Kidney Failure Risk Equation Research Articles

Validation of the kidney failure risk equation in predicting the risk of progression to kidney failure in a multi-ethnic Singapore chronic kidney disease cohort.

The Kidney Failure Risk Equation (KFRE) was developed to predict the risk of progression to end-stage kidney disease (ESKD). Although the KFRE has been validated in multinational cohorts, the Southeast Asian population was under-represented. This study aimed to validate the KFRE in a multi-ethnic Singapore chronic kidney disease (CKD) cohort. Stage 3-5 CKD patients referred to the renal medicine department at Singapore General Hospital in 2009 were included. The primary outcome (time to ESKD) was traced until 30 June 2017. The eight- and four-variable KFRE (non-North America) models using age, gender, estimated glomerular filtration rate, urine albumin-creatinine ratio, serum albumin, phosphate, bicarbonate and calcium were validated in our cohort. Cox regression, likelihood ratio (Χ2), adequacy index, Harrell's C-index and calibration curves were calculated to assess the predictive performance, discrimination and calibration of these models on the cohort. A total of 1,128 patients were included. During the study period, 252 (22.3%) patients reached ESKD at a median time to ESKD of 84.8 (range 0.1-104.7) months. Both the eight- and four-variable KFRE models showed excellent predictive performance and discrimination (eight-variable: C-index 0.872, 95% confidence interval [CI] 0.850-0.894, adequacy index 97.3%; four-variable: C-index 0.874, 95% CI 0.852-0.896, adequacy index 97.9%). There was no incremental improvement in the prediction ability of the eight-variable model over the four-variable model in this cohort. The KFRE was validated in a multi-ethnic Singapore CKD cohort. This risk score may help to identify patients requiring early renal care.

Impact of Kidney Failure Risk Prediction Clinical Decision Support on Monitoring and Referral in Primary Care Management of CKD: A Randomized Pragmatic Clinical Trial

Rationale & ObjectiveTo design and implement clinical decision support incorporating a validated risk prediction estimate of kidney failure in primary care clinics and to evaluate the impact on stage-appropriate monitoring and referral.Study DesignBlock-randomized, pragmatic clinical trial.Setting & ParticipantsTen primary care clinics in the greater Boston area. Patients with stage 3-5 chronic kidney disease (CKD) were included. Patients were randomized within each primary care physician panel through a block randomization approach. The trial occurred between December 4, 2015, and December 3, 2016.InterventionPoint-of-care noninterruptive clinical decision support that delivered the 5-year kidney failure risk equation as well as recommendations for stage-appropriate monitoring and referral to nephrology.OutcomesThe primary outcome was as follows: Urine and serum laboratory monitoring test findings measured at one timepoint 6 months after the initial primary care visit and analyzed only in patients who had not undergone the recommended monitoring test in the preceding 12 months. The secondary outcome was nephrology referral in patients with a calculated kidney failure risk equation value of >10% measured at one timepoint 6 months after the initial primary care visit.ResultsThe clinical decision support application requested and processed 569,533 Continuity of Care Documents during the study period. Of these, 41,842 (7.3%) documents led to a diagnosis of stage 3, 4, or 5 CKD by the clinical decision support application. A total of 5,590 patients with stage 3, 4, or 5 CKD were randomized and included in the study. The link to the clinical decision support application was clicked 122 times by 57 primary care physicians. There was no association between the clinical decision support intervention and the primary outcome. There was a small but statistically significant difference in nephrology referral, with a higher rate of referral in the control arm.LimitationsContamination within provider and clinic may have attenuated the impact of the intervention and may have biased the result toward null.ConclusionsThe noninterruptive design of the clinical decision support was selected to prevent cognitive overload; however, the design led to a very low rate of use and ultimately did not improve stage-appropriate monitoring.FundingResearch reported in this publication was supported by the National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health under award K23DK097187.Trial RegistrationClinicalTrials.gov Identifier: NCT02990897.

Machine learning to predict end stage kidney disease in chronic kidney disease

The purpose of this study was to assess the feasibility of machine learning (ML) in predicting the risk of end-stage kidney disease (ESKD) from patients with chronic kidney disease (CKD). Data were obtained from a longitudinal CKD cohort. Predictor variables included patients’ baseline characteristics and routine blood test results. The outcome of interest was the presence or absence of ESKD by the end of 5 years. Missing data were imputed using multiple imputation. Five ML algorithms, including logistic regression, naïve Bayes, random forest, decision tree, and K-nearest neighbors were trained and tested using fivefold cross-validation. The performance of each model was compared to that of the Kidney Failure Risk Equation (KFRE). The dataset contained 748 CKD patients recruited between April 2006 and March 2008, with the follow-up time of 6.3 ± 2.3 years. ESKD was observed in 70 patients (9.4%). Three ML models, including the logistic regression, naïve Bayes and random forest, showed equivalent predictability and greater sensitivity compared to the KFRE. The KFRE had the highest accuracy, specificity, and precision. This study showed the feasibility of ML in evaluating the prognosis of CKD based on easily accessible features. Three ML models with adequate performance and sensitivity scores suggest a potential use for patient screenings. Future studies include external validation and improving the models with additional predictor variables.

FC079: The Race-Free EGFR Equation, The Kidney Failure Risk Equation and Referrals to Specialist Renal Clinics

Abstract BACKGROUND AND AIMS The new race-free estimated glomerular filtration rate (eGFR) was developed in 2021. Recently in the UK in keeping with similar initiatives elsewhere, the kidney failure risk equation (KFRE) to predict the risk of kidney failure has been incorporated into clinical guidelines. Referral from primary care to a specialist renal clinic is recommended if eGFR falls to < 30 mL/min/1.73 m2 and/or if the 5-year KFRE is greater than 5%. We investigate the impact of using the race-free eGFR equation and KFRE on CKD diagnosis in primary care and potential referrals to the renal clinic. METHOD Primary care records for 79% of the population of Wales (UK) are held in the electronic health records repository Secure Anonymised Information Linkage Databank (SAIL). We studied serum creatinine values and urine albumin-creatinine ratios (uACRs) from 1 January 2013 to 31 December 2020. We calculated eGFR values using three equations: MDRD, CKD-EPI 2009 and (race-free) CKD-EPI 2021. Using the different equations, we compared the numbers of patients with incident eGFR <60 mL/min/1.73 m2 and incident eGFR < 30 mL/min/1.73 m2 (i.e. their eGFR fell from above to below these values for more than 3 months). For each year from 2013 to 2020, we identified the patients with prevalent eGFR 30–60 mL/min/1.73 m2 those with annual uACR testing and those who met referral criteria by A) eGFR decline and B) KFRE without eGFR decline. RESULTS There were 121 471 patients with prevalent CKD between 2013 and 2020. eGFR values were lowest using the MDRD equation (median 47.1 mL/min/1.73 m2 IQI 39.7–51.9) and highest with the CKD-EPI 2021 equation (median 50.0 mL/min/1.73 m2 IQI 41.6–55.3). Changing between these two equations would have led to a 17.6% reduction in incident eGFR < 60 mL/min/1.73 m2 and a 7.5% reduction in incident eGFR < 30 between 2013 and 2020 (Figure 1). The rate of annual uACR testing fell from 46.3% in 2013 to 25.3% in 2019 (Figure 2). eGFR and uACR testing were reduced further in 2020 during the COVID-19 pandemic. Patients without diabetes and older patients were the least likely to have had uACR testing at any time: for example, amongst those aged 60–70 years, 90.0% of those with diabetes had uACR testing at any time compared to 42.7% of those without diabetes; amongst those aged over 80 years, 79.1% of those with diabetes were tested compared to 32.7% of those without diabetes. In 2019 (the last year before the COVID-19 pandemic), 787/61 721 (1.3%) patients with CKD stage 3 met referral criteria by eGFR decline and an additional 587 (1.0%) by KFRE without eGFR decline. CONCLUSION Using the race-free eGFR equation will reduce diagnoses of incident eGFR < 30 warranting referral to specialist renal clinics. KFRE can be used to identify a significant number of patients at heightened risk of kidney failure, and these numbers may be higher if more uACR testing was performed. Annual uACR testing rates are low, especially in those without diabetes and in older adults. eGFR and uACR testing were markedly reduced during the COVID-19 pandemic in 2020 as most routine disease monitoring stopped. Expanding uACR testing in primary care (particularly in those without diabetes and in older adults) and using KFRE may improve the identification of individuals at risk of progressive kidney disease, but this is challenging during the COVID-19 pandemic.

MO483: Validation of The Kidney Failure Risk Equation in a Portuguese Cohort

Abstract BACKGROUND AND AIMS In chronic kidney disease (CKD) patients, the risk of kidney replacement therapy (KRT) is highly variable. In 2011, Tangri et al. developed the kidney failure risk equations (KFRE) to predict the 2- and 5-year probability of requiring kidney replacement therapy (KRT). The KFRE is an easily calculated 4-variable equation that has been extensively validated in multiple cohorts. The aim of this study was to validate this risk score in a Portuguese cohort. METHOD We conducted a retrospective analysis of CKD patients stage 3–5 referred for nephrology consult at Centro Hospitalar Universitário Lisboa Norte during the first 6 months of 2018. Age, gender, estimated glomerular filtration rate (eGFR) and albuminuria were assessed. The four-variable Kidney Failure Risk Equation (KFRE) calibrated for a non-North American population was calculated. Requirement for KRT was assessed in a 2-year follow-up. We assessed the Cox logistic regression method of the KFRE to predict KRT requirement, and the discriminatory ability was determined using the receiver operating characteristic (ROC) curve. A cut-off value was defined as that with the highest validity. RESULTS A total of 360 patients were included, and 54.4% were male. The mean age was 74.9 ± 12.2 years, serum creatinine was 1.97 ± 0.84 mg/dL, eGFR was 33.4 ± 12.13 mL/min/1.73 m2 and albuminuria was 571.1 ± 848.3 mg/g. The mean calculated risk score was 6.2 ± 11.2%. A total of 23 patients required KRT (6.4%) in the 2-year follow-up. The hazard ratio was 1.1 [95% confidence interval (95% CI) 1.06–1.12; P < .001] for the 2-year risk of KRT. The KFRE predicted progression to KRT requirement with an auROC of 0.903, (95% CI 0.86–0.95; P < 0.001), with a sensitivity of 91.3% and specificity of 71.8%. The optimal KFRE cut-off was >4.5% for 2-year nephrologist referral, with a hazard ratio of HR 26.7 [95% CI 6.15–116.3; P < .001] for 2-year risk of KRT. requirement. CONCLUSION We have independently externally validated the 2-year KFRE and shown that it has excellent discrimination. The KFRE should be incorporated into the clinical care of patients with CKD to improve patient-clinician dialogue and provide guidance on the timing of referral for nephrology evaluation and planning for dialysis access.

FC079: The Race-Free EGFR Equation, The Kidney Failure Risk Equation and Referrals to Specialist Renal Clinics

BACKGROUND AND AIMSThe new race-free estimated glomerular filtration rate (eGFR) was developed in 2021. Recently in the UK in keeping with similar initiatives elsewhere, the kidney failure risk equation (KFRE) to predict the risk of kidney failure has been incorporated into clinical guidelines. Referral from primary care to a specialist renal clinic is recommended if eGFR falls to < 30 mL/min/1.73 m2 and/or if the 5-year KFRE is greater than 5%. We investigate the impact of using the race-free eGFR equation and KFRE on CKD diagnosis in primary care and potential referrals to the renal clinic.METHODPrimary care records for 79% of the population of Wales (UK) are held in the electronic health records repository Secure Anonymised Information Linkage Databank (SAIL). We studied serum creatinine values and urine albumin-creatinine ratios (uACRs) from 1 January 2013 to 31 December 2020. We calculated eGFR values using three equations: MDRD, CKD-EPI 2009 and (race-free) CKD-EPI 2021. Using the different equations, we compared the numbers of patients with incident eGFR <60 mL/min/1.73 m2 and incident eGFR < 30 mL/min/1.73 m2 (i.e. their eGFR fell from above to below these values for more than 3 months). For each year from 2013 to 2020, we identified the patients with prevalent eGFR 30–60 mL/min/1.73 m2 those with annual uACR testing and those who met referral criteria by A) eGFR decline and B) KFRE without eGFR decline.RESULTSThere were 121 471 patients with prevalent CKD between 2013 and 2020. eGFR values were lowest using the MDRD equation (median 47.1 mL/min/1.73 m2 IQI 39.7–51.9) and highest with the CKD-EPI 2021 equation (median 50.0 mL/min/1.73 m2 IQI 41.6–55.3). Changing between these two equations would have led to a 17.6% reduction in incident eGFR < 60 mL/min/1.73 m2 and a 7.5% reduction in incident eGFR < 30 between 2013 and 2020 (Figure 1). The rate of annual uACR testing fell from 46.3% in 2013 to 25.3% in 2019 (Figure 2). eGFR and uACR testing were reduced further in 2020 during the COVID-19 pandemic. Patients without diabetes and older patients were the least likely to have had uACR testing at any time: for example, amongst those aged 60–70 years, 90.0% of those with diabetes had uACR testing at any time compared to 42.7% of those without diabetes; amongst those aged over 80 years, 79.1% of those with diabetes were tested compared to 32.7% of those without diabetes. In 2019 (the last year before the COVID-19 pandemic), 787/61 721 (1.3%) patients with CKD stage 3 met referral criteria by eGFR decline and an additional 587 (1.0%) by KFRE without eGFR decline.CONCLUSIONUsing the race-free eGFR equation will reduce diagnoses of incident eGFR < 30 warranting referral to specialist renal clinics. KFRE can be used to identify a significant number of patients at heightened risk of kidney failure, and these numbers may be higher if more uACR testing was performed. Annual uACR testing rates are low, especially in those without diabetes and in older adults. eGFR and uACR testing were markedly reduced during the COVID-19 pandemic in 2020 as most routine disease monitoring stopped. Expanding uACR testing in primary care (particularly in those without diabetes and in older adults) and using KFRE may improve the identification of individuals at risk of progressive kidney disease, but this is challenging during the COVID-19 pandemic.FIGURE 1:Incident CKD 2013–2020.FIGURE 1:CKD stage 3 monitoring and potential renal clinic referrals by year.

MO187: Healthcare Costs based on Risk of Progression in Patients with Chronic Kidney Disease

Abstract BACKGROUND AND AIMS Nephrologists follow patients with chronic kidney disease (CKD) stage G3 and G4 as a homogeneous group with the assumption that everyone had similar rates of progression with scheduled visits and lab investigations based on the stage of the disease. We now recognize that not all patients progress at similar rates to kidney failure and treatment and follow-up needs vary. The Kidney Failure Risk Equation (KFRE) identifies patients at different risks of progression to kidney failure (low, medium and high risk) in each stage of the disease. Previous studies had looked at resource utilization of patients based on the stage of the CKD. The purpose of our analysis was to examine resource utilization and associated costs based on the risk of progression by KFRE in the setting of a universal healthcare system. METHOD We conducted a retrospective cohort study of adults with CKD G3 and G4 enrolled in multidisciplinary CKD clinics in the province of Saskatchewan, Canada. Data was collected from January 2004 through December 2012 and patients were followed for 5 years. The predicted risk of kidney failure for each patient was calculated using the 8-variable KFRE. The equation used clinical and routine laboratory data, to stratify patients into three risk categories (low, medium and high risk) of progression. We compared the number and cost of hospital admissions, physician visits and prescription drugs by risk within G3 and G4. Negative binomial regression and generalized linear model were used to compare healthcare utilization and cost between the groups respectively (α = 0.05). RESULTS A total of 1003 adults with CKD G3 and G4 were included in the study. In patients with stage G3 CKD, 311 (59%), 150 (28%) and 68 (13%) were in low, medium and high-risk categories, respectively. Amongst patients with CKD stage G4, 275 (58%), 86 (18%) and 113 (24%) were in similar categories respectively. The cost of hospital admissions, physician visits and drug dispensations in stage G4 high risk in comparison to low risk over the 5-year study period was CAD $89 265 versus $48 374 (P = .008), $23 423 versus $11 231 (P &amp;lt; .001) and $21 853 versus $16 757 (P = .01), respectively. In stage G3, the cost of hospital admissions was CAD $55 944 versus $36 740 (P = 0.10), physician visits $13 414 versus $10 370 (P = .08) and prescription drugs $20 394 versus $14 902 (P = .02) in high-risk patients in comparison to low-risk patients (Figure 1). CONCLUSION In patients followed in multidisciplinary clinics with CKD stages G3 and G4, the cost of hospital admissions, physician visits and prescription drugs were higher in high-risk patients compared to patients in low-risk category. In our study, the KFRE, designed to predict the risk of progression to dialysis in patients with CKD, also assisted in identifying patients with higher health resource utilization and healthcare costs compared to those with lower health resource use. We additionally suggest that patients who are in medium and high-risk categories be followed in multidisciplinary clinics rather than individual physician offices to delay the trajectory of decline to kidney failure.

FC078: Impact of Removing Race from the CKD-EPI Equation: Analysis of 1.6 Million Swedish Adults

Abstract BACKGROUND AND AIMS The Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) recently developed a novel creatinine-based eGFR equation without a race coefficient. While American nephrology societies recommend using this novel equation, its implications are unknown. METHOD We included 1.6 million adult individuals with routine outpatient serum creatinine testing during 2007–2018 in Stockholm, Sweden. First, we calculated reclassification across KDIGO eGFR categories when changing from the 2009 to 2021 CKD-EPI equation. Second, for both equations, the association between eGFR and (1) kidney failure with replacement therapy (KFRT), (2) all-cause mortality, (3) cardiovascular mortality and (4) major adverse cardiovascular events was estimated with Cox regression. Third, prognostic accuracy of both eGFR equations within the Kidney Failure Risk Equation was assessed with discrimination and calibration. RESULTS Compared with the 2009 equation, the 2021 equation yielded a higher eGFR by a median (IQR) of 3.9 (2.9–4.8) mL/min/1.73 m2, decreasing prevalence of CKD G3–G5 from 5.1 to 3.8%. The 2021 equation reclassified 9.9% of the total population and 36.2% of the CKD G3–G5 population to a less severe eGFR category. Individuals who were reclassified to less severe eGFR categories were older and therefore exhibited higher crude risks of all-cause/cardiovascular death and major adverse cardiovascular events, and lower risk of kidney replacement therapy compared with nonreclassified participants of similar eGFR. eGFR by both equations strongly predicted study outcomes, with similar discrimination and calibration for the Kidney Failure Risk Equation. CONCLUSION Implementing the 2021 CKD-EPI equation in predominantly white European populations raises eGFR by a modest amount (larger at older age and men) and shifts a major proportion of CKD patients to a higher eGFR category, with eGFR by both equations strongly predicting outcomes.

FC078: Impact of Removing Race from the CKD-EPI Equation: Analysis of 1.6 Million Swedish Adults

Abstract BACKGROUND AND AIMS The Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) recently developed a novel creatinine-based eGFR equation without a race coefficient. While American nephrology societies recommend using this novel equation, its implications are unknown. METHOD We included 1.6 million adult individuals with routine outpatient serum creatinine testing during 2007–2018 in Stockholm, Sweden. First, we calculated reclassification across KDIGO eGFR categories when changing from the 2009 to 2021 CKD-EPI equation. Second, for both equations, the association between eGFR and (1) kidney failure with replacement therapy (KFRT), (2) all-cause mortality, (3) cardiovascular mortality and (4) major adverse cardiovascular events was estimated with Cox regression. Third, prognostic accuracy of both eGFR equations within the Kidney Failure Risk Equation was assessed with discrimination and calibration. RESULTS Compared with the 2009 equation, the 2021 equation yielded a higher eGFR by a median (IQR) of 3.9 (2.9–4.8) mL/min/1.73 m2, decreasing prevalence of CKD G3–G5 from 5.1 to 3.8%. The 2021 equation reclassified 9.9% of the total population and 36.2% of the CKD G3–G5 population to a less severe eGFR category. Individuals who were reclassified to less severe eGFR categories were older and therefore exhibited higher crude risks of all-cause/cardiovascular death and major adverse cardiovascular events, and lower risk of kidney replacement therapy compared with nonreclassified participants of similar eGFR. eGFR by both equations strongly predicted study outcomes, with similar discrimination and calibration for the Kidney Failure Risk Equation. CONCLUSION Implementing the 2021 CKD-EPI equation in predominantly white European populations raises eGFR by a modest amount (larger at older age and men) and shifts a major proportion of CKD patients to a higher eGFR category, with eGFR by both equations strongly predicting outcomes.

Association of Clonal Hematopoiesis of Indeterminate Potential with Worse Kidney Function and Anemia in Two Cohorts of Patients with Advanced Chronic Kidney Disease.

Clonal hematopoiesis of indeterminate potential (CHIP) is an inflammatory premalignant disorder resulting from acquired genetic mutations in hematopoietic stem cells. This condition is common in aging populations and associated with cardiovascular morbidity and overall mortality, but its role in CKD is unknown. We performed targeted sequencing to detect CHIP mutations in two independent cohorts of 87 and 85 adults with an eGFR<60 ml/min per 1.73m2. We also assessed kidney function, hematologic, and mineral bone disease parameters cross-sectionally at baseline, and collected creatinine measurements over the following 5-year period. At baseline, CHIP was detected in 18 of 87 (21%) and 25 of 85 (29%) cohort participants. Participants with CHIP were at higher risk of kidney failure, as predicted by the Kidney Failure Risk Equation (KFRE), compared with those without CHIP. Individuals with CHIP manifested a 2.2-fold increased risk of a 50% decline in eGFR or ESKD over 5 years of follow-up (hazard ratio 2.2; 95% confidence interval, 1.2 to 3.8) in a Cox proportional hazard model adjusted for age, sex, and baseline eGFR. The addition of CHIP to 2-year and 5-year calibrated KFRE risk models improved ESKD predictions. Those with CHIP also had lower hemoglobin, higher ferritin, and higher red blood cell mean corpuscular volume versus those without CHIP. In this exploratory analysis of individuals with preexisting CKD, CHIP was associated with higher baseline KFRE scores, greater progression of CKD, and anemia. Further research is needed to define the nature of the relationship between CHIP and kidney disease progression.

Editorial introductions.

Current Opinion in Nephrology and Hypertension was launched in 1992. It is part of a successful series of review journals whose unique format is designed to provide a systematic and critical assessment of the literature as presented in the many primary journals. The field of nephrology and hypertension is divided into 12 sections that are each reviewed once a year. Each section is assigned a Section Editor, a leading authority in the area, who identifies the most important topics at that time. Here we are pleased to introduce the Section Editors for this issue. SECTION EDITORS Agnes B. FogoAgnes B. FogoProfessor Agnes B. Fogo completed her medical training at Vanderbilt University Medical School, USA. She has remained at Vanderbilt for residency in pathology and fellowship in renal pathology and experimental kidney disease models. Her main research interests are focused on potential regression of chronic kidney disease and tubular-glomerular crosstalk in this process, pursued with funding from the National Institutes of Health and other sources. She is also dedicated to diagnostic renal pathology and has created an online Atlas of Renal Pathology with the American Journal of Kidney Disease and the National Kidney Foundation. She has also authored two books on renal pathology. Prof. Fogo enjoys teaching renal pathology and received the Robert G. Narins award from the American Society of Nephrology in 2011 as well as the Roscoe R. Robinson award from the International Society of Nephrology in 2019, for her contributions in teaching. She is currently President of the ISN and the John L. Shapiro Professor of Pathology, Microbiology and Immunology, Professor of Medicine and Paediatrics and Director of the Renal/Electron Microscopy Laboratory at Vanderbilt University Medical Center, Nashville, Tennessee, USA. Navdeep TangriNavdeep TangriDr Tangri is an attending physician and Associate Professor in the Division of Nephrology, Department of Medicine and the Department of Community Health Sciences at the University of Manitoba, Canada. Dr Tangri's research program is clinical and translational and focused on improving clinical decision making for patients with advanced chronic kidney disease. He developed and validated the Kidney Failure Risk Equation (KFRE) to predict the need for dialysis in patients with Chronic Kidney Disease (CKD), and is presently engaged in multiple validation and implementation exercises to increase the uptake of the KFRE. In addition, Dr Tangri is conducting a large prospective study on frailty, physical and cognitive function in advanced CKD, as well as leading a multinational randomized trial on the safety and efficacy of a new therapy in this population. He has published over 250 manuscripts and presented at multiple national and international scientific meetings. He is also a recipient of the CIHR New Investigator Award and a CIHR Foundation grant.

In CKD, the Kidney Failure Risk Equation predicted 2-y risk for ESKD better than eGFR alone.

Bundy JD, Mills KT, Anderson AH, et al. Prediction of end-stage kidney disease using estimated glomerular filtration rate with and without race: a prospective cohort study. Ann Intern Med. 2022;175:305-13. 35007146.

A prediction model with lifestyle factors improves the predictive ability for renal replacement therapy: a cohort of 442 714 Asian adults.

ABSTRACTBackgroundThere are limited renal replacement therapy (RRT) prediction models with good performance in the general population. We developed a model that includes lifestyle factors to improve predictive ability for RRT in the population at large.MethodsWe used data collected between 1996 and 2017 from a medical screening in a cohort comprising 442 714 participants aged 20 years or over. After a median follow-up of 13 years, we identified 2212 individuals with end-stage renal disease (RRT, n: 2091; kidney transplantation, n: 121). We built three models for comparison: model 1: basic model, Kidney Failure Risk Equation with four variables (age, sex, estimated glomerular filtration rate and proteinuria); model 2: basic model + medical history + lifestyle risk factors; and model 3: model 2 + all significant clinical variables. We used the Cox proportional hazards model to construct a points-based model and applied the C statistic.ResultsAdding lifestyle factors to the basic model, the C statistic improved in model 2 from 0.91 to 0.94 (95% confidence interval: 0.94, 0.95). Model 3 showed even better C statistic value i.e., 0.95 (0.95, 0.96). With a cut-off score of 33, model 3 identified 3% of individuals with RRT risk in 10 years. This model detected over half of individuals progressing to RRT, which was higher than the sensitivity of cohort participants with stage 3 or higher chronic kidney disease (0.53 versus 0.48).ConclusionsOur prediction model including medical history and lifestyle factors improved the predictive ability for end-stage renal disease in the general population in addition to chronic kidney disease population.

Validation of the kidney failure risk equation in a Portuguese cohort

IntroductionIn chronic kidney disease (CKD) patients, the risk of kidney replacement therapy (KRT) is highly variable. In 2011, Tangri et al. developed the kidney failure risk equations (KFRE) to predict the 2 and 5-year probability of requiring kidney replacement therapy (KRT). The KFRE is an easily calculated 4-variable equation which has been extensively validated in multiple cohorts. The aim of this study was to validate this risk score in a Portuguese cohort. MethodsWe conducted a retrospective analysis of CKD patients stage 3–5 referred for nephrology consult at Centro Hospitalar Universitário Lisboa Norte during the first 6 months of 2018. Age, gender, estimated glomerular filtration rate (eGFR) and albuminuria were assessed. The 4-variable kidney failure risk equation (KFRE) calibrated to a non-North American population was calculated. Requirement of KRT was assessed in a 2-year follow-up. We assessed the Cox logistic regression method of the KFRE to predict KRT requirement and the discriminatory ability was determined using the receiver operating characteristic (ROC) curve. A cut-off value was defined as that with the highest validity. Results360 patients were included and 54.4% were male. Mean age was 74.9±12.2 years, serum creatinine was 1.97±0.84mg/dL, eGFR was 33.4±12.13ml/min/1.73m2 and albuminuria was 571.1±848.3mg/g. Mean calculated risk score was 6.2±11.2%. Twenty-three patients required KRT (6.4%) in the two-year follow-up. The hazard ratio was 1.1 [95% CI (1.06–1.12), p<0.001] for the 2-year risk of KRT. The KFRE predicted progression to KRT requirement with an auROC of 0.903, [95% CI (0.86–0.95), p<0.001], with a sensitivity 91.3% and specificity of 71.8%. The optimal KFRE cut-off was >4.5% for 2-year nephrologist referral, with an hazard ratio of HR 26.7 [95% CI (6.15–116.3), p<0.001] for 2-year risk of KRT requirement. DiscussionWe have independently externally validated the 2-year KFRE and shown that it has excellent discrimination. The KFRE should be incorporated in clinical care of patients with CKD to improve patient-clinician dialogue and provide guidance on timing of referral for nephrology evaluation and planning for dialysis access.

Risk-based versus GFR threshold criteria for nephrology referral in chronic kidney disease.

Chronic kidney disease (CKD) and kidney failure are global health problems associated with morbidity, mortality and healthcare costs, with unequal access to kidney replacement therapy between countries. The diversity of guidelines concerning referral from primary care to a specialist nephrologist determines different outcomes around the world among patients with CKD where several guidelines recommend referral when the glomerular filtration rate (GFR) is<30mL/min/1.73m2 regardless of age. Additionally, fixed non-age-adapted diagnostic criteria for CKD that do not distinguish correctly between normal kidney senescence and true kidney disease can lead to overdiagnosis of CKD in the elderly and underdiagnosis of CKD in young patients and contributes to the unfair referral of CKD patients to a kidney specialist. Non-age-adapted recommendations contribute to unnecessary referral in the very elderly with a mild disease where the risk of death consistently exceeds the risk of progression to kidney failure and ignore the possibility of effective interventions of a young patient with long life expectancy. The opportunity of mitigating CKD progression and cardiovascular complications in young patients with early stages of CKD is a task entrusted to primary care providers who are possibly unable to optimally accomplish guideline-directed medical therapy for this purpose. The shortage in the nephrology workforce has classically led to focused referral on advanced CKD stages preparing for kidney replacement, but the need for hasty referral to a nephrologist because of the urgent requirement for kidney replacement therapy in advanced CKD is still observed and changes are required to move toward reducing the kidney failure burden. The Kidney Failure Risk Equation (KFRE) is a novel tool that can guide wiser nephrology referrals and impact patients.

Is there a role of the kidney failure risk equation in optimizing timing of vascular access creation in pre-dialysis patients?

The aims of this study were to assess the utility of using the Kidney Failure Risk Equation (KFRE) as an indicator to guide timing of vascular access creation in pre-dialysis patients. Patients referred for vascular access creation had KFRE calculated at the time of assessment and compared to standard criteria for referral. Receiver operating characteristic curves were produced for each parameter. The outcomes at 3 months, 6 months, and 1 year were used as time points for analysis. Two hundred and three patients were assessed, and full data sets were available on 190 (94.6%). Access was created in 156 patients (82.1%) with a fistula in 153 (98.7%). Only 65.7% initiated dialysis within the follow up period. Those patients with an AV access created (n = 156) 37 (23.7%) did not reach end stage over the entire follow up period. Of the remaining patients (n = 119) that reached end stage 72.2% (n = 86) started on an AVF/AVG and 27.7% (n = 33) on a CVC. Using ROC analysis for referral eGFR, ACR and KFRE predicting dialysis initiation predictors resulted in C statistics for eGFR, ACR, and KFRE2 of 0.68 (0.58-0.79), 0.75 (0.65-0.84), and 0.72 (0.62-0.81) at 3 months; 0.73 (0.65-0.81), 0.70 (0.62-0.78), and 0.75 (0.67-0.81) at 6 months; and 0.65 (0.57-0.72); 0.67 (0.59-0.75), and 0.68 (0.61-0.77) at 12 months. In a group of patients referred for vascular access creation the predictive models are relatively poor when applied to initiation of dialysis. The application of current guidelines to fistula creation appears to result in a high rate of unnecessary fistula formation and non-use. The study requires further evaluation in a test set of patients to confirm these findings and also identify where such risk based approaches may need modification.

76 Performance of the Kidney Failure Risk Equation by Race in Advanced CKD

76 Performance of the Kidney Failure Risk Equation by Race in Advanced CKD

Machine Learning Improves Upon Clinicians' Prediction of End Stage Kidney Disease.

Background and ObjectivesChronic kidney disease progression to ESKD is associated with a marked increase in mortality and morbidity. Its progression is highly variable and difficult to predict.MethodsThis is an observational, retrospective, single-centre study. The cohort was patients attending hospital and nephrology clinic at The Canberra Hospital from September 1996 to March 2018. Demographic data, vital signs, kidney function test, proteinuria, and serum glucose were extracted. The model was trained on the featurised time series data with XGBoost. Its performance was compared against six nephrologists and the Kidney Failure Risk Equation (KFRE).ResultsA total of 12,371 patients were included, with 2,388 were found to have an adequate density (three eGFR data points in the first 2 years) for subsequent analysis. Patients were divided into 80%/20% ratio for training and testing datasets.ML model had superior performance than nephrologist in predicting ESKD within 2 years with 93.9% accuracy, 60% sensitivity, 97.7% specificity, 75% positive predictive value. The ML model was superior in all performance metrics to the KFRE 4- and 8-variable models.eGFR and glucose were found to be highly contributing to the ESKD prediction performance.ConclusionsThe computational predictions had higher accuracy, specificity and positive predictive value, which indicates the potential integration into clinical workflows for decision support.

Independent External Validation and Comparison of Death and Kidney Replacement Therapy Prediction Models in Advanced CKD

Rationale & ObjectiveThe Kidney Failure Risk Equation (KFRE) is widely used to predict the risk of kidney replacement therapy (KRT) initiation in chronic kidney disease (CKD) stages G3-G5. The new Grams calculator developed for advanced CKD (stage G4+) predicts KRT initiation, cardiovascular events, and death by uniquely incorporating the competing risk of death. We aimed to validate this tool in a stage G4+ cohort for death and KRT.Study DesignRetrospective cohort study.Setting & Participants442 patients with CKD stage G4+ (mean ± SD age, 73 ± 12 years; mean ± SD estimated glomerular filtration rate, 20 ± 6.2 mL/min/1.73 m2) who visited the multidisciplinary CKD clinic at Kingston Health Sciences Center in Ontario, Canada.Outcomes & Analytical ApproachDiscrimination and calibration were examined for the outcome of death using the 2- and 4-year Grams scores. The 2- and 5-year KFRE and 2- and 4-year Grams scores were compared in terms of discrimination and calibration for KRT.ResultsThere were 91, 161, and 206 death events and 90, 145, and 159 KRT events in our cohort at 2, 4, and 5 years, respectively. The Grams model demonstrated modest discrimination for death at 4 years (area under the curve [AUC] 0.70; 95% CI, 0.65-0.75) and performed worse at 2 years (AUC, 0.63; 95% CI, 0.57-0.70). It only overpredicted death by approximately 10% across most of the predicted range. Both models had similar discrimination for KRT at 2 years (KFRE AUC, 0.83; 95% CI, 0.78-0.88 and Grams AUC, 0.8; 95% CI, 0.76-0.87), 4 years (Grams AUC, 0.82; 95% CI, 0.77-0.86), and 5 years (KFRE AUC, 0.81; 95% CI, 0.76-0.85). There was excellent calibration for KRT using the 2-year KFRE and Grams values for predicted risk thresholds of ≤15% and using the 5-year KFRE and 4-year Grams values for predicted risk thresholds of ≤20%. At higher risk ranges, KFRE overpredicts and Grams underpredicts the KRT risk.LimitationsThis is a single-center study with a primarily White cohort limited by smaller sample sizes at the higher ranges of the predicted risks, particularly for the Grams calculator.ConclusionsThe Grams model provides moderately accurate death predictions, and consideration should be given to its incorporation into patient education and advanced care planning. Both the Grams and KFRE models remain clinically useful for determining KRT risks in advanced CKD.

Contemporary risk prediction models in chronic kidney disease: when less is more.

Clinicians have an ever-increasing number of prediction tools at their disposal for estimating the risk of kidney failure in their patients. This review aims to summarize contemporary evidence for chronic kidney disease (CKD) risk prediction models across the spectrum of kidney function, and explore nuances in the interpretation of risk estimates. A European study using predominantly laboratory data has extended kidney failure prediction to patients with more preserved estimated glomerular filtration rate. For older patients with advanced CKD, prediction tools that censor for death (such as the Kidney Failure Risk Equation) overestimate the risk of kidney failure, especially over time horizons longer than 2 years. This problem can be addressed by accounting for the competing risk of death, as shown in well designed validation studies. The clinical utility of kidney failure risk prediction tools is being increasingly tested at a population level to inform policy and referral guidelines. There is welcome trend to validate existing prediction tools in diverse clinical settings and identify their role in clinical practice. Clinicians should be cognizant of overestimating kidney failure risk in older patients with advanced CKD due to the competing risk of death. For moderate CKD and for short-term predictions, the Kidney Failure Risk Equation remains the most widely validated prediction tool.