Prenatal Diagnosis and Prognosis of Fetal Hyperechogenic Kidney: A Study of 80 Cases

Background Prenatal diagnosis of fetal hyperechogenic kidney poses a challenge. The aim of this study was to investigate the genetic reasons and prognosis of fetal hyperechogenic kidney diagnosed on prenatal ultrasonography. Methods We retrospectively reviewed the clinical data of 80 cases of prenatally diagnosed fetal hyperechogenic kidney by the obstetric ultrasound. The genetic characteristics and pregnancy outcomes were analyzed using chromosome karyotype analysis, chromosome microarray analysis, and whole-exome sequencing. Results Of the 80 cases, 48 (60%) were those of isolated fetal hyperechogenic kidney and 32 (40%) were those of non-isolated cases, including 4 cases (5%) of urinary system abnormalities, 7 (8.75%) of central nervous system abnormalities, 5 (6.25%) of cardiac abnormalities, and 16 (20%) of multiple abnormalities. Chromosome karyotype analysis and microarray analysis revealed 17 (21.25%) abnormalities, including isolated fetal hyperechogenic kidney (9, 11.25%) and chromosome microdeletion microduplication (17q12 microdeletion syndrome, Williams-Beuren syndrome, 4p16.3-p16.1 microduplication syndrome) (8, 10%). Moreover, 9 patients had single gene mutations, including those of BBS2, BBS7, HNF1B, ACE, CEP290, COL4A5, and PKHD1. Total 48 pregnancies were terminated (57.3%), and the remaining 32 fetuses survived and grew normally, the neonatal renal function tests were normal. non-isolated fetal hyperechogenic kidneys. All the examinations were performed using a Voluson Expert E8 (GE Healthcare, USA), using curvilinear 2.0 to 5.0-MHz transducers. The whole body and accessory structures of the fetus were examined systematically; routine biometric measurements were performed. Fetal kidney examination parameters included bilateral kidney size, renal echogenicity, amniotic uid volume, corticomedullary differentiation (CMD), and cyst characteristics. The diagnostic marker of fetal hyperechogenic kidney is that the echo of the kidney is stronger than that of the liver in the normal fetus during (including BBS2, CEP200) and one mutation gene (COL4A5) with variants of unknown signicance (VOUS). In isolated fetal hyperechogenic kidneys, three pathogenic heterozygous mutation genes (HNF1B, ACE and BBS2) were detected and one hemizygous mutation gene (COL4A5) with VOUS, which was X-linked dominant inheritance. However, in non-isolated cases, three pathogenic mutations (BBS7, HNF1B and PKHD1) and three likely pathogenic mutations (BBS2, CEP290) were found.


Introduction
Fetal kidneys are generally considered hyperechogenic if the renal cortex has greater echogenicity than the liver or spleen and this becomes more ascertained with associated absence of corticomedullary differentiation [1] . In physiological conditions, fetal kidney echo also shows changes during the pregnancy. 40% normal cases were observed renal hyperechogenic during the early second trimester, but progressively were replaced by iso-and hypo-echogenicity thereafter [2] . Hyperechogenic kidneys, may be the rst indicator of underlying renal disease. Part of the fetal with hyperechogenic kidney is combined with chromosome aneuploidy, chromosome microdeletion syndrome (chromosome 17q12 deletion), autosomal dominant (ADPKD) and autosomal recessive polycystic kidney disease, single gene disorders (Bardet-biedl syndrome, Meckel-Gruber syndrome, Perlman syndrome), or even with obstructive uropathy, teratogenic exposure, other renal dysplasia [3] . However, it is impossible to determine the renal function of the fetus during pregnancy. The clari cation and prognosis establishment are challenging in the clinical treatment and prenatal consultation, because of a wide range of outcomes from completely normal renal function to early endstage renal failure and/or pulmonary hypoplasia in the childhood.
As a signi cant development in next-generation sequencing, whole exome sequencing (WES) has proven to be a useful tool for evaluating the fetus with structural abnormalities as the chromosomal microarray analysis (CMA) [4] . Genetic results could be a reference to provide the predicted information about the fetal health risks to pregnant women. Not only this, the need of designing prenatal counseling based on the results of studies on prenatal cohorts and a longterm postnatal follow-up are also important.
In the present study, we retrospectively analyzed the results of ultrasonography, karyotype analysis, chromosome microarray examination, whole-exome sequencing, pregnancy outcome, and follow-up results of 80 cases of fetal hyperechogenic kidney diagnosed on prenatal ultrasonography. We further explored the prenatal diagnosis and clinical signi cance of fetal hyperechogenic kidney.

Study Participants and sample collection
This retrospective study on the prenatal diagnosis of fetal hyperechogenic kidney was performed in the Guangzhou Women and Children's Medical Center between January 2013 and May 2019. We included 80 women who underwent fetal ultrasonography at our center and were diagnosed with fetal hyperechogenic kidney. We divided the patients as those with normal-size or slightly enlarged isolated fetal hyperechogenic kidneys and those with nonisolated fetal hyperechogenic kidneys. All the examinations were performed using a Voluson Expert E8 (GE Healthcare, USA), using curvilinear 2.0 to 5.0-MHz transducers. The whole body and accessory structures of the fetus were examined systematically; routine biometric measurements were performed. Fetal kidney examination parameters included bilateral kidney size, renal echogenicity, amniotic uid volume, corticomedullary differentiation (CMD), and cyst characteristics. The diagnostic marker of fetal hyperechogenic kidney is that the echo of the kidney is stronger than that of the liver in the normal fetus during the second and third trimester [5] . The examinations were performed by experienced maternal fetal medicine specialists and ultrasonography technicians.
( Fig. 1) Karyotype Analysis G-banded chromosome analysis was performed for the 80 cases as per the standard protocols. All fetuses with a hyperechogenic kidney were referred for chromosomal microarray analysis (CMA).

Chromosomal Microarray Analysis
The CMA was performed by using CytoScan HD Array (Affymetrix, Santa Clara, CA, USA) as per the manufacturer's instructions, and the reporting threshold of the copy number variations (CNVs) was set at 100 kb with a marker count of ≥ 50. The results were analyzed using the Chromosome Analysis Suite software.
The detected copy number losses or gains were aligned with the known CNVs listed in the publicly available databases, including the Database of Genomic Variants, UCSC, ClinGen ,OMIM, the DECIPHER database, As per the guidelines [6,7,8] , the CNVs were classi ed as benign, likely benign, likely pathogenic, pathogenic, variants of unknown signi cance (VOUS). Disease-associated analysis and biological analysis were also performed.

Whole exome sequencing
Considering the economic situation and the severity of the ultrasonography examination, the whole-exome sequencing was performed for 45 fetuses. All the trio genomic DNA samples were extracted, and DNA libraries were prepared using a NEXT ex tm Rapid DNA Sequencing kit (5144-02) (Bioo Scienti c, Austin, TX, USA) as per the manufacturer's instructions. A HiSeq2500 sequencer (version 3, Illumina, Inc, San Diego, CA, USA) was used for sample sequencing. MES was performed with at least 200-fold average coverage. Sequencing reads were mapped to the reference human genome version hg19. Variants from the proband and parents were processed together, and the source of each variant was annotated. The MAFs (minor allele frequencies) of all known variants were also reported according to their presence in the dbSNP, 1000 Genomes Project, Exome Aggregation Consortium. Databases such as OMIM, ClinVar and All whole-exome variants were subjected to biological effects analysis, which included the use of databases such as Mutation Taster, SIFT, PolyPhen-2, REVEL, Human Splicing Finder and PROVEAN to predict whether an amino acid substitution or indel has an important biological effect. All of the above selected variants were then classi ed as pathogenic, likely pathogenic, variant of unknown signi cance, likely benign or benign according to the American College of Medical Genetics and Genomics guidelines (ACMG) [9,10] .
During the postnatal follow-up, we evaluated the renal function of neonates on the 42nd day after birth. All the parents received a written explanation of the study and signed a consent form before study participation. Ethical approval was not required because of the retrospective nature of the study.

Results
Total 80 fetuses diagnosed with hyperechogenic kidney using fetal ultrasonography were included. The age of the pregnant women ranged 20-37 years, with a gestational age ranging 13-35 weeks. Among the 80 cases, 48 (60%) were diagnosed with isolated fetal hyperechogenic kidneys, and 32 (40%) comprised the non-isolated group. The primary associated malformation that accompanied non-isolated fetal hyperechogenic kidneys included urinary abnormality (4 cases, 5%), central nervous system abnormality (7 cases, 8.75%), cardiac abnormality (5 cases, 6.25%), and multiple congenital anomalies (16 cases, 20%) ( Table 1). Amniotic volume as well as the size and number of affected kidneys are often used as indicators for prenatal ultrasonographic diagnosis. As shown in Table 2, in the patients with isolated fetal hyperechogenic kidneys, hydramnios, oligohydramnios, and normal amniotic uid were seen in 8 (16.7%), 6 (12.5%), and 34 cases (70.8%), respectively. Unilateral 12 cases (25%), bilateral 36 cases (75%). The size of the kidney was increased in 14 cases (29.1%), reduced in 3 cases (6.25%), and normal in 31 cases (64.6%).   Table 3). Isolated fetal kidney hyperechoic in 9 cases (11.25%), urinary system malformation in 1 case (1.25%), and cardiac abnormality in 2 cases (2.5%), 5 (6.25%) cases with multiple malformations (Table 1). Further analysis showed chromosomal abnormalities in 9 cases (11.25%), including trisomy 21 syndrome, trisomy 13 syndrome, 46, XX, add (12)   Samples from 63 fetuses with normal chromosomal array and karyotyping results were subjected to whole exon sequencing (WES). Data from fetus-fathermother trios were analyzed. The characteristics of structural malformations by WES are summarized in Tables 3 and 4. The mean depth of coverage for the coding regions targeted with the WES was 10²×A mean of 99.56% of bases in the targeted coding regions were covered by at least 10 reads. Moreover, total 9 gene mutations were found in the whole exon sequencing. As shown in table, we detected pathogenic variants of ve genes (including BBS2, BBS7, HNF1B, ACE and PKHD1), likely pathogenic variants of three genes (including BBS2, CEP200) and one mutation gene (COL4A5) with variants of unknown signi cance (VOUS). In isolated fetal hyperechogenic kidneys, three pathogenic heterozygous mutation genes (HNF1B, ACE and BBS2) were detected and one hemizygous mutation gene (COL4A5) with VOUS, which was X-linked dominant inheritance. However, in non-isolated cases, three pathogenic mutations (BBS7, HNF1B and PKHD1) and three likely pathogenic mutations (BBS2, CEP290) were found.   (Table 1). After birth, the subjects were followed up for 42 days; no abnormalities were found during the monitoring of renal function.

Discussion
With the development of prenatal ultrasonography, hyperechogenic kidneys are occasionally observed during routine ultrasonography examinations.
Obstetricians are increasingly facing the challenge of counseling pregnant women with fetal hyperechogenic kidneys. Fetal hyperechogenic kidney is diagnosed after 17 weeks of gestation when the kidneys appear more echogenic than the spleen and/or the liver. A recent research has shown that such hyperechogenic kidney can be a result of autosomal recessive polycystic kidney diseases, autosomal dominant polycystic kidney diseases, and cystic dysplasia. The remaining causes include tubulopathies, tubular dysgenesis, transient hyperechogenicity, tuberous sclerosis, and miscellaneous diseases [11,12] . The differential diagnosis should consider the family history and the presence of associated anomalies. Recent studies have shown that some fetal abnormalities are associated with hyperechogenic kidneys, including Meckel-Gruber syndrome, Joubert syndrome, Bardet-Biedl syndrome, and VACTERL syndrome [13] . Our study also showed abnormalities, such as cystic renal dysplasia, enlarged lateral ventricles, choroid plexus cyst, congenital heart disease, polydactyly, and talipes varus. (Tables 1 and 3).
Prenatal ultrasonography examination plays an important role in the detection of fetal renal dysplasia. In the case of a fetus with renal cysts associated with hyperechogenic kidneys, detailed ultrasonography examination is useful, with careful attention to the brain, heart, hands, feet, spine, posterior fossa, etc. Fetal kidneys must be measured for renal echogenicity, CMD, cyst characteristics (size, location, and number), and amniotic uid volume. In the subjects with isolated fetal hyperechogenic kidneys, amniotic uid abnormalities were present in about 29.2% of the fetuses (8 cases of hydramnios and 6 cases of oligohydramnios) and about 35.4% had kidney size changes; the bilateral and unilateral ratio was 3:1 (Table 2). However, ultrasonography examination does not provide 100% sensitivity, especially for severe oligohydramnios; therefore, in such cases, fetal magnetic resonance imaging can be used [14,15] .
Studies have shown that about 10% of fetal urinary system abnormalities associated with other systemic malformations is associated with chromosomal abnormalities, including trisomy 21 syndrome, trisomy 18 syndrome, trisomy 13 syndrome, and chromosomal microdeletions [16] . Our retrospective analysis found that 9 fetuses had chromosomal abnormalities, accounting for about 11.25% of all cases. Trisomy 21 syndrome was present in 3 subjects, and trisomy 13 syndrome was present in 2 subjects. In isolated fetal hyperechogenic kidneys, we also detected fetuses with chromosomal abnormalities, including trisomy 21 syndrome and trisomy 13 syndrome, similar to previous reports. Moreover, eight cases of chromosomal microdeletion microduplication syndrome were also detected, including 17q12 microdeletion syndrome (6 cases), Williams-Beuren syndrome, and 4p16.3-p16.1 duplication syndrome.
Chromosomal 17q12 microdeletions and microduplications syndrome have been associated with a wide range of clinical phenotypes. In the prenatal setting, deletion of 17q12 is associated with renal cysts and echogenicity [17] , developmental delay [18,19] , autism, and schizophrenia [20] . 17q12 microdeletions encompassing the hepatocyte nuclear factor 1-beta (HNF1B) gene, also referred to as transcription factor 2 (TCF 2). HNF1B plays a crucial role in early development, including causing renal pathology as a result of haploinsu ciency within the commonly deleted region [21] . In humans, mutations in HNF1B cause congenital anomalies of the kidney and the urinary tract, maturity-onset diabetes of the young type 5, and genital malformations [22] . Gondra L, et al.
reported that HNF1B mutation is the leading cause of isolated hyperechogenic fetal kidneys with normal or moderately large size and that HNF1B can be associated with polyhydramnios in the absence of maternal diabetes [23] . In their study, Loirat C, found that 3 out of the 53 children had cystic or hyperechogenic kidneys and heterozygous 17q12 deletion encompassing HNF1B mutation [24] . In our study, we found 6 fetuses with hyperechogenic kidneys, including isolated cases with 17q12 deletion and detected HNF1B mutation (which cause Renal cysts and diabetes syndrome, OMIM:137920, autosomal dominance inheritance); 17q12 microdeletion syndrome has a variety of phenotypes. Even if the fetus only has kidney dysplasia without diabetes mellitus or neurocognitive impairment, future occurrence cannot be ruled out. Therefore, phenotypic prediction and consulting after birth are challenging in cases with prenatal diagnosis of fetal 17q12 microdeletion syndrome.
Furthermore, we detected Williams-Beuren syndrome through chromosome microarray analysis. Williams-Beuren syndrome is a common chromosome microdeletion syndrome characterized by a speci c dysmorphic face and habitus, postnatal growth deceleration, mild to moderate psychomotor retardation, and multiple organ dysfunction [25,26] . Sammour et al. have reported a prevalence of 75% for urinary tract abnormalities [27] . We also detected a fetus with prenatal ultrasound showed oligohydramnios. Mutations in BBS7 could cause Bardet-Biedl syndrome 7 (OMIM: 615984). Bhowmik A, et al [28] reported that ACE gene mutation could cause Renal tubular dysgenesis (OMOM:267430). Moreover, BBS2 gene mutation also could cause Bardet-Biedl syndrome 2(OMIM:615981), which is manifested as severe retinitis pigmentosa, obesity, polydactyly, renal malformation and mental retardation. All these variants were considered to be pathogenic according to the ACMG criteria and these diseases are inherited in autosomal recessive patterns.
Prenatal diagnosis of hyperechogenic kidneys could bene t fetuses or neonates if the renal disorder is detected and treated early. However, prenatal diagnosis of a fetal renal abnormality may be extremely stressful for the parents, can make prenatal counseling challenging, and could even make the parents decide to terminate the pregnancy, especially when the long-term outcome is uncertain. Estroff et al. reported that the survival rate of non-isolated fetuses with hyperechoic kidneys was signi cantly lower than that of isolated fetuses, and oligohydramnios predicted a poor prognosis [29] . Kumar et al. suggested that the factors associated with poor prognosis included bilateral disease, absence of amniotic uid, and presence of associated other malformation [30] . Our study also showed that the pregnancy outcome depends on whether other structural abnormalities are present. Among fetuses with isolated hyperechogenic kidneys, 11.25% had chromosomal abnormalities and 45.8% (22/48) terminated the pregnancy; some parents of fetuses with normal karyotype analysis also chose to terminate the pregnancy. In addition, among fetuses with other structural abnormalities, 25% had chromosomal abnormalities, and 81.25% (26/32) chose to terminate the pregnancy. Further study of fetuses with isolated hyperechogenic kidneys revealed that amniotic uid volume, the size of the kidney, and the number of kidneys involved were important factors that affected the choice of pregnancy outcomes. Parents of fetuses with oligohydramnios and enlarged kidney volume were more likely to terminate the pregnancy. Moreover, no abnormal growth and development were observed in the follow-up of live births, and we found no abnormalities in the renal function of the newborns; this may be associated with an insu cient follow-up duration; additional regular follow-up may be required at a later stage. We also suggested that kidney size and amniotic uid volume were the best prenatal predictors of outcome and found that patients with large kidneys and oligohydramnios are likely to have poor outcomes.

Conclusions
In summary, fetal hyperechogenic kidney is an important ultrasonographic manifestation of congenital kidney dysplasia. When combined with other system abnormalities, the survival rate is poor, and the incidence of chromosomal abnormalities with isolated hyperechogenic fetal kidneys is low. For fetuses with other systemic abnormalities, it is crucial to improve the prenatal gene detection rate by combining multiple monitoring techniques, such as chromosome karyotype analysis, chromosome microarray, and/or whole-exome sequencing, to obtain more reliable evidence for guiding prenatal counseling and providing preventive treatment. In addition, whole-exome sequencing has potential value as a rst-line prenatal diagnostic tool for the diagnosis of disease population, especially in the identi cation of incidental discoveries unrelated to phenotypic presentation. Therefore, it is also very important to improve the accuracy of interpretation of variant types.
Abbreviations CAKUT= Congenital anomalies of the kidney and the urinary tract, CMA= chromosomal microarray analysis, CNVs= copy number variants, VOUS= variants of unknown signi cance, CMD= corticomedullary differentiation, WES= whole exome sequencing Figures Figure 1 Transverse scan shows bilateral hyperechogenic kidneys with normal size. B. Coronal scan of fetus shows bilateral enlarged and hyperechogenic kidneys. C.
Coronal US image shows bilateral renal enlargement. Both kidneys show the multiple tiny cysts within the medulla giving the appearance of dilated tubules with relative preservation of the columns. D. Fetal US image shows an enlarged kidney with multiple tubulars, cystic structures. E and F. Section from the kidneys using hematoxylin and eosin stain (H and E) shows sub-capsular nephrogenic zone with glomeruli. Lower cortex and medulla show some cysts of waring sizes lined by cuboidal epithelium.