The Neuroscience of Drives for Food, Water, and Salt

by | May 2, 2019

The New England Journal of Medicine


For 30 years, end-of-life experts, including bioethicists1 and clinical researchers,2 have maintained that dying without artificial hydration is more comfortable than having dehydration relieved by intravenous fluids. The review article by Lowell (Jan. 31 issue)3 suggests that this is not true.

By identifying the brain regions and neuronal pathways involved in the state of dehydration, not confined to dry mouth, neuroscience answers in the affirmative the old question, “Is dehydration painful?”4 By identifying the cortical feed-forward reward mechanism, neuroscience answers a newer question — “Can low-volume fluids serve as placebo?”5 — in the negative. Advances detailed in Lowell’s review make clear that the robust positive response to even low-volume fluids is no artifact, but a very real benefit.

It seems that both clinical evidence and now evidence from basic science support the provision of supplemental hydration for patient comfort. Accordingly, I think that the Physician’s Orders for Life-Sustaining Treatment and other end-of-life documentation should be modified to include intravenous fluids as part of basic comfort care.

Nancy W. Weisman, Ph.D.
5415 W. Cedar Ln., Bethesda, MD

No potential conflict of interest relevant to this letter was reported.


The author replies: The suggestion that dehydration during end-of-life care is aversive, and should be treated, is an interesting extrapolation from the research described in my review. Indeed, state-of-the-art research in mice on the motivational drive of thirst supports this view. Dehydration strongly activates thirst-inducing neurons in the median preoptic nucleus of the hypothalamus (MnPO). Importantly, we now know that activation of these thirst neurons, by means of either dehydration or artificial optogenetic activation in mice that are otherwise euhydrated, induces an aversive emotional state.1-3 Mice clearly don’t like this state because they will engage in a variety of tasks to avoid it. Indeed, the MnPO neuron-induced aversive emotional state is what motivates mice to seek water. If dehydration generates the same aversive emotional state in end-of-life patients, which seems probable owing to conservation of basic homeostatic circuits, then correcting dehydration and “deactivating” these MnPO thirst neurons would be expected to alleviate the aversive emotional state that they induce. On the basis of this observation, and all other things being equal, such an intervention could improve quality of life.

Bradford B. Lowell, M.D., Ph.D.
Beth Israel Deaconess Medical Center, Boston, MA

Since publication of his article, the author reports no further potential conflict of interest.

Leave a Reply

Your email address will not be published. Required fields are marked *