Comparison of genetic profile and kidney outcomes between very early-onset and non-very-early-onset autosomal dominant polycystic kidney disease in children: a systematic review and meta-analysis

Abstract

Background

Autosomal dominant polycystic kidney disease (ADPKD) is the most common inherited kidney disorder and a leading cause of kidney failure. Although typically considered an adult-onset condition, a subset of patients present in utero or within the first 18 months of life (very early-onset ADPKD; VEO-ADPKD). This subgroup may experience a more severe disease course.

Methods

A systematic search of MEDLINE was conducted to identify studies reporting clinical, genetic, radiologic and kidney outcome data in children (< 18 years) with VEO-ADPKD. Case series including fewer than three VEO-ADPKD patients were excluded. ADPKD diagnosis was based on 2019 international consensus radiologic criteria, including age-appropriate cystic and/or enlarged kidneys, with supportive family history and/or genetic confirmation. Children with non-VEO-ADPKD reported within the same studies were included for comparison. Outcomes of interest included inheritance patterns, genetic profile, hypertension, proteinuria, chronic kidney disease (CKD), kidney failure, and perinatal death. Descriptive analyses and Bayesian model-averaged random-effects meta-analyses were performed using JASP (version 0.9.13) with a conditional model specification. A non-informative Beta (1,1) prior was used.

Results

Nineteen studies comprising 736 ADPKD children (VEO: 335, non-VEO: 401) met inclusion criteria. Bayesian model-averaged meta-analysis (8 studies) showed higher odds in VEO-ADPKD for hypertension (OR 2.08, 95% CrI 1.00–2.70; BF₁₀ 18.2), proteinuria (OR 1.58, 1.00–2.60; BF₁₀ 4.0), CKD (OR 1.77, 1.00–2.66; BF₁₀ 7.4) and kidney failure (OR 1.99, 1.00–2.69; BF₁₀ 10.6). Evidence for maternal inheritance in VEO-ADPKD was weak (OR 1.31, 1.00–2.44; BF₁₀ 1.6). Biallelic PKD1 mutations (38/158 vs. 0/147) and perinatal death (34/320 vs. 0/401) occurred exclusively in the VEO group and are reported descriptively, as complete separation precluded pooled estimation.

Conclusion

VEO-ADPKD represents a distinct, high-risk pediatric phenotype characterized by earlier and more severe kidney involvement. These findings emphasize the importance of early diagnosis, comprehensive genetic evaluation, and close monitoring. Genetic testing is essential to identify children with biallelic mutations, representing a clinical entity more akin to mild autosomal recessive polycystic kidney disease. Standardized phenotyping and prospective pediatric studies are needed to guide early therapeutic interventions and improve long-term outcomes in this rare population.

Clinical trial number

NA.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12882-026-05182-8.

Background

Autosomal dominant polycystic kidney disease (ADPKD) is the most common genetic kidney disease with an estimated prevalence of 1 in 400 to 1000 individuals and one of the leading causes of kidney failure worldwide [1]. It is characterized by the progressive development of multiple kidney cysts, which can steadily lead to loss of kidney function and subsequently kidney failure around the sixth decade of life [2–5]. Genetic mutations in PKD1(polycystin-1) and PKD2 (polycystin-2), which encode proteins in primary cilia of renal tubular epithelial cells, are responsible for most cases of ADPKD.

Although for decades ADPKD was considered to be a disease of adulthood with favorable prognosis, with recent improvements in prenatal follow up and radiological modalities, a few studies have reported severe ADPKD presentation within the first 18 months of life (VEO) or even early in utero with similar findings to those seen in neonates with autosomal recessive polycystic kidney disease (ΑRΡΚD). VEO-ADPKD patients often experience severe kidney manifestations, yet the spectrum of disease expression remains highly variable, creating substantial uncertainty for families and paediatric nephrologists. Specifically, children with VEO-ADPKD may present rapidly progressive disease resulting in chronic kidney disease (CKD) and consequently kidney failure in ϲhilԁhοod has been reported [6–10].

The aim of this article is to systematically review for the first time the genetic profile and kidney outcomes of children with VEO-ADPKD across paediatric studies and compare these findings with those of non-VEO-ADPKD via meta-analysis. This comparison seeks to offer further insight into this rare pediatric population with respect to genetic counseling, prognostic stratification, recommended follow-up strategies and therapeutic management.

Methods

Study selection and inclusion criteria

Studies considered eligible for inclusion in this systematic review were those reporting data on the clinical characteristics, radiological characteristics, genetic profile and kidney outcomes of children with the diagnosis of VEO-ADPKD < 18 years. No automation tools were used. Non-VEO-ADPKD children included in the aforementioned eligible VEO-ADPKD studies were also analyzed to enable comparisons between both groups. All study designs were considered if they reported data on a minimum of 3 children. VEO-ADPKD was defined as a diagnosis of ADPKD within the first 18 months of life [10]. Non-VEO-ADPKD patients were diagnosed using the same criteria after 18 months of age. Diagnosis of ADPKD was established based on radiologic criteria in accordance with the 2019 international consensus statement on the diagnosis and management of ADPKD, defined by the presence of one or more kidney cysts in children younger than 15 years and/or hyperechogenic and/or enlarged kidneys (> 2 SD) in fetuses or neonates, in the context of a positive family history of ADPKD and/or confirmatory genetic testing. Patient demographics and genetic profiles were reviewed here, where applicable, in detail. Outcomes of interest were inheritance pattern, the presence of biallelic mutations in PKD1/PKD2, hypertension (BP > 95th centile for age, gender and height was required), proteinuria (defined by the local measurement method), CKD (GFR < 90 ml/min/1.73m²), kidney failure (GFR < 15 ml/min/1.73m²) and perinatal mortality (stillbirth or neonatal death). The search was restricted to full text articles published in English. Studies which did not include VEO-ADPKD patients were excluded. Case report studies including less than three VEO-ADPKD patients were also excluded. The titles of all papers retrieved from this search were screened by two independent authors (CK and GD) for potential inclusion in this review.

Study design

This systematic review was conducted in accordance with the PRISMA 2020 guidelines illustrated in the Supplementary file 1. A comprehensive systematic search was performed by two independent reviewers (CK, GD) searching MEDLINE to discern published articles assessing the clinical characteristics and long-term kidney outcomes of children with VEO-ADPKD using relevant keywords depicted in the Supplementary file 2.

Statistical analysis

Descriptive comparisons between VEO and non-VEO ADPKD patients were performed using study-level data in all 19 studies included in this review. A Bayesian framework was selected given the small number of studies, expected heterogeneity, and presence of zero-event data to compare studies, which included VEO and non-VEO-ADPKD children. A Bayesian model-averaged random-effects meta-analysis was conducted to estimate the pooled odds ratio (OR) across the included studies. The analysis was performed in JASP (version 0.9.13) using the conditional model specification. Study-level data consisted of the number of events and total sample size in the VEO and non-VEO ADPKD groups. A Beta (1,1) prior was used for the study-specific OR, reflecting a non-informative prior with equal weight across the 0–1 range. Between-study heterogeneity (τ) was modeled with a Half-Cauchy (0, 0.5) prior, allowing for moderate variability while constraining extreme values. Posterior estimates included pooled ORs with 95% credible intervals and the posterior distribution of heterogeneity. Bayes factor (BF₁₀) was additionally calculated to quantify the relative evidence supporting the alternative hypothesis over the null hypothesis, with larger BF₁₀ values indicating stronger evidence for an association. A forest plot was generated to visualize the results. Zero-event studies were included without continuity correction, as Bayesian modeling naturally accommodates sparse data.

Risk of bias

Quality assessment performed by two independent reviewers (CK, GD) is depicted in the Supplementary file 3. Comparative VEO vs. non-VEO studies were evaluated using the ROBINS-I tool. Studies were judged to have moderate risk of bias when they included clearly defined comparison groups, standardized outcome assessment, and adequate follow-up. Studies were judged to have serious risk of bias when key ROBINS-I domains were affected, most commonly confounding (limited adjustment for disease severity, genotype, or familial clustering), selection bias and missing or incomplete follow-up, particularly in older retrospective studies. Non-comparative VEO studies were assessed using the Joanna Briggs Institute (JBI) Critical Appraisal Checklist for Case Series. Studies were evaluated as moderate risk of bias when cases were clearly defined, outcomes were reported systematically, and follow-up was described, but limitations remained due to retrospective design. Studies were rated as serious risk of bias when there was unclear or heterogeneous outcome definitions and limited follow-up, which is common in early prenatal and neonatal cases.

Results

The initial search included records for 350 published articles in Medline. Nineteen studies (Table 1) including 736 ADPKD children (VEO: 335, non-VEO: 401) met the inclusion criteria for this review (Prisma flow chart, supplementary file 4). Table 2 illustrates the number and percentage of VEO and non-VEO ADPKD children included in all 19 studies comparing the frequency of maternal inheritance, type of genetic mutations in PKD1/PKD2, hypertension, proteinuria, CKD, kidney failure and perinatal death. Eight out of 19 studies, including 572 ADPKD children (VEO: 171, non-VEO: 401), provided comparative data for both groups and were therefore eligible for meta-analysis comparing abovementioned parameters.

Table 1.

Overview of ADPKD studies included in the systematic review

ID	Author	Year	Study design	Country	Aim of study	Follow up time	VEO and non-VEO patients	Total n (VEO/non VEO)	Sex (M/F)
1	Demir [6]	2021	Observational retrospective	Turkey	To describe the clinical radiological and genetic characteristics in ADPKD children	4.9y	YES	41 (5/36)	18/23
2	Durkie [7]	2021	Observational retrospective	UK	To describe the prevalance of PKD1/PKD2 mutations in a large paediatric cohort	NA	NO	30 (30/0)	NA
3	Garel [8]	2019	Cross sectional- retrospective	France-Belgium	To describe prenatal ultrasound patterns of ADPKD mimicking ARPKD	NA	NO	12 (12/0)	NA
4	Al-Hamed [9]	2019	Cross-sectional - retrospective	Saoudi Arabia	To identify biallelic PKD1 mutations causing early-onset ADPKD	NA	NO	4 (4/0)	NA
5	Novak [10]	2016	Lungitudinal-retrospective	USA	To compare long-term outcomes between VEO and non-VEO-ADPKD children	Until the age of 16 years	YES	140 (70/70)	58/82
6	Audrezet [11]	2015	Cross-sectional- retrospective	France-Tunisia	To investigate the genetic basis of prenatal and VEO-ADPKD	NA	NO	42 (42/0)	20/22
7	Fencl [12]	2008	Cross-sectional- retrospective	Chech Republic	To compare genotype–phenotype correlations (PKD1 vs. PKD2) in children with ADPKD.	NA	YES	60 (7/53)	30/30
8	Boyer [13]	2007	Longitudinal-retrospective	France	To assess the prognosis and clinical outcomes of children with ADPKD diagnosed prenatally or at birth	76 m	NO	26 (26/0)	15/11
9	Shamshirsaz [14]	2005	Longitudinal-prospective	USA	To compare disease progression between VEO and non-VEO children	Up to 29y	YES	199 (46/153)	92/107
10	Mashiach [15]	2005	Cross-sectional retrospective	Israel	To evaluate the postnatal outcomes of fetuses with hyperechogenic kidneys	NA	NO	3 (3/0)	NA
11	Brun [16]	2004	Longitudinal-retrospective	France-Belgium	To describe specific prenatal ultrasound patterns associated with autosomal dominant polycystic kidney disease	32 m	NO	27 (27/0)	17/8/2 unknown
12	Tee [17]	2004	Longitudinal- retrospective	Canada	To describe specific prenatal ultrasound patterns associated with ADPKD	4.9y	YES	55 (10/45)	27/28
13	Mac Dermot [18]	1998	Cross sectional retrospective	UK	To describe prenatal diagnosis of PKD1-related ADPKD presenting in utero and to assess prognosis of VEO-ADPKD	NA	NO	3 (3/0)	2/1
14	Zerres [19]	1993	Multicenter-longitudinal prospective	Germany	To analyse the clinical features of childhood-onset ADPKD and to estimate the recurrence risk of early disease manifestation among siblings	Up to ~ 25y	YES	37 (19/18)	18/19
15	Gal [20]	1989	Cross- sectional retrospective	Germany	To determine whether childhood-onset ADPKD represents a genetically distinct form from adult-onset ADPKD	NA	NO	6 (6/0)	2/4
16	Cole [21]	1987	Cross- sectional retrospective	USA	To determine the frequency of ARPKD versus ADPKD in infants and to compare their clinical progression.	NA	NO	6 (6/0)	NA
17	Kaariainen [22]	1987	Cross-sectional retrospective	Finland	To classify childhood PKD	NA	YES	17 (9/8)	8/8/1 unknown
18	Pretorius [23]	1987	Cross-sectional-retrospective	USA	To describe prenatal ultrasound findings and assess postnatal findings of cystic kidney disease	NA	NO	5 (5/0)	NA
19	Sedman [24]	1987	Longitudinal-rospective	USA	To describe progression of ADPKD children	7.6y	YES	23 (5/18)	13/10
								736 (335/401)

Bold indicates studies included for meta-analyses between VEO and non-VEO ADPKD

ARPKD stands for autosomal recessive polycystic kidney disease

Table 2.

Genetic profile and Kidney outcomes of VEO and non-VEO ADPKD children included in all 19 studies

Genetics/Outcome	VEO n / N (%)	Non-VEO n / N (%)
Maternal inheritance	167 / 277 (60.2%)	154 / 286 (53.8%)
Heterozygous PKD1	68/158 (43%)	87/147 (59.2%)
Biallelic PKD1	38 / 158 (24.1%)	0 / 147 (0%)
Homozygous PKD1	5/158 (3.2%)	0/147 (0%)
Heterozygous PKD2	2/158 (1.3%)	10/147 (6.8%)
Biallelic PKD2	1/158 (0.6%)	0/147 (0%)
Homozygous PKD2	1/158 (0.6%)	0/147 (0%)
Digenic PKD1/PKD2 mutations	1/158 (0.6%)	0/147 (0%)
Hypertension	106 / 268 (39.6%)	91 / 357 (25.5%)
Proteinuria	30 / 95 (31.6%)	71 / 251 (28.3%)
CKD	35 / 250 (14.0%)	17 / 340 (5.0%)
Kidney failure	15 / 276 (5.4%)	1 / 393 (0.3%)
Perinatal death	34 / 320 (10.6%)	0 / 401 (0%)

Aggregated event counts are presented for descriptive purposes only; inference is based on pooled odds ratios. Data are presented as number (%). For each outcome, only studies reporting that outcome were included. Denominators correspond to the total number of VEO or non-VEO children (VEO_n or non_VEO_n) in studies reporting the outcome. VEO: very early onset; ADPKD: autosomal dominant polycystic kidney disease

Table 2 summarizes data in regards to the genetic profile and kidney outcomes of VEO and non-VEO ADPKD children across all 19 included studies. Maternal inheritance was more frequent in the VEO group compared with non-VEO children (60.2% vs. 53.8%). VEO-ADPKD children demonstrated a markedly higher prevalence of biallelic and homozygous PKD1/PKD2 mutations, whereas heterozygous PKD1 and PKD2 variants were more commonly seen in the non-VEO group. Specifically, biallelic PKD1 mutations were observed in 24.1% of VEO children but were absent in the non-VEO cohort. In terms of clinical outcomes VEO-ADPKD was associated with higher rates of hypertension (39.6% vs. 25.5%), proteinuria (31.6% vs. 28.3%), CKD (14.0% vs. 5.0%), kidney failure (5.4% vs. 0.3%), and perinatal death (10.6% vs. 0%). Only one case of kidney failure and no cases of perinatal death were reported among non-VEO children.

Meta-analyses (VEO vs. non-VEO ADPKD)

Supplementary File 5 summarizes the total number of VEO and non-VEO ADPKD patients of each study included in the meta-analyses for each outcome of interest.

Maternal inheritance

Figure 1 illustrates the meta-analyses of maternal inheritance between both ADPKD groups. The Bayesian model-averaged random-effects meta-analysis yielded a pooled effect size of log (OR) = 0.27 (posterior median = 0.19; 95% CrI: 0.00 to 0.89), corresponding to an OR of 1.31 (95% CrI: 1.00 to 2.44). The inclusion Bayes factor provided only anecdotal evidence supporting an association between maternal inheritance and VEO-ADPKD (BF₁₀ = 1.58; posterior model probability = 0.61). The lower bound of the model-averaged credible interval reaches the null, and although the point estimate is directionally consistent with more frequent maternal inheritance in the VEO group, the meta-analytic evidence is weak and inconclusive. The heterogeneity parameter (τ) showed a posterior mean of 0.26 (median 0.00; 95% CrI: 0.00 to 1.39), indicating limited and uncertain between-study variability (Fig. 1).

Fig. 1.

Forest plot of log odds ratios for exhibiting maternal inheritance across studies in the VEO-ADPKD group

Biallelic mutations (PKD1)

Biallelic PKD1 mutations were identified in 38 of 158 VEO-ADPKD children (24.1%) compared with 0 of 147 non-VEO-ADPKD children (0%) across the studies reporting this outcome. As no biallelic PKD1 mutations were observed in the non-VEO group, the data exhibit complete separation, and a pooled odds ratio could not be meaningfully estimated; we therefore report this outcome descriptively. The presence of biallelic PKD1 mutations exclusively in the VEO group is consistent with the more severe genotype reported in very early-onset disease, although the retrospective nature and limited size of the contributing studies warrant caution.

Hypertension

The Bayesian model-averaged random-effects meta-analysis produced a pooled effect size of log (OR) = 0.73 (posterior median = 0.81; 95% CrI: 0.00 to 0.99), corresponding to an OR of 2.08 (95% CrI: 1.00 to 2.70). The inclusion Bayes factor provided strong evidence for an association (BF₁₀ = 18.19; posterior model probability = 0.95), indicating that children with VEO-ADPKD had approximately twice the odds of hypertension compared with those with non-VEO ADPKD. Between-study heterogeneity was modest under model averaging (τ posterior median = 0.12; 95% CrI: 0.00 to 1.16; inclusion BF = 1.35), indicating limited and uncertain variability across the included studies (Fig. 2).

Fig. 2.

Forest plot of log odds ratios for exhibiting hypertension across studies in the VEO-ADPKD group

Proteinuria

Regarding proteinuria, the Bayesian model-averaged random-effects meta-analysis yielded a pooled effect size of log (OR) = 0.45 (posterior median = 0.50; 95% CrI: 0.00 to 0.96), corresponding to an OR of 1.58 (95% CrI: 1.00 to 2.60). The inclusion Bayes factor provided moderate evidence for an association (BF₁₀ = 4.02; posterior model probability = 0.80), indicating a higher risk of proteinuria in the VEO group, although the evidence is weaker than that observed for hypertension. Between-study heterogeneity was modest under model averaging (τ posterior median = 0; 95% CrI: 0.00 to 0.92; inclusion BF = 0.60), indicating limited and uncertain variability across the three contributing studies (Fig. 3).

Fig. 3.

Forest plot of log odds ratios for exhibiting proteinuria across studies in the VEO-ADPKD group

CKD – CKD

The Bayesian model-averaged random-effects meta-analysis indicated a pooled effect size of log (OR) = 0.57 (posterior median = 0.63; 95% CrI: 0.00 to 0.98), corresponding to an OR of 1.77 (95% CrI: 1.00 to 2.66). The inclusion Bayes factor provided moderate-to-strong evidence for an association (BF₁₀ = 7.42; posterior model probability = 0.88), indicating a higher risk of CKD in the VEO group than in the non-VEO group. The heterogeneity parameter (τ) showed a posterior mean of 0.18 (median 0; 95% CrI: 0.00 to 1.25), indicating limited and uncertain between-study variability. Overall, these results support a moderately higher risk of CKD in VEO-ADPKD children (Fig. 4a).

Fig. 4.

(a) Forest Plot of Log Odds Ratios for exhibiting CKD across studies in the VEO-ADPKD group. (b) Forest Plot of Log Odds Ratios for exhibiting kidney failure across studies in the VEO-ADPKD group

Kidney failure

The Bayesian model-averaged random-effects meta-analysis indicated a pooled effect size of log (OR) = 0.69 (posterior median = 0.79; 95% CrI: 0.00 to 0.99), corresponding to an OR of 1.99 (95% CrI: 1.00 to 2.69). The inclusion Bayes factor provided strong evidence for an association (BF₁₀ = 10.61; posterior model probability = 0.91), indicating higher odds of kidney failure in the VEO compared to the non-VEO group.The heterogeneity parameter (τ) showed a posterior mean of 0.71 (median 0.29; 95% CrI: 0.00 to 3.30), indicating substantial and highly uncertain between-study variability. Overall, these results support a higher risk of kidney failure in VEO-ADPKD children, although the wide heterogeneity interval warrants caution (Fig. 4b).

Perinatal death

Perinatal death was reported in 34 of 320 VEO-ADPKD children (10.6%) compared with 0 of 401 non-VEO-ADPKD children (0%) across the studies reporting this outcome. Because no perinatal deaths were recorded in the non-VEO group, the data exhibit complete separation, and a pooled odds ratio could not be meaningfully estimated; any such estimate would be determined by the choice of continuity correction or prior rather than by the observed data. We therefore report this outcome descriptively. The complete absence of perinatal deaths in the non-VEO group, against 34 events among VEO children, represents a marked difference, although the small number of events and the retrospective nature of the contributing studies warrant caution in interpretation.

Discussion

In this systematic review of 19 studies comprising 736 children with ADPKD, we compared the genetic profile and kidney outcomes between VEO and non-VEO-ADPKD children. Meta-analyses directly comparing VEO and non-VEO studies were feasible in 8 out of the 19 included studies. Of note, this is the first systematic review and meta-analyses to date comparing the genetic profile and kidney outcomes between both ADPKD groups.

Although the aggregated descriptive proportions of maternal inheritance were slightly higher among children with VEO-ADPKD than among those with non-VEO ADPKD (Table 2), the Bayesian meta-analysis (Fig. 1) provided only weak evidence for this association (OR 1.31, 95% CrI 1.00–2.44; BF₁₀ = 1.58). Although previous pediatric cohorts have reported a similar trend, ADPKD is an autosomal dominant disorder without an established parent-of-origin effect for PKD1 or PKD2 mutations [19, 21, 22]. The apparent association may therefore reflect ascertainment bias due to increased prenatal surveillance in affected mothers, as well as the contribution of possible maternal genetic or intrauterine factors that may influence disease expression. Further studies are definitely required to determine whether this association reflects a true biological effect or differences in early disease detection.

Additionally, this study confirms previous reports in regards to the severity of homozygous and compound heterozygous variants in VEO-ADPKD children. Although PKD1 and PKD2 mutations classically show an autosomal dominant inheritance pattern, many biallelic variants are responsible for VEO disease with unfavorable prognosis and outcome [13]. It has been already well described that children with biallelic PKD variants may show a clinical course more akin to mild cases of ARPKD [25]. A high percentage of VEO-ADPKD children in this review were found to carry biallelic or homozygous PKD1 or PKD2 mutations when compared to the non-VEO population, which showed a higher percentage of heterozygous PKD1/PKD2 mutations. Occasionally, digenic (PKD1/PKD2) inheritance was also reported rarely in the VEO group. Notably, the marked enrichment of biallelic PKD variants and the more severe kidney outcomes observed in VEO-ADPKD warrant consideration of this phenotype within the broader spectrum of pediatric cystic kidney diseases, particularly ARPKD. Of note, both disorders may present antenatally or during infancy with enlarged hyperechogenic kidneys, early-onset hypertension and proteinuria, and impaired kidney function. However, the clinical course of ARPKD is typically more aggressive, with earlier progression to advanced CKD and kidney failure [25]. Although non-VEO ADPKD is generally considered a slowly progressive kidney disease [2], our findings indicate that VEO-ADPKD constitutes a distinct high-risk phenotype with significantly worse kidney outcomes than non-VEO ADPKD. Importantly, these adverse outcomes may reflect not only earlier disease presentation but also the enrichment of genetically more severe forms of disease within the VEO population. Hence, these findings highlight the importance of comprehensive genetic testing in infants and young children presenting with severe cystic kidney disease, as it may not only confirm the diagnosis of ADPKD but also improve prognostic stratification and help differentiate severe ADPKD from other hereditary cystic kidney disorders. Moreover, studies published before 2008 lacked genetic information for children with VEO-ADPKD, highlighting the need for further research in the future to better define genotype-phenotype correlations in this population. Additionally, this high-risk group may represent a population in whom early administration of tolvaptan, which is now investigated for use in ADPKD children, could be considered beneficial in future studies [26].

Children with VEO ADPKD exhibited significantly higher rates of hypertension compared with those diagnosed later in childhood (Fig. 2). Early-onset hypertension has been repeatedly identified as a marker of disease severity in both pediatric and adult ADPKD cohorts and is thought to reflect early cyst burden, kidney enlargement, and activation of renin-angiotensin-aldosterone pathways [27–29]. Data on the age at onset of hypertension are truly limited among studies included in this review. Novak et al. study, which analyzed 140 ADPKD children (VEO: 70, non-VEO: 70) highlighted a median age at hypertension onset of 20 years in the VEO group compared with 37 years in the non-VEO group [10], highlighting the difference in age onset of hypertension between both groups. However, small case reports studies have described VEO children with earlier onset of hypertension compared to the non-VEO group, occurring under the age of 10 years [19, 24]. Hence, our findings reinforce previous observations that hypertension may develop early in the disease course of children with VEO-ADPKD and suggest recommendations for early and systematic blood pressure monitoring in this high-risk population, in accordance with the latest International Consensus Statement on the diagnosis and management of ADPKD in children and young people [30].

Additionally, our meta-analysis showed that VEO-ADPKD patients had significantly increased rate of proteinuria when compared to the non-VEO group (Fig. 3). This finding suggests that VEO-disease may be associated with earlier or more pronounced kidney involvement. Notably, our results are consistent with those reported by Boyer et al., who similarly observed an increased rate of proteinuria in patients with VEO-ADPKD [12]. Collectively, these findings support the hypothesis that VEO-ADPKD represents a more severe clinical phenotype, characterized by early markers of kidney damage.

Kidney outcomes were also considerably worse among children with VEO disease (Fig. 4). The prevalence of CKD and progression to kidney failure was significantly higher in the VEO group compared with non-VEO patients. These findings align with earlier cohort studies and latest 2025 KDIGO guidelines for the evaluation, management and treatment of patients with ADPKD reporting accelerated loss of kidney function in children presenting with antenatal or infancy-onset ADPKD [7, 10, 31]. Novak et al. study, which is one of the strongest studies in this systematic review reported a median age at CKD onset of 22 years and 25 years in the VEO and non-VEO group respectively; however, data from smaller studies have documented the occurrence of CKD in VEO children at much younger ages (< 5 years) [10, 13, 14, 17]. Interestingly, a few studies reported kidney failure onset even at smaller ages (Range: 3.5-20y) in the VEO group. Only 1 out of 410 non-VEO patients was reported with kidney failure at the age of 10 years among all 19 studies.

Finally, perinatal death was found in significantly higher rate in the VEO-ADPKD children (34/320), which also highlights the severity and worse outcome of antenatal detected ADPKD. Genetic information was unavailable for 20 of these 34 cases; however, among the genetically characterized patients, the majority carried biallelic PKD1 mutations. Notably, no cases of perinatal death were reported in the non-VEO group, highlighting the markedly more severe clinical course associated with VEO-ADPKD.

Strengths and limitations

To our knowledge, this is the first systematic review and meta-analysis, which provides a comprehensive review of the genetic profile and kidney outcomes of paediatric VEO and non-VEO-ADPKD children.

This systematic review has several limitations. Most included studies were retrospective and observational, with moderate to high risk of bias. There was substantial heterogeneity between studies in design, patient selection, diagnostic criteria, follow-up duration, genetic testing availability, and outcome definitions. Differences in referral patterns, surveillance intensity, and diagnostic practices may also have contributed to variability between VEO and non-VEO cohorts. Although a Bayesian random-effects model was used to account for between-study heterogeneity, the pooled estimates should be interpreted cautiously.

Moreover, the limited availability of genetic testing in older studies may have introduced diagnostic heterogeneity, including the potential inclusion of phenocopies or alternative hereditary cystic kidney disorders. This limitation may have affected the accuracy of genotype-phenotype correlations and the classification of some patients, especially in old cohorts. Additionally, VEO-ADPKD was defined by age at diagnosis, which may have been influenced by differences in prenatal imaging, screening practices, and study era, introducing potential ascertainment bias. Hence, mildly affected or asymptomatic VEO-ADPKD children may have remained undiagnosed or underreported especially in older studies, potentially leading to underrecognition of VEO cases and enrichment of more severe phenotypes within reported VEO cohorts. Nevertheless, diagnosis of ADPKD within the first 18 months of life still remains the most consistently reported and widely accepted definition of VEO-ADPKD in the current literature [10].

Outcomes such as hypertension, CKD, and kidney failure are age-dependent and may have been influenced by differences in follow-up duration and surveillance practices across studies. In addition, limited and inconsistent reporting of age at outcome onset precluded formal time-to-event analyses. Therefore, pooled estimates should be interpreted cautiously, particularly for long-term kidney outcomes, and reflect overall associations between VEO and non-VEO ADPKD rather than precise estimates of disease progression over time.

Pooled effect estimates could not be generated for biallelic PKD1 mutations and perinatal death because these outcomes occurred exclusively in the VEO group, resulting in complete separation and precluding meta-analytic estimation.

Finally, genetic analyses should be interpreted with caution, as comprehensive genetic testing was unavailable in many older studies and only a minority of included studies (7/19) reported detailed molecular data. In addition, genetic testing strategies and reporting of variant characteristics varied substantially across cohorts, limiting precise genotype-phenotype analyses.

Conclusion

This systematic review and meta-analyses provide a detailed comparison of genetic profiles and kidney outcomes between VEO and non-VEO ADPKD children for the first time. Our findings demonstrate that VEO-ADPKD may be associated with a more severe clinical course, characterized by a higher burden of maternal inheritance, biallelic PKD1 variants, earlier and more frequent hypertension and proteinuria as well as accelerated progression to CKD and kidney failure in young ages. However, these findings should be interpreted cautiously given the retrospective and heterogeneous nature of the available evidence.

In conclusion, these results support that VEO-ADPKD is a high-risk group requiring early recognition with genetic counselling and confirmation when clinically suspected. This review emphasizes the need for larger studies with genotype-phenotype correlation including prospective clinical follow-up to guide early therapeutic interventions where applicable, and ultimately improve outcomes for children with VEO-ADPKD.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

CK conceived the idea for the study, designed the study methodology, selected articles for inclusion and participated in the evaluation of the studies. CK collected all data and drafted the initial manuscript. GD participated in the collection and the evaluation of the studies. GD performed the statistical analysis in this systematic review and reviewed the initial draft. AK reviewed and revised the initial manuscript. MM revised the manuscript and approved the final manuscript as submitted.

Funding

The authors did not receive support from any organization for the submitted work.

Data availability

Most of the data analysed in this systematic review and meta-analysis are included within the manuscript or the supplementary files. Rest of the data can be provided by the corresponding author upon request.

Declarations

Ethics approval and consent to participate

NA.

Consent for publication

NA.

Competing interests

The authors declare no competing interests.