Some have suggested that since salt intakes are related to blood pressure, and since cardiovascular risks are also related to blood pressure, that, surely, salt intake levels are related to cardiovascular risk. This is the “salt hypothesis” or “sodium hypothesis.” Data are needed to confirm or reject hypotheses.
Blood pressure is a sign. When it goes up (or down) it indicates an underlying health concern. Changes result from many variables, often still poorly-understood. High blood pressure is treated with pharmaceuticals and with lifestyle interventions such as diet and exercise. The anti-hypertensive drugs are all approved by regulatory authorities such as the U.S. Food and Drug Administration. To be approved, these drugs must prove they work to lower blood pressure. Whether they also work to lower the incidence of heart attacks and strokes has not been the test to gain approval (it would take too long to develop new drugs), but the National Heart, Lung and Blood Institute has invested heavily in such “health outcomes” studies.
The ALLHAT study was funded by the National Heart, Lung and Blood Institute (NHLBI) to compare the health outcomes of four classes of anti-hypertensive drugs, all of which had demonstrated their ability to reduce blood pressure in relative safety. The idea is that blood pressure is only a “surrogate outcome,and we should be more concerned with clinically meaningful endpoints. Dr. Jeffrey R. Cutler, who supervised the study for the National Heart, Lung and Blood Institute (NHLBI) explained its importance: “Trials are based on the notion that different antihypertensive regimes, despite similar efficacy in lowering blood pressure, have other beneficial or harmful effects that modify their net effect on cardiovascular or all-cause morbidity and mortality.”
Lifestyle interventions are “antihypertensive regimes” too. For years, the same situation prompting the ALLHAT trial applied to lifestyle interventions designed to improve blood pressure — they were untested regarding health outcomes. Certain dietary and lifestyle interventions reduced blood pressure, at least in sensitive sub-populations. Whether they also reduced the incidence of heart attacks and strokes had never been tested. Thus, until the 1990s, scientists had never tested the “salt hypothesis” by documenting whether reducing dietary salt actually reduces a person’s chances of having a heart attack or a stroke. As in the drug “health outcomes” trials, this is now changing. The results have vast public health policy implications. We should not be recommending that everyone change their diets without evidence of some overall health benefit.
Even documenting an association of, for example, low-sodium diets with reduced incidence of heart attacks would only be the first step. Association is not the same as causation. Nevertheless, unless an association is established, we have no reason to think that a causal link is possible. Of the first seventeen “health outcomes” studies of sodium reduction, four have found an association in the general population between low-sodium diets and reduced incidence of cardiovascular events like stroke or heart attack (and two of those were in exceptionally high salt-consuming societies). The medical literature does not show a health benefit from reduced-salt diets. Here’s what scientists have found (citations):
1985. A ten-year study of nearly 8,000 Hawaiian Japanese men concluded: “No relation was found between salt intake and the incidence of stroke.”
1995. An eight-year study of a New York City hypertensive population stratified for sodium intake levels found those on low-salt diets had more than four times as many heart attacks as those on normal-sodium diets – the exact opposite of what the “salt hypothesis” would have predicted.
1997. An analysis by NHLBI’s Dr. Cutler of the first six years’ data from the MRFIT database documented no health outcomes benefits of lower-sodium diets.
1997. A ten-year follow-up study to the huge Scottish Heart Health Study found no improved health outcomes for those on low-salt diets.
1998. An analysis of the health outcomes over twenty years from those in the massive US National Health and Nutrition Examination Survey (NHANES I) documented a 20% greater incidence of heart attacks among those on low-salt diets compared to normal-salt diets
1998. A health outcomes study in Finland, reported to the American Heart Association that no health benefits could be identified and concluded “…our results do not support the recommendations for entire populations to reduce dietary sodium intake to prevent coronary heart disease.”
1999. A further analysis of the MRFIT database, this time using fourteen years’ data, confirmed no improved health benefit from low-sodium diets. Its author conceded that there is “no relationship observed between dietary sodium and mortality.”
1999. A study of Americans found that less sodium-dense diets did reduce the cardiovascular mortality of one population sub-set, overweight men – the article reporting the findings did not explain why this obese group actually consumed less sodium than normal-weight individuals in the study.
2001. A Finnish study reported an increase in cardiovascular events for obese men (but not women or normal-weight individuals of either gender) – the article, however, failed to adjust for potassium intake levels which many researchers consider a key associated variable.
2002. In September, 2002, the prestigious Cochrane Collaboration produced the latest and highest-quality meta-analysis of clinical trials. It was published in the British Medical Journal and confirmed earlier meta-analyses’ conclusions that significant salt reduction would lead to very small blood pressure changes in sensitive populations and no health benefits.
2003. In June 2003, Dutch researchers using a massive database in Rotterdam concluded that “variations in dietary sodium and potassium within the range commonly observed in Westernized societies have no material effect on the occurrence of cardiovascular events and mortality at old age.”
2004. In July 2004, the first “outcomes” study identifying a population risk appeared in Stroke magazine. Researchers found that in a Japanese population, “low” sodium intakes (about 20% above Americans’ average intake) had one-third the incidence of fatal strokes of those consuming twice as much sodium as Americans.
2006. A March 2006 analysis of the federal NHANES II database in The American Journal of Medicine found a 37% higher cardiovascular mortality rate for low-sodium dieters
2007. A February 2007 reported in the International Journal of Epidemiology studied 40,547 Japanese over seven years and found “the Japanese dietary pattern was associated with a decreased risk of CVD mortality, despite its relation to sodium intake and hypertension.”
2007. An April 2007 article in the British Medical Journal found a 25% lower risk of CV events in a group which years earlier had achieved significant sodium reduction during two clinical trials (TOHP I and TOHP II).
2007. An October 2007 analysis of a large Dutch database published in the European Journal of Epidemiology documented no benefit of low-salt diets in reducing stroke or heart attack incidence nor lowering death rates.
2008. A May 2008 examination of NHANES II (the largest US federal database of nutrition and health) published in the Journal of General Internal Medicine confirmed two earlier studies of earlier NHANES surveys that there is no health benefit (CVD or all-cause mortality) for those on low-sodium diets.
From my own research I have found an association between low sodium diets and increased tissue and organ acidity. Since sodium or salt is necessary in the production of sodium bicarbonate to manage dietary and/or metabolic acid a low sodium diet would increase risk of latent tissue acidosis and cononary vascular disease.
Do low-salt diets improve health outcomes? The evidence
There have been relatively few studies of the fundamental question of whether reducing an individual’s — or a population’s — salt intake will improve their health outcomes. Usually only one risk factor is considered: blood pressure. Other impacts confound blood pressure, itself a rather herterogeneous response. None of the outcomes studies is a controlled trial. Thus, policies embracing universal salt (or sodium) reduction have a weak foundation in the medical literature.
Here are the health outcomes studies reported publicly with links to the original sources where available:
Cutler, J.R., Presented May 30, 1997, at American Society of Hypertension annual meeting, San Francisco, CA. (unpublished).
Valkonen, V-P. “Sodium and potassium excretion and the risk of acute myocardial infarction” Presented October 15, 1998 to the American Heart Association Scientific Sessions, Dallas, TX (unpublished).
Cohen, J.D. presentation to NHLBI Workshop on Sodium and Blood Pressure, January 28, 1999, Bethesda, MD
Grobbee, D.E. et al. “Sodium and potasium intake and risk of cardiovascular events and all-cause mortality: the Rotterdam Study” presented to the 13th European Meeting on Hypertension in Milan, Italy, June 13-17, 2003 (published abstract)