Risk Prediction Models for Atherosclerotic Cardiovascular Disease in Patients with Chronic Kidney Disease: The CRIC Study

Abstract

BackgroundIndividuals with CKD may be at high risk for atherosclerotic cardiovascular disease (ASCVD). However, there are no ASCVD risk prediction models developed in CKD populations to inform clinical care and prevention.MethodsWe developed and validated 10-year ASCVD risk prediction models in patients with CKD that included participants without self-reported cardiovascular disease from the Chronic Renal Insufficiency Cohort (CRIC) study. ASCVD was defined as the first occurrence of adjudicated fatal and nonfatal stroke or myocardial infarction. Our models used clinically available variables and novel biomarkers. Model performance was evaluated based on discrimination, calibration, and net reclassification improvement.ResultsOf 2604 participants (mean age 55.8 years; 52.0% male) included in the analyses, 252 had incident ASCVD within 10 years of baseline. Compared with the American College of Cardiology/American Heart Association pooled cohort equations (area under the receiver operating characteristic curve [AUC]=0.730), a model with coefficients estimated within the CRIC sample had higher discrimination (P=0.03), achieving an AUC of 0.736 (95% confidence interval [CI], 0.649 to 0.826). The CRIC model developed using clinically available variables had an AUC of 0.760 (95% CI, 0.678 to 0.851). The CRIC biomarker-enriched model had an AUC of 0.771 (95% CI, 0.674 to 0.853), which was significantly higher than the clinical model (P=0.001). Both the clinical and biomarker-enriched models were well-calibrated and improved reclassification of nonevents compared with the pooled cohort equations (6.6%; 95% CI, 3.7% to 9.6% and 10.0%; 95% CI, 6.8% to 13.3%, respectively).ConclusionsThe 10-year ASCVD risk prediction models developed in patients with CKD, including novel kidney and cardiac biomarkers, performed better than equations developed for the general population using only traditional risk factors.

A 4-Variable Model to Predict Cardio-Kidney Events and Mortality in Chronic Kidney Disease: The Chronic Renal Insufficiency Cohort (CRIC) Study

“…Regarding the prediction of atherosclerotic events, Bundy et al [4] published an enriched model of 10 variables, including NT-proBNP and hs-cTnT, in the CRIC Study. Although the performance of the Bundy model (c-statistic = 0.77) is slightly better than that of our 4-variable model (c-statistic = 0.75), we believe that the greater simplicity favours the clinical implementation of our tool.…”

“…Current prognostic models for CKD are complex (i.e., they include multiple variables) and were designed to predict a single outcome (i.e., there are different models to predict ischaemic events, HF outcomes, CKD progression, and mortality) [4][5][6]. Therefore, simple models that capture the combined risk of CV events, HF, CKD progression, and mortality are lacking.…”

“…However, the KFRE was not validated to predict the risk of CV events that frequently occur before reaching the kidney failure stage [9]. Additional models have been developed to predict atherosclerotic CV disease, including 7–10 variables [4], and HF, with 10 or more variables of various clinical, biomarker, and echocardiographic domains [5, 6].…”

Development of machine learning-based models to predict 10-year risk of cardiovascular disease: a prospective cohort study

“…Compared with conventional prediction methods, the machine learning (ML)-based approach has outstanding advantages, as it can effectively handle massive amounts of time-to-event data featuring multidimensional space 12–21. However, they were commonly restricted to limited follow-up time,12 13 inclusion of too many covariates,14 or utilisation of prespecified domain of variables15–18 or participants,19 20 narrowing their applications to research or expertise settings. All these limitations underscore the need to construct novel cardiovascular prediction to boost better risk prediction and stratification in real-world clinical practice.…”

Cardiovascular Risk Prediction Scores in CKD: What Are We Missing?

“…In this issue of JASN , Bundy et al 5 . used the Chronic Renal Insufficiency Cohort (CRIC) to evaluate the performance of the PCEs in the CKD population and construct new models using CKD-specific risk factors, including routinely available laboratory values and novel biomarkers for the prediction of incident ASCVD events.…”