregnancies. K.W. Brigatti, F.D. Malone / Obstet Gynecol Clin N Am 31 (2004) 1–20 Gasiorek-Wiens et al [25] Zoppi et al [26] Brizot et al [27] Wayda et al [28] Schuchter et al [29] Murta and Franca [30] Rozenberg et al [31] Crossley et al [32] Lam et al [33] Bindra et al [34] Comas et al [35] Wald et al [36] 5 6 K.W. Brigatti, F.D. Malone / Obstet Gynecol Clin N Am 31 (2004) 1–20 positive rate, equivalent to a 77% detection rate for a 5% false-positive rate [12]. Investigators in that study calculated that based on the maternal age and gestational age distribution of the enrolled subjects, in the absence of any screening, 266 live Down syndrome births would have resulted in their study group. Assuming that as many as 40% of first-trimester Down syndrome cases spontaneously demise in utero, the 266 live births with Down syndrome suggest that at least 443 fetuses with Down syndrome were viable at 10 to 14 weeks gestation (40% of 443 = 177; 433 to 177 = 266 term live births). The quoted detection rate of 268 (82%) per 326 should have been stated more correctly as 268 (60%) per 443 [41]. Underascertainment of true cases of Down syndrome in this study most likely masks a true sensitivity between 60% and 77% for a 5% false-positive rate [12,41]. Indeed, this issue may be one of the reasons the Fetal Medicine Foundation group has revised the performance characteristics of NT-based screening five times over the past 6 years, with detection rates varying from 73% to 84% for a false-positive rate of 5% [12,34,43– 45]. Another limitation of the current literature on NT-based screening is the lack of information on the success rate at obtaining an NT measurement [10 – 12, 14 – 16,19 – 22,24,28,30,35]. Some studies suggest a 100% success rate at obtaining an NT measurement [17,25 –27,34] but none provide any information on the Box 1. Criteria to maximize good quality of NT ultrasound 1. NT ultrasound should only be performed by sonographers certified in the technique. 2. Transabdominal or transvaginal approach should be left to the sonographer’s discretion, based on maternal body habitus, gestational age, and fetal position. 3. Gestation should be limited between 10 and 14 weeks (Crown Rump Length (CRL) 36 to 80 mm). 4. Fetus should be examined in a mid-sagittal plane. 5. Fetal neck should be in a neutral position. 6. Fetal image should occupy at least 75% of the viewable screen. 7. Fetal movement should be awaited to distinguish between amnion and overlying fetal skin. 8. Calipers should be placed on the inner borders of the nuchal fold. 9. Calipers should be placed perpendicular to the fetal body axis. 10. At least three NT measurements should be obtained, with the mean value of those used in risk assessment and patient counseling. 11. At least 20 minutes may need to be dedicated to the NT measurement before abandoning the effort as failed.