Genetic diagnosis in fetuses with biliary tract system abnormalities: a meta-analysis and systematic review Authors: Hanyu Cao, Yanping Qian, Zhu Zhang Published: 2026 ABSTRACT BackgroundBiliary tract system (BTS) anomalies in fetuses are rare, but carry a wide range of clinical prognoses. The present meta-analysis aims to determine the prevalence and spectrum of genetic etiologies in fetuses with BTS anomalies and evaluate their clinical implications.MethodsPubMed and Ovid Medline were searched up to June 2025. Studies reporting prenatal genetic results and outcomes of fetuses with BTS anomalies were included. The spectrum of genetic etiologies detected across included studies was systematically summarized. Pooled prevalence of genetic etiology was calculated using random-effects models, with stratified analyses by isolation status, associated anomaly types, specific BTS phenotypes, genetic testing methods, and abnormality categories. We also assessed the incidence of biliary atresia (BA) in fetuses with normal genetic test results.ResultsSeventeen studies comprising 1,105 fetuses were included. The pooled prevalence of genetic etiology was 9.1% (95% CI, 3.3–16.4%). The risk differed sharply between isolated versus nonisolated status: 0% (95% CI, 0.0-0.2%) in isolated and 32.6% (95% CI, 18.1–48.6%) in nonisolated cases. Among genetically positive cases, numerical chromosomal abnormalities (including aneuploidies and triploidies) accounted for 41.3% (mainly trisomy 18), chromosomal structural abnormalities (including copy number variations, CNVs) accounted for 33.6%, and monogenic disorders (e.g., cystic fibrosis, CF) for 25.0%. The detection rate was higher with chromosomal microarray analysis (CMA)/copy number variation sequencing (CNV-seq) than with karyotype alone (13.5% vs. 9.4%). In fetuses with normal genetic results, postnatal BA occurred in 1.1% (95% CI, 0.0%-8.5%) of the nonisolated group and 0.0% (95% CI, 0.0%-1.3%) of the isolated group.ConclusionsIn fetal BTS anomalies, genetic risk depends largely on associated findings. The 0% diagnostic yield in isolated cases comes mainly from karyotyping and CMA; sequencing was rarely used, so rare monogenic disorders cannot be fully excluded. Nonisolated cases carry a high probability of genetic anomalies and benefit from genetic evaluation. Postnatal surveillance for BA is still needed even with normal genetic results.Trial registrationNot applicable. This systematic review and meta-analysis was registered in PROSPERO (CRD420251128401).Supplementary InformationThe online version contains supplementary material available at 10.1186/s12884-026-09355-z. Background The development of the gallbladder commences as early as the fourth embryonic week, with all elements of the biliary tree becoming recognizable by the fifth week [1, 2] after which fetal hepatic cells begin producing bile at 12 weeks of gestation and bile staining becomes visible by 14 weeks [3]. However, the fetal gallbladder can’t easily be visualized until the second trimester in most cases. In 1985, Hata et al. first reported the normal sonographic characteristics and measurements of fetal gallbladder in the second and third trimesters, demonstrating a detection rate ranging from 37.5% (20–23 weeks) to 64.7% (24–27 weeks) [4]. In 2002, Blazer reported a visualization rate of over 99% at 14 to 16 weeks of gestation [5]. Advances in prenatal imaging have improved the detection of various biliary tract system (BTS) malformations, including nonvisualization of the fetal gallbladder (NVFGB), enlarged gallbladder (GB), small GB and choledochal cyst (CDC), among others [6, 7]. Although rare, these findings carry significant prognostic implications, ranging from benign variants such as gallbladder agenesis to severe conditions like biliary atresia (BA) or cystic fibrosis (CF). While these ultrasound features provide clues for prenatal diagnosis, distinguishing between benign conditions and poor outcomes still remains challenging, which may cause substantial anxiety or even lead to unnecessary termination of pregnancy [8]. Several studies have described postnatal outcomes following prenatal detection of abnormal fetal gallbladder. Di Pasquo et al. reported that among fetuses with persistent non-visualization of the gallbladder, BA occurred in about 5% of isolated cases and 18% of non-isolated cases, while CF was seen in 3% and 23%, respectively [9]. Markova et al. confirmed these findings in a larger cohort, showing that persistent non-visualization carried risks of BA (8.5%), CF (12.5%), and chromosomal anomalies (6.3%), whereas transient NVFGB generally predict favorable outcomes [9, 10]. Accordingly, a better understanding of the genetic etiology and postnatal outcomes associated with fetal biliary tract anomalies is critically needed to guide clinical practice. However, most published evidence predates the widespread clinical implementation of prenatal exome or genome sequencing. Current data remain largely limited to karyotype and chromosomal microarray analysis (CMA)/copy number variation sequencing (CNV-seq), which may underestimate the full spectrum of monogenic causes underlying fetal BTS anomalies. In the present study, we performed a systematic review with meta-analysis of 17 studies to investigate the genetic associations and spectrum of genetic etiology in fetuses with BTS anomalies, aiming to inform more tailored prenatal management and genetic counseling. Methods Protocol The systematic review protocol (No. CRD420251128401) specified the research objectives, eligibility criteria, and intended analytical methods. During implementation, the registered protocol was followed with only minor adjustments. This systematic review and meta-analysis was conducted and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (see Supplementary Material 1). All methodological processes including search strategy, data extraction, and statistical analysis were conducted in accordance with the registered protocol. Search strategy A comprehensive literature search on the BTS anomalies was conducted in databases of PubMed and Ovid Medline, up to 30 June 2025 with the search limited to English and Chinese. The following search terms and their combinations were used: “fetal”, “prenatal”, “gallbladder”, ”biliary”, ”choledochal” (see Supplementary Material 2). Two authors independently conducted the initial search and study selection according to the inclusion and exclusion criteria, including identifying and deleting any duplicate papers, screening the title and abstract, and identifying studies for full-text review. We also reviewed the bibliographies of the retrieved articles to find further relevant trials. PICOS framework Population Fetuses with prenatal ultrasound diagnosis of BTS anomalies, including NVFGB, enlarged gallbladder, small gallbladder, and other morphological abnormalities. Intervention/exposure Genetic testing (karyotype, CMA/CNV-seq, CFTR gene analysis, or genome sequencing) was performed to identify genetic abnormalities. Comparator Not applicable for prevalence meta-analysis; for stratified analyses, comparisons were made between isolated vs. nonisolated BTS anomalies, different types of associated anomalies, different BTS phenotypes, and different genetic testing methodologies. Outcomes Prevalence and spectrum of genetic etiologies (numerical chromosomal abnormalities, chromosomal structural abnormalities, monogenic disorders). Postnatal incidence of BA in fetuses with normal genetic test results. Study design Observational studies and case series with ≥ 3 cases that reported prenatal genetic results and outcomes of fetuses with BTS anomalies. Eligibility criteria All selected studies must meet the following criteria: (1) studies that investigated the detection rate of genetic etiology with or without postnatal outcomes of fetal BTS anomalies; (2) results were reported with sufficient data for the quantitative evaluation; (3) English or Chinese language; (4) full text articles; (5) when there were duplicated data in studies by the same authors, only the largest or latest one was selected. The major exclusion criteria were as follows: (1) conference abstracts, reviews, case reports or case series involving fewer than 3 patients were excluded; (2) presented data were not sufficient for analysis. Data extraction Two investigators (Hanyu Cao, Yanping Qian) independently collected the information below from all qualified studies based on the inclusion standard listed above: first author, year of publication, patient country, number of patients, study type, gestational age at diagnosis, follow-up time, methods for prenatal imaging, type of BTS abnormalities and their specific definition, genetic testing method, evaluated outcome. Quality assessment All included studies were evaluated using the Newcastle-Ottawa Scale (NOS) [11]. The same two investigators (Hanyu Cao and Yanping Qian) who performed data extraction also independently assessed the quality of each included study. Studies were categorized as follows: high quality (≥ 7 stars), moderate quality (5–6 stars), and low quality (≤ 4 stars). Any disagreements arising at any stage of the process were first resolved through discussion between the two investigators. If a consensus could not be reached, a final decision was made by a third investigator (Zhu Zhang). Statistical analysis First, this meta-analysis systematically evaluated the prevalence and spectrum of genetic anomalies in fetuses with BTS abnormalities. The pooled prevalence of genetic anomalies was calculated, and stratified analyses were performed based on the isolation status of the BTS finding, the type of associated anomalies, the specific BTS phenotype, and the genetic testing methodology employed. Additionally, the incidence of BA among fetuses with normal genetic results was assessed. All selected studies, including those reporting zero genetic events, were included in the meta-analysis to minimize selection bias. A random-effects model was applied to all pooled estimates to account for potential heterogeneity among the observational studies. Study heterogeneity was evaluated with the I² statistic. All analyses were performed using the software Stata 12.0 (Stata Corp, TX, USA). Results Description of included studies The selection process is shown in Fig. 1. A total of 17 selected studies with 1105 patients were selected from the 1231 articles identified in the database [2, 5, 9, 12–25]. The main characteristics of these studies are summarized in Table 1. The publication year spanned from 1993 to 2025, with sample sizes ranging from 4 to 271 cases. Of these studies, six were from Israel [2, 5, 12, 13, 16, 22], four were from China [15, 21, 23, 24], three were from France [17, 18, 25], and one each was from Germany, America, UK and Turkey [9, 14, 19, 20]. Three studies were published between 1993 and 2000, three between 2001 and 2010, six between 2011 and 2020, three between 2021 and 2022, two between 2023 and 2025. A detailed summary of prenatal imaging modalities, diagnostic criteria for each BTS anomaly, and postnatal confirmation status for each included study is provided in supplementary Table 1 (see Supplementary Material 3). All included studies reported prenatal genetic diagnostic results utilizing the karyotype (12 studies) [2, 5, 9, 12–14, 17–19, 21, 22, 25], CMA or CNV-seq (6 studies) [15, 16, 20, 21, 23, 24], CFTR gene analysis (7 studies) [5, 9, 12, 13, 17, 18, 25], exome sequencing (1 study)24, deployed individually or in combination. Specifically, CMA alone was employed in four studies [15, 16, 20, 23], karyotype alone in four [2, 14, 19, 22], karyotype plus CMA in one21, karyotype plus CFTR gene analysis in seven [5, 9, 12, 13, 17, 18, 25], and CMA/CNV-seq plus genome sequencing in one24. All cases of fetal BTS anomalies included in the primary studies were diagnosed by prenatal ultrasonography. Supplementary Table 2 (Supplementary Material 4) was added to specify the genetic testing methods used in each study, along with a complete list of associated anomalies as well as the corresponding diagnostic outcomes in cases with nonisolated BTS abnormalities. For chromosomal structural abnormalities, only those reported as pathogenic or likely pathogenic (P/LP) were included in the primary analysis when clinical pathogenicity classification was provided by the original studies. For the majority of studies that did not provide a pathogenicity classification, all reported copy number variations (CNVs) were conservatively included in the genetic abnormality group, as their clinical impact could not be reliably ascertained from the published data. Furthermore, twelve out of the 17 studies were rated as high quality, five as moderate quality and none as low quality (see Supplementary Material 5). Fig. 1PRISMA flow of information selection and exclusion PRISMA flow of information selection and exclusion Table 1Main characteristics of the included studiesStudyYearPatient CountryNumberof patientsStudy TypeGA at diagnosis(weeks)Follow-up timeType of BTS abnormalitiesEvaluated outcomeKaratas2025Turkey78retrospective244yNVFGB and atypical gallbladderchomosomal anomalies; additional anomaly; neonatal outcomeZeng2025China227retrospective27.8 ± 3.4N/ANVFGB and gallbladder size anomalieschomosomal anomalies; single nucleotide variants; associated findings; neonatal outcomeQian2022China271retrospective26 + 4(19 + 1–37 + 1)1-6yisolated and nonisolated BTS anomaliesultrasound soft markers; chomosomal anomalies; pregnancy outcomeTing2022China19retrospective21.3(20.3–22.3)6yNVFGBadditional findings; amnio result; final diagnosis on GBZhang2022China65retrospective26.22N/ANVFGBchromosome analysis; associated anomalies; neonatal outcomeDi Pasquo2019Germany16retrospective23–338yNVFGBassociated anomalies; neonatal outcome; chromosome analysisSagi-Dain2018Israel51retrospective19.8 ± 3.8N/ANVFGBchromosomal anomaliesBardin2016Israel32retrospective17–22N/ANVFGBadditional findings; chromosome analysis; final diagnosisDreux2012France102retrospective23(median)N/ANVFGBadditionnal ultrasound abnormalities; final diagnosisDuguépéroux2011France37retrospectiveN/AN/ANVFGBadditionnal ultrasound abnormalities; final diagnosisShen2011Israel21prospective16.1(14–22)4 m-2.5yNVFGBadditional anomalies; chromosome analysis; final diagnosisBoughanim2008France4case series22–262yNVFGBchromosome analysis; final diagnosisOchshorn2007Israel22prospective17–22N/ANVFGBchromosome analysis; additional malformation; final diagnosisBlazer2002Israel34prospective14–16N/ANVFGBchromosome analysis; additional findings; fetal or neonatal outcomeHertzberg1996USA101prospective30.771d(1-268d)NVFGBchromosome analysis; additional findings; neonatal outcomeSepulveda1995UK8retrospective24.6(19–31)N/Aenlarged gallbladderchromosome analysis; additional US findings; perinatal outcomeBronshtei1993Israel17retrospective14–29N/Agallbladder anomalieschromosome analysis; associated anomalies; neonatal outcomeBTS Biliary tract system, GA Gestational age, GB Gallbladder, NVFGB Non-visualisation of fetal gallbladder, US Ultrasound, N/A Not Available Table 1Main characteristics of the included studiesStudyYearPatient CountryNumberof patientsStudy TypeGA at diagnosis(weeks)Follow-up timeType of BTS abnormalitiesEvaluated outcomeKaratas2025Turkey78retrospective244yNVFGB and atypical gallbladderchomosomal anomalies; additional anomaly; neonatal outcomeZeng2025China227retrospective27.8 ± 3.4N/ANVFGB and gallbladder size anomalieschomosomal anomalies; single nucleotide variants; associated findings; neonatal outcomeQian2022China271retrospective26 + 4(19 + 1–37 + 1)1-6yisolated and nonisolated BTS anomaliesultrasound soft markers; chomosomal anomalies; pregnancy outcomeTing2022China19retrospective21.3(20.3–22.3)6yNVFGBadditional findings; amnio result; final diagnosis on GBZhang2022China65retrospective26.22N/ANVFGBchromosome analysis; associated anomalies; neonatal outcomeDi Pasquo2019Germany16retrospective23–338yNVFGBassociated anomalies; neonatal outcome; chromosome analysisSagi-Dain2018Israel51retrospective19.8 ± 3.8N/ANVFGBchromosomal anomaliesBardin2016Israel32retrospective17–22N/ANVFGBadditional findings; chromosome analysis; final diagnosisDreux2012France102retrospective23(median)N/ANVFGBadditionnal ultrasound abnormalities; final diagnosisDuguépéroux2011France37retrospectiveN/AN/ANVFGBadditionnal ultrasound abnormalities; final diagnosisShen2011Israel21prospective16.1(14–22)4 m-2.5yNVFGBadditional anomalies; chromosome analysis; final diagnosisBoughanim2008France4case series22–262yNVFGBchromosome analysis; final diagnosisOchshorn2007Israel22prospective17–22N/ANVFGBchromosome analysis; additional malformation; final diagnosisBlazer2002Israel34prospective14–16N/ANVFGBchromosome analysis; additional findings; fetal or neonatal outcomeHertzberg1996USA101prospective30.771d(1-268d)NVFGBchromosome analysis; additional findings; neonatal outcomeSepulveda1995UK8retrospective24.6(19–31)N/Aenlarged gallbladderchromosome analysis; additional US findings; perinatal outcomeBronshtei1993Israel17retrospective14–29N/Agallbladder anomalieschromosome analysis; associated anomalies; neonatal outcomeBTS Biliary tract system, GA Gestational age, GB Gallbladder, NVFGB Non-visualisation of fetal gallbladder, US Ultrasound, N/A Not Available Main characteristics of the included studies BTS Biliary tract system, GA Gestational age, GB Gallbladder, NVFGB Non-visualisation of fetal gallbladder, US Ultrasound, N/A Not Available Types and rates of genetic etiology The genetic etiology identified across the included studies are summarized in Table 2. Among the 92 cases with abnormal genetic testing results, numerical chromosomal abnormalities (including aneuploidies and triploidies) were found in 38 fetuses (41.3%, 38/92), including 16 trisomy 18 (17.4%, 16/92), 7 sex chromosome aneuploidies (7.6%, 7/92) (three 47, XXX, two Turner syndrome, one Klinefelter’s syndrome, one 47, XYY), 5 trisomy 21 (5.4%, 5/92), 3 trisomy 13 (3.3%, 3/92), two cases both involving mosaic trisomy 9 (2.2%, 2/92) (one 47, XX + 9[44]/46, XX [34], one mos 47, XN, + 9[7]/46, XN[15], and the remaining two studies reported five cases of triploidy (5.4%, 5/92) in total. Moreover, 31 (33.6%, 31/92) cases with chromosomal structural abnormalities (including CNVs) were reported, including recurrent genomic disorders (e.g., 22q11.2 deletion syndrome). However, some of these cases were reported without sufficient details (e.g., precise breakpoints, gene counts, or overlap with known pathogenic regions), which prevented a reliable assessment of their pathogenicity. Moreover, CF was identified in 20 cases (21.7%, 20/92), representing a significant proportion of the monogenic findings. Among the 20 CF cases, 10 were associatedwith isolated echogenic bowel, 1 with isolated bowel dilation, and 2 with both echogenic bowel and bowel dilation. The remaining 7 casespresented as isolated NVFGB. In addition, pathogenic single-nucleotide variants (SNVs) were detected in three other cases (3.3%, 3/92) by genome sequencing, involving the KMT2D, PTEN, and COL1A2 genes. Additionally, five cases of α0 thalassemia with the homozygous --SEA/--SEA genotype were reported in one study. These cases were not included in the 92 fetuses with an identified genetic etiology in our main analysis, as this finding likely reflects the high endemicity of α⁰ thalassemia in the study region. Table 2Spectrum of genetic abnormalities detected in the included studiesChromosome AneuploidiesChromosomal Structural AbnormalitiesSNVT21T18T13Sex Chromosome AneuploidiesOther Chromosome AneuploidiesTriploidy Karatas 2025 [20]2411(one Turner syndrome)3(2q37.3 mutation;46XX, inv.(p15.3q14.2);5p dup) Zeng 2025 [24]33(one Klinefelter’s syndrome, one Turner syndrome, one 47, XXX)1(one mos 47, XN, + 9[7]/46, XN[15])9(P/LP CNVs: one 1p36.33p36.31 microdeletion, one 17p12 microdeletion, one 1p22.2p22.1 microdeletion, one 16p13.3 microdeletion, two 2q13 microdeletion, one 16p12.2p11.2 microdeletion, one 17q12 microdeletion and one 17q12 microduplicatioin)3(KMT2D c.11851 C > T; PTEN c.635-1G > C;COL1A2 c.3034G > A) Qian 2022 [23]31(47, XXX)12 (P/LP CNVs: one 1q21.1q21.2 microdeletion, one 7q11.23 microduplication, two 7q11.23 microduplication, one 14q11.2q12 microduplication, one 17q12 microdeletion, one 17p12 microdeletion, one 22q11.21 microduplication, one 2q36.1q37.3 microduplication and 14q32.33 microdeletion, one 20q13.2q13.33 microduplication and one 12p13.33 microdeletion, one trisomy of the short arm of chromosome 9, one 9p24.3p22.1 deletion) Ting 2022 [15]111(22q11.2 microduplication) Zhang 2022 [21]1(47,XX + 9[44]/46,XX[34]) Di Pasquo 2019 [9] Sagi-Dain 2018 [16]112(one 16p11.2 microduplication, one Xp22.31 microdeletion) Bardin 2016 [2]1(14q32.1 microduplication) Dreux 2012 [25]1(DG)10(CF) Duguépéroux 2011 [17]2(one DG, one 5q-16p balanced translocation)5(CF) Shen 2011 [12]211(CF) Boughanim 2008 [18]1(CF) Ochshorn 2007 [13]1(47, XXX)1(CF) Blazer 2002 [5]1(47, XYY)42(CF) Hertzberg 1996 [19]1 Sepulveda 1995 [14]21Bronshtei 1993 [22]1T21 Trisomy 21, T18 Trisomy 18, T13 Trisomy 13, CF Cystic Fibrosis, CNV Copy Number Variantions, SNV Single-Nucleotide Variant, DG DiGeorge Syndrome Table 2Spectrum of genetic abnormalities detected in the included studiesChromosome AneuploidiesChromosomal Structural AbnormalitiesSNVT21T18T13Sex Chromosome AneuploidiesOther Chromosome AneuploidiesTriploidy Karatas 2025 [20]2411(one Turner syndrome)3(2q37.3 mutation;46XX, inv.(p15.3q14.2);5p dup) Zeng 2025 [24]33(one Klinefelter’s syndrome, one Turner syndrome, one 47, XXX)1(one mos 47, XN, + 9[7]/46, XN[15])9(P/LP CNVs: one 1p36.33p36.31 microdeletion, one 17p12 microdeletion, one 1p22.2p22.1 microdeletion, one 16p13.3 microdeletion, two 2q13 microdeletion, one 16p12.2p11.2 microdeletion, one 17q12 microdeletion and one 17q12 microduplicatioin)3(KMT2D c.11851 C > T; PTEN c.635-1G > C;COL1A2 c.3034G > A) Qian 2022 [23]31(47, XXX)12 (P/LP CNVs: one 1q21.1q21.2 microdeletion, one 7q11.23 microduplication, two 7q11.23 microduplication, one 14q11.2q12 microduplication, one 17q12 microdeletion, one 17p12 microdeletion, one 22q11.21 microduplication, one 2q36.1q37.3 microduplication and 14q32.33 microdeletion, one 20q13.2q13.33 microduplication and one 12p13.33 microdeletion, one trisomy of the short arm of chromosome 9, one 9p24.3p22.1 deletion) Ting 2022 [15]111(22q11.2 microduplication) Zhang 2022 [21]1(47,XX + 9[44]/46,XX[34]) Di Pasquo 2019 [9] Sagi-Dain 2018 [16]112(one 16p11.2 microduplication, one Xp22.31 microdeletion) Bardin 2016 [2]1(14q32.1 microduplication) Dreux 2012 [25]1(DG)10(CF) Duguépéroux 2011 [17]2(one DG, one 5q-16p balanced translocation)5(CF) Shen 2011 [12]211(CF) Boughanim 2008 [18]1(CF) Ochshorn 2007 [13]1(47, XXX)1(CF) Blazer 2002 [5]1(47, XYY)42(CF) Hertzberg 1996 [19]1 Sepulveda 1995 [14]21Bronshtei 1993 [22]1T21 Trisomy 21, T18 Trisomy 18, T13 Trisomy 13, CF Cystic Fibrosis, CNV Copy Number Variantions, SNV Single-Nucleotide Variant, DG DiGeorge Syndrome Spectrum of genetic abnormalities detected in the included studies T21 Trisomy 21, T18 Trisomy 18, T13 Trisomy 13, CF Cystic Fibrosis, CNV Copy Number Variantions, SNV Single-Nucleotide Variant, DG DiGeorge Syndrome Stratified analysis of genetic etiology prevalence Prevalence by isolation status Of the 17 studies, 16 reported data on genetic etiology allowing classification into isolated (n = 417) and nonisolated (n = 453) BTS anomalies. One study was excluded from this comparison as it did not specify the isolation status for cases with genetic etiology [19]. The pooled prevalence of genetic etiology was 9.1% with 95% confidence interval (CI) of 3.3–16.4%, with the isolated subgroup demonstrating a prevalence of 0%(95%CI, 0.0%−0.2%), compared to 32.6% (95%CI, 18.1–48.6%) in those with nonisolated anomalies (Fig. 2). Fig. 2Pooled prevalence of genetic etiology by isolation status Pooled prevalence of genetic etiology by isolation status Prevalence by type of associated anomalies in nonisolated cases Of all included studies, sixteen reported nonisolated BTS anomalies (n = 415) and one study was excluded due to the lack of clear description regarding the specific nonisolated type from further evaluation [20]. We further stratified the genetic risk by the nature of associated findings, including soft markers, structural abnormalities, polyhydramnios/oligohydramnios. Among the remaining 16 studies, only three reported all three categories of associated findings (soft markers, structural abnormalities, and polyhydramnios/oligohydramnios). The other 13 studies reported only one or two of these categories. Consequently, when we performed separate meta-analyses for each category, studies that did not contribute cases to a specific category were automatically excluded from that particular analysis. The overall pooled prevalence of genetic etiology was 13.1% (95%CI, 4.4%−24.2%). The prevalence of genetic etiology was significantly higher in fetuses with major structural anomalies (19.6%; 95%CI, 7.1%−34.9%) compared to those with isolated amniotic fluid abnormalities (11.9%; 95% CI, 0.0%−45.9%) or only soft markers (6.4%; 95% CI, 0.0%−26.7%) (Fig. 3). Fig. 3Prevalence of genetic etiology by type of associated anomalies in nonisolated cases Prevalence of genetic etiology by type of associated anomalies in nonisolated cases Prevalence by Type of BTS Anomalies Data on specific type of BTS anomalies-including gallbladder enlargement, small gallbladder, NVFGB and other morphological anomalies were available across all included studies. Among these, three studies reported the rate of genetic etiology in cases with gallbladder enlargement, two in cases with small gallbladder, fifteen in cases with NVFGB, and three in cases of other structural anomalies such as atypical gallbladder morphology, choledochal cyst, and bile duct dilatation. When stratified by phenotype category, the prevalence of genetic etiology was highest among fetuses with other BTS anomalies (16.0%; 95%CI, 0.0%−61.8%), followed by those with gallbladder enlargement (9.8%; 95%CI, 1.6%−22.4%) and small gallbladder (9.0%; 95%CI, 3.5%−16.4%).The lowest prevalence was observed in cases with NVFGB (8.3%; 95%CI, 4.3%−13.3%), as illustrated in Fig. 4. Fig. 4Prevalence of genetic etiology by BTS phenotype Prevalence of genetic etiology by BTS phenotype Prevalence by genetic testing methodology Studies were grouped by the highest level of chromosomal genetic testing performed. Six studies employed CMA or CNV-seq, whereas the remaining eleven relied on karyotype alone. The pooled prevalence of genetic etiology was substantially higher in studies utilizing advanced genomic techniques of CMA/CNV-seq: (13.5%; 95% CI, 5.5%−23.8%) compared with those using karyotype only (9.4%; 95% CI, 3.6%−17%), as summarized in Fig. 5. This difference may reflect the enhanced detection capability of CMA and CNV-seq for submicroscopic chromosomal anomalies that are not identifiable by karyotype. Notably, targeted CFTR gene analysis and genome sequencing were used in some included studies and were not included in this grouping, as these methods detect SNVs, which CMA/CNV-seq and karyotype can not identify. Fig. 5Prevalence of genetic etiology by genetic testing methodology Prevalence of genetic etiology by genetic testing methodology Spectrum of genetic etiology by category Of the 92 fetuses with an abnormal genetic diagnosis, numerical chromosomal abnormalities (including aneuploidies and triploidies) constituted the most frequently identified category, representing 3.6% (95% CI, 0.8%−7.6%) of all evaluated cases. Structural chromosomal abnormalities (including CNVs) were the second most prevalent finding (1.2%; 95% CI, 0.2%−2.7%), followed by pathogenic SNVs (0.8%; 95% CI, 0.0%−3.4%). The distribution of these genetic anomaly subtypes is presented in Fig. 6. Fig. 6Spectrum of genetic etiology in fetuses with BTS anomalies Spectrum of genetic etiology in fetuses with BTS anomalies Incidence of biliary atresia in fetuses with normal genetic results This analysis included 15 studies that provided follow-up data on fetuses with BTS anomalies and normal genetic testing results. Two studies were excluded due to insufficient data for pooling [19, 24]. Among the remaining cases, a total of 24 cases of BA were identified among fetuses with normal genetic testing results, the incidence of BA was 1.1% (95% CI, 0.0%−8.5%) in the nonisolated group and 0.0% (95% CI, 0.0%−1.3%) in the isolated group, as detailed in Fig. 7. Notably, all BA cases had normal genetic findings, with only one exception carrying a 5p14.3 microdeletion of 768 kb that was classified as a variant of uncertain significance (VUS). Fig. 7Incidence of biliary atresia in fetuses with normal genetic results Incidence of biliary atresia in fetuses with normal genetic results Discussion Recent advances in prenatal ultrasound have improved the detection of biliary tract abnormalities [7, 8]. These uncommon anomalies demonstrate a wide prognostic spectrum, ranging from benign variants such as gallbladder agenesis to severe conditions including BA and CF9,10. However, the genetic etiology remains poorly characterized, impeding evidence-based decisions regarding the necessity of prenatal genetic testing [16, 23]. This meta-analysis is the first and the largest comprehensive synthesis providing a comprehensive quantitative synthesis of the associated genetic risk by pooling data based on 1105 patients from 17 studies. The spectrum of identified genetic anomalies is informative. Among 92 genetically abnormal cases, 41.3% involved numerical chromosomal abnormalities (including aneuploidies and triploidies), predominantly trisomy 18, which was in accordance with previous studies [24]. As anticipated, all 16 trisomy 18 cases universally exhibited additional sonographic anomalies [20]. Sex chromosome aneuploidies (18.4%) constituted the second most prevalent category, with the 47, XXX being the most common subtype. Among the seven sex chromosome aneuploidy cases, four exhibited additional structural anomalies, while three demonstrated isolated manifestations. Triploidies (13.2%) were less common. Chromosomal structural abnormalities (including CNVs) were identified in 33.6% of all cases, with recurrent genomic disorders at 22q11.2 (DiGeorge syndrome region), 17q12 (HNF1B deletion syndrome), and 7q11.23 (Williams syndrome region) being most prevalent. These regions are associated with recognized genomic disorders that often include visceral and developmental anomalies, making them plausible findings in fetuses presenting with complex phenotypes that include BTS abnormalities. CF (21.7%) was the leading monogenic finding, emphasizing that BTS anomalies can be a prenatal presentation of this condition. This warrants consideration of targeted CFTR analysis, especially in cases with echogenic bowel or a family history. The identification of pathogenic SNVs in other genes like KMT2D (Kabuki syndrome), PTEN (PTEN hamartoma tumor syndrome), and COL1A2 (osteogenesis imperfecta) were reported in only one study by trio whole-exome sequencing. While the fetal phenotype was partly concordant with established syndromes (Kabuki syndrome, Cowden syndrome, osteogenesis imperfecta), none of these conditions include BTS anomalies in the phenotype spectrum [26–28]. The authors themselves concluded that the association between these variants and gallbladder anomalies is likely coincidental. Furthermore, the contribution of CMA/CNV-seq is evident: studies utilizing these advanced techniques reported a higher prevalence of genetic anomalies (13.5%) compared to those relying solely on karyotype (9.4%). This difference likely reflects the enhanced detection of chromosomal structural abnormalities, which accounted for 33.6% of positive findings in our analysis. The inclusion of studies predating the widespread use of CMA/CNV-seq, while providing valuable longitudinal data, inevitably leads to an underestimation of the true prevalence of submicroscopic CNVs. Therefore, our pooled estimates should be interpreted as conservative regarding the risk from structural genomic variants. Targeted gene testing (such as CFTR gene analysis or exome/genome sequencing) may be considered in select cases, especially those with echogenic bowel, a family history of CF, or persistent NVFGB. The predominance of numerical chromosomal abnormalities (41.3%), underscores their central role and also reflects the historical reliance on karyotype in a majority of the included studies. Therefore, this finding should also be interpreted within the context of the diagnostic technologies available across the study period. In this study, the pooled results revealed a genetic anomalies prevalence of 9.1%, higher than the 6.9% reported in a previous systematic review [24]. Furthermore, this risk is overwhelmingly concentrated in fetuses with nonisolated BTS anomalies (32.6%), as opposed to those with isolated findings (0%). The negligible risk in isolated BTS anomalies group is reassuring and suggests that routine invasive genetic testing may not be warranted in this subgroup when a detailed anatomical survey is otherwise normal. Sixteen studies presented associated findings in extrahepatic systems such as ultrasound soft markers (e.g., echogenic bowel), abnormal amniotic fluid volume, or structural anomalies detected by prenatal ultrasound scan. The prevalence of genetic conditions was highest in fetuses with concurrent major structural anomalies (19.6%), intermediate in those with amniotic fluid disorders (11.9%), and lowest when only soft markers were present (6.4%). This gradient underscores the importance of a meticulous, detailed ultrasound examination in determining genetic risk. Among BTS phenotypes, NVFGB was the most common. In some cases, the gallbladder may become visible on later scans. A study by Markova et al. reported a 24.6% (128/521) gallbladder visualization rate during follow-up of mid-trimester NVFGB10. In addition, a retrospective cohort by Zhang et al. reported that 21 out of 49 fetuses (42.9%) showed gallbladder visualization in later pregnancy [21]. This underscores the importance of serial ultrasound examinations to confirm persistent NVFGB, rather than making a definitive diagnosis based on a single sonographic finding. In addition, serial ultrasonographic surveillance is important as additional structural anomalies were identified in almost 45.5% of included NVFGB cases included in our meta-analysis, which is higher than the 20% reported in previous studies [9, 12, 25]. This discrepancy is likely attributable to the fact that NVFGB cases with associated anomalies were more likely to undergo amniocentesis. For other BTS phenotypes such as gallbladder enlargement and small gallbladder, the pooled prevalence of genetic etiology was 9.8% and 9.0%, respectively. It is critical to emphasize that these estimates were derived from studies in which these findings were prenatally detected anomalies that specifically prompted genetic investigation; they do not represent the risk in the general fetal population or for incidental findings. Most importantly, the genetic risk was profoundly modified by the presence or absence of additional ultrasound anomalies, a consistent finding across all BTS phenotypes that underscores the necessity of a stratified approach to risk assessment. In our phenotype‑stratified analysis, anomalies beyond classic NVFGB, such as choledochal cysts and atypical gallbladder morphology, carried the highest associated genetic risk (16.0%), though the CIs were wide. The aneuploidies identified in this study, including trisomy 18, trisomy 21, trisomy 13 can generally be detected by standard non-invasive prenatal testing (NIPT). Triploidy, however, is not reliably detected by conventional NIPT. Sex chromosome aneuploidies and common pathogenic microdeletions, such as those involving 22q11.2, 17q12, and 7q11.23, are frequently incorporated into many expanded NIPT panels. Furthermore, single-gene NIPT may theoretically identify common CF causing CFTR mutations [29]. Therefore, when invasive testing is declined, expanded NIPT may provide supportive preliminary information. Nevertheless, NIPT cannot replace invasive diagnostic testing, and it cannot rule out most rare chromosomal or monogenic disorders. Diagnostic confirmation remains essential for clinical decision-making. BA ccurred in 1.1% (95% CI, 0.0%−8.5%) of the nonisolated group and 0.0% (95% CI, 0.0%−1.3%) of the isolated group with normal genetic results. BA cases were almost exclusively seen in fetuses without an identified genetic etiology, with only one case carrying a CNV classified as VUS. This suggests that the etiology of BA is complex. However, most included studies relied on conventional karyotyping or CMA, the absence of identified genetic causes does not fully exclude a genetic contribution. Recent family-based and population-based exome studies in children with isolated BA have demonstrated a low diagnostic yield for pathogenic monogenic variants, suggesting that monogenic causes are unlikely to explain most BA cases [30, 31]. Nevertheless, the persistence of this risk underscores the importance of continued prenatal surveillance and systematic postnatal follow-up for all fetuses with BTS anomalies, regardless of their genetic test results. Most fetal BTS anomalies are clinically isolated and sporadic events [32]. Isolated gallbladder agenesis and transient NVFGB carry no increased familial recurrence risk. Similarly, BA is predominantly sporadic, with rare familial cases and frequent phenotypic discordance among monozygotic twins [32, 33]. Nevertheless, a positive family history of BA, choledochal cyst, gallbladder agenesis raises the likelihood of an underlying genetic etiology [34, 35]. Well‑documented familial clustering of choledochal cyst further supports a heritable component in a small subset of affected kindreds. For CF, autosomal recessive inheritance and CFTR pathogenic variants are established causes of prenatal biliary tract findings, and family history or parental carrier status can directly guide targeted genetic testing [36]. Of note, familial cases are often managed with clinical surveillance alone and are less likely to undergo invasive prenatal diagnostic procedures, which can lead to underascertainment of genetic conditions and bias pooled prevalence estimates in observational studies and meta‑analyses. A detailed family history should be routinely obtained for risk stratification and individualized genetic counseling in pregnancies affected by fetal BTS anomalies. Several potential limitations should be addressed. Significant heterogeneity was present, as reflected by high I² values in some of our pooled analyses. This heterogeneity likely arises from differences in study design (mostly retrospective), sample sizes, a broad range of BTS phenotypes, diagnostic criteria, genetic testing methods, and reporting standards. Variability in ultrasound criteria for NVFGB and gallbladder size thresholds may have affected case selection and isolation classification. Moreover, the influence of older, karyotype only studies is acknowledged as a factor that contributes to heterogeneity and results in a conservative bias for CNV prevalence. Furthermore, nonisolated BTS anomalies are more likely to prompt amniocentesis, introducing ascertainment bias that may underestimate it in isolated cases. Accordingly, our finding of 0% in isolated cases should be interpreted with this bias in mind. Beyond CFTR variants, a few other SNVs identified in one study were considered incidental, highlighting the challenge of distinguishing causative from incidental findings when sequencing fetuses with multiple anomalies, especially when the identified genes have no known link to biliary tract development. In addition, the inconsistent reporting of variant pathogenicity across included studies is another limitation of this study. A proportion of genetic findings lacked formal classification, making it impossible to definitively distinguish incidental variants from those causally related to fetal biliary tract phenotypes. In this sense, more robust and validated evidence is required to further reduce heterogeneity and validate our findings. Conclusions Genetic risk in fetal BTS anomalies is strongly associated with additional ultrasound findings. In isolated cases, the yield of karyotyping and CMA is almost none, suggesting that invasive testing may not be necessary, although rare monogenic disorders such as CF cannot be fully excluded. In contrast, nonisolated cases have a high likelihood of genetic abnormalities and benefit from genetic evaluation, including CMA/CNV-seq and targeted gene testing when indicated. BA occurred almost exclusively in fetuses without an identified genetic cause. Despite the low risk, postnatal follow-up remains important for all fetuses with a prenatal diagnosis of BTS anomaly, regardless of genetic test results. Supplementary Information Supplementary Material 1. PRISMA checklist. Supplementary Material 1. PRISMA checklist. Supplementary Material 2. Search strategy. Supplementary Material 2. Search strategy. Supplementary Material 3.Supplementary Table 1. Supplementary Material 3.Supplementary Table 1. Supplementary Material 4. Supplementary Table 2. Supplementary Material 4. Supplementary Table 2. Supplementary Material 5. Quality assessment (NOS) of included studies. Supplementary Material 5. Quality assessment (NOS) of included studies.