The relationship between repeated measurement of casual and 24-hour urinary 1
sodium-to-potassium ratio in patients with chronic kidney disease 2
3
Yuka Okuyama, MD1, Haruhito A. Uchida, MD, PhD1,2, Toshiyuki Iwahori, PhD3,4, 4
Hiroyoshi Segawa, MD5, Ayako Kato MD1, Hidemi Takeuchi, MD, PhD1, 5
Yuki Kakio MD, PhD1, Ryoko Umebayashi, MD, PhD1, Masashi Kitagawa, MD, PhD1, 6
Hitoshi Sugiyama, MD, PhD6, Katsuyuki Miura, MD, PhD3,5, 7
Hirotsugu Ueshima, MD, PhD3,5, Jun Wada, MD, PhD1 8
9
1Department of Nephrology, Rheumatology, Endocrinology and Metabolism, Okayama 10
University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama, 11
Japan 12
2Department of Chronic Kidney Disease and Cardiovascular Disease, Okayama University 13
Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama, Japan 14
3Department of Public Health, Shiga University of Medical Science, Otsu, Japan 15
4Graduate School of Science, Technology and Innovation, Kobe University, Kobe, Japan 16
5Center for Epidemiologic Research in Asia, Shiga University of Medical Science, Otsu, 17
Japan 18
6Department of Human Resource Development of Dialysis Therapy for Kidney Disease, 19
Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, 20
Okayama, Japan 21
22 23
Running title: Estimation of daily urinary sodium/potassium ratio in patients with CKD 24
Corresponding author: Haruhito A. Uchida, MD, PhD 25
Department of Chronic Kidney Disease and Cardiovascular Disease, Okayama University 26
Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, 2-5-1 Shikata-cho 27
Kita-ku, Okayama, 700-8558, Japan 28
29
E-mail address: [email protected] 30
31
Abstract 32
This study aimed to clarify the relationship between repeated measurements of casual (spot) 33
and 24-hour urinary sodium-to-potassium (Na/K) ratios in patients with chronic kidney 34
disease (CKD). A total of 61 inpatients with CKD, 31 in stage 1-3 (eGFR [estimated 35
glomerular filtration rate] ≥ 30 ml/min/1.73m2) and 30 in stage 4-5 (eGFR<30 36
ml/min/1.73m2), aged 20 to 85 consuming a low-sodium diet (NaCl [sodium chloride] 6 37
g/day) were recruited. Urinary Na, K, and Na/K ratios were measured in both casual urine 38
samples and 2-day, 24-hour urine samples, and then analyzed by correlation and 39
Bland-Altman analyses. Mean 24-hour urine Na/K ratio was higher in participants in stage 40
4-5 (5.1) than in participants in stage 1-3 (4.1) CKD. Casual urine Na/K ratio was strongly 41
correlated with 2-day, 24-hour urine Na/K ratio by sampling 4 casual urine specimens every 42
morning and evening in participants in stage 1-3 (r=0.69-0.78), but not in stage 4-5 43
(r=0.12-0.19). The bias for mean Na/K ratio between 2-day, 24-hour urine and the 4 casual 44
urine sampling ranged from -0.86 to 0.16 in participants in stage 1-3, and the quality of 45
agreement for the mean of this casual urine sampling was similar to that of sampling 8 casual 46
urine samples for estimating 2-day, 24-hour values. Methods using repeated casual urine 47
Na/K ratios may provide a reasonable estimation of 24-hour urine Na/K ratio in normotensive 48
and hypertensive as well as individuals with stage 1-3, but not stage 4-5 CKD.
49
(236 words) 50
Introduction 52
International guidelines for managing chronic kidney disease (CKD) have focused 53
on dietary modifications of individual nutrients [1-2]. Lowering sodium (Na) intake is 54
important to delay the progression of CKD to end-stage renal disease (ESRD) and reduce 55
cardiovascular disease (CVD) risk [1-2]. Dietary potassium (K) restriction is advised for 56
those with impaired kidney function in advanced stage CKD to avoid hyperkalemia [1-2].
57
Thus, diet therapy is essential for many patients with CKD.
58
Individual awareness of dietary intake remains poor as there remains a large gap 59
between recommended and actual consumption of Na and K [3-7].To measure individual 60
dietary intake, repeated high quality 24-hour urine collection is required [8-14]; however, it is 61
neither easy nor practical to collect. Measurement of casual urine is much easier than 24-hour 62
urine collection; however, the most common used formulas for estimating 24-hour urine 63
excretions contain problematic biases in low and high Na/K consumption and less bias in 64
population studies. Thus, these estimates are more applicable to populations rather than 65
individuals. [13,15-19]. Since these formulas depend on other parameters such as body 66
weight and creatinine levels, these measurements make it difficult for patients to know their 67
values [15-17].
68
Emerging evidence on Na/K ratios show benefits for Na reduction and K increase 69
compared to Na and K separately [20].Epidemiological studies suggest that urinary Na/K 70
ratio may be a more superior metric than Na and K in relation to blood pressure and CVD 71
risks [20-27]. Recent reports suggest elevated urinary Na or high Na with low K, i.e.
72
increased urinary Na/K ratio may be a risk factor of progression to CKD and greater 73
estimated glomerular filtration rate (eGFR) decline [1-2,28]. Self-monitoring devices for 74
urinary Na/K ratio provide prompt onsite feedback, and their evaluation supports an 75
individual approach for measuring Na and K changes [29]. Higher correlations are observed 76
for the casual Na/K ratio versus individual casual Na or K when compared with the 24-hour 77
urine values [30]. Casual urine measurements provide reliable estimates of 24-hour urine 78
value with less bias for Na/K ratio for not only population estimates, but also for individual 79
normotensive and hypertensive subjects, especially in repeated measurements [30-32].
80
However, these findings have not been validated in CKD patients. The present study aimed to 81
assess utility of casual urine specimens to estimate 24-hour urinary Na/K ratio in individuals 82
with CKD.
83
Methods 84
Participants and Measurements 85
CKD was defined as follows [1]. 1) Structural or functional abnormalities, defined as 86
abnormal findings on histological examination, urinalysis, biochemical examination, or 87
imaging studies for a duration of 3 months or longer regardless of eGFR. 2) eGFR < 60 88
mL/min/1.73m2 regardless of the primary disease, using the Modification of Diet in Renal 89
Disease Study equation [33].
90
A total of 61 inpatients with CKD were recruited at Okayama University hospital.
91
Patient characteristics were: 35 men and 26 women, aged 20 to 85 years, 31 in stage 1-3 92
(eGFR ≥ 30 ml/min/1.73m2) and 30 in stage 4-5 (eGFR < 30 ml/min/1.73m2) consuming a 93
low-Na diet (NaCl 6 g/day). Laboratory data, such as blood urea nitrogen (BUN), creatinine 94
(Cr), eGFR, uric acid (UA), Na, K, serum total cholesterol concentration, urinary 95
N-acetyl-β-D-glucosaminidase (NAG) and urinary β2 microglobulin (β-2MG) were measured 96
before study enrollment in the University Hospital (using Bio Majesty JCA-BM8040G and 97
JCA-BM6050, JEOL, Tokyo, Japan). Baseline blood pressure was measured in patients’
98
room by clinical staff using automated sphygmomanometer after 5 minutes of resting in a 99
sitting position. Body mass index was calculated as weight divided by height squared [kg/m2].
100
Primary kidney disease, prescriptions, and complications from each patient were obtained in 101
accordance with physician’s charts. Patients unable to collect urine samples or who were 102
recognized inappropriate for this study by attending physicians were excluded.
103
Participants were instructed to collect and measure their total urine volume with a 104
standardized measuring cup at each void during 2 consecutive days as a 24-hour urine sample.
105
In addition, they collected casual urine at 4 points per day (first voids after rising, breakfast, 106
lunch and dinner) during 2 consecutive days, unless urine collection was unsuccessful or 107
contaminated by feces. A 24-hour urine collection began at 7:00 AM and ended at 6:59 AM 108
the next morning. If participants declared that they failed to complete their urine collections, 109
they asked to repeat the process. Na and K concentrations (mmol/L) of casual and 24-hour 110
urine samples were measured at Okayama University hospital. The Na/K ratio of casual urine 111
samples was calculated using Na and K concentrations from each casual urine sample; while 112
the 24-hour urinary Na/K ratio was calculated using 24-hour Na and K urinary excretion 113
values. The mean Na/K ratio in casual urine was calculated for the first day or 2 consecutive 114
days at each or several time points per day.
115
116
Ethics 117
This study followed the Declaration of Helsinki (seventh revision, 2013) on medical 118
protocol and ethics. The ethics committees of Okayama University Institutional Review 119
Board (accredited ISO9001/2000), Okayama, Japan, approved this protocol, #2771. Written 120
informed consents were obtained from all patients.
121
122
Statistical Analysis 123
Mean Na/K ratio in casual urine samples were calculated from concentrations of Na 124
and K for the first day, and combined 2 days after beginning urine collection from each 125
individual. Pearson’s correlation coefficients for Na/K ratio were calculated to examine 126
correlation between specific values for casual urine and corresponding mean values for 127
24-hour urine samples over a 2-day period as the gold standard. For example, correlation 128
coefficients for Na/K ratio of casual urine sampled atfirst void after rising (value of the first 129
1 day and mean of combined 2 days) were made with the 2-day, 24-hour urine Na/K ratio.
130
The calculations were performed for daily casual urine specimens of first voids after rising, 131
breakfast, lunch and dinner samples.
132
Agreement between the casual urine Na/K ratio and 2-day, 24-hour urine Na/K ratio 133
was examined using the method proposed by Bland and Altman [34]. The “agreement”, 134
estimated by mean difference (bias) and the 95% of difference (defined by the width between 135
upper and lower limits of agreement [mean difference ± 1.96 × standard deviation of 136
difference]) between the casual urine method and 2-day, 24-hour urine method were 137
evaluated.
138
Correlation and agreement were also compared with mean Na/K ratios of multiple 139
casual urine sampling in a day and 2-day, and 2-day 24-hour urine. For example, correlation 140
and agreement analyses for mean casual urine Na/K ratio sampled from first voids after rising 141
and dinner (value of the first 1 day and mean of all 2 days) were made with mean 2-day, 142
24-hour urine Na/K ratios.
143
Results 144
The basic characteristics and urinary findings of the study participants are shown in 145
Table 1. The mean age of participants was 64.3 ± 15.0 years. Twenty-six participants (42.6%) 146
were women. Mean 24-hour urinary volume for 2 days was 1,857 ± 721 mL. Mean 24-hour 147
Na excretion over 2 days was 93.3 ± 44.0 mmol/24 h; whereas mean 24-hour K excretion was 148
22.5 ± 9.7 mmol/24 h. Mean 24-hour Na and K excretion was lower in participants in stage 149
4-5 (Na: 87.5 ± 32.9 mmol/24h, K: 18.8 ± 7.4 mmol/24h) than in participants in stage 1-3 150
(Na: 99.0 ± 52.2 mmol/24h, K: 26.1 ± 10.4 mmol/24h). Mean Na/K ratio of 24-hour urine 151
was 4.6 ± 2.4. Mean 24-hour urine Na/K ratio was higher in participants in stage 4-5 (5.1 ± 152
2.3) than in participants in stage 1-3 (4.1 ± 2.3). Casual urine Na concentration and casual 153
urinary Na/K ratios were highest in the first void after rising, but casual urine K concentration 154
was lowest. Na concentration of casual urine collections at all 4 points (first voids after rising, 155
breakfast, lunch and dinner) showed higher values than the 24-hour value, and its value 156
decreased from first void after rising toward first void after dinner. Casual urinary Na/K ratio 157
was highest in first void after rising and lowest in the first void after dinner.
158
Correlation coefficients of the casual urine Na/K ratio with 2-day, 24-hour Na/K 159
ratio in the 61 individuals are shown in Table 2. Correlation between casual urine Na/K ratio 160
and 2-day, 24-hour Na/K ratios generally became stronger as the number of days and the 161
daily casual urine sampling frequency increased in participants in stage 1-3; however, not in 162
stage 4-5 CKD. Mean Na/K ratio of all 8 casual urine specimens obtained in 2 days strongly 163
correlated with 2-day, 24-hour Na/K ratio in participants in stage 1-3 (r=0.79), but not in 164
stage 4-5 CKD (r=0.16). In participants in stage 1-3, correlation between 2-day, 24-hour 165
urinary Na/K ratios and mean Na/K ratios of 4 specimens (by sampling 2 casual urine 166
specimens per day for 2 days) were strong (r=0.69-0.78), but not as strong as correlation 167
between 2-day, 24-hour urinary Na/K ratios and mean Na/K ratios of 2 specimens (by 168
sampling 1 casual urine specimens per day for 2 days: r=0.58-0.72, and by sampling 2 casual 169
urine specimens in a day: r=0.47-0.77). In terms of the time of casual urine collectionwhen 170
sampling 2 casual urine specimens per day, combination of specimens sampled in the 171
morning and evening (first voids after rising and dinner or first voids after breakfast and 172
dinner) showed generally higher correlation coefficients as compared to the other 173
combinations (Table 2).
174
Because the mean Na/K ratio of casual urine in participants in stage 1-3 CKD 175
correlated more strongly with the 24-hour Na/K ratio for 2 days, assessment of agreement by 176
the Bland-Altman method was performed for the Na/K ratio in participants in stage 1-3 177
(Table 2). Since combinations of casual urine sampling of all 8 specimens, and 4 specimens 178
sampled in the morning and evening (first voids after rising and dinner or first voids after 179
breakfast and dinner, sampled for 2 days) showed generally higher correlation with 2-day, 180
24-hour urine collections than other combinations. Assessment of agreement was analyzed 181
for these combinations as these showed generally better agreement quality as compared to 182
others (Table 2). The bias for mean Na/K ratio between all 8 casual urine specimens and 183
2-day, 24-hour urine was -0.35 and 95% of difference was 5.75 (lying between -3.23 and 184
2.52), whereas bias ranged from -0.35 and 95% of difference ranged 4.84 (lying between 185
-2.77 and 2.07) for mean Na/K ratio between 4 casual urine specimens (first voids after rising 186
and dinner for 2 days) and 2-day, 24-hour urine (Figure 1).
187
Discussion 188
Main findings from the present study are that mean Na/K ratios of more than 4 189
casual urine samples (mainly sampled at morning and evening for 2-days) and 2-day, 24-hour 190
urine showed high correlation and reasonable agreement quality in patients with 191
mild-to-moderate stage CKD, which were consistent with previous findings in normotensive 192
and hypertensive individuals [31-32]. Thus, assessment of multiple casual urine specimens 193
may serve as a reasonable estimate of daily urinary Na/K ratios for individuals with 194
mild-to-moderate stage CKD.
195
Valid estimations of Na intake are challenging since random and systematic errors 196
are common [10,13,18-19]. To reduce both random errors from high day-to-day variability of 197
Na within an individual and systematic error due to incomplete urine collection, the use of 198
high quality repeated 24-hour urine collection has become the gold standard for estimating 199
individual daily Na intake [8-14]. However, collecting high quality repeated 24-hour urine is 200
a substantial burden for study participants [13]. Casual urine is easier to measure and 201
suboptimal methods have been proposed to estimate 24-hour urine values [15-17]; however, 202
the most commonly used formulas for estimating 24-hour urine Na excretion contain 203
problematic bias (overestimates in the low salt ranges and underestimates in the high salt 204
ranges) which lead to incorrect conclusions in association studies between salt intake and 205
CVD [18-19]. Since these formulas aim for estimating population mean values in 206
epidemiological studies, applying these formulas in clinical practices for individual estimates 207
results in gross measurement errors [15-17].
208
High correlation (r=0.80-0.88) and reasonable agreement is observed between the 209
mean value of the 4-7 repeated casual urine Na/K ratio and the 7-day, 24-hour urine Na/K 210
ratio in normotensive and hypertensive individuals [31-32]. There are several benefits for 211
using the casual urine estimate to determine the 24-hour urine Na/K ratio. These benefits 212
include, estimation is independent of urine volume, creatinine excretion and body weight;
213
self-monitoring devices provide prompt feedback [29]; and repeated random sampling 214
minimizes systemic error caused by diurnal and day-to-day variation of Na/K ratios with less 215
bias from low to high salt range in meals [10,30-32,35]. Thus, our analyses here provide a 216
reliable method to identify individual Na/K ratio levels, and support Na reduction and K 217
augmentation in individuals with CKD. However, this method is only applicable in 218
individuals with CKD in stage 1-3, but not in stage 4-5. By virtue of its ease and convenience, 219
repeated casual urine Na/K ratio measurement can support self-monitoring of an individual’s 220
Na/K ratio even at home without substantial burden. It is reasonable to infer that our findings 221
may provide a useful method to screen mild-to-moderate CKD patients with high urinary Na 222
or high Na with low K in order to prevent further eGFR decline and progressive CKD stages.
223
The greater correlation and stronger agreement quality among repeated casual and 224
7-day 24-hour urine Na/K ratio in normotensive and hypertensive individuals were 225
previously explained by the diurnal variation of urinary Na/K ratio [31-32,35]. Thus, diurnal 226
variation of urinary Na/K ratio in individuals with CKD in stage 1-3 may be similar to that of 227
normotensive and hypertensive individuals, but not in stages 4-5. The results that the highest 228
Na/K ratio observed in first void after rising and decreased Na/K ratio observed in first void 229
after breakfast were similar to the previous findings in normotensive and hypertensive 230
individuals [31-32]; however, lower urinary Na/K ratio compared to the 24-hour value in the 231
first void after dinner was different. The results of casual urine Na and K concentrations 232
measured at 4 points suggest that CKD individuals may show similar diurnal variation pattern 233
for urinary K concentration compared to normotensive and hypertensive individuals, but not 234
for urinary Na concentration [35]. Casual urinary Na and K concentrations have the lowest 235
value in first void after rising, and increase their values toward late afternoon in normotensive 236
and hypertensive individuals [35]; however, Na concentration tended to decrease its value 237
from first void after rising to first void after dinner in the present study. Thus, lower Na/K 238
ratio in daytime toward night hours might be one of the specific patterns of urinary Na/K 239
ratio in individuals with CKD. Further investigations are needed to assess diurnal variation of 240
urinary electrolytes in individuals with CKD.
241
The kidney functions precisely to regulate serum Na and K concentration [36-37].
242
When proximal tubules are damaged, distal tubules compensate for loss of Na reabsorption in 243
the proximal tubule; in addition, collecting ducts also resorb excess Na in urine [36-37].
244
Kidney function is usually estimated by serum creatinine concentration or glomerular 245
filtration rate; however, these values do not identify the precise site of pathological 246
abnormality (for example the glomerular injury, the tubular disorder, or co-existence of these 247
damages) or the degree of damage. In addition, many patients with CKD stage 4-5 are 248
prescribed drugs that effect regulation of sodium and potassium excretion, such as diuretics, 249
potassium sparing drugs, potassium lowering drugs and sodium hydrogen carbonate. Thus, 250
estimation of sodium and potassium excretion in urine at various times is complicated 251
especially in patients with CKD stage 4-5.
252
In the present study, participants who also had nephritis and nephrosclerosis and/or 253
taking loop diuretics usually had a more severe stage of CKD. Thus, factors among CKD 254
stages, nephritis, nephrosclerosis and specific medications may act as confounders, so 255
cautious interpretations of the results are needed for these individuals. However, the power of 256
this observational study was limited since it was not planned to make causal inference.
257
Further investigations are needed to identify the diurnal variation of urinary Na/K ratio and 258
the factors contributing to the weaker correlation among 2-day 24-hour and multiple casual 259
urine Na/K ratio in these subgroups.
260
A limitation of this study was the relatively small sample size that was monitored 261
only in inpatients consuming a low Na diet. Hence, the data obtained here did not reflect 262
actual dietary consumptions at home which may have a broader range of Na and K intake.
263
Thus, the correlation analyses might have underestimated in this study.
264
In conclusion, repeated casual urine Na/K ratio measurements may provide a 265
reasonable estimate of 24-hour urine Na/K ratio in moderate stage CKD individuals, as well 266
as normotensive and hypertensive individuals; however, not in advanced stage CKD 267
individuals.
268
269
Acknowledgement 270
This study was conducted by Okayama University hospital in cooperation with the 271
Department of Public Health, Shiga University of Medical Science. HAU and TI contributed 272
to the design of the study. HAU, YO, and AK participated in data collection. HAU, TI and 273
YO participated in data analysis and contributed to manuscript drafting. All authors 274
participated in critical revision of the manuscript. All authors approved the final version of 275
the manuscript for submission. The authors thank Mrs. Debra L. Rateri (University of 276
Kentucky) for skillful English editing.
277
278
Source of funding 279
This work was supported by a scientific research grants from Japanese Study Group for 280
Physiology and Management of Blood Pressure.
281
282
Conflict of Interest 283
Dr. Haruhito A. Uchida belongs to the Department of Chronic Kidney Disease and 284
Cardiovascular Disease which is endowed by Chugai pharmaceutical, MSD, Boehringer 285
Ingelheim, and Kawanishi Holdings. Dr. Toshiyuki Iwahori was an employee of OMRON 286
HEALTHCARE Co., Ltd. until March, 2018. Dr. Hirotsugu Ueshima served as a consultant 287
for this project. Dr. Katsuyuki Miura received a research grant from OMRON 288
HEALTHCARE Co., Ltd.
289
290
Summary Table 291
What is known about topic?
292
Casual urine sampling provides reliable estimates of 24-hour urine values with less bias for 293
Na/K ratio not only for population estimates, but also estimates in normotensive and 294
hypertensive individuals 295
What this study adds?
296
This study demonstrates that repeated casual urine Na/K ratio measurements may provide a 297
reasonable estimate of 24-hour urine Na/K ratios, in patients with CKD stage 1 to 3 as well as 298
normotensive and hypertensive individuals; however, this finding does not apply to patients 299
with CKD stage 4 to 5.
300
301
References 302
1. Levey AS, Coresh J, Balk E, Kausz AT, Levin A, Steffes MW, et al. National Kidney 303
Foundation. National Kidney Foundation practice guidelines for chronic kidney disease:
304
evaluation, classification, and stratification. Ann Intern Med. 2003 Jul 15;139(2):137-47.
305
Erratum in: Ann Intern Med. 2003 Oct 7;139(7):605.
306
2. Kidney Disease Outcomes Quality Initiative (K/DOQI). K/DOQI clinical practice guidelines 307
on hypertension and antihypertensive agents in chronic kidney disease. Am J Kidney Dis.
308
2004 May;43(5 Suppl 1):S1-290.
309
3. Brown IJ, Tzoulaki I, Candeias V, Elliott P. Salt intakes around the world: implications for 310
public health. Int J Epidemiol. 2009 Jun;38(3):791-813.
311
4. Drewnowski A, Rehm CD, Maillot M, Mendoza A, Monsivais P. The feasibility of meeting 312
the WHO guidelines for sodium and potassium: a cross-national comparison study. BMJ 313
Open. 2015 Mar 20;5(3):e006625.
314
5. WHO Guideline: Sodium intake for adults and children. In. Geneva: World Health 315
Organization (WHO); 2012.
316
6. WHO Guideline: Potassium intake for adults and children. In. Geneva: World Health 317
Organization (WHO); 2012.
318
7. Whelton PK, Appel LJ, Sacco RL, Anderson CA, Antman EM, Campbell N, et al. Sodium, 319
blood pressure, and cardiovascular disease: further evidence supporting the American 320
Heart Association sodium reduction recommendations. Circulation. 2012 Dec 321
11;126(24):2880-9.
322
8. Hunter D. Biochemical indicators of dietary intake. In: Willet W, Nutritional Epidemiology, 323
Second Edition. Oxford University Press: New York, NY, USA, 1998, pp 174-243.
324
9. Pietinen PI, Findley TW, Clausen JD, Finnerty FA, Altschul AM. Studies in Community 325
Nutrition: Estimation of Sodium Output. Prev. Med. 1976 Sep;5(3):400-407.
326
10. Liu K, Stamler J. Assessment of sodium intake in epidemiological studies on blood 327
pressure. Ann Clin Res. 1984;16 Suppl 43:49-54.
328
11. Cogswell ME, Mugavero K, Bowman BA, Frieden TR. Dietary Sodium and Cardiovascular 329
Disease Risk - Measurement Matters. N Engl J Med. 2016 Aug 11;375(6):580-6.
330
12. Dennis B, Stamler J, Buzzard M, Conway R, Elliott P, Moag-Stahlberg A, et al.;
331
INTERMAP Research Group. INTERMAP: the dietary data--process and quality control.
332
J Hum Hypertens. 2003;17:609-622.
333
13. Cobb LK, Anderson CA, Elliott P, Hu FB, Liu K, Neaton JD, et al.; American Heart 334
Association Council on Lifestyle and Metabolic Health. Methodological issues in cohort 335
studies that relate sodium intake to cardiovascular disease outcomes: a science advisory 336
from the American Heart Association. Circulation. 2014 Mar 11;129(10):1173-86.
337
14. Lerchl K, Rakova N, Dahlmann A, Rauh M, Goller U, Basner M, et al. Agreement 338
between 24-hour salt ingestion and sodium excretion in a controlled environment.
339
Hypertension. 2015 Oct;66(4):850-7.
340
15. Kawasaki T, Itoh K, Uezono K, Sasaki H. A simple method for estimating 24 h urinary 341
sodium and potassium excretion from second morning voiding urine specimen in adults.
342
Clin Exp Pharmacol Physiol. 1993 Jan;20(1):7-14.
343
16. Tanaka T, Okamura T, Miura K, Kadowaki T, Ueshima H, Nakagawa H, et al. A simple 344
method to estimate populational 24-h urinary sodium and potassium excretion using a 345
casual urine specimen. J Hum Hypertens 2002;16:97-103.
346
17. Brown IJ, Dyer AR, Chan Q, Cogswell ME, Ueshima H, Stamler J, et al. INTERSALT 347
Co-Operative Research Group. Estimating 24-hour urinary sodium excretion from casual 348
urinary sodium concentrations in Western populations: the INTERSALT study. Am J 349
Epidemiol. 2013 Jun 1;177(11):1180-1192.
350
18. Polonia J, Lobo MF, Martins L, Pinto F,Nazare J. Estimation of populational 24-h urinary 351
sodium and potassium excretion from spot urine samples: evaluation of four formulas in 352
a large national representative population. J Hypertens. 2017 Mar;35(3):477-486.
353
19. Huang L, Crino M, Wu JH, Woodward M, Barzi F, Land MA, et al. Mean population salt 354
intake estimated from 24-h urine samples and spot urine samples: a systematic review 355
and meta-analysis. Int J Epidemiol. 2016 Feb;45(1):239-50.
356
20. Iwahori T, Miura K, Ueshima H. Time to Consider Use of the Sodium-to-Potassium Ratio 357
for Practical Sodium Reduction and Potassium Increases. Nutrients. 2017 Jul 5;9(7). pii:
358
E700.
359
21. INTERSALT Co-operative Research Group. INTERSALT: an international study of 360
electrolyte excretion and blood pressure. Results for 24-hr urinary sodium and potassium 361
excretion. BMJ. 1988 Jul 30;297(6644):319-328.
362
22. Stamler J, Rose G, Stamler R, Elliott P, Dyer A, Marmot M. INTERSALT study findings.
363
Public health and medical care implications. Hypertension 1989 Nov;14(5):570-577.
364
23. Tzoulaki I, Patel CJ, Okamura T, Chan Q, Brown IJ, Miura K, et al. A nutrient-wide 365
association study on blood pressure. Circulation. 2012 Nov 20;126(21):2456-2464.
366
24. Perez V, Chang ET. Sodium-to-Potassium Ratio and Blood Pressure, Hypertension, and 367
Related Factors. Adv Nutr. 2014 Nov 14;5(6):712-741.
368
25. Cook NR , Obarzanek E, Cutler JA, Buring JE, Rexrode KM, Kumanyika SK, et al.; Trials 369
of Hypertension Prevention Collaborative Research Group. Joint effects of sodium and 370
potassium intake on subsequent cardiovascular disease: the Trials of Hypertension 371
Prevention follow-up study. Arch Intern Med. 2009 Jan 12;169(1):32-40.
372
26. Cook NR, Appel LJ, Whelton PK. Sodium Intake and All-Cause Mortality Over 20 Years 373
in the Trials of Hypertension Prevention. J Am Coll Cardiol. 2016 Oct 374
11;68(15):1609-1617.
375
27. Cook NR , Cutler JA, Obarzanek E, Buring JE, Rexrode KM, Kumanyika SK, et al. Long 376
term effects of dietary sodium reduction on cardiovascular disease outcomes:
377
observational follow-up of the trials of hypertension prevention (TOHP). BMJ. 2007 Apr 378
28;334(7599):885-8. Epub 2007 Apr 20.
379
28. Kieneker LM, Bakker SJ, de Boer RA, Navis GJ, Gansevoort RT, Joosten MM. Low 380
potassium excretion but not high sodium excretion is associated with increased risk of 381
developing chronic kidney disease. Kidney Int 2016; 90:888–896.
382
29. Iwahori T, Ueshima H, Ohgami N, Yamashita H, Miyagawa N, Kondo K, et al.
383
Effectiveness of a self-monitoring device for urinary sodium/potassium ratio on dietary 384
improvement in free-living adults: a randomized controlled trial. J Epidemiol. 2018 Jan 385
5;28(1):41-47 386
30. Iwahori T, Miura K, Ueshima H, Chan Q, Dyer AR, Elliott P, et al.; INTERSALT Research 387
Group. Estimating 24-hour urinary sodium/potassium ratio from casual (“spot”) urinary 388
sodium/potassium ratio: The INTERSALT Study. Int J Epidemiol. 2017 Oct 389
1;46(5):1564-1572 390
31. Iwahori T, Ueshima H, Miyagawa N, Ohgami N, Yamashita H, Ohkubo T, et al. Six 391
sodium/potassium ratio as compared to 7-day 24-hr urine collections. Hypertension Res.
393
2014 Aug;37(8):765-771.
394
32. Iwahori T, Ueshima H, Torii S, Saito Y, Fujiyoshi A, Ohkubo T, et al. Four to seven 395
random casual urine specimens are sufficient to estimate 24-hr urinary 396
sodium/potassium ratio in individuals with high blood pressure. J Hum Hypertens. 2016 397
May;30(5):328-334 398
33. Matsuo S, Imai E, Horio M, Yasuda Y, Tomita K, Nitta K, et al.; Collaborators developing 399
the Japanese equation for estimated GFR. Revised equations for estimated GFR from 400
serum creatinine in Japan. Am J Kidney Dis 2009;53:982-92.
401
34. Bland J, Altman DG. Statistical methods for assessing agreement between two methods 402
of clinical measurement. Lancet 1986;1(8476):307-10.
403
35. Iwahori T, Ueshima H, Torii S, Saito Y, Kondo K, Tanaka-Mizuno S, et al. Diurnal 404
variation of urinary sodium-to-potassium ratio in free-living Japanese individuals.
405
Hypertens Res. 2017 Jul;40(7):658-664.
406
36. Koeppen BM, Stanton BA. Renal Transport Mechanisms: NaCl and Water Reabsorption 407
Along the Nephron. In: Koeppen BM, Stanton BA, Renal Physiology, Fifth Edition.
408
Elsevier Inc.: Amsterdam, Netherlands, 2013, pp 45-71.
409
37. Dhondup T, Qian Q. Electrolyte and Acid-Base Disorders in Chronic Kidney Disease and 410
End-Stage Kidney Failure. Blood Purif. 2017;43(1-3):179-188.
411
412
Figure legends 413
Figure 1 414
Plot of Na/K ratio of casual urine versus 2-day 24-hour urine, and Bland-Altman plot in CKD 415
inpatients. (○ inpatients with CKD in stage 1-3, x: inpatients with CKD in stage 4-5) 416
Pearson’s correlation coefficient between mean casual urine Na/K ratio of 4 specimens 417
sampled twice a day (at first voids after rising and dinner) for 2 days and 2-day 24-hour urine 418
Na/K ratio ranged 0.37 for the overall 61 inpatients with CKD, 0.75 for 31 inpatients with 419
CKD in stage 1-3 and 0.12 for 30 inpatients with CKD in stage 4-5 (A). The bias between the 420
2-day casual urine Na/K ratio sampled twice a day (at first voids after rising and dinner) and 421
2-day 24-hour Na/K ratio by Bland-Altman method was 0.02, with the limits of differences 422
10.14 lying between -5.05 and 5.09 in overall 61 inpatients with CKD, was -0.35, with the 423
limits of differences 4.84 lying between -2.77 and 2.07 in inpatients with CKD in stage 1-3, 424
and the bias 0.40with the limits of differences 13.57 lying between -6.38 and 7.19 in 425
inpatients with CKD in stage 4-5 (B). Pearson’s correlation coefficient between mean casual 426
urine Na/K ratio of 4 specimens sampled twice a day (at first voids after breakfast and dinner) 427
for 2 days and 2-day 24-hour urine Na/K ratio ranged 0.36 for the overall 61 inpatients with 428
CKD, 0.76 for 31 inpatients with CKD in stage 1-3 and 0.12 for 30 inpatients with CKD in 429
stage 4-5 (C). The bias between the 2-day casual urine Na/K ratio sampled twice a day (at 430
first voids after breakfast and dinner) and 2-day 24-hour Na/K ratio by Bland-Altman method 431
was 0.82, with the limits of differences 10.18 lying between -4.26 and 5.91 in overall 61 432
inpatients with CKD, was 0.16, with the limits of differences 5.93 lying between -2.81 and 433
3.12 in inpatients with CKD in stage 1-3, and the bias 1.51with the limits of differences 434
12.77 lying between -4.87 and 7.90 in inpatients with CKD in stage 4-5 (D). Pearson’s 435
correlation coefficient between mean casual urine Na/K ratio of all 8 specimens sampled 436
twice a day (at first voids after rising, breakfast, lunch and dinner) for 2 days and 2-day 437
24-hour urine Na/K ratio ranged 0.40 for the overall 61 inpatients with CKD, 0.79 for 31 438
inpatients with CKD in stage 1-3 and 0.16 for 30 inpatients with CKD in stage 4-5 (E). The 439
bias between the 2-day casual urine Na/K ratio sampled twice a day (at first voids after rising, 440
breakfast, lunch and dinner) and 2-day 24-hour Na/K ratio by Bland-Altman method was 441
0.28, with the limits of differences 9.99 lying between -4.71 and 5.28 in overall 61 inpatients 442
with CKD, was -0.35, with the limits of differences 5.75 lying between -3.23 and 2.52 in 443
inpatients with CKD in stage 1-3, and the bias 0.94with the limits of differences 12.61 lying 444
between -5.36 and 7.24 in inpatients with CKD in stage 4-5 (F).
445
ble 1-A. Basic characteristics of study participants stratified into CKD stages Variables CKD patients (n=61) Stage 1-3 (n=31) Stage 4-5 (n=30) Overall (n=61) meanStandard deviationmeanStandard deviationmeanStandard deviation Male / Female 18 / 1317 / 1335 / 26 Age (years) 61.6 16.9 67.1 12.3 64.3 15.0 Height (cm) 161.1 11.0 159.6 9.1 160.3 10.1 Weight (kg) 63.7 10.5 60.2 13.1 62.0 11.9 Body mass index (kg/m2 ) 24.7 4.7 23.6 4.5 24.2 4.6 Blood pressure (mmHg) 125 / 7313 / 9133 / 7720 / 12129 / 7617 / 11 Pulse (bpm) 71 9 68 10 70 10 BUN (mg/dl) 17.9 5.8 50.4 21.8 33.9 22.7 Cr (mg/dl) 0.98 0.27 4.04 1.51 2.49 1.87 eGFR (ml/min/1.73m2 ) 59.3 19.7 13.3 6.5 36.7 27.4 Serum Na (mmol/l) 138.6 1.9 138.6 3.38 138.6 2.7 Serum K (mmol/l) 4.0 0.4 4.34 0.7 4.17 0.6 T-chol (mg/dl) 195 41 171 49 183 46
2
Proteinuria (g/day) 1.48 2.43 2.43 2.55 1.95 2.51 U-NAG (U/l)11.64 11.84 6.966.259.389.77 U-β2MG (mg/l) 0.41 0.49 11.04 11.47 5.82 9.73 Diagnosis n % n % n % Diabetic nephropathy 14 45.2 10 33.3 24 39.3 Nephrosclerosis 1 3.2 12 40.0 13 21.3 Glomerulonephritis 13 41.9 6 20.0 19 31.1 Others 5 16.1 6 20.0 11 18.0 Complications Hypertension 21 67.7 28 93.3 49 80.3 Diabetes mellitus21 67.7 16 53.3 37 60.7 Dyslipidemia22 71.0 15 50.0 37 60.7 Hyperuricemia7 22.6 22 73.3 29 47.5
3
Drug treatment Loop 4 12.9 13 43.3 17 27.9 thiazide 3 9.7 4 13.3 7 11.5 tolvaptan0 0 1 3.3 1 1.6 Spironolactone1 3.2 1 3.3 2 3.3 DRI/ARB/ACE-I 26 83.9 17 56.7 43 70.5 Potassium lowering drug 3 9.7 8 26.7 11 18.0 Oral prednisolone 14 45.2 8 26.7 22 36.1 Sodium hydrogen carbonate 0 0 12 40.0 12 19.7 BUN : blood urea nitrogen, Cr : creatinine, eGFR : estimated glomerular filtrating ratio, U-NAG : N-acetyl-β-D glucosaminidase in urine, U-β2MG : microglobulin in urine, Urine RBC count : red blood cell count in urine, Na : sodium, K : potassium, AST : aspirate aminotransferase, ALT : alanine notransferase, G-GT : gamma glutamyltranspeptidase, T-cho : total cholesterol, TG :triglyceride, LDL : low-density lipoprotein cholesterol , HDL: ity lipoprotein cholesterol, FPG :fasting plasma glucose, HbA1c : Hemoglobin A1c, PRA : plasma renin activity, PAC: plasma aldosterone ncentration n values of 2 times of 24-hour urine.
4
ble 1-B. Urinary findings of study participants stratified into CKD stages Variables CKD patients (n=61) Stage 1-3 (n=31) Stage 4-5 (n=30) Overall (n=61) meanStandard deviationmeanStandard deviationmeanStandard deviation 24-hour urine volume (ml) 1878.5 725.9 1835.8 723.1 1857.5 721.8 24-hour Na excretion (mmol/24hr)
99.0 52.2 87.5 32.9 93.3 44.0 24-hour K excretion (mmol/24hr)
26.1 10.4 18.8 7.4 22.5 9.7 Na concentration (mmol/l) 24-hour urine 54.1 19.0 49.2 12.2 51.7 16.1 Spot urine First void after rising67.5 24.8 51.3 18.3 59.6 23.2
5
First void after breakfast64.7 26.6 50.2 19.8 57.5 24.5 First void after lunch 57.9 28.6 48.9 14.7 53.5 23.2 First void after dinner59.5 32.2 46.5 16.3 53.1 26.4 K concentration (mmol/l) 24-hour urine 14.9 6.4 11.2 4.8 13.0 5.9 Spot urine First void after rising16.9 9.4 9.8 4.7 13.4 8.2 First void after breakfast26.8 19.2 15.2 7.7 21.1 15.8 First void after lunch 22.4 18.7 16.3 10.1 19.4 15.4 First void after dinner25.2 14.7 14.8 7.3 20.1 12.7 Na/K ratio 24-hour urine 4.1 2.3 5.1 2.3 4.6 2.4 Spot urine First void after rising4.9 2.6 6.1 3.0 5.5 2.9
6
First void after breakfast3.8 3.2 3.9 2.1 3.9 2.7 First void after lunch 4.1 4.1 4.0 2.5 4.1 3.4 First void after dinner3.1 2.6 4.0 2.5 3.5 2.6 Na : sodium, K : potassium, # Mean values of 2 times of 24-hour urine.
7
ble 2. Correlation coefficients of casual urine Na/K ratio with 2-day 24-hour Na/K ratio in CKD inpatients stratified into CKD stages Time of casual urine
Number of days to calculate mean
24-hour Na/K ratioa CKD patients (n=61) Stage 1-3 (n=31)Stage 4-5 (n=30)Overall (n=61) Correlation coefficients
Agreement with Bland-Altman plot Correlation coefficients Agreement with Bland-Altman plotCorrelation coefficients
Agreement with Bland-Altman plot Bias 95% of differenceBias 95% of differenceBias 95% of difference First void after rising (1 specimen/day)
1 day 0.27-0.889.46 0.07-0.52 15.24 0.22 -0.71 12.55 2 days 0.58-1.247.15 0.13-0.66 14.86 0.32 -0.95 11.56 First void after breakfast (1 specimen/day)
1 day 0.540.367.800.121.4113.350.300.8810.99 2 days 0.72-0.227.86 0.131.5612.560.330.6610.94 First void after lunch (1 specimen/day)
1 day 0.54-0.6214.61 0.151.4113.510.260.3814.52 2 days 0.68-0.4911.54 0.211.3912.830.350.4412.65 First void after dinner (1 specimen/day)
1 day 0.700.369.750.081.4614.190.340.9012.23 2 days 0.660.546.450.091.4613.670.330.9910.70 First void after rising & dinner (2 specimens/day) 1 day 0.74-0.26 5.21 0.080.4714.14 0.34 0.10 10.60 2 days 0.75-0.35 4.84 0.120.4013.57 0.37 0.02 10.14
8
First void after breakfast & dinner (2 specimens/day)
1 day 0.770.365.880.111.4413.27 0.37 0.89 10.34 2 days 0.760.165.930.121.5112.77 0.36 0.82 10.18 First void after rising, breakfast, lunch and dinner (4 specimens/day)
1 day 0.74-0.205.83 0.120.9413.15 0.36 0.36 10.27 2 days 0.79-0.355.75 0.160.9412.61 0.40 0.28 9.99 a Means of all 2 days Na, sodium; K, potassium.
Figure1.
A B
r=0.37
r=0.75
r=0.12
Upper limit 5.09
Bias 0.02
Lower limit -5.05
Upper limit 2.07 Bias -0.35
Lower limit -2.77
Upper limit 7.19
Bias 0.40
Lower limit -6.38
〇:inpatients with CKD in stage 1-3
×:inpatients with CKD in stage 4,5
×:inpatients with CKD in stage 4,5
〇:inpatients with CKD in stage 1-3
〇:inpatients with CKD in stage 1-3
×:inpatients with CKD in stage 4,5
〇:inpatients with CKD in stage 1-3
×:inpatients with CKD in stage 4,5
Figure1.
C D
r=0.36
r=0.76
r=0.12
Upper limit 5.91
Bias 0.82
Lower limit -4.26
Upper limit 3.12 Bias 0.16
Lower limit -2.81
Upper limit 7.90
Bias 1.51
Lower limit -4.87
〇:inpatients with CKD in stage 1-3
×:inpatients with CKD in stage 4,5
〇:inpatients with CKD in stage 1-3
×:inpatients with CKD in stage 4,5
〇:inpatients with CKD in stage 1-3
×:inpatients with CKD in stage 4,5
〇:inpatients with CKD in stage 1-3
×:inpatients with CKD in stage 4,5
Figure1.
E F
r=0.40
r=0.79
r=0.16
Upper limit 5.28
Bias 0.28
Lower limit -4.71
Upper limit 2.52 Bias -0.35
Lower limit -3.23
Upper limit 7.24
Bias 0.94
Lower limit -5.36
〇:inpatients with CKD in stage 1-3
×:inpatients with CKD in stage 4,5
〇:inpatients with CKD in stage 1-3
×:inpatients with CKD in stage 4,5
〇:inpatients with CKD in stage 1-3
×:inpatients with CKD in stage 4,5
〇:inpatients with CKD in stage 1-3
×:inpatients with CKD in stage 4,5
