Drinking Water: What's the Science?

Jeffrey S. Berns, MD; Stanley Goldfarb, MD


June 18, 2014

Editorial Collaboration

Medscape &

This feature requires the newest version of Flash. You can download it here.

Jeffrey S. Berns, MD: Hello. This is Jeffrey Berns, Editor-in-Chief of Medscape Nephrology. With me today is Dr. Stan Goldfarb, who has a special interest in water. I have asked him to join me to talk about water, in particular how much water we should be drinking.

There is a theory in the lay press that we should be drinking at least 6 or 8 glasses of water a day, so let us begin with that. Where did that advice come from?

Stanley Goldfarb, MD: It is hard to say where it came from. People have tried to trace its origins. More than anything else, it may relate to the fact that people actually do consume about 6-8 glasses of water per day, including water in their food and all the liquids they consume. That is based on studies[1] that have been carried out by the National Academy of Sciences.

The issue is that the lay press has promoted the "urban myth" that not only should you drink 6-8 glasses of water a day as a typical intake, but in addition you should drink another 6 or 8 glasses a day in order to have some sort of improvement in your health. For that, there is virtually no basis at all.

Dr. Berns: Is it harmful, do you think?

Dr. Goldfarb: It is not harmful unless, of course, you drink it more rapidly than your renal excretory capacity. The excretory capacity is something on the order of 15-20 L over 24 hours as long as it is consumed at that rate. But if a person consumes several liters during a brief period of time, some have developed acute hyponatremia on that basis. So there is that risk, but it is a minor risk. The bigger issue is that it is expensive to drink bottled water, and the bottles are additional items that fill up the dumps that blight the environment.

Dr. Berns: So there is an environmental question, too.

Dr. Goldfarb: Absolutely.

Dr. Berns: Is there any physiologic reason to think that drinking more water than our thirst dictates may be beneficial?

Dr. Goldfarb: One hypothesis, primarily based on animal studies, is that animals that consume large amounts of fluid, usually stimulated by adding some sugar to the drinking water, have better outcomes in terms of experimental disease. A few studies have suggested this.[2] The explanation is that vasopressin is believed to be a culprit and somehow is producing alterations in the glomerular hemodynamics and perhaps also other effects. On the other hand, human studies have not shown similar findings.

A couple of human studies have suggested that individuals who drink very, very small amounts of fluid, significantly less than 800-900 mL/day, may be at higher risk for some cardiovascular diseases, perhaps bladder cancer, and strokes.[3] On the other hand, there is no evidence that people who consume normal amounts of water and have urine outputs of over 1 L/day are healthier when their outputs are 2-3 L/day. The most recent epidemiologic studies[4] that have been conducted have suggested that increased water intake is not associated with better cardiovascular outcomes or improved mortality.

Dr. Berns: Two patient populations come to mind that might benefit from a higher water intake: people who have a history of kidney stones and people who have polycystic kidney disease.

Dr. Goldfarb: Yes, absolutely. For people with kidney stones, certainly the evidence is incontrovertible. As a matter of fact, there was a literature review in Kidney International fairly recently, "The Medicinal Uses of Water in Renal Diseases,"[5] about water therapy, and most of the article is devoted to discussing the multiple studies that show the benefits of increasing water intake above 2 L/day for people with recurrent stone disease. It is quite clear that it is beneficial in that group.

With polycystic kidney disease, there is the observation that using vasopressin antagonists may reduce the cyst growth. As you know, studies have suggested that there may be some side effects that make the medications less useful. The logic was, if we can block cyst growth with vasopressin antagonists, perhaps high water intake will suppress vasopressin physiologically, and it would be beneficial in that way.

Some studies have shown that, depending on the level of kidney function, anywhere from perhaps 3 to as high as 4 or 5 L of water intake a day can keep the vasopressin levels relatively low and get urine osmolality to the same level as plasma osmolality. It is pretty hard to get it below plasma osmolality. But there is no evidence yet that that kind of strategy will actually reduce the growth of cysts.

Dr. Berns: Is your advice to our patients and to ourselves to drink as thirst dictates?

Dr. Goldfarb: I think you should drink when you are thirsty. I should point out that the one retrospective study of high fluid intake in people with kidney disease that came out regarding nondiabetic renal disease showed fairly impressively that individuals who had the highest urine output actually had the most rapid deterioration of renal function.[6] That study has been criticized because the study population may have had more kidney disease, and that is why they had less concentrating ability. However, the group of patients with the highest urine output also had the lowest serum sodium, suggesting that, in fact, they were driving the high urine output with a high fluid intake.

I believe the advice nephrologists should give patients with chronic kidney disease is that other than in certain circumstances -- foreign bodies in the urinary tract such as calculi and so on -- patients should not push a high fluid intake. There is no evidence that it is beneficial, and there may be some evidence that it's harmful.

Dr. Berns: And, finally, tap water vs bottled water.

Dr. Goldfarb: Tap water is a lot cheaper. Water is tested several times a day by most water departments. And I do not believe there is any rationale other than being a victim of really effective marketing techniques to drink bottled water.

Dr. Berns: Very good. Thanks. I have been speaking with Dr. Stanley Goldfarb, one of my colleagues from the Perelman School of Medicine at the University of Pennsylvania in Philadelphia. I am Jeff Berns, Editor-in-Chief of Medscape Nephrology.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.