Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Jan 11.
Published in final edited form as: Eur J Prev Cardiol. 2023 Jan 11;30(1):8–16. doi: 10.1093/eurjpc/zwac176

Including Measures of Chronic Kidney Disease to Improve Cardiovascular Risk Prediction by SCORE2 and SCORE2-OP

Kunihiro Matsushita 1,*, Stephen Kaptoge 2,*, Steven HJ Hageman 3,*, Yingying Sang 4, Shoshana H Ballew 5, Morgan E Grams 6, Aditya Surapaneni 7, Luanluan Sun 8, Johan Arnlov 9, Milica Bozic 10, Hermann Brenner 11, Nigel J Brunskill 12, Alex R Chang 13, Rajkumar Chinnadurai 14, Massimo Cirillo 15, Adolfo Correa 16, Natalie Ebert 17, Kai-Uwe Eckardt 18, Ron T Gansevoort 19, Orlando Gutierrez 20, Farzad Hadaegh 21, Jiang He 22, Shih-Jen Hwang 23, Tazeen H Jafar 24, Simerjot K Jassal 25, Takamasa Kayama 26, Csaba P Kovesdy 27, Gijs W Landman 28, Andrew S Levey 29, Donald M Lloyd-Jones 30, Rupert W Major 31, Katsuyuki Miura 32, Paul Muntner 33, Girish N Nadkarni 34, Christoph Nowak 35, Takayoshi Ohkubo 36, Michelle J Pena 37, Kevan R Polkinghorne 38, Toshimi Sairenchi 39, Elke Schaeffner 40, Markus P Schneider 41, Varda Shalev 42, Michael G Shlipak 43, Marit D Solbu 44, Nikita Stempniewicz 45, James Tollitt 46, José M Valdivielso 47, Joep van der Leeuw 48, Angela Yee-Moon Wang 49, Chi-Pang Wen 50, Mark Woodward 51, Kazumasa Yamagishi 52, Hiroshi Yatsuya 53, Luxia Zhang 54, Jannick AN Dorresteijn 55, Emanuele Di Angelantonio 56, Frank LJ Visseren 57,†, Lisa Pennells 58,†, Josef Coresh 59,†; Chronic Kidney Disease Prognosis Consortium
PMCID: PMC9839538  NIHMSID: NIHMS1844083  PMID: 35972749

Abstract

Aims

The 2021 ESC guideline on cardiovascular disease (CVD) prevention categorizes moderate and severe chronic kidney disease (CKD) as high and very-high CVD risk status regardless of other factors like age and does not include estimated glomerular filtration rate (eGFR) and albuminuria in its algorithms, SCORE2 and SCORE2-OP, to predict CVD risk. We developed and validated an “Add-on” to incorporate CKD measures into these algorithms, using a validated approach.

Methods

In 3,054,840 participants from 34 datasets, we developed three Add-ons (eGFR only, eGFR + urinary albumin-to-creatinine ratio [ACR] [the primary Add-on], and eGFR + dipstick proteinuria) for SCORE2 and SCORE2-OP. We validated c-statistics and net reclassification improvement (NRI), accounting for competing risk of non-CVD death, in 5,997,719 participants from 34 different datasets.

Results

In the target population of SCORE2 and SCORE2-OP without diabetes, the CKD Add-on (eGFR only) and CKD Add-on (eGFR + ACR) improved c-statistic by 0.006 (95%CI 0.004-0.008) and 0.016 (0.010-0.023), respectively, for SCORE2 and 0.012 (0.009-0.015) and 0.024 (0.014-0.035), respectively, for SCORE2-OP. Similar results were seen when we included individuals with diabetes and tested the CKD Add-on (eGFR + dipstick). In 57,485 European participants with CKD, SCORE2 or SCORE2-OP with a CKD Add-on showed a significant NRI (e.g., 0.100 [0.062-0.138] for SCORE2) compared to the qualitative approach in the ESC guideline.

Conclusion

Our Add-ons with CKD measures improved CVD risk prediction beyond SCORE2 and SCORE2-OP. This approach will help clinicians and patients with CKD refine risk prediction and further personalize preventive therapies for CVD.

Keywords: chronic kidney disease, cardiovascular disease, risk prediction, meta-analysis

Introduction

Chronic kidney disease (CKD) affects more than 10% of the adult population globally and is widely recognized as an important risk factor for cardiovascular disease (CVD).1,2 Indeed, in the 2021 European Society of Cardiology (ESC) guideline on CVD prevention,3 individuals with moderate and severe CKD (according to the KDIGO staging system based on reduced glomerular filtration rate [GFR] and elevated albuminuria4) are regarded as high and very high-risk of CVD, respectively. However, such a qualitative approach misses an opportunity to personalize CVD preventive therapies according to quantitative measures of CKD, which are often readily available in clinical practice, in addition to traditional CVD risk factors.

We recently developed and validated a new approach, “CKD Add-on”,5 that allows the inclusion of information on the two CKD measures, GFR and albuminuria, into existing prediction models. With this approach, the original predicted risk of CVD is calibrated in the individual participant having GFR (or albuminuria) that differs from their expected GFR based upon the profile of their demographic and risk factor characteristics. Using this approach, the two CKD measures have significantly improved CVD risk prediction beyond two reference CVD risk prediction models, the Pooled Cohort Equation (PCE)6 and SCORE.5,7

Here, we sought to develop and validate a CKD Add-on for SCORE2 and SCORE2-OP (i.e., the risk prediction algorithms adopted by the 2021 ESC CVD prevention guideline), using data from the CKD Prognosis Consortium (CKD-PC). We also compared risk classification between our quantitative approach with a CKD Add-on and the qualitative approach proposed in the 2021 ESC guideline.

Methods

Study populations

The data sources were 68 datasets taking part in CKD-PC with individual-level data necessary for this specific study (namely, GFR, albuminuria, traditional CVD risk factors, and CVD outcomes defined below). These cohorts included both prospective research cohorts and health system datasets and enrolled participants from 41 countries from Europe, the Middle East, Asia, Australasia, and the Americas. These cohorts represented general population cohorts (no specific selection of some clinical conditions), high-risk cohorts (selection of some specific clinical conditions but not exclusively CKD), and CKD cohorts (explicit inclusion of individuals with CKD). This project included cohorts with 50 or more CVD outcomes and 95th percentile of follow-up time longer than 5 years among eligible participants without a history of CVD at baseline. This study was approved for use of de-identified data by the institutional review board at the Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA (#IRB00003324). The need for informed consent was waived by the institutional review board.

Both SCORE2 and SCORE2-OP were designed for adults aged 40-69 years and those aged ≥70 years, respectively, but were derived from datasets including individuals with broader age ranges. Such an age margin is advantageous to obtain reliable coefficients of the interaction terms between age and predictors at relevant age thresholds. Thus, for the development of the CKD Add-on, we applied an age margin of 10 years and included all eligible adults aged ≥30 years for SCORE2 and those aged ≥60 years for SCORE2-OP.8 Nonetheless, as detailed below, the validation of the CKD Add-on was restricted to individuals in the target age range of SCORE2 (40-69 years) and SCORE2-OP (≥70 years).

The 2021 ESC guideline classifies all individuals with diabetes mellitus as moderate to very high risk according to the disease duration and the presence of end organ damage.3 SCORE2 algorithms are therefore proposed for individuals without diabetes.8 However, the development of SCORE2 algorithms included diabetes as a covariate, to facilitate recalibration of the models using CVD incidence rates from the general population that included individuals with diabetes.8 Thus, we also included individuals with diabetes in the development of the CKD Add-on. Nonetheless, to match the proposed target population of SCORE2 algorithms, our primary validation was focused on the population without diabetes, and we secondarily explored data from the entire population including diabetes.

CKD measures

We focused on the two key CKD measures used for CKD staging in nephrology clinical guidelines, GFR and albuminuria.9 Estimated GFR (eGFR) was calculated using the 2021 CKD Epidemiology Collaboration (CKD-EPI) creatinine-based equation (but results were similar when an Add-on was developed for the 2009 CKD-EPI eGFR creatinine-based equation).10 Albuminuria was ascertained primarily as urine albumin-to-creatinine ratio (ACR) 9 but secondarily included dipstick proteinuria. Data on urine protein-to-creatinine ratio was converted to ACR using a validated equation when ACR information was not available.11

Traditional CVD risk factors

We considered the following predictors in SCORE2 and SCORE2-OP as traditional CVD risk factors: age, sex, smoking status (current vs. non-current), diabetes, systolic blood pressure, total cholesterol, and high-density lipoprotein cholesterol.

CVD outcome

Following the development process of SCORE2 and SCORE2-OP,8 CVD outcome of interest was a composite of myocardial infarction, stroke, and CVD mortality. Web Appendix 1 summarizes details of how each cohort defined CVD events.

Statistical analysis

We first summarized characteristics (e.g., continuous variables as mean [SD] or median [IQI] and categorical variables as proportion or counts) in development and validation datasets. In general, we conducted two-stage meta-analysis in which each cohort was analyzed separately, and then the relevant estimates were pooled using random-effects models.12,13

Following the process of developing the CKD Add-ons for PCE and SCORE,5 we used 34 datasets able to share de-identified individual-level data with the CKD-PC Data Coordinating Center as development datasets. These datasets represented a wide range of populations, including the general population. The remaining 33 datasets, which could not share individual-level data or included highly selected populations (e.g., only CKD patients), were included as validation datasets. An exception was that we randomly split the OptumLabs® Data Warehouse (OLDW) cohorts into equal halves for the development and validation in order to have a good representation of health system databases for validation. The OLDW is a longitudinal, real-world data asset with de-identified administrative claims and electronic health record data. Our datasets also included clinical trial cohorts, and we confirmed the results are consistent after excluding these cohorts. Even in those studies that could not share individual-level data, collaborators ran a statistical code specific for the present study and shared relevant estimates and variance-covariance with the CKD-PC Data Coordinating Center, and thus the present study should be considered as individual-level data meta-analysis.

Using the previously published method,5,14 we first developed the “CKD Add-on” using the development datasets. The CKD Add-on method consists of the following three steps: 1: linear regression models to estimate expected levels of eGFR and log-ACR according to traditional CVD risk factors; 2: subdistribution hazard ratios (sub-HRs) of CVD outcome for eGFR and log-ACR adjusted for traditional risk factors; and 3: the calibration of predicted CVD risk based on the deviation between actual eGFR and log-ACR and expected eGFR and log-ACR (from the first step) and their adjusted sub-HRs (from the second step) in every individual. In the first two steps, we included all possible two-way interaction terms with age. One exception was log-ACR in the second step since age did not statistically significantly modify the association of log-ACR with CVD risk (p=0.12). In the second step, log-sub-HRs for traditional CVD risk factors were fixed according to the original SCORE2 or SCORE2-OP coefficients, and eGFR was modeled with two knots at 60 and 90 ml/min/1.73m2 to reflect well-known J-shaped associations between eGFR and CVD risk.2 Since the main purpose of a CKD Add-on is to enhance the predicted risk related to reduced eGFR (but not necessarily high eGFR), we only applied sub-HRs for eGFR below 90 ml/min/1.73m2 when we implemented CKD Add-ons. Following the development process of SCORE2 and SCORE2-OP,8 we used sub-HRs based on Fine and Gray models accounting non-CVD death as a competing outcome. In studies with only data on dipstick proteinuria, we secondarily developed a CKD Add-on for dipstick proteinuria and eGFR. Given that eGFR is more widely available than albuminuria in clinical practice, as we did previously,5 we developed a CKD Add-on with eGFR only first (expressed as CKD Add-on [eGFR only] below). Subsequently, we developed a CKD Add-on with eGFR and measures of albuminuria (CKD Add-on [eGFR + ACR] and CKD Add-on [eGFR + dipstick], with the former as our primary Add-on).

Using the validation datasets, we assessed the following prediction statistics after applying CKD Add-ons: Harrel’s c-statistic as a measure of risk discrimination15 and categorical net reclassification improvement (NRI).16 According to the 2021 ESC guideline,3 we categorized predicted risk into age-specific categories of low/moderate, high, and very high CVD risk. The corresponding 10-year risk thresholds were 2.5% and 7.5% in age <50 years, 5% and 10% in 50-69 years, and 7.5% and 15% in ≥70 years. We used normal approximations to calculate 95% confidence intervals of c-statistics and NRI. We primarily used the study-specific recalibrated baseline risk of each cohort since the evaluation of the improvement of an established risk equation like SCORE2 is predicated on the assumption that the established equation is well-calibrated in the relevant cohort. We, a priori, selected the Clinical Practice Research Datalink (CPRD) for the validation of calibration, since both SCORE2 and SCORE2-OP were well-calibrated in this UK dataset.8 As done previously,5 in CKD cohorts, as the expected values of CKD measures, we used the mean of eGFR and albuminuria in each cohort given overestimation of expected eGFR and underestimation of expected ACR when relying on linear regression models from non-CKD cohorts.

We conducted additional analyses to evaluate the public health and clinical implications of the CKD Add-ons. First, we described the median ratio of newly predicted risk with a CKD Add-on to originally predicted risk without a CKD Add-on; we took the median and IQI of median ratios from individual datasets. Second, we explored four clinical scenarios with a specific combination of traditional CVD risk factors and described the changes in predicted risk before and after applying a CKD Add-on for two sets of levels of eGFR and ACR representing moderate and severe CKD (eGFR 45 ml/min/1.73m2 + ACR 150 mg/g and eGFR 25 ml/min/1.73m2 + ACR 500 mg/g, respectively). Finally, we evaluated NRI when we applied SCORE2 or SCORE2-OP, as appropriate, with a CKD Add-on instead of the approaches recommended in the 2021 ESC guideline on CVD prevention (i.e., qualitative classification in moderate and severe CKD and quantitative risk prediction using SCORE2 or SCORE2-OP in mild CKD).

All analyses used complete datasets and were conducted with STATA 16 (College Station, TX). We followed the TRIPOD statement for reporting.17

Results

Study Characteristics

Development datasets and validation datasets included 3,054,840 individuals and 5,997,719 individuals, respectively. Summary characteristics were largely similar between development and validation datasets, although the proportion of men was greater in the validation datasets than in the development datasets (Table 1). Characteristics across individual studies are summarized in Web Table 1.

Table 1.

Overall baseline characteristics for development and validation datasets.

Development datasets Validation datasets
Number of datasets 34 34
N of participants 3,054,840 5,997,719
Age (SD), y 54 (14) 55 (14)
Male sex, % 43 56
Current smokers, % 7.1 19
Systolic BP (SD), mmHg 126 (17) 127 (17)
Diabetes, % 18 18
Total cholesterol (SD), mmol/L 4.8 (0.9) 4.9 (0.9)
HDL cholesterol (SD), mmol/L 1.4 (0.4) 1.3 (0.4)
eGFR (SD), ml/min/1.73m2 90 (19) 91 (19)
N for ACR 625,531 (21%) 1,429,373 (26%)
ACR (IQI), mg/g 11 (6-28) 9 (4-29)
N for dipstick 947,323 (36%) 1,229,141 (40%)
Dipstick ≥1+, % 9.1 8.1
 
Follow-up (SD), y 3.7 (3.6) 4.6 (3.6)
Number of CVD events 90,650 142,379
10-y baseline risk (IQI)*
 Men 0.059 (0.031-0.069) 0.050 (0.034-0.064)
 Women 0.030 (0.017-0.042) 0.029 (0.021-0.041)
 Older men 0.202 (0.128-0.257) 0.155 (0.135-0.206)
 Older women 0.141 (0.088-0.174) 0.108 (0.083-0.149)
Open in a new tab

Values indicated count, proportion, mean (SD), or median (IQI).

*

Baseline risk was estimated in the 10-year time frame with each predictor centered at age 60 years, systolic blood pressure 120 mmHg, total cholesterol 6 mmol/L, HDL cholesterol 1.3 mmol/L, never smokers, and no diabetes for younger age scenario and 73 years, 150 mmHg, 6 mmol/L, and 1.4 mmol/L for older age scenarios (smoking status and diabetes stayed the same). For cohorts with only 5-year follow-up time, 5-year baseline risk was converted to 10-year by 1-(1-risk)2. Between-study difference was considerable even within the development or validation datasets, and the IQIs for the baseline risk in the validation studies overlap the estimates in the development datasets.

Development of CKD Add-ons in the Development Datasets

The coefficients of traditional CVD risk factors for estimating expected eGFR and log-ACR are displayed in Web Table 2. Older age and lower HDL cholesterol were associated with lower baseline eGFR. Higher systolic blood pressure, diabetes, and lower eGFR were the major correlates of higher baseline log-ACR. As anticipated,2,5 both lower eGFR and higher ACR were significantly associated with elevated CVD risk (Table 2), in the context of both SCORE2 and SCORE2-OP. Sub-HR per 15 ml/min/1.73m2 lower eGFR below 60 ml/min/1.73m2 was greater when we investigated adults aged ≥30 years compared to when we restricted to older adults aged ≥60 years (1.74 [1.64, 1.84] at age 55 vs. 1.33 [1.25, 1.40] at age 75). Sub-HR for higher ACR was similar regardless of age. Dipstick proteinuria also demonstrated a dose-response relationship with CVD risk. When we excluded a cluster randomized community-level intervention trial in the development datasets, results were almost identical (Web Table 3).

Table 2.

Meta-analyzed sub hazard ratios (95% CI) in development datasets

Variables Sub hazard ratio (95% CI) Sub hazard ratio (95% CI)
CKD Add-on (eGFR only) Age 30+* CKD Add-on (eGFR only) Age 60+**
eGFR <60 at age 55, per −15 ml 1.74 (1.64, 1.84) eGFR <60 at age 75, per −15 ml 1.33 (1.25, 1.40)
eGFR 60-89 at age 55, per −15 ml 1.09 (1.00, 1.19) eGFR <90 at age 75, per −15 ml 1.08 (1.05, 1.11)
eGFR 90+ at age 55, per −15 ml 0.75 (0.70, 0.82) eGFR 90+ at age 75, per −15 ml 0.62 (0.52, 0.74)
eGFR <60 × age, per −15 ml × 5y 0.92 (0.91, 0.94) eGFR <60 × age, per −15 ml × y 0.99 (0.98, 0.99)
eGFR 60-89 × age, per −15 ml × 5y 1.01 (0.98, 1.03) eGFR <90 × age, per −15 ml × y 0.99 (0.98, 1.00)
eGFR 90+ × age, per −15 ml × 5y 0.98 (0.95, 1.00) eGFR 90+ × age, per −15 ml × y 0.99 (0.98, 1.01)
CKD Add-on (eGFR+ACR) CKD Add-on (eGFR+ACR)
ACR, per 8 fold 1.28 (1.21, 1.34) ACR, per 8 fold 1.27 (1.21, 1.33)
CKD Add-on (eGFR+dipstick) CKD Add-on (eGFR+dipstick)
Trace 1.30 (1.22, 1.39) Trace 1.29 (1.20, 1.37)
+ 1.51 (1.37, 1.66) + 1.47 (1.34, 1.62)
++ or more 1.61 (1.50, 1.73) ++ or more 1.52 (1.42, 1.64)
Open in a new tab
*

Age 30+, all population including diabetes and no diabetes (in the context of SCORE2)

**

Age 60+, all population including diabetes and no diabetes (in the context of SCORE2-OP)

We confirmed the improvement in c-statistics with both the CKD Add-on (eGFR only) and the CKD Add-on (eGFR + ACR) in the development datasets in the context of both SCORE2 and SCORE2-OP (Web Table 4). For example, in the study population aged ≥30 years, the CKD Add-on (eGFR only) and the CKD Add-on (eGFR + ACR) for SCORE2 improved c-statistic by 0.004 (0.003-0.006) and 0.015 (0.011-0.019), respectively. Similarly, the CKD Add-on (eGFR only) and the CKD Add-on (eGFR + ACR) for SCORE2-OP demonstrated c-statistic improvement (0.008 [0.006-0.010] and 0.022 [0.016-0.027], respectively) in the study population aged ≥60 years. We also observed positive overall NRIs in all comparisons with the CKD Add-on (eGFR only) and the CKD Add-on (eGFR + ACR) (Web Table 3). The CKD Add-on (eGFR + dipstick) also improved risk prediction. Results across individual datasets are shown in Web Tables 5 and 6 (CKD Add-on [eGFR only]) and 7 and 8 (CKD Add-on [eGFR + ACR]).

Validation of CKD Add-ons in the Validation Datasets

Both the CKD Add-on (eGFR only) and the CKD Add-on (eGFR + ACR) improved c-statistics in the target populations for SCORE2 and SCORE2-OP in the validation datasets (Table 3). In the study population aged 40-69 years without diabetes, the CKD Add-on (eGFR only) and the CKD Add-on (eGFR + ACR) for SCORE2 improved c-statistic by 0.006 (0.004-0.008) and 0.016 (0.010-0.023), respectively. The corresponding estimates of the CKD Add-on (eGFR only) and the CKD Add-on (eGFR + ACR) for SCORE2-OP were 0.012 (0.09, 0.015) and 0.024 (0.014, 0.035) in the study population aged 70 years or older without diabetes. Overall NRI was also significantly positive in all comparisons (e.g., 0.039 [0.018-0.059] with the CKD Add-on [eGFR + ACR] for SCORE2). The CKD Add-on (eGFR + dipstick) also improved the risk prediction (Table 3). The results were largely consistent when we focused on individuals at high risk of CVD, as defined in the ESC 2021 CVD prevention guideline3 and noted above (Web Table 9). The improvement of risk prediction was generally more evident when we included individuals with diabetes (Web Table 10) as well as when we removed the two clinical trials in individuals with diabetes from the analyses (Web Table 11). The vast majority of individual studies demonstrated improvement in c-statistic and positive NRIs with the CKD Add-on (eGFR only) (Web Table 12 and 13) and the CKD Add-on (eGFR + ACR) (Web Table 14 and 15). When we focused on general population cohorts, the results were largely consistent (Web Table 16). In CPRD, the application of the CKD Add-on (eGFR only) or the CKD Add-on (eGFR + ACR) did not alter the calibration of SCORE2 and SCORE2-OP much (Web Figure 1).

Table 3.

C-statistics and NRI with the CKD Add-ons in the SCORE2 and SCORE2-OP populations from the validation datasets

CKD Add-on (eGFR only) CKD Add-on (eGFR+ACR) CKD Add-on (eGFR+dipstick)
Overall SCORE2 in age 40-69, non-diabetics population
N 2817487 510622 684170
Base C-statistic (IQI) 0.686 (0.658, 0.719) 0.634 (0.604, 0.697) 0.688 (0.671, 0.715)
ΔC-statistic (95% CI) 0.006 (0.004, 0.008) 0.016 (0.010, 0.023) 0.019 (0.013, 0.025)
Category NRI (95% CI) Overall 0.030 (0.023, 0.037) 0.039 (0.018, 0.059) 0.095 (0.071, 0.120)
Event 0.050 (0.039, 0.060) 0.104 (0.069, 0.139) 0.124 (0.093, 0.154)
Non-event −0.012 (−0.014, −0.010) −0.041 (−0.053, −0.029) −0.027 (−0.034, −0.021)
Overall SCORE2-OP in age 70+, non-diabetics population
N 556887 57696 121312
Base C-statistic (IQI) 0.641 (0.601, 0.656) 0.613 (0.568, 0.661) 0.640 (0.626, 0.670)
ΔC-statistic (95% CI) 0.012 (0.009, 0.015) 0.024 (0.014, 0.035) 0.024 (0.017, 0.031)
Category NRI (95% CI) Overall 0.033 (0.024, 0.042) 0.046 (0.019, 0.074) 0.068 (0.044, 0.093)
Event 0.088 (0.065, 0.111) 0.150 (0.101, 0.200) 0.214 (0.165, 0.262)
Non-event −0.044 (−0.057, −0.032) −0.077 (−0.100, −0.055) −0.146 (−0.191, −0.100)
Open in a new tab

C-statistic was calculated within each gender group, no comparison between men and women

Risk category was defined as low/moderate risk (<2.5% for age <50, <5% for age 50-69 and <7.5% for age 70+), high risk (2.5-7.5% for age <50, 5-10% for age 50-69 and 7.5-15% for age 70+), very high risk (>7.5% for age <50, >10% for age 50-69 and >15% for age 70+).

Implications of CKD Add-ons

The median predicted risk ratio (i.e., with a CKD Add-on over without a CKD Add-on) across the validation datasets by different stages of CKD is shown in Figure 1. In the study population aged 40-69 without diabetes, the median predicted risk ratio was ~2.8 in severe CKD (cross-categories of eGFR and ACR in red in Figure 1), ~1.7 in moderate CKD (cross-categories in orange), and ~1.3 in mild CKD (cross-categories in yellow). The corresponding ratios were ~1.6, ~1.3 and ~1.1 in the study population aged ≥70 years without diabetes. We observed largely similar patterns for the CKD Add-on with dipstick (Web Figure 2). The results were similar in the study population including diabetes (Web Figure 3). Figure 2 demonstrates the extent to which the CKD Add-on (eGFR + ACR) influences predicted risk based on SCORE2 and SCORE2-OP in a few hypothetical scenarios (details in Appendix 1).

Figure 1. CKD staging and risk ratio of the CKD Add-on (eGFR+ACR) in the SCORE2 and SCORE2-OP populations from the validation datasets.

Figure 1.

Open in a new tab

Figure 2. The CKD Add-on (eGFR+ACR) impact on predicted risk based on SCORE2 and SCORE2-OP in 4 hypothetical scenarios.

Figure 2.

Open in a new tab

CVD risk classification was defined as low/moderate risk (<2.5% for age <50, <5% for age 50–69 and <7.5% for age 70+), high risk (2.5–7.5% for age <50, 5–10% for age 50–69 and 7.5–15% for age 70+), very high risk (>7.5% for age <50, >10% for age 50–69 and >15% for age 70+).

In 13 European datasets in CKD-PC including 57,485 participants with CKD, according to the approach in the 2021 ESC CVD prevention guideline (i.e., qualitative classification of severe and moderate CKD to very-high and high CVD risk and SCORE2 or SCORE2-OP in mild CKD), the proportion of individuals in the CVD risk of low/moderate, high, and very-high was 40.9%, 38.0%, and 21.2%, respectively. The corresponding proportion was 44.2%, 35.5%, and 20.3% when using a CKD Add-on. Compared to the approach in the 2021 ESC guideline, the new approach of augmenting SCORE2/SCORE2-OP with a CKD Add-on in this CKD population in Europe resulted in 13.8% (4524 out of 32,703) of the individuals reclassified upward to a higher CVD risk group and 14.6% (4788 out of 32,703) downward to a lower risk group, with overall positive NRI in the study populations aged 40-69 years (0.100 [0.062-0.138]) and ≥70 years (0.063 [0.014-0.112]) (Web Table 17).

Discussion

Using data from >9 million individuals from 68 datasets, we have developed and validated CKD Add-ons for SCORE2 and SCORE2-OP, the latest risk algorithms designed for primary CVD prevention in Europe.8 The improvement of risk prediction was generally greater with the CKD Add-on (eGFR + ACR) than the CKD Add-on (eGFR only). For example, in the target population of SCORE2 (age 40-69 years without diabetes) in the validation datasets, increases in c-statistics were 0.017 (95%CI 0.011-0.023) vs. 0.007 (0.005-0.008), respectively. NRI also supported the risk prediction improvement with either CKD Add-on. The improvement in risk prediction with the CKD Add-on was confirmed when we used dipstick proteinuria instead of ACR, included populations with diabetes, and focused on the high CVD risk group.

It is not easy to appreciate clinical values of specific risk prediction models from changes in c-statistics or NRI, and thus we have comprehensively evaluated other matrices such as a ratio of the predicted risk after an Add-on to the originally predicted risk, which demonstrated the impact of accounting (or not accounting) for the CKD measures. For example, in the target population of SCORE2, the median ratio in our validation datasets was ~1.7 in moderate CKD (e.g., eGFR 45-59 ml/min/1.73m2 plus ACR 30-299 mg/g) and ~2.8 in severe CKD (e.g., eGFR 30-44 ml/min/1.73m2 plus ACR 300+ mg/g). The corresponding ratios were slightly smaller in the targeted population for SCORE2-OP, ~1.3 and ~1.6, respectively. Importantly, in both target populations, the ratio was ~1 in individuals without CKD, confirming that those without CKD can simply rely on SCORE2 or SCORE2-OP. Of note, in CKD populations from 13 European cohorts, SCORE2 or SCORE2-OP with a CKD Add-on demonstrated a better risk classification than the quantitative approach proposed in the ESC 2021 CVD prevention guideline.

The discussion of the value of a novel predictor intrinsically includes the concept of whether that predictor should be newly measured or not. However, the situation of CKD measures is quite different in this regard since the assessment of eGFR and albuminuria is already recommended in several clinical scenarios. In fact, in the US, serum creatinine is measured ~300 million times annually.18 Likewise, the evaluation of albuminuria is recommended in patients with diabetes, hypertension, and reduced eGFR. Thus, in many individuals, the data on these CKD measures are readily available, and their omission is a critical missed opportunity to further personalize risk prediction and prevention approaches of CVD. Therefore, our CKD Add-ons would provide a validated means for clinicians and patients to incorporate existing CKD measures into SCORE2 algorithms and personalize CVD preventive therapies.

A few recent studies have shown that measures of albuminuria are less likely to be assessed compared to eGFR even when it is clinically indicated (e.g., patients with diabetes or hypertension). For example, in a US clinical database study, eGFR was measured at least once in a 1-year period among most patients with diabetes, whereas only half of them had measures of albuminuria.19 Our data further support the importance of taking into account albuminuria for CVD risk assessment. Importantly, the present study has validated a CKD Add-on using dipstick proteinuria as well for improving risk prediction of CVD, which adds to the applicability of our findings.

Several limitations of the present study should be acknowledged. The assessment of eGFR, albuminuria, and traditional CVD predictors and the ascertainment of CVD events were not necessarily standardized across all the cohorts. In addition, the data availability of albuminuria in clinical database cohorts is limited to a subsample, reflecting clinical indications. However, the overall consistent results across most of the cohorts, with diverse demographic and clinical characteristics, support the robustness of our study. Also, although we included 13 datasets from Europe, all are from low- or moderate-risk regions. Also, we have not included information on primary causes of CKD.

In conclusion, our CKD Add-ons improved CVD risk prediction according to SCORE2 and SCORE2-OP. This approach will help clinicians and patients refine risk prediction and further personalize preventive therapies for CVD when information on the CKD measures is available and indicates CKD.

Supplementary Material

Supplementary Materials

Funding:

The CKD Prognosis Consortium (CKD-PC) Data Coordinating Center is funded in part by a program grant from the US National Kidney Foundation (NKF funding sources include Boehringer Ingelheim) and the National Institute of Diabetes and Digestive and Kidney Diseases (R01DK100446). A variety of sources have supported enrollment and data collection including laboratory measurements, and follow-up in the collaborating cohorts of the CKD-PC (eAppendix 3). These funding sources include government agencies such as national institutes of health and medical research councils as well as foundations and industry sponsors. The funders of the study had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. In addition, the funders had no right to veto publication or to control the decision regarding to which journal the paper would be submitted.

Acknowledgements

CKD-PC investigators/collaborators (cohort acronyms/abbreviations are listed in eAppendix 2 in the Supplement:

ADVANCE: John Chalmers, Mark Woodward; Aichi: Hiroshi Yatsuya, Koji Tamakoshi, Yuanying Li, Yoshihisa Hirakawa; ARIC: Josef Coresh, Kunihiro Matsushita, Jung-Im Shin, Junichi Ishigami; AusDiab: Kevan Polkinghorne, Steven Chadban, Robert Atkins; BIS: Elke Schaeffner, Natalie Ebert; CARDIA: Donald Lloyd-Jones, Orlando Gutierrez; China NS: Luxia Zhang, Minghui Zhao, Fang Wang, Bixia Gao, Jinwei Wang; CIRCS: Kazumasa Yamagishi, Isao Muraki, Yuji Shimizu, Hiroyasu Iso; COBRA: Tazeen Jafar, Imtiaz Jehan, Neil Poulter, Nish Chaturvedi; CRIC: Jiang He, Wei Yang, Matthew Weir, Stephanie Toth-Manikowski, Christopher Jepson; ESTHER: Hermann Brenner, Dietrich Rothenbacher, Ben Schöttker, Bernd Holleczek; Framingham: Daniel Levy, Shih-Jen Hwang; GCKD: Markus P Schneider, Anna Köttgen, Heike Meiselbach, Kai-Uwe Eckardt; Geisinger: Alex R. Chang, Gurmukteshwar Singh, Jamie A Green, H. Lester Kirchner; Gubbio: Massimo Cirillo; Hong Kong CKD: Angela Yee-Moon Wang, Hoi Ching Cheung, Hailey Yee Tsui, Victoria Ngai; IPHS: Fujiko Irie, Toshimi Sairenchi; JHS: Adolfo Correa, Casey M. Rebholz, Bessie Young, L. Ebony Boulware, April Carson; LCC: Nigel Brunskill, Laura Gray, Rupert Major, James Medcalf; Maccabi: Varda Shalev, Gabriel Chodick; MESA: Michael Shlipak; Mt Sinai BioMe: Girish N Nadkarni, Erwin P Bottinger, Ruth JF Loos, Stephen B Ellis; NEFRONA: José M Valdivielso, Marcelino Bermúdez-López, Milica Bozic, Serafí Cambray; NHANES: Yingying Sang; NIPPON DATA90: Hirotsugu Ueshima, Tomonori Okamura, Katsuyuki Miura; Ohasama: Takayoshi Ohkubo, Hirohito Metoki, Michihiro Satoh, Masahiro Kikuya; OLDW: John Cuddeback, Elizabeth Ciemins, Emily Carbonara, Stephan Dunning; PREVEND: Ron T. Gansevoort, Lyane M Kieneker, Stephan JL Bakker, Hans L Hillege, Pim van der Harst; Rancho Bernardo: Simerjot K. Jassal, Jacklyn Bergstrom, Joachim Ix; RCAV: Csaba P. Kovesdy, Keiichi Sumida, Miklos Z. Molnar, Praveen Potukuchi; REGARDS: Orlando M Gutierrez, Paul Muntner, David Warnock; RENAAL: Dick de Zeeuw, Michelle J. Pena, Hiddo J. L. Heerspink; SKS: Philip A Kalra, Rajkumar Chinnadurai, James Tollitt, Darren Green; SMART: Frank Visseren, Joep van der Leeuw; Taiwan MJ: Chi-Pang Wen, Min-Kuang Tsai; Takahata: Takamasa Kayama, Tsuneo Konta; TLGS: Mohammadhassan Mirbolouk, Fereidoun Azizi, Farzad Hadaegh, Maryam Tohidi; Tromso: Marit Dahl Solbu, Bjørn Odvar Eriksen, Trond Geir Jenssen, Anne Elise Eggen; UK Biobank: Christoph Nowak, Johan Ärnlöv; ULSAM: Lars Lannfelt, Anders Larsson, Johan Ärnlöv; ZODIAC: Henk J.G. Bilo, Gijs W.D. Landman, Kornelis J.J. van Hateren, Nanne Kleefstra

CKD-PC Steering Committee: Josef Coresh (Chair), Shoshana H Ballew, Alex R. Chang, Ron T Gansevoort, Morgan E. Grams, Orlando Gutierrez, Tsuneo Konta, Anna Köttgen, Andrew S Levey, Kunihiro Matsushita, Kevan Polkinghorne, Elke Schäffner, Mark Woodward, Luxia Zhang

CKD-PC Data Coordinating Center: Shoshana H Ballew (Assistant Project Director), Jingsha Chen (Programmer), Josef Coresh (Principal Investigator), Morgan E Grams (Director of Nephrology Initiatives), Kunihiro Matsushita (Director), Yingying Sang (Lead Programmer), Aditya Surapaneni (Programmer), Mark Woodward (Senior Statistician)

Footnotes

Declaration of Interests: All authors will complete the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author).

Contributor Information

Kunihiro Matsushita, Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD.

Stephen Kaptoge, Department of Public Health and Primary Care, University of Cambridge, Cambridge, United Kingdom.

Steven HJ Hageman, Department of Vascular Medicine, University Medical Center Utrecht, Utrecht, The Netherlands.

Yingying Sang, Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD.

Shoshana H Ballew, Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD.

Morgan E Grams, Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD.

Aditya Surapaneni, Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD.

Luanluan Sun, Department of Public Health and Primary Care, University of Cambridge, Cambridge, United Kingdom.

Johan Arnlov, Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Stockholm, Sweden.

Milica Bozic, Vascular & Renal Translational Research Group, IRBLleida, Spain and Spanish Research Network for Renal Diseases (RedInRen. ISCIII) , Lleida , Spain.

Hermann Brenner, Division of Clinical Epidemiology and Aging Research, German Cancer Research Center (DKFZ) and Network Aging Research, University of Heidelberg, Heidelberg, Germany.

Nigel J Brunskill, John Walls Renal Unit, Leicester General Hospital, University Hospitals of Leicester NHS Trust, Leicester, United Kingdom, Department of Cardiovascular Sciences, University of Leicester, Leicester, United Kingdom.

Alex R Chang, Department of Nephrology and Kidney Health Research Institute, Geisinger Medical Center, Danville, Pennsylvania.

Rajkumar Chinnadurai, Department of Renal Medicine, Salford Care Organisation, Northern Care Alliance NHS Foundation Trust, Salford, UK.

Massimo Cirillo, Department of Public Health, University of Naples “Federico II”, Italy.

Adolfo Correa, University of Mississippi Medical Center, Jackson.

Natalie Ebert, Institute of Public Health, Charité - Universitätsmedizin Berlin, Berlin, Germany.

Kai-Uwe Eckardt, Department of Nephrology and Hypertension, Friedrich-Alexander Universität Erlangen-Nürnberg, Erlangen, Germany; Department of Nephrology and Medical Intensive Care, Charité – Universitätsmedizin Berlin, Berlin, Germany.

Ron T Gansevoort, Department of Nephrology, University Medical Center Groningen, University of Groningen, Groningen, the Netherlands.

Orlando Gutierrez, Departments of Epidemiology and Medicine, University of Alabama at Birmingham, Birmingham, AL.

Farzad Hadaegh, Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran.

Jiang He, Department of Epidemiology, Tulane University School of Public Health and Tropical Medicine, New Orleans, LA.

Shih-Jen Hwang, National Heart, Lung, and Blood Institute, Framingham, Massachusetts.

Tazeen H Jafar, Program in Health Services and Systems Research, Duke-National University of Singapore Medical School, Singapore, Department of Medicine, Aga Khan University, Karachi, Pakistan, and Duke Global Health Institute, Durham, Duke University, North Carolina.

Simerjot K Jassal, Division of General Internal Medicine, University of California, San Diego and VA San Diego Healthcare, San Diego, California.

Takamasa Kayama, Global Center of Excellence, Yamagata University Faculty of Medicine, Yamagata, Japan; Department of Public Health Medicine, Faculty of Medicine, and Health Services Research and Development Center, University of Tsukuba, Japan.

Csaba P Kovesdy, Medicine-Nephrology, Memphis Veterans Affairs Medical Center and University of Tennessee Health Science Center, Memphis, Tennessee.

Gijs W Landman, Gelre hospital location, Apeldoorn, The Netherlands.

Andrew S Levey, Division of Nephrology, Tufts Medical Center, Boston, Massachusetts.

Donald M Lloyd-Jones, Department of Preventive Medicine, Northwestern University, Chicago, Illinois.

Rupert W Major, John Walls Renal Unit, Leicester General Hospital, University Hospitals of Leicester NHS Trust, Leicester, United Kingdom, Department of Cardiovascular Sciences, University of Leicester, Leicester, United Kingdom.

Katsuyuki Miura, NCD Epidemiology Research Center, Shiga University of Medical Science, Shiga, Japan.

Paul Muntner, Department of Epidemiology, University of Alabama at Birmingham, Birmingham, Alabama.

Girish N Nadkarni, Department of Medicine, Division of Nephrology, Icahn School of Medicine at Mount Sinai, New York, New York.

Christoph Nowak, Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Stockholm, Sweden.

Takayoshi Ohkubo, Department of Hygiene and Public Health, Teikyo University School of Medicine, Tokyo, Japan.

Michelle J Pena, Department of Clinical Pharmacy and Pharmacology, University Medical Center Groningen, University of Groningen, Groningen, the Netherlands.

Kevan R Polkinghorne, Monash University, Clayton, Australia.

Toshimi Sairenchi, Medical Science of Nursing, Dokkyo Medical University School of Nursing, Mibu, Japan.

Elke Schaeffner, Institute of Public Health, Charité - Universitätsmedizin Berlin, Berlin, Germany.

Markus P Schneider, Department of Nephrology and Hypertension, Friedrich-Alexander Universität Erlangen-Nürnberg, Erlangen, Germany.

Varda Shalev, Institute for Health and Research and Innovation, Maccabi Healthcare Services and Tel Aviv University, Tel Aviv, Israel.

Michael G Shlipak, Kidney Health Research Collaborative, University of California, San Francisco, and San Francisco VA Healthcare System, San Francisco.

Marit D Solbu, Section of Nephrology, University Hospital of North Norway, Tromsø, Norway and UiT The Arctic University of Norway, Tromsø, Norway.

Nikita Stempniewicz, AMGA (American Medical Group Association), Alexandria, Virginia and OptumLabs Visiting Fellow.

James Tollitt, Department of Renal Medicine, Salford Care Organisation, Northern Care Alliance NHS Foundation Trust, Salford, UK; Renal Department, University of Manchester, Oxford Road, Manchester, United Kingdom.

José M Valdivielso, Vascular & Renal Translational Research Group, IRBLleida, Spain and Spanish Research Network for Renal Diseases (RedInRen. ISCIII) , Lleida , Spain.

Joep van der Leeuw, Department of Vascular Medicine, University Medical Center Utrecht, Utrecht, The Netherlands.

Angela Yee-Moon Wang, Department of Medicine, Queen Mary Hospital, The University of Hong Kong, Hong Kong.

Chi-Pang Wen, China Medical University Hospital, Taichung, Taiwan.

Mark Woodward, Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD.

Kazumasa Yamagishi, George Institute for Global Health, Australia, and George Institute for Global Health, Imperial College, London, United Kingdom.

Hiroshi Yatsuya, Department of Public Health, Fujita Health University School of Medicine, Aichi, Japan and Department of Public Health and Health Systems, Nagoya University Graduate School of Medicine, Aichi, Japan.

Luxia Zhang, Peking University First Hospital and Peking University, Beijing, China.

Jannick AN Dorresteijn, Department of Vascular Medicine, University Medical Center Utrecht, Utrecht, The Netherlands.

Emanuele Di Angelantonio, Department of Public Health and Primary Care, University of Cambridge, Cambridge, United Kingdom; Health Data Science Centre, Human Technopole, Milan, Italy.

Frank LJ Visseren, Department of Vascular Medicine, University Medical Center Utrecht, Utrecht, The Netherlands.

Lisa Pennells, Department of Public Health and Primary Care, University of Cambridge, Cambridge, United Kingdom.

Josef Coresh, Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD.

Data Availability Statement:

Under agreement with the participating cohorts, CKD-PC cannot share individual data with third parties. Inquiries regarding specific analyses should be made to ckdpc@jhmi.edu. Investigators may approach the original cohorts regarding their own policies for data sharing (e.g., https://sites.cscc.unc.edu/aric/distribution-agreements for the Atherosclerosis Risk in Communities Study).

References

  • 1.Gansevoort RT, Correa-Rotter R, Hemmelgarn BR, Jafar TH, Heerspink HJL, Mann JF, Matsushita K, Wen CP. Chronic kidney disease and cardiovascular risk: epidemiology, mechanisms, and prevention. The Lancet 2013;382:339–352. [DOI] [PubMed] [Google Scholar]
  • 2.Matsushita K, Coresh J, Sang Y, Chalmers J, Fox C, Guallar E, Jafar T, Jassal SK, Landman GWD, Muntner P, Roderick P, Sairenchi T, Schöttker B, Shankar A, Shlipak M, Tonelli M, Townend J, Zuilen Av, Yamagishi K, Yamashita K, Gansevoort R, Sarnak M, Warnock DG, Woodward M, Ärnlöv J. Estimated glomerular filtration rate and albuminuria for prediction of cardiovascular outcomes: a collaborative meta-analysis. Lancet Diabetes-Endocrinol 2015;3:514–525. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Visseren FLJ, Mach F, Smulders YM, Carballo D, Koskinas KC, Bäck M, Benetos A, Biffi A, Boavida JM, Capodanno D, Cosyns B, Crawford C, Davos CH, Desormais I, Di Angelantonio E, Franco OH, Halvorsen S, Hobbs FDR, Hollander M, Jankowska EA, Michal M, Sacco S, Sattar N, Tokgozoglu L, Tonstad S, Tsioufis KP, van Dis I, van Gelder IC, Wanner C, Williams B. 2021 ESC Guidelines on cardiovascular disease prevention in clinical practice. Eur Heart J 2021. [Google Scholar]
  • 4.Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group. KDIGO 2012 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. Kidney International. Supplement 2013;3:1–150. [Google Scholar]
  • 5.Matsushita K, Jassal SK, Sang Y, Ballew SH, Grams ME, Surapaneni A, Arnlov J, Bansal N, Bozic M, Brenner H, Brunskill NJ, Chang AR, Chinnadurai R, Cirillo M, Correa A, Ebert N, Eckardt KU, Gansevoort RT, Gutierrez O, Hadaegh F, He J, Hwang SJ, Jafar TH, Kayama T, Kovesdy CP, Landman GW, Levey AS, Lloyd-Jones DM, Major RW, Miura K, Muntner P, Nadkarni GN, Naimark DM, Nowak C, Ohkubo T, Pena MJ, Polkinghorne KR, Sabanayagam C, Sairenchi T, Schneider MP, Shalev V, Shlipak M, Solbu MD, Stempniewicz N, Tollitt J, Valdivielso JM, van der Leeuw J, Wang AY, Wen CP, Woodward M, Yamagishi K, Yatsuya H, Zhang L, Schaeffner E, Coresh J. Incorporating kidney disease measures into cardiovascular risk prediction: Development and validation in 9 million adults from 72 datasets. EClinicalMedicine 2020;27:100552. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Grundy SM, Stone NJ, Bailey AL, Beam C, Birtcher KK, Blumenthal RS, Braun LT, de Ferranti S, Faiella-Tommasino J, Forman DE, Goldberg R, Heidenreich PA, Hlatky MA, Jones DW, Lloyd-Jones D, Lopez-Pajares N, Ndumele CE, Orringer CE, Peralta CA, Saseen JJ, Smith SC Jr., Sperling L, Virani SS, Yeboah J. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation 2019;139:e1082–e1143. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Piepoli MF, Hoes AW, Agewall S, Albus C, Brotons C, Catapano AL, Cooney MT, Corra U, Cosyns B, Deaton C, Graham I, Hall MS, Hobbs FDR, Lochen ML, Lollgen H, Marques-Vidal P, Perk J, Prescott E, Redon J, Richter DJ, Sattar N, Smulders Y, Tiberi M, van der Worp HB, van Dis I, Verschuren WMM, Binno S, Group ESCSD. 2016 European Guidelines on cardiovascular disease prevention in clinical practice: The Sixth Joint Task Force of the European Society of Cardiology and Other Societies on Cardiovascular Disease Prevention in Clinical Practice (constituted by representatives of 10 societies and by invited experts)Developed with the special contribution of the European Association for Cardiovascular Prevention & Rehabilitation (EACPR). Eur Heart J 2016;37:2315–2381. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.SCORE2 Working Group and ESC Cardiovascular Risk Collaboration. SCORE2 risk prediction algorithms: new models to estimate 10-year risk of cardiovascular disease in Europe. Eur Heart J 2021;42:2439–2454. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group. KDIGO 2012 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. Kidney inter., Suppl 2013;3:1–150. [Google Scholar]
  • 10.Inker LA, Eneanya ND, Coresh J, Tighiouart H, Wang D, Sang Y, Crews DC, Doria A, Estrella MM, Froissart M, Grams ME, Greene T, Grubb A, Gudnason V, Gutierrez OM, Kalil R, Karger AB, Mauer M, Navis G, Nelson RG, Poggio ED, Rodby R, Rossing P, Rule AD, Selvin E, Seegmiller JC, Shlipak MG, Torres VE, Yang W, Ballew SH, Couture SJ, Powe NR, Levey AS, Chronic Kidney Disease Epidemiology C. New Creatinine- and Cystatin C-Based Equations to Estimate GFR without Race. N Engl J Med 2021;385:1737–1749. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Sumida K, Nadkarni GN, Grams ME, Sang Y, Ballew SH, Coresh J, Matsushita K, Surapaneni A, Brunskill N, Chadban SJ, Chang AR, Cirillo M, Daratha KB, Gansevoort RT, Garg AX, Iacoviello L, Kayama T, Konta T, Kovesdy CP, Lash J, Lee BJ, Major RW, Metzger M, Miura K, Naimark DMJ, Nelson RG, Sawhney S, Stempniewicz N, Tang M, Townsend RR, Traynor JP, Valdivielso JM, Wetzels J, Polkinghorne KR, Heerspink HJL. Conversion of Urine Protein-Creatinine Ratio or Urine Dipstick Protein to Urine Albumin-Creatinine Ratio for Use in Chronic Kidney Disease Screening and Prognosis : An Individual Participant-Based Meta-analysis. Ann Intern Med 2020;173:426–435. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Matsushita K, Coresh J, Sang Y, Chalmers J, Fox C, Guallar E, Jafar T, Jassal SK, Landman GW, Muntner P, Roderick P, Sairenchi T, Schottker B, Shankar A, Shlipak M, Tonelli M, Townend J, van Zuilen A, Yamagishi K, Yamashita K, Gansevoort R, Sarnak M, Warnock DG, Woodward M, Arnlov J, Consortium CKDP. Estimated glomerular filtration rate and albuminuria for prediction of cardiovascular outcomes: a collaborative meta-analysis of individual participant data. Lancet Diabetes Endocrinol 2015;3:514–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Matsushita K, van der Velde M, Astor BC, Woodward M, Levey AS, de Jong PE, Coresh J, Gansevoort RT. Association of estimated glomerular filtration rate and albuminuria with all-cause and cardiovascular mortality in general population cohorts: a collaborative meta-analysis. Lancet 2010;375:2073–2081. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Matsushita K, Sang Y, Chen J, Ballew SH, Shlipak M, Coresh J, Peralta CA, Woodward M. Novel "Predictor Patch" Method for Adding Predictors Using Estimates From Outside Datasets- A Proof-of-Concept Study Adding Kidney Measures to Cardiovascular Mortality Prediction. Circ J 2019;83:1876–1882. [DOI] [PubMed] [Google Scholar]
  • 15.Pencina MJ, D'Agostino RB. Overall C as a measure of discrimination in survival analysis: model specific population value and confidence interval estimation. Stat Med 2004;23:2109–2123. [DOI] [PubMed] [Google Scholar]
  • 16.Pencina MJ, D'Agostino RB Sr., Steyerberg EW. Extensions of net reclassification improvement calculations to measure usefulness of new biomarkers. Stat Med 2011;30:11–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Collins G, Reitsma J, Altman D, Moons K. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): The TRIPOD statement. https://www.equator-network.org/reporting-guidelines/tripod-statement/. [DOI] [PubMed] [Google Scholar]
  • 18.Inker LA, Levey AS. Pro: Estimating GFR using the chronic kidney disease epidemiology collaboration (CKD-EPI) 2009 creatinine equation: the time for change is now. Nephrol Dial Transplant 2013;28:1390–1396. [DOI] [PubMed] [Google Scholar]
  • 19.Stempniewicz N, Vassalotti JA, Cuddeback JK, Ciemins E, Storfer-Isser A, Sang Y, Matsushita K, Ballew SH, Chang AR, Levey AS, Bailey RA, Fishman J, Coresh J. Chronic Kidney Disease Testing Among Primary Care Patients With Type 2 Diabetes Across 24 U.S. Health Care Organizations. Diabetes Care 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Supplementary Materials
NIHMS1844083-supplement-Supplementary_Materials.docx (247.4KB, docx)

Data Availability Statement

Under agreement with the participating cohorts, CKD-PC cannot share individual data with third parties. Inquiries regarding specific analyses should be made to ckdpc@jhmi.edu. Investigators may approach the original cohorts regarding their own policies for data sharing (e.g., https://sites.cscc.unc.edu/aric/distribution-agreements for the Atherosclerosis Risk in Communities Study).

RESOURCES