Significant kidney impairment in pediatric inflammatory bowel disease and practice points for pediatric gastroenterologists

Abstract

Objectives

Children with inflammatory bowel disease (IBD) have an increased risk of developing kidney disorders, which may cause significant kidney function impairment (SKI) or lead to chronic kidney disease (CKD). In this study we aimed to provide insights in causes and diagnoses of SKI cases and to provide recommendations for pediatric gastroenterologists for children with IBD and SKI.

Methods

Cases of SKI in children with IBD (<19 years) were collected from the international, prospective PIBD‐SETQuality Safety Registry. A monthly survey was sent to participating pediatric gastroenterologists to report cases of SKI (defined as an estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73m²). Additionally, a panel consisting of 16 members (including experts in pediatric IBD and nephrology) rated the most likely cause of the cases and formulated recommendations for screening, follow‐up and referral for children with IBD and SKI.

Results

Between November 1, 2016 and December 31, 2023, 42 cases of SKI were eligible for analysis. Tubulo‐interstitial nephritis (TIN) was the most common diagnosis (n = 15). Sixteen (38%) cases were confirmed with renal biopsy (10 cases of TIN). Twelve patients developed CKD. IBD medication was the most frequently reported cause (n = 15), however, there was low concordance between panelists about the most likely etiology (Fleiss' kappa 0.143 and 0.102).

Conclusions

This is the first prospective study to report cases of SKI in children with IBD. SKI may lead to CKD. Confirming the etiology of the SKI proved to be very challenging. The study provides recommendations for screening and follow‐up of SKI in children with IBD.

Trial Registration

NCT03571373, date of registration: June 27, 2018, URL: https://clinicaltrials.gov/study/NCT03571373

What is Known

• Children with inflammatory bowel disease (IBD) may have a higher risk of developing kidney disorders.

• Pediatric IBD guidelines lack clear recommendations for kidney function monitoring.

What is New

• This is the first study reporting detailed information of prospectively collected cases of significant kidney impairment (SKI) in children with IBD, including diagnoses and outcomes.

• Recommendations for screening and referral of kidney function impairment were created by a panel of experts in pediatric IBD and nephrology.

• Even with detailed review by experts identifying the most likely cause of SKI and chronic kidney disease is challenging.

1. INTRODUCTION

Children with inflammatory bowel disease (IBD) may have an increased risk of developing kidney disorders. ¹ , ² Several studies suggest a higher risk of kidney function impairment in children and adults with IBD. ³ , ⁴ , ⁵ , ⁶ , ⁷ The prevalence of kidney disorders in children with IBD, as reported by two retrospective studies, ranges between 4% and 15%. ⁵ , ⁸ Data from the prospective international Pediatric Inflammatory Bowel Diseases Network for Safety, Efficacy, and Treatment and Quality improvement of care (PIBD‐SETQuality) Safety Registry estimated the incidence rate of significant kidney impairment (SKI) was estimated to be 3.75 per 10,000 person‐years (95% confidence interval 2.74–4.99), making renal failure the second most reported adverse event (AE). ⁹ Kidney disorders in pediatric IBD are diverse and can include nephrolithiasis, glomerular disorders (e.g., IgA glomerulonephritis), tubulo‐interstitial nephritis (TIN), and secondary renal amyloidosis. ¹ , ²

While exact causes are often unclear, possible contributors include genetic susceptibility to autoimmune disease, systemic inflammation, immune dysregulation, disease complications, or side effects of IBD treatments such as 5‐aminosalicylates (5‐ASA) or biologics. ¹ , ² , ¹⁰ , ¹¹ , ¹² Kidney involvement can progress to acute or chronic kidney disease (CKD), underscoring the need for timely recognition and intervention. Despite this risk, there is limited guidance on kidney function monitoring in children with IBD, largely due to a lack of large‐scale prospective studies. ¹³ , ¹⁴

In this study we described frequency and characteristics of kidney disorders resulting in SKI in children with IBD. Based on the knowledge learnt we planned to provide practice points for pediatric gastroenterologists for screening, follow‐up, and referral of children with IBD and kidney function impairment.

2. METHODS

2.1. Ethics statement

The PIBD‐SETQuality project received approval from the Erasmus Medical Centre's research ethics committee in Rotterdam (MEC‐2016‐321). Ethical clearance for other participating regions or centers was obtained in line with local regulations and governance standards. Since no personally identifiable information was gathered, the committee granted a waiver for informed consent. ⁹ Patients and/or the public were not involved in the design, conduct, reporting or dissemination plans of this research.

2.2. Reporting of cases with significant kidney function impairment

Data of incident cases of kidney disorders were collected from the ongoing prospective PIBD‐SETQuality Safety Registry. ⁹ The registry commenced on November 1, 2016, and data for this analysis were collected until December 31, 2023. During this period, pediatric gastroenterologists from 168 hospitals worldwide participated in the registry (Supporting Information: Figure 1). For each reported kidney‐related case, physicians were asked whether it resulted in SKI, defined as an estimated glomerular filtration rate (eGFR) < 60 ml/min/1.73 m², corresponding to Kidney Disease: Improving Global Outcomes (KDIGO) CKD stages G3–G5. ¹⁵ The eGFR was calculated by the Schwartz formula. ¹⁶

2.3. Detailed assessment of significant kidney function impairment

Once SKI was reported, an automated complication‐specific follow‐up form was sent to the reporting physician (Supporting Information: Methods S12). This form, developed with input from a pediatric nephrologist, collected further clinical details on IBD status, kidney function, and medications. IBD‐related treatments were not classified as nephrotoxic medication for this analysis, though medications known to cause drug‐induced acute TIN, such as NSAIDs, allopurinol, antibiotics, and PPIs, were. ¹⁷ Physicians were able to state whether the creatinine was still elevated and elevated for more than 3 months at last available follow‐up. Duration of SKI was defined from the first abnormal creatinine (creatinine leading to eGFR < 60 mL/min/1.73 m²) until normalization (return of serum creatinine around baseline creatinine) or the last available measurement. CKD was defined as eGFR < 60 mL/min/1.73 m² lasting > 3 months. Patients who had an eGFR between 60 and 90 mL/min/1.73 m² for >3 months during follow‐up were labeled as “uncertain CKD” due to missing albuminuria data, which was not collected in our study. Cases with follow‐up < 12 weeks or unknown normalization status were excluded from CKD classification.

Physicians were asked to report the likely diagnosis behind SKI (e.g., pre‐renal SKI, TIN, glomerulonephritis, amyloidosis, nephrolithiasis, other, or unknown) and identify the most probable cause from a predefined list: (1) IBD itself: presence of IBD or an increased intestinal disease activity was the direct cause of the renal injury; (2) IBD medication: IBD medication was the direct cause of renal injury; (3) IBD related: IBD was considered as an indirect or underlying cause of renal failure, non‐IBD related; (4) unknown; or (5) other. They also rated causality likelihood using terms adapted from the Liverpool Causality Assessment Tool: definite, probable, possible, very unlikely, or unclassifiable. ¹⁸ Data were collected anonymously via the REDCap. ¹⁹

2.4. Panel meetings

A panel was established with as main goals (i) to assess the likely causes and likelihood of SKI, and (ii) to develop consensus‐based practice points for screening, follow‐up, and referral to pediatric nephrologist. The panel consisted of 4 of the study team, a pediatric nephrologist (FV), a pediatrician with a special interest in nephrology (SM), and 10 pediatric IBD experts who had reported cases (TS, AG, SH, MC, PM, PD, MW, NM, SvB, and MF). Two virtual meetings were held (May 24 and July 5, 2024).

In the first meeting, background information on kidney disorders in children with IBD was provided and followed by a discussion of the definitions of the most likely causes. After this meeting, the panelists were divided into two groups (group A and B). Each member was asked to individually rate the most likely cause and likelihood of half of the reported cases with SKI. Cases were anonymized, and relevant information on patient demographics, medication, IBD disease activity, renal diagnostics (including creatinine values over time) were provided. The forms were filled out by 14 of the 16 panel members. Before the second meeting, the core team drafted proposed clinical recommendations based on literature and expert experience. These were discussed and refined during the second panel meeting, which was attended by 10/16 panel members.

2.5. Statistical analysis

Continuous variables were summarized as medians with interquartile ranges and compared using the Mann–Whitney U or Kruskal–Wallis tests. Categorical variables were reported as proportions and compared using Chi‐square or Fisher's exact tests, as appropriate. To compare the sex and disease type distribution of SKI patients with the general pediatric IBD population, binomial proportion tests were conducted using reference values from the EUROKIDS registry. ²⁰ Concordance between reporting physician and panelists was calculated based on the Fleiss' Kappa coefficient. ²¹ , ²² All analyses were performed using R Statistical Software (version 4.3.3; R development Core Team 2021, Vienna). Relevant R packages are listed in the Supporting Information: Methods S12.

3. RESULTS

3.1. Demographics and diagnosis of cases with significant kidney function impairment

Between November 1, 2016, and December 31, 2023, 42 cases of SKI were included (Table 1 and Supporting Information: Figure 2). The proportion of males and patients with CD were not significantly different compared to the general pediatric IBD population (p = 0.69, p = 0.62, respectively). The most likely diagnosis were TIN (n = 15), pre‐renal kidney injury (n = 8), glomerulonephritis (n = 6), unknown (n = 6), other (n = 5), and nephrolithiasis (n = 2). In 16/42 cases, the diagnosis was confirmed by renal biopsy: 10/15 cases of TIN, 5/6 cases of glomerulonephritis, and a sole case of acute tubular necrosis (ATN). Previous medication and medication prescribed at the time of first abnormal creatinine are provided in Supporting Information: Table 1.

Table 1.

Demographics and IBD characteristics at the time of significant kidney impairment.

	Total [n = 42]	CD [n = 26]	UC/IBD‐U [n = 16]	p‐value (CD vs. UC/IBD‐U)
Male sex	29 [69%]	22 [85%]	7 [44%]	0.01
White ethnicity	36 [86%]	22 [85%]	14 [88%]	0.65
Age at significant kidney impairment [y]	15.2 [13.0–16.9]	15.3 [13.1–17.0]	15.0 [13.0–16.6]	0.82
BMI	20.0 [16.4–23.0]	20.2 [16.6–24.2]	18.9 [16.2–22.0]	0.51
History of kidney diseasea , d	6 [14%]	3 [12%]	3 [20%]	0.75
Administration of nephrotoxic medication	7 [17%]	3 [12%]	4 [25%]	0.40
History of intestinal surgeryd	5 [12%]	5 [19%]	0	0.14
Duration of IBD at onset of significant kidney impairment [m]d	31.0 [7.0–68.0]	35.5 [13.3–73.3]	26.0 [3.5–51.0]	0.19
Active EIM at time of significant kidney impairmentb,d	3 [7%]	2 [8%]	1 [7%]	1.0
Disease location/extent		L1 4 [15%]	E1 0	n/a
		L2 9 [35%]	E2 3 [19%]
		L3 13 [50%]	E3 1 [6%]
		L4a/L4b 9 [35%]	E4 12 [75%]
Disease behavior		B1 20 [77%]	n/a	n/a
		B2 3 [12%]
		B3 3 [12%]
Perianal disease		6 [23%]	n/a	n/a
Disease activityd				0.52
Remission	16 [39%]	12 [46%]	4 [27%]
Mild	11 [27%]	7 [27%]	4 [27%]
Moderate	7 [17%]	4 [15%]	3 [20%]
Severe	7 [17%]	3 [12%]	4 [27%]
Fecal calprotectin [μg/g]c	431 [121–1147]	416 [80–971]	500 [341–1328]	0.56

Note: Values are presented as mean (SD), median (25th percentile – 75th percentile) or number (percentage).

Abbreviations: 5‐ASA, 5‐aminosalicylic acids; BMI, body‐mass index; CD, Crohn's disease; EIM, extra‐intestinal manifestation; IBD‐U, inflammatory bowel disease unclassified; m, months; n/a, not available; SD, standard deviation; UC, ulcerative colitis; y, years.

^a Solitary right kidney (n = 1), multicystic dysplastic right kidney (n = 1), renal amyloidosis (n = 1), focal bilateral pyelonephritis (n = 1), right pelvic dilatation (n = 1), episode(s) of acute kidney injury (n = 1),

^b Arthritis (n = 1), peristomal pyoderma gangrenosum (n = 1), primary sclerosing cholangitis (n = 1),

^c 19 missing values.

^d One missing value.

3.2. Reported causes of significant kidney function impairment

3.2.1. Concordance between treating physician and panelists

The most frequently reported cause by the physician was IBD medication (n = 15), followed by non IBD‐related causes (n = 7) and IBD‐related causes (n = 6) (Figure 1). The most frequently reported cause by the majority of panel members was IBD‐related (n = 14), followed by IBD medication (n = 10).

Figure 1.

Overview of the most likely cause of significant kidney function impairment stated by the physician and the majority of the panel members. The most likely cause identified by the panel was determined by selecting the cause that received the most votes from the panel members. In case of a tie or when 3 causes scored the same, the cause was valued as 0.5 or 0.3 respectively, instead of 1. IBD, inflammatory bowel disease; TIN, tubulo‐interstitial nephritis.

There was very low concordance among the reporting physician and individual panelists (Fleiss' kappa = 0.143 for Group A and Fleiss' kappa = 0.102 for Group B). This did not improve when including cases in which the most likely cause was TIN, pre‐renal kidney injury, glomerulonephritis, or when including only cases with CKD or those confirmed with renal biopsy (Supporting Information: Table 2).

3.2.2. Details on inflammatory bowel disease medication as most likely cause reported by treating physician

The most frequently suspected medications by physicians were 5‐ASA (n = 7), tacrolimus (n = 2), azathioprine (n = 2), and (not yet FDA approved for pediatric IBD) vedolizumab (n = 2) (Supporting Information: Table 3). Other suspected drugs included vitamin D/calcium, adalimumab, ciprofloxacin, infliximab, cyclosporine and methotrexate (all reported once). In several cases multiple drugs were suspected as cause of SKI. In 8/13 (62%) cases kidney impairment recovered after cessation of the drug. Of the cases most likely caused by IBD itself (n = 4), disease activity around time of SKI was remission (n = 2) or mild (n = 2). Detailed information on each case by diagnosis is presented in Supporting Information: Tables 4–9.

3.2.3. Most likely cause of tubulointerstitial nephritis, pre‐renal kidney injury and glomerulonephritis reported by treating physician

For cases of TIN (n = 15, Supporting Information: Table 4), the most frequently reported cause was IBD medication (n = 9), with 5‐ASA being the most commonly mentioned (n = 6). Other drugs reported to have possibly cause TIN included azathioprine (n = 2), vedolizumab (n = 2), and single cases of infliximab, cyclosporine A, and methotrexate; 4 out of 7 of these patients also used concomitant 5‐ASA (Supporting Information: Table 3). Patients with TIN attributed to other causes than IBD medication, had not used 5‐ASA. For cases of pre‐renal kidney injury (n = 8, Supporting Information: Table 5), non‐IBD‐related causes (n = 3) and other causes (n = 3) were most common. Other causes included gastrointestinal losses due to infection, suspected dehydration, and shock caused by fulminant colitis. For glomerulonephritis (n = 6, Supporting Information: Table 6), both IBD itself (n = 2) and IBD medication (n = 2, adalimumab and 5‐ASA, Supporting Information: Table 3) were the most likely causes.

3.3. Significant kidney function impairment details

For the majority of cases with SKI (n = 30/41, 73%) the indication for testing creatinine was routine follow‐up of IBD. Specifically, in 12/15 (80%) of TIN cases, 3/8 (38%) pre‐renal kidney injury cases and 6/6 (100%) of cases with glomerulonephritis, indication for testing creatinine was routine follow‐up of IBD. The time from IBD diagnosis until SKI varied from 0 to 180 months (median 31 months [interquartile range {IQR} 7–68]). SKI had resolved in 21/39 (54%) patients, with a median duration of kidney function impairment of 3 weeks (IQR 0.0–13.0), whereas 18/39 (46%) patients did not have creatinine normalization after a median follow‐up time until the last recorded creatinine of 121 weeks [IQR 55–157], and in three patients it was unknown whether creatinine normalized (one case of TIN, one case of other and one case with an unknown diagnosis).

3.4. Outcome of significant kidney function impairment

In cases with TIN, SKI resolved in 3/14 (21%) patients after a median duration of 6 weeks [IQR 3–23], in these three cases IBD medication was the most likely cause and was ceased, in 1/3 patients corticosteroids were also started. In 11/14 (79%) creatinine remained elevated after a median follow‐up of 121 weeks [IQR 70–146], of which IBD medication was the most likely cause in 5/11 cases and was ceased, 9/11 patients received corticosteroids. In cases with pre‐renal kidney injury, SKI resolved in 7/8 (88%) patients after a median duration of 2 days [IQR 1–20] and in 1/8 (12%) creatinine remained elevated after a follow‐up of 3 weeks. In cases with glomerulonephritis, SKI resolved in 3/6 (50%) cases after a median duration of 16 weeks [IQR 11–18], in which IBD medication was most likely cause in 2/3 patients and 2/3 patients received corticosteroids. In 3/6 (50%) cases creatinine remained elevated after a median follow‐up of 59 weeks [42–95 weeks], of which 1/3 patients received a course of corticosteroids.

3.4.1. Chronic kidney disease

Four patients were excluded from the classification of CKD: in one patient, it was known that creatinine levels were not normalized, but follow‐up was shorter than 3 months. In another patient, there was no data reported on whether the SKI had recovered, and in two patients, it was unclear if renal failure had resolved. Twelve out of 38 (32%) patients developed CKD (abnormal creatinine > 3 months and eGFR < 60 mL/kg/1.73 m²), two of them requiring kidney replacement therapy (KRT): one temporary KRT (diagnosed with ATN) and one kidney transplantation (diagnosed with TIN). Five patients had an abnormal creatinine for more than 3 months, but their eGFR was between 60 and 90 mL/kg/1.73 m² without information on markers of kidney damage, and so it was uncertain whether these patients had CKD. Characteristics of patients who developed CKD compared to patients who did not develop CKD are shown in Table 2. In patients with CKD, there was a significantly higher proportion of TIN diagnoses, and a significantly lower proportion of pre‐renal kidney impairment diagnoses compared to patients without CKD. Specifically, within patients with TIN in whom CKD could be evaluated (n = 11), eight patients (73%) developed CKD.

Table 2.

Comparison of IBD characteristics and details of SKI between patients who developed CKD versus patients who did not.

	CKD [n = 12]	No CKD [n = 21]	Uncertain CKD [n = 5]	p‐value
IBD type				0.642
Crohn's disease	9 [75%]	12 [57%]	3 [60%]
Ulcerative colitis/IBD‐U	3 [25%]	9 [43%]	2 [40%]
Male sex	11 [92%]	11 [52%]	5 [100%]	0.021
Duration of IBD at onset of SKI in months, median (IQR)	47 [12–72]	24 [7–44]	52 [23–119]	0.148
Disease activity at time of SKI				0.352
Remission	8 [67%]	5 [24%]	2 [40%]
Mild	2 [17%]	8 [38%]	1 [20%]
Moderate	1 [8%]	3 [14%]	1 [20%]
Severe	1 [8%]	5 [24%]	1 [20%]
Underlying diagnosis (confirmed with renal biopsy)				0.008
Pre‐renal SKI	0	7 [33%]	0
Glomerulonephritis	3 [25%] (3 [25%])	3 [14%] (2 [10%])	0
Tubulo‐interstitial nephritis	8 [67%] (7 [58%])	3 [14%] (1 [5%])	[60%] (2 [40%])
Nephrolithiasis	0	2 [10%]	0
Othera	0	4 [19%] (1 [5%])	0
Unknown	1 [8%]	2 [10%]	2 [40%]
Most likely cause (original/majority panel)				0.113/0.002b
IBD itself	[25%]/2.5 [21%]	1 [5%]/1 [5%]	0/0.3 [6%]
IBD medication	[17%]/1 [8%]	8 [38%]/6 [29%]	3 [60%]/3 [60%]
Non‐IBD related	1 [8%]/2.5 [21%]	5 [24%]/6.5 [31%]	1 [20%]/0.3 [6%]
IBD‐related	4 [33%]/6 [50%]	1 [5%]/6.5 [31%]	0/0
Other	0/0	4 [19%]/0	0/0
Unknown	2 [17%]/0	2 [10%]/1 [5%]	1 [20%]/1.3 [26%]

Note: p‐value represents overall comparison of the three groups.

Abbreviations: CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; IBD, inflammatory bowel disease; IBD‐U, IBD‐unclassified; IQR, interquartile range; SKI, significant kidney impairment.

^a One case of post‐renal SKI, acute tubular necrosis, tacrolimus induced nephrotoxicity, nephrocalcinosis. CKD was defined as eGFR < 60 mL/kg/1.73 m² for >3 months. Uncertain CKD was defined as a follow‐up >3 months and eGFR between 60 and 90 mL/kg/1.73 m².

^b Most likely cause identified by the panel. Cases with ties (n = 5) were left out of the analysis of the p‐value.

3.5. Recommendations

Practice points regarding frequency of screening, tools to assess kidney function, areas of attention in medical history and physical examination, and when to consult a pediatric nephrologist are shown in Figure 2. Detailed background information and rationales behind the formulated practice points are provided in the Supporting Information: Results S13.

Figure 2.

Recommendations for screening of kidney function in children with IBD, follow‐up on (suspicion of) kidney impairment and considerations when to consult a pediatric nephrologist. ^†For 5‐ASA, first monitoring after 3 months, then every 6 months. ^‡When positive, perform a repeat urinalysis to: (1) measure the urine protein‐to‐creatinine ratio (PCR) if positive for protein; (2) perform a microscopic evaluation of hematuria and RBC morphology (casts, dysmorphic RBCs) if positive for blood (in this case a fresh urine sample is required). 5‐ASA; 5‐aminosalic acids; ACR, albumin‐to‐creatinine ratio; EIM, extra‐intestinal manifestations; eGFR, estimated glomerular filtration rate; IBD, inflammatory bowel disease; PCR, protein‐to‐creatinine ratio.

4. DISCUSSION

To our knowledge, this is the first prospective multicenter study describing SKI in children with IBD. Tubulointerstitial nephritis (TIN) was the most frequently reported kidney disorder leading to SKI. IBD medication, most commonly 5‐ASA, was often suspected as the cause. Notably, SKI progressed to CKD in 32% of cases, and in 73% of patients with TIN.

In this study and others, various IBD therapies have been implicated in the development of both SKI and TIN. Among these, 5‐ASA has been most commonly associated with TIN, although its exact nephrotoxic risk remains controversial. ²³ , ²⁴ , ²⁵ TIN can occur irrespective of 5‐ASA treatment duration; approximately 50% of cases arise within the first year, but delayed onset after several years has also been reported. ²³ , ²⁶ Several studies suggest no clear dose‐response relationship between 5‐ASA and nephrotoxicity. ²³ When drug‐induced TIN is suspected, timely discontinuation of the suspected drug is essential. High‐dose steroids may be considered, though evidence of their benefit is limited. ¹⁷ One study showed that discontinuing 5‐ASA within 10 months resulted in renal recovery in 40%–85% of cases, while recovery dropped to one‐third when detected after 18 months. ²³ , ²⁷

Although less frequently reported, several case reports describe kidney function impairment in pediatric and adult patients with IBD possibly caused by anti‐TNF treatment or vedolizumab. ¹¹ , ²⁸ , ²⁹ , ³⁰ In our registry, two SKI cases were attributed to vedolizumab; one patient was also on 5‐ASA, while the other had abnormal creatinine levels before vedolizumab was initiated. Calcineurin inhibitors like cyclosporine and tacrolimus are known nephrotoxic agents and can cause TIN or ATN. ¹⁰ , ³¹ Some antibiotics, such as fluoroquinolones (e.g., ciprofloxacin), are generally safe but may also lead to TIN or ATN. ³² , ³³

Importantly, IBD itself was also considered the likely cause of kidney impairment in several cases. This aligns with prior literature suggesting that IBD can result in kidney involvement as an extra‐intestinal manifestation (EIM). ² , ¹⁰ Although the pathophysiology of (kidney) EIMs in IBD remains unclear, one suggested pathway involves increased systemic IBD disease activity. ¹² Some cases occurred at or shortly after IBD diagnosis, suggesting an association with active disease. However, EIMs may also develop independently of IBD activity, due to separate inflammatory processes or genetic susceptibility. ¹² Our findings support this, as four cases attributed to IBD occurred during remission or mild disease.

Identifying the cause of SKI remains challenging. To aid this, a panel of experts reviewed cases and voted on likely causes. Despite access to detailed clinical data, significant variation was seen among panel members, highlighting the complexity in determining the causative factor. In many cases, it remains unclear whether medication or IBD itself is the cause of SKI. Even after the suspected medication was discontinued, kidney function sometimes failed to recover or worsened again, suggesting the medication was a less likely cause. Identifying factors that make IBD medication or IBD itself more likely to be the cause is difficult. The difficulty is often compounded by inconsistent creatinine monitoring before and after new treatments. Regular monitoring could help identify temporal relationships between medication initiation and kidney injury. When deciding to discontinue potentially nephrotoxic IBD therapy, clinicians must balance renal risks against the limited treatment options in pediatric IBD. Nonetheless, early discontinuation in suspected drug‐induced SKI may reverse the damage and reduce CKD risk. ²⁷

Among patients who developed CKD, most had TIN or glomerulonephritis. Two required temporary kidney replacement therapy. By regularly monitoring creatinine, kidney function impairment can hopefully be timely detected, adequately treated and if required, medication can be ceased thereby reducing the risk of developing CKD. In 73% of our cases, the first abnormal creatinine level was found during standard IBD follow‐up. Current European guidelines do not recommend routine kidney function monitoring for pediatric CD, while in UC, such recommendations are limited to those receiving calcineurin inhibitors. ¹³ , ¹⁴ In this study, we provide practical recommendations for screening and follow‐up, aligning closely with recent adult IBD guidelines. ³⁴

This study has several strengths. First, data were collected prospectively, thereby making it less prone to bias. Additionally, the Safety Registry is the largest international registry to collect detailed data on kidney disorders in children with IBD in a homogenous manner. To our knowledge, we are the first study to propose guidance on screening, follow‐up, and referral of pediatric IBD patients at risk for kidney dysfunction. We hope this helps raise awareness and inform future study designs. However, several limitations must be acknowledged. Our recommendations were developed through expert consensus rather than formal methodologies such as DELPHI or RAND processes. Additionally, our definition of SKI (eGFR <60 mL/min/1.73 m²) was based on 2016 criteria and did not align fully with the 2019 KDIGO definitions. ³⁵ Although our definition of SKI eGFR <60 mL/min/1.73 m² is in line with CKD Stage G3‐G5, we were unable to classify other cases of kidney failure according to these guidelines, for example because information on markers of kidney damage (albuminuria) was not systematically documented. Therefore, we might have underestimated the number of cases which developed CKD. Additionally, we have included cases leading to both AKD and/or CKD. Prerenal cases of SKI often resolve quickly upon fluid therapy and as these usually have a less severe impact, it is very likely that not all cases of pre‐renal SKI meeting the criteria have been reported to the registry. We believe that further research should focus on patients developing CKD. We used the Schwartz formula, which is dependent on serum creatinine. Since creatinine levels are influenced by muscle mass, this method can overestimate kidney function in patients with low muscle mass and underestimate it in those with relatively high muscle mass, such as adolescent males. ³⁶

5. CONCLUSION

In conclusion, SKI may occur in children with IBD, as an EIM or as complication of the treatment or the disease itself, and may lead to CKD. Identifying the most likely cause of CKD is challenging. Regular screening for kidney function aims to ensure timely identification of kidney function impairment, may help in identifying the most likely cause, and may lead to prompt consultation with pediatric nephrologists, reducing the risk of further deterioration.

CONFLICT OF INTEREST STATEMENT

Lissy de Ridder reports grants from ZonMW, Stichting Autoimmuun Onderzoek (SAIO), Stichting Vrienden van het Sophia (Erasmus MC Sophia Foundation), ECCO, Pfizer and consultancy fees from Abbvie, during the conduct of the study. Nicholas M. Croft has received institutional research support from Pfizer, Takeda, Janssen, Abbvie and Eli Lilly, all unrelated to the content of this manuscript. Stephanie A. Vuijk report grants from ZonMw and Crocokids, and grants and non‐financial support from Pfizer. Renz C.W. Klomberg reports grants from Stichting Autoimmuun Onderzoek (SAIO), and Stichting Vrienden van het Sophia (Erasmus MC Sophia Foundation). Anne M. Griffiths reports consultancy fees from Abbvie, Janssen, Lilly; speaker fees from Abbvie, Takeda. Séamus Hussey received research grants from Janssen. Matthew W. Carroll has received speakers honorarium from Abbvie. All other authors declare no conflicts of interest. The PIBD‐SETQuality Safety Registry was funded by the European Union Horizon 2020 grant (grant agreement number 668023). The funders had no role in the study design, data collection or analysis, the preparation of the manuscript, or decision to submit the results.

Supporting information

Supplementary Figure 1. Overview of countries represented by physicians who have participated in the PIBD‐SETQuality Safety Registry. Paediatric gastroenterologists have actively contributed from 168 centres across 36 different countries. Supplementary Figure 1A. Global overview of countries represented by physicians reporting for the PIBD‐SETQuality Safety Registry. Supplementary Figure 1B. Overview of countries represented by physicians reporting for the PIBD‐SETQuality Safety Registry in Europe.

Supplementary Figure 2. Flow chart of reported and included cases of significant kidney function impairment. TIN; Tubulo‐interstitial nephritis, CKD; chronic kidney disease.

Supplementary Table 1. Overview of previous and current administered medication in patients with significant kidney impairment. Previous medication was considered if it was started and stopped before the first abnormal creatinine measurement. Current medication was considered if it was started and still ongoing at the time of the first abnormal creatinine measurement. Antibiotics, (exclusive) enteral nutrition and corticosteroids were not taken into account in this table. ^†One missing value. 5‐ASA; 5‐aminosalicylic acids.

Supplementary Table 2. Concordance among reporting physician and individual panellists in scoring the most likely cause of significant kidney impairment based on the Fleiss Kappa coefficient. Two panel members (SV and RK) filled out the most likely cause for both groups. TIN; Tubulo‐interstitial nephritis.

Supplementary Table 3. Details SKI cases where reporting physician stated IBD medication as most likely cause. ^†Start medication until first abnormal creatinine. ^‡First abnormal creatinine until first normal creatinine or last reported creatinine. ^§Confirmed by renal biopsy. ^¶Stop date cyclosporine unknown, assumed stopped at start SKI. ^††First abnormal creatinine observed 2 weeks before start vedolizumab, but vedolizumab was deemed the most likely cause by the reporting physician.^‡‡Calcium lowered, not ceased. ^§§Tacrolimus dose lowered, not ceased. SKI, significant kidney function impairment; IBD, inflammatory bowel disease; CD, Crohn's disease; UC, ulcerative colitis; IBD‐U, IBD‐unclassified; 5‐ASA, 5‐aminosalicylic acids; eGFR, estimated glomerular filtration rate.

Supplementary Table 4. Cases diagnosed with tubulo‐interstitial nephritis. Corticosteroids, antibiotics and (exclusive) enteral therapy were not reported in this table. IBD, inflammatory bowel disease, CD, Crohn's disease; UC, ulcerative colitis; EIM, extra‐intestinal manifestation; AKI, acute kidney injury; MTX, methotrexate; CKD, Chronic kidney disease; 5‐ASA, 5‐aminosalycate acids; ↑, above normal limits; n/a, not applicable, i.e. test has not been performed; m, months.

Supplementary Table 5. Cases diagnosed with pre‐renal kidney impairment. ^†Pre‐renal kidney impairment due to intravascular volume depletion due to worsening of nephrotic syndrome due to infection of unknown origin. IBD, inflammatory bowel disease, CD, Crohn's disease; UC, ulcerative colitis; EIM, extra‐intestinal manifestation; AKI, acute kidney injury; MTX, methotrexate; m, months, n/a; not available.

Supplementary Table 6. Cases diagnosed with glomerulonephritis. IBD, inflammatory bowel disease, CD, Crohn's disease; UC, ulcerative colitis; EIM, extra‐intestinal manifestation; m; months, 5‐ASA; 5‐aminosalycate acids, n/a; not available.

Supplementary Table 7. Cases diagnosed with nephrolithiasis. IBD; Inflammatory bowel disease, 5‐ASA: 5‐aminosalicates.

Supplementary Table 8. Cases diagnosed with other cause. ^†Details of this case have been previously published (Regev et al. Oct 2020, Pediatr. Nephrol PMID: 32232639). IBD; Inflammatory bowel disease, 5‐ASA: 5‐aminosalicates.

Supplementary Table 9. Cases diagnosed with unknown cause. IBD; Inflammatory bowel disease, 5‐ASA: 5‐aminosalicates.

Supplementary_Methods_renalfailure_JPGN.

Supplementary_Results_JPGN.

Vuijk SA, Klomberg RCW, Schwerd T, et al. Significant kidney impairment in pediatric inflammatory bowel disease and practice points for pediatric gastroenterologists. J Pediatr Gastroenterol Nutr. 2026;82:1251‐1260. 10.1002/jpn3.70391

DATA AVAILABILITY STATEMENT

The data underlying this article are available upon reasonable request to the corresponding author.