Fig 3 from Yang et al. © National Kidney Foundation

In recent years, several studies have attempted to identify risk factors for progression of chronic kidney disease (CKD). The majority of these studies utilize end-stage renal disease (ESRD) as an outcome, defined as either initiation of renal replacement therapy or kidney transplantation. Other studies utilize measures of kidney function, eg, creatinine, in reporting CKD progression.  Because of these differing endpoints, there remains significant uncertainty regarding the predictability of CKD progression in relation to the risk factors. In an attempt to better define such relations, Yang and colleagues used separate models for outcomes based on kidney disease events and death to address 2 specific questions: (1) are risk factor–outcome associations for CKD progression sensitive to the definition of an event of CKD progression? and (2) are the risk factor–death associations similar to those for various kidney disease outcomes?  The authors hope that their findings may better inform the selection of outcomes in studies of risk factors for CKD progression in the CRIC (Chronic Renal Insufficiency Cohort) Study and future prospective studies of CKD. Author Dr. Dawei Xie (DX) discusses this study with Dr. Edgar Lerma (eAJKD), eAJKD Contributor.

eAJKD: Please give us a summary of the CRIC Study – when was it started and what are the original objectives of the study?

DX: In order to advance our understanding of the epidemiology of mild-to-moderate CKD, the National Institute of Diabetes and Digestive and Kidney Disease (NIDDK) established the CRIC Study in 2001. The objective of the CRIC Study is to explore and elucidate the natural history, etiologic determinants, and associated morbidity and mortality of mild-to-moderate CKD. From 2003 to 2008, the CRIC Study successfully recruited 3939 participants from seven national clinical centers. Annual follow-up continues to-date, and recruitment of an additional 1500 participants began in July 2013. The cohort’s generalizability is enriched by a broad representation of age, race, and diabetic patients, as well as a supplemental Hispanic cohort (H-CRIC). Through several areas of scientific inquiry, it is CRIC’s goal to develop predictive models for high-risk subgroups with CKD, as well to unveil novel etiologic factors that could potentially become intervention targets for future randomized control trials to curb progression of kidney disease and associated comorbidity.

eAJKD: What are the major findings of your study?

DX: For this study, we demonstrated that kidney disease outcomes and death had substantially different risk factor profiles in persons with CKD, and that the association of risk factors, other than baseline eGFR level, did not vary significantly across different kidney disease outcome definitions, including ESRD alone and in combination with various estimated glomerular filtration rate (eGFR)-based events. Risk factors that are significantly associated with higher ESRD risk include older age, male gender, black race and other races/ethnicity, diabetes, lower eGFR, and proteinuria.

eAJKD: Please comment on the limitations of your study, and how do you think did they influence the results?

DX: CRIC subjects were recruited from seven centers representing 13 clinical sites.  These were not a representative sample of all CKD patients nationwide. Thus, risk factor relationships explored in this analysis may not reproduce in all CKD populations. Nonetheless, given the diversity of CRIC subjects, the findings from this study, which are mainly based on associations, should be applicable to CKD population in general. Second, some eGFR events were defined with respect to the baseline eGFR values, eg, eGFR halving from baseline. It is possible that the findings may be different in other CKD cohorts if the distribution of baseline eGFR is different. Third, we selected only a few established CKD risk factors to compare their associations with different outcomes. It is possible that some CKD risk factors may have different associations with different kidney disease outcome definitions.

eAJKD: You mentioned that analyzing CKD progression in this cohort is particularly challenging because “death is a competing risk for kidney disease outcomes.” Please elaborate on this concept.

DX: When a patient with CKD dies before experiencing a kidney disease outcome, the estimated probability of kidney disease outcomes is altered.  When death and kidney disease outcomes are considered separately, death is a competing risk for kidney disease outcomes as one can’t have kidney disease outcomes after death. There are a lot of ways to approach competing risk. As the main interest of this study is to estimate the cause-specific role of certain risk factors, we thought it appropriate to treat death as a censoring event in the standard cox proportional hazard models of kidney disease outcomes. Nonetheless, we did examine the same risk factor-renal outcome relationships using competing risk models, and the results were nearly identical to our main findings.

eAJKD: As clinicians, we are always trying to identify which patients are going to benefit from screening and early interventions by identifying risk factors for CKD progression. From a clinician’s perspective, what are the potential implications of this study? 

DX: The primary goal of this study was to inform the selection of renal outcomes for CKD studies (both experimental and observational). Our study provides guidance to researchers to examine associations of new risk factors separately for renal endpoints and death, and that choice of eGFR-based outcomes should not materially alter the findings of any given study. Analyzing renal endpoints and death separately will provide better evidence of risk factors for CKD progression that will be meaningful and useful for patients and their health care providers.

To view the article abstract or full-text (subscription required), please visit AJKD.org.