This retrospective cohort study investigated the associations between various maternal diseases and CAKUT detected antenatally in offspring. Statistical analysis showed that gestational diabetes, thyroid dysfunction, neuropsychiatric disease, anemia, and ovarian and uterine disorders are associated with antenatally detected CAKUT. The associations showed differing patterns between patients with isolated UTD and KA. The risk of isolated UTD was increased in the fetuses of mothers with gestational diabetes, gestational hypertension, thyroid dysfunction, or hepatic disease. In contrast, the risk of KA was increased in the fetuses whose mothers had kidney disease, thyroid dysfunction, gestational diabetes, or ovarian/uterine disorders.

The antenatal detection ratio of CAKUT in the present study was 5.8 per 1000 pregnancies, which is within the lower range reported previously [8,9,10]. The prevalence of CAKUT in recent population studies was reported to range from 0.4% to 1.1% in term infants and 2.0% in preterm infants [5]. Variations between different reports may be associated with the criteria of CAKUT diagnosis, the database used for information retrieval, variations in sample size, and ethnic differences. From 2020 to 2022, the birth rate in China changed from 8.52‰ to 6.77‰ [11]. At the end of 2022, Tai’an had a permanent population of 1.25 million with a birth rate of 4.05 per 1000 population. Based on the estimation, the cohort of 19,656 pregnant women who participated in this study represented a significant proportion of the pregnant population in Tai’an City from 2020 to 2022.

Antenatal UTD is one of the most common fetal anomalies detected during pregnancy [1, 12]. It is commonly observed that the majority of dilatations tend to resolve without intervention, which is referred to as physiological or transient dilatation [12]. Nevertheless, there are some instances where UTD may be the initial indication of an underlying urinary tract anomaly [13, 14]. In the present study, the antenatal detection rate of isolated UTD was found to be 4.5‰, which was comparable with previous reports [15, 16]. These results provided further consultations for postnatal follow-up among the offspring. A meta-analysis found that 12% of infants with mild dilatation, 45% with moderate dilatation, and 88% with severe fetal dilatation were postnatally diagnosed with CAKUT [17].

The KA detected antenatally in this study were predominantly characterized by the malformations of hypo/dysplasia, ectopia, and lower urinary tract obstruction, with a prevalence of 1.3‰. Individuals with reduced nephron endowment, which is a manifestation of CAKUT, can be identified as having chronic kidney disease (CKD) in adulthood. We provided perinatal close follow-up for pregnant mothers and fetuses [1]. Postnatal monitoring of kidney function, electrolyte level, and ultrasound scanning were prescribed for the newborns. KA are high-risk factors associated with acute kidney injury in neonates [18].

In our study, maternal gestational diabetes was associated with isolated UTD as well as with KA in the fetus. Several epidemiological studies have provided evidence of a correlation between maternal diabetes and congenital malformations overall, as well as kidney malformations [8, 19]. The correlation may be attributed to maternal hyperglycemia. In animal models, maternal hyperglycemia has been shown to have a negative impact on fetal kidney development, including ureteric bud branching morphogenesis and nephron endowment [20, 21]. The significance of hyperglycemia during the early stage of pregnancy, when a process of active nephrogenesis occurs, may be supported by the significantly high likelihood of KA (adjusted OR of 6.8 shown in our data) in fetuses of mothers with gestational diabetes. A recent study showed the association of hemoglobin A1c with complicated CAKUT with extrarenal phenotypes [8]. The levels of hemoglobin A1c during early pregnancy may be lower, even in cases of gestational diabetes, compared to the nonpregnant state [8]. In addition, exposure to such levels could be sufficient to increase the risk of adverse pregnancy outcomes [22].

The risk of isolated UTD or KA was also increased in the fetuses of mothers with thyroid dysfunction. The expected incidence of gestational hypothyroidism or hyperthyroidism ranges from 0.3% to 11%, depending on the definition and geographical area [23]. We have presented a prevalence of 2.2% for maternal thyroid dysfunction. As is well known about the association of maternal hypothyroidism and preterm birth [24], the findings of preterm birth associated with CAKUT [5] may be due to maternal thyroid dysfunction. A cohort from another city in China documented the associations between maternal hypothyroidism and postnatal CAKUT in either preterm infants or term infants [6]. Further investigation should be conducted on the potential teratogenic effects of drugs which are contraindicated for pregnant women but might be used if the pregnancy was unplanned or suboptimally managed.

The nomogram presents a novel approach to providing personalized predictive information, based on risk factors that have demonstrated significant differences in multivariate analysis. Currently, in the field of CAKUT, most studies have utilized logistic regression to identify risk factors, yet a predictive model has yet to be established. Here, we built a nomogram for predicting antenatal CAKUT using a new statistical method, which can extract as much information as possible to provide the most accurate predictions from maternal factors. The nomogram prediction model could be a useful tool for estimating the maternal risks of CAKUT in offspring. It would provide an avenue for targeted screening and close follow-up of individuals with a high risk of CAKUT.

Several limitations of our study should be noted. Firstly, the risk factors associated with CAKUT during the fetal stage may be missed in situations such as miscarriage, termination of pregnancy, stillbirth, or death of fetuses prenatally diagnosed with KA. Secondly, since the diagnoses of maternal diseases were partially obtained from self-reported questionnaires, some diagnoses may have been overlooked, biased, or inconsistent, especially in cases of preexisting chronic hypertension, thyroid dysfunction, or diabetes. Furthermore, the diagnosis of maternal kidney disease was not specified (e.g., CAKUT, CKD, chronic nephritis, and proteinuria were possible for “kidney disease”). There was recall bias due to the reliance on self-reported maternal health conditions. In addition, maternal health conditions such as neurological disease and anemia might be underreported. Collection of medication history would help with the identification of maternal diseases. Thirdly, we were able to examine the association of maternal diseases with UTD and KA detected antenatally. It has been demonstrated that children with any degree of antenatal hydronephrosis are at greater risk of postnatal pathology compared to the normal population [12, 17]. Postnatal follow-up for the offspring in another ongoing study would provide further knowledge on the associations of the risk factors and postnatal diagnosis of CAKUT. Finally, the risk of KA was increased in fetuses of mothers with kidney disease in this study. This finding is significant because 10%–15% of cases with CAKUT show familial aggregation [1]. In the absence of genetic testing, it is not possible to definitively ascertain the veracity of consolidating all instances of isolated CAKUT.

In conclusion, maternal diseases, especially gestational diabetes mellitus and thyroid dysfunction, are highly associated with CAKUT in offspring. Various subtypes of CAKUT were associated with different maternal diseases. Our results may facilitate the early identification of women at the highest risk of bearing children with CAKUT and the provision of more specific preconception counseling and subsequent follow-up for newborns.