IgE‐mediated food allergy and upper urinary tract disorders in children: A matched study

Abstract

Background

Food allergy involves type 2 immune activation and epithelial‐barrier dysfunction beyond the gut and airway. Whether IgE‐mediated food allergy is associated with upper urinary tract disorders in children, and whether any association is anatomically specific is unknown.

Methods

In a frequency‐matched cross‐sectional study using Israeli health‐maintenance organization records (2001–2025), 879 children (<16 years) with allergist‐confirmed IgE‐mediated food allergy were matched 1:4 to 3516 controls on age, sex, population sector, socioeconomic status, and body mass index. The primary outcome was a composite of upper urinary tract disorders (vesicoureteral reflux [VUR], hydronephrosis, or acute pyelonephritis); lower urinary tract and genital conditions served as negative controls. Odds ratios were estimated by logistic regression, adjusted for atopic dermatitis and asthma.

Results

The upper‐tract composite was more prevalent with food allergy (4.0% vs. 1.1%; OR 3.9, 95% CI 2.4–6.2). VUR (OR 12.1), hydronephrosis (OR 2.6), and acute pyelonephritis (OR 9.4) were each increased; lower urinary tract (OR 1.2) and genital conditions were not. Urology visit rates were similar between groups. After adjustment for atopic dermatitis and asthma, VUR (adjusted OR 9.2, 95% CI 4.5–16.5) and the composite (aOR 2.1) remained elevated, whereas hydronephrosis and pyelonephritis attenuated. Early‐life antibiotic exposure was only modestly higher (aOR 1.2).

Conclusion

IgE‐mediated food allergy was associated with upper, but not lower, urinary tract disorders; the VUR association was independent of atopic comorbidity. Because these anomalies are congenital, the pattern likely reflects shared susceptibility rather than causation, warranting prospective study.

Abbreviations

aOR

adjusted odds ratio

BMI

body mass index

CAKUT

congenital anomalies of the kidney and urinary tract

CI

confidence interval

EHR

electronic health record

FDR

false discovery rate

IgE

immunoglobulin E

LHS

Leumit Health Services

OR

odds ratio

SMD

standardized mean difference

UTI

urinary tract infection

VUR

vesicoureteral reflux

Key message.

In a matched study of 4395 children, IgE‐mediated food allergy was associated with upper, but not lower, urinary tract disorders (vesicoureteral reflux, hydronephrosis, pyelonephritis). The vesicoureteral reflux association remained strong after adjustment for atopic comorbidity, whereas hydronephrosis and pyelonephritis were largely explained by it. Because the structural components are congenital, this pattern points to shared developmental or immunologic susceptibility rather than causation, and merits prospective study.

1. INTRODUCTION

IgE‐mediated food allergy affects up to 8% of children and results from a loss of immune tolerance to food allergens, with allergen‐specific IgE production and type 2 immune activation. ¹ , ² Epithelial‐barrier dysfunction, particularly in the skin and gastrointestinal tract, is increasingly recognized as an important contributor to allergic sensitization and disease expression. ³ However, whether susceptibility involving epithelial and immune regulation is associated with disorders of the urinary tract remains largely unexplored.

The urinary tract is lined by a specialized urothelium that functions not only as a physical barrier but also as an active component of innate host defense through tight‐junction integrity, pathogen recognition, antimicrobial‐peptide production, and inflammatory signaling. ⁴ Pediatric upper urinary tract disorders encompass both congenital structural abnormalities, including vesicoureteral reflux (VUR) and hydronephrosis, and acquired infections such as acute pyelonephritis. VUR and many causes of hydronephrosis are included within the spectrum of congenital anomalies of the kidney and urinary tract and arise during fetal development, whereas reflux and obstruction may subsequently increase susceptibility to upper urinary tract infection. ⁵ , ⁶ Consequently, an association between food allergy and these structural conditions would not indicate that food allergy causes urinary tract malformation, but could instead reflect shared antecedent developmental, genetic, environmental, or immunologic susceptibility.

Empirical evidence linking food allergy to pediatric urinary tract disease is limited. Earlier reports proposed relationships between dietary hypersensitivity and bladder symptoms, but contemporary controlled data are sparse. A recent case–control study of 83 children found no significant difference in lower urinary tract symptoms or bladder‐related quality of life between children with laboratory‐confirmed food allergy and controls. ⁷ That study did not evaluate congenital upper urinary tract abnormalities or acute pyelonephritis. It therefore remains unknown whether IgE‐mediated food allergy is associated specifically with upper urinary tract disorders rather than with genitourinary conditions more broadly. Any observed association could also reflect the broader atopic phenotype, differences in antibiotic exposure, residual confounding, or greater contact with healthcare services.

We therefore conducted a population‐based, frequency‐matched study to examine the association between allergist‐confirmed IgE‐mediated food allergy and urinary tract disorders in children. The prespecified primary outcome was a composite of VUR, hydronephrosis, and acute pyelonephritis, with each component also evaluated separately. Anatomically distinct lower urinary tract and genital conditions were examined as comparison outcomes, and specialist‐contact rates were assessed to explore differential ascertainment.

2. METHODS

2.1. Study design and data source

We conducted a population‐based, frequency‐matched cross‐sectional study using electronic health records from Leumit Health Services (LHS), one of Israel's nationwide health‐maintenance organizations. LHS provides comprehensive medical services to approximately 750,000 members, and its centralized database contains longitudinal information on demographic characteristics, diagnoses, laboratory investigations, medication dispensations, and specialist consultations.

The study period extended from January 1, 2001, through December 31, 2025. Diagnoses recorded during each participant's period of enrollment in LHS were used to estimate the prevalence of urinary tract conditions. Because urinary tract diagnoses could have been recorded before or after the documentation of food allergy, the study was designed to assess prevalence associations and not temporal or causal relationships.

The study was approved by the LHS institutional review board (LEU 0014‐23), which waived the requirement for informed consent because the analysis used de‐identified routinely collected data. The study was conducted in accordance with the Declaration of Helsinki and is reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) recommendations.

2.2. Identification of IgE‐mediated food allergy

Eligible participants were children aged <16 years with a recorded diagnosis of food allergy during the study period. Potential cases were initially identified using International Classification of Diseases, Ninth Revision (ICD‐9) diagnostic codes and subsequently underwent structured manual chart review by board‐certified allergists using a standardized abstraction protocol.

IgE‐mediated food allergy was confirmed in accordance with European Academy of Allergy and Clinical Immunology diagnostic criteria. ⁸ Confirmation required all of the following: a documented immediate reaction compatible with food allergy occurring within 2 h after ingestion of the suspected food; objective evidence of sensitization to the implicated food, demonstrated by a positive skin‐prick test or elevated serum food‐specific IgE; and a physician‐documented diagnosis of IgE‐mediated food allergy accompanied by a recommendation for avoidance of the implicated food.

Oral food‐challenge results, when available, were considered confirmatory. Children with exclusively non–IgE‐mediated food allergy, food sensitization without a compatible clinical reaction, or insufficient documentation to confirm the diagnosis were excluded.

2.3. Control selection and matching

Each child with confirmed food allergy was frequency‐matched 1:4 to children without any food allergy diagnosis on age, sex, population sector (general, ultra‐orthodox Jewish, or Arab), socioeconomic status (1–20 area‐based scale), and body mass index (BMI). Sex and population sector were matched on identical categories, and age, socioeconomic status, and BMI were matched by distribution, so that the marginal distributions of all matching variables were closely balanced between groups. Matching on these factors, together with shared enrolment in a single‐payer system, was intended to minimize differences in baseline characteristics and access to care. Adequacy of covariate balance after matching was quantified with standardized mean differences (SMDs); an absolute SMD <0.10 was prespecified as indicating negligible residual imbalance. ⁹

2.4. Outcomes

The prespecified primary outcome was a composite of any upper urinary tract disorder, defined as the presence of ≥1 of vesicoureteral reflux (VUR; ICD‐9593.7), hydronephrosis (591), or acute pyelonephritis (590.1). These conditions were combined a priori because they are clinically linked along a single reflux–obstruction–infection axis and frequently co‐occur in the same child. Because the composite nonetheless combines two congenital structural anomalies (VUR, hydronephrosis) with an acquired infectious event (acute pyelonephritis), which have distinct pathogenesis, we prespecified that the three components would also be reported individually and that the structural anomalies and the infectious outcome would be interpreted as separate entities rather than as manifestations of a single mechanism.

Two negative‐control outcome groups ¹⁰ were specified to test anatomic specificity. The lower urinary tract group comprised conditions sharing the urothelial epithelium with the upper tract: cystitis (595); urethritis and urethral disorders (597, 598.91, 599); and lower urinary tract symptoms, including dysuria, urinary frequency, and incontinence (788). The genital group comprised embryologically and anatomically distinct conditions (balanitis, hydrocele, phimosis, testicular torsion, vulvovaginitis). Urinary tract infection of unspecified site (599.0) was analyzed separately because it cannot be localized to the upper or lower tract. All outcomes were ascertained over each child's entire observation period.

2.5. Assessment of differential ascertainment

To explore whether children with food allergy were more likely to receive specialist evaluation through which urinary tract abnormalities could be detected, we compared recorded contacts with pediatric urology, urologic surgery, and pediatric nephrology services between the study groups. These analyses were used to assess the possibility of differential diagnostic surveillance but were not considered sufficient to exclude all differences in imaging or diagnostic intensity.

2.6. Observation period and outcome ascertainment

All outcomes were identified from diagnoses recorded during each child's available period of enrollment in LHS. A participant with more than one diagnosis within an outcome category was counted once in the corresponding analysis. For composite outcomes, the presence of any component qualified the participant as having the outcome.

Because the study evaluated cumulative diagnoses across the available observation period, the analyses estimated associations with recorded disease prevalence. They did not determine whether urinary tract abnormalities preceded or followed the diagnosis of food allergy.

2.7. Statistical analysis

Baseline characteristics were compared using the Student t‐test or Mann–Whitney U test for continuous variables and the χ ² or Fisher exact test for categorical variables, and covariate balance was summarized with standardized mean differences (SMDs) as described above. The association between food allergy and each outcome was estimated as an odds ratio (OR) with a 95% confidence interval (CI). Because frequency matching does not create matched sets requiring a conditional analysis, unconditional (binary) logistic regression was used; primary (crude) estimates used food‐allergy status as the sole predictor. Because several outcomes were rare, CIs for these unadjusted estimates were computed using the Haldane–Anscombe continuity correction and Fisher exact methods rather than uncorrected Wald approximations, which are unstable when event counts are low. The false discovery rate was controlled within the genitourinary outcome family using the Benjamini–Hochberg procedure, and FDR‐adjusted q values are reported. Adjusted associations (Table 2 and the sensitivity analyses below) were estimated with Firth penalized logistic regression (logistf package in R), which remains stable for rare outcomes and in the presence of a zero cell, with profile penalized‐likelihood 95% CIs.

TABLE 2.

Association between IgE‐mediated food allergy and genitourinary conditions, unadjusted and adjusted for atopic comorbidity.

Condition (ICD‐9)	Food allergy, n (%)	Controls, n (%)	Unadjusted OR (95% CI)	Adjusted OR (95% CI) a	FDR q
Upper urinary tract composite b (primary)	35 (4.0)	37 (1.1)	3.9 (2.4–6.2)	2.1 (1.5–2.9)	<.001
Congenital structural anomalies
Vesicoureteral reflux (593.7)	9 (1.0)	3 (0.1)	12.1 (3.2–37.4)	9.2 (4.5–16.5)	<.001
Hydronephrosis (591)	21 (2.4)	33 (0.9)	2.6 (1.5–4.5)	1.2 (0.8–1.8)	.003
Other kidney/ureter disorders (593)	9 (1.0)	6 (0.2)	6.1 (2.2–16.0)	1.4 (0.7–2.5)	.003
Acquired infection
Acute pyelonephritis (590.1)	7 (0.8)	3 (0.1)	9.4 (2.4–30.8)	3.3 (0.9–6.4)	.003
Site‐unspecified
UTI, site unspecified (599.0)	47 (5.4)	137 (3.9)	1.4 (1.0–2.0)	1.0 (0.8–1.4)	.12
Lower urinary tract (negative control)
Lower urinary tract composite c	72 (8.2)	240 (6.8)	1.2 (0.9–1.6)	1.1 (0.9–1.3)	.30
Cystitis (595)	2 (0.2)	6 (0.2)	1.3 (0.4–6.6)	1.3 (0.3–3.8)	.73
Urethral disorders (597, 598.91, 599)	49 (5.6)	147 (4.2)	1.4 (1.0–1.9)	1.2 (0.9–1.6)	.30
Dysuria (788.1)	51 (5.8)	162 (4.6)	1.3 (0.9–1.8)	1.2 (0.9–1.6)	.30
Urinary incontinence (788.3)	16 (1.8)	71 (2.0)	0.9 (0.5–1.6)	0.7 (0.4–1.1)	.79
Genital conditions (negative control)
Balanitis (607.10)	5 (0.6)	7 (0.2)	2.9 (1.0–8.9)	2.0 (0.8–4.3)	.69
Hydrocele (603.9)	10 (1.1)	43 (1.2)	0.9 (0.5–1.9)	0.8 (0.4–1.4)	1.00
Phimosis (605.2)	0 (0.0)	13 (0.4)	0.0 (0.0–2.5)	n/e	.73
Testicular torsion (608.2)	2 (0.2)	3 (0.1)	2.7 (0.6–14.5)	2.2 (0.5–6.1)	.33
Vulvovaginitis (112.10)	5 (0.6)	6 (0.2)	3.3 (1.1–10.6)	2.0 (0.8–4.3)	.10

Note: Unadjusted odds ratios were estimated by logistic regression, with Haldane–Anscombe continuity‐corrected/Fisher exact intervals for rare outcomes.

Abbreviations: CI, confidence interval; OR, odds ratio; UTI, urinary tract infection.

^a Adjusted OR = food‐allergy odds ratio from Firth penalized logistic regression adjusting for asthma and atopic dermatitis, with profile‐likelihood 95% CIs; n/e, not estimable (0 food‐allergy cases). The false discovery rate (FDR) was controlled within the genitourinary family by the Benjamini–Hochberg method.

^b Primary outcome: ≥1 of vesicoureteral reflux, hydronephrosis, or acute pyelonephritis.

^c ≥1 of cystitis, urethral disorders, dysuria, urinary frequency, or incontinence.

Three prespecified sensitivity analyses addressed the principal alternative explanations and are presented as sensitivity rather than primary analyses because each conditions on a post‐exposure or downstream covariate (atopic comorbidity, antibiotic exposure) or on a subset of the cohort, which can introduce over‐adjustment or collider bias if treated as the primary estimand; the crude matched comparison therefore remains the primary analysis, with the adjusted models used to probe robustness. First, to address confounding by the broader atopic phenotype, models were adjusted for atopic dermatitis and asthma. Second, to address the possibility that early‐life antibiotic exposure and consequent dysbiosis drive both food allergy and urinary tract outcomes, we compared early‐life systemic antibiotic exposure (≥1 systemic antibiotic prescription and the number of prescriptions during the first year of life) between groups and considered it a potential confounder. Third, to address differential ascertainment (surveillance bias), ¹¹ we compared rates of pediatric urology, urologic surgery, and pediatric nephrology contact between groups. Analyses were performed in R; two‐sided p < .05 was considered significant.

3. RESULTS

3.1. Cohort characteristics

The cohort comprised 879 children with confirmed IgE‐mediated food allergy and 3516 matched controls (1:4 ratio; Table 1 and Figure 1). Matching produced groups that were balanced on age (mean 4.6 years), sex (37.8% female), population sector, socioeconomic status, and BMI (all absolute SMDs <0.10). Sex and population sector were matched exactly and were therefore identical between groups by design. As expected, children with food allergy had a higher prevalence of atopic dermatitis (31.6% vs. 9.5%; OR 4.4, 95% CI 3.7–5.3) and asthma (18.8% vs. 6.0%; OR 3.6, 95% CI 2.9–4.5) and higher total IgE and blood eosinophil levels (p < .001), consistent with a type 2–polarized phenotype.

TABLE 1.

Demographic and clinical characteristics of the study population.

Characteristic	Food allergy (n = 879)	Controls (n = 3516)	SMD	p
Age, mean ± SD, years	4.6 ± 4.2	4.6 ± 4.3	−0.01	.87
0–2 years, n (%)	376 (42.8)	1460 (42.0)	—	.52
3–9 years, n (%)	366 (41.6)	1456 (41.9)	—	.91
10–16 years, n (%)	137 (15.6)	559 (16.1)	—	.88
Female sex, n (%) a	332 (37.8)	1328 (37.8)	0.00	1.00
Population sector, n (%) a
General	460 (52.3)	1840 (52.3)	0.00	1.00
Ultra‐orthodox Jewish	348 (39.6)	1392 (39.6)	0.00	1.00
Arab	71 (8.1)	284 (8.1)	0.00	1.00
Socioeconomic status (1–20), mean ± SD	7.9 ± 3.5	7.9 ± 3.3	0.01	.85
BMI, mean ± SD, kg/m2	17.6 ± 4.3	17.6 ± 3.7	0.00	.81
Atopic dermatitis, n (%)	278 (31.6)	334 (9.5)	0.57	<.001
Asthma, n (%)	165 (18.8)	212 (6.0)	0.39	<.001
Total IgE, median [IQR], IU/mL	256.5 [136.8–618.0]	79.8 [25.6–196.0]	—	<.001
Blood eosinophils, median [IQR], ×109/L	0.4 [0.2–0.7]	0.3 [0.2–0.5]	—	<.001
≥1 systemic antibiotic, first year, n (%)	418 (47.6)	1519 (43.2)	0.09	.02
Systemic antibiotic Rx, first year, mean ± SD	1.5 ± 2.1	1.2 ± 1.9	—	<.001

Abbreviations: BMI, body mass index; IgE, immunoglobulin E; IQR, interquartile range; SD, standard deviation.

^a Frequency‐matched (identical category proportions). Continuous variables were compared by Student t‐test or Mann–Whitney U test; categorical variables by χ ² or Fisher exact test. Groups were matched 1:4 on age, sex, population sector, socioeconomic status, and BMI. Standardized mean differences (SMD) summarize post‐matching balance; |SMD| < 0.10 indicates negligible imbalance.

FIGURE 1.

Study flow diagram. Identification of children (<16 years) with an ICD‐9 food‐allergy code, allergist confirmation of IgE‐mediated food allergy, exclusions, and 1:4 matching to controls without food allergy, yielding 879 cases and 3516 controls.

3.2. Upper urinary tract disorders (primary outcome)

The composite upper urinary tract outcome was nearly four times as prevalent among children with food allergy (4.0% vs. 1.1%; unadjusted OR 3.9, 95% CI 2.4–6.2; p < .001) (Table 2). Considered as the two biologically distinct entities that make up this composite, both the congenital structural anomalies and the acquired infection were individually elevated and significant after FDR correction. Among the structural anomalies, VUR was strongly associated (1.0% vs. 0.1%; OR 12.1, 95% CI 3.2–37.4) as was hydronephrosis (2.4% vs. 0.9%; OR 2.6, 95% CI 1.5–4.5); the broader category of other disorders of the kidney and ureter (ICD‐9593), capturing additional upper‐tract structural pathology, was likewise increased (1.0% vs. 0.2%; OR 6.1, 95% CI 2.2–16.0). The infectious outcome, acute pyelonephritis, was also more frequent (0.8% vs. 0.1%; OR 9.4, 95% CI 2.4–30.8). With continuity‐corrected intervals, all primary associations excluded the null.

3.3. Anatomic specificity (negative controls)

Associations were confined to the upper tract. The lower urinary tract composite was not associated with food allergy (8.2% vs. 6.8%; OR 1.2, 95% CI 0.9–1.6), nor were its individual components, including cystitis (OR 1.3, 95% CI 0.4–6.6), urethral disorders (OR 1.4, 95% CI 1.0–1.9), dysuria (OR 1.3), and urinary incontinence (OR 0.9) (Table 2). Site‐unspecified UTI showed only a weak, non‐significant signal (5.4% vs. 3.9%; OR 1.4, 95% CI 1.0–2.0). Genital conditions—balanitis, hydrocele, phimosis, testicular torsion, and vulvovaginitis—were not significantly associated after FDR correction.

3.4. Assessment of ascertainment

Differential specialist evaluation did not explain the findings. Rates of pediatric urology visits (1.9% vs. 1.3%; OR 1.5, 95% CI 0.9–2.6; p = .21) and urologic‐surgery visits (2.8% vs. 2.3%; OR 1.2, 95% CI 0.8–2.0; p = .39) did not differ significantly between groups (Table 3). Thus, children with food allergy were not more likely to undergo the specialist evaluations through which VUR and hydronephrosis are typically detected, yet these diagnoses were markedly more frequent among them.

TABLE 3.

Pediatric urology and nephrology specialist contact, by group.

Specialist contact	Food allergy, n (%)	Controls, n (%)	OR (95% CI); p
Pediatric urology	17 (1.9)	47 (1.3)	1.5 (0.9–2.6); .21
Urologic surgery	25 (2.8)	81 (2.3)	1.2 (0.8–2.0); .39
Pediatric nephrology	3 (0.3)	2 (0.1)	6.0 (0.7–72.3); .07

Note: Comparable urologic specialist contact between groups argues against differential ascertainment as the explanation for the upper urinary tract associations.

Abbreviations: CI, confidence interval; OR, odds ratio.

3.5. Sensitivity analyses

After adjustment for atopic dermatitis and asthma (Firth penalized logistic regression), the associations diverged by outcome. The upper‐tract composite (aOR 2.1, 95% CI 1.5–2.9) and, most strikingly, vesicoureteral reflux (aOR 9.2, 95% CI 4.5–16.5) remained strongly and significantly elevated, indicating that the reflux signal in particular is largely independent of the broader atopic phenotype. In contrast, hydronephrosis (aOR 1.2, 95% CI 0.8–1.8), other kidney/ureter disorders (aOR 1.4, 95% CI 0.7–2.5), and acute pyelonephritis (aOR 3.3, 95% CI 0.9–6.4) attenuated toward the null and were no longer statistically significant. All negative‐control outcomes remained null after adjustment (Table 2). Early‐life systemic antibiotic exposure was only modestly more frequent among children with food allergy: 47.6% (418/879) versus 43.2% (1519/3516) had ≥1 systemic antibiotic prescription in the first year of life (crude OR 1.2; adjusted OR 1.2, 95% CI 1.1–1.4; SMD 0.09), with a mean ± SD of 1.5 ± 2.1 versus 1.2 ± 1.9 prescriptions. An exposure difference of this magnitude is quantitatively incapable of generating associations of the observed size (e.g., VUR OR 12.1), making antibiotic‐mediated dysbiosis an implausible sole explanation.

4. DISCUSSION

In this population‐based, matched study of 4395 children, IgE‐mediated food allergy was associated with a higher prevalence of upper urinary tract disorders—VUR, hydronephrosis, and acute pyelonephritis—captured by a prespecified composite outcome. The association was anatomically restricted: lower urinary tract and genital conditions showed no relationship. Three features strengthen the internal validity of this pattern. First, the upper‐tract components were directionally concordant, consistent with clustering along a reflux–obstruction–infection axis ⁵ rather than three independent claims. Second, the null associations across anatomically and embryologically distinct negative‐control outcomes argue against indiscriminate over‐detection, and comparable rates of urology and nephrology contact between groups argue specifically against surveillance bias ¹⁰ , ¹¹ as the principal driver. Third, the vesicoureteral reflux association and the upper‐tract composite persisted after adjustment for atopic dermatitis and asthma, whereas hydronephrosis and pyelonephritis attenuated toward the null—so the reflux signal specifically is not explained by the broader atopic phenotype, even though the other components are substantially atopy‐related.

4.1. Temporality and the congenital nature of the structural outcomes

An important consideration is the fundamentally different biology of the structural and infectious upper urinary tract outcomes. VUR and many cases of hydronephrosis belong to the spectrum of congenital anomalies of the kidney and urinary tract (CAKUT), the most common congenital malformations in children and a leading cause of chronic kidney disease during childhood and early adult life. ⁶ These abnormalities arise during fetal development through disturbances in ureteric bud induction, branching morphogenesis, nephrogenesis, or ureterovesical junction formation, and are frequently detected prenatally or during the first months of life. ⁶ In contrast, IgE‐mediated food allergy is an acquired immune‐mediated disease that typically develops after postnatal exposure to dietary antigens. ¹ , ² Consequently, food allergy cannot plausibly cause congenital malformations, and our cross‐sectional design does not permit inference about temporal sequence or causality.

Instead, our findings are more consistent with the hypothesis that congenital upper urinary tract abnormalities and IgE‐mediated food allergy may represent parallel manifestations of a shared developmental or constitutional susceptibility. Recent advances in developmental nephrology have shown that CAKUT is genetically heterogeneous, with pathogenic variants affecting genes involved in nephrogenesis, epithelial differentiation, ureteric‐bud branching, and urinary tract morphogenesis accounting for a substantial proportion of susceptibility. ⁶ , ¹² Gene–environment interactions and epigenetic regulation further contribute to phenotypic variability. ¹²

Although direct genetic links between CAKUT and food allergy have not been established, epithelial development and immune maturation are closely interconnected during fetal and early postnatal life. ³ , ¹³ Genome‐wide association studies of IgE‐mediated food allergy have identified susceptibility loci involved not only in adaptive immune regulation but also in epithelial barrier integrity, innate immune signaling, and tissue repair. ¹⁴ These observations raise the possibility that inherited abnormalities of epithelial development or epithelial–immune communication could increase susceptibility to both congenital urinary tract abnormalities and allergic disease, even though the two become clinically apparent at different stages of life. ¹³ , ¹⁴ At present, this hypothesis remains speculative and requires confirmation through integrated genetic, developmental, and mechanistic studies.

The persistence of the association between food allergy and VUR after adjustment for asthma and atopic dermatitis further suggests it is not solely explained by the broader atopic phenotype. Nevertheless, adjustment for coexisting allergic diseases cannot establish biological independence, and residual confounding by shared genetic or environmental factors cannot be excluded. ¹⁰ Accordingly, our findings should be interpreted as an epidemiologic association rather than proof of a shared pathogenic mechanism.

Acute pyelonephritis warrants separate consideration because its pathophysiology differs fundamentally from congenital structural anomalies. Pyelonephritis is an acquired bacterial infection, most commonly caused by ascending uropathogenic Escherichia coli, and its occurrence is strongly influenced by underlying urinary tract abnormalities, particularly VUR and obstruction. ⁵ , ¹⁵ Thus, part of the excess risk observed among children with food allergy may be mediated indirectly through the higher prevalence of coexisting congenital upper urinary tract abnormalities rather than representing an independent infectious susceptibility.

Beyond structural predisposition, the renal collecting system and upper urinary tract constitute an active component of the innate immune system rather than a passive conduit for urine. ⁴ Renal tubular epithelial cells and urothelial cells recognize invading uropathogens through pattern‐recognition receptors, produce antimicrobial peptides, cytokines, and chemokines, and coordinate both innate and adaptive immune responses during pyelonephritis. ⁴ , ¹⁵ Whether the epithelial immune dysregulation characteristic of allergic diseases contributes to susceptibility to upper urinary tract infection remains unknown, but represents a biologically plausible hypothesis for future investigation. ⁴ , ¹³ , ¹⁶ Future longitudinal studies integrating prenatal data, genomic analyses, and functional immune profiling will be required to determine whether these associations reflect common developmental pathways or independent processes that converge within susceptible individuals.

4.2. Antibiotic exposure and early‐life dysbiosis

An important non‐immunologic pathway is early‐life antibiotic exposure, which has been repeatedly associated with later allergic disease through disruption of the developing gut microbiome and immune maturation. ¹⁷ , ¹⁸ Antibiotics are also used to treat febrile urinary tract infection and, in selected children with vesicoureteral reflux, as continuous prophylaxis. ¹⁹ Antibiotic exposure could therefore confound the association in two ways: by promoting allergic disease through microbiome perturbation, and by increasing detection of congenital anomalies during urinary‐infection work‐up.

We examined this directly. Children with food allergy were only modestly more likely than controls to have received a systemic antibiotic during the first year of life (47.6% vs. 43.2%; adjusted OR 1.2, 95% CI 1.1–1.4), with a small difference in the mean number of prescriptions (1.5 vs. 1.2). A differential of this magnitude cannot account for associations of the size observed: a 1.2‐fold difference in exposure prevalence could not generate relative risks of 9–12. Early‐life dysbiosis is therefore unlikely to be the primary driver, although it may contribute to the acquired infectious component (pyelonephritis).

Antibiotic exposure may nonetheless act simultaneously as a confounder, mediator, and marker of healthcare utilization, depending on the temporal sequence of events—altering immune maturation, lying on a causal path to food allergy, or merely flagging children whose recurrent infections prompted the imaging that revealed a congenital anomaly. Because our design was cross‐sectional, we could not order these events. Prospective birth‐cohort studies integrating microbiome development, antibiotic exposure, allergic sensitization, and urinary tract imaging will be required to distinguish these possibilities and establish temporality.

4.3. Clinical context

Our findings are hypothesis‐generating and do not, on their own, justify a change in urologic practice or support broader urinary tract imaging in children with food allergy. Current guidelines already direct upper‐tract evaluation in children with recurrent or atypical urinary infections regardless of atopic history, ²⁰ and the absolute prevalence of upper‐tract pathology in our food‐allergy cohort, although elevated in relative terms, remained low. The principal value of these results lies in identifying an unexpected, anatomically specific association that warrants mechanistic and prospective investigation rather than immediate clinical action.

4.4. Strengths and limitations

Strengths include allergist‐adjudicated confirmation of IgE‐mediated food allergy, population‐based matched sampling within a single payer, prespecified composite and negative‐control outcomes, and direct measurement of specialist contact to probe ascertainment. Several limitations temper interpretation. First, and most importantly, the cross‐sectional design precludes inference about temporal sequence or causation; because the structural outcomes are congenital, the association cannot represent an effect of food allergy on urinary tract formation and must be read as reflecting shared susceptibility or, for the infectious outcome, downstream risk. Second, despite a large overall cohort, key outcomes were based on few events; even with continuity‐corrected intervals, the precision of individual rare‐outcome estimates (notably VUR) is limited, and point estimates should be interpreted with caution. Third, outcomes derived from administrative coding are subject to misclassification, and residual confounding from unmeasured factors—including incompletely captured early‐life antibiotic exposure—cannot be excluded. Fourth, although urologic specialist contact was comparable between groups, we cannot fully exclude subtler differences in imaging intensity. These limitations reinforce that our findings should be regarded as a signal for further study rather than as definitive evidence.

4.5. Conclusion

In this matched pediatric cohort, IgE‐mediated food allergy was associated with upper—but not lower—urinary tract disorders. The vesicoureteral reflux association persisted after adjustment for atopic dermatitis and asthma, whereas hydronephrosis and pyelonephritis were largely explained by atopic comorbidity. Because these structural anomalies are congenital, the pattern probably reflects shared susceptibility rather than causation and warrants prospective, mechanistically oriented study.

AUTHOR CONTRIBUTIONS

Eli Magen: Conceptualization; investigation; writing – original draft; methodology; writing – review and editing; formal analysis; data curation; validation; supervision. Shlomo Vinker: Methodology; validation; software; formal analysis; project administration; resources; writing – review and editing. Ilan Green: Investigation; methodology; validation; writing – review and editing; formal analysis; data curation. Eugene Merzon: Investigation; methodology; validation; writing – review and editing; data curation; resources. Avivit Golan‐Cohen: Investigation; methodology; validation; software; formal analysis; data curation; writing – review and editing. Israel Magen: Conceptualization; investigation; writing – original draft; methodology; writing – review and editing; formal analysis; visualization. Ariel Israel: Methodology; validation; visualization; writing – review and editing; software; formal analysis; project administration; resources; supervision.

FUNDING INFORMATION

This research received no specific grant from any funding agency in the public, commercial, or not‐for‐profit sectors.

CONFLICT OF INTEREST STATEMENT

The authors have no conflicts of interest to declare.

ETHICS STATEMENT

The study was approved by the institutional review board of Leumit Health Services (LEU 0014‐23). The board waived the requirement for informed consent for analysis of de‐identified data. The study conformed to the principles of the Declaration of Helsinki.

Magen E, Magen I, Merzon E, et al. IgE‐mediated food allergy and upper urinary tract disorders in children: A matched study. Pediatr Allergy Immunol. 2026;37:e70481. doi: 10.1111/pai.70481

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are not publicly available because they are derived from electronic health records governed by Leumit Health Services' privacy regulations. Aggregated data are available from the corresponding author upon reasonable request and subject to institutional approval.