The American Journal of Medicine
Volume 120, Issue 1 , Page e7, January 2007

The Reply

  • Hillel W. Cohen, MPH, DrPH

      Affiliations

    • Department of Epidemiology and Population Health, Albert Einstein College of Medicine, Bronx, NY
    • ,
  • Susan M. Hailpern, MS, DrPH

      Affiliations

    • Department of Epidemiology and Population Health, Albert Einstein College of Medicine, Bronx, NY
    • ,
  • Michael H. Alderman, MD

      Affiliations

    • Department of Epidemiology and Population Health, Albert Einstein College of Medicine, Bronx, NY
    • ,
  • Jing Fang, MD

      Affiliations

    • Centers for Disease Control and Prevention, Atlanta, Ga

Article Outline

 

We found, in a nationally representative observational follow-up study, a robust, significant, and consistent significant inverse association between dietary sodium and cardiovascular mortality.1 Lack of statistical significance for a subset of outcomes with fewer events does not, as He et al assert,2 invalidate the overall findings. On the contrary, the similar point estimates, despite much smaller numbers of specific coronary or stroke events, show consistency. Further, these findings are also consistent with the other study results in populations consuming <200 mmol/24 hours of sodium, except for one obese subgroup.3

He et al assert that a 24-hour urine measure of daily sodium is generally better than a dietary recall, and we agree. But when citing one observational study4 with the measure that supports their view, these authors omit another study, also with that measure but with the opposite finding.5

The contradictory findings among all the available observational studies may reflect heterogeneity among individuals in the effect of different levels of sodium, as well as differences in methodology and the inherent limitations of observational studies. Observational data, particularly when they are inconsistent, do not provide an adequate basis for clinical or public health recommendations, much less recommendations based on the supposition that one size fits all. Hypotheses generated from observational data are appropriately tested through randomized clinical trials (RCT). The feasibility of an RCT has been demonstrated through the experience of the Treatment to Prevent Hypertension Study with 2300 subjects, in which those randomized to a lower sodium diet achieved a 40 mmol/24 hour reduction in urinary sodium after 36 months.6 Rather than further debate the relative merits of inconsistent observational studies, we urge that He et al join us in a call for an RCT to determine the impact of varied salt intake on morbidity and mortality.

Back to Article Outline

References 

  1. Cohen HW, Hailpern SM, Fang J, Alderman MH. Sodium intake and mortality in the NHANES II follow-up study. Am J Med. 2006;119(3):275.e7;275.e14
  2. He FJ, Wardener HE, MacGregor GA. Salt intake and cardiovascular mortality (letter). Am J Med. 2007;120:e5
  3. He J, Ogden LG, Vupputuri S, Bassano LA, Loria C, Whelton PK. Dietary sodium intake and subsequent risk of cardiovascular disease in overweight adults. JAMA. 1999;282:2027–2034
  4. Tuomilehto J, Jousilahti P, Rastenyte D, et al. Urinary sodium excretion and cardiovascular mortality in Finland: a prospective study. Lancet. 2001;347:848–851
  5. Alderman MH, Madhavan S, Cohen H, Sealey JE, Laragh JH. Low urinary sodium is associated with greater risk of myocardial infarction among treated hypertensive men. Hypertension. 1995;25:1144–1152
  6. The Trials of Hypertension Prevention Collaborative Research Group. Effects of weight loss and sodium reduction intervention on blood pressure and hypertension incidence in overweight people with high-normal blood pressure (The Trials of Hypertension Prevention, phase II). Arch Intern Med. 1997;157(6):657–667

 M.H.A. has been an unpaid consultant to the Salt Institute, a trade organization. Neither he, nor the other authors, has ever received research support, consulting fees, or speaker honoraria from either the Salt Institute or any other commercial entity related to use of sodium.

PII: S0002-9343(06)01132-6

doi:10.1016/j.amjmed.2006.08.027

The American Journal of Medicine
Volume 120, Issue 1 , Page e7, January 2007

Article Tools