Risk score for first-screening of prevalent undiagnosed chronic kidney disease in Peru

The CRONICAS-CKD risk score

Rodrigo M. Carrillo-Larco, J. Jaime Miranda, Robert H. Gilman, Josefina Medina-Lezama, Julio A. Chirinos-Pacheco, Paola V. Muñoz-Retamozo, Liam Smeeth, William Checkley, Antonio Bernabe-Ortiz, Juan P. Casas, George Davey Smith, Shah Ebrahim, Luis Huicho, Germán Málaga, Victor Manuel Montori, Víctor M. Montori, Gregory B. Diette, Fabiola León-Velarde, Robert A. Wise, Héctor H. García & 1 others J. Jaime Miranda

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Background: Chronic Kidney Disease (CKD) represents a great burden for the patient and the health system, particularly if diagnosed at late stages. Consequently, tools to identify patients at high risk of having CKD are needed, particularly in limited-resources settings where laboratory facilities are scarce. This study aimed to develop a risk score for prevalent undiagnosed CKD using data from four settings in Peru: a complete risk score including all associated risk factors and another excluding laboratory-based variables. Methods: Cross-sectional study. We used two population-based studies: one for developing and internal validation (CRONICAS), and another (PREVENCION) for external validation. Risk factors included clinical- and laboratory-based variables, among others: sex, age, hypertension and obesity; and lipid profile, anemia and glucose metabolism. The outcome was undiagnosed CKD: eGFR < 60 ml/min/1.73m2. We tested the performance of the risk scores using the area under the receiver operating characteristic (ROC) curve, sensitivity, specificity, positive/negative predictive values and positive/negative likelihood ratios. Results: Participants in both studies averaged 57.7 years old, and over 50% were females. Age, hypertension and anemia were strongly associated with undiagnosed CKD. In the external validation, at a cut-off point of 2, the complete and laboratory-free risk scores performed similarly well with a ROC area of 76.2% and 76.0%, respectively (P = 0.784). The best assessment parameter of these risk scores was their negative predictive value: 99.1% and 99.0% for the complete and laboratory-free, respectively. Conclusions: The developed risk scores showed a moderate performance as a screening test. People with a score of ≥ 2 points should undergo further testing to rule out CKD. Using the laboratory-free risk score is a practical approach in developing countries where laboratories are not readily available and undiagnosed CKD has significant morbidity and mortality.

Original languageEnglish (US)
Article number343
JournalBMC Nephrology
Volume18
Issue number1
DOIs
StatePublished - Nov 29 2017

Fingerprint

Peru
Chronic Renal Insufficiency
ROC Curve
Anemia
Hypertension
Developing Countries
Obesity
Cross-Sectional Studies
Morbidity
Lipids
Glucose
Sensitivity and Specificity
Mortality
Health
Population

Keywords

  • Chronic kidney disease
  • Kidney
  • Latin America
  • Risk assessment

ASJC Scopus subject areas

  • Nephrology

Cite this

Carrillo-Larco, R. M., Miranda, J. J., Gilman, R. H., Medina-Lezama, J., Chirinos-Pacheco, J. A., Muñoz-Retamozo, P. V., ... Jaime Miranda, J. (2017). Risk score for first-screening of prevalent undiagnosed chronic kidney disease in Peru: The CRONICAS-CKD risk score. BMC Nephrology, 18(1), [343]. https://doi.org/10.1186/s12882-017-0758-4

Risk score for first-screening of prevalent undiagnosed chronic kidney disease in Peru : The CRONICAS-CKD risk score. / Carrillo-Larco, Rodrigo M.; Miranda, J. Jaime; Gilman, Robert H.; Medina-Lezama, Josefina; Chirinos-Pacheco, Julio A.; Muñoz-Retamozo, Paola V.; Smeeth, Liam; Checkley, William; Bernabe-Ortiz, Antonio; Casas, Juan P.; Smith, George Davey; Ebrahim, Shah; Huicho, Luis; Málaga, Germán; Montori, Victor Manuel; Montori, Víctor M.; Diette, Gregory B.; León-Velarde, Fabiola; Wise, Robert A.; García, Héctor H.; Jaime Miranda, J.

In: BMC Nephrology, Vol. 18, No. 1, 343, 29.11.2017.

Research output: Contribution to journalArticle

Carrillo-Larco, RM, Miranda, JJ, Gilman, RH, Medina-Lezama, J, Chirinos-Pacheco, JA, Muñoz-Retamozo, PV, Smeeth, L, Checkley, W, Bernabe-Ortiz, A, Casas, JP, Smith, GD, Ebrahim, S, Huicho, L, Málaga, G, Montori, VM, Montori, VM, Diette, GB, León-Velarde, F, Wise, RA, García, HH & Jaime Miranda, J 2017, 'Risk score for first-screening of prevalent undiagnosed chronic kidney disease in Peru: The CRONICAS-CKD risk score', BMC Nephrology, vol. 18, no. 1, 343. https://doi.org/10.1186/s12882-017-0758-4
Carrillo-Larco RM, Miranda JJ, Gilman RH, Medina-Lezama J, Chirinos-Pacheco JA, Muñoz-Retamozo PV et al. Risk score for first-screening of prevalent undiagnosed chronic kidney disease in Peru: The CRONICAS-CKD risk score. BMC Nephrology. 2017 Nov 29;18(1). 343. https://doi.org/10.1186/s12882-017-0758-4
Carrillo-Larco, Rodrigo M. ; Miranda, J. Jaime ; Gilman, Robert H. ; Medina-Lezama, Josefina ; Chirinos-Pacheco, Julio A. ; Muñoz-Retamozo, Paola V. ; Smeeth, Liam ; Checkley, William ; Bernabe-Ortiz, Antonio ; Casas, Juan P. ; Smith, George Davey ; Ebrahim, Shah ; Huicho, Luis ; Málaga, Germán ; Montori, Victor Manuel ; Montori, Víctor M. ; Diette, Gregory B. ; León-Velarde, Fabiola ; Wise, Robert A. ; García, Héctor H. ; Jaime Miranda, J. / Risk score for first-screening of prevalent undiagnosed chronic kidney disease in Peru : The CRONICAS-CKD risk score. In: BMC Nephrology. 2017 ; Vol. 18, No. 1.
@article{974746eb564c424a801b496ea450ff90,
title = "Risk score for first-screening of prevalent undiagnosed chronic kidney disease in Peru: The CRONICAS-CKD risk score",
abstract = "Background: Chronic Kidney Disease (CKD) represents a great burden for the patient and the health system, particularly if diagnosed at late stages. Consequently, tools to identify patients at high risk of having CKD are needed, particularly in limited-resources settings where laboratory facilities are scarce. This study aimed to develop a risk score for prevalent undiagnosed CKD using data from four settings in Peru: a complete risk score including all associated risk factors and another excluding laboratory-based variables. Methods: Cross-sectional study. We used two population-based studies: one for developing and internal validation (CRONICAS), and another (PREVENCION) for external validation. Risk factors included clinical- and laboratory-based variables, among others: sex, age, hypertension and obesity; and lipid profile, anemia and glucose metabolism. The outcome was undiagnosed CKD: eGFR < 60 ml/min/1.73m2. We tested the performance of the risk scores using the area under the receiver operating characteristic (ROC) curve, sensitivity, specificity, positive/negative predictive values and positive/negative likelihood ratios. Results: Participants in both studies averaged 57.7 years old, and over 50{\%} were females. Age, hypertension and anemia were strongly associated with undiagnosed CKD. In the external validation, at a cut-off point of 2, the complete and laboratory-free risk scores performed similarly well with a ROC area of 76.2{\%} and 76.0{\%}, respectively (P = 0.784). The best assessment parameter of these risk scores was their negative predictive value: 99.1{\%} and 99.0{\%} for the complete and laboratory-free, respectively. Conclusions: The developed risk scores showed a moderate performance as a screening test. People with a score of ≥ 2 points should undergo further testing to rule out CKD. Using the laboratory-free risk score is a practical approach in developing countries where laboratories are not readily available and undiagnosed CKD has significant morbidity and mortality.",
keywords = "Chronic kidney disease, Kidney, Latin America, Risk assessment",
author = "Carrillo-Larco, {Rodrigo M.} and Miranda, {J. Jaime} and Gilman, {Robert H.} and Josefina Medina-Lezama and Chirinos-Pacheco, {Julio A.} and Mu{\~n}oz-Retamozo, {Paola V.} and Liam Smeeth and William Checkley and Antonio Bernabe-Ortiz and Casas, {Juan P.} and Smith, {George Davey} and Shah Ebrahim and Luis Huicho and Germ{\'a}n M{\'a}laga and Montori, {Victor Manuel} and Montori, {V{\'i}ctor M.} and Diette, {Gregory B.} and Fabiola Le{\'o}n-Velarde and Wise, {Robert A.} and Garc{\'i}a, {H{\'e}ctor H.} and {Jaime Miranda}, J.",
year = "2017",
month = "11",
day = "29",
doi = "10.1186/s12882-017-0758-4",
language = "English (US)",
volume = "18",
journal = "BMC Nephrology",
issn = "1471-2369",
publisher = "BioMed Central",
number = "1",

}

TY - JOUR

T1 - Risk score for first-screening of prevalent undiagnosed chronic kidney disease in Peru

T2 - The CRONICAS-CKD risk score

AU - Carrillo-Larco, Rodrigo M.

AU - Miranda, J. Jaime

AU - Gilman, Robert H.

AU - Medina-Lezama, Josefina

AU - Chirinos-Pacheco, Julio A.

AU - Muñoz-Retamozo, Paola V.

AU - Smeeth, Liam

AU - Checkley, William

AU - Bernabe-Ortiz, Antonio

AU - Casas, Juan P.

AU - Smith, George Davey

AU - Ebrahim, Shah

AU - Huicho, Luis

AU - Málaga, Germán

AU - Montori, Victor Manuel

AU - Montori, Víctor M.

AU - Diette, Gregory B.

AU - León-Velarde, Fabiola

AU - Wise, Robert A.

AU - García, Héctor H.

AU - Jaime Miranda, J.

PY - 2017/11/29

Y1 - 2017/11/29

N2 - Background: Chronic Kidney Disease (CKD) represents a great burden for the patient and the health system, particularly if diagnosed at late stages. Consequently, tools to identify patients at high risk of having CKD are needed, particularly in limited-resources settings where laboratory facilities are scarce. This study aimed to develop a risk score for prevalent undiagnosed CKD using data from four settings in Peru: a complete risk score including all associated risk factors and another excluding laboratory-based variables. Methods: Cross-sectional study. We used two population-based studies: one for developing and internal validation (CRONICAS), and another (PREVENCION) for external validation. Risk factors included clinical- and laboratory-based variables, among others: sex, age, hypertension and obesity; and lipid profile, anemia and glucose metabolism. The outcome was undiagnosed CKD: eGFR < 60 ml/min/1.73m2. We tested the performance of the risk scores using the area under the receiver operating characteristic (ROC) curve, sensitivity, specificity, positive/negative predictive values and positive/negative likelihood ratios. Results: Participants in both studies averaged 57.7 years old, and over 50% were females. Age, hypertension and anemia were strongly associated with undiagnosed CKD. In the external validation, at a cut-off point of 2, the complete and laboratory-free risk scores performed similarly well with a ROC area of 76.2% and 76.0%, respectively (P = 0.784). The best assessment parameter of these risk scores was their negative predictive value: 99.1% and 99.0% for the complete and laboratory-free, respectively. Conclusions: The developed risk scores showed a moderate performance as a screening test. People with a score of ≥ 2 points should undergo further testing to rule out CKD. Using the laboratory-free risk score is a practical approach in developing countries where laboratories are not readily available and undiagnosed CKD has significant morbidity and mortality.

AB - Background: Chronic Kidney Disease (CKD) represents a great burden for the patient and the health system, particularly if diagnosed at late stages. Consequently, tools to identify patients at high risk of having CKD are needed, particularly in limited-resources settings where laboratory facilities are scarce. This study aimed to develop a risk score for prevalent undiagnosed CKD using data from four settings in Peru: a complete risk score including all associated risk factors and another excluding laboratory-based variables. Methods: Cross-sectional study. We used two population-based studies: one for developing and internal validation (CRONICAS), and another (PREVENCION) for external validation. Risk factors included clinical- and laboratory-based variables, among others: sex, age, hypertension and obesity; and lipid profile, anemia and glucose metabolism. The outcome was undiagnosed CKD: eGFR < 60 ml/min/1.73m2. We tested the performance of the risk scores using the area under the receiver operating characteristic (ROC) curve, sensitivity, specificity, positive/negative predictive values and positive/negative likelihood ratios. Results: Participants in both studies averaged 57.7 years old, and over 50% were females. Age, hypertension and anemia were strongly associated with undiagnosed CKD. In the external validation, at a cut-off point of 2, the complete and laboratory-free risk scores performed similarly well with a ROC area of 76.2% and 76.0%, respectively (P = 0.784). The best assessment parameter of these risk scores was their negative predictive value: 99.1% and 99.0% for the complete and laboratory-free, respectively. Conclusions: The developed risk scores showed a moderate performance as a screening test. People with a score of ≥ 2 points should undergo further testing to rule out CKD. Using the laboratory-free risk score is a practical approach in developing countries where laboratories are not readily available and undiagnosed CKD has significant morbidity and mortality.

KW - Chronic kidney disease

KW - Kidney

KW - Latin America

KW - Risk assessment

UR - http://www.scopus.com/inward/record.url?scp=85036564975&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85036564975&partnerID=8YFLogxK

U2 - 10.1186/s12882-017-0758-4

DO - 10.1186/s12882-017-0758-4

M3 - Article

VL - 18

JO - BMC Nephrology

JF - BMC Nephrology

SN - 1471-2369

IS - 1

M1 - 343

ER -