Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Sep 1.
Published in final edited form as: Clin Gastroenterol Hepatol. 2019 Nov 1;18(10):2262–2268. doi: 10.1016/j.cgh.2019.10.043

Inflammatory Bowel Diseases Are Associated With an Increased Risk for Chronic Kidney Disease, Which Decreases With Age

Ravy K Vajravelu *,‡, Lawrence Copelovitch §, Mark T Osterman *, Frank I Scott ‡,∥, Ronac Mamtani ‡,, James D Lewis *,‡, Michelle R Denburg ‡,§
PMCID: PMC7569504  NIHMSID: NIHMS1542338  PMID: 31683056

Abstract

BACKGROUND & AIMS:

It is not clear what factors affect risk of chronic kidney disease (CKD) in patients with inflammatory bowel disease (IBD); increased risk has been inconsistently associated with use of 5-aminosalicylates (5-ASAs). We aimed to calculate the relative hazard of CKD among patients with IBD, adjusted for CKD risk factors, and to determine whether IBD medications are associated with change in estimated glomerular filtration rate (eGFR).

METHODS:

We performed a retrospective cohort study of data from The Health Improvement Network. Patients with IBD (n = 17,807) were matched for age, sex, and practice to individuals without IBD (n = 63,466). The relative hazard of CKD, stages 3 through 5D, in patients with IBD was calculated using a Cox proportional hazards model adjusted for common CKD risk factors. We also evaluated the association of 5-ASAs, azathioprine, and methotrexate with change in eGFR using a longitudinal model.

RESULTS:

After we controlled for risk factors associated with CKD, we found IBD to be associated with development of CKD in patients 16–77 years old. As patient age increased, the adjusted hazard ratio for CKD decreased monotonically, from 7.88 (95% CI, 2.56–24.19) at age 16 to 1.13 (95% CI, 1.01–1.25) at age 77. In the longitudinal analysis, exposure to 5-ASAs or methotrexate was not associated with change in eGFR, whereas azathioprine was associated with a slightly higher eGFR (0.32 mL/min/1.73 m2; 95% CI, 0.16–0.48).

CONCLUSIONS:

In a retrospective study of more than 80,000 persons, we found that IBD is associated with increased risk of CKD, and the hazard ratio is highest among younger patients. Commonly used non-biologic therapeutic agents were not associated with lower eGFR.

Keywords: Inflammatory Bowel Disease, Chronic Kidney Disease, 5-Aminosalicylates

Crohn’s disease and ulcerative colitis, collectively known as inflammatory bowel disease (IBD), are characterized by inflammation of the gastrointestinal tract and are associated with several disease-related complications.1 Whether chronic kidney disease (CKD) is one of these complications is uncertain, because prior studies assessing whether IBD is associated with CKD have had conflicting results.2–6 Hypothesized mechanisms linking CKD and IBD include mesalamine use,7–11 interstitial nephritis,12–15 IgA nephropathy,16 secondary amyloidosis,17 and nephrolithiasis.18–21

The existing literature regarding the association of IBD with CKD includes small, single-center studies with low statistical power3,4 and population-based studies that relied on CKD diagnosis codes to ascertain the outcome.2,5,6 Although definitions of CKD based solely on diagnosis codes are highly specific, they lack sensitivity. For example, in one study, the sensitivity of diagnostic codes for CKD was only 48.8% when compared with CKD based on the gold standard of estimated glomerular filtration rate (eGFR).22 Furthermore, patients identified to have CKD by diagnosis codes reflect a subset of the general CKD population that is more ill.23 Additionally, most prior studies have not controlled for traditional risk factors for CKD, such as coronary artery disease, diabetes mellitus, and hypertension.2–5 As such, prior studies may have suffered from outcome misclassification, selection bias, and confounding, thereby misestimating the risk of CKD among patients with IBD.

It is important to accurately estimate the incidence of CKD among patients with IBD because early detection and management of CKD is associated with reduced mortality.24 Furthermore, determining risk factors for CKD among patients with IBD may help inform decision making about therapeutics and monitoring. To determine whether patients with IBD are at increased risk of CKD and to address the limitations of the prior literature, we conducted a retrospective cohort study of individuals with IBD and matched control subjects using The Health Improvement Network (THIN), a patient-level electronic medical record database of general practices in the United Kingdom. Using creatinine-based measures of eGFR and CKD diagnosis codes from this cohort of patients, we determined the incidence of CKD among patients with IBD, the relative hazard of developing CKD among patients with IBD versus matched individuals without IBD, and whether mesalamine and other non-biologic medications commonly used to treat IBD are associated with eGFR decline.

Methods

Study Population

Individuals in THIN who were diagnosed with IBD during follow-up were identified by the presence of at least 1 diagnostic code consistent with IBD (Supplementary Table 1, database details in Supplementary Methods). To be considered for inclusion in the cohort, an individual’s first IBD code must have occurred at least 2 years subsequent to enrollment with the practice and after the medical practice enrolled in THIN.25 The date of the first IBD code was considered the index date. The positive predictive value of diagnostic codes for IBD is 85%–91%.26 Control subjects were selected from all individuals in the database who never had a diagnosis of IBD during follow-up. Individuals with IBD were matched to up to 4 control subjects on age (at the time of IBD diagnosis in 3-year age groups from ages 2–30 and 5-year age groups from ages 30–90), sex, and practice. Control subjects were assigned the index date of their match, and like individuals in the IBD cohort, the matched index date for potential control subjects had to occur at least 2 years after enrollment with the practice and after the medical practice enrolled in THIN. Both individuals with and without IBD who were diagnosed with CKD (criteria below) before the index date were excluded. Children with IBD younger than 2 years old were excluded because they are likely to have very early onset IBD, which has a different pathogenesis compared with the general form of IBD,27 and because the eGFR of children does not reach adult levels until after age 2.28,29 Data on patients diagnosed with IBD after 90 years of age were excluded to avoid potential reidentification.

Outcome Definition

Development of CKD stages 3–5D (referred to from this point forward as CKD) was the primary outcome of interest. CKD status was identified if an individual had an eGFR <60 mL/min/1.73 m2 on at least 2 occasions measured at least 90 days apart30,31 or had a diagnostic code consistent with CKD (eGFR calculation details in Supplementary Methods, specific CKD codes used in Supplementary Table 2). The validity of this outcome definition was assessed in several sensitivity analyses (see later). The date of CKD onset was the earliest of the first date when the eGFR was <60 mL/min/1.73 m2 or the first documented CKD code.

Other Covariates

Covariates of interest were selected because of their potential association with CKD. Conditions of interest included coronary artery disease, diabetes mellitus, hypertension, peripheral artery disease, stroke, nephrolithiasis, urologic obstruction, and current or former tobacco smoking. These were identified by the presence ≥1 representative diagnostic code during follow-up. All diagnosis covariates were incorporated into the analyses as unidirectional time-dependent variables that updated on the earliest date of diagnosis documentation. Medication exposures of interest included aspirin, nonsteroidal anti-inflammatory drugs, proton-pump inhibitors, and statins. IBD medication exposures were mesalamine, azathioprine (including 6-mercaptopurine), methotrexate, and glucocorticoids. Because general practitioners in the United Kingdom do not usually prescribe biologic medications, the effect of tumor necrosis factor-α and integrin inhibitors on eGFR could not be assessed. All medication exposures were bidirectional time-dependent variables. A patient was considered exposed to a medication at the time of creatinine measurement if he or she received a prescription for the medication with a supply long enough to provide at least 1 dose of medication in the 60 days before the creatinine measurement. We also assessed whether a patient was experiencing a flare of IBD activity at the time of creatinine measurement. An individual was considered to be in flare if he or she had a prescription for an oral or intravenous systemic glucocorticoid after a period of 120 days without any glucocorticoid exposure. The validity of this definition of flare has been previously evaluated to have a positive predictive value of 91%.32,33 Health care use intensity was estimated from the total number of general practitioner interactions, outpatient consultations, emergency department visits, and inpatient hospitalizations. This was incorporated into the analysis as a time-updating covariate. Race/ethnicity was not included as a covariate because it is not available in THIN.

Statistical Analysis

Stata version 15 (Stata Corp, College Station, TX) was used for all statistical analyses. Continuous data are reported as medians with interquartile range (IQR). Categorical data are reported as counts and percentages. Pearson chi-square test was used for comparisons of categorical variables, and the Wilcoxon rank sum test was used for unadjusted comparisons of continuous variables. All tests were 2-tailed and evaluated at a type I error rate of 0.05.

Incidence Rate of Chronic Kidney Disease and Relative Hazard of Inflammatory Bowel Disease for Development of Chronic Kidney Disease

Incidence rates (IRs) of CKD were directly age-standardized to the 2013 European Standard population. Unadjusted cumulative IRs for CKD were calculated using Kaplan-Meier survival analysis. The adjusted hazard ratio (aHR) of CKD among patients with IBD was calculated using Cox proportional hazards regression. The model was adjusted for the following risk factors for CKD: coronary artery disease, diabetes mellitus, ever-smoker status, hypertension, peripheral artery disease, and stroke. Because of a nonlinear trend identified in the relationship between the CKD IR ratio and age between persons with and without IBD (Supplementary Figure 1), several IBD-age interactions were tested using quadratic terms and splines (details in Supplementary Methods). In a subanalysis, the risk of developing CKD was stratified by IBD subtype. Sensitivity analyses were performed to test key study assumptions, including studying more advanced CKD and studying stricter eGFR criteria for CKD to assess for possible misclassification from repeated episodes of acute kidney injury (details in Supplementary Methods).

Risk Factors for Estimated Glomerular Filtration Rate Change Between Measurements Among Patients With Inflammatory Bowel Disease

Risk factors for eGFR change between measurements were evaluated in an analysis limited to patients with IBD using generalized estimating equations to estimate the coefficients of a longitudinal linear regression model describing the relationship between eGFR and the diagnosis and medication covariates of interest. In this model, the outcome of interest was the change in the eGFR from one measurement to another. Observations were clustered by patient, and an autoregressive correlation structure was used. The exposures of interest were IBD subtype, mesalamine, azathioprine, 6-mercaptopurine, methotrexate, and IBD flare status. The coefficients for each of these exposures were adjusted for age; sex; comorbid diagnoses of hypertension (diagnosed at least 2 years prior), coronary artery disease, diabetes mellitus, peripheral artery disease, or stroke; current or former tobacco smoking; and exposure to nonsteroidal anti-inflammatory drugs, aspirin, proton-pump inhibitors, and statins. Patients were considered to be exposed to a given medication if they had at least 1 dose prescribed in the 60 days before eGFR measurement. The observations in this analysis were the eGFRs calculated from the creatinine measurements of patients in the IBD cohort after the IBD index date. Patients were censored from the analysis at the time of renal transplant or dialysis. Several sensitivity analyses were performed to test key study assumptions, including several adjustments of the medication exposure window (details in Supplementary Methods).

Results

Cohort Characteristics

A total of 19,076 patients meeting inclusion criteria for IBD during follow-up were identified, and they were age-, sex-, and practice-matched to 70,666 control subjects. After excluding patients with preexisting CKD, 1269 patients with IBD (and their corresponding 3463 control subjects) and 3737 control subjects were excluded from the analysis. This left 17,807 patients with IBD and 63,466 control subjects in the analysis (median number of control subjects per IBD patient, 4; IQR, 3–4). For patients with IBD, the median age at the time of diagnosis was 44 (IQR, 30–60), and 49.7% of patients in the cohort were female. For control subjects, the median age at matching was 44 (IQR, 30–58), and 50.5% of patients were female. Additional characteristics of the cohorts are presented in Table 1.

Table 1.

Cohort Characteristics, IBD Cohort Versus Age-, Sex-, and Practice-Matched Individuals Without IBD

IBD n = 17,807 Matched non-IBD n = 63,466 P value
Female 8856 (49.7) 32,073 (50.5) .58
Age at index date 44 (30–60) 44 (30–58) < .01
Follow-up time after index date (y) 4.0 (1.5–7.6) 4.7 (2.1–8.5) < .01
Health care encounters 299 (165–502) 139 (53–298) < .01
eGFR measurements 6 (2–13) 2 (0–6) < .01
eGFR measurements among those with at least 1 eGFR 7 (3–14) 4 (2–9) < .01
CKD diagnoses
 By eGFR definition 571 (3.2) 1450 (2.3) < .01
 By diagnostic code 678 (3.8) 1741 (2.7) < .01
 By either eGFR or diagnostic code 910 (5.1) 2224 (3.5) < .01
Comorbidity exposure years (%)
 Total exposure years 97,424.4 352,489.8
 Coronary artery disease 3219.3 (3.3) 8470.3 (2.5) < .01
 Diabetes mellitus 8426.5 (8.6) 23,506.4 (7.0) < .01
 Ever smoker 59,353.0 (60.9) 195,173.8 (58.2) < .01
 Glomerulonephropathy 406.4 (0.4) 902.4 (0.3) < .01
 Hypertension 17,804.4 (18.3) 65,029.3 (19.4) < .01
 Nephrolithiasis 1548.3 (1.6) 3729.0 (1.1) < .01
 Peripheral artery disease 252.5 (0.3) 813.3 (0.2) .42
 Stroke 1608.7 (1.7) 4344.1 (1.3) < .01
 Systemic lupus erythematosus 154.7 (0.2) 570.8 (0.2) .40
 Urologic obstruction 631.5 (0.6) 1417.6 (0.4) < .01
Open in a new tab

Note. Continuous variables presented as median (interquartile range) and categorical variables presented as count (%).

CKD, chronic kidney disease stages 3–5D; eGFR, estimated glomerular filtration rate; IBD, inflammatory bowel disease.

Chronic Kidney Disease Incidence and Risk Factors

Of the patients in the IBD cohort, 910 (5.1%) were diagnosed with CKD during follow-up. The median age at CKD diagnosis was 70 (IQR, 62–77; minimum, 16). The crude and age-standardized IRs were 933.4 per 100,000 person-years (95% confidence interval [CI], 874.7–996.0) and 130.3 per 100,000 person-years (95% CI, 121.9–139.0), respectively. Of the matched non-IBD group, 2224 (3.5%) were diagnosed with CKD during follow-up. The median age at CKD diagnosis was 71 (IQR, 64–78; minimum, 22). The crude and age-standardized IRs were 662.2 per 100,000 person-years (95% CI, 635.2–690.3) and 91.3 per 100,000 person-years (95% CI, 87.6–95.2), respectively. The absolute rates of CKD by age among persons with and without IBD are presented in Figure 1.

Figure 1.

Figure 1.

Open in a new tab

Unadjusted incidence rate ratios by age of persons with IBD and matched individuals without IBD. Solid lines represent smoothed lines of best fit.

An unadjusted Kaplan-Meier curve for the diagnosis of CKD by IBD status is presented in Supplementary Figure 2. The unadjusted cumulative incidence of CKD at 1, 5, and 10 years was 1.2% (95% CI, 1.1%–1.4%), 5.0% (95% CI, 4.6%–5.4%), and 8.6% (95% CI, 8.0%– 9.2%), respectively, for patients with IBD and 0.5% (95% CI, 0.5%–0.6%), 3.3% (95% CI, 3.1%–3.4%), and 6.6% (95% CI, 6.3%–6.9%) for those without IBD.

The unadjusted HR of CKD among patients with IBD was 1.41 (95% CI, 1.30–1.52). Based on information criteria and likelihood ratio testing, the model with an IBD-age interaction with 1 knot at age 60 was selected as the best-fitting model (P value vs quadratic model with no knots < .001 and P value vs model with knots at age 40 and 60 = .38). In this model, after controlling for risk factors associated with CKD and the number of health care encounters, IBD was associated with development of CKD from ages 16 to 77, with the aHR declining monotonically from aHR 7.88 (95% CI, 2.56–24.19) to aHR 1.13 (95% CI, 1.01–1.25) with increasing age (Figure 2). aHRs and 95% CIs for covariates are presented in Supplementary Table 3. The aHRs at specific ages are presented in Supplementary Table 4. Sub-analyses assessing the hazard of CKD associated with Crohn’s disease or ulcerative colitis relative to no IBD were similar to the main analysis, although the aHR for ulcerative colitis was slightly lower than the aHR for Crohn’s disease for all ages younger than 82 (Supplementary Figure 3). All sensitivity analyses confirmed the main study findings, including those assessing stricter eGFR criteria for CKD, those assessing more severe CKD, and those limiting the analysis to individuals with at least 5 years of follow-up before IBD diagnosis (Supplementary Results).

Figure 2.

Figure 2.

Open in a new tab

Adjusted hazard ratio of IBD by age (black solid line). Gray shaded area represents 95% confidence intervals. Horizontal dashed line indicates adjusted hazard ratio = 1. Vertical dotted line indicates quadratic spline at age 60. 95% confidence interval crosses 1 at age 78. Graph left-truncated at age 16 because no diagnoses of CKD occurred before this age. Adjusted hazard ratio values are presented in Supplementary Table 4.

Risk Factors for Estimated Glomerular Filtration Rate Change Among Patients With Inflammatory Bowel Disease

In the longitudinal analysis of eGFR change among patients with IBD, 11,839 patients with IBD and at least 2 eGFR measurements were identified. The median number of eGFR measurements was 6 (IQR, 3–12). Medication exposure among the cohort is presented in Supplementary Table 5. Exposure to mesalamine and methotrexate was not associated with a change in eGFR (−0.04 mL/min/1.73 m2, 95% CI, −0.17 to 0.09 and 0.29 mL/min/1.73 m2, 95% CI, −0.07 to 0.66, respectively). Azathioprine was associated with a slightly higher eGFR (0.32 mL/min/1.73 m2; 95% CI, 0.16–0.48). Active IBD flare at the time of eGFR measurement was not associated with a change in eGFR (−0.10 mL/min/1.73 m2; 95% CI, −0.33 to 0.48). Coefficients for confounding diagnoses and medication prescriptions are presented in Supplementary Table 6. The main study results were confirmed in the sensitivity analyses (Supplementary Results).

Discussion

This large population-based retrospective cohort study demonstrates that IBD is associated with development of CKD stages 3–5D after adjustment for common CKD risk factors. The age-standardized IR of CKD among patients with IBD is 130.3 per 100,000 person-years (95% CI, 121.9–139.0) compared with 91.3 per 100,000 person-years (95% CI, 87.6–95.2) in age-, sex-, and practice-matched individuals without IBD. We also show that the association between IBD and CKD attenuates with age, because the aHR declines from 7.88 (95% CI, 2.56–24.19) at age 16 to aHR of 1.13 (95% CI, 1.01–1.25) at age 77. Additionally, among persons with IBD, common IBD medications, such as mesalamine, azathioprine, and methotrexate, are not associated with lower eGFR compared with not using those medications. These findings have important implications for clinical practice, because gastroenterologists often monitor the renal function of patients receiving mesalamine to avoid nephrotoxicity.34 Because this study indicates that all patients with IBD are at increased risk for CKD and that the risk is not conferred by mesalamine use, it may be reasonable to periodically monitor renal function in all patients with IBD, not just those taking mesalamine.

Our finding that the association between IBD and CKD declines with increasing age is supported by the recent study by Park et al6 of the association between IBD and end-stage kidney disease among South Koreans. In subgroup analyses in that study, age younger than 40 conferred a higher hazard of end-stage kidney disease compared with age 40 or older. The mechanism of declining strength of association of IBD and CKD with increasing age demonstrated in our study and Park’s study is uncertain. If the IBD-CKD association is related to autoimmune-mediated glomerulopathy, it is possible that the declining association with age is related to decreased incidence of autoimmune disease in the elderly from age-related immune dysfunction.35 For example, a recent cohort study of patients with IBD in Switzerland demonstrated that those diagnosed with IBD in childhood are more likely to develop arthralgias and primary sclerosing cholangitis compared with those diagnosed in adulthood.36 Another possible explanation for the association between IBD and CKD is that CKD develops secondary to repeated renal injury from dehydration during IBD flares. Although this study did not show an association between an acute flare of IBD and lower eGFR, it was not designed to determine whether acute flare is associated with acute kidney injury, and this should be the topic of further investigation. However, if this was the mechanism of the association, one would expect that the IBD-CKD association increases with age as an individual patient experiences more dehydration episodes over time.

Strengths and Limitations

A strength of this study was the use of a patient-level electronic medical record database that allowed us to identify nearly 18,000 persons with IBD. Using these administrative data allowed for granular assessment of several time-varying confounders, which had not previously been evaluated in studies of the IBD-CKD association.2–5 Additionally, by supplementing CKD diagnosis codes with eGFR data, this study likely provides a more accurate estimate of the incidence of and risk factors for CKD among patients with IBD. Finally, the main study conclusions were upheld in several sensitivity analyses testing key assumptions.

There are potential caveats to consider when interpreting these study results. First, like all observational studies, there may have been unmeasured confounding. If a potential unmeasured confounder was associated with aging and inversely associated with IBD, it is possible that the declining IBD-CKD association with age is caused by the unmeasured confounder, because absolute CKD risk increases with age in the general population. However, given that our study accounted for the most prevalent risk factors for CKD in the general population, we do not expect that an unmeasured confounder strongly influenced the results. Similarly, we cannot entirely ensure that there was no misclassification in the assessment of CKD risk factors that we used as potential confounders, but these variables have been widely used in other studies using THIN.37–39 Another limitation is that because biologic medications, such as infliximab, adalimumab, and vedolizumab, are not usually prescribed by general practitioners in the United Kingdom, we could not assess the effect of these medications on eGFR change. However, these medications have not been associated with CKD among patients with IBD in prior studies,40–42 so their inclusion is unlikely to alter the results. Furthermore, like other studies of IBD using administrative data, we could not assess markers of disease severity, such as stool frequency, endoscopic mucosal appearance, abdominal pain, or surgery. However, we used a validated definition of IBD flare based on corticosteroid use to determine that acute flare was not associated with lower eGFR. Finally, in the analysis of the association between IBD medication use and eGFR change, we could not account for medication dose in the analysis because dose has not been validated in THIN. However, several sensitivity analyses, including changing the exposure window to 30 and 120 days and considering cumulative days of medication use, did not change the study conclusions.

Conclusions

This study illustrates that IBD is a risk factor for CKD, and that the relative hazard of CKD caused by IBD declines with increasing age. Additionally, this study demonstrates that the use of common nonbiologic IBD medications, including mesalamine, azathioprine, or methotrexate, is not associated with lower eGFR. Given the overall association of IBD with CKD, intermittent monitoring of renal function in all patients with IBD may be indicated as part of usual care. Future studies should focus on prospectively identifying mediators of the CKD-IBD association.

Supplementary Material

1

What You Need to Know.

Background

We investigated what factors affect risk of chronic kidney disease (CKD) in patients with inflammatory bowel diseases (IBD).

Findings

In a retrospective study of more than 80,000 persons, we found that IBD is associated with increased risk of CKD, and the hazard ratio is highest among younger patients. Common non-biologic therapies for IBD are not associated with lower estimated glomerular filtration rates.

Implications for patient care

Young patients with IBD should be monitored for CKD; mesalamine treatment does not seem to increase risk for CKD.

Acknowledgments

Funding

Ravy K. Vajravelu received funding from National Institutes of Health (NIH)/National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK; T32-DK007066, K08-DK119475). Frank I. Scott received funding from NIH/NIDDK (K08-DK095951). Ronac Mamtani received funding from NIH/National Cancer Institute (K23-CA187185). James D. Lewis received funding from NIH/NIDDK (K24-DK078228). Michelle R. Denburg receive funding from NIH/NIDDK (K23-DK093556).

Conflicts of interest

These authors disclose the following: Mark T. Osterman has served as a consultant to AbbVie, Janssen, Lycera, Merck, Pfizer, Takeda, and UCB, and the scope of work was unrelated to this research. He has received research funding from UCB, also unrelated to this research. Frank I. Scott has served as a consultant to Janssen and Merck. The scope of work was unrelated to this research. He has received research funding from Takeda and Janssen, also unrelated to this research. James D. Lewis has served as a consultant to AbbVie, Arena Pharmaceuticals, Bristol-Myers Squibb, Bridge Biotherapeutics, Celgene, Gilead, Janssen, Johnson & Johnson Consumer Inc, Lilly, Merck, Nestle Health Science, Pfizer, Samsung Bioepis, Takeda, and UCB. The scope of work for each of these was unrelated to this research. He has received research funding from Janssen, Nestle Health Science, and Takeda, also unrelated to this research. He received honorarium for participation in a CME activity sponsored by Nestle Health Science, also unrelated to this research. Michelle R. Denburg has received research funding from Mallinckrodt Pharmaceuticals, unrelated to this research.

Abbreviations used in this paper:

aHR

adjusted hazard ratio

CI

confidence interval

CKD

chronic kidney disease stages 3–5

eGFR

estimated glomerular filtration rate

IBD

inflammatory bowel disease

IQR

interquartile range

IR

incidence rate

THIN

The Health Improvement Network

Footnotes

Supplementary Material

Note: To access the supplementary material accompanying this article, visit the online version of Clinical Gastroenterology and Hepatology at www.cghjournal.org, and at https://doi.org/10.1016/j.cgh.2019.10.043.

References

  • 1.Peyrin-Biroulet L, Loftus EV Jr, Colombel JF, et al. Long-term complications, extraintestinal manifestations, and mortality in adult Crohn’s disease in population-based cohorts. Inflamm Bowel Dis 2011;17:471–478. [DOI] [PubMed] [Google Scholar]
  • 2.Van Staa TP, Travis S, Leufkens HGM, et al. 5-aminosalicylic acids and the risk of renal disease: a large British epidemiologic study. Gastroenterology 2004;126:1733–1739. [DOI] [PubMed] [Google Scholar]
  • 3.Lewis B, Mukewar S, Lopez R, et al. Frequency and risk factors of renal insufficiency in inflammatory bowel disease inpatients. Inflamm Bowel Dis 2013;19:1846–1851. [DOI] [PubMed] [Google Scholar]
  • 4.Primas C, Novacek G, Schweiger K, et al. Renal insufficiency in IBD: prevalence and possible pathogenetic aspects. J Crohns Colitis 2013;7:e630–e634. [DOI] [PubMed] [Google Scholar]
  • 5.Bernstein CN, Wajda A, Blanchard JF. The clustering of other chronic inflammatory diseases in inflammatory bowel disease: a population-based study. Gastroenterology 2005;129:827–836. [DOI] [PubMed] [Google Scholar]
  • 6.Park S, Chun J, Han KD, et al. Increased end-stage renal disease risk in patients with inflammatory bowel disease: a nationwide population-based study. World J Gastroenterol 2018;24:4798–4808. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Novis BH, Korzets Z, Chen P, et al. Nephrotic syndrome after treatment with 5-aminosalicylic acid. Br Med J (Clin Res Ed) 1988;296:1442. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Barbour VM, Williams PF. Nephrotic syndrome associated with sulphasalazine. BMJ 1990;301:818. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Dwarakanath AD, Michael J, Allan RN. Sulphasalazine induced renal failure. Gut 1992;33:1006–1007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Fornaciari G, Maccari S, Borgatti PP, et al. Nephrotic syndrome from 5-ASA for ulcerative colitis? Complicated by carcinoma of the colon and sclerosing cholangitis. J Clin Gastroenterol 1997; 24:37–39. [DOI] [PubMed] [Google Scholar]
  • 11.Schreiber S, Hamling J, Zehnter E, et al. Renal tubular dysfunction in patients with inflammatory bowel disease treated with aminosalicylate. Gut 1997;40:761–766. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Thuluvath PJ, Ninkovic M, Calam J, et al. Mesalazine induced interstitial nephritis. Gut 1994;35:1493–1496. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Witte T, Olbricht CJ, Koch KM. Interstitial nephritis associated with 5-aminosalicylic acid. Nephron 1994;67:481–482. [DOI] [PubMed] [Google Scholar]
  • 14.Mahmud N, McDonald GS, Kelleher D, et al. Microalbuminuria correlates with intestinal histopathological grading in patients with inflammatory bowel disease. Gut 1996;38:99–103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Kreisel W, Wolf LM, Grotz W, et al. Renal tubular damage: an extraintestinal manifestation of chronic inflammatory bowel disease. Eur J Gastroenterol Hepatol 1996;8:461–468. [PubMed] [Google Scholar]
  • 16.Ambruzs JM, Walker PD, Larsen CP. The histopathologic spectrum of kidney biopsies in patients with inflammatory bowel disease. Clin J Am Soc Nephrol 2014;9:265–270. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Greenstein AJ, Sachar DB, Panday AK, et al. Amyloidosis and inflammatory bowel disease. A 50-year experience with 25 patients. Medicine (Baltimore) 1992;71:261–270. [DOI] [PubMed] [Google Scholar]
  • 18.Parks JH, Worcester EM, O’Connor RC, et al. Urine stone risk factors in nephrolithiasis patients with and without bowel disease. Kidney Int 2003;63:255–265. [DOI] [PubMed] [Google Scholar]
  • 19.Cury DB, Moss AC, Schor N. Nephrolithiasis in patients with inflammatory bowel disease in the community. Int J Nephrol Renovasc Dis 2013;6:139–142. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Alexander RT, Hemmelgarn BR, Wiebe N, et al. Kidney stones and kidney function loss: a cohort study. BMJ 2012;345:e5287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Denburg MR, Jemielita TO, Tasian GE, et al. Assessing the risk of incident hypertension and chronic kidney disease after exposure to shock wave lithotripsy and ureteroscopy. Kidney Int 2016;89:185–192. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Denburg MR, Haynes K, Shults J, et al. Validation of The Health Improvement Network (THIN) database for epidemiologic studies of chronic kidney disease. Pharmacoepidemiol Drug Saf 2011;20:1138–1149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Smith DH, Shetterly S, Flory J, et al. Diagnosis-based cohort augmentation using laboratory results data: the case of chronic kidney disease. Pharmacoepidemiol Drug Saf 2018;27:872–877. [DOI] [PubMed] [Google Scholar]
  • 24.Chan MR, Dall AT, Fletcher KE, et al. Outcomes in patients with chronic kidney disease referred late to nephrologists: a meta-analysis. Am J Med 2007;120:1063–1070. [DOI] [PubMed] [Google Scholar]
  • 25.Lewis JD, Bilker WB, Weinstein RB, et al. The relationship between time since registration and measured incidence rates in the General Practice Research Database. Pharmacoepidemiol Drug Saf 2005;14:443–451. [DOI] [PubMed] [Google Scholar]
  • 26.Lewis JD, Brensinger C, Bilker WB, et al. Validity and completeness of the General Practice Research Database for studies of inflammatory bowel disease. Pharmacoepidemiol Drug Saf 2002;11:211–218. [DOI] [PubMed] [Google Scholar]
  • 27.Pigneur B, Escher J, Elawad M, et al. Phenotypic characterization of very early-onset IBD due to mutations in the IL10, IL10 receptor alpha or beta gene: a survey of the Genius Working Group. Inflamm Bowel Dis 2013;19:2820–2828. [DOI] [PubMed] [Google Scholar]
  • 28.Atiyeh BA, Dabbagh SS, Gruskin AB. Evaluation of renal function during childhood. Pediatr Rev 1996;17:175–180. [DOI] [PubMed] [Google Scholar]
  • 29.Heilbron DC, Holliday MA, al-Dahwi A, et al. Expressing glomerular filtration rate in children. Pediatr Nephrol 1991;5:5–11. [DOI] [PubMed] [Google Scholar]
  • 30.Schwartz GJ, Munoz A, Schneider MF, et al. New equations to estimate GFR in children with CKD. J Am Soc Nephrol 2009; 20:629–637. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009; 150:604–612. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Lewis JD, Aberra FN, Lichtenstein GR, et al. Seasonal variation in flares of inflammatory bowel disease. Gastroenterology 2004; 126:665–673. [DOI] [PubMed] [Google Scholar]
  • 33.Grainge MJ, West J, Card TR. Venous thromboembolism during active disease and remission in inflammatory bowel disease: a cohort study. Lancet 2010;375:657–663. [DOI] [PubMed] [Google Scholar]
  • 34.Zallot C, Billioud V, Frimat L, et al. 5-Aminosalicylates and renal function monitoring in inflammatory bowel disease: a nationwide survey. J Crohns Colitis 2013;7:551–555. [DOI] [PubMed] [Google Scholar]
  • 35.Vadasz Z, Haj T, Kessel A, et al. Age-related autoimmunity. BMC Med 2013;11:94. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Herzog D, Fournier N, Buehr P, et al. Age at disease onset of inflammatory bowel disease is associated with later extraintestinal manifestations and complications. Eur J Gastroenterol Hepatol 2018;30:598–607. [DOI] [PubMed] [Google Scholar]
  • 37.Vajravelu RK, Scott FI, Mamtani R, et al. Medication class enrichment analysis: a novel algorithm to analyze multiple pharmacologic exposures simultaneously using electronic health record data. J Am Med Inform Assoc 2018; 25:780–789. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Ogdie A, Yu Y, Haynes K, et al. Risk of major cardiovascular events in patients with psoriatic arthritis, psoriasis and rheumatoid arthritis: a population-based cohort study. Ann Rheum Dis 2015;74:326–332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Vajravelu ME, Keren R, Weber DR, et al. Incidence and risk of celiac disease after type 1 diabetes: a population-based cohort study using the health improvement network database. Pediatr Diabetes 2018;19:1422–1428. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Infliximab package insert. Available at: http://www.janssenlabels.com/package-insert/product-monograph/prescribing-information/REMICADE-pi.pdf07/24/18. Accessed July 24, 2018.
  • 41.Adalimumab package insert. Available at: https://www.rxabbvie.com/pdf/humira.pdf07/24/18. Accessed July 24, 2018.
  • 42.Vedolizumab package insert. Available at: https://general.takedapharm.com/ENTYVIOPI 07/24/18. Accessed July 24, 2018.

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1542338-supplement-1.pdf (1,006.2KB, pdf)

RESOURCES