Dietary reference values for sodium

View full-text article in PMC

. 2019 Sep 4;17(9):e05778. doi: 10.2903/j.efsa.2019.5778

References (country)	Design	Baseline characteristics mean ± SD (N)	Na intake assessment method	UNa (mg/day) Mean ± SD (N)	Outcomes assessed	Confounders adjusted for	Key findings
Devine et al. (1995); Prince et al. (1995) (Australia)	RCT in postmenopausal women (≥ 10 years after menopause) Group 1: Ca suppl. (1 g/day) Group 2: Ca suppl. (1 g/day) + exercise programme Group 3: Placebo Group 4: ~200 mL milk/day (1 g Ca/day) N = 196 screened N = 168 randomised (42/group) Duration: 2 years	Age: 63 (168) bw: 66 ± 10 (191) Nutrient intakes based on 4‐day weighed diet records (190): Protein: 76 ± 16 g/day Ca (food only): 805 ± 320 mg/day P: 1,269 ± 311 mg/day Energy: 1,632 ± 348 kcal/day	Single 24‐h urine samples collected at baseline, year 1 and year 2; excretions averaged over the 2‐year period. No information on instructions to participants or on measures to check completeness of collections.	Baseline UNa: 2,783 ± 1,081 (196) Average year 1 and 2 UNa: 3,049 ± 808 (127)	BMD (total hip, intertrochanter, femoral neck, ultradistal ankle, lumbar spine) By DEXA at baseline, 1 year and 2 years	Body weight, change in METs/day over the 2‐year period, average daily Ca intake	Multiple regression analysis: Negative associations between average UNa and 2‐year change in BMD for total hip (SRC = −0.20, p = 0.002) and ultradistal ankle (SRC = −0.18, p = 0.016)
Ilich et al. (2010) (USA)	RCT in postmenopausal women (≥ 5 years after menopause); all supplemented with Ca (630 mg/day) and vitamin D (~ 400 IU/day) Group 1: instructed to reduce Na intake to 1,500 mg/day Group 2: usual Na intake (~ 3,000 mg/day) N = 136 randomised (68/group) N = 97 completed Duration: 3 years	Age: 68.6 ± 7.1 (136) bw: 68.0 ± 11.3 (136) BMI: 26.0 ± 3.8 (136) Nutrient intakes based on 4‐day weighed diet records (136): Protein: 70.6 ± 18.6 g/day Ca (food only): 872 ± 365 mg/day P: 1,077 ± 351 mg/day Energy: 1,691 ± 382 kcal/day	Single 24‐h urine samples collected every 6 months. Cumulative Na/Cr used as variable. Careful instructions to participants and use of Cr to screen for errors or incomplete collection	Baseline UNa: 2,404 ± 963 (136) Baseline UNa/g Cr: 2,465 ± 903 (136)	BMD (composite femur, forearm, lumbar spine, total body) By DEXA at baseline and every 6 months	Age, height, cumulative lean and/or fat tissue, cumulative total calcium intake and cumulative modes of physical activity	Multiple regression analysis (ITT): Positive associations between cumulative UNa/Cr and BMD at 36 months at LS (Coeff. = 5.85^e−5, p = 0.056), at forearm (Coeff. = 8.12^e−5, p = 0.076) and total body BMD (Coeff. = 2.81^e−5, p = 0.041)
	Random‐effects regression analysis (accounting for repeated measures and missing data): Main effect of UNa in the forearm: higher UNa associated with higher BMD at baseline and subsequent time points (t = 2.63, p = 0.0089). Similar observation in the spine. No effect on total body or femoral BMD

References (country)

Design

Baseline characteristics mean ± SD (N)

Na intake assessment method

UNa (mg/day) Mean ± SD (N)

Outcomes

assessed

Confounders adjusted for

Key findings

Devine et al. (1995); Prince et al. (1995)

(Australia)

RCT in postmenopausal women (≥ 10 years after menopause)

Group 1: Ca suppl. (1 g/day)

Group 2: Ca suppl. (1 g/day) + exercise programme

Group 3: Placebo

Group 4: ~200 mL milk/day (1 g Ca/day)

N = 196 screened

N = 168 randomised (42/group)

Duration: 2 years

Age: 63 (168)

bw: 66 ± 10 (191)

Nutrient intakes based on 4‐day weighed diet records (190):

Protein: 76 ± 16 g/day

Ca (food only): 805 ± 320 mg/day

P: 1,269 ± 311 mg/day

Energy: 1,632 ± 348 kcal/day

Single 24‐h urine samples collected at baseline, year 1 and year 2; excretions averaged over the 2‐year period.

No information on instructions to participants or on measures to check completeness of collections.

Baseline UNa:

2,783 ± 1,081 (196)

Average year 1 and 2 UNa:

3,049 ± 808 (127)

BMD (total hip, intertrochanter, femoral neck, ultradistal ankle, lumbar spine)

By DEXA at baseline, 1 year and 2 years

Body weight, change in METs/day over the 2‐year period, average daily Ca intake

Multiple regression analysis:

Negative associations between average UNa and 2‐year change in BMD for total hip (SRC = −0.20, p = 0.002) and ultradistal ankle (SRC = −0.18, p = 0.016)

Ilich et al. (2010)

(USA)

RCT in postmenopausal women (≥ 5 years after menopause); all supplemented with Ca (630 mg/day) and vitamin D (~ 400 IU/day)

Group 1: instructed to reduce Na intake to 1,500 mg/day

Group 2: usual Na intake (~ 3,000 mg/day)

N = 136 randomised (68/group)

N = 97 completed

Duration: 3 years

Age: 68.6 ± 7.1 (136)

bw: 68.0 ± 11.3 (136)

BMI: 26.0 ± 3.8 (136)

Nutrient intakes based on 4‐day weighed diet records (136):

Protein: 70.6 ± 18.6 g/day

Ca (food only): 872 ± 365 mg/day

P: 1,077 ± 351 mg/day

Energy: 1,691 ± 382 kcal/day

Single 24‐h urine samples collected every 6 months. Cumulative Na/Cr used as variable.

Careful instructions to participants and use of Cr to screen for errors or incomplete collection

Baseline UNa:

2,404 ± 963 (136)

Baseline UNa/g Cr:

2,465 ± 903 (136)

BMD (composite femur, forearm, lumbar spine, total body)

By DEXA at baseline and every 6 months

Age, height, cumulative lean and/or fat tissue, cumulative total calcium intake and cumulative modes of physical activity

Multiple regression analysis (ITT):

Positive associations between cumulative UNa/Cr and BMD at 36 months at LS (Coeff. = 5.85^e−5, p = 0.056), at forearm (Coeff. = 8.12^e−5, p = 0.076) and total body BMD (Coeff. = 2.81^e−5, p = 0.041)

Random‐effects regression analysis (accounting for repeated measures and missing data):

Main effect of UNa in the forearm: higher UNa associated with higher BMD at baseline and subsequent time points (t = 2.63, p = 0.0089). Similar observation in the spine. No effect on total body or femoral BMD

BMD: bone mineral density; BMI: body mass index; bw: body weight; Ca: calcium; Cr: creatinine; DEXA: dual‐energy X‐ray technology; MET: metabolic equivalent activity; IU: international unit; N: number; Na: sodium; NR: not reported; P: phosphorus; RCT: randomised controlled trial; SD: standard deviation; SRC: standardised regression coefficient; UNa: urinary sodium excretion.