Introduction
Regular and timely antenatal follow-up, along with the widespread use of detailed ultrasonography, has improved the detection rate of congenital diseases during pregnancy. Congenital abnormalities of the kidney and urinary tract (CAKUTs) encompass a broad spectrum of structural disorders affecting the kidneys and collecting system [1]. With advances in antenatal screening, many cases can now be identified via ultrasonography at 18–20 weeks of gestation. CAKUTs occur in approximately 0.4–4 per 1,000 pregnancies and represent the leading cause of chronic kidney disease (CKD) in children (35%–60%) in developed countries, accounting for approximately 30% of end-stage kidney failure. Among CAKUT cases, 62% involve urinary system dilatation, 15% involve renal cystic dilatation, and the remaining 23% comprise other subgroups [2].
CAKUTs can lead to growth retardation, cognitive and psychosocial impairments, and increased morbidity and mortality due to multisystem complications. The lifelong need for costly treatments and the associated employment challenges faced by both patients and parents impose a significant economic burden on healthcare and insurance systems. These consequences underscore the importance of diagnostic, preventive, and therapeutic research—both experimental and clinical. However, data on the long-term outcomes of infants diagnosed with CAKUTs remain limited in the literature. This study aimed to elucidate the 1-year clinical course of infants diagnosed with CAKUTs during the antenatal period and to identify significant prognostic factors influencing outcomes. In this retrospective, observational cohort study, we aimed to establish a bridge between the antenatal and postnatal periods by assessing the predictive value of antenatal findings for postnatal morbidities, including acute kidney injury (AKI), CKD, urinary tract infections (UTIs), and the need for surgical interventions.
Methods
This single-center, retrospective cohort study included fetuses diagnosed with CAKUTs on antenatal ultrasound during Perinatology Council meetings between January 2014 and December 2020. Data were obtained from medical files and electronic hospital records. Infants born from these pregnancies who had at least 1 year of follow-up at our hospital were included, whereas those lacking adequate follow-up data were excluded.
Data collection
Maternal demographic and clinical characteristics—including age and presence of maternal and obstetric conditions such as hypertensive disorders, diabetes mellitus, and obesity—were recorded. Prenatal data included the presence of oligohydramnios, polyhydramnios, and intrauterine growth restriction (IUGR); a positive family history of CAKUTs; consanguinity; and results of antenatal genetic analyses. Natal data included gestational age (GA), sex, birth weight, mode of delivery, fifth-minute Apgar score, and the need for aggressive resuscitation at birth (defined as positive-pressure ventilation via bag and mask or endotracheal tube, chest compressions, or drug administration). GA was determined based on the maternal history of the last menstrual period and confirmed via ultrasonographic examination. Infants were classified as small for GA (SGA) if their birth weight was below the 10th percentile according to Fenton’s 2013 growth charts [3]. Admission to the neonatal intensive care unit (NICU), need for respiratory support (noninvasive or invasive), need for inotropic support, and exposure to nephrotoxic medications were recorded. Postnatal anthropometric measurements (weight, length, and head circumference) and corresponding percentiles were recorded at birth, 1 month, 6 months, and 1 year of age.
Antenatal and postnatal CAKUT findings were categorized into three groups: (1) collecting system anomalies (CSAs; ureteropelvic junction obstruction [UPJO], ureterovesical obstruction, ureterocele, ectopic ureter, bladder diverticulum, bladder exstrophy, urethral atresia, posterior urethral valves [PUVs], vesicoureteral reflux [VUR], and hydronephrosis [HN]); (2) renal parenchymal malformations (RPMs; renal agenesis, renal hypoplasia, renal dysplasia, multicystic dysplastic kidney [MCDK], autosomal recessive polycystic kidney disease, and autosomal dominant polycystic kidney disease; and (3) embryonic migration anomalies (ectopic kidney and horseshoe kidney).
The timing and laterality (right, left, or bilateral) of CAKUT diagnosis, along with antenatal and postnatal ultrasonographic findings—including kidney size, parenchymal thinning, renal pelvis anteroposterior diameter (RPAPD), presence of HN relative to GA, caliectasis, ureteral dilatation, and good prognostic criteria (absence of cortical cysts, normal renal echogenicity, and preservation of corticomedullary differentiation)—were recorded. According to the internationally accepted urinary tract dilation (UTD) criteria published in 2014 as a multidisciplinary consensus, RPAPD values below 4 mm between 16 and 27 weeks of gestation (second trimester) and below 7 mm at ≥28 weeks (third trimester) were considered normal [4]. Based on these criteria, antenatal HN cases were classified as either low risk or high risk. Low risk was defined as an RPAPD of 4–7 mm at 16–27 weeks or 7–10 mm at ≥28 weeks, with or without central calyceal dilatation. High risk was defined as an RPAPD of ≥7 mm at 16–27 weeks or ≥10 mm at ≥28 weeks, accompanied by peripheral calyceal dilatation, parenchymal thinning, ureteral or bladder anomalies, or unexplained oligohydramnios [4].
During the 1-year follow-up, ultrasonographic findings obtained at the first week, first month, sixth month, and first year were reviewed. Infants were categorized into low-, moderate-, or high-risk groups based on postnatal assessments. Cases with HN and an RPAPD of 10–14 mm with isolated central calyceal dilatation were classified as low risk. Cases with an RPAPD of >15 mm accompanied by central calyceal dilatation and a ureteral anomaly were classified as moderate risk. Cases with an RPAPD of >15 mm associated with parenchymal thinning, parenchymal anomalies, or bladder anomalies were categorized as high risk [4].
Kidney size was evaluated according to GA [5]. Voiding cystourethrography was performed in infants with bilateral HN, ureteral dilatation, PUVs, or recurrent UTIs. The severity of VUR was graded according to the criteria of the International Reflux Study Committee [6]. The presence of obstruction and ureteral involvement was further assessed using MAG3 (technetium-99m mercaptoacetyltriglycine) renal scintigraphy.
Assessment of renal function
Renal function was evaluated based on urine output (UO; mL/kg/day); estimated glomerular filtration rate (eGFR; calculated using the Schwartz formula); and biochemical parameters including blood urea nitrogen (BUN) levels, serum creatinine (SCr) levels, and urinalysis findings such as hematuria and proteinuria. These parameters were measured at the first week, first month, sixth month, and first year of life. SCr levels were determined using the uncompensated Jaffe method, a colorimetric assay based on the reaction of creatinine with alkaline picrate [7]. eGFR was calculated using the Schwartz formula as follows: eGFR (mL/min/1.73 m2)=0.45×length (cm)/plasma creatinine (mg/100 mL) [8]. UO was assessed over a 24-hour period and expressed in mL/kg/hr. Oliguria was defined as a UO of <1 mL/kg/hr. Because low UO is physiologic within the first 24 hours after birth, reductions during this period were not considered a criterion for oliguria. Hematuria and proteinuria were identified based on urinalysis results.
Definitions of AKI and CKD
AKI during the neonatal period was defined using the neonatal Risk, Injury, Failure, Loss, and End-stage Kidney Disease (nRIFLE) criteria [9]. For infants aged 1–12 months, the pediatric-modified RIFLE (pRIFLE) criteria were applied [10]. The only distinction between pRIFLE and nRIFLE lies in the UO thresholds, which are summarized in Supplementary Table 1.
CKD was diagnosed according to the Kidney Disease: Improving Global Outcomes classification [11]. During follow-up, the occurrence and frequency of UTIs and the need for prophylaxis were recorded. UTI was defined as the growth of ≥100,000 colony-forming units of a single microorganism in a urine culture obtained through transurethral catheterization, regardless of symptoms [12]. Cases of asymptomatic bacteriuria were also included. Recurrent UTI was defined as two or more episodes of pyelonephritis, one episode of pyelonephritis with one episode of cystitis, or three or more episodes of cystitis [13]. The primary outcomes included the incidences of AKI, CKD, and UTI as well as the identification of prognostic indicators linking the antenatal and postnatal periods.
Statistical analysis
Statistical analyses were performed using the Statistical Package for the Social Sciences for Windows (version 23.0). Data normality was assessed using the Shapiro–Wilk test. Results are expressed as mean ± standard deviation (SD), median (interquartile range), frequency, or percentage, as appropriate. For descriptive statistics, normally distributed data are presented as mean±SD, non-normally distributed data as median (interquartile range), and categorical variables as percentages. Group comparisons were conducted using the t test for normally distributed data and the Mann-Whitney U test for nonparametric data. Categorical variables were analyzed using the chi-square test. Repeated-measures analysis of variance was used to assess within-group differences over time, with pairwise comparisons performed using the least significant difference test. Univariate analyses were conducted using the chi-square or Mann-Whitney U tests, as appropriate. Variables with a P-value of <0.05 in the univariate analysis were entered into a multivariate logistic regression model to identify independent predictors of AKI. Odds ratios (ORs) with 95% confidence intervals (CIs) were reported, and a P-value of <0.05 was considered statistically significant.
Results
The study included fetuses diagnosed with CAKUTs between January 2014 and December 2020 who were delivered at our hospital and followed postnatally for at least 1 year. Fifty-five infants met the inclusion criteria: 30 in the CSA group and 25 in the RPM group. One infant with an embryonic migration anomaly was evaluated within the RPM group. The study flowchart is presented in Fig. 1.
Antenatal and postnatal CAKUT types are summarized in Table 1. Antenatally, 30 infants were classified in the CSA group and 25 in the RPM group; postnatally, these numbers were 31 and 24, respectively. Among the 30 fetuses with antenatal CSAs, 27 (90%) were confirmed postnatally, whereas 21 of 25 (84%) with antenatal RPMs remained in the same group. Misclassifications included three cases of MCDK initially assigned to the CSA group and two cases of VUR, one case of PUV, and one case of UPJO initially assigned to the RPM group. Fetuses identified with antenatal HN were further subclassified postnatally as HN without obstructive uropathy, UPJO, PUV, or VUR.
Antenatally, bilateral involvement was more frequent in the CSA group (56.7%) than in the RPM group (12%; P=0.002). In the CSA group, 68.9% of fetuses with HN were classified as high risk, compared with two of three in the RPM group. Notably, 77% of antenatal high-risk cases remained high risk postnatally. Conversely, 50% of antenatal low-risk fetuses were reclassified as high risk after birth, despite no antenatal interventions. The demographic and clinical characteristics of the study groups are summarized in Table 2. No significant differences were observed between the groups with respect to maternal characteristics; birth weight; sex; GA; mode of delivery; fifth-minute Apgar score; GA at CAKUT diagnosis; or the incidences of AKI, CKD, UTI, and NICU admission. Although not statistically significant, NICU admission and the need for respiratory support were more frequent in infants with RPM. A notable finding was that CKD was more prevalent in the RPM group (95.8%) than in the CSA group (61.3%; P=0.008). CKD stage classification remained consistent at both 6 and 12 months of age. Another significant difference was that infants in the CSA group received prophylaxis after a UTI more frequently (70%) than those in the RPM group (25%; P=0.004). Additionally, 61.3% of infants in the CSA group required surgical intervention within the first year of life due to the nature of their CAKUT type, compared with 16.6% in the RPM group (P=0.002). Four patients were diagnosed with syndromic disorders, including VACTERL association, Alagille syndrome, Bardet–Biedl syndrome, and Ochoa syndrome.
Analysis of anthropometric measurements (weight, length, and head circumference) at birth, 1 month, 6 months, and 1 year revealed no significant differences in body weight or head circumference between the groups. However, the mean birth length of the CSA group (49.23±2.52 cm) was significantly greater than that of the RPM group (47.37±3.25 cm; P=0.022), and this difference remained significant at 1 year of age (P=0.027) (Fig. 2).
Among the 32 fetuses with antenatal HN, 10 were classified as low risk and 22 as high risk (Table 2). When patients with HN were evaluated over 1 year according to antenatal risk groups, the differences in RPAPDs were statistically significant from the antenatal period onward, except at 6 months. Although 13 of the 22 high-risk patients (59.1%) underwent surgical intervention within the first year, the mean RPAPD at 1 year remained significantly higher than that of low-risk patients (P=0.001). A comparison of RPAPDs in antenatal low- and high-risk patients with HN is presented in Fig. 3.
No significant between-group differences were observed in BUN or SCr levels or eGFR throughout follow-up. Within-group analysis showed higher SCr levels during the first week compared with later time points, with no further significant changes thereafter. Trends in SCr levels over 1 year are shown in Fig. 4. Additionally, no correlation was found between maternal SCr levels and the initial SCr levels of the infants (P>0.05).
During the neonatal period, AKI was observed in four patients: three in the RPM group (two with MCDK and one with polycystic kidney disease) and one in the CSA group (with VUR). Antenatal oligohydramnios was present in three of these four patients. Among the 11 fetuses diagnosed with oligohydramnios, three developed neonatal AKI (P=0.022). Notably, none of the patients who developed AKI had favorable prognostic indicators identified antenatally.
Across the entire cohort of 55 patients, 8 developed AKI within the first year of life, all belonging to the high-risk subgroup. Infants with AKI were more likely to have antenatal oligohydramnios than those without AKI (P=0.042) and had a significantly lower median birth length (P=0.019). Postnatal ultrasonography revealed that AKI was more frequently associated with bladder trabeculation (P=0.034) and loss of corticomedullary differentiation (P=0.018). Furthermore, MAG3 scintigraphy demonstrated a nonobstructive pattern in five of the eight patients with AKI (P=0.018) (Table 3). These findings indicate that GA, presence of oligohydramnios, and specific ultrasonographic findings (including loss of corticomedullary differentiation, bladder trabeculation, and cysts) are important factors associated with the development of AKI. In multivariate logistic regression analysis, loss of corticomedullary differentiation remained an independent predictor of AKI (OR, 13.5; 95% CI, 1.8–100.0; P=0.011), whereas oligohydramnios, bladder trabeculation, and other factors did not retain statistical significance.
Discussion
Congenital anomalies of the urinary system range from asymptomatic lesions detected incidentally to severe cases progressing to end-stage kidney disease requiring dialysis. Early diagnosis and timely intervention in progressive cases are critical. In our cohort of 6,991 pregnant women evaluated over 7 years, CAKUTs were identified in 167 fetuses, corresponding to an incidence of 2.38%. In comparison, a Danish population-based study reported an incidence of 0.39% among 50,193 births and 24 terminated fetuses [14]. The higher incidence in our study likely reflects our role as a tertiary referral center, which concentrates high-risk pregnancies from surrounding provinces. In our study, the most frequent antenatal anomaly was HN (47.2%), followed by MCDK (38.2%). This pattern is consistent with findings from a Swiss study, wherein HN was the most common CAKUT subtype among 174 fetuses [15].
Intrauterine environmental factors, genetics, and maternal health are believed to contribute to the development of CAKUTs. In our cohort, six patients were diagnosed with IUGR, five of whom were in the RPM group, where the incidence of CKD was notably high (95.8%). This finding may reflect fetal programming, as previously suggested [16]. Because nephrogenesis is completed by the 36th gestational week, preterm birth, IUGR, and low birth weight are all associated with an increased risk of CKD from infancy into adulthood [17]. Consistent with this, Hirano et al. [18] reported that low birth weight is linked to a higher risk of kidney failure.
Although genetic factors may contribute to CAKUTs, maternal health remains a key determinant. Diabetes, which affected over 422 million people globally in 2014 and is projected to reach 642 million by 2040, remains highly prevalent among pregnant women [19,20]. In our study, 16.4% of the mothers had diabetes. Population-based data report CAKUT prevalence rates of 3.7% in the United States, 4% in England, and 14.4% among Asians living in England [21].
A high concordance was observed between antenatal and postnatal diagnoses of CAKUTs—84% for RPMs and 90% for CSAs. Similar concordance rates have been reported in Switzerland (82% and 87%) and by Policiano et al. (88.8%), while another study reported an 87.2% predictive value for prenatal detection [15,22,23]. These findings support the reliability of antenatal diagnosis while emphasizing the importance of postnatal reevaluation, given the potential for false positives or negatives. Classification changes in our cohort were within acceptable limits, and all patients remained under CAKUT follow-up.
Among the 12 infants with HN, postnatal risk levels increased in four, decreased in three, and remained unchanged in five. Over a 1-year follow-up, RPAPD measurements differed significantly between the low- and high-risk groups, except at 6 months (Fig. 3). Although 59.1% of high-risk infants underwent surgery, their 1-year RPAPD remained significantly higher than that of low-risk infants. Although the UTD criteria include additional features beyond RPAPD, these differences are clinically significant.
A Swedish study that followed 71 children with antenatal UTD for 13–15 years reported no adverse outcomes in those with an RPAPD of ≤7 mm and no associated abnormalities, suggesting that follow-up may be unnecessary in such cases. However, among children with an RPAPD of >7 mm and/or additional urinary tract pathology, 15% had persistent dilation and 32%–39% developed kidney damage, with CAKUTs increasing the risk 14-fold [24]. In our study, the rates of AKI and CKD were higher (9.6% and 61.3%, respectively), likely reflecting the predominance (83.8%) of high-risk HN cases based on UTD criteria. These results emphasize the need for careful follow-up, even in infants who initially appear to be at low risk.
In the abovementioned Swedish cohort, 37.5% of children with an RPAPD of >7 mm received antibiotic prophylaxis, and febrile UTIs occurred in only 2.5% of cases [24]. In contrast, 38.7% of infants in our CSA group received prophylaxis, but the overall UTI incidence during the first year was substantially higher at 74.2%. This likely reflects the impact of routine monthly urine testing, which detected both symptomatic and asymptomatic infections. For comparison, a Brazilian study of 822 children followed for 43 months reported a 29.8% UTI rate, with recurrent UTIs occurring in 6.8%–16.4% of cases [25]. Despite our smaller sample size, recurrent UTIs were common: 30% overall, 42% in the CSA group, and 16.6% in the RPM group, with no significant difference between sexes.
Among the 55 infants, oligohydramnios was detected in 11 (20%) and emerged as a significant prognostic factor for AKI. Two of the four infants who developed AKI were preterm and had significantly lower mean GA compared with those without AKI (P=0.03), suggesting that reduced nephron endowment due to prematurity may contribute to AKI in addition to congenital anomalies. Antenatally, five fetuses met favorable prognostic criteria, and none developed AKI during follow-up. Conversely, all eight infants who developed AKI lacked these criteria. Although the sample size was limited, our findings indicate that favorable antenatal prognostic indicators are reliable predictors of positive outcomes, providing reassurance to clinicians.
In our cohort, loss of corticomedullary differentiation and bladder trabeculation were more frequently observed in infants who developed AKI during the first year. Multivariate logistic regression analysis revealed loss of corticomedullary differentiation as an independent predictor of AKI, whereas bladder trabeculation and other ultrasonographic findings did not reach statistical significance, likely due to the limited sample size. Previous studies have reported several prenatal and postnatal factors as predictors of poor renal outcomes in CAKUTs. For example, in patients with PUVs, loss of corticomedullary differentiation and increased parenchymal echogenicity predicted progression to stage 5 CKD [26]. In our study, loss of corticomedullary differentiation was associated only with AKI, possibly reflecting the smaller sample size. Other reported prognostic indicators include oligohydramnios, UTI, VUR, PUV, and proteinuria [1] as well as bilateral involvement [27]. Long-term follow-up studies have shown that 25% of patients with bilateral CAKUTs progress to end-stage renal disease within 20 years [28].
In our 1-year follow-up of 55 infants with CAKUTs, SCr levels were highest during the first postnatal week and generally stabilized by 1 month, reflecting normal postnatal renal adaptation and maturation of the eGFR [29]. Although previous studies have reported correlations between neonatal and maternal SCr levels [30,31], we observed no consistent association in our cohort, likely because most infants with CAKUTs had CKD despite normal maternal renal function. These findings suggest that altered SCr levels in infants with CAKUTs primarily reflects intrinsic fetal renal programming rather than maternal influence.
During the 1-year follow-up, body weight and head circumference were similar between groups, whereas infants in the CSA group had significantly greater length at birth and at 1 year compared with those in the RPM group. This difference may be explained by the higher prevalence of IUGR (20%) and SGA (12%) among infants with parenchymal malformations. Growth retardation in infants with CAKUTs may also be influenced by recurrent UTIs, surgical interventions, and hospitalizations; however, the absence of a healthy control group limits this assessment. As no prior studies have specifically evaluated growth in infants with CAKUTs, our findings on weight, length, and head circumference represent a novel contribution.
This study has several limitations, including its single-center, retrospective design and relatively small sample size. Its primary strength lies in its originality and comprehensive scope. To the best of our knowledge, it is the first study to categorize CAKUT types and compare their clinical courses from the antenatal period to the first year of life, integrating biochemical, radiological, and anthropometric data. Antenatal HN cases were further stratified by risk to enhance prognostic interpretation. The neonatal period represents a critical transitional phase—bridging antenatal development and postnatal renal maturation—during which kidney function evolves substantially, characterized by declining creatinine levels and increasing eGFR in infants with CAKUTs. By identifying practical prognostic markers, this study provides valuable insights for early risk assessment and long-term management.
In conclusion, loss of corticomedullary differentiation emerged as a strong predictor of AKI during the first year, surpassing the predictive value of creatinine levels alone. Given the low regression rates of high-grade HN, proactive multidisciplinary collaboration—including pediatric nephrologists and urologists within perinatology councils—together with timely family education, is essential for vigilant follow-up. Management of infants with CAKUTs should address not only renal function but also growth and overall development, with careful avoidance of nephrotoxic exposures to optimize long-term outcomes. These findings provide a practical framework for guiding patient care while acknowledging inherent uncertainties.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement
Print
