Meta-analysis or Systematic Review

M - Click here for explanation of classification scheme.

POSITIVE: See Quality Criteria Checklist below.

To establish the time for achievement of maximal blood pressure (BP) efficacy of a sodium reduction (SR) intervention and the dose-response relation between the amount of SR and the BP response in individuals with hypertension and normal BP.

• Relevant studies retrieved from a pool of 167 randomized controlled trials (RCTs) published in the period from 1973 to 2010
• For estimation of maximal efficacy: Studies with at least two measurements of BP effects between week one and week four after initiation of SR intervention
• For estimation of dose-response relation: Studies randomly allocating participants to at least three different amounts of SR.

Studies from the pool of 167 RCTs that did not meet inclusion criteria.

Recruitment

This study is a supplementary analysis of a previous meta-analysis. Relevant studies were retrieved from a pool of 167 randomized controlled trials (RCTs) published in the period between 1973 to 2010 and identified in a Cochrane review in 2011. Search methods and the general eligibility criteria of the Cochrane review are described previously. Fifteen relevant RCTs were included in the maximal efficacy analysis and eight were included in the dose-response analysis.

Design

Most data were retrieved as explained previously. The following data were recorded:

• Mean baseline systolic blood pressure (SBP) and diastolic blood pressure (DBP) (mm Hg)
• Mean effect measurements (change in SBP and DBP) obtained weekly until termination of study
• Effect measurements (change in SBP and DBP) for different doses of SR
• Cochrane risk of bias sources.

Statistical Analysis

• Maximal efficacy analysis: Studies were integrated irrespective of baseline BP. The effect of SR on SBP and DBP was calculated for each week. In the comparison of week one vs. week two, only those studies that had measurement at both week one and week two were included. The two summary estimates for weeks one and two were then compared. The same procedure was used for all week comparisons. For each week investigated, data were categorized as generic inverse variance data and the mean difference of the week compared with baseline was calculated by means of the inverse variance method. The summary estimates (mean difference in SBP and DBP) for each week were then categorized as continuous data and compared by means of the inverse variance method. Normotensive participants (BP less than 140/90mm Hg) and hypertensive participants (BP of 140/90mm Hg or more) were combined in the analyses.
• Dose-response analysis:
  • The sodium doses were categorized as
    • Low: Less than 90mmol per day (2,070mg per day)
    • Low usual: From 90mmol to 159mmol per day (2,070mg to 3,657mg per day)
    • High usual: From 159mmol to 248mmol per day (3,657mg to 5,700mg per day)
    • High: Higher than 248mmol per day (higher than 5,700mg per day).
  • However, if a study had two groups within one category, we allowed the one closest to one of the cut-off values to be moved to the neighboring category. Blood pressures at different sodium doses were recorded and depicted as a function of sodium intake. The effects of higher sodium intake categories were compared with the low-sodium category. Normotensive and hypertensive studies were integrated separately.
  • Normotensive and hypertensive participants were compared separately in Review Manager. Because there is general agreement about the effect of SR on BP in hypertensive participants but not in normotensive participants, mixed groups of participants of normal BP and hypertension were classified as hypertensive, even if the mean BP was below 140/90mm Hg, in order to be able to evaluate a clean normotensive population. The significance level was defined to be 5%.
• The Cochrane Collaboration tool for assessing risk of bias was used to assess the risk of bias in the individual randomized trials. Funnel plots were used to estimated publication bias across studies. Review Manager was used to synthesize data and heterogeneity was estimated by I² ( I² is the proportion of the total variability explained by heterogeneity and is expressed as a percentage).

Dependent Variables

Change in SBP and DBP, obtained weekly until termination of study and for different doses of SR.

Independent Variables

Sodium reduction (categorized into four sodium doses for the dose-response analysis).

Control Variables

• Baseline BP
• Study design (crossover vs. parallel)
• Washout period
• Other BP-lowering intervention (medication).

 

Study Description

• Ten studies included healthy normotensive participants
• Ten hypertensive participants
• Three studies included both
• Eighteen studies specifically mentioned that participants with diseases were excluded
• One study included regular drinkers (three drinks per day) who were otherwise healthy
• Five studies did not mention exclusion criteria
• None of the studies described the random assignment procedure. There was lack of observer blinding (detection bias) in five of the maximal efficacy studies and five of the dose-response studies.

Maximal Efficacy Analysis (15 Studies)

• Seven studies included hypertensive participants
• Seven included normotensive participants, and one included both
• Three studies included treated hypertensive participants
• Eight studies used crossover design
• Seven used parallel design
• Because of the few studies with measurements at weeks three, five, seven, eight and beyond, separate analyses of these measurements compared with weeks one, two, four and six were not performed
• In two studies that measured BP at weeks one, two and three, the week three measurement was included as a week four measurement
• In one study that measured BP at weeks two and five, the week five measurement was included as a week four measurement
• In another study that measured BP at weeks two, four, eight, 12 and 14, the week eight measurement was included as a week six measurement.

Dose-response Analysis (Eight Studies)

• Four studies included hypertensive participants who were all untreated
• All studies but one used a crossover design.

Key Findings

• The maximal efficacy analysis showed that, after initiation of SR (range of 55mmol to 118mmol per day), there were no significant differences in SBP or DBP between measurements at weeks one and two, weeks one and four, weeks two and four, weeks two and six, and weeks four and six.

Early Week vs. Successive Week	Mean Difference, mm Hg (95% CI)
Week one vs. two (DBP) Week one vs. two (SBP)	0.12 (-2.53, 2.77) -0.18 (-3.03, 2.67)
Week one vs. four (DBP) Week one vs. four (SBP)	0.35 (-2.02, 2.72) -0.50 (-3.20, 2.20)
Week two vs. four (DBP) Week two vs. four (SBP)	-0.20 (-1.12, 0.72) -0.10 (-1.88, 1.68)
Week two vs. six (DBP) Week two vs. six (SBP)	-0.42 (-1.69, 0.85) -0.50 (-2.66, 1.66)
Week four vs. six (DBP) Week four vs. six (SBP)	-0.22 (-1.50, 1.06) 0.39 (-1.77, 2.55)

 

• The dose-response analysis, which showed that within the established usual range of sodium intake [less than 248mmol per day (5,700mg per day)], there was no relation between the amount of SR and BP outcome in normotensive populations. In contrast, pre-hypertensive and hypertensive populations showed a significant dose-response relation.

Normotensive Populations
 

Sodium Dose	Change in SBP Compared With the Low-sodium Dose  mm Hg (95% CI)	P-value	Change in DBP Compared With the Low-sodium Dose   mm Hg (95% CI)	P-value
Low-usual-sodium dose	Not applicable	Not applicable	-0.49 (-4.00, 3.03)	0.79
High-usual-sodium dose	0.99 (-2.12, 4.10)	0.53	-1.67 (-4.17, 0.82)	0.19
High-sodium dose	3.06 (-1.47, 7.60)	0.19	0.38 (-4.03, 4.79)	0.87

 

Prehypertensive and Hypertensive Populations

Sodium Dose	Change in SBP Compared With the Low-sodium Dose mm Hg (95% CI)	P-value	Change in DBP Compared With the Low-sodium Dose  mm Hg (95% CI)	P-value
Low-usual-sodium dose	4.65 (3.39, 5.91)	<0.00001	2.44 (1.57, 3.31)	0.00001
High-usual-sodium dose	6.87 (5.61, 8.12)	<0.00001	3.61 (2.83, 4.39)	0.00001
High-sodium-dose	10.03 (4.12, 15.95)	0.0009	5.55 (1.53, 9.56)	0.007

Other Findings

• Because lack of blinding was the most obvious source of bias, separate maximal efficacy analyses of 10 studies with blinded BP measurements were performed. Eight of the studies were double blind. Theses analyses were in accordance with the main analysis.
• Five analyses showed significant heterogeneity. In all five analyses, heterogeneity disappeared when separate analyses of participants with untreated hypertension and participants with a baseline BP below 140/90mm Hg (normotensive and treated hypertensive participants) were performed.

• The present data may be considered to be insufficient to definitively answer the question on the time of maximal efficacy of SR and the dose-response relation
• Although available data are limited, this present analysis indicates that the effect of SR is evident primarily in persons with borderline BP elevation and hypertension. Specifically, the effect of SR on BP appears to reach maximal efficacy at week one and remain stable over subsequent time intervals. At present, it is reasonable to assert that there is no statistically significant evidence to show that the time of maximal efficacy requires more than one week to be established.
• The importance of kinetic and dynamic properties of sodium reduction as well as baseline BP, should probably be considered when establishing a policy of sodium reduction.

Not-for-profit

A.P. Møller Foundation for the Advancement of Medical Science