Académique Documents
Professionnel Documents
Culture Documents
DOI: 10.1111/jch.13610
Correspondence
Sang‐Ho Jo, MD, PhD, Division of Abstract
Cardiology, Department of Internal The objective of this study was to evaluate the association between sodium intake and
Medicine, Hallym University Sacred Heart
Hospital, 896 Pyeongchon‐dong, Dongan‐ blood pressure (BP) control in hypertensive patients taking antihypertensive medica‐
gu, Anyang‐si, Gyeonggi‐do, 14103, Korea. tions by using 24‐hour urine collection and 24‐hour ambulatory BP. This is a cross‐
Email: sophi5neo@gmail.com
and sectional community‐based study and conducted in 2011 and 2012. A total of 1128
Moo‐Yong Rhee, MD, PhD, Cardiovascular participants were recruited from five cities in Korea. Among them, 740 participants
Center, Dongguk University Ilsan Hospital,
27 Dongguk‐ro, Ilsandong‐gu, Goyang‐si, who had complete 24‐hour urine collection and valid 24‐hour ambulatory BP data were
Gyeonggi‐do, 10326, Korea. included in this study. Participants were divided into four groups: normotensives (NT,
Email: mooyong_rhee@dumc.or.kr
n = 441), untreated hypertensive patients (UTHT, n = 174), controlled hypertensive pa‐
Funding information
tients (CHT, n = 62), and uncontrolled hypertensive patients (UCHT, n = 63). UCHT and
This study was supported by the Research
Program funded by the Ministry of Food and CHT groups showed higher mean age than NT and UTHT groups. UCHT and UTHT
Drug Safety in 2011 (11162KFDA162) and
groups showed higher 24‐hour systolic BP (SBP) and diastolic BP (DBP) than NT and
2012 (12162MFDS103).
CHT groups. UCHT group had the highest level of 24‐hour urine sodium. Multivariate
analysis adjusted with age, gender, body mass index, estimated glomerular filtration
rate, and use of diuretics showed higher level of 24‐hour urine sodium in UCHT group
than that in CHT group. Multivariate logistic regression analysis revealed independent
association of the amount of 24‐hour urine sodium with uncontrolled BP in hyperten‐
sive patients on antihypertensive drug treatment. Higher level of 24‐hour urine sodium
excretion in uncontrolled hypertensive patients suggests that excessive sodium intake
could be associated with blunted BP lowering efficacy of antihypertensive medications.
1 | I NTRO D U C TI O N serum sodium, and potassium levels were measured after overnight
fasting for at least 8 hours.
Hypertension is a well‐known risk factor of cardiovascular disease, The 24‐hour urine collection was validated by both self‐reported
and poor blood pressure (BP) control is associated with increased urine diary and 24‐hour urine creatinine‐based determination. If self‐
risk of cardiovascular events.1 reported loss of a urine sample was more than 100 mL at a time or
Among risk factors of hypertension, salt intake is an import‐ more than once, or if the creatinine index [24‐hour urine creatinine,
ant lifestyle factor. Excessive sodium intake is associated with mg/dL / (21 × body weight)] was <0.7, the collected urine sample
elevated blood pressure (BP). 2 It can attribute to poor BP con‐ was considered to be incomplete for a 24‐hour period collection.13,15
trol despite the use of antihypertensive medications. However, The 24‐hour ambulatory BP was measured simultaneously with
3-7
only a few short‐term controlled trials with small sample size 24‐hour urine collection (Mobil‐O‐Graph; IEM GmbH).16 More
8,9
and cross‐sectional studies have evaluated the association be‐ than 70% of the attempted measurement, at least 14 measure‐
tween high sodium intake and poor BP control. These cross‐sec‐ ments during the daytime (9 am‐9 pm) and at least seven measure‐
tional studies measured sodium intake by dietary survey method ments at night time (00:00‐06:00), were regarded as satisfactory
and spot urine method known to have limitations.10,11 Moreover, measurement.17
they measured casual BP, not ambulatory BP known to have Hypertension was defined based on average of 24‐hour sys‐
stronger association with future cardiovascular events than ca‐ tolic BP (SBP) ≥130 mm Hg and/or diastolic BP (DBP) ≥80 mm Hg
sual BP.12 or current use of antihypertensive medication.18 The Institutional
Therefore, the objective of this study was to evaluate the asso‐ Review Board of each participating hospital approved the study
ciation between sodium intake and BP control in hypertensive pa‐ protocol.
tients taking antihypertensive medications by using 24‐hour urine
collection and 24‐hour ambulatory BP.
2.3 | Statistical analysis
n 441 174 62 63
a b c c
Age (y) 45.9 ± 10.3 49.8 ± 8.9 57.3 ± 8.0 53.9 ± 8.0 <0.001
Men, n (%) 121 (27.4) 118 (67.8) 26 (41.9)c 39 (61.9) <0.001
BMI, kg/m2 22.9 ± 2.8a 24.6 ± 3.0 b 25.0 ± 3.3b 26.3 ± 3.2c <0.001
Diabetes, n (%) 14 (3.2) 21 (12.1) 12 (19.4) 9 (14.3) <0.001
No. of antihypertensive medications
Average, n 1.51 1.67
1, n 31 27 0.515
2, n 27 28
3, n 2 6
4, n 1 1
Antihypertensive medications
Calcium channel blocker, n 32 32
Angiotensin receptor blockers, n 36 38
Angiotensin‐converting enzyme 2 3
inhibitors, n
Beta blockers, n 6 10
Diuretics, n 18 21
Alpha blockers, n 1 0
a b c b
24‐h SBP, mm Hg 110.6 ± 7.5 126.7 ± 9.0 114.4 ± 7.1 125.9 ± 7.1 <0.001
24‐h DBP, mm Hg 70.3 ± 6.3a 87.1 ± 6.0 b 72.0 ± 5.2a 86.7 ± 5.2b <0.001
Serum Na (mmol/dL) 141.2 ± 2.4 141.1 ± 2.3 141.7 ± 2.3 141.3 ± 2.5 0.345
Serum K (mmol/dL) 4.3 ± 0.4 4.3 ± 0.3 4.3 ± 0.4 4.2 ± 0.4 0.278
Serum creatinine (mg/dL) 0.74 ± 0.14 0.82 ± 0.16 0.82 ± 0.22 0.86 ± 0.18 <0.001
24‐h urine creatine (mg/24h) 1191.3 ± 434.8a 1376.1 ± 372.8b 1189.6 ± 337.2a 1426.6 ± 321.2b <0.001
24‐h urine volume (mL) 1489.6 ± 592.1 1565.2 ± 572.3 1535.8 ± 572.7 1691.5 ± 659.1 0.062
eGFR (mL/min/1.73 m2) 95.8 ± 16.7a 93.6 ± 15.6a 86.4 ± 16.1b 86.6 ± 14.7b <0.001
Total cholesterol, mg/dL 193.7 ± 34.5a,b 203.7 ± 36.7a 192.9 ± 37.3a,b 186.9 ± 35.3b 0.002
LDL cholesterol, mg/dL 119.3 ± 31.7a,b 127.2 ± 32.8a 117.3 ± 35.7a,b 112.9 ± 31.0 0.007
HDL cholesterol, mg/dL 59.2 ± 14.9a 52.1 ± 14.3b 54.8 ± 14.1a,b 49.8 ± 13.9b <0.001
Triglyceride, mg/dL 108.1 ± 69.6a 165.4 ± 109.0 b,c 147.8 ± 87.9b 184.8 ± 196.0 c <0.001
a b b b
Fasting glucose, mg/dL 93.7 ± 15.0 103.4 ± 27.5 103.5 ± 18.8 108.4 ± 28.8 <0.001
Note: Different alphabets represent significant difference between groups at alpha level of 0.05 by post hoc Tukey HSD analysis.
Abbreviations: BMI, body mass index; CHT, controlled hypertension group; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate
by Modification of Diet in Renal Disease (MDRD) formula; HDL, high‐density lipoprotein; K, potassium; LDL, low‐density lipoprotein; Na, sodium; NT,
normotensives group; SBP, systolic blood pressure; UCHT, uncontrolled hypertension group; UTHT, untreated hypertension group.
*P values obtained by analysis of variance or chi‐square test for difference between four groups.
than those of NT and UTHT groups. The BMI of UCTH group was the The UCHT group had the highest level of 24‐hour urine sodium
highest among all groups. The 24‐hour SBP and DBP of UTHT and among all groups in unadjusted and adjusted analyses (Table 2, P < 0.001
UCHT groups were significantly higher than those of NT and CHT and P = 0.023, respectively). The level of 24‐hour urine sodium in the
groups (P < 0.05). There was no significant difference in 24‐hour SBP UCHT group was higher than in the CHT group in adjusted analysis
and DBP between UTHT and UCHT groups. The number of antihy‐ (P = 0.048). There was no significant difference in 24‐hour urine po‐
pertensive drugs was 1.67 in UCHT group and 1.51 in CHT group, tassium and sodium‐to‐potassium ratio between groups. Multivariate
showing no difference between the two groups. There was no signif‐ logistic regression analysis revealed the independent association of the
icant difference in classes of antihypertensive drugs between CHT amount of 24‐hour urine sodium and age with uncontrolled BP in hy‐
and UCHT groups either. pertensive patients on antihypertensive drug treatment (Table 3).
|
1060 RHEE et al.
Abbreviations: CHT, controlled hypertension group; K, potassium; Na, sodium; Na/K, sodium‐to‐potassium ratio; NT, normotensives group; UCHT,
uncontrolled hypertension group; UTHT, untreated hypertension group.
*P value by analysis of variance, and different alphabets represent significant difference at alpha level of 0.05 by post hoc Tukey HSD analysis.
**P value by analysis of covariance between four groups and adjusted for age, gender, body mass index, estimated glomerular filtration rate, and use
of diuretics.
***P value by analysis of covariance between uncontrolled hypertension group vs controlled hypertension group, adjusted for age, gender, body mass
index, estimated glomerular filtration rate, and use of diuretics.
attributing factor to poor BP control in patients who are taking an‐ HDL cholesterol 0.983 (0.954‐1.014) 0.272
tihypertensive medications. However, majority of intervention stud‐ 24‐h urine sodium 1.008 (1.001‐1.015) 0.027
ies have evaluated the effect of high sodium intake on BP elevation, Note: Multivariate logistic regression analysis. Age, gender, body mass
but not on BP control, because they evaluated the effect of sodium index, use of diuretics, fasting blood glucose, HDL cholesterol, and 24‐h
intake on BP in patients not taking antihypertensive drugs. urine sodium were included in the model.
Abbreviation: HDL, high‐density lipoprotein.
Only a few studies have evaluated the effect of high sodium in‐
take on BP control.3-9 Intervention trials for the effect of sodium
intake in patients taking antihypertensive medication have been relatively small population. A real world may be different from the
performed in short‐term studies, and most of them were performed environment of controlled trial although RCTs provide the best evi‐
with small sample sizes. In patients with treatment‐resistant hy‐ dence. Long‐term compliance to low sodium diet in the real world is
pertension, lowering of sodium intake from 250 mmol/d for 7 days quite different from that in RCTs.22 Cohort study or cross‐sectional
to 50 mg/d for 7 days has reduced SBP by 22.7 mm Hg and DBP epidemiologic study may be close to real world. A relatively long‐
by 9.1 mm Hg. 19
Education for salt intake reduction in 150 treated term study of lifestyle education including salt intake reduction has
hypertensive patients reduced SBP by 5.3 mm Hg and DBP by failed to reduce sodium intake and BP in patients with antihyper‐
2.9 mm Hg after 2 months. The reduction was significantly larger tensive drugs treatment. 23 Only a few cross‐sectional studies have
than that in the control group. 4 evaluated the association of sodium intake on BP control. Chinese
Sodium intake reduction can potentiate effects of antihyper‐ hypertensive patients consuming more than 6 g/d salt which was
tensive drugs. Such effects are more obvious for renin‐angioten‐ estimated by using food frequency questionnaire had lower BP con‐
sin‐aldosterone system‐related drugs. 3,5-7
Regarding mechanisms trol rate compared with patients consuming less than 6 g/d salt.8
by which low sodium intake can potentiate BP lowering effects of In the analysis of Korean National Health and Nutritional Survey,
5
antihypertensive drugs, Weir et al have suggested that lower so‐ among patients with cardiovascular disease, hypertension group
dium intake may reverse to angiotensin II–dependent form of hyper‐ (≥140/90 mm Hg) had higher 24‐hour urinary sodium excretion
tension, that is, more susceptible to angiotensin‐converting enzyme which was estimated from spot urine sodium, compared with the
inhibitor. High sodium intake can reduce the bioavailability of antihy‐ normal BP group.9
pertensive drugs20,21 by modulating intestinal transporters. Reduced The distinguishing feature of our study from previous studies is
bioavailability of antihypertensive drugs by high sodium intake may the measurement method of salt intake and BP. We used 24‐hour
attenuate the BP lowering efficacy of ingested drugs. urine collection method in the estimation of salt intake to avoid er‐
Although randomized controlled trials (RCTs) showed beneficial rors of dietary survey method10,24 and spot urine method.11,25 The
effect of sodium intake reduction in treatment of hypertension with 24‐hour urine collection method is widely regarded as the gold stan‐
antihypertensive drugs, most studies were short‐term trials with dard method in the assessment of salt intake. In the measurement
RHEE et al. |
1061
of BP, we measured 24‐hour ambulatory BP. Although casual BP is prepared methodology of the study. All authors read and approved
significantly related to future cardiovascular events, ambulatory BP the manuscript.
is better than casual BP in the prediction of cardiovascular events.12
No previous study has evaluated the association between BP control
ORCID
and salt intake by measuring 24‐hour urine electrolyte and ambula‐
tory BP in the general population. With these distinguished features, Sang‐Ho Jo https://orcid.org/0000-0002-2063-1542
we found a significant association between high sodium intake and
poor BP control, providing an indirect evidence that high sodium
intake might be an important barrier in the optimal treatment of REFERENCES
as markers of the completeness of collection. Hum Nutr Clin Nutr. randomized controlled trial of lifestyle intervention against hyper‐
1986;40(3):229‐237. tension in eastern Finland. J Hypertens. 2002;20(12):2505‐2512.
16. Wei W, Tolle M, Zidek W, van der Giet M. Validation of the mobil‐O‐ 24. Yoshita K, Miura K, Okayama A, et al. A validation study on
Graph: 24 h‐blood pressure measurement device. Blood Press Monit. food composition tables for the international coopera‐
2010;15(4):225‐228. tive INTERMAP study in Japan. Environ Health Prev Med.
17. O'Brien E, Asmar R, Beilin L, et al. European Society of Hypertension 2005;10(3):150‐156.
recommendations for conventional, ambulatory and home blood 25. Polonia J, Lobo MF, Martins L, Pinto F, Nazare J. Estimation of pop‐
pressure measurement. J Hypertens. 2003;21(5):821‐848. ulational 24‐h urinary sodium and potassium excretion from spot
18. Mancia G, De Backer G, Dominiczak A, et al. 2007 guidelines for urine samples: evaluation of four formulas in a large national repre‐
the management of arterial hypertension: the task force for the sentative population. J Hypertens. 2017;35(3):477‐486.
management of arterial hypertension of the European Society of 26. Luft FC, Fineberg NS, Sloan RS. Estimating dietary sodium intake
Hypertension (ESH) and of the European Society of Cardiology in individuals receiving a randomly fluctuating intake. Hypertension.
(ESC). J Hypertens. 2007;25(6):1105‐1187. 1982;4(6):805‐808.
19. Pimenta E, Gaddam KK, Oparil S, et al. Effects of dietary sodium re‐ 27. Sun Q, Bertrand KA, Franke AA, Rosner B, Curhan GC, Willett WC.
duction on blood pressure in subjects with resistant hypertension: Reproducibility of urinary biomarkers in multiple 24‐h urine sam‐
results from a randomized trial. Hypertension. 2009;54(3):475‐481. ples. Am J Clin Nutr. 2017;105(1):159‐168.
20. Azizi M, Blanchard A, Charbit B, et al. Effect of contrasted sodium
diets on the pharmacokinetics and pharmacodynamic effects of renin‐
angiotensin system blockers. Hypertension. 2013;61(6):1239‐1245.
How to cite this article: Rhee M‐Y, Jo S‐H, Kim J‐H, et al.
21. Darbar D, Fromm MF, Dell’Orto S, et al. Modulation by di‐
etary salt of verapamil disposition in humans. Circulation.
Difference in 24‐hour urine sodium excretion between
1998;98(24):2702‐2708. controlled and uncontrolled patients on antihypertensive drug
22. Luft FC, Morris CD, Weinberger MH. Compliance to a low‐salt diet. treatment. J Clin Hypertens. 2019;21:1057–1062. https://doi.
Am J Clin Nutr. 1997;65(2 Suppl):698S‐703S. org/10.1111/jch.13610
23. Kastarinen MJ, Puska PM, Korhonen MH, et al. Non‐pharmacologi‐
cal treatment of hypertension in primary health care: a 2‐year open