Spot urine sodium measurements do not accurately estimate dietary sodium intake in chronic kidney disease

Carly E. Dougher; Dena E. Rifkin; Cheryl A M Anderson; Gerard Smits; Martha S. Persky; Geoffrey A. Block; Joachim H. Ix

doi:10.3945/ajcn.115.127423

Spot urine sodium measurements do not accurately estimate dietary sodium intake in chronic kidney disease

Carly E. Dougher, Dena E. Rifkin, Cheryl A M Anderson, Gerard Smits, Martha S. Persky, Geoffrey A. Block, Joachim H. Ix

Research output: Contribution to journal › Article › peer-review

23 Scopus citations

Abstract

Sodium intake influences blood pressure and proteinuria, yet the impact on long-term outcomes is uncertain in chronic kidney disease (CKD). Accurate assessment is essential for clinical and public policy recommendations, but few large-scale studies use 24-h urine collections. Recent studies that used spot urine sodium and associated estimating equations suggest that they may provide a suitable alternative, but their accuracy in patients with CKD is unknown. Objective: We compared the accuracy of 4 equations [the Nerbass, INTERSALT (International Cooperative Study on Salt, Other Factors, and Blood Pressure), Tanaka, and Kawasaki equations] that use spot urine sodium to estimate 24-h sodium excretion in patients with moderate to advanced CKD. Design: We evaluated the accuracy of spot urine sodium to predict mean 24-h urine sodium excretion over 9 mo in 129 participants with stage 3-4 CKD. Spot morning urine sodium was used in 4 estimating equations. Bias, precision, and accuracy were assessed and compared across each equation. Results: The mean age of the participants was 67 y, 52% were female, and the mean estimated glomerular filtration rate was 31 ± 9mL · min^-1 · 1.73 m^-2. The mean ± SD number of 24-h urine collections was 3.5 ± 0.8/participant, and the mean 24-h sodium excretion was 168.2 ± 67.5 mmol/d. Although the Tanaka equation demonstrated the least bias (mean:-8.2 mmol/d), all 4 equations had poor precision and accuracy. The INTERSALT equation demonstrated the highest accuracy but derived an estimate only within 30% of mean measured sodium excretion in only 57% of observations. Bland-Altman plots revealed systematic bias with the Nerbass, INTERSALT, and Tanaka equations, underestimating sodium excretion when intake was high. Conclusion: These findings do not support the use of spot urine specimens to estimate dietary sodium intake in patients with CKD and research studies enriched with patients with CKD.

Original language	English (US)
Pages (from-to)	298-305
Number of pages	8
Journal	American Journal of Clinical Nutrition
Volume	104
Issue number	2
DOIs	https://doi.org/10.3945/ajcn.115.127423
State	Published - Aug 1 2016
Externally published	Yes

Keywords

24-hour urine
Cardiovascular disease
Chronic kidney disease
Hypertension
Sodium intake
Spot urine

ASJC Scopus subject areas

Medicine (miscellaneous)
General Medicine
Nutrition and Dietetics

Access to Document

10.3945/ajcn.115.127423

Cite this

@article{d7728b5f208c42259b172d72276bde05,

title = "Spot urine sodium measurements do not accurately estimate dietary sodium intake in chronic kidney disease",

abstract = "Sodium intake influences blood pressure and proteinuria, yet the impact on long-term outcomes is uncertain in chronic kidney disease (CKD). Accurate assessment is essential for clinical and public policy recommendations, but few large-scale studies use 24-h urine collections. Recent studies that used spot urine sodium and associated estimating equations suggest that they may provide a suitable alternative, but their accuracy in patients with CKD is unknown. Objective: We compared the accuracy of 4 equations [the Nerbass, INTERSALT (International Cooperative Study on Salt, Other Factors, and Blood Pressure), Tanaka, and Kawasaki equations] that use spot urine sodium to estimate 24-h sodium excretion in patients with moderate to advanced CKD. Design: We evaluated the accuracy of spot urine sodium to predict mean 24-h urine sodium excretion over 9 mo in 129 participants with stage 3-4 CKD. Spot morning urine sodium was used in 4 estimating equations. Bias, precision, and accuracy were assessed and compared across each equation. Results: The mean age of the participants was 67 y, 52% were female, and the mean estimated glomerular filtration rate was 31 ± 9mL · min-1 · 1.73 m-2. The mean ± SD number of 24-h urine collections was 3.5 ± 0.8/participant, and the mean 24-h sodium excretion was 168.2 ± 67.5 mmol/d. Although the Tanaka equation demonstrated the least bias (mean:-8.2 mmol/d), all 4 equations had poor precision and accuracy. The INTERSALT equation demonstrated the highest accuracy but derived an estimate only within 30% of mean measured sodium excretion in only 57% of observations. Bland-Altman plots revealed systematic bias with the Nerbass, INTERSALT, and Tanaka equations, underestimating sodium excretion when intake was high. Conclusion: These findings do not support the use of spot urine specimens to estimate dietary sodium intake in patients with CKD and research studies enriched with patients with CKD.",

keywords = "24-hour urine, Cardiovascular disease, Chronic kidney disease, Hypertension, Sodium intake, Spot urine",

author = "Dougher, {Carly E.} and Rifkin, {Dena E.} and Anderson, {Cheryl A M} and Gerard Smits and Persky, {Martha S.} and Block, {Geoffrey A.} and Ix, {Joachim H.}",

year = "2016",

month = aug,

day = "1",

doi = "10.3945/ajcn.115.127423",

language = "English (US)",

volume = "104",

pages = "298--305",

journal = "American Journal of Clinical Nutrition",

issn = "0002-9165",

publisher = "American Society for Nutrition",

number = "2",

}

TY - JOUR

T1 - Spot urine sodium measurements do not accurately estimate dietary sodium intake in chronic kidney disease

AU - Dougher, Carly E.

AU - Rifkin, Dena E.

AU - Anderson, Cheryl A M

AU - Smits, Gerard

AU - Persky, Martha S.

AU - Block, Geoffrey A.

AU - Ix, Joachim H.

PY - 2016/8/1

Y1 - 2016/8/1

N2 - Sodium intake influences blood pressure and proteinuria, yet the impact on long-term outcomes is uncertain in chronic kidney disease (CKD). Accurate assessment is essential for clinical and public policy recommendations, but few large-scale studies use 24-h urine collections. Recent studies that used spot urine sodium and associated estimating equations suggest that they may provide a suitable alternative, but their accuracy in patients with CKD is unknown. Objective: We compared the accuracy of 4 equations [the Nerbass, INTERSALT (International Cooperative Study on Salt, Other Factors, and Blood Pressure), Tanaka, and Kawasaki equations] that use spot urine sodium to estimate 24-h sodium excretion in patients with moderate to advanced CKD. Design: We evaluated the accuracy of spot urine sodium to predict mean 24-h urine sodium excretion over 9 mo in 129 participants with stage 3-4 CKD. Spot morning urine sodium was used in 4 estimating equations. Bias, precision, and accuracy were assessed and compared across each equation. Results: The mean age of the participants was 67 y, 52% were female, and the mean estimated glomerular filtration rate was 31 ± 9mL · min-1 · 1.73 m-2. The mean ± SD number of 24-h urine collections was 3.5 ± 0.8/participant, and the mean 24-h sodium excretion was 168.2 ± 67.5 mmol/d. Although the Tanaka equation demonstrated the least bias (mean:-8.2 mmol/d), all 4 equations had poor precision and accuracy. The INTERSALT equation demonstrated the highest accuracy but derived an estimate only within 30% of mean measured sodium excretion in only 57% of observations. Bland-Altman plots revealed systematic bias with the Nerbass, INTERSALT, and Tanaka equations, underestimating sodium excretion when intake was high. Conclusion: These findings do not support the use of spot urine specimens to estimate dietary sodium intake in patients with CKD and research studies enriched with patients with CKD.

AB - Sodium intake influences blood pressure and proteinuria, yet the impact on long-term outcomes is uncertain in chronic kidney disease (CKD). Accurate assessment is essential for clinical and public policy recommendations, but few large-scale studies use 24-h urine collections. Recent studies that used spot urine sodium and associated estimating equations suggest that they may provide a suitable alternative, but their accuracy in patients with CKD is unknown. Objective: We compared the accuracy of 4 equations [the Nerbass, INTERSALT (International Cooperative Study on Salt, Other Factors, and Blood Pressure), Tanaka, and Kawasaki equations] that use spot urine sodium to estimate 24-h sodium excretion in patients with moderate to advanced CKD. Design: We evaluated the accuracy of spot urine sodium to predict mean 24-h urine sodium excretion over 9 mo in 129 participants with stage 3-4 CKD. Spot morning urine sodium was used in 4 estimating equations. Bias, precision, and accuracy were assessed and compared across each equation. Results: The mean age of the participants was 67 y, 52% were female, and the mean estimated glomerular filtration rate was 31 ± 9mL · min-1 · 1.73 m-2. The mean ± SD number of 24-h urine collections was 3.5 ± 0.8/participant, and the mean 24-h sodium excretion was 168.2 ± 67.5 mmol/d. Although the Tanaka equation demonstrated the least bias (mean:-8.2 mmol/d), all 4 equations had poor precision and accuracy. The INTERSALT equation demonstrated the highest accuracy but derived an estimate only within 30% of mean measured sodium excretion in only 57% of observations. Bland-Altman plots revealed systematic bias with the Nerbass, INTERSALT, and Tanaka equations, underestimating sodium excretion when intake was high. Conclusion: These findings do not support the use of spot urine specimens to estimate dietary sodium intake in patients with CKD and research studies enriched with patients with CKD.

KW - 24-hour urine

KW - Cardiovascular disease

KW - Chronic kidney disease

KW - Hypertension

KW - Sodium intake

KW - Spot urine

UR - http://www.scopus.com/inward/record.url?scp=84980325618&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84980325618&partnerID=8YFLogxK

U2 - 10.3945/ajcn.115.127423

DO - 10.3945/ajcn.115.127423

M3 - Article

C2 - 27357090

AN - SCOPUS:84980325618

SN - 0002-9165

VL - 104

SP - 298

EP - 305

JO - American Journal of Clinical Nutrition

JF - American Journal of Clinical Nutrition

IS - 2

ER -

Spot urine sodium measurements do not accurately estimate dietary sodium intake in chronic kidney disease

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this