Autosomal Recessive Polycystic Kidney Disease and Caroli Disease

An autosomal recessive hepatorenal fibrocystic syndrome, caused predominantly by biallelic PKHD1 variants, with dilatation of the renal collecting ducts and hepatic ductal plate malformation; includes the spectrum of Caroli intrahepatic biliary dilatations (disease and syndrome) and congenital hepatic fibrosis.

Literature and regulatory information updated: 10 October 2026.

0. Initial Reference Table

1. Introduction

1.1. Historical overview

The association between polycystic kidneys and liver disease in children was systematised by Blyth and Ockenden in 1971, in a study entitled “polycystic disease of kidneys and liver presenting in childhood” [19], to which GeneReviews attributes the clinical–histological classification into perinatal, neonatal, infantile and juvenile forms [1]. For three decades, this classification shaped the designation “infantile polycystic kidney disease”, still present in reference texts and in the term “Potter syndrome type I” [1, 20]. The description of the biliary component is earlier: in 1958, Jacques Caroli and colleagues published in Paris two descriptions of congenital cystic dilatation of the intrahepatic bile ducts, one with an attempted classification and another with the unilobar form, which gave the disease its eponym [21, 22].

In the pre-molecular era, distinction between the recessive and dominant forms depended on family history, imaging and histology. The Finnish series of Kääriäinen and colleagues (1988), involving 93 children with polycystic disease, illustrates that period: 17 cases were classified as dominant and 32 as recessive through family studies, and of the 44 sporadic cases, 41 were attributed to the recessive form; early death occurred in 55 of 73 children with the recessive form and in 4 of 18 with the dominant form, and no laboratory parameter distinguished the two entities, although urinary concentrating ability was always reduced in the recessive form [23]. In 1998, Zerres and colleagues estimated an incidence of around 1:20,000 live births in Germany, a figure that remains the most widely cited [1, 10].

The gene was identified in 2002 by two independent groups. Ward and colleagues started from the PCK rat, a model of recessive polycystic disease, demonstrated the orthologous relationship with the human region and found, in 14 probands, 6 truncating and 12 missense variants; they described a transcript of around 16 kb expressed in the kidney, liver and pancreas and a large receptor protein, which they called fibrocystin [24]. Onuchic and colleagues described a gene of at least 469 kb with at least 86 exons assembled into alternative transcripts, whose longest open reading frame encodes a protein of 4074 amino acids, which they called polyductin [25]. This accounts for the dual name fibrocystin/polyductin (FPC). Genetic heterogeneity was only conclusively demonstrated in 2017, with the identification of biallelic variants in DZIP1L, a ciliary transition zone protein [26], followed in 2021 by the description of the first patient with biallelic variants in CYS1, the gene of the cpk mouse model [27], and in 2026 by two new families with CYS1 variants [28].

1.2. Nomenclature and reclassifications

GeneReviews uses “ARPKD” as a broad term for all patients with a clinical diagnosis, with or without a molecular diagnosis, and proposes nomenclature that adds the genetic aetiology to the phenotype and mode of inheritance: ARPKD-PKHD1, ARPKD-DZIP1L, ARPKD-CYS1 and ARPKD-PKD1 [1]. In this article, ARPKD denotes autosomal recessive polycystic kidney disease in the broad sense and ARPKD-PKHD1 the classical form. Most cases are due to biallelic PKHD1 variants; in the larger cohorts, the PKHD1 variant detection rate is around 85% [1], and the 2014 consensus indicated 80–85% in cases with strong clinical or histopathological evidence [15].

The position of PKD1 differs between sources. The 2025 KDIGO (Kidney Disease: Improving Global Outcomes) guideline on autosomal dominant polycystic kidney disease (ADPKD) includes PKHD1, DZIP1L, CYS1 and PKD1 in the “recessive PKD” row of its table of cystic diseases and states that biallelic PKD1 alterations can cause disease from very early onset to adulthood [6]. GeneReviews adopts the same logic with the designation ARPKD-PKD1 [1]. In contrast, the analysis of the international ARegPKD registry (an ARPKD registry based in Cologne) published by Halawi and colleagues in 2023 classifies patients with PKD1 variants as ARPKD phenocopies, together with HNF1B, TMEM67, PKD2 and NPHP3 [29]. The divergence is mainly conceptual: this article treats the ARPKD phenotype due to biallelic hypomorphic PKD1 as an entity within the spectrum (section 2.6) and very early-onset ADPKD forms, monoallelic or digenic, as a differential diagnosis.

1.3. Definitions of the entities within scope

ARPKD is defined by Orphanet as a rare genetic hepatorenal fibrocystic syndrome, with cystic dilatation and ectasia of the renal collecting tubules and hepatic ductal plate malformation with congenital hepatic fibrosis; the typical presentation is in utero or at birth, and the most severe cases feature Potter sequence, oligohydramnios, pulmonary hypoplasia and massively enlarged echogenic kidneys [10]. Hepatic involvement is obligatory: ductal plate malformation is present in all patients, even when not clinically detectable [1, 30].

CHF is a histopathological diagnosis of a developmental disorder of the portobiliary system: defective ductal plate remodelling (ductal plate malformation, DPM), abnormal branching of the intrahepatic portal veins and progressive fibrosis of the portal tracts [12]. Depending on the level of the biliary tree affected, DPM manifests as CHF without macroscopic dilatations, or as CHF with cysts in continuity with the bile ducts, corresponding to Caroli syndrome; Caroli disease, which is much rarer, denotes dilatations contiguous with the biliary tree without CHF [12]. The 2022 guideline of the European Association for the Study of the Liver (EASL) describes Caroli disease as a Todani type V lesion, limited to dilatation of the larger intrahepatic ducts, and Caroli syndrome as the combination of small-duct dilatation with CHF and, frequently, renal cysts [5]. The two forms sometimes coexist in the same family, suggesting a continuum [12]. The 2024 GeneReviews inconsistently uses “Caroli disease” for the biliary dilatations accompanying ARPKD [1]; this article adopts EASL and Orphanet terminology, in which dilatations associated with CHF constitute Caroli syndrome. Isolated CHF (ORPHA:485426) is a parenchymal liver disease with progressive portal fibrosis due to arrested ductal plate maturation, with portal hypertensive, cholangitic, mixed or latent forms, and symptoms mainly in adolescence and young adulthood [4].

2. Molecular Genetics and Pathophysiology

2.1. PKHD1 and fibrocystin

PKHD1 is located at 6p12.3-p12.2 and is one of the largest human genes: it spans around 470–500 kb, has at least 86 exons and an extensive alternative splicing pattern, with several transcripts; the longest open reading frame, with 66 coding exons, produces fibrocystin/polyductin of 4074 amino acids [15, 20, 25]. The 2026 French National Diagnosis and Care Protocol (PNDS) of the Haute Autorité de Santé (HAS) describes the gene as more than 470 kb with 67 exons; the difference from Onuchic's at least 86 exons arises because that author also counts the exons of alternative transcripts [9, 25]. The MANE (Matched Annotation from NCBI and EMBL-EBI) reference transcript is NM_138694.4, corresponding to protein NP_619639.3 [1, 7]. The protein has a large extracellular domain with IPT (immunoglobulin-like, plexin, transcription factor) domains and parallel β-helix repeats, a single transmembrane domain near the carboxyl terminus and a short cytoplasmic tail; some transcripts do not include the transmembrane domain and may encode secreted forms [25]. Expression is concentrated in the kidney, bile ducts and pancreas [1, 24]. The 2014 consensus emphasises that a critical amount of full-length protein appears necessary for normal function [15].

Fibrocystin is localised to the primary cilium, cell junctions and extracellular vesicles, where it forms a complex with polycystin-1 (the product of PKD1) and polycystin-2 (the product of PKD2); fragments of the cytoplasmic tail have been observed in the nucleus and mitochondria [1]. The GeneGuidance record describes it as a single-pass protein in the ciliary membrane of renal and biliary epithelia, promoting ciliogenesis and tubular lumen architecture and regulating centrosomal duplication, the mitotic spindle and cell division oriented according to planar polarity [11]. There is no biochemical assay to measure the deficiency [11]. ClinGen classified the PKHD1–ARPKD relationship as “definitive” in November 2020 [11].

2.2. Mechanism and renal and hepatic pathophysiology

The mechanism is loss of function: biallelic variants cause partial or complete loss of fibrocystin [1]. In preclinical models, fibrocystin regulates signalling cascades shared with ADPKD — intracellular cyclic adenosine monophosphate (cAMP), protein kinase A (PKA), the SRC-STAT3 axis and Myc — and cellular metabolism, which control epithelial proliferation and morphogenesis [1]. The primary cilium functions as a sensory organelle coordinating signalling pathways such as Hedgehog and Wnt, and ARPKD belongs to the ciliopathies, which also encompass nephronophthisis, Joubert, Meckel and Bardet–Biedl syndromes and short-rib thoracic dysplasias [31].

In the kidney, ARPKD cysts originate only in the collecting ducts, as ubiquitous fusiform dilatations of the distal nephron producing numerous microcysts; in ADPKD, cysts arise in all nephron segments and are fewer and larger [20, 30]. Early renal development is normal, and the kidneys retain their reniform shape as they increase in volume [20, 30]. Impaired urinary dilution, activation of the intrarenal renin–angiotensin–aldosterone system without systemic elevation of angiotensin I and II, and dysregulation of the collecting duct epithelial sodium channel contribute to arterial hypertension and fluid-overload hyponatraemia [15]. The urinary concentrating defect is present in all children with ARPKD [16].

In the liver, failure of ductal plate remodelling preserves embryonic biliary structures that may dilate, accompanied by abnormal portal vein branching and periportal fibrosis; the resulting intrahepatic resistance explains PHT with preserved hepatocellular function [1, 12]. In experimental models, fibrocystin deficiency promotes fibrosis through macrophage recruitment [1]. Renal and hepatic progression are independent: in the prospective National Institutes of Health (NIH) cohort, renal disease severity did not correlate with hepatic disease severity [32, 33].

2.3. Variant spectrum and founder variants

Most patients are compound heterozygotes, with private variants throughout the gene, and variants of uncertain significance (VUS) are frequent [1, 30]. There are truncating and non-conservative missense variants, most in the coding region or canonical splicing sites; deep intronic variants detectable only by genome sequencing or RNA analysis have been described [1, 20]. Sequencing detects more than 95% of known pathogenic variants, and deletions or duplications of one or more exons account for less than 5% [1]. The following table presents variants of population or educational relevance, described according to Human Genome Variation Society (HGVS) nomenclature and normalised in VariantValidator for the MANE transcript.

The frequency of c.107C>T is presented using different metrics: 13–20% of patients carry at least one allele, whereas around 10% of the alleles analysed correspond to this variant [16, 30]. GeneReviews presents, without reconciling them, the mutational hotspot and single European origin hypotheses [1]. The Bedouin intronic variant has some of the most benign splicing prediction scores among known PKHD1 splicing variants, explaining why it escaped conventional methods in 12 patients from 5 families; it was identified by short- and long-read genome sequencing and confirmed in urinary tubular epithelial cells [36, 37]. The PKD1 variants described in ARPKD-PKD1 (p.Val1045Met and p.Thr1570Met) were not normalised because the source does not specify the transcript. The French PNDS counts more than 700 pathogenic PKHD1 variants classified according to ACMG criteria, of which around 60% produce a truncated protein (nonsense, frameshift, splicing), around 40% are missense and copy number variants are rare; it attributes c.107C>T p.(Thr36Met) to 20% of European patients, a value at the upper limit of the 13–20% range given above [9].

2.4. Genotype–phenotype correlations in PKHD1

Variant class has prognostic value, but with limitations. Almost all patients with two truncating variants have a severe phenotype with perinatal or neonatal death, and neonatal survivors generally have at least one missense variant; however, some missense variants are as severe as truncating variants and there are survivors with two truncating variants [15, 20, 30]. The study by Burgmaier and colleagues (2021), involving 304 neonatal survivors, predominantly of European ancestry and mainly carrying missense variants, refined these correlations in two independent cohorts [38]. The GeneReviews tables derived from that study are summarised below.

Source: GeneReviews, tables 3a and 3b, based on Burgmaier 2021 and earlier studies [1, 38].

Despite these differences, GeneReviews notes that there were no statistically significant differences in renal survival during childhood and adolescence between subgroups stratified solely by molecular diagnosis [1]. Kidney volume follows genotype: in ARegPKD, median height-adjusted estimated total kidney volume (haeTKV) was 793 mL/m in null/null patients, 403 mL/m in null/missense and 230 mL/m in missense/missense patients [39]. A Spanish series from 2026, involving 40 patients with molecular confirmation according to American College of Medical Genetics and Genomics and Association for Molecular Pathology (ACMG/AMP) criteria, with phase and segregation, showed an allele dosage effect on the kidney: neonatal or infantile onset in 88% of patients with two loss-of-function variants, 56% with one and 29% with none (p<0.001), and KRT in 65%, 31% and 0% (p=0.002); hepatobiliary disease was highly prevalent in all classes, without association with the burden of loss-of-function variants [40]. In the NIH cohort, patients with biallelic non-truncating variants and those with one truncating variant had similar progression [33]. The 2014 consensus summarises the limitation of these correlations: caution is required when predicting the clinical course from the genotype [15].

2.5. Intrafamilial and interfamilial variability

Historically, 20–42% of sibling pairs were discordant (perinatal death in one, survival through childhood in the other) [1]. In ARegPKD, among 70 siblings from 35 families, marked variability in renal function occurred in only three pairs, and in 17 of 20 families of neonatal survivors the siblings had minimal differences in renal function at comparable ages, indicating a strong genotype effect [41]. Variability between families with the same genotype is demonstrated by homozygotes for the Afrikaner founder variant, whose phenotype varies, suggesting genetic or environmental modifiers [1, 34].

2.6. Additional genes: DZIP1L, CYS1 and biallelic PKD1

DZIP1L encodes a ciliary transition zone protein, localised to centrioles and the distal end of basal bodies, which interacts with septin-2 at the periciliary diffusion barrier; its loss impairs translocation of polycystins 1 and 2 to the ciliary membrane [26]. Lu and colleagues (2017) described 7 patients from 4 families, with kidney failure at 12, 18, 20 and 26 years; Hertz and colleagues (2022) described 4 children from 3 consanguineous families homozygous for c.193T>C p.(Cys65Arg), with kidney failure at 9 years [1]. The phenotype includes enlarged hyperechogenic kidneys with multiple small cysts, some macrocysts and poor corticomedullary differentiation; the liver had no cysts and elastography was normal, although one person had hepatosplenomegaly [1]. Orphanet associates DZIP1L with ORPHA:731 [10].

CYS1 encodes cystin-1, a myristoylated ciliary and nuclear protein whose deficiency in the cpk mouse causes polycystic disease; cystin regulates Myc through necdin [27]. The first patient was a 5-year-old child homozygous for c.318+5G>A, with multiple bilateral medullary and cortical cysts and mild CHF [1, 27]. In 2026, two new families were described: one with a homozygous deletion of around 24 kb encompassing the regulatory region and exon 1, and two siblings homozygous for c.4G>A p.(Gly2Ser) in the myristoylation motif, with evidence of a myristoyl-electrostatic switch regulated by phosphorylation of serine 17 by PKA and by the phosphatase PPM1A [28]. CYS1 is not yet listed in Orphanet [10].

Biallelic hypomorphic PKD1 variants cause an early phenotype similar to ARPKD-PKHD1, with prenatally or perinatally enlarged hyperechogenic kidneys and a liver without cysts or CHF in early childhood, although isolated cases of ductal plate malformation have been described; in one patient with p.Val1045Met and p.Thr1570Met, kidney failure occurred at 8.5 years [1]. In ARegPKD, PKD1 phenocopies had 10-year KRT-free survival of 85%, higher than phenocopies due to other genes (47%) [29]. KDIGO 2025 notes that biallelic PKD1 alterations can cause disease from very early onset to adulthood [6].

2.7. Heterozygotes and allelic diseases

PKHD1 heterozygotes do not develop ARPKD, but mild abnormalities occur in a small subgroup. In an NIH study of 110 parents from 64 families, 6 (5.5%) had medullary hyperechogenicity compatible with nephrocalcinosis and 10 (9%) multiple small liver cysts, all asymptomatic and with normal function; complete PKHD1 sequencing in the 16 with abnormalities confirmed only the transmitted variant [42]. KDIGO 2025 includes monoallelic PKHD1 among the “suspected, with limited evidence” genes of the ADPKD and autosomal dominant polycystic liver disease (ADPLD) spectrum, in around 1% of families screened, with very mild renal cystic disease and preserved function into old age and frequent liver cysts, which may occur without renal cysts; this is a low-penetrance genotype [6]. Heterozygous mice show focal tubular ectasia resembling medullary sponge kidney [35]. These abnormalities should not lead to a carrier being classified as having atypical ARPKD.

Caroli syndrome forms part of the ARPKD phenotypic spectrum and PKHD1 variants have been discovered in some patients [5]; Orphanet associates PKHD1 with Caroli disease [2]. CHF and biliary dilatations also occur in other ciliopathies, such as Joubert and COACH (cerebellar vermis hypoplasia, oligophrenia, congenital ataxia, coloboma and hepatic fibrosis), Meckel, Bardet–Biedl, OFD1, Ellis–van Creveld, Jeune and renal–hepatic–pancreatic dysplasia [12]. Smith's cites WDR19 variants as an allelic cause of cranioectodermal dysplasia, Jeune, retinal dystrophy, Senior–Løken and Caroli disease [43]. In ADPKD, CHF is rare: 19 patients from 14 families with CHF and PHT have been documented, without PKHD1 variants as a modifier and without vertical transmission of CHF, suggesting an autosomal modifier [12].

3. Clinical Spectrum and Manifestations by System

3.1. Presentation by age

ARPKD primarily affects the kidneys and liver, and manifestations in other organs are mainly secondary [1]. GeneReviews distinguishes three ages of initial presentation: perinatal (from the prenatal period to 4 weeks) and infantile (from 4 weeks to 1 year), the most frequent, both with the classical enlarged kidneys and differing mainly in the pulmonary involvement of the perinatal form; and presentation in childhood after the first year or in young adulthood, which is less common and may be dominated by hepatobiliary disease [1]. Survival has improved with neonatal respiratory support, dialysis, kidney transplantation, liver transplantation and combined liver–kidney transplantation [1]. Some natural history data predate recognition of genetic heterogeneity and come from “ARPKD” cohorts without molecular stratification; the phenotype is similar regardless of the gene, but aetiology may influence renal survival [1, 29].

Source: GeneReviews, table of features by age at presentation [1].

The relative frequency of manifestations by age, compiled by GeneReviews from general cohorts and PKHD1 cohorts, shows how the burden shifts from the kidney and lung to the liver over the course of life.

Key: +++ always; ++ frequent; + rare; (+) very rare. Source: GeneReviews, table 2 [1].

3.2. Fetal period and respiratory system

Fetal urine constitutes most amniotic fluid from 14–16 weeks onwards, and lack of urine production leads to oligohydramnios [16, 20]. When volume is very low between 16 and 24 weeks, the canalicular stage of lung development, restriction produces Potter sequence: pulmonary hypoplasia, micrognathia, characteristic facies and postural abnormalities such as clubfoot [30, 44]. The earlier oligohydramnios occurs, the more severe the pulmonary effect [44]. Pulmonary hypoplasia of variable severity is the main cause of neonatal morbidity and mortality; massive kidneys restrict diaphragmatic excursion, worsening hypoventilation, and persistent pulmonary hypertension may be prominent; pneumothorax is an additional complication [1, 30]. In a large retrospective European study, 23% of infants (78/333) required respiratory support [1]. Long-term pulmonary function appears good, except when neonatal mechanical ventilation was required [1]. Prematurity is frequent: in an international study of 288 children, mean gestational age was 37.5 ± 2.7 weeks, and 36.1 ± 2.4 weeks in those who required dialysis in the first year [1].

3.3. Kidney

Examination of the newborn reveals palpable bilateral flank masses [1]. Urine output does not usually decrease, because the concentrating defect causes polyuria and polydipsia; oliguria and acute kidney injury may, however, occur in the first weeks [1]. Hyponatraemia occurs in 6–26% of patients and probably results from impaired urinary dilution with fluid overload, at a stage when caloric and fluid intake are coupled; it tends to resolve in the first year, except in severe CKD [1, 15, 30]. Renal function in the neonatal period ranges from mild impairment to failure, and may improve during the first year, comparably to normal renal maturation [1, 30]. Kidney failure is not the main cause of neonatal death [30].

In the initial phase, kidney involvement has a tubulointerstitial profile: disorganisation of the parenchyma and progressive interstitial fibrosis result in an almost constant urinary concentrating defect, with a polyuria–polydipsia syndrome and, in children, delayed acquisition of bladder control, in salt wasting and in possible distal tubular acidosis; proteinuria is low-grade (urine protein/creatinine ratio below 1 g/g, or 100 mg/mmol) and microscopic haematuria is classically absent [9]. Metabolic acidosis, often early, appears in the first months in more than half of patients with neonatal and infantile forms [9]. As the disease progresses, the initially preserved glomeruli become sclerotic through ischaemia and increased intracapsular pressure, and albuminuria increases [9]. Kidney biopsy is almost never performed; when it is, it mainly shows non-specific interstitial fibrosis and fusiform dilatations of the tubules and collecting ducts in the cortex and medulla [9]. The PNDS recommends estimating GFR with the Schwartz equation in children or CKD-EPI in adults and, preferably, with equations combining creatinine and cystatin C (CKiD 2012, FAS U25) [9].

After the first year, functional decline is slow on average. In the American CKiD (Chronic Kidney Disease in Children) cohort, 22 children with ARPKD lost an average of 1.4 mL/min/1.73 m² (−6%) of glomerular filtration rate (GFR) per year, with greater decline from 10 years onwards (−11.5% per year) [1]. In ARegPKD, which includes up to 658 patients with a clinical diagnosis, the mean annual decline in estimated GFR (eGFR) after the first year was 1.3 mL/min/1.73 m² in childhood and adolescence (0.5–2.2 in subgroups), and the estimated probability of kidney failure at 20 years was 50.1% (95% confidence interval, 95% CI, 42.2–57.0) [45]. Kidney volume is related to progression in the early years: haeTKV in the first 18 months correlates with renal survival, which at 10 years was 20% in the highest quartile and 94% in the lowest [39]. After that age, haeTKV remains relatively stable (mean increase of 4.5 mL/m per year), while height-adjusted kidney length decreases [16, 39]. This observation contradicts the 2014 consensus statement that kidney size does not correlate with function [15]. In the NIH cohort, patients with corticomedullary involvement on high-resolution ultrasound had creatinine clearance of 61 ± 32 mL/min/1.73 m², compared with 131 ± 46 in those with medullary involvement only (p<0.0001), and an eGFR slope of −1.6 versus −0.6 mL/min/1.73 m² per year; no patient progressed from medullary to corticomedullary disease [33, 46]. Urinary tract infections are frequent, especially in the first years, and pyelonephritis may be the presenting feature; in a series of 25 patients published in 2010, 40% had had a urinary tract infection before diagnosis, and the proportion reached 52% after a median follow-up of 12 years [1, 9]. Nephrocalcinosis and lithiasis associated with low urinary pH and hypocitraturia have been described; the spongy, hyperechoic appearance of the medulla, due to cystic tubular dilatations, mimics the precalyceal canalicular ectasia of Cacchi-Ricci disease (medullary sponge kidney), which is never accompanied by liver disease, so the combination of precalyceal ectasia and liver disease should point towards ARPKD [9]. In the presence of lithiasis or nephrocalcinosis, the PNDS recommends a complete urinary metabolic work-up (24-hour urine or spot sample) and regular imaging surveillance [9].

Kidney cysts are frequent but not obligatory, unlike in ADPKD, and are usually smaller than 1 cm; in adults, most patients have more than 10 cysts in each kidney, but some have only one or two, unilateral [9]. The ratio of kidney size to height remains stable or decreases during childhood, and kidney growth slows or ceases in adulthood [9]. Nephromegaly is present in 35–90% of patients, depending on the age of the population studied and the degree of renal impairment, and mean kidney length in non-dialysed adults is around 12.6 and 13.2 cm; enlargement is greater when the dilatations are corticomedullary than when they are exclusively medullary [9]. In adults, kidneys may shrink as function is lost, because of fibrosis, and resemble those in medullary cystic disease [1, 35].

3.4. Arterial hypertension and cardiovascular system

Systemic arterial hypertension (HTN) has a prevalence of 33–75% in the series included in the 2014 consensus, is marked in the first months and is associated with the need for mechanical ventilation (p<0.0001) [15]. More recent reviews describe it as almost universal in infants and children with ARPKD, frequently requiring several drugs, compared with around 20% of children with ADPKD [16]. It may be severe and resistant to multidrug therapy and improve over time [1, 30]. The French PNDS reports HTN in the first months of life in 80% of children, a paediatric prevalence of 55–100% depending on the study and HTN in 50% of patients diagnosed in adulthood, usually less severe; it recommends screening at every visit, supplemented by home blood pressure monitoring or ABPM, and cardiological assessment when HTN is confirmed, because severe HTN may cause left ventricular failure [9]. Rates of altered left ventricular geometry with systolic dysfunction are high [1]. In a retrospective American series of 30 patients with a median age of 7.2 years, 93% were receiving antihypertensives and left ventricular hypertrophy (LVH) affected 23% or 50% depending on the ventricular mass index criterion; ventricular mass was not associated with blood pressure after adjustment, but was negatively associated with eGFR [47]. Cerebral and extracerebral aneurysms have been described in 9 people with ARPKD, rarely in children; no association has been demonstrated and screening of asymptomatic patients is not recommended [1, 15].

3.5. Liver and bile ducts

Hepatic fibrosis is invariably present at birth, but clinical, imaging or laboratory signs may be absent in newborns and infants [1]. In the Zerres series (115 children, mean age at diagnosis 29 days), 46% had clinical evidence of hepatic involvement over a mean follow-up of 4.9 years, with a positive relationship with age [1]. In most patients, the liver becomes clinically manifest later than the kidney, and progression in the two organs may be parallel or dissociated [30]. Hepatocellular synthetic and detoxification function remains normal, transaminases are normal or slightly elevated and gamma-glutamyltransferase (GGT) may be markedly elevated [1, 15]. The French PNDS estimates clinical liver involvement in 45% of patients and hepatomegaly on examination in 33%, with a firm or hard liver, sometimes without imaging correlate; liver severity does not depend on kidney severity [9]. According to data from the European Burgmaier cohort cited by the protocol, 11% of patients had raised ALT, 4% raised AST and one third GGT above normal, with hyperbilirubinaemia in only 10%; abnormalities, when present, are usually less than 2 times the upper limit of normal, and imaging shows heterogeneous parenchyma in more than half of patients, cysts, mostly small, in 23% and biliary dilatation in 14%, sometimes limited to one lobe, usually the left [9]. Liver biopsy is not generally necessary and is not specific for ARPKD [9].

The NIH cohort (2003–2009) assessed 73 patients aged 1 to 56 years (mean 12.7 years): initial symptoms were hepatic in 26%, the left lobe was enlarged on magnetic resonance imaging (MRI) in 69%, hepatic echogenicity was increased in 92% and splenomegaly was present in 65% (60% of children under 5 years); platelet count was the best predictor of splenic volume (area under the curve 0.889) [32]. Varices were found in 22 of 31 patients undergoing endoscopy, with 5 variceal bleeds and 2 portosystemic shunts; around 70% had biliary abnormalities, with Caroli syndrome in 40% and isolated common bile duct dilatation in 30% [32]. GeneReviews cites a series in which biliary dilatations were observed in more than 60% of patients [1]. In prospective NIH follow-up, PHT was present in two-thirds, stable in 39% and worsened in 61%, and was associated with lower platelets and relatively higher aspartate aminotransferase, GGT, direct bilirubin and ammonia levels, always with intact synthetic function [33].

CHF leads to PHT with splenomegaly, hypersplenism (thrombocytopenia and leukopenia), portosystemic collaterals, oesophageal and gastric varices, haemorrhoids, protein-losing enteropathy and gastrointestinal bleeding; variceal bleeding contributes substantially to morbidity and mortality [1, 12]. In a Polish series of 17 patients with molecular confirmation (2.5–42 years), 71% had some sign of PHT, 59% splenomegaly, 47% hypersplenism and 92% of those assessed by endoscopy had varices; 4 had gastrointestinal bleeding and 9 underwent band ligation [48]. Portopulmonary hypertension and hepatopulmonary syndrome are rare; ascites and encephalopathy are less frequent than in cirrhosis [12]. In ARegPKD, among 304 neonatal survivors observed up to 15 years of age, 60–65% developed signs of PHT and around 25% substantial hepatic complications (variceal bleeding, liver transplantation, portocaval shunt) [1].

The French PNDS adds quantitative and screening data. PHT affects up to 70% of patients followed into adulthood and splenomegaly more than 50%; hypersplenism mainly affects platelets (thrombocytopenia in 42%) and haemoglobin (anaemia in 25%) and may lower some coagulation factors through splenic consumption, but thrombocytopenia resulting solely from PHT does not increase bleeding risk [9]. In the Burgmaier cohort cited by the protocol, 10% of patients had oesophageal varices and none had gastric varices; ascites is rarely clinical and more often only seen on imaging [9]. It is estimated that half of patients with hepatic fibrosis will have at least one upper gastrointestinal bleed during their lifetime, mostly between 5 and 20 years of age, sometimes as the presenting event [9]. Protein-losing enteropathy complicates childhood PHT in severe forms, with oedema and ascites, hypoalbuminaemia and malabsorption, and is a major risk factor for malnutrition; hepatopulmonary syndrome and portopulmonary hypertension, which are rare, are potentially fatal and require regular screening (section 11.2) [9]. Hepatocellular failure is very rare, with prothrombin time, factor V and INR broadly normal throughout follow-up; hepatic encephalopathy occurs in fewer than 1% of patients, mainly in kidney failure or shortly after kidney transplantation, or as a complication of advanced PHT with portosystemic shunts, with subtle signs in children (sleep–wake reversal, difficulty concentrating, drowsiness or agitation) [9]. Chronic fibrosis and biliary abnormalities may lead to nodular transformation, with rare occurrence of adenoma, cholangiocarcinoma and hepatocellular carcinoma [9].

Abnormal biliary drainage predisposes to recurrent or persistent ascending bacterial cholangitis with sepsis, a major cause of morbidity and mortality, particularly after isolated kidney transplantation under immunosuppression [1]. The classical triad of fever, jaundice and right upper quadrant pain is rare in children, and cholangitis may present as recurrent bacteraemia due to enteric bacteria [1, 15]. According to the Srinath review cited by the PNDS, 7% of patients have at least one episode of cholangitis, and more than half of these will have recurrent episodes, which may progress to septic shock and indicate liver transplantation [9, 49]. There is no clear correlation between the severity of biliary abnormalities and cholangitis risk [15]. The risks of cholangiocarcinoma and hepatocellular carcinoma are increased in CHF, but they have not been described before 40 years of age and are not characteristic of childhood ARPKD [15]. In the review of 1230 patients with CHF, hepatobiliary malignancies (21 cases, 19 cholangiocarcinomas) occurred mainly after 40 years of age in patients with Caroli syndrome or isolated CHF, with a mean age of 60.1 years at cholangiocarcinoma diagnosis and the youngest case at 33 years [12, 49]. In the same review, only one liver cancer was found among 788 patients with ARPKD, but targeted cancer screening had been performed in only 17, which precludes taking this figure as a risk estimate [9, 49].

3.6. Growth, nutrition, neurodevelopment and quality of life

Malnutrition, favoured by gastric compression from massive hepatomegaly and aggravated by ascites and protein loss, is a severity criterion and a priority criterion when liver transplantation is indicated [9]. Feeding difficulties result from gastric compression by the kidneys, liver and spleen, pulmonary hypoplasia, uraemia, CKD-related anorexia and peritoneal dialysis [1, 30]. Aggressive nutritional support in the first two years improved growth even with moderate to severe CKD and PHT; in a CKiD subanalysis (mean age 7.9 years), ARPKD had no specific effect on growth compared with other congenital causes of CKD [1, 30]. Neurocognition in children with mild to moderate ARPKD is comparable to that of children with CKD of similar severity [1, 15]. Severe neurological complications are concentrated in children undergoing bilateral nephrectomy in the first three months and those who have had severe hypotensive episodes (section 7.4) [50]. A German case–control study of 43 children (age 9.0 ± 4.8 years) showed that health-related quality of life follows renal function, is worse on dialysis and, for hepatic symptoms, is not captured by kidney disease-specific questionnaires; developmental delay was the most important extrarenal risk factor, and psychosocial problems and caregiver burden were equal to or greater than in controls with advanced CKD [51].

3.7. Adolescents and young adults

A small but significant proportion of patients retain corticomedullary differentiation into young adulthood and have no CKD until middle age; the disease may present in adulthood with minimal renal involvement and predominantly hepatic disease [1, 16]. In ARegPKD, 45 young adults with a mean age of 21.4 ± 3.3 years had native renal function at CKD stages 1–3 in more than half of cases; 14 (31%) had received a kidney transplant and 6 (13%) a liver or combined transplant [52]. In the NIH cohort, 25% of patients with perinatal presentation (n=31) required KRT at 11 years, whereas in patients symptomatic after the first month (n=42), the same proportion of 25% was reached, for kidney transplantation, only at 32 years [46]. In women, pregnancy is possible; the limited data are in section 7.10 [53].

The French PNDS considers presentation in adulthood probably underestimated. Diagnosis is often late or initially wrong, with patients followed as atypical ADPKD, Caroli disease, CHF or chronic interstitial nephropathy of undetermined cause [9]. In adults, liver disease predominates, in the form of late-discovered CHF revealed by complications of PHT or of Caroli syndrome; kidney involvement is moderate or absent, with slowly progressive CKD, normal-sized kidneys and a few small corticomedullary cysts, calcifications or a Cacchi-Ricci-like appearance [9]. Renal function is often normal at diagnosis, and progression to kidney failure occurs mainly in the fourth decade, although several reports describe preserved function [9]. Among patients with late-presenting ARPKD, more than 40% have a diagnosis of Caroli disease and almost all have concomitant kidney involvement [9]. Circumstances of discovery include renal (cysts, calcifications) or hepatic imaging findings (hepatomegaly, dysmorphism, fibrosis, signs of PHT, more rarely cysts or biliary dilatation), hepatosplenomegaly, pyelonephritis, rupture of oesophageal varices, cholangitis, renal impairment or cholestasis found on blood tests, nephrocalcinosis or lithiasis, kidney cyst infection, aetiological investigation of HTN and family screening [9].

3.8. Caroli disease, Caroli syndrome and isolated congenital hepatic fibrosis

Caroli disease and syndrome share multifocal segmental, saccular or fusiform dilatation of the intrahepatic ducts, which may reach 5 cm; extrahepatic dilatations are frequent in Caroli disease and are attributed to recurrent cholangitis and passage of stones [5]. The main presentation is cholestasis, with stone formation, biliary obstruction, cholangitis and abscesses [5]. In the systematic review by Fahrner and colleagues (12 retrospective studies, 561 patients, 53% women, mean age 41.6 years), most had episodes of cholangitis, sepsis, fever or abdominal pain [14]. Caroli syndrome additionally features CHF manifestations (PHT, splenomegaly, varices) and, in most patients, renal cystic lesions due to fusiform dilatation of the collecting ducts and distal tubules, without correlation between renal and hepatic severity [3, 5].

Cholangiocarcinoma is the most feared complication. Its prevalence is around 7% in several studies, compared with 0.05% in the general population [5]; in Fahrner's review, incidence ranged from 2.7% to 37.5% depending on the study, with an overall value of 6.6%, and the tumour was usually an incidental finding in the surgical specimen [14]. These figures come from surgical series and adult referral centres, tending to overestimate risk compared with unselected patients.

Isolated CHF, without renal disease or macroscopic biliary dilatation, manifests as PHT with splenomegaly, hypersplenism and varices, with preserved hepatocellular function; the liver is large and abnormally shaped, with the left lobe palpable below the xiphoid process [4, 12]. In the review of 1230 patients with CHF from 155 articles, 64% had ARPKD with CHF and 9.5% isolated CHF; over follow-up of 0–38 years (mean 7.5), PHT sequelae predominated (409 patients, 164 with varices, 74 with bleeding and 81 with a portosystemic shunt), cholangitis occurred in 152 and was fatal in 3 of 23 children infected after kidney transplantation [12, 49]. Excessive alcohol, steatohepatitis, hepatotoxic agents and viral hepatitis may accelerate fibrosis [12].

4. Clinical Recognition and Warning Signs

4.1. Prenatal period

Suspicion usually arises at the second-trimester anomaly scan (18–20 weeks) or in the third trimester, with bilaterally enlarged, symmetrical, hyperechogenic kidneys (“bright”, that is, more echogenic than the liver or spleen), poor corticomedullary differentiation and progressive oligohydramnios [1, 16, 20]. Individual cysts are not usually visible in the fetus; discrete cysts in the second trimester are not characteristic of ARPKD and should prompt consideration of congenital anomalies of the kidney and urinary tract (CAKUT), such as obstruction or multicystic dysplastic kidney [16, 20]. Nevertheless, bilateral cysts of 5–7 mm have been described in 29% of cases, and macrocysts larger than 10 mm suggest multicystic dysplastic kidney [15]. The French PNDS describes hyperechogenicity that is usually heterogeneous and medulla-predominant, sometimes global and homogeneous, with rare macrocysts, visible from 18 weeks of gestation in early forms, and nephromegaly that is sometimes massive, occupying the whole abdominal cavity and increasing fetal abdominal circumference [9]. Hepatic abnormalities are rarely visible on prenatal ultrasound, but fetal MRI may show hepatomegaly, heterogeneous parenchyma and dilated bile ducts, and is useful when oligohydramnios hampers ultrasound [1, 44]. A deepest vertical fluid pocket below 3 cm defines oligohydramnios in the third trimester, and values below 1 cm are associated with perinatal mortality above 95%, particularly between 16 and 24 weeks [44].

Not all fetuses with large echogenic kidneys have ARPKD. In a series of 316 fetuses with hyperechogenic kidneys, ARPKD accounted for 10 cases (3%), fewer than ADPKD (4%), CAKUT (31%) or aneuploidies (13%) [16]. In a large cohort, almost 80% of fetuses with isolated renal hyperechogenicity had normal renal function after birth, and all those with echogenic kidneys and normal amniotic fluid had normal function [30]. Prenatal warning signs are oligohydramnios or anhydramnios, marked renal enlargement (more than 2 standard deviations for gestational age) and the simultaneous presence of the three risk factors described in section 11.1 [6, 54].

4.2. Newborns and infants

In newborns, suspicion is supported by abdominal distension due to massive enlargement of the kidneys and liver, palpable flank masses, respiratory distress with pulmonary hypoplasia on radiography, severe HTN in the first weeks, hyponatraemia and polyuria [1, 30]. Figures 1 to 3 in the open-access review by Liebau (2021) show ultrasound and MRI appearances of the enlarged kidneys (Fig. 1), pathology with ubiquitous cysts and tubular dilatations (Fig. 2) and the distended abdomen with pulmonary hypoplasia on radiography (Fig. 3) [30]. In infants, high-resolution ultrasound shows enlarged kidneys retaining their shape, with innumerable very small cysts (rarely above 1–2 mm) in the cortex and medulla; the term “salt and pepper” is considered outdated by GeneReviews because it described the speckled appearance obtained with older transducers, although the 2026 French PNDS still uses it for the heterogeneous pattern of corticomedullary microcysts [1, 9]. Among neonatal signs, the PNDS adds left ventricular failure secondary to severe HTN, respiratory failure with pulmonary arterial hypertension and failure to thrive; in infants and children, pyelonephritis with suggestive renal imaging, signs of PHT (splenomegaly, collateral circulation, ascites, gastrointestinal bleeding, digital clubbing), cholangitis and family screening [9].

4.3. Children, adolescents and adults

After the first year, suspicion arises with childhood HTN, nephromegaly with macrocysts and reduced corticomedullary differentiation, a urinary concentrating defect, or signs of PHT: splenomegaly, thrombocytopenia, varices or gastrointestinal bleeding [1]. Thrombocytopenia is the best laboratory predictor of PHT and splenomegaly [1, 32]. ARPKD should be considered in the differential diagnosis of fibrocystic hepatorenal disease in young adults, including patients with non-cirrhotic PHT and mildly abnormal kidneys [52]. Normal urinary concentrating ability virtually excludes ARPKD [16]. For the adult nephrologist, the PNDS recommends considering the diagnosis in any kidney disease with corticomedullary cysts, with or without renal enlargement, particularly if there are biochemical (cholestasis, cytolysis, thrombocytopenia) or morphological (fibrosis, hepatosplenomegaly) liver abnormalities or signs of PHT; for the hepatologist, in any CHF or Caroli disease, with renal assessment by imaging and blood and urine tests, even without evident kidney disease [9]. Age or an atypical renal presentation does not exclude the diagnosis (section 3.7) [9].

4.4. Emergency signs

Emergencies include neonatal respiratory distress (pulmonary hypoplasia, pulmonary hypertension, pneumothorax), severe HTN with organ damage, oligoanuria in the first days, fever without a focus in a child with ARPKD (cholangitis until proven otherwise, particularly in the first months after kidney transplantation or under intense immunosuppression), and haematemesis, haematochezia or melaena, requiring immediate assessment in a service with transfusion capability [1, 15]. Parents should know the signs of variceal bleeding and cholangitis (fever, abdominal pain, jaundice) [12]. The PNDS adds left ventricular failure due to severe HTN in the newborn and cholangitis with systemic signs, which requires hospital admission because it may progress to septic shock [9].

4.5. Caroli disease and syndrome

Caroli disease is suspected in an adult or child with recurrent cholangitis, intrahepatic stones or abscesses, and imaging showing segmental saccular or fusiform dilatations of the intrahepatic ducts communicating with the biliary tree [5]. The central dot sign, corresponding to fibrovascular bundles (portal radicles) within the dilated ducts, is typical [1, 5]. Splenomegaly, thrombocytopenia, varices or renal cysts point towards Caroli syndrome [3, 5].

4.6. Newborn screening

There is no biochemical or genetic newborn screening for ARPKD in the sources consulted, and no biochemical assay of fibrocystin deficiency [11]. Early detection depends on obstetric ultrasound and examination of the newborn.

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Clinical criteria

The 2014 consensus considers that bilateral large echogenic kidneys with poor corticomedullary differentiation and coexisting liver disease on conventional ultrasound are usually sufficient for diagnosis, that high-resolution ultrasound increases sensitivity in mild forms and that, in some patients, only serial observation allows the correct diagnosis [15]. The modified Zerres criteria, presented by Liebau (2021), require typical renal imaging and at least one additional criterion, as shown in the table [30].

The same elements (biliary ectasia, parents without cysts, signs of periportal fibrosis, fibrosis on biopsy, affected sibling) were used as inclusion criteria in the tesevatinib trial [55]. The 2014 consensus defines, in known ARPKD, PHT as splenomegaly (spleen palpable more than 2 cm below the left costal margin or more than 1 cm above the upper limit of normal for age) with thrombocytopenia below 150,000/mm³, or known varices, ascites or hepatopulmonary syndrome; and cholangiopathy as non-invasive radiological demonstration of biliary abnormalities, noting that cholangitis may exist with normal imaging [15]. With PHT and transaminases below 2 times the upper limit of normal, CHF is presumed and neither liver biopsy nor extensive investigation of other causes is indicated, but portal vein thrombosis should be excluded by ultrasound [15]. A compatible family history (affected siblings, unaffected parents, consanguinity) supports diagnosis, but its absence does not exclude it [1]. There is no widely accepted diagnostic algorithm [16]. The French PNDS bases diagnosis on clinical and imaging criteria and genetic confirmation, and requires, after renal diagnosis, systematic investigation of liver involvement (blood tests and imaging) with hepatology follow-up even in the absence of manifestations; conversely, any CHF or Caroli disease should prompt renal assessment and genetic testing [9].

5.2. Imaging

Ultrasound is the method of choice prenatally and at all paediatric ages, because it is accessible and requires neither radiation nor sedation [1]. High-resolution ultrasound (5–7 MHz probe or higher) is superior for mild disease: in the NIH cohort, 35% of patients with medullary abnormalities only had normal conventional ultrasound [12, 46]. Initial assessment includes kidney volumes and height-adjusted total kidney volume [1]. Liver assessment looks for hepatomegaly, heterogeneous parenchyma, periportal fibrosis, cysts and biliary dilatation, and signs of PHT: enlargement of the liver or spleen, a dilated portal vein, biphasic or retrograde portal flow, retrograde flow in the splenic vein, portosystemic collaterals and ascites [1]. MRCP is the most sensitive examination for biliary anatomy: dilatations are hypointense on T1 and hyperintense on T2; contrast enhances the central portal radicles, but should be used cautiously when renal function is reduced [1]. Biliary abnormalities, combined with renal imaging, are diagnostic and have largely replaced liver biopsy [1]. ARFI (acoustic radiation force impulse) elastography, with liver and spleen shear-wave velocity, detects fibrosis and PHT, and MR elastography quantifies stiffness [1]. The French PNDS stresses that ultrasound is more sensitive for cysts in infants and young children, especially with high-frequency linear probes; in adults, CT or MRI better visualise microcysts and fusiform medullary dilatations and are the imaging of choice to complement ultrasound; in children, MRI and CT are used only in particular situations, and CT has no role in detection, save in exceptional cases [9]. For the liver, the protocol recommends Doppler ultrasound, MRI with MR cholangiography sequences and liver and spleen elastography (transient elastography or ARFI), and screening for varices by upper gastrointestinal endoscopy or, above 5 years of age, by oesophageal capsule endoscopy [9]. Kidney biopsy should not be performed to diagnose ARPKD [1, 9]. Liver biopsy, when performed, shows abundant bile ducts in a primitive ductal plate configuration, abnormal portal branching, non-inflammatory periportal fibrosis with portal–portal bridges (rather than portal–central, as in cirrhosis), von Meyenburg complexes and normal parenchyma [12]; it rarely diagnoses cholangitis and carries risk with biliary dilatation [15].

5.3. Genetic testing

Molecular diagnosis is established with suggestive findings and biallelic pathogenic or likely pathogenic (P/LP) PKHD1 variants, clinically equivalent terms under ACMG/AMP classification [1]. Two VUS, or a P variant with a VUS, neither establish nor exclude diagnosis [1]. Single-gene testing (PKHD1 sequencing with deletion and duplication analysis) is rarely useful and is not recommended first-line, given the number of phenocopies [1, 15]. The preferred approach is a cystic kidney disease multigene panel including PKHD1, DZIP1L, CYS1, PKD1, PKD2, HNF1B, nephronophthisis genes (NPHP3, TMEM67) and PMM2, or exome or genome sequencing [1, 16]. The 2025 review recommends the panel in early-onset or severe bilateral cystic disease, atypical renal imaging and cystic disease with a negative or unknown family history, always with shared decision-making after genetic counselling [16]. KDIGO 2025 recommends offering genetic testing to children with very early-onset ADPKD or atypical presentation [6]. Testing is strongly advised after birth or post mortem in severe cases, because of its impact on counselling for future pregnancies [16]. The French PNDS proposes immediate testing in six situations: suggestive prenatal forms (fetal nephromegaly, hyperechogenic kidneys, oligohydramnios), where it is a prognostic emergency in severe forms; severe neonatal forms; infantile or juvenile forms with compatible kidney or liver involvement; known family history or screening of relatives; atypical or borderline forms (fibrosing liver diseases, nephronophthisis-type tubulointerstitial nephropathies); and plans for prenatal or preimplantation diagnosis [9]. Confirmation requires two P/LP variants in trans; a single variant or VUS requires multidisciplinary interpretation (geneticist, nephrologist, molecular biologist) and may justify family studies, and the protocol places the diagnostic yield of next-generation sequencing of coding regions at around 80–90%, depending on inclusion criteria [9]. When only one pathogenic variant is identified in an affected patient, it proposes genome sequencing and/or RNA studies of urinary or blood cells to search for rare intronic variants and PKHD1 splicing abnormalities [9].

Source: GeneReviews, table 1 [1].

The table values represent the proportion of known variants detectable by each method, not the clinical sensitivity of the test. The detection rate for two PKHD1 variants in patients with a clinical diagnosis is 80–90%, with at least one variant in up to 95% of families [15, 30]. In ARegPKD, only around 55% of the initial cohort of 665 patients had genetic testing; among those tested, 243 had at least one relevant PKHD1 variant and 31 (5% of the total cohort, 11% of those with an identified variant) had a cause in another gene [29].

Technical limitations have practical consequences. The gene is large and has complex splicing; private variants and novel missense variants generate frequent VUS, and there are no standardised functional assays [1]. Deep intronic variants escape exome sequencing and panels, as demonstrated by the Bedouin founder variant, detected only by genome sequencing and confirmed by RNA analysis [36]. Exome and genome sequencing detect deletions and duplications with lower sensitivity than targeted methods [1]. Apparent homozygosity may represent a deletion of one or more exons on the other allele (false homozygosity) or uniparental isodisomy, requiring parental studies [1]. PKD1 analysis is hampered by highly homologous pseudogenes, and panels should be designed to maximise detection in that region [1]. A negative result or a result with only one variant does not exclude diagnosis [1]. The report should identify the gene, transcript with version (NM_138694.4 for PKHD1), DNA and protein HGVS nomenclature, zygosity, ACMG/AMP class and, for two heterozygous variants, phase (cis or trans) determined by parental studies [1, 40]. The Spanish series from 2026 illustrates the effect of this discipline: of 68 patients referred, 40 had confirmed ARPKD (biallelic P/LP variants or VUS favouring pathogenicity with segregation in trans), 10 had biallelic variants of uncertain pathogenicity and 18 were monoallelic carriers [40].

5.4. Additional assessment and biomarkers

Laboratory assessment includes urea, creatinine and cystatin C, electrolytes and blood gas analysis (hyponatraemia, hyperkalaemia, acidosis), osmolality and proteinuria, full blood count with platelets, prothrombin time, albumin, transaminases, GGT, bile acids and 25-hydroxyvitamin D [1]. The French PNDS adds factor V, INR, bilirubin, alkaline phosphatase and fibrinogen to assess synthesis, cytolysis and cholestasis, and alpha-fetoprotein, carcinoembryonic antigen and CA 19-9 in situations of high neoplastic risk (section 11.2) [9]. The urinary peptidome analysed by capillary electrophoresis coupled to mass spectrometry identified a signature of 77 peptides, 83% of which were collagen fragments: in matched validation, sensitivity was 84.2% (95% CI 60.4–96.6) and specificity 100% (92.0–100), but specificity fell to 76.1% (72.4–79.5) in 591 unmatched CKD samples; it is a research tool without clinical validation [56]. In CHF without a diagnosis, assessment includes a three-generation pedigree, ultrasound of parents and siblings, ophthalmological examination, brain MRI (molar tooth sign), skeletal radiography and echocardiography to identify syndromic ciliopathies [12, 31].

5.5. Prenatal diagnosis

In a fetus with hyperechogenic kidneys, assessment includes detailed anomaly scanning for extrarenal abnormalities, a three-generation pedigree (adult renal disease, neonatal deaths from respiratory or renal failure, teratogens such as angiotensin-converting enzyme inhibitors), parental renal ultrasound to screen for presymptomatic ADPKD or HNF1B-related disease, and amniocentesis for karyotyping and cytomegalovirus PCR testing [15, 20, 44]. Chromosomal microarray is the preferred fetal genetic test because it detects CNVs such as the 17q12 deletion including HNF1B, a relatively frequent cause of echogenic fetal kidneys; if non-diagnostic, a multigene panel or exome sequencing follows [16]. Non-invasive prenatal testing using cell-free fetal DNA is not currently recommended for CNVs or monogenic diseases [16]. Isolated fetal hyperechogenicity and renal enlargement are insufficient for a definitive prenatal diagnosis, and failure to detect two variants does not exclude ARPKD [1]. If a pregnancy with hyperechogenic kidneys ends, fetal and parental DNA should be stored and a complete autopsy or kidney biopsy obtained, enabling subsequent counselling [20, 44]. Reliable early prenatal diagnosis is possible only through molecular analysis when familial variants are known (section 8) [15].

5.6. Caroli disease and syndrome

EASL recommends that discovery of multiple segmental cystic or saccular biliary dilatations should trigger assessment for CHF, to distinguish Caroli disease from Caroli syndrome (level of evidence 4, strong recommendation, 100% consensus) [5]. Computed tomography (CT) and MRI show the saccular or fusiform dilatations and the central dot sign, and contrast demonstrates continuity between the lesions and ducts; MRCP has the greatest accuracy, and confirmation should ideally take place in experienced centres [5]. Fibrosis can be assessed by liver elastography, and renal imaging looks for cysts [5]. Endoscopic retrograde cholangiopancreatography (ERCP) is not recommended for diagnosis or screening because of the risk of biliary infection [5].

The role of genetic testing differs between specialties. EASL considers that, given the size of PKHD1, its complex splicing and interpretative difficulties, genetic testing has little role in routine clinical practice for Caroli disease and syndrome [5]. Paediatric nephrology and KDIGO recommend a genetic panel in early-onset or atypical renal cystic diseases, and Caroli syndrome with renal cysts frequently meets these criteria [6, 16]. The two positions answer different questions: in an adult with isolated Caroli disease, the result rarely changes hepatic management; in a child or young adult with Caroli syndrome and cystic kidneys, molecular diagnosis defines recurrence risk for siblings and offspring and enables prenatal and preimplantation diagnosis.

5.7. Practical algorithm

Sources: [1, 5, 15, 16, 30].

6. Differential Diagnosis

6.1. Enlarged hyperechogenic kidneys in fetuses and children

ARPKD belongs to the congenital hepatorenal fibrocystic syndromes, and the differential diagnosis includes ARPKD forms due to less frequent genes, genetic phenocopies, syndromic ciliopathies and non-genetic causes [1]. The table brings together the main entities.

The most important differential diagnoses, because of frequency and familial implications, are very early-onset ADPKD and HNF1B-related disease. In ARegPKD, among 31 patients with a clinical diagnosis of ARPKD and a cause in another gene, PKD1 explained 17 cases (54.8%), HNF1B and TMEM67 5 each, PKD2 and NPHP3 2 each; no patient with very early-onset ADPKD had a family history of renal cysts [29]. Compared with ARPKD-PKHD1, phenocopies more frequently had prenatal cysts (42.3% versus 21.2%), polyuria (12.9% versus 2.5%) and growth delay, and fewer radiological signs of hepatic fibrosis at diagnosis (13.3% versus 30.4%); only PKD1 phenocopies had height-adjusted kidney length above 10 cm/m after the first year [29]. KDIGO 2025 defines very early-onset ADPKD (VEO-ADPKD) by prenatal diagnosis of enlarged hyperechogenic kidneys (more than 2 standard deviations for gestational age) with oligohydramnios, or by enlarged cystic kidneys between birth and 18 months with hypertension or reduced eGFR; the most severe phenotypes are frequently due to combinations of variants in more than one gene, and ultrasound of parents, and grandparents when parents are under 40 years, forms part of assessment of a child with cysts without a family history [6, 57]. Normal parental ultrasound supports ARPKD, but PKHD1 heterozygotes may have small liver cysts or medullary hyperechogenicity (section 2.7) [1, 42]. Pei ultrasound criteria for ADPKD apply to parents from 15 years of age and not to the affected child [8].

HNF1B-related disease is, in some studies, the most common cause of hyperechogenic fetal kidneys and has better renal survival; it frequently leads to misdiagnosis as ARPKD [1, 15, 30]. In young adults, ARPKD calcifications may make distinction from medullary sponge kidney difficult [1]; the French PNDS stresses that Cacchi-Ricci disease is never associated with liver disease, so precalyceal ectasia with liver disease points towards ARPKD [9]. The same protocol includes in the differential diagnosis bilateral cystic renal dysplasias associated with PAX2, in addition to HNF1B, glutaric aciduria type II and carnitine palmitoyltransferase II deficiency, the PMM2 promoter variant (with glomerular cysts and sometimes liver cysts) and biallelic MPI variants in CHF; in adults, because kidneys may be normal-sized or small, any other cause of advanced CKD enters the differential diagnosis [9].

6.2. Congenital hepatic fibrosis and biliary dilatations

A few small liver cysts may be a normal variant that increases with age [12]. Caroli disease is distinguished from the syndrome by the absence of CHF, and the syndrome from ARPKD only by the relative intensity of renal involvement, because they share the same spectrum [5]. Biliary dilatation without CHF and without renal disease is, by definition, Caroli disease; CHF without dilatation or renal disease corresponds to isolated CHF, whose genetic cause remains undefined in most cases [4, 12].

7. Treatments and Risk Management Strategies

7.1. Current guidelines and divergences

There is no curative treatment or approved disease-modifying therapy for ARPKD; management is supportive and interdisciplinary [1, 16]. The 2014 international consensus (Guay-Woodford and colleagues) remains the reference document for paediatric diagnosis and management, complemented by the 2018 European recommendation for perinatal cystic kidney diseases (Gimpel and colleagues) [15, 61]. The updated ARPKD clinical practice recommendations, prepared by the European Society for Paediatric Nephrology (ESPN) and ERKNet, had been submitted for publication but were not published at the date of this review [62]. The table records the normative sources used in this article.

Four divergences influence decisions. First, TIPS: the retired 2014 chapter allowed it only in emergencies, whereas the 2024 GeneReviews considers it an option, particularly with good residual renal function; the current position takes precedence [1, 12]. Second, ursodeoxycholic acid: EASL recommends it in Caroli disease and syndrome, even in asymptomatic patients, whereas the 2014 paediatric consensus and the 2024 GeneReviews do not recommend it as a choleretic in ARPKD [1, 5, 15]. Third, genetic testing in Caroli disease (section 5.6). Fourth, classification of biallelic PKD1 (section 1.2).

The 2026 French PNDS is the most recent normative document and the only one covering adults, and it introduces further divergences [9]. For ursodeoxycholic acid, it proposes 13–15 mg/kg/day in 1–2 doses in the chronic cholestasis of ARPKD, a position close to EASL and contrary to the 2014 consensus and GeneReviews; the French SmPC for ursodeoxycholic acid indicates 20–30 mg/kg/day in genetic cholestasis and contraindicates it in acute cholangitis [5, 9, 15, 64]. For neonatal HTN, it prefers beta-blockers (acebutolol, propranolol) and amlodipine as first line, allowing ACE inhibitors introduced very gradually, whereas the 2014 consensus favours renin–angiotensin system inhibitors at all ages; the acebutolol SmPC contraindicates it in metabolic acidosis, present in more than half of neonatal and infantile forms [9, 15, 65]. For dialysis, it accepts peritoneal dialysis or haemodialysis and considers peritoneal dialysis difficult or contraindicated with lack of abdominal space or intra-abdominal hypertension due to nephromegaly, whereas the consensus and GeneReviews treat peritoneal dialysis as the modality of choice [1, 9, 15]. For liver surveillance, it requires Doppler ultrasound at least annually and screening for PHT every 6–12 months, compared with 2–3 years in the 2014 consensus with negative ultrasound (section 11.2) [9, 15]. For endoscopic screening in adults, it waives endoscopy with elastography up to 15 kPa and platelets of at least 150,000/mm³; Baveno VII uses this pair of values to exclude clinically significant PHT in compensated cirrhosis, indicates endoscopic screening from 20 kPa or with platelets up to 150 × 10⁹/L in patients who cannot receive beta-blockers and considers that non-invasive criteria do not apply to porto-sinusoidal vascular disease; CHF is not cirrhosis, and these thresholds have not been validated in it [9, 63]. For prenatal diagnosis, it accepts non-invasive prenatal diagnosis of the paternal allele in maternal plasma and indirect diagnosis by polymorphic markers when only one variant is known, whereas the 2025 review does not recommend non-invasive testing in monogenic diseases and the 2014 consensus abandoned linkage analysis as the method of choice (sections 8.1 and 8.3) [9, 15, 16]. For carrier counselling, it does not routinely propose testing of the unrelated partner of a carrier, whereas KDIGO 2025 recommends offering it when there is reproductive planning [6, 9]. In adults, it sets the systolic pressure target below 130 mmHg and, if tolerated, 120 mmHg, as in proteinuric nephropathies of young adults [9].

7.2. Initial assessment after diagnosis

GeneReviews recommends, in the perinatal and infantile periods, delivery in a centre with neonatal intensive care and paediatric nephrology; respiratory assessment (examination, oximetry, blood gas analysis, chest radiograph, echocardiogram if indicated); regular blood pressure measurement, with parental training for home measurement if elevated; renal function, electrolytes and blood gas analysis; urinary concentration and proteinuria; testing for urinary tract infection if fever is present; clinical and ultrasound assessment of volume status; renal ultrasound with volumes; abdominal ultrasound for CHF and PHT, with indications for elastography and MRCP; nutritional assessment with a dietitian; transaminases, bile acids, albumin, coagulation, 25-hydroxyvitamin D and other fat-soluble vitamins and full blood count; initial cranial ultrasound through the fontanelle; and genetics, psychology and social work consultations [1]. When decisions on palliative approaches or limitation of therapeutic modalities may arise, interdisciplinary shared decision-making should begin early [1]. After the first year, home blood pressure measurement to distinguish sustained from white-coat hypertension, neurodevelopmental assessment with electroencephalography, cranial ultrasound or MRI when indicated, and respiratory function tests when indicated are added [1].

7.3. Perinatal period and respiratory support

The 2014 consensus recommends delivery in a centre with a level IV neonatal intensive care unit, multidisciplinary prenatal consultation (maternal–fetal medicine, neonatology, paediatric nephrology) and a delivery plan anticipating caesarean section for abdominal dystocia caused by nephromegaly [15]. Intervention intensity should consider family preferences, and the decision whether to offer dialysis is shared between doctors and parents [15, 30]. Serial amnioinfusions lack sufficient data for general recommendation; in bilateral renal agenesis they attenuated lethal pulmonary hypoplasia with poor long-term outcomes, and in ARPKD there is no evidence of benefit because nephromegaly independently contributes to thoracic restriction [1, 16, 61]. Antenatal corticosteroids may help lung maturation in late-preterm fetuses [1, 30].

The French PNDS recommends that the couple be seen at a multidisciplinary prenatal diagnosis centre (CPDPN) with a paediatric nephrology opinion, delivery at a neonatal centre able to treat pulmonary hypoplasia and renal failure and, given the risk of pulmonary hypoplasia, prevention of respiratory syncytial virus (RSV) infection, by maternal vaccination (Abrysvo®) between 32 and 36 weeks of gestation or by nirsevimab (Beyfortus®) in the infant [9]. The European SmPC for Abrysvo® provides for a single 0.5 mL intramuscular dose in the deltoid between 24 and 36 weeks of gestation; the PNDS window of 32–36 weeks is narrower [9, 66]. The European SmPC for Beyfortus® provides, in the first RSV season, a single intramuscular dose of 50 mg below 5 kg and 100 mg from 5 kg, from birth in those born during the season; in children up to 24 months who remain vulnerable in their second season, 200 mg (two 100 mg injections); dosing between 1.0 and 1.6 kg is extrapolated and there are no clinical data below 32 weeks of postmenstrual age [67]. In the first days, the protocol recommends monitoring blood pressure, feeding and weight gain; measuring the kidneys (long axis and volume) and describing renal echostructure, the liver, spleen and portal system; checking renal function and electrolytes; and holding paediatric nephrology and hepatology consultations while still in the maternity unit [9]. Infants with significant renal failure or massive nephromegaly may need nutritional support [9].

In newborns, hypoxaemic or hypercapnic respiratory failure is treated with non-invasive or invasive support; high-frequency ventilation is used in mild hypoplasia, persistent pulmonary hypertension may respond to inhaled nitric oxide, and extracorporeal membrane oxygenation (ECMO) may be appropriate in selected cases where respiratory failure is not due to lethal hypoplasia [1, 15]. Pneumothorax and pulmonary hypertension are treated as indicated [1]. Prenatal prediction of lethal hypoplasia is imprecise: an MRI lung volume below 0.90 of expected after 26 weeks had sensitivity of 77.8% and specificity of 95% for non-survival in 46 fetuses with severe genitourinary abnormalities, and the consensus concludes that prenatal imaging does not accurately predict lethal pulmonary hypoplasia [15].

7.4. Kidney: dialysis, nephrectomy and chronic kidney disease

Oliguria or anuria in the first days may require dialysis, and peritoneal dialysis is the modality of choice [1, 15, 30]. The French PNDS considers both techniques, peritoneal dialysis and haemodialysis, possible and decides case by case, because peritoneal dialysis may be complex or contraindicated owing to lack of space or intra-abdominal hypertension due to nephromegaly [9]. In the international paediatric peritoneal dialysis registry (IPPN), 79 children with ARPKD had overall survival similar to children with congenital nephrotic syndrome and lower than children with CAKUT, without a difference in technique survival; ARPKD required only minor adaptations, and some patients had greater glucose-driven ultrafiltration, possibly related to PHT [30]. KRT initiated in the neonatal period or first year is associated with survival of around 80% at 2 years in an international cohort and 80–85% at 3 years in a North American cohort [30]. The decision to start, withhold or discontinue KRT is shared with the family and multidisciplinary team, potentially within a formal ethical framework [30].

Unilateral or bilateral nephrectomy is advocated by some when massive kidneys prevent diaphragmatic excursion, cause severe feeding intolerance or uncontrollable hypertension [1]. The 2014 consensus emphasises that the rationale for unilateral nephrectomy rests on a few small nutritional studies, that there is no evidence of respiratory improvement after nephrectomy and that there is no evidence for nephrectomy for severe hypertension [15]. The 2025 review describes its utility as highly controversial [16]. Risks are concrete. Bilateral nephrectomy eliminates the possibility of native function maturing and imposes lifelong KRT; after unilateral nephrectomy, the contralateral kidney may enlarge rapidly and markedly, sometimes requiring a second nephrectomy soon afterwards [1, 30]. In ARegPKD (504 patients), very early bilateral nephrectomy (up to 3 months; n=19) was associated with more severe neurological complications than all control groups, and very early bilateral nephrectomy and documented severe hypotensive episodes were independent risk factors [50]. Indications should be approached very cautiously in the first months, when cardiovascular autonomic control is still maturing, especially if residual renal function exists, and severe hypotension should be avoided [1, 50].

General paediatric recommendations for anaemia, acidosis and mineral and bone disorder apply in CKD, and growth hormone follows paediatric CKD recommendations [1]. The French PNDS details nephroprotection: controlled protein intake, salt adapted to needs and blood pressure, generous hydration, avoidance of tobacco, iodinated contrast and non-steroidal anti-inflammatory drugs, preservation of the antecubital veins (for future vascular access) and strict control of blood pressure and proteinuria, with a target urine protein/creatinine ratio below 0.5 g/g (50 mg/mmol), using ACE inhibitors or ARBs as first line [9]. CKD complications are treated with iron and erythropoietin for anaemia, native and active vitamin D (alfacalcidol, in advanced renal failure) and phosphate binders according to calcium–phosphate metabolism, enteral nutrition when needed in infants and growth hormone for growth failure with CKD [9]. The French SmPC for alfacalcidol (Un-Alfa®) indicates, in renal osteodystrophy, a starting dose of 0.05 µg/kg/day below 20 kg and 0.5–1 µg/day in adults and children above 20 kg, with a therapeutic dose of 1 µg/day below 20 kg and 1–2 µg/day in adults, adjusted to the calcium–phosphate product, with no maximum dose set in the SmPC (hypercalcaemia requires discontinuation) [68]. The European SmPC for somatropin (Omnitrope®) provides, for growth failure in chronic renal insufficiency, 0.045–0.050 mg/kg/day (1.4 mg/m²/day), with renal function below 50% of normal and one year of growth observation before starting [69]. Urinary tract infections should be recognised and treated early, avoiding aminoglycosides whenever possible [1]. In kidney failure, kidney transplantation is the treatment of choice and peritoneal dialysis the preferred chronic dialysis method; native nephrectomy before or simultaneously with transplantation may be necessary [1]. According to the French REIN registry cited by the PNDS, pre-emptive transplantation (before dialysis) accounts for 12% of annual transplants in France, but for 54% in ARPKD, 36% of them from a living related donor; the disease does not recur in the graft, and it is not uncommon for the same patient to need several kidney transplants, especially in childhood-onset forms [9].

7.5. Arterial hypertension

Early recognition is essential, and hypertension in the first months may require several drugs [1, 30]. The 2014 consensus considers angiotensin-converting enzyme inhibitors (ACE inhibitors) or angiotensin receptor blockers (ARBs) the mainstays of treatment and advises against combining ACE inhibitors with ARBs because of more adverse effects without clear benefit [15]. Liebau (2021) considers renin–angiotensin system inhibitors the treatment of choice, based on experimental observations and pathophysiology [30]. A blood pressure target at the 50th percentile for age, sex and height, derived from the ESCAPE trial in paediatric CKD and KDIGO CKD recommendations, appears reasonable, although not established specifically for ARPKD [1, 15]. KDIGO 2025 for paediatric ADPKD sets a target at or below the 50th percentile, or at or below 110/70 mmHg in adolescents, recommends annual 24-hour ambulatory blood pressure monitoring (ABPM) from 5 years and ACE inhibitors or ARBs as first-line treatment [6, 57].

In newborns and young infants, the recommendation for ACE inhibitors conflicts with the label and neonatal practice. The American label for enalapril oral solution (Epaned®) does not recommend it in newborns (up to 1 month), premature infants who have not reached 44 weeks postconceptional age or with GFR below 30 mL/min/1.73 m² [70], and lisinopril oral solution (Qbrelis®) is not recommended below 6 years or with GFR below 30 [71]. Starr and Flynn usually avoid ACE inhibitors until 44 weeks corrected postmenstrual age, because captopril may cause excessive blood pressure falls, particularly in premature infants, and hyperkalaemia, acute kidney injury and severe hypotension have been described; they frequently start with isradipine or amlodipine [72]. In practice, treatment of severe neonatal hypertension in ARPKD is off-label for most drugs and should be managed by paediatric nephrology. In severe hypertension with organ damage, intravenous infusions are used, avoiding rapid falls because of the risk of cerebral ischaemia or haemorrhage, particularly in premature infants [72]. In patients with PHT, a beta-blocker may be chosen, as illustrated by the case in section 10, where propranolol was preferred to ACE inhibitors or ARBs [35].

Neonatal doses come from table 5 of Starr and Flynn, a review without ARPKD trials, rather than labels [72].

The French PNDS sets the target around the 50th percentile for height and keeps renin–angiotensin–aldosterone system blockers as first line [9]. In newborns, in whom HTN may be severe and difficult to control, it prefers as first line beta-blockers (acebutolol oral suspension, Sectral®; propranolol oral suspension, in France under compassionate access authorisation) and amlodipine (Amlor®), allowing ACE inhibitors introduced very gradually and cautiously [9]. In children, it starts an ACE inhibitor at very gradually increasing doses and, in case of intolerance (mainly cough and headache) or insufficient efficacy, replaces it with an ARB; HTN may require multidrug therapy [9]. In adults, it recommends an ACE inhibitor titrated to the maximum tolerated dose, with a systolic pressure target below 130 mmHg and, if tolerated, 120 mmHg [9]. At all ages, blood pressure (hypotension, asthenia, malaise), potassium and creatinine are monitored, and renin–angiotensin–aldosterone system blockers are not used with hypovolaemia, previous allergy or previous bradykinin-mediated angioedema on these drugs [9]. The table compares the PNDS doses, which are the reference, with the French SmPCs.

The PNDS notes that several of these drugs are used outside the terms of the French marketing authorisation (AMM) [9]. There are no trials comparing antihypertensive classes in ARPKD, and optimal HTN treatment is undefined [16]. Some authors suggest that HTN may improve as renal function declines [30].

7.6. Hyponatraemia, hydration, nutrition and growth

In hyponatraemia with euvolaemia or hypervolaemia, the consensus recommends minimising fluid intake without compromising nutrition, for example by concentrating feeds, and tolerating low-normal serum sodium; sodium supplementation worsens HTN and is justified only with hypovolaemia; vasopressin receptor antagonists (vaptans) are not recommended because the mechanism of hyponatraemia is uncertain [15, 30]. The French PNDS agrees: in hyponatraemia in the first weeks, even without a fall in glomerular filtration, it limits fluid intake by concentrating milk feeds (sometimes poorly tolerated gastrointestinally) in the euvolaemic or hypervolaemic child, and gives sodium chloride only with hypovolaemia or documented salt wasting, in compounded capsules (the only form available in France) with the dose adjusted to laboratory results [9]. In older children, the concentrating defect requires generous hydration and attention to extrarenal losses (diarrhoea, vomiting, intense heat), and the protocol recommends rotavirus vaccination [9]. For metabolic acidosis, it proposes sodium bicarbonate in compounded capsules or as Bicafres® 1000 mg, off-label below 14 years of age [9]. The French SmPC for Bicafres® (gastro-resistant tablet) indicates, in metabolic acidosis of CKD in adults, 2 to 3 tablets daily in divided doses, adjusted to maintain plasma bicarbonate at 22 mmol/L or above, with efficacy demonstrated up to 8 g/day, and does not establish safety or efficacy below 14 years; in children, the dose of the compounded preparation is titrated to the same plasma bicarbonate target, by extrapolation [79]. Dehydration, poor weight gain and poor growth should be recognised and treated, with supplementary feeding or fluids via a nasogastric tube or gastrostomy; in the first months caloric intake depends on fluid intake, requiring dietitian follow-up [1].

Gastrostomy is considered relatively contraindicated in ARPKD because of the risk of splenic injury with splenomegaly and stomal varices [30]. In an international survey of around 200 participants from 39 countries, with retrospective data on 38 patients with ARPKD, most supported gastrostomy when necessary, complications were comparable to those in patients without ARPKD and there were very few bleeding complications, even with PHT [30, 80]. Early involvement of paediatric gastroenterology and surgery is recommended [30].

7.7. Hepatobiliary disease in ARPKD: cholangitis and portal hypertension

Hepatobiliary management focuses on ascending cholangitis and PHT [1]. Cholangitis is underdiagnosed: it should be considered in any child with ARPKD and unexplained fever, with the highest suspicion in the first months after kidney transplantation or with high immunosuppression [15]. Alkaline phosphatase and GGT may rise in acute cholangitis, but normal values do not exclude it; imaging may support diagnosis, and liver biopsy is rarely diagnostic [1, 15]. Treatment frequently requires prolonged intravenous antibiotics [15]. The French PNDS sets out the regimen: systematic hospital admission in the presence of systemic signs; amoxicillin–clavulanic acid as the most usual choice, for 7–14 days; fluoroquinolones reserved for infections with bacteriological documentation and not recommended first line in children; piperacillin–tazobactam or a third-generation cephalosporin in severe or nosocomial infections; and systematic morphological assessment looking for biliary stricture or abscess, with multidisciplinary discussion of endoscopic treatment if there is a stricture [9]. Doses come from the French SmPCs, by extrapolation from complicated intra-abdominal infections, because the PNDS does not give them (following table). Recurrent cholangitis causes biliary damage and cholestasis and may indicate liver transplantation [1, 15]. Routine antibiotic prophylaxis is not indicated; it may be considered for 6–12 weeks after a cholangitis episode, immediately after transplantation or with intensified immunosuppression, without a general recommendation or defined regimen [1, 15]; the PNDS considers the data insufficient and decides case by case [9]. Cholestasis causes malabsorption of fats and fat-soluble vitamins, treated according to clinical and biochemical findings [1]. In chronic cholestasis, the French PNDS proposes supplementation with vitamins A, D, E and K (for example DEKAs Essential®, Vedrop® or injectable forms), caloric intake increased up to 150% of recommended requirements and adjustment of medium-chain triglycerides; for cholestatic pruritus, off-label rifampicin and, as second line after discussion at a hepatology centre, ileal bile acid transporter inhibitors (IBATi) [9]. In hepatic encephalopathy, investigated by ammonia, electroencephalography and brain imaging to exclude other diagnoses, treatment combines lactulose, rifaximin and limitation of protein intake, and recurrent episodes with loss of quality of life may indicate liver transplantation [9].

Ursodeoxycholic acid is contraindicated in acute cholangitis according to the SmPC, so the PNDS reserves it for chronic cholestasis [9, 64].

Ursodeoxycholic acid as a choleretic cannot be recommended in ARPKD, according to the 2014 consensus and the 2024 GeneReviews: its theoretical effect on chronic cholestasis does not apply to most CHF cases, and randomised placebo-controlled trials in adults with sclerosing cholangitis identified toxicity of unknown relevance to CHF [1, 15]. This position differs from the EASL recommendation for Caroli disease and syndrome (section 7.9). The two positions apply to different populations (children with ARPKD and CHF on the one hand; adults with Caroli disease or syndrome and cholestasis or stones on the other), and neither rests on trials.

The French PNDS requires regular, systematic screening for signs of PHT every 6–12 months, by ultrasound, blood tests and elastography (transient or ARFI), and upper gastrointestinal endoscopy or oesophageal capsule endoscopy according to the results; in adults it waives endoscopy with stiffness up to 15 kPa and platelets of at least 150,000/mm³ and, in other cases, repeats it every 2 years without varices and annually with grade 1 varices (divergence from Baveno VII in section 7.1) [9, 63]. Management of complications follows the Baveno recommendations used in other chronic liver diseases: a non-selective beta-blocker in the presence of oesophageal varices (for example carvedilol, off-label in paediatrics); band ligation or sclerotherapy of grade II or III varices in children and, in adults, when the beta-blocker is contraindicated or not tolerated; and, for rebleeding despite medical and endoscopic treatment, TIPS in adults or children or, in children in whom TIPS is not feasible and bleeding risk is high, a surgical portosystemic shunt (portocaval or splenorenal) after the opinion of a specialist centre, very effective against bleeding but without effect on, or with worsening of, the other consequences of PHT (encephalopathy, pulmonary arterial hypertension), which requires Doppler surveillance for thrombosis or stenosis [9]. Baveno VII prefers carvedilol to traditional beta-blockers in compensated cirrhosis and recommends, in porto-sinusoidal vascular disease, management similar to that of cirrhosis for lack of specific data [63]. None of these documents gives doses; the French SmPC for propranolol provides, for prevention of variceal bleeding in adults with cirrhosis, a usual starting dose of 160 mg/day, adapted to reduce heart rate by around 25%, with no maximum dose specific to this indication, and that for carvedilol has only a heart failure indication in adults, starting at 3.125 mg; use in children with PHT is off-label extrapolation [78, 90]. Pulmonary arterial hypertension and hepatopulmonary syndrome are serious complications that contraindicate a surgical shunt and may indicate liver transplantation; the former may cause exertional malaise, asthma-like dyspnoea or cyanosis and requires specialist cardiology follow-up, and the latter exertional fatigue, desaturation with orthodeoxia and digital clubbing [9]. Screening is by supine and standing oximetry at every visit (ideally every 6–12 months) and annual echocardiography; a difference of 2–4% between the two positions leads to contrast echocardiography, which cannot be interpreted if there is an intracardiac shunt, and to scintigraphy with labelled macroaggregated albumin, which quantifies the shunt [9]. Variceal prevention and treatment follow the principles used for other causes of paediatric PHT and are decided by paediatric hepatology: endoscopic band ligation or sclerotherapy, according to age and local procedures [1, 15]. Sclerotherapy is frequently used in small children because ligation devices do not fit smaller-calibre endoscopes [12]. Ligation was effective in the Polish series, and surgery, including liver transplantation, was rarely necessary [48]. There is no formal paediatric PHT guideline equivalent to the adult Baveno VII consensus; a 2026 European survey of paediatric hepatology centres found consensus on definitions and the use of endoscopy for primary and secondary prophylaxis, but notable divergence in non-selective beta-blocker use, elastography without standardised paediatric thresholds and divergent practices in indications for shunts and interventional radiology [18]. Haematemesis, haematochezia or melaena require immediate assessment in a service with transfusion capability [15]. Hypersplenism rarely has consequences: leukopenia does not typically increase infection risk, isolated splenectomy is rarely indicated and is considered contraindicated by the historical chapter because it does not treat PHT and frequently worsens it, and restricting contact sports with a palpable spleen is controversial and not evidence-based [12, 15].

Progressive PHT may require an interventional or surgical portosystemic shunt [1]. In a Belgian series of 5 children with CHF and ARPKD and severe PHT treated with TIPS using an expanded polytetrafluoroethylene-covered prosthesis, followed for a mean of 7 years and 2 months, oesophageal varices and ascites disappeared in all, the reduction in spleen size and platelet increase were not significant, hepatic and renal function remained stable and there was no encephalopathy; endotipsitis was suspected in 2 and in-stent stenosis occurred in 2 [91]. Based on this series, the 2024 GeneReviews considers TIPS an option, particularly in patients with good residual renal function [1]. The retired 2014 chapter allowed TIPS only in emergencies because of occlusion rates, repeated revisions and potentially nephrotoxic iodinated contrast, and preferred surgical shunts for recurrent bleeding, particularly from gastric varices [12]. The current position takes precedence, but caveats remain: contrast requires caution in CKD, patency requires surveillance, and recurrent hepatic encephalopathy and death after a portocaval shunt have been described in 2 patients with renal failure [1, 15]. Intractable PHT or severe disease of both organs justifies liver or combined transplantation [1].

7.8. Kidney, liver and combined transplantation

Isolated kidney transplantation is the usual treatment for kidney failure, with graft survival comparable to other renal diseases [1, 92]. Cholangitis is the main threat after transplantation: 64–80% of deaths after kidney transplantation in ARPKD were attributed to cholangitis or sepsis [1]. In an American series of 3 children with mild to moderate liver disease (fibrosis and biliary dilatation without PHT, cholangitis or bleeding), isolated kidney transplantation had patient and graft survival of 100% at 2 years, without cholangitis [93]. In the French series of 50 children, 9 isolated kidney transplants proceeded without hepatic complications, and indications for liver or combined transplantation were limited to recurrent cholangitis (one combined transplant at 20 years) and uncontrollable PHT (one isolated liver transplant at 15 years for variceal bleeding despite a shunt) [94].

ARPKD is, with primary hyperoxaluria type 1, one of the two main indications for paediatric combined liver–kidney transplantation, without definitive recommendations [30, 95]. The liver from the same donor immunologically protects the kidney, with less rejection, but only when transplantation is simultaneous [1, 95]. In the ESPN/ERA-EDTA registry (36 European countries, 1995–2012), 202 patients with ARPKD up to 19 years of age underwent transplantation after a median of 0.4 years of dialysis, at a median age of 9 years: 32 (15.8%) received a combined transplant and 163 (80.7%) an isolated kidney transplant [92]. Patient survival at 5 years was 97.4% (95% CI 94.9–100) after kidney transplantation and 87.0% (75.8–99.8) after combined transplantation, with a 6.7-fold higher risk of death (95% CI 1.8–25.4; p=0.005) after combined transplantation; death-censored kidney graft survival was similar (92.1% versus 85.9%; p=0.4) [92]. The registry lacked hepatic data for isolated kidney recipients, precluding comparison of patients with equivalent liver disease [92]. Conversely, data from the American UNOS (United Network for Organ Sharing) database showed superior kidney graft survival at 5 years after combined transplantation from 6 months onwards [1]. In the Essen series, 6 children with ARPKD received simultaneous combined transplantation, and cumulative 10-year survival for the total of 21 combined or sequential transplants was 78.4% [96]. The choice between simultaneous and sequential transplantation depends on the dominant organ, pathophysiology and regional allocation systems [97]. GeneReviews proposes considering combined transplantation when significant CKD is accompanied by substantial hepatic involvement (severe PHT or recurrent cholangitis), with an individual decision because clear criteria are lacking [1].

The French PNDS cites, from the ESPN/ERA-EDTA registry, 5-year kidney graft survival of 85.9% after isolated kidney transplantation in ARPKD, similar to that in CAKUT (86.8%), and of 92.1% after combined transplantation, similar to that in hyperoxaluria (83.7%) [9, 92]. It lists as indications for liver transplantation, which are rare, recurrent cholangitis, PHT uncontrolled medically with repeated gastrointestinal bleeding, pulmonary arterial hypertension, hepatopulmonary syndrome and, exceptionally, hepatocellular failure or encephalopathy; liver transplantation may be isolated, before, after or simultaneous with kidney transplantation, depending on renal severity at the time [9]. Some paediatric teams discuss pre-emptive liver transplantation combined with kidney transplantation when there is severe biliary disease, because of the risk of repeated cholangitis under immunosuppression, with morbidity, mortality and kidney graft loss due to sepsis [9]. After isolated liver transplantation, renal function usually falls by around 25 mL/min/1.73 m² in the first year, and calcineurin inhibitors are nephrotoxic, so pre-transplant assessment should include systematic screening for kidney disease and measurement of GFR [9].

7.9. Caroli disease and syndrome

Initial treatment is symptomatic: cholangitis is treated with antibiotics or ERCP, abscesses are drained radiologically, endoscopically or surgically, and cholestasis and stones are treated with ursodeoxycholic acid; osteoporosis should be monitored because of cholestasis-associated vitamin D deficiency [5]. EASL adds, in the supporting text and without a separate graded recommendation, that patients with Caroli disease or syndrome should receive ursodeoxycholic acid at 13–15 mg/kg/day, similar to the dose used in primary biliary cholangitis, even when asymptomatic [5]. This dose is based on analogy and use in this indication is off-label. The divergence from paediatric ARPKD sources was described in section 7.7.

When symptomatic treatment fails, liver resection is the first surgical step in symptomatic monolobar disease without fibrosis or PHT [5]. EASL recommends considering referral for liver transplantation in patients with recurrent cholangitis and bilobar involvement, or monolobar involvement associated with hepatic fibrosis or PHT, when resection has been excluded (level of evidence 3, strong recommendation, 96% consensus) [5]. Survival at 5 years after transplantation ranges from 62% to 100% for patients and 72% to 100% for grafts; cholangiocarcinoma is a contraindication, with a possible exception for tumours up to 2 cm in cirrhosis; there is no consensus on listing timing, which should occur outside acute cholangitis and before physical deterioration [5]. In Caroli syndrome with kidney failure, combined transplantation may have advantages over sequential transplantation, because immunosuppression after kidney transplantation increases the risks of cholangiocarcinoma and cholangitis and the liver confers immunoprotection to the kidney only when transplantation is simultaneous [5].

The series by Ulrich and colleagues (Charité, Berlin, 1989–2002) included 40 patients with Caroli disease or syndrome operated on after conservative treatment failed, with a median age of 49 years, 80% with monolobar disease (left:right 2.6:1) and a median of 26.5 months between symptoms and surgery: 33 (82.5%) underwent liver resection, 4 (10%) transplantation for diffuse disease and 3 (7.5%) isolated biliodigestive anastomosis; minor complications occurred in 37.5% and 2 transplant recipients were reoperated on for bleeding; over a median follow-up of 86.5 months there was significant improvement in symptoms, biliary complications and antibiotic use, and 3 unrelated deaths [98]. In a Chinese series of 20 transplant recipients with Caroli syndrome (Beijing, 2014–2024; 1.23% of 1623 liver transplants; 3–56 years), 16 had polycystic kidney disease; the median interval between the first major complication and transplantation was 12 months, 6 patients developed chronic renal failure during follow-up, and survival at 1, 3 and 5 years was 90%, 85% and 78.5%, with 4 deaths (20%) [99]. The authors recommend attention to renal preservation and individualised immunosuppression [99]. Fahrner's review concludes, as the authors' opinion, that definitive surgical treatment should be offered early, because 1-year survival after cholangiocarcinoma was 36% and recurrence reaches 75% [14]; EASL notes that this recurrence value probably overestimates the true figure [5]. For extrahepatic cysts with recurrent infection or stones, the historical chapter described common bile duct excision with Roux-en-Y hepaticojejunostomy [12].

7.10. General measures, agents to avoid, pregnancy and perioperative context

Sympathomimetics in HTN, nephrotoxic agents (non-steroidal anti-inflammatory drugs and aminoglycosides) unless clinically justified, a high-salt diet, smoking and obesity should be avoided, and hepatotoxic agents (paracetamol above 30 mg/kg/day, herbal supplements, alcohol) minimised [1]. Caffeine, methylxanthines such as theophylline and calcium channel blockers worsened cysts in cellular and animal models, but have not been rigorously studied in ARPKD and clinical relevance is uncertain [1]. Non-steroidal anti-inflammatory drugs should also be avoided in patients with varices because of bleeding risk and platelet dysfunction [12]. The French PNDS assigns the reinforcement of these rules to the attending physician: an adapted diet, contraindication of non-steroidal anti-inflammatory drugs, venepuncture away from the antecubital fossa and caution with iodinated contrast [9].

With severe PHT and splenic dysfunction, vaccination against encapsulated bacteria (pneumococcus, Haemophilus influenzae type b, meningococcus) is recommended, in addition to the routine schedule for healthy and premature children [1]. The historical CHF chapter mentions the Centers for Disease Control and Prevention (CDC) recommendation for hepatitis A and B vaccination in chronic liver disease [12].

Very little is known about pregnancy in women with ARPKD and there are no comprehensive recommendations [1]. CKD generally increases the risk of renal decline, pre-eclampsia and poor fetal outcomes, and PHT may increase variceal bleeding risk, with sparse data in non-cirrhotic PHT [1]. In an NIH series of 12 adult women, 4 had pregnancies with live births: one, diagnosed at 6 months, had transient worsening of creatinine from 1.15 to 1.78 mg/dL at 21 years; the other three were uncomplicated, and two of these women were diagnosed only after delivery (at 23 and 46 years) [53]. Individual renal and hepatic counselling before pregnancy, more frequent nephrology and hepatology consultations and follow-up in a high-risk pregnancy clinic are recommended [1]. ACE inhibitors and ARBs have fetal toxicity and should be discontinued as soon as pregnancy is detected [70]. The sources consulted contain no data on breastfeeding in ARPKD.

The French PNDS includes pregnancy follow-up and perioperative management among the objectives of follow-up, without specific recommendations [9]. In the perioperative context, the critical points are bleeding risk associated with thrombocytopenia and varices, iodinated contrast nephrotoxicity, hypotension in infants (associated with neurological injury after early bilateral nephrectomy) and cholangitis risk under immunosuppression after transplantation [1, 15, 50]. Gastrostomy requires prior assessment of the spleen and varices [30].

7.11. Transition to adult care

The 2025 review recommends introducing a transition plan at 12–14 years, gradually, with paediatric and adult coordinators, joint clinics, coordination with hepatology, access to support groups, insurance education, genetic counselling on offspring and confirmation that the patient has established care in adult nephrology; transfer should not occur during a crisis [16]. Adult nephrologists have less ARPKD experience, and hepatology follow-up should continue [16]. The French PNDS recommends anticipating the move to adult care with therapeutic education of the young person and joint or alternating paediatric and adult nephrology and hepatology consultations [9]. Therapeutic education covers the disease, treatments, nephroprotection and warning signs of renal and hepatic complications; after transition, it includes cofactors that worsen liver disease (viruses, alcohol, excess weight, diabetes) [9]. It recommends psychological support for the child, parents and siblings from the time the diagnosis is disclosed, written information and referral to patient associations, and a team coordinated by nephrologists and hepatologists, preferably with a nurse coordinator, including paediatric urology, anaesthesia and radiology, genetics, psychology, social work, dietetics, advanced practice nursing, educators and teachers and, as appropriate, psychomotor therapy, physiotherapy and speech and language therapy [9]. The 2014 consensus recommends a team including a psychologist or neuropsychologist, systematic developmental surveillance annually or biennially and neuropsychological assessment for learning, attention or behavioural problems [15].

7.12. Regulatory status

No medicine has an approved indication for ARPKD or Caroli disease. Tolvaptan, a vasopressin V2 receptor antagonist, is the only drug with ongoing trials in ARPKD and is authorised only for adult ADPKD.

The sources consulted did not allow verification of national access (Portugal, Brazil) to these medicines or of paediatric formulations authorised in the European Union.

7.13. Clinical trials and investigational therapies

Tolvaptan is being evaluated based on the efficacy of V2 antagonists in the PCK rat [16]. In Study 307, suspension and tablets are administered in a weight-based split regimen (for example 15+7.5 mg, 30+15 mg or 45+15 mg on waking and 8 hours later), and Study 204 planned suspension doses of 0.15 to 1 mg/kg/day; this is experimental dosing without an approved indication [17, 105]. Planned monitoring includes hepatic and renal function, fluid balance, electrolytes and growth after initiation and each dose escalation [104]. Tesevatinib, a multikinase inhibitor, markedly slowed renal and hepatobiliary cystic disease in two animal models [1]. No randomised trial of disease-modifying therapy has been completed in ARPKD, and experimental agents should not be offered outside trials [11].

Interpretation of these trials has limitations. Study 307 has safety rather than efficacy as its primary outcome, Study 204 was withdrawn, and comparison with historical controls is subject to selection and era bias. Excluding patients with PHT excludes precisely those with significant liver disease, and renal outcomes (KRT) do not capture the liver; the 2025 review identifies outcome selection, enrichment strategies and patient-reported outcomes as challenges for future trials [16].

Developments over the last twelve months, with their respective level of evidence, are as follows. A cohort study (ARegPKD, 2025) published a prognostic score applicable at 2 months of age, with five predictors, described in section 11.1 [45]. In an exploratory diagnostic study (2025), a urinary peptide signature distinguished ARPKD from other causes of CKD, without clinical validation [56]. A population genetic study (2025) identified the Bedouin deep intronic founder variant [36, 37]. In a genetic and functional study (2026), two new families with CYS1 variants consolidated this gene as a cause of ARPKD phenocopy [28]. In a clinical series (2026), classification with phasing and segregation defined a core of molecularly confirmed ARPKD with an allele dosage effect on the kidney [40]. A case–control study (2026) quantified quality of life and caregiver burden [51]. In the trial registry, Study 204 was withdrawn in December 2025 and Study 307 remains recruiting [17, 105]. In preclinical research, there are studies on urinary extracellular vesicles and the carboxy-terminal domain of fibrocystin, without human data. No new regulatory opinion or authorisation for ARPKD was identified in the sources consulted.

8. Reproductive Options and Family Counselling

8.1. Recurrence risks

ARPKD is autosomal recessive, and recurrence risk is 25% in each pregnancy of heterozygous parents, regardless of sex; each sibling has a 25% probability of being affected, 50% of being heterozygous and 25% of being a non-carrier [1, 15]. Molecular testing of parents is recommended to confirm heterozygosity and variant phase [1]. If only one parent is a carrier, with parentage confirmed, a de novo variant in the proband or germline or postzygotic mosaicism in a parent should be considered; apparent homozygosity requires exclusion of a deletion on the other allele and uniparental isodisomy [1]. Pseudodominance, with disease in two generations, has been described in rare families [1]. Children of an affected individual are obligate heterozygotes unless the other parent is affected or a carrier; the proband's aunts and uncles have a 50% probability of being carriers [1]. Carrier frequency in Europe is around 1:70, and KDIGO 2025 recommends counselling PKHD1 variant carriers and offering partner testing when planning children, emphasising the importance of this frequency [1, 6]. Partner carrier testing is particularly relevant if ancestry is shared, notably Afrikaner, Ashkenazi Jewish or Negev Bedouin [1, 36]. The French PNDS restricts this offer: it looks for consanguineous unions in the maternal and paternal families and, if present, proposes heterozygosity testing to the related family member and, if that person is a carrier, to their spouse; testing of the unrelated spouse of a carrier is not routinely proposed and is decided case by case [9]. Molecular study of the patient and both parents may be requested by the nephrologist, geneticist, obstetrician or fetal pathologist, by renal ciliopathy panel or exome [9].

In phenocopies, risk depends on the gene: forms due to DZIP1L, CYS1 and biallelic PKD1 are autosomal recessive; very early-onset ADPKD is dominantly inherited and, when a severe prenatal form occurs in an ADPKD family, the risk of further children with early disease is higher and may result from co-inheritance of a hypomorphic allele transmitted by the apparently unaffected parent [8, 20]. HNF1B-related disease is dominant and frequently de novo [16]. In CHF and Caroli syndrome associated with ciliopathies, counselling depends on the specific genetic diagnosis: most syndromes are autosomal recessive, OFD1 is X-linked and ADPKD dominant [12].

8.2. Cascade testing and sibling assessment

Expression varies between siblings, and abdominal imaging of siblings through to adulthood may be appropriate; genetic testing of apparently asymptomatic siblings is useful when both proband variants are identified [15]. GeneReviews allows early screening and genetic testing of asymptomatic siblings to identify those benefiting from examinations and preventive measures, preceded by formal counselling on the capabilities and limitations of predictive testing and socioeconomic issues; abdominal ultrasound and blood pressure measurement in siblings are informative [1]. In minor siblings, testing is not limited to carrier status: it identifies presymptomatic affected individuals who benefit from blood pressure, renal function and liver surveillance, justifying testing in childhood after counselling [1, 15]. Carrier testing requires prior identification of the familial variants [1]. Parental renal and hepatic ultrasound is useful both to exclude ADPKD and to identify mild abnormalities in heterozygotes [1, 42]. The French PNDS recommends screening of siblings, including molecular screening of the couple's other children [9].

8.3. Prenatal and preimplantation diagnosis

When familial variants are known, molecular prenatal diagnosis (chorionic villus sampling or amniocentesis) and preimplantation genetic testing (PGT) after in vitro fertilisation are possible [1, 30]. Ultrasound often detects disease too late for a pregnancy decision, and there are no systematic data on prenatal ultrasound sensitivity and specificity in pregnancies at 25% risk; haplotype linkage analysis is no longer the method of choice because of error risk [1, 15]. PGT should consider medical, legal and ethical aspects, including parental autonomy and values [30]. Non-invasive prenatal testing is not currently recommended for monogenic diseases [16]. In France, requests for prenatal and preimplantation diagnosis are discussed case by case at the CPDPN, and the PNDS accepts non-invasive prenatal diagnosis by detection of the paternal variant in maternal plasma DNA when the mother does not carry it; when, after termination of pregnancy for medical reasons, only one variant has been identified, expert fetal pathology examination allows discussion of indirect prenatal or preimplantation diagnosis using polymorphic markers close to PKHD1 [9]. These two options diverge from the North American sources cited (section 7.1). In low-risk pregnancies with enlarged cystic kidneys, the pathway is that described in section 5.5 [1].

8.4. Prenatal counselling and prognosis

Termination of pregnancy may be considered with bilateral enlarged hyperechogenic kidneys, but not all these fetuses have ARPKD, and most fetuses with isolated hyperechogenicity have normal renal function after birth [30]. In ARegPKD, the predicted probability of dialysis in the first year was 1.5% (95% CI 0.5–4.1) without prenatal ultrasound abnormalities and 32.3% (95% CI 22.2–44.5) with oligohydramnios or anhydramnios, renal cysts and enlarged kidneys; even with all three factors, most infants did not require early dialysis, and at 36 months the probability was 35% [16, 54]. Oligohydramnios alone is an imprecise indicator: in 74 live births with oligohydramnios of renal origin, almost all 26 neonatal deaths had anhydramnios, but more than half of the 48 survivors also had it [16]. These data argue against an automatically gloomy prognosis. Counselling should be non-directive and multidisciplinary (fetal medicine, paediatric nephrology, neonatology and genetics) and include place and timing of delivery, family implications, contact with other families and palliative care when relevant [16]. Perspectives on prenatal diagnosis differ between families, and the decision is personal [1]. The risk of abdominal dystocia in vaginal delivery should feature in the birth plan [1, 15]. The French PNDS recommends that prenatal disclosure take place in a joint paediatric nephrology and obstetrics consultation, if possible with a psychologist, in a dedicated setting, preferably with both parents and in several stages, addressing natural history, renal and hepatobiliary management, complications and transmission [9]. In the most severe cases, with fetal imaging indicating a high risk of pulmonary hypoplasia or early severe renal failure, termination of pregnancy for medical reasons may be discussed, the request being assessed collegially at the CPDPN according to the French legal criterion of a high probability of a condition of particular severity recognised as incurable at the time of diagnosis; after termination, fetal pathology examination is proposed for definitive diagnosis and counselling [9].

8.5. Family planning for affected adults

Family planning should ideally precede pregnancy, and counselling should be offered to affected young adults, carriers or those at risk [1]. For affected women, the maternal risks described in section 7.10 also apply [1, 53]. In the NIH series, of the 12 adult women, 6 had children: 4 through pregnancies with live births, 1 through surrogacy after medically assisted reproduction and 1 through adoption [53].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Incidence and prevalence: what each figure measures

The most widely cited figure, around 1:20,000 live births in Europe with a carrier frequency of 1:70, comes from Zerres and colleagues (1998) in Germany and is birth incidence, not point prevalence [1, 10, 11]. Orphanet records a validated birth prevalence of 1–9/100,000 (mean value 5.0/100,000), and considers point prevalence in Europe unknown [10]. The 2026 French PNDS gives an incidence of 1/20,000 to 1/30,000 live births, without its own French estimate [9]. In a disease with high perinatal mortality, point prevalence is much lower than birth incidence [11].

In the United States, the Cerner Health Facts database (more than 500 institutions, 2010–2014) enabled estimation of an annualised incidence of 1:26,485 live births (annual range 1:20,933–1:33,098), neonatal survival of 79% and around 120 new neonatal survivors per year; annualised prevalence between 0 and 29 years was 1.17/100 000 (0.88–1.45), corresponding to around 1500 patients in that age group [13]. Limitations are the use of diagnostic codes from electronic records without clinical or molecular confirmation and extrapolation from the database's proportion relative to vital statistics [13].

CHF has no prevalence data; the estimate of 1/10 000–1/20 000, based on the prevalence of associated ciliopathies, is indirect [12]. Caroli disease has an annual incidence of 1–9/1 000,000 in Orphanet, and the point prevalence of the disease and syndrome is unknown [2, 3]. In a Beijing liver transplant series, Caroli syndrome accounted for 1.23% of 1623 transplants, a figure reflecting selection at one centre rather than population frequency [99].

9.2. Founder populations

In the Afrikaner population of South Africa, 25 of 28 families with polycystic disease seen in a paediatric nephrology clinic had ARPKD, 92% of whom were Afrikaans-speaking; point prevalence between 0 and 19 years was estimated at 1:26,000 (carriers 1:83) and, based on a birth incidence of 1:11,000, carriers would be 1:53 [108]. The p.(Met627Lys) variant was found in homozygosity in 27 patients from 24 of 36 Afrikaner families (67%), on a common haplotype, enabling targeted testing in at-risk pregnancies [34]. In Ashkenazi Jews, c.107C>T and c.3761_3762delinsG are founder variants [1]. In Negev Bedouins of Arabian Peninsula ancestry, the deep intronic variant c.8643-2945G>A has a carrier frequency of 2 in 100 controls (1:50), and a further 12 PKHD1 variants have been identified in this population [36]. GeneReviews mentions a higher frequency in isolated populations, citing Finland, but the abstract of the Finnish study consulted does not provide an incidence estimate [1, 23].

9.3. Regional data

North American data come from the prospective NIH cohort (73 patients aged 1 to 56 years), the HRFDCC (Hepatorenal Fibrocystic Disease Core Center) database, CKiD and hospital series [1, 13, 32, 47]. In Europe, ARegPKD (international, based in Cologne) brings together the largest cohort, with up to 658 patients [45]. There are national series from France (50 patients, 1985–2017; PHT in 58%, cholangitis in 16%, normal liver function in all), Slovenia (13 patients, 1980–2020; prenatal diagnosis in 23%, HTN in all, PHT in 54%, dialysis in 46% at a mean age of 15.3 years and 2 deaths), Poland (17 patients; section 3.5) and Spain (68 referred, 40 confirmed) [40, 48, 94, 109]. In the French series, diagnosis was prenatal in 24% and remaining patients were diagnosed at a mean age of 1.8 years [94]. The French PNDS cites the national REIN registry, in which 54% of kidney transplants for ARPKD were pre-emptive (section 7.4) [9]. The sources consulted contain no data from Portugal or Brazil; no geographically targeted searches were performed, and this absence does not demonstrate that published data do not exist. The North American HRFDCC database is open to patients from North, Central and South America [13].

9.4. Special populations

Premature infants with ARPKD have a greater risk of severe respiratory failure and pulmonary hypertension and are more sensitive to ACE inhibitor-induced hypotension [1, 72]. Kidney transplant recipients under immunosuppression have the highest risk of fatal cholangitis [15, 49]. Women of childbearing age require counselling on pregnancy and the fetal toxicity of ACE inhibitors and ARBs (section 7.10). Adults with late diagnosis and predominantly hepatic disease risk being followed only in hepatology without renal and genetic assessment [16, 52].

9.5. Reference networks and patient associations

In Europe, ERKNet (European Reference Network for Rare Kidney Diseases, https://www.erknet.org) is the relevant network for ARPKD, and ERN RARE-LIVER (https://rare-liver.eu) for CHF, Caroli disease and syndrome and paediatric PHT [1, 18]. The ARegPKD registries and HRFDCC/ARPKD database (NCT01401998, currently based at Children's Hospital of Philadelphia, recruiting until 2030) support research [54, 110]. Patient associations cited in the sources are the ARPKD/CHF Alliance (United States, https://www.arpkdchf.org), the PKD Foundation (https://www.pkdcure.org), the PKD Charity (United Kingdom, https://www.pkdcharity.org.uk), PKD International, the Ciliopathy Alliance (United Kingdom), the National Kidney Foundation, the Kidney Foundation of Canada and, in the prenatal context, Antenatal Results and Choices (ARC, United Kingdom) [1, 30, 31, 44].

In France, the organisation described in the 2026 PNDS is based on the ORKiD rare kidney disease health network (filière) (https://www.filiereorkid.com), with three reference centres: MaReGe (Hôpital Femme Mère Enfant, Bron, Lyon; protocol coordinator), SORARE (Toulouse) and MARHEA (Hôpital Necker-Enfants Malades, Paris), and with paediatric nephrology competence centres in each of the 28 university hospital centres [9]. Care is coordinated by a paediatric or adult nephrologist or hepatologist at a reference or competence centre, in liaison with the attending physician, and the protocol records off-label uses [9]. The disease qualifies for the long-term condition (ALD) scheme for chronic kidney diseases, with exemption from patient co-payment; depending on severity, an application to the Maison Départementale des Personnes Handicapées (MDPH) allows the child disability education allowance (AEEH), the adult disability allowance (AAH) or the disability compensation benefit (PCH) to be requested, and parents may receive the daily parental presence allowance (AJPP), with support from a hospital social worker; therapeutic education programmes are authorised by the regional health agency [9]. The French associations are AIRG-France (Association for Information and Research on Genetic Kidney Diseases, https://www.airg-france.fr, with its own ARPKD brochure), France REIN (https://www.francerein.org) and Polykystose France (https://www.polykystose.org); the protocol also refers to the ARPKD/CHF Alliance, PKD International (Geneva), the ERKNet French-language leaflet on ARPKD and the ERN RARE-LIVER page on CHF and polycystic liver disease, and to the Maladies Rares Info Services telephone service [9].

10. Illustrative Clinical Case

Arriola-Montenegro and colleagues, from the Mayo Clinic, published in 2026 the case of a 35-year-old woman with an atypical ARPKD course, in whom the kidneys decreased in volume as function was lost [35]. Diagnosis was made at 6 years during assessment of hypertension (155/78 mmHg) and nephromegaly: creatinine was 0.5 mg/dL, creatinine clearance 92 mL/min and ultrasound showed bilaterally enlarged kidneys (12.2 cm each) with medullary hyperechogenicity. Genetic testing revealed compound heterozygosity for two PKHD1 missense variants, one inherited from each parent: NM_138694.4:c.107C>T p.(Thr36Met), pathogenic, and NM_138694.4:c.10414T>G p.(Cys3472Gly), likely pathogenic; the absence of truncating variants was suggested by the authors as a possible explanation for the milder course [7, 35]. Hypertension was treated with propranolol, chosen instead of ACE inhibitors or ARBs because of parenchymal liver disease and PHT evident on imaging [35].

At 14 years, creatinine clearance was 80 mL/min and systolic HTN above 140 mmHg persisted; the kidneys measured 15.1 and 14.5 cm, with loss of corticomedullary differentiation and small cysts, and there was hepatic fibrosis with PHT, hypersplenism and oesophageal varices. At 16 years, automated three-dimensional volumetry measured a total kidney volume of 723 cm³ and a total liver volume of 1863 cm³. At 22 years the kidneys began to shrink (14.2 and 14.4 cm), with creatinine of 1.4 mg/dL and proteinuria of 492 mg/24 h; at 25 years the patient had CKD stage G3 (creatinine 1.7 mg/dL), with kidney volume of 608 cm³. At 32 years, at stage G4 (creatinine 3.54 mg/dL), with hyperphosphataemia, metabolic acidosis and hypokalaemia, the kidneys measured 9.9 and 11.6 cm (387 cm³); cerebral MR angiography revealed a 2 mm aneurysm of the left cavernous internal carotid artery. At 35 years, at stage G5 without dialysis (creatinine 5.02 mg/dL), total kidney volume was 238 cm³ (142 cm³ on the right and 96 cm³ on the left) and liver volume 1609 cm³ [35]. Figure 1 of the article shows longitudinal changes in kidney and liver volumes, three-dimensional reconstructions at 35 years and serial coronal MRI and CT images [35].

The authors draw four lessons: in adults with ARPKD, kidney volumetry does not predict prognosis because fibrosis leads to renal involution; eGFR, proteinuria and blood pressure should be prioritised; hepatic complications (PHT, cholangitis) progress independently of kidney size; and early assessment for combined liver–kidney transplantation should be considered [35]. This pattern contrasts with childhood, when haeTKV in the first 18 months has prognostic value [39]. Two practices in the report do not correspond to current routine recommendations: cerebral MR angiography in an asymptomatic patient, because the 2014 consensus does not recommend aneurysm screening in ARPKD [15]; and selection of a beta-blocker as an antihypertensive, which differs from the consensus preference for renin–angiotensin system inhibitors, although it is an individualised decision justified by PHT [15, 30]. The case also illustrates renal–hepatic dissociation and the importance of maintaining hepatology follow-up in adults.

11. Prognosis and Long-Term Follow-Up

11.1. Survival and prognostic factors

Around 20–30% of affected infants die in the neonatal period or first year, mainly from respiratory failure or superimposed pulmonary infections; older estimates of 30–50% neonatal mortality predate advances in neonatal intensive care [1, 8, 30]. In the United States, neonatal survival was 79% in 2010–2014 [13]. The French PNDS places perinatal mortality in severe prenatally diagnosed forms, due to pulmonary hypoplasia, at 30–40%, a figure that refers to this subgroup and not to all patients [9]. Among those surviving the first year, subsequent survival is 85–87% at 1 year and 82% at 10 years [1]. In a study of 164 neonatal survivors, survival without treated kidney failure or death from kidney failure was 86% at 5 years, 71% at 10 years and 42% at 20 years [1]. In ARegPKD, 15-year renal survival among 304 neonatal survivors with PKHD1 variants was 65–75%, and the probability of kidney failure at 20 years, in up to 658 patients, was 50.1% [1, 45]. In the phenocopy analysis, survival without KRT or death in ARPKD-PKHD1 fell abruptly in the postnatal period and then declined steadily, reaching 75% at 10 years [29]. The PNDS summarises these data as renal survival of around 50% at 20 years, with a median renal survival age of 20 years in the European series of 304 patients, and survival without severe hepatic complication of 78% at 20 years [9].

Predictors of dialysis in the first year in ARegPKD (385 patients, 9.4% with dialysis in the first year) were oligohydramnios or anhydramnios, prenatal renal enlargement, low Apgar score and the need for postnatal respiratory support; the effects of Apgar and ventilation disappeared after 5 and 8 months, respectively [54]. HaeTKV in the first 18 months, perinatal presentation and corticomedullary involvement on high-resolution ultrasound are associated with faster progression [1, 39, 46]. In 2025, ARegPKD published a score applicable at 2 months of age, with five predictors: gestational age at onset of oligohydramnios or anhydramnios, gestational age at birth, functional genotype, serum creatinine and documented HTN at 2 months; performance was good particularly in the first 3 years and reliably identified patients without risk of early failure [45]. The editorial commentary emphasises that the score applies to children over 2 months without kidney failure, estimates relative rather than absolute risk and requires external validation [111]. The French PNDS adds that prematurity and the intensity of neonatal ventilatory support are associated with lower renal survival, that diagnosis before 1 year is associated with kidney failure before 20 years in 85% of patients, compared with 35% when diagnosis is later, that HTN reduces renal survival, that raised creatinine at 2 months appears to be associated with early failure and that a GFR below 60 mL/min/1.73 m² at 1 year indicates rapid progression [9].

Hepatic prognosis is independent of renal prognosis. Among ARegPKD neonatal survivors, 60–65% developed signs of PHT and around 25% substantial hepatic complications by 15 years [1]. In an American study, around 7% of children surviving the neonatal period required liver transplantation; among ARegPKD young adults, 13% had already received liver or combined transplantation, indicating that hepatobiliary disease tends to become the dominant feature as KRT improves [1, 52]. In the NIH cohort, 23 patients received a kidney (mean age 17.5 years) and 10 a liver (mean age 20.3 years) [33]. After kidney transplantation, 64–80% of deaths are due to cholangitis or sepsis [1]. After combined transplantation, patient survival at 5 years is lower than after isolated kidney transplantation (87.0% versus 97.4%) [92].

In Caroli disease and syndrome, prognosis depends on recurrent cholangitis and cholangiocarcinoma (prevalence around 7%; 1-year survival after tumour diagnosis 36%) [5, 14]. Liver resection for localised disease and transplantation for diffuse disease have good long-term outcomes in surgical series [98, 99].

11.2. Surveillance

In paediatric ARPKD, cholangiocarcinoma is not a concern because it has not been described before 40 years [15]; annual MRCP surveillance applies to Caroli disease and syndrome diagnosed in adults and, by extension, to adults with ARPKD and Caroli syndrome, although the recommendation is weak and opinion-based [5]. Screening for intracranial aneurysms in asymptomatic patients is not indicated [15]. The French PNDS justifies hepatology surveillance in all patients, even without hepatic manifestations, because biliary abnormalities and PHT arise throughout life and can be anticipated, notably through surveillance of varices [9]. The divergence in liver ultrasound intervals between the PNDS (annual) and the 2014 consensus (2–3 years) reflects the inclusion of adults and transition in the French protocol; neither source is based on trials. In adults, kidney volumetry loses prognostic value, and eGFR, proteinuria and blood pressure should guide follow-up [16, 35].

12. Summary and Final Considerations

ARPKD is an autosomal recessive hepatorenal fibrocystic ciliopathy with birth incidence of around 1:20,000 in Europe and 1:26,485 in the United States, caused in most patients by biallelic PKHD1 variants and, in a minority, by DZIP1L or CYS1 variants or biallelic hypomorphic PKD1 variants. The phenotype combines, in variable proportions, enlarged kidneys due to collecting duct dilatation, with a concentrating defect, early hypertension and progressive CKD, and obligatory ductal plate malformation progressing to CHF with PHT and frequently Caroli-type biliary dilatations with cholangitis risk. Perinatal severity depends on pulmonary hypoplasia secondary to oligohydramnios and nephromegaly; after the neonatal period, around half of patients reach kidney failure by 20 years and liver disease becomes progressively dominant, independently of renal disease.

Diagnosis rests on renal and hepatic imaging, parental ultrasound and a multigene panel with CNV analysis, whose limitations (deep intronic variants, deletions, VUS, PKD1 and HNF1B phenocopies) require cautious interpretation of a negative or incomplete result. Treatment is supportive and interdisciplinary: respiratory support, peritoneal dialysis, renin–angiotensin system inhibitors with neonatal restrictions, management of hyponatraemia without sodium supplementation or vaptans, intensive nutrition, haematological and ultrasound surveillance of PHT, early cholangitis treatment, endoscopic management of varices, TIPS in selected patients with good renal function, and kidney, liver or combined transplantation according to the dominant organ, weighing the higher mortality of combined transplantation. Early bilateral nephrectomy requires caution because of neurological risk.

Caroli disease and syndrome lie within the same spectrum of ductal plate malformation; distinction depends on assessment for CHF, adult cholangiocarcinoma risk is around 7% and management includes antibiotics, ursodeoxycholic acid according to EASL, liver resection for localised disease and transplantation for diffuse disease or disease with PHT. Sources diverge on ursodeoxycholic acid, genetic testing in Caroli disease, TIPS and classification of biallelic PKD1, and these divergences reflect different populations and questions. There is no approved disease-modifying therapy: tolvaptan is being evaluated in a phase 3b trial focused on safety, another trial was withdrawn and tesevatinib has no published results. The 2026 French PNDS, the most recent normative document and the only one covering adults, adds paediatric antihypertensive doses, antibiotic regimens for cholangitis, ursodeoxycholic acid in chronic cholestasis, structured screening for PHT and its pulmonary complications and follow-up intervals by severity, with divergences from the 2014 consensus and Baveno VII. The European update of clinical practice recommendations, still unpublished, is expected to replace the 2014 consensus.

13. Bibliographic References

1. Burgmaier K, Gimpel C, Schaefer F, Liebau M. Autosomal Recessive Polycystic Kidney Disease – PKHD1. 2001 Jul 19 [updated 2024 Apr 4]. In: Adam MP, Feldman J, Mirzaa GM, Pagon RA, Wallace SE, Amemiya A, editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1326/
2. Orphanet. Caroli disease (ORPHA:53035). Available from: https://www.orpha.net/en/disease/detail/53035
3. Orphanet. Caroli syndrome (ORPHA:480520). Available from: https://www.orpha.net/en/disease/detail/480520
4. Orphanet. Isolated congenital hepatic fibrosis (ORPHA:485426). Available from: https://www.orpha.net/en/disease/detail/485426
5. European Association for the Study of the Liver. EASL Clinical Practice Guidelines on the management of cystic liver diseases. J Hepatol. 2022;77(4):1083-1108. doi:10.1016/j.jhep.2022.06.002. PMID: 35728731.
6. Kidney Disease: Improving Global Outcomes (KDIGO) ADPKD Work Group. KDIGO 2025 Clinical Practice Guideline for the Evaluation, Management, and Treatment of Autosomal Dominant Polycystic Kidney Disease (ADPKD). Kidney Int. 2025;107(2S):S1-S239. doi:10.1016/j.kint.2024.07.009. PMID: 39848759.
7. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: Accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-68. doi:10.1002/humu.23348. PMID: 28967166; PMC5765404.
8. Firth HV, Hurst JA. Autosomal dominant polycystic kidney disease (ADPKD). In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 354–356.
9. Haute Autorité de Santé. Protocole National de Diagnostic et de Soins (PNDS): Polykystose rénale autosomique récessive. Saint-Denis La Plaine: HAS; 2026 Mar 12 (texte du Centre de référence MaReGe, filière ORKiD, février 2026). Available from: https://www.has-sante.fr/jcms/p_3892968
10. Orphanet. Autosomal recessive polycystic kidney disease (ORPHA:731). Available from: https://www.orpha.net/en/disease/detail/731
11. GeneGuidance. PKHD1 — gene record (clinical curation; version of 2026-09-26).
12. Gunay-Aygun M, Gahl WA, Heller T. Congenital Hepatic Fibrosis Overview [RETIRED CHAPTER, FOR HISTORICAL REFERENCE ONLY]. 2008 Dec 9 [updated 2014 Apr 24; retired 2020 May 14]. In: Adam MP, Bick S, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993-2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK2701/
13. Alzarka B, Morizono H, Bollman JW, Kim D, Guay-Woodford LM. Design and Implementation of the Hepatorenal Fibrocystic Disease Core Center Clinical Database: A Centralized Resource for Characterizing Autosomal Recessive Polycystic Kidney Disease and Other Hepatorenal Fibrocystic Diseases. Front Pediatr. 2017;5:80. doi:10.3389/fped.2017.00080. PMID: 28473971; PMC5397503.
14. Fahrner R, Dennler SG, Inderbitzin D. Risk of malignancy in Caroli disease and syndrome: A systematic review. World J Gastroenterol. 2020;26(31):4718-4728. doi:10.3748/wjg.v26.i31.4718. PMID: 32884228; PMC7445861.
15. Guay-Woodford LM, Bissler JJ, Braun MC, Bockenhauer D, Cadnapaphornchai MA, Dell KM, et al. Consensus expert recommendations for the diagnosis and management of autosomal recessive polycystic kidney disease: report of an international conference. J Pediatr. 2014;165(3):611-7. doi:10.1016/j.jpeds.2014.06.015. PMID: 25015577; PMC4723266.
16. Cadnapaphornchai MA, Dell KM, Gimpel C, Guay-Woodford LM, Gulati A, Hartung EA, et al. Polycystic Kidney Disease in Children: The Current Status and the Next Horizon. Am J Kidney Dis. 2025;86(3):383-392. doi:10.1053/j.ajkd.2025.01.022. PMID: 40113156; PMC12476570.
17. Otsuka Pharmaceutical. Tolvaptan in children and adolescents with ARPKD (Study 156-201-00307). ClinicalTrials.gov ID: NCT04782258. Available from: https://clinicaltrials.gov/study/NCT04782258
18. Mercadal-Hally M, Ilhardt T, Ackermann O, Bruyne R, Doef HV, Fitzpatrick E, et al. Beyond consensus: Identifying divergences in the care of children with portal hypertension. J Pediatr Gastroenterol Nutr. 2026;. doi:10.1002/jpn3.70530. PMID: 42596615.
19. Blyth H, Ockenden BG. Polycystic disease of kidney and liver presenting in childhood. J Med Genet. 1971;8(3):257-84. doi:10.1136/jmg.8.3.257. PMID: 5097134; PMC1469189.
20. Firth HV, Hurst JA. Urinary tract and renal anomalies (congenital anomalies of the kidney and urinary tract—CAKUT). In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 796–799.
21. Caroli J, Soupault R, Kossakowski J, Plocker L, Paradowska. [Congenital polycystic dilation of the intrahepatic bile ducts; attempt at classification]. Sem Hop. 1958;34:488-95. French. PMID: 13543375.
22. Caroli J, Couinaud C, Soupault R, Porcher P, Eteve J. [A new disease, undoubtedly congenital, of the bile ducts: unilobar cystic dilation of the hepatic ducts]. Sem Hop. 1958;34:496-502. French. PMID: 13543376.
23. Kääriäinen H, Koskimies O, Norio R. Dominant and recessive polycystic kidney disease in children: evaluation of clinical features and laboratory data. Pediatr Nephrol. 1988;2(3):296-302. doi:10.1007/BF00858681. PMID: 3153029.
24. Ward CJ, Hogan MC, Rossetti S, Walker D, Sneddon T, Wang X et al. The gene mutated in autosomal recessive polycystic kidney disease encodes a large, receptor-like protein. Nat Genet. 2002;30(3):259-69. doi:10.1038/ng833. PMID: 11919560.
25. Onuchic LF, Furu L, Nagasawa Y, Hou X, Eggermann T, Ren Z et al. PKHD1, the polycystic kidney and hepatic disease 1 gene, encodes a novel large protein containing multiple immunoglobulin-like plexin-transcription-factor domains and parallel beta-helix 1 repeats. Am J Hum Genet. 2002;70(5):1305-17. doi:10.1086/340448. PMID: 11898128; PMC447605.
26. Lu H, Galeano MCR, Ott E, Kaeslin G, Kausalya PJ, Kramer C et al. Mutations in DZIP1L, which encodes a ciliary-transition-zone protein, cause autosomal recessive polycystic kidney disease. Nat Genet. 2017;49(7):1025-1034. doi:10.1038/ng.3871. PMID: 28530676; PMC5687889.
27. Yang C, Harafuji N, O'Connor AK, Kesterson RA, Watts JA, Majmundar AJ et al. Cystin genetic variants cause autosomal recessive polycystic kidney disease associated with altered Myc expression. Sci Rep. 2021;11(1):18274. doi:10.1038/s41598-021-97046-4. PMID: 34521872; PMC8440558.
28. Yang C, Harafuji N, Watts JA, Tao B, Moran C, Clements J et al. Disruption of the human cystin-1 myristoyl-electrostatic switch causes polycystic kidney disease that phenocopies autosomal recessive polycystic kidney disease. Kidney Int. 2026;109(5):939-956. doi:10.1016/j.kint.2026.01.023. PMID: 41720266; PMC13056400.
29. Halawi AA, Burgmaier K, Buescher AK, Dursun I, Erger F, Galiano M et al. Clinical Characteristics and Courses of Patients With Autosomal Recessive Polycystic Kidney Disease-Mimicking Phenocopies. Kidney Int Rep. 2023;8(7):1449-1454. doi:10.1016/j.ekir.2023.04.006. PMID: 37441483; PMC10334384.
30. Liebau MC. Early clinical management of autosomal recessive polycystic kidney disease. Pediatr Nephrol. 2021;36(11):3561-3570. doi:10.1007/s00467-021-04970-8. PMID: 33594464; PMC8497312.
31. Firth HV, Hurst JA. Ciliopathies. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 366–368.
32. Gunay-Aygun M, Font-Montgomery E, Lukose L, Tuchman Gerstein M, Piwnica-Worms K, Choyke P et al. Characteristics of congenital hepatic fibrosis in a large cohort of patients with autosomal recessive polycystic kidney disease. Gastroenterology. 2013;144(1):112-121.e2. doi:10.1053/j.gastro.2012.09.056. PMID: 23041322; PMC4162098.
33. Abdul Majeed N, Font-Montgomery E, Lukose L, Bryant J, Veppumthara P, Choyke PL et al. Prospective evaluation of kidney and liver disease in autosomal recessive polycystic kidney disease-congenital hepatic fibrosis. Mol Genet Metab. 2020;131(1-2):267-276. doi:10.1016/j.ymgme.2020.08.006. PMID: 32919899; PMC7749036.
34. Lambie L, Amin R, Essop F, Cnaan A, Krause A, Guay-Woodford LM. Clinical and genetic characterization of a founder PKHD1 mutation in Afrikaners with ARPKD. Pediatr Nephrol. 2015;30(2):273-9. doi:10.1007/s00467-014-2917-1. PMID: 25193386.
35. Arriola-Montenegro J, Mohan A, Gregory AV, Kline TL, Harris PC, Dahl NK. ARPKD With Shrinking Kidneys and Progressive Loss of Kidney Function. Kidney Int Rep. 2026;11(3):103755. doi:10.1016/j.ekir.2025.103755. PMID: 41624451; PMC12860245.
36. Agam N, Wormser O, Biller A, Hadar N, Dolgin V, Freund O et al. Intronic and Coding Genetic Variants in Autosomal Recessive Polycystic Kidney Disease Among Israeli Bedouins of Arabian Peninsula Ancestry. Am J Kidney Dis. 2025;86(6):730-739.e1. doi:10.1053/j.ajkd.2025.06.011. PMID: 40816622.
37. Sayer JA, Pillai J. A Novel Founder PKHD1 Disease Causing Variant in Israeli Bedouins With Autosomal Recessive Polycystic Kidney Disease. Am J Kidney Dis. 2025;86(6):719-720. doi:10.1053/j.ajkd.2025.09.001. PMID: 41110000.
38. Burgmaier K, Brinker L, Erger F, Beck BB, Benz MR, Bergmann C et al. Refining genotype-phenotype correlations in 304 patients with autosomal recessive polycystic kidney disease and PKHD1 gene variants. Kidney Int. 2021;100(3):650-659. doi:10.1016/j.kint.2021.04.019. PMID: 33940108.
39. Burgmaier K, Kilian S, Arbeiter K, Atmis B, Büscher A, Derichs U et al. Early childhood height-adjusted total kidney volume as a risk marker of kidney survival in ARPKD. Sci Rep. 2021;11(1):21677. doi:10.1038/s41598-021-00523-z. PMID: 34737334; PMC8568977.
40. Lapunzina-Soler P, Shabaka A, Peces R, Alonso Á, Cuesta E, Mena R et al. Next-Generation Sequencing Defines a Molecularly Confirmed ARPKD Core Within the Broader PKHD1-Associated Disease Spectrum. Genes (Basel). 2026;17(2):229. doi:10.3390/genes17020229. PMID: 41751613; PMC12940483.
41. Ajiri R, Burgmaier K, Akinci N, Broekaert I, Büscher A, Dursun I et al. Phenotypic Variability in Siblings With Autosomal Recessive Polycystic Kidney Disease. Kidney Int Rep. 2022;7(7):1643-1652. doi:10.1016/j.ekir.2022.04.095. PMID: 35812281; PMC9263410.
42. Gunay-Aygun M, Turkbey BI, Bryant J, Daryanani KT, Gerstein MT, Piwnica-Worms K et al. Hepatorenal findings in obligate heterozygotes for autosomal recessive polycystic kidney disease. Mol Genet Metab. 2011;104(4):677-81. doi:10.1016/j.ymgme.2011.09.001. PMID: 21945273; PMC3224207.
43. Jones KL, Jones MC, del Campo M. Cranioectodermal dysplasia (Sensenbrenner syndrome). In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 774.
44. Firth HV, Hurst JA. Oligohydramnios (including Potter/oligohydramnios sequence). In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 770–771.
45. Burgmaier K, Kilian S, Arbeiter K, Atmis B, Boyer O, Buescher A et al. A risk score to predict kidney survival in patients with autosomal recessive polycystic kidney disease at the age of two months. Kidney Int. 2025;107(5):903-915. doi:10.1016/j.kint.2025.01.023. PMID: 39922379.
46. Gunay-Aygun M, Font-Montgomery E, Lukose L, Tuchman M, Graf J, Bryant JC et al. Correlation of kidney function, volume and imaging findings, and PKHD1 mutations in 73 patients with autosomal recessive polycystic kidney disease. Clin J Am Soc Nephrol. 2010;5(6):972-84. doi:10.2215/CJN.07141009. PMID: 20413436; PMC2879301.
47. Lin M, Rubin J, Palermo RA, Zee J, Hartung EA. Determinants of left ventricular mass in children with autosomal recessive polycystic kidney disease. J Nephrol. 2025;38(9):2949-2960. doi:10.1007/s40620-025-02426-y. PMID: 41053527; PMC12712119.
48. Wicher D, Grenda R, Teisseyre M, Szymczak M, Halat-Wolska P, Jurkiewicz D et al. Occurrence of Portal Hypertension and Its Clinical Course in Patients With Molecularly Confirmed Autosomal Recessive Polycystic Kidney Disease (ARPKD). Front Pediatr. 2020;8:591379. doi:10.3389/fped.2020.591379. PMID: 33282801; PMC7690924.
49. Srinath A, Shneider BL. Congenital hepatic fibrosis and autosomal recessive polycystic kidney disease. J Pediatr Gastroenterol Nutr. 2012;54(5):580-7. doi:10.1097/MPG.0b013e31824711b7. PMID: 22197937; PMC4369775.
50. Burgmaier K, Ariceta G, Bald M, Buescher AK, Burgmaier M, Erger F et al. Severe neurological outcomes after very early bilateral nephrectomies in patients with autosomal recessive polycystic kidney disease (ARPKD). Sci Rep. 2020;10(1):16025. doi:10.1038/s41598-020-71956-1. PMID: 32994492; PMC7525474.
51. Gimpel C, Schaefer S, Schaefer F. Health-related quality of life, mental health and caregiver burden in children with autosomal recessive polycystic kidney disease. Pediatr Nephrol. 2026;41(1):135-150. doi:10.1007/s00467-025-06795-1. PMID: 40965583; PMC12686003.
52. Burgmaier K, Kilian S, Bammens B, Benzing T, Billing H, Büscher A et al. Clinical courses and complications of young adults with Autosomal Recessive Polycystic Kidney Disease (ARPKD). Sci Rep. 2019;9(1):7919. doi:10.1038/s41598-019-43488-w. PMID: 31138820; PMC6538621.
53. Banks N, Bryant J, Fischer R, Huizing M, Gahl WA, Gunay-Aygun M. Pregnancy in autosomal recessive polycystic kidney disease. Arch Gynecol Obstet. 2015;291(3):705-8. doi:10.1007/s00404-014-3445-8. PMID: 25214022; PMC4320663.
54. Burgmaier K, Kunzmann K, Ariceta G, Bergmann C, Buescher AK, Burgmaier M et al. Risk Factors for Early Dialysis Dependency in Autosomal Recessive Polycystic Kidney Disease. J Pediatr. 2018;199:22-28.e6. doi:10.1016/j.jpeds.2018.03.052. PMID: 29753540.
55. Kadmon Corporation. Tesevatinib single ascending dose in children with ARPKD (KD019-103). ClinicalTrials.gov ID: NCT03096080. Available from: https://clinicaltrials.gov/study/NCT03096080
56. Burgmaier K, Buffin-Meyer B, Klein J, Becknell B, McLeod D, Boeckhaus J et al. Urinary peptide signature distinguishes autosomal recessive polycystic kidney disease from other causes of chronic kidney disease. Clin Kidney J. 2025;18(5):sfaf093. doi:10.1093/ckj/sfaf093. PMID: 40322675; PMC12044329.
57. Subhash S, Safadi A, Bajpai R, Raina R. 2025 KDIGO Executive Summary: Diagnosis and Management of Pediatric Autosomal Dominant Polycystic Kidney Disease. Kidney Blood Press Res. 2026;51(1):101-106. doi:10.1159/000550103. PMID: 41569965; PMC12948374.
58. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Asphyxiating thoracic dysplasia. In: Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 380.
59. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Mainzer-Saldino syndrome. In: Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 403.
60. Jones KL, Jones MC, del Campo M. Oral-facial-digital syndrome type I. In: Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 380.
61. Gimpel C, Avni FE, Bergmann C, Cetiner M, Habbig S, Haffner D et al. Perinatal Diagnosis, Management, and Follow-up of Cystic Renal Diseases: A Clinical Practice Recommendation With Systematic Literature Reviews. JAMA Pediatr. 2018;172(1):74-86. doi:10.1001/jamapediatrics.2017.3938. PMID: 29181500.
62. European Society for Paediatric Nephrology (ESPN), Working Group Inherited Kidney Disorders. Report of activities 2025 [Internet]. 2026 Mar. Available from: https://www.espn-online.org/wp-content/uploads/2026/03/espn-ikd-working-group-report-of-activities-2025.pdf
63. de Franchis R, Bosch J, Garcia-Tsao G, Reiberger T, Ripoll C; Baveno VII Faculty. Baveno VII – Renewing consensus in portal hypertension. J Hepatol. 2022;76(4):959-974. doi:10.1016/j.jhep.2021.12.022. PMID: 35120736; PMC11090185.
64. ANSM. DELURSAN 250 mg, comprimé pelliculé — Résumé des caractéristiques du produit (CIS 68764346), mis à jour le 15/09/2026. Available from: https://base-donnees-publique.medicaments.gouv.fr/affichageDoc.php?specid=68764346&typedoc=R
65. ANSM. SECTRAL 40 mg/mL, solution buvable en flacon — Résumé des caractéristiques du produit (CIS 61969626), mis à jour le 22/04/2025. Available from: https://base-donnees-publique.medicaments.gouv.fr/affichageDoc.php?specid=61969626&typedoc=R
66. European Medicines Agency. Abrysvo (respiratory syncytial virus vaccine, bivalent, recombinant) — Summary of product characteristics. First authorised 23 August 2023. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/abrysvo
67. European Medicines Agency. Beyfortus (nirsevimab) — Summary of product characteristics. First authorised 31 October 2022. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/beyfortus
68. ANSM. UN-ALFA 0,10 microgramme, solution buvable en gouttes — Résumé des caractéristiques du produit (CIS 65401076), mis à jour le 24/03/2026. Available from: https://base-donnees-publique.medicaments.gouv.fr/affichageDoc.php?specid=65401076&typedoc=R
69. European Medicines Agency. Omnitrope (somatropin) — Summary of product characteristics. First authorised 12 April 2006. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/omnitrope
70. EPANED (enalapril maleate) oral solution — Prescribing Information. NDA 208686; effective 2025-12-16. DailyMed setid e6dbda02-779e-420c-893d-d0fea09432be. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=e6dbda02-779e-420c-893d-d0fea09432be
71. QBRELIS (lisinopril) oral solution — Prescribing Information. NDA 208401; effective 2025-12-16. DailyMed setid 9f6e4e57-a489-4b36-b093-b93865d3717c. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=9f6e4e57-a489-4b36-b093-b93865d3717c
72. Starr MC, Flynn JT. Neonatal hypertension: cases, causes, and clinical approach. Pediatr Nephrol. 2019;34(5):787-799. doi:10.1007/s00467-018-3977-4. PMID: 29808264; PMC6261698.
73. KATERZIA (amlodipine) oral suspension — Prescribing Information. NDA 211340; effective 2025-12-16. DailyMed setid df673a4d-acb8-444c-a472-c87ab8cbd366. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=df673a4d-acb8-444c-a472-c87ab8cbd366
74. Agence nationale de sécurité du médicament et des produits de santé (ANSM). NOYADA 5 mg/5 ml, solution buvable — Résumé des caractéristiques du produit (CIS 62231576), mis à jour le 31/07/2026. Base de données publique des médicaments. Available from: https://base-donnees-publique.medicaments.gouv.fr/affichageDoc.php?specid=62231576&typedoc=R
75. ANSM. TRIATEC 1,25 mg, comprimé — Résumé des caractéristiques du produit (CIS 67103203), mis à jour le 01/07/2023. Available from: https://base-donnees-publique.medicaments.gouv.fr/affichageDoc.php?specid=67103203&typedoc=R
76. ANSM. TAREG 3 mg/ml, solution buvable — Résumé des caractéristiques du produit (CIS 62518793), mis à jour le 29/12/2025. Available from: https://base-donnees-publique.medicaments.gouv.fr/affichageDoc.php?specid=62518793&typedoc=R
77. ANSM. COZAAR 50 mg, comprimé pelliculé — Résumé des caractéristiques du produit (CIS 69197614), mis à jour le 13/03/2025. Available from: https://base-donnees-publique.medicaments.gouv.fr/affichageDoc.php?specid=69197614&typedoc=R
78. ANSM. PROPRANOLOL TEVA 40 mg, comprimé sécable — Résumé des caractéristiques du produit (CIS 67954061), mis à jour le 29/09/2025. Available from: https://base-donnees-publique.medicaments.gouv.fr/affichageDoc.php?specid=67954061&typedoc=R
79. ANSM. BICAFRES 1000 mg, comprimé gastro-résistant — Résumé des caractéristiques du produit (CIS 64610621), mis à jour le 14/09/2026. Available from: https://base-donnees-publique.medicaments.gouv.fr/affichageDoc.php?specid=64610621&typedoc=R
80. Burgmaier K, Brandt J, Shroff R, Witters P, Weber LT, Dötsch J, et al. Gastrostomy Tube Insertion in Pediatric Patients With Autosomal Recessive Polycystic Kidney Disease (ARPKD): Current Practice. Front Pediatr. 2018;6:164. doi:10.3389/fped.2018.00164. PMID: 29915780; PMC5994991.
81. ANSM. AMOXICILLINE/ACIDE CLAVULANIQUE PANPHARMA 500 mg/50 mg nourrissons et enfants, poudre pour solution injectable (I.V.) — Résumé des caractéristiques du produit (CIS 61119706), mis à jour le 08/02/2023. Available from: https://base-donnees-publique.medicaments.gouv.fr/affichageDoc.php?specid=61119706&typedoc=R
82. ANSM. AUGMENTIN 100 mg/12,50 mg par ml ENFANTS, poudre pour suspension buvable en flacon — Résumé des caractéristiques du produit (CIS 66853013), mis à jour le 16/09/2026. Available from: https://base-donnees-publique.medicaments.gouv.fr/affichageDoc.php?specid=66853013&typedoc=R
83. ANSM. TAZOCILLINE 4 g/0,5 g, poudre pour solution pour perfusion — Résumé des caractéristiques du produit (CIS 60445784), mis à jour le 13/11/2025. Available from: https://base-donnees-publique.medicaments.gouv.fr/affichageDoc.php?specid=60445784&typedoc=R
84. ANSM. ROCEPHINE 1 g/10 mL, poudre et solvant pour solution injectable (IV) — Résumé des caractéristiques du produit (CIS 68732658), mis à jour le 15/04/2025. Available from: https://base-donnees-publique.medicaments.gouv.fr/affichageDoc.php?specid=68732658&typedoc=R
85. European Medicines Agency. Vedrop (tocofersolan) — Summary of product characteristics. First authorised 24 July 2009. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/vedrop
86. ANSM. RIFADINE 2 POUR CENT, suspension buvable — Résumé des caractéristiques du produit (CIS 64085981), mis à jour le 08/04/2026. Available from: https://base-donnees-publique.medicaments.gouv.fr/affichageDoc.php?specid=64085981&typedoc=R
87. European Medicines Agency. Livmarli (maralixibat) — Summary of product characteristics. First authorised 9 December 2022. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/livmarli
88. ANSM. DUPHALAC 66,5 POUR CENT, solution buvable en flacon — Résumé des caractéristiques du produit (CIS 62729121), mis à jour le 12/02/2025. Available from: https://base-donnees-publique.medicaments.gouv.fr/affichageDoc.php?specid=62729121&typedoc=R
89. ANSM. TIXTAR 550 mg, comprimé pelliculé — Résumé des caractéristiques du produit (CIS 63467355), mis à jour le 21/05/2026. Available from: https://base-donnees-publique.medicaments.gouv.fr/affichageDoc.php?specid=63467355&typedoc=R
90. ANSM. KREDEX 6,25 mg, comprimé sécable — Résumé des caractéristiques du produit (CIS 64233832), mis à jour le 12/10/2023. Available from: https://base-donnees-publique.medicaments.gouv.fr/affichageDoc.php?specid=64233832&typedoc=R
91. Verbeeck S, Mekhali D, Cassiman D, Maleux G, Witters P. Long-term outcome of transjugular intrahepatic portosystemic shunt for portal hypertension in autosomal recessive polycystic kidney disease. Dig Liver Dis. 2018;50(7):707-712. doi:10.1016/j.dld.2018.03.009. PMID: 29622386.
92. Mekahli D, van Stralen KJ, Bonthuis M, Jager KJ, Balat A, Benetti E et al. Kidney Versus Combined Kidney and Liver Transplantation in Young People With Autosomal Recessive Polycystic Kidney Disease: Data From the European Society for Pediatric Nephrology/European Renal Association-European Dialysis and Transplant (ESPN/ERA-EDTA) Registry. Am J Kidney Dis. 2016;68(5):782-788. doi:10.1053/j.ajkd.2016.06.019. PMID: 27555106.
93. Acharya R, Upadhyay K. Short-Term Outcome of Isolated Kidney Transplantation in Children with Autosomal Recessive Polycystic Kidney Disease: A Case Series and Literature Review. Clin Pract. 2023;14(1):24-30. doi:10.3390/clinpract14010003. PMID: 38300123; PMC10887803.
94. Dorval G, Boyer O, Couderc A, Delbet JD, Heidet L, Debray D et al. Long-term kidney and liver outcome in 50 children with autosomal recessive polycystic kidney disease. Pediatr Nephrol. 2021;36(5):1165-1173. doi:10.1007/s00467-020-04808-9. PMID: 33165639.
95. Grenda R, Kaliciński P. Combined and sequential liver-kidney transplantation in children. Pediatr Nephrol. 2018;33(12):2227-2237. doi:10.1007/s00467-017-3880-4. PMID: 29322327; PMC6208698.
96. Büscher R, Büscher AK, Cetiner M, Treckmann JW, Paul A, Vester U et al. Combined liver and kidney transplantation and kidney after liver transplantation in children: Indication, postoperative outcome, and long-term results. Pediatr Transplant. 2015;19(8):858-65. doi:10.1111/petr.12595. PMID: 26341656.
97. Yi NJ, Kim J, Hong SY, Kang HG. Combined liver-kidney transplantation in pediatric patients. Pediatr Transplant. 2024;28(1):e14666. doi:10.1111/petr.14666. PMID: 38059323.
98. Ulrich F, Pratschke J, Pascher A, Neumann UP, Lopez-Hänninen E, Jonas S, et al. Long-term outcome of liver resection and transplantation for Caroli disease and syndrome. Ann Surg. 2008;247(2):357-64. doi:10.1097/SLA.0b013e31815cca88. PMID: 18216545.
99. Hou F, Xiong HF, Liu Y, Zhao S, Zhang YZ, Kang Q et al. Clinical outcomes of liver transplantation in Caroli syndrome: a retrospective analysis. Orphanet J Rare Dis. 2025;20(1):497. doi:10.1186/s13023-025-03943-6. PMID: 41039418; PMC12492817.
100. Otsuka America Pharmaceutical. JYNARQUE (tolvaptan) tablets — Prescribing Information. Revised 9/2026. DailyMed setid 3febc0a1-9e5a-4ce0-843d-210f21d862c4; Drugs@FDA NDA 204441. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=3febc0a1-9e5a-4ce0-843d-210f21d862c4
101. European Medicines Agency. Jinarc (tolvaptan) — EPAR. Authorised 2015-05-27; last EC decision 2026-08-18. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/jinarc
102. Otsuka America Pharmaceutical. SAMSCA (tolvaptan) tablets — Prescribing Information. NDA 022275; label 2026-05-07. Available from: https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=022275
103. European Medicines Agency. Samsca (tolvaptan) — EPAR. Available from: https://www.ema.europa.eu/en/medicines/human/EPAR/samsca
104. Mekahli D, Liebau MC, Cadnapaphornchai MA, Goldstein SL, Greenbaum LA, Litwin M et al. Design of two ongoing clinical trials of tolvaptan in the treatment of pediatric patients with autosomal recessive polycystic kidney disease. BMC Nephrol. 2023;24(1):33. doi:10.1186/s12882-023-03072-x. PMID: 36782137; PMC9926647.
105. Otsuka Pharmaceutical. Tolvaptan in infants with ARPKD (Study 156-12-204). ClinicalTrials.gov ID: NCT04786574. Available from: https://clinicaltrials.gov/study/NCT04786574
106. Cleveland Clinic. Advanced MRI for kidney disease progression in ARPKD. ClinicalTrials.gov ID: NCT07201025. Available from: https://clinicaltrials.gov/study/NCT07201025
107. MacAskill CJ, Smothers JB, Kretzler ME, Heilman J, Pepin K, Kuehn B, et al. Advanced MRI methods for children and young adults with autosomal recessive polycystic kidney disease (ARPKD) to support future multi-center clinical trials. Pediatr Radiol. 2026;. doi:10.1007/s00247-026-06737-5. PMID: 42517885.
108. Lombard EH, Kromberg JG, Thomson PD, Milner LS, van Biljon I, Jenkins T. Autosomal recessive polycystic kidney disease. Evidence for high frequency of the gene in the Afrikaans-speaking population. S Afr Med J. 1989;76(7):321-3. PMID: 2799576.
109. Fon Gabršček A, Meglič A, Novljan G, Kersnik Levart T, Rus R. Clinical characteristics of Slovenian pediatric patients with autosomal recessive polycystic kidney disease. Clin Nephrol. 2021 Suppl;96(1):56-61. doi:10.5414/CNP96S10. PMID: 34643492.
110. University of Alabama at Birmingham / Children's Hospital of Philadelphia. ARPKD/Congenital Hepatic Fibrosis (HRFD) clinical database. ClinicalTrials.gov ID: NCT01401998. Available from: https://clinicaltrials.gov/study/NCT01401998
111. Ng DK, Matheson MB, Hartung EA. Progress, challenges, and pragmatic concessions in predicting relative risk of kidney survival in ARPKD. Kidney Int. 2025;107(5):788-791. doi:10.1016/j.kint.2025.02.018. PMID: 40254358; PMC12101604.
112. Gimpel C, Avni EF, Breysem L, Burgmaier K, Caroli A, Cetiner M, et al. Imaging of Kidney Cysts and Cystic Kidney Diseases in Children: An International Working Group Consensus Statement. Radiology. 2019;290(3):769-782. doi:10.1148/radiol.2018181243. PMID: 30599104; PMC6394734.