Advertisement
Mayo Clinic Proceedings Home

In reply—Low-Sodium Intake: A Risk Factor for Stroke?

      To the Editor:
      We are grateful to Drs Musso and Dotto
      • Musso N.
      • Dotto A.
      Low-sodium intake: a risk factor for stroke?.
      for the appraisal of our article on low urinary sodium excretion (UNaV) as an indicator of low sodium intake and increased risk of stroke.
      • Kieneker L.M.
      • Eisenga M.F.
      • Gansevoort R.T.
      • et al.
      Association of low urinary sodium excretion with increased risk of stroke.
      We agree that the mentioned earlier study by O’Donnell and colleagues,
      • O'Donnell M.
      • Mente A.
      • Rangarajan S.
      • et al.
      Urinary sodium and potassium excretion, mortality, and cardiovascular events.
      although representative of various populations by including more than 100,000 participants from 17 countries, has some limitations. In particular, the assessment of sodium intake via a single spot urine sample is a major limitation. Actual measurement of 24-hour UNaV in multiple urine collections (to account for day-to-day variability), as we did in the Prevention of Renal and Vascular End-Stage Disease (PREVEND) study, has been shown to be a more accurate method for the assessment of usual sodium intake.
      • Holbrook J.T.
      • Patterson K.Y.
      • Bodner J.E.
      • et al.
      Sodium and potassium intake and balance in adults consuming self-selected diets.
      • McLean R.M.
      Measuring population sodium intake: a review of methods.
      In their letter, Drs Musso and Dotto note potential concerns regarding our study. First, they noted that the number of strokes among the participants of PREVEND study is low when compared with other European populations. For this, it is important to realize that, in our study, we excluded all participants with cardiovascular events (including strokes) at baseline. Therefore, in our analyses we are investigating incidence rates of stroke rather than rates of prevalence. But, still, the stroke incidence rate of 0.21% per year in the PREVEND study may seem relatively low compared with the incidence rates of other European populations, which reportedly range from 0.08% to 2.54%, with highest rates in Eastern and Northern Europe (Croatia, Estonia, Lithuania, Sweden).
      King’s College of London
      The burden of stroke in Europe.
      These differences among countries can partly be explained by the presence of risk factors of stroke such as the number of smokers, elevated blood pressure, elevated cholesterol levels, and treatment of stroke. However, some of this variation is likely also due to the different criteria—such as inclusion of transient ischemic attacks—and methods used to collect the data.
      • Levi F.
      • Chatenoud L.
      • Bertuccio P.
      • Lucchini F.
      • Negri E.
      • La Vecchia C.
      Mortality from cardiovascular and cerebrovascular diseases in Europe and other areas of the world: an update.
      Important to note is that the incidence of stroke in the PREVEND study (0.21% per year) is very similar to the overall stroke incidence reported for the Netherlands (0.23% per year).
      Rijksinstituut voor Volksgezondheid en Milieu (RIVM)
      Prevalentie en nieuwe gevallen van beroerte.
      Second, the participants of the PREVEND study had relatively low sodium intake when compared with other study populations.
      • Powles J.
      • Fahimi S.
      • Micha R.
      • et al.
      Global, regional and national sodium intakes in 1990 and 2010: a systematic analysis of 24 h urinary sodium excretion and dietary surveys worldwide.
      When looking at the lowest quintile of UNaV, we observed that the median intake of this quintile was 83 mmol per 24 hours. Assuming that approximately 90% of ingested sodium is excreted in the urine,
      • Holbrook J.T.
      • Patterson K.Y.
      • Bodner J.E.
      • et al.
      Sodium and potassium intake and balance in adults consuming self-selected diets.
      subjects in this lowest quintile of UNaV consumed approximately 5.4 grams of salt per day (equivalent to 92 mmol of sodium per day). This is slightly below the current recommendation for maximum salt intake of 6.0 grams of salt per day set by the Dutch Health Council
      Gezondheidsraad
      Natrium: Achtergronddocument bij Richtlijnen goede voeding 2015.
      and slightly above the recommendation for a maximum intake of 5.0 grams of salt per day set by the World Health Organization.
      • World Health Organization
      Guideline: Sodium Intake for Adults and Children.
      However, when comparing the median intake of the PREVEND study of 8 grams per day (equivalent to 137 mmol of sodium per day) with the global mean intake of salt of approximately 10 grams per day (equivalent to 168 mmol of sodium per day),
      • Powles J.
      • Fahimi S.
      • Micha R.
      • et al.
      Global, regional and national sodium intakes in 1990 and 2010: a systematic analysis of 24 h urinary sodium excretion and dietary surveys worldwide.
      we can conclude that the intake of sodium in the PREVEND study is relatively low. One might suppose that the association observed between low intake of sodium and risk of stroke may be explained by reverse causality, wherein a change in dietary choices is prompted by disease status. However, at baseline we already excluded subjects with history of cardiovascular disease, and, in sensitivity analyses, we tried to limit the chance of reverse causality by excluding all subjects with malignancies, type 2 diabetes, and chronic kidney disease at baseline. Exclusion of these patients did not materially change the results (hazard ratio [HR] per 1 standard deviation [SD] [51 mmol/24h] decrement, 1.45; 95% confidence interval [CI], 1.10-1.92), making reverse causality unlikely. However, as our study is observational in nature, reverse causality cannot completely be ruled out.
      Third, we agree with Drs Musso and Dotto that sodium intake is closely linked to potassium intake and that high potassium has protective cardiovascular effects. For this reason, we included 24-hour urinary potassium excretion as a marker of potassium intake and as a potential confounder in the survival analyses (multivariable adjusted model 3). The association of low sodium intake with increased risk of stroke remained independent of adjustment for urinary potassium excretion (HR per 1 SD [51 mmol/24h] decrement, 1.44; 95% CI, 1.14-1.82), supporting the notion that the increased risk of stroke observed is not due to harm induced by low potassium intake.
      Fourth, the same holds true for adjustments for blood pressure and use of antihypertensive medication. We examined whether these variables were potential mediators of the association between sodium intake and stroke by including these variables in the multivariable model. However, the association of UNaV with risk of stroke remained materially unchanged (HR per 1 SD [51 mmol/24h] decrement, 1.47; 95% CI, 1.14-1.89). We therefore did not include these variables in the model on which Figure 1 is based, as these variables did not materially influence the association between urinary sodium intake and risk of stroke.
      We thank Drs Musso and Dotto for appraising our paper as the first to correctly address a possible inverse association between sodium intake and risk of stroke and agree that, as in any study of observational nature, a risk of residual confounding will remain. As Drs Musso and Dotto implicitly suggest, more and higher-quality evidence on potential harmful effects of low sodium intake is certainly needed.

      References

        • Musso N.
        • Dotto A.
        Low-sodium intake: a risk factor for stroke?.
        Mayo Clin Proc. 2019; 94: 728-729
        • Kieneker L.M.
        • Eisenga M.F.
        • Gansevoort R.T.
        • et al.
        Association of low urinary sodium excretion with increased risk of stroke.
        Mayo Clin Proc. 2018; 93: 1803-1809
        • O'Donnell M.
        • Mente A.
        • Rangarajan S.
        • et al.
        Urinary sodium and potassium excretion, mortality, and cardiovascular events.
        N Engl J Med. 2014; 371: 612-623
        • Holbrook J.T.
        • Patterson K.Y.
        • Bodner J.E.
        • et al.
        Sodium and potassium intake and balance in adults consuming self-selected diets.
        Am J Clin Nutr. 1984; 40: 786-793
        • McLean R.M.
        Measuring population sodium intake: a review of methods.
        Nutrients. 2014; 6: 4651-4662
        • King’s College of London
        The burden of stroke in Europe.
        • Levi F.
        • Chatenoud L.
        • Bertuccio P.
        • Lucchini F.
        • Negri E.
        • La Vecchia C.
        Mortality from cardiovascular and cerebrovascular diseases in Europe and other areas of the world: an update.
        Eur J Cardiovasc Prev Rehabil. 2009; 16: 333-350
        • Rijksinstituut voor Volksgezondheid en Milieu (RIVM)
        Prevalentie en nieuwe gevallen van beroerte.
        • Powles J.
        • Fahimi S.
        • Micha R.
        • et al.
        Global, regional and national sodium intakes in 1990 and 2010: a systematic analysis of 24 h urinary sodium excretion and dietary surveys worldwide.
        BMJ Open. 2013; 3 (e003733-2013-003733)
        • Gezondheidsraad
        Natrium: Achtergronddocument bij Richtlijnen goede voeding 2015.
        2015 (A15/15)
        • World Health Organization
        Guideline: Sodium Intake for Adults and Children.
        (Geneva, Switzerland)2012

      Linked Article

      • Association of Low Urinary Sodium Excretion With Increased Risk of Stroke
        Mayo Clinic ProceedingsVol. 93Issue 12
        • Preview
          The positive relationship between sodium intake and blood pressure is well established. However, results of observational studies on dietary sodium intake and risk of stroke are inconsistent. Moreover, prospective studies with multiple 24-hour urine samples for accurate estimation of habitual sodium intake are scarce. We examined the association of urinary sodium excretion (UNaV) as an accurate estimate of intake with risk of stroke. We studied 7330 individuals free of cardiovascular events at baseline in the Prevention of Renal and Vascular End-stage Disease (PREVEND) study, a prospective, population-based cohort of Dutch men and women.
        • Full-Text
        • PDF
      • Low-Sodium Intake: A Risk Factor for Stroke?
        Mayo Clinic ProceedingsVol. 94Issue 4
        • Preview
          The recent findings by Kieneker et al1 represent the first convincing evidence that a low sodium intake might be linked to an increased cerebrovascular risk. The highly cited papers by a Canadian group2 already claimed to show a sort of J-shaped association between sodium intake and cardiovascular events, but there are limitations to these studies, as has been discussed previously.3
        • Full-Text
        • PDF