SEARCH

SEARCH BY CITATION

Keywords:

  • isolated fetal pyelectasis;
  • prenatal diagnosis;
  • risk adjustment;
  • trisomy 21

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

Objectives

We performed a meta-analysis to examine the performance of second-trimester (14–24 weeks' gestation) isolated fetal pyelectasis as a marker for trisomy 21 and to calculate its associated weighted pooled likelihood ratios.

Methods

PubMed, Ovid MEDLINE and Cochrane databases were searched using the terms ‘pyelectasis’ and ‘pelviectasis’. Studies were included if fetuses with isolated pyelectasis were reported separately from fetuses with other soft markers of aneuploidy and/or structural anomalies and if knowledge of the fetal karyotype was unknown at the time of ultrasound examination.

Results

Individual study statistics were pooled as weighted positive and negative likelihood ratios with 95% CIs, using a random-effects model. Ten observational studies were included (2148 cases of isolated pyelectasis). Isolated fetal pyelectasis was defined in seven out of 10 studies as a renal pelvis anteroposterior diameter of ≥ 4 mm. Isolated fetal pyelectasis was associated with pooled positive and negative likelihood ratios of 2.78 (95% CI, 1.75–4.43) and 0.99 (95% CI, 0.98–1.00), respectively.

Conclusions

The detection of isolated fetal pyelectasis on mid-trimester ultrasound is associated with an increased likelihood of trisomy 21. If the finding of isolated fetal pyelectasis is used to adjust the trisomy 21 risk from maternal serum screening tests, a positive likelihood ratio of 2.78 should be used in the calculation. Copyright © 2013 ISUOG. Published by John Wiley & Sons Ltd.


INTRODUCTION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

Isolated fetal pyelectasis is identified in 1–3% of fetuses during second-trimester targeted ultrasound[1, 2]. Studies have shown that fetal pyelectasis in the presence of other congenital anomalies and/or sonographic soft markers is associated with an increased likelihood of aneuploidy, particularly trisomy 21[3, 4]. When specific sonographic soft markers are identified, likelihood ratios are commonly used to adjust a patient's a priori aneuploidy risk from maternal serum screening and/or nuchal translucency test results. However, controversy exists as to the significance of isolated fetal pyelectasis as a soft marker for trisomy 21[1, 2, 5-7]. Some authors promote adjustment of the trisomy 21 risk when isolated fetal pyelectasis is identified during targeted ultrasound[7], whereas others argue against risk adjustment[8, 9]. The American College of Obstetricians and Gynecologists (ACOG) recommends that risk adjustment based on the identification of sonographic soft markers should be limited to centers with expertise in ultrasound and/or those centers engaged in clinical research to develop a standardized approach to evaluating these markers[10]. In a survey of Society for Maternal Fetal Medicine practitioners, 76% of respondents revise a patient's trisomy 21 risk result from first- and/or second-trimester screening tests in the presence of one or more sonographic soft markers[11]. However, there is no consensus as to which positive likelihood ratio should be used in the risk-adjustment calculation because different likelihood ratios have been reported for each soft marker[6, 8]. The objective of this meta-analysis was to evaluate the performance of isolated fetal pyelectasis as a soft marker for trisomy 21 and to calculate its associated weighted pooled likelihood ratios.

METHODS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

We performed a systematic review to evaluate studies that identified fetuses with isolated pyelectasis in the second trimester and reported trisomy 21 as an outcome. Electronic searches were performed using PubMed, Ovid MEDLINE and Cochrane databases using the terms ‘pyelectasis’ and ‘pelviectasis’. Articles had to be available online or in print between 1966 and December 2011 to be included. No geographic restrictions were applied, but articles had to be published in the English language. The reference lists of included studies were reviewed for identification of additional articles not captured by the electronic search. Each potentially eligible article was examined to prevent inclusion of articles with duplication of data. Authors were contacted for additional information when needed.

Study selection

The abstract of each electronic search result was reviewed for relevance, and studies were excluded if they did not meet the selection criteria. The full manuscript was obtained and reviewed for the remaining studies. Final exclusion or inclusion criteria determination was based on examination of the entire manuscript.

Studies were eligible for inclusion in the meta-analysis if the diagnosis of pyelectasis was made by prenatal ultrasound in the second trimester (14–24 weeks' gestation) and knowledge of the fetal karyotype was unknown at the time of the sonographic evaluation. Studies had to report fetuses with isolated pyelectasis separately from fetuses with pyelectasis in the presence of other sonographic abnormalities. Pyelectasis was considered as isolated if no other major or minor congenital anomalies were identified on ultrasound, including other sonographic soft markers for aneuploidy. In addition, the incidence of trisomy 21 had to be a reported outcome in fetuses with and without isolated pyelectasis, allowing for construction of a 2 × 2 table to calculate likelihood ratios. We excluded review articles, case reports and abstracts that were not published as manuscripts. Prospective, retrospective and case–control studies were included if eligibility criteria were met. In cases in which there was duplication of study subjects or data, we included the most current study that met the inclusion criteria. Our objective was to use studies that included pregnant women of all ages with isolated fetal pyelectasis to derive a meta-analysis of the positive and negative likelihood ratios for trisomy 21.

Two reviewers (V.B. and K.O.) independently extracted data on the study characteristics, outcome and quality. Disagreement regarding inclusion of studies was resolved by consensus between the authors (V.B. and K.O.). Both authors independently assessed the quality of the included studies via the Methodological Index for Non-Randomized Studies (MINORS)[12]. Each of the 12 categories assessed can earn a score of 0–2 points for a maximum total of 24 points. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement guidelines were followed in the preparation of this manuscript[13]. This study was exempt from review by the Institutional Review Board because it is a meta-analysis of previously published studies.

The primary outcome was the incidence of trisomy 21 among fetuses with and without isolated pyelectasis on second-trimester ultrasound. Positive and negative likelihood ratios with 95% CIs were calculated for each individual study to assess the performance of isolated fetal pyelectasis as a soft marker for trisomy 21. Clinical heterogeneity was evaluated and reported in the table of included studies. The presence of between-study statistical heterogeneity was assessed using the Cochran Q-test statistic, and the amount of heterogeneity was quantified using the Higgins I2 statistic. Likelihood ratios were transformed on a log scale, and the weighted summary likelihood ratio with 95% CI was calculated using a random-effects model that incorporates both within-study and between-study variation. Forest plots were generated to illustrate the weighted likelihood ratios. Statistical analysis was performed using Meta-DiSc software[14].

RESULTS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

The search yielded 220 possible citations, of which 181 were excluded after review of the abstract because they did not meet selection criteria. The full manuscript was reviewed in the remaining 39 studies. An additional 29 articles were subsequently excluded because they did not meet the selection criteria. The remaining 10 studies were included in this review and meta-analysis (Figure 1). Each included study received high marks in more than 75% of the categories assessed in MINORS (Figure 2).

image

Figure 1. Flow diagram of the study selection process.

Download figure to PowerPoint

image

Figure 2. Quality assessment using the Methodological Index for Non-Randomized Studies (MINORS). Items scored as: 0 points = item not reported (image), 1 point = item reported but inadequate (image) or 2 points = item reported and adequate (image).

Download figure to PowerPoint

The characteristics of each study are shown in Table 1[1-6, 15-18]. Most study populations comprised high-risk women referred for advanced maternal age or abnormal maternal serum screening results who had targeted anatomy ultrasound examination in the second trimester (Table 1). The primary outcome was based on fetal karyotype, postnatal physical examination or karyotype, physician interview and postnatal records (Table 1).

Table 1. Characteristics of included studies
ReferenceStudy designGA (weeks)Study populationDefinition of pyelectasisa (AP diameter)Other soft markers evaluatedDiagnosis of trisomy 21
  1. a

    Diameter of fetal renal pelvis measured on ultrasound. 2VC, two-vessel umbilical cord; AMA, advanced maternal age; AP, anteroposterior; BPD, biparietal diameter; CM, cisterna magna; CPC, choroid plexus cyst; EIF, echogenic intracardiac focus; FL, femur length; GA, gestational age; NSF, nuchal skin fold.

Carbone et al. (2011)2Retrospective16–22Women from high-risk perinatal center (indications for referral not specified)≥ 4 mmThickened NSF, cystic hygroma, echogenic bowel, EIF, shortened long bones (< 5th percentile for gestation)Prenatal chromosome analysis or postnatal examination
Aagaard-Tillery et al. (2009)[15]Retrospective15–23Women with singleton pregnancy who had first- and second-trimester aneuploidy screening and genetic sonogram performed in same center≥ 3 mmShortened long bones, NSF ≥ 6 mm, CPC, CM > 10 mm, ventriculomegaly ≥ 10 mm, EIF, pericardial effusion, hydrops, echogenic bowel, liver calcification, 2VC, polydactyly, clinodactyly, sandal gap and club footAmniocentesis; neonatal cord blood; tissue sampling (if pregnancy loss, termination or stillbirth)
Bottalico et al. (2009)[16]Retrospective15–23Women referred for AMA, positive second-trimester maternal serum screen, detection of major congenital anomaly or sonographic markers of fetal aneuploidy, or positive family history of aneuploidy≥ 4 mmShortened long bones, NSF ≥ 6 mm, CPC ≥ 3 mm, echogenic bowel, EIF, absence or hypoplasia of 5th middle phalanxAmniocentesis; postnatal examination or from pregnancy termination cytogenetic results
Bornstein et al. (2007)[17]Retrospective2nd trimesterWomen from high-risk perinatal center undergoing amniocentesis after sonographic detection of pyelectasis compared to women with normal maternal serum screen and genetic sonogram who underwent amniocentesis for maternal anxiety or AMA≥ 4 mmNot specifiedAmniocentesis
Smith-Bindman et al. (2007)[6]Prospective16–23Women from prenatal diagnostic centers with an abnormal second-trimester triple screen result≥ 4 mmCPC, NSF thickening, EIF, echogenic bowel, short long bones, clenched hands, clinodactyly, 2VCPrenatal chromosome analysis; California chromosome abnormality database; California Birth Defect database; prenatal database; California Birth, Deaths & Fetal Deaths database
Weisz et al. (2007)[18]Retrospective> 15Women with abnormal integrated screen test result (risk ≥ 1 in 150)≥ 5 mmEIF, echogenic bowel, NSF ≥ 6 mm, shortened FLAmniocentesis or postnatal ascertainment
Coco and Jeanty (2005)1Retrospective16–23Women routinely referred by obstetricians to perinatal center for anatomy ultrasound; ‘low risk’ population≥ 4 mmCPC, EIF, 2VC, NSF thickening, shortening of limbs, radial aplasia, overlapping fingers, talipes, rocker bottom feet, clinodactyly, brachymeso-phalangia of 5th digitPrenatal chromosome analysis, infant postnatal examination or via contact with pediatrician
Bromley et al. (2002)3Prospective15–20Women referred with abnormal maternal serum screening, AMA or family history of aneuploidy≥ 4 mmShortened long bones, NSF ≥ 6 mm, EIF, cystic hygroma, echogenic bowelAmniocentesis
Nyberg and Souter (2001)4Retrospective14–20Women from high-risk perinatal center (indications for referral not specified)≥ 3 mmNSF ≥ 5 mm, echogenic bowel, EIF, shortened long bonesCytogenetic, birth and pathologic records matched with sonographic records
Corteville et al. (1992)5Retrospective14–24Women seen for AMA, dating, abnormal maternal serum alpha fetal protein, risk for biochemical disorder, suspected anomaly, teratogen exposure, obstetric or maternal complications≥ 4 mmShortened long bones, NSF ≥ 6 mm, BPD:FL > 1.5 SD above meanNot specified

Table 2 shows the frequency of isolated pyelectasis in fetuses with trisomy 21 for each study, along with the sensitivity and specificity of isolated pyelectasis for the detection of trisomy 21. Random-effects analysis resulted in a weighted pooled positive likelihood ratio of 2.78 (95% CI, 1.75–4.43) and a pooled negative likelihood ratio of 0.99 (95% CI, 0.98–1.00) for isolated fetal pyelectasis as a marker for trisomy 21 (Figure 3). Significant between-study heterogeneity was identified (Cochran Q = 17.26; P < 0.05) and accounted for nearly half of the variability among included studies (Higgins I2 = 47.8%).

Table 2. Incidence of trisomy 21 in fetuses with and those without isolated pyelectasis
ReferenceTrue positive (n)False positive (n)False negative (n)True negative (n)Sensitivity (95% CI)Specificity (95% CI)
Carbone et al. (2011)[2]9104620960 5610.04 (0.02–0.08)0.98 (0.98–0.98)
Aagaard-Tillery et al. (2009)[15]41035176740.07 (0.02–0.18)0.99 (0.99–0.99)
Bottalico et al. (2009)[16]3996190.25 (0.07–0.52)0.99 (0.98–0.99)
Bornstein et al. (2007)[17]331726690.60 (0.17–0.93)0.68 (0.68–0.68)
Smith-Bindman et al. (2007)[6]59524086120.02 (0.01–0.05)0.99 (0.99–0.99)
Weisz et al. (2007)[18]0191223010.00 (0.00–0.28)0.99 (0.99–0.99)
Coco and Jeanty (2005)113041012 3570.09 (0.00–0.41)0.98 (0.97–0.98)
Bromley et al. (2002)35131596430.03 (0.01–0.06)0.98 (0.98–0.99)
Nyberg and Souter (2001)4518715084880.03 (0.01–0.07)0.98 (0.98–0.98)
Corteville et al. (1992)511911657560.01 (0.00–0.05)1.00 (0.99–1.00)
image

Figure 3. Forest plot of pooled positive (a) and negative (b) likelihood ratios (LRs) from studies of isolated fetal pyelectasis as a marker for trisomy 21. Only first author of each study is given.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

We evaluated the performance of isolated fetal pyelectasis as a soft marker for trisomy 21. In this study, isolated fetal pyelectasis more than doubled the odds of trisomy 21 from maternal serum screening tests, with a pooled positive likelihood ratio of 2.78 (95% CI, 1.75–4.43). The pooled negative likelihood ratio was 0.99 (95% CI, 0.98–1.00), indicating that the absence of isolated fetal pyelectasis does not decrease the odds of trisomy 21 (Figure 3). Our results are consistent with the findings from other studies[4, 6, 8]. Positive likelihood ratios from the 10 included studies ranged from 1.50 to 17.44, with 95% CIs that always included our pooled likelihood ratio of 2.78 (Figure 3).

One of the strengths of our study was the inclusion of articles that specified the number of cases of trisomy 21 among fetuses with isolated fetal pyelectasis separately from fetuses with pyelectasis in association with other anomalies, to determine the most accurate positive and negative likelihood ratios. Furthermore, in this meta-analysis of observational studies, 70% defined fetal pyelectasis as the anteroposterior (AP) measurement of the renal pelvis ≥ 4 mm in the second trimester (two studies used an AP measurement of ≥ 3 mm and one study used an AP measurement of ≥ 5 mm[4, 15, 18]). The use of a random-effects model also strengthened this study because a random-effects analysis accounts for between-study heterogeneity.

One of the most important strengths of this meta-analysis is the use of likelihood ratios as the pooled meta-analysis summary statistic. The positive likelihood ratio (sensitivity/1 – specificity) estimates the odds of disease among subjects who test positive, whereas the negative likelihood ratio (1 – sensitivity/specificity) estimates the odds of disease among subjects who test negative. Likelihood ratios are not influenced by disease prevalence rates and can capture the magnitude of the abnormality for a particular test result. In general, a positive likelihood of 2 indicates a small increase from pretest to post-test probability of disease[19]. Although small positive likelihood ratios are not useful in improving disease detection rates at the population level, they are clinically useful with respect to individual risk assessment[20]. For example, assume a woman's trisomy 21 risk is 1 in 300 based on second-trimester maternal serum screening, and isolated fetal pyelectasis is subsequently detected on targeted ultrasound performed at 20 weeks' gestation. Using a positive likelihood ratio of 2.78, the modified odds of trisomy 21 becomes 1 in 108 in the presence of isolated pyelectasis. This increases her odds above the commonly accepted positive threshold of 1 in 270 used in the USA to report a positive serum screening test, and after proper counseling she may opt for further testing. The concept of individual risk modification is controversial because this practice only slightly improves the trisomy 21 detection rate at the expense of increasing the false-positive rate[21, 22]. Therefore, women must be properly counseled on the risk/benefit ratio for further testing, either non-invasive (i.e. cell-free fetal DNA) or invasive (i.e. amniocentesis). Despite the controversy that exists, aneuploidy risk modification for one or more sonographic soft markers has become increasingly common among practitioners[11]. This meta-analysis provides a pooled positive likelihood ratio for isolated fetal pyelectasis that can be used in the calculation for risk modification if desired.

The limitations of this study relate to the many potential sources of bias inherent in a meta-analysis of observational studies. It is possible that publication bias may have distorted the pooled likelihood ratios as published studies may not be truly representative of all studies performed. Exclusion of non-English language studies could also result in a biased meta-statistic, although this is unlikely in our analysis because only one study was excluded for this reason. Another potential source of bias is the inclusion of older studies, the results of which may have been limited by primitive ultrasound capabilities and inadequate knowledge of soft markers for aneuploidy. An additional constraint is that the assessment of quality of the included studies was limited by a lack of clear reporting in some of the studies, particularly older studies. Furthermore, 30% of included studies defined fetal pyelectasis differently from the AP measurement of the renal pelvis ≥ 4 mm; this limits comparison between studies and can result in biased pooled likelihood ratios. Our results also show a large degree of heterogeneity amongst included studies, which could be caused by many factors, including different populations studied, different observational study designs, as well as the use of different definitions of fetal pyelectasis and ‘isolated’ fetal pyelectasis (Table 1). We used a random-effects model in order to take into account both within- and between-study variability, but despite this, flaws in study design could have affected our results.

The results from this meta-analysis should be interpreted with caution because most studies included women at high risk for aneuploidy based on maternal age, abnormal maternal serum screening and/or abnormal nuchal translucency test results. Therefore, our pooled likelihood ratios may have limited generalizability in a population of women at low risk for aneuploidy.

Given the relatively low incidence of trisomy 21, properly designed prospective studies should include an unselected cross-sectional population, a clear definition of isolated fetal pyelectasis, a clearly defined gestational age range (e.g. 16–23 weeks' gestation), proper ascertainment of trisomy 21 in all fetuses/neonates and an adequately powered sample size. Until such a large, well-designed study is conducted, this meta-analysis represents the largest, most accurate evaluation of the association of isolated fetal pyelectasis with trisomy 21. Our meta-analysis shows that isolated fetal pyelectasis detected on targeted ultrasound more than doubles the odds of trisomy 21 from maternal serum screening test results, particularly in a population of women at high risk for aneuploidy. For those clinicians who choose to modify a woman's a priori risk for trisomy 21 from maternal serum screening tests in the presence of isolated fetal pyelectasis, a positive likelihood ratio of 2.78 should be used in the calculation.

REFERENCES

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES
  • 1
    Coco C, Jeanty P. Isolated fetal pyelectasis and chromosomal abnormalities. Am J Obstet Gynecol 2005; 193: 732738.
  • 2
    Carbone JF, Tuuli MG, Dicke JM, Macones GA, Odibo AO. Revisiting the risk for aneuploidy in fetuses with isolated pyelectasis. Prenat Diagn 2011; 31: 566570.
  • 3
    Bromley B, Lieberman E, Shipp TD, Benacerraf BR. The genetic sonogram a method of risk assessment for Down syndrome in the second trimester. J Ultrasound Med 2002; 21: 10871096.
  • 4
    Nyberg DA, Souter VL. Sonographic markers of fetal trisomies: second trimester. J Ultrasound Med 2001; 20: 655674.
  • 5
    Corteville JE, Dicke JM, Crane JP. Fetal pyelectasis and Down syndrome: is genetic amniocentesis warranted? Obstet Gynecol 1992; 79: 770772.
  • 6
    Smith-Bindman R, Chu P, Goldberg JD. Second trimester prenatal ultrasound for the detection of pregnancies at increased risk of Down syndrome. Prenat Diagn 2007; 27: 535544.
  • 7
    Havutcu AE, Nikolopoulos G, Adinkra P, Lamont RF. The association between fetal pyelectasis on second trimester ultrasound scan and aneuploidy among 25,586 low risk unselected women. Prenat Diagn 2002; 22: 12011206.
  • 8
    Smith-Bindman R, Hosmer W, Feldstein VA, Deeks J, Goldberg JD. Second-trimester ultrasound to detect fetuses with Down syndrome a meta-analysis. JAMA 2001; 285: 10441055.
  • 9
    Bethune M. Literature review and suggested protocol for managing ultrasound soft markers for Down syndrome: thickened nuchal fold, echogenic bowel, shortened femur, shortened humerus, pyelectasis and absent or hypoplastic nasal bone. Australas Radiol 2007; 51: 218225.
  • 10
    ACOG Committee on Practice Bulletins-Obstetrics. ACOG Practice Bulletin: Screening for Fetal Chromosomal Anomalies number 77, January 2007: Obstet Gynecol 2007; 109: 217227.
  • 11
    Fang YMV, Benn P, Egan JFX. Use of the Genetic Sonogram in the United States in 2001 and 2007. J Ultrasound Med 2008; 27: 15431548.
  • 12
    Slim K, Nini E, Forestier D, Kwiatkowski F, Panis Y, Chipponi J. Methodological index for non-randomized studies (minors): development and validation of a new instrument. ANZ J Surg 2003; 73: 712716.
  • 13
    Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009). Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. Ann Intern Med 2009; 151: 4.
  • 14
    Zamora J, Abraira V, Muriel A, Khan KS, Coomarasamy A. Meta-DiSc a software for meta-analysis of test accuracy data. BMC Med Res Methodol 2006; 6: 31.
  • 15
    Aagaard-Tillery KM, Malone FD, Nyberg DA, Flint Porter T, Cuckle HS, Fuchs K, Sullivan L, Comstock CH, Saade G, Eddleman K, Gross S, Dugoff L, Craigo SD, Timor-Tritsch E, Carr SE, Wolfe HM, Bianchi DW, D'Alton ME for the First and Second Trimester Evaluation of Risk (FASTER) Research Consortium. Role of Second-Trimester Genetic Sonography After Down Syndrome Screening. Obstet Gynecol 2009; 114: 11891196.
  • 16
    Bottalico JN, Chen X, Tartaglia M, Rosario B, Yarabothu D, Nelson L. Second-trimester genetic sonogram for detection of fetal chromosomal abnormalities in a community-based antenatal testing unit. Ultrasound Obstet Gynecol 2009; 33: 161168.
  • 17
    Bornstein E, Barnhard Y, Donnenfeld A, Ferber A, Divon MY. The risk of a major trisomy in fetuses with pyelectasis: the impact of an abnormal maternal serum screen or additional sonographic markers. Am J Obstet Gynecol 2007; 196: e24e26.
  • 18
    Weisz B, Pandya PP, David AL, Huttly W, Jones P, Rodeck CH. Ultrasound findings after screening for Down syndrome using the integrated test. Obstet Gynecol 2007; 109: 10461052.
  • 19
    Chien PF, Khan KS. Evaluation of a clinical test. II: assessment of validity. BJOG 2001; 108: 568572.
  • 20
    Akobeng A. Understanding diagnostic tests 2: likelihood ratios, pre- and post-test probabilities and their use in clinical practice. Acta Paediatrica 2006; 96: 487491.
  • 21
    Krantz DA, Hallahan TW, Macri VJ, Macri JN. Genetic sonography after first-trimester Down syndrome screening. Ultrasound Obstet Gynecol 2007; 29: 666670.
  • 22
    Martinez-Zamora MA, Borrell A, Borobio V, Gonce A, Perez M, Botet F, Nadal A, Albert A, Puerto B, Fortuny A. False positives in the prenatal ultrasound screening of fetal structural anomalies. Prenat Diagn 2007; 27: 1822.