Using large-scale two-sample MR analysis, we systematically evaluated the causal relationship between genetically predicted IBD and kidney function or disease in the present study. Our results show that UC increases the levels of uACR, and CD increases the risk of urolithiasis. These significant results solidify the findings of prior observational studies suggesting that renal and urinary involvement are extraintestinal manifestations of IBD, hence highlighting the existence of the gut-kidney axis.
Proteinuria is central to the diagnosis and management of kidney diseases. Measurement of uACR is a reliable method to predict renal events in patients with nephropathy. The most common type of glomerulonephritis and the primary cause of ESRD in patients with primary glomerular disease is IgA nephropathy [
21]. A retrospective study of kidney biopsy in patients with IBD found that the most frequent diagnosis was IgA nephropathy, followed by interstitial nephritis, arterionephrosclerosis, and acute tubular injury. Notably, proteinuria is the common indication for kidney biopsy in patients with IBD [
22]. Here, we demonstrate that UC causally increases the levels of uACR, which is consistent with earlier findings in observational studies [
8,
9]. However, there was no causal relationship between the IBD and IgA nephropathy in our study. Considering IBD susceptibility loci were shared with IgA nephropathy and the assumption 3 in our MR analysis ruled out the risk factor of an IgA nephropathy outcome, IgA nephropathy probably is a co-occurring disease rather than a secondary manifestation of IBD. Moreover, several studies have demonstrated that there is a positive correlation between the disease activity of IBD and tubulointerstitial damage defined by elevated tubule marker proteins, such as N-acetyl-beta-D-glucosaminidase, alpha-1-microglobulin, and beta-2-microglobulin [
8,
23,
24].
The prevalence of urolithiasis in patients with IBD has been shown to be approximately 3–5% [
25]. In a recent nationwide cohort study consisting of 75,236 patients with IBD and 767,403 non-IBD individuals, it was discovered that patients diagnosed with IBD had twice the risk of urolithiasis [
10]. Similarly, our study found that CD increases the risk of urolithiasis. Indeed, the risk of urolithiasis is higher in CD compared with subjects with UC [
26], probably due to frequent ileocolic involvement in CD. Disruption of the gut-kidney axis could play an important role in the development of urolithiasis in patients with IBD. Patients with IBD primarily develop renal stones consisting of uric acid or calcium oxalate.
Oxalobacter formigenes, a member of the human colonic microbiota, may have an antilithogenic effect in calcium oxalate urolithiasis, through modulation of colonic oxalate transport and secretion [
27]. Indeed, patients with both IBD and urolithiasis rarely exhibit
Oxalobacter formigenes in their stools when compared to controls, which may lead to hyperoxaluria in patients with IBD [
28]. In particular, CD patients with urolithiasis had significantly higher levels of urinary oxalate when compared to those without [
29]. A recent study based on multi-omics data showed that enteric oxalate levels are elevated in patients with IBD, with highest levels in CD patients with ileocolic involvement. They also demonstrated that microbiota oxalate degradation is decreased in patients with IBD, potentially contributing to the disruption of oxalate homeostasis [
30]. Notably, bile acid malabsorption in CD patients with ileocolic involvement may lead to fatty acids reaching the colon, thus free fatty acids in the colon may compete with oxalate to bind calcium. As a result, an increased levels of free oxalate might be reabsorbed via colonocytes and excreted in the urine, leading to oxalate urolithiasis [
31]. In general, IBD represents a newly appreciated cause for the development of urolithiasis which may relate to gut-kidney axis disruption.uACR and eGFRcrea are the two key markers for CKD, and urolithiasis is a risk factor for CKD. However, our MR study has demonstrated that IBD is not the cause for increased levels of eGFRcrea and risk for CKD, despite the relations that UC and CD affect the levels of uACR and the risk of urolithiasis, separately. A single-cohort study showed that renal dysfunction is a rare complication (2%) in CD, and recurrent urolithiasis appears to be the primary causative factor [
32]. A case–control study demonstrated that the frequency of renal dysfunction in IBD inpatients was 15.9% [
33], this high prevalence of renal dysfunction may be associated with severity of IBD inpatients. Moreover, a recent study of over 80,000 individuals showed that IBD is correlated with an increased risk of CKD, and exposure to 5-aminosalicylates or methotrexate was not associated with the change in eGFRcrea [
5]. A population-based study conducted across Korea showed that patients with CD had a substantially higher risk of ESRD in comparison with matched controls (HR = 6.33). Conversely, there was no significant difference in the risk of ESRD observed between UC and control groups [
34]. GWAS data typically involves exposures and outcomes that are measured at a single timepoint. The interpretation of the causal estimates assumes that the effect of the genetic instruments on IBD remains stable over time. Nonetheless, the etiology of IBD involves a complex interaction between the genetic, the immune responses and environmental factors, thus environmental or epigenetic influences may alter the correlation between a genetic instrument and the exposure throughout the lifespan. In addition, our MR study cannot entirely exclude the hypothesis that IBD is the cause of the elevated levels of eGFRcrea and risk of CKD, since the disease activity and disease duration of IBD have not been extensively studied.
To the best of our knowledge, this is the first study to perform a MR analysis to address causality between IBD (including UC and CD) and kidney function/disease (including eGFRcrea, uACR, CKD, urolithiasis, and IgA nephropathy) using large-scale GWAS data. Importantly, our results show that UC increases the levels of uACR, and CD increases the risk of urolithiasis. The current study, however, has several limitations. First, the enrolled individuals were almost European, thus the causal effect of IBD on the levels of uACR and the risk of urolithiasis in other ethnic populations remains unknown. Second, the GWAS of IgA nephropathy were from a trans-ethnic population (73% European and 27% East Asian). Hence, we need to cautiously interpret the conclusion that there is no causal relationship between IBD and IgA nephropathy. Third, our findings only report alterations in kidney function and increased risk of urolithiasis in patients with IBD, thus further investigations are required to address underlying mechanisms. Notably, microphysiological systems could be an ideal approach to investigating the role of the gut-kidney axis in IBD [
35]. Besides, IBD drug-induced nephrotoxicity should be taken into consideration in clinical practice of IBD as described in previous studies.
In conclusion, UC and CD affect the levels of uACR and increase the risk of urolithiasis, separately. The renal involvement in patients with IBD can be categorized into secondary diseases caused by IBD, co-occurring disease with IBD, and IBD drug-induced nephrotoxicity. As the rate of renal involvement in patients with IBD may be currently underestimated or even overlooked in clinics, monitoring of kidney function should be considered in clinical practice in patients with IBD.