Hydration

Water balance is controlled mainly through the concentration of the blood. When blood becomes more concentrated or blood volume falls, osmoreceptors in the brain and elsewhere trigger thirst and the release of arginine vasopressin, a hormone that makes the kidneys retain water; receptors in the mouth, throat, gut and liver also register drinking and adjust thirst before blood composition changes. Water deficits can build up over a few hours through reduced intake or increased losses from physical activity and heat, but on a day-to-day basis thirst and drinking with meals maintain normal hydration in healthy people. Because the kidneys can excrete considerable excess water, the US reference intakes set no upper limit for water [1].

Common markers of hydration include serum sodium and plasma osmolality (the concentration of dissolved particles in the blood), which rise when the body is short of water, and urine colour and volume.

The US Institute of Medicine set adequate intakes for total water in 2005 at the median intakes observed in the national NHANES III survey, rather than at a level shown to prevent disease. For adults aged 19 to 30, the adequate intake is 3.7 litres a day for men and 2.7 litres for women. Drinking water and other beverages supplied about 3.0 and 2.2 litres a day, roughly 13 and 9 cups, or about 81% of total water intake; the remainder came from water in food [1]. Needs rise with heat and physical activity.

A 2002 review sought the origin of the advice to drink at least eight 8-ounce glasses of water a day, often called "8 x 8". It found no scientific studies supporting it. Surveys of healthy adults suggested such large amounts are not needed by healthy people in temperate climates with largely sedentary lives, and published studies showed that caffeinated drinks, and to a lesser extent moderate beer, may be counted toward the daily total. The review noted that larger intakes are advisable in vigorous work and exercise, especially in hot climates, and in the treatment or prevention of some diseases [3].

The main long-term evidence comes from the Atherosclerosis Risk in Communities study, which enrolled 15,752 adults aged 45 to 66 and followed them for 25 years. Researchers used middle-age serum sodium as a proxy for habitual hydration and calculated a biological age from age-related biomarkers. Serum sodium above 142 mmol/L was associated with 39% higher risk of chronic disease (hazard ratio 1.39, 95% CI 1.18 to 1.63) and up to 50% higher odds of being biologically older than chronological age; serum sodium above 144 mmol/L was associated with 21% higher risk of premature death (hazard ratio 1.21, 95% CI 1.02 to 1.45). The study was motivated partly by findings in mice, in which lifelong water restriction shortened lifespan [2].

Serum sodium reflects kidney function, hormones and illness as well as fluid intake, and the study was observational. The authors stated that intervention studies are needed to confirm a link between hydration and aging [2].

Short-term effects of mild fluid loss have been tested in controlled experiments. In a placebo-controlled study, 25 women aged about 23 completed three 8-hour sessions: exercise-induced dehydration, exercise-induced dehydration with a diuretic, and normal hydration. Dehydration averaged 1.36% of body mass. Compared with normal hydration, it worsened mood (lower vigour, more fatigue and greater total mood disturbance), increased perceived task difficulty, lowered concentration and brought on headache, both at rest and during exercise. Most measures of cognitive performance were not affected [4].

Hyponatremia is a serum sodium concentration of 135 mmol/L or less, and can occur when people drink far more than they lose. In the 2002 Boston Marathon, 488 runners gave a usable blood sample at the finish. Of these, 13% had hyponatremia and 0.6% had critical hyponatremia (120 mmol/L or less). Hyponatremia was associated with drinking more than 3 litres during the race, drinking at every mile and weight gain during the race. In the adjusted analysis, weight gain (odds ratio 4.2), a racing time over 4 hours (odds ratio 7.4 compared with under 3 hours 30 minutes) and extremes of body mass index were associated with hyponatremia, while sex and the type of fluid drunk were not [5]. Hyponatremia has become a recognised cause of race-related death and life-threatening illness among marathon runners [5].

The type of drink has a better-documented relationship with long-term health than the amount of water drunk. Among 37,716 men and 80,647 women followed for up to 34 years, consumption of sugar-sweetened drinks was associated with higher total mortality in a graded way, reaching 21% higher at two or more servings a day than at less than one a month, mainly through cardiovascular deaths [6]. Caffeinated drinks count toward fluid intake [3], and coffee and tea are described under coffee, tea and energy drinks. Sugary drinks in the context of whole diets are covered under dietary patterns.

Full deep dive: Hydration.