Comfort Measures for the Terminally III

The decision to administer fluids to a dehydrated dying patient is often discussed primarily as an ethical issue, but sensible therapeutic choices should be based first on a clear appreciation of the clinical manifestations of water arid electrolyte disorders and of the potential benefits of treatment. Does dehydration cause symptoms? Are these symptoms distressing to the patient? Do fluid and electrolyte administration and other treatments help make dying patients comfortable? Dehydration, defined here as a loss of normal body water, is a term that is often used imprecisely to describe conditions with differing causes, symptoms, and management.1, 2 In general, research on fluid and electrolyte disorders has not been aimed at defining how water-depleted conditions of various types, severity, and duration are manifested symptomatically, nor at how symptoms respond to treatments. Currently, when fluid and electrolyte disorders are studied or reviewed, sodium metabolism tends to receive much greater attention than water depletion. Textbook presentations of the signs and symptoms of dehydration are often subsumed under discussions of hyponatremia (or hypoosmolality) and hypernatremia (or hyperosmolality). A questionable assumption is made that hyponatremia or hypernatremia associated with dehydration causes the same clinical picture as when these electrolyte disorders develop in patients with normal or increased body water. Classic studies by McCance3 and Nadal et al.4 of short-term, experimentally induced dehydration in a small series of normal subjects appear to be the only widely cited and somewhat systematic observations of the clinical manifestations of dehydration in adults. In the light of contemporary research design criteria, these studies (and many subsequent reports on the subjective response to dehydration) can be severely faulted. Nadal et al.,4 however, suggested the important notion that quite different clinical syndromes could be associated with two prototypical forms of dehydration: sodium depletion and pure water loss. One experimental form of water loss, pure salt or sodium depletion, develops when subjects undergo salt and water depletion in a setting in which only water is restored. This condition can be produced experimentally and clinically when lost sweat or gastroinestinal secretions are replaced with salt-free fluids. A similar pattern of clinical findings, perhaps best called hyponatremic or hypotonic dehydration, develops when salt and water in body fluids are lost, but either the loss of sodium is proportionately greater than the loss of fluid or, as occasionally occurs among patients on diuretics, the repletion of salt is inadequate. Salt is the principal cation of the extracellular fluid, so hyponatremic dehydration leads predominantly to salt and fluid loss from the intravascular and interstitial compartments. Sodium depletion is thus sometimes called volume depletion, a term that also is imprecise but that points to the prominence in this condition of signs of circulatory insufficiency. Azotemia, hyponatremia, and hemoconcentration (decreased intravascular fluid and increased red cell volume) are the principal laboratory findings in pure salt depletion. Patients with pure salt depletion exhibit weight loss, poor skin turgor, dry mucous membranes, diminished sweat, and postural hypotension. Neuropsychiatric manifestations, such as weakness, apathy, lethargy, restlessness, confusion, delirium, stupor, coma, and seizures, also occur, and primarily may reflect hyponatremia, as these symptoms also are seen when hyponatremia develops in the absence of volume depletion. The neuropsychiatric symptoms have been attributed to overhydration of brain cells, and are similar to clinical findings in other hypoosmolar states. Neuropsychiatric disorders occur especially when hyponatremia is severe or develops rapidly. A prompt lowering of serum sodium to 128 mmol/L can produce marked symptoms, whereas a gradual lowering to 110 mmol/L may go unnoticed. Hyponatremia also has been reported to cause psychosis and such localized neurologic findings as aphasia, ataxia, and focal weakness. Subjects engaged in physical labor or undergoing relatively rapid onset of hyponatremia may complain of muscle cramps. Anorexia, nausea, vomiting, and loss of taste have sometimes been noted in experimental and clinical subjects with hyponatremic dehydration. However, the frequency and severity of these symptoms with various degrees of hyponatremic dehydration has not been adequately characterized, and the possibility remains that they usually are contributing causes rather than the effects of salt and water deprivation. Notably, if hyponatremic dehydration produces symptoms that lead to diminished salt intake or increased salt loss through emesis, a vicious cycle of worsening salt depletion will occur.1 Thirst is provoked primarily by hyperosmolar states, and thus may be absent or mild in patients with hyponatremic dehydration. Although modest reductions in blood volume and blood pressure do not elicit thirst, marked volume loss and hypotension are generally regarded as stimuli for antidiuretic hormone (ADH) release and water craving, as is apparent in the early stages of hypovolemic shock.5 In a second experimental situation, subjects lose salt and water but have limited access to water while continuing to ingest salt. Clinically, this pattern of dehydration typically develops in confused or somnolent patients who have lost water disproportionately to salt or for whom water has been inadequately replaced. Similar findings may occur in patients with an osmotic or postobstructive diuresis or with fluid loss from burns. Because two-thirds of body water is in the intracellular fluid compartment, equilibration from losses of extracellular fluid will result in water loss from cells, a trigger for thirst and the release of ADH from the hypothalamus. Thus, pure water loss (pure water deprivation or hypernatremic or hypertonic dehydration) is characterized early in its development by intense thirst. Indeed, the perpetuation of this condition in situations in which water is available suggests a blunting of presumed normal mechanisms of thirst or the response to thirst. Extracellular fluid volume is fairly well maintained in states of pure water deprivation, so skin turgor, blood pressure, and pulse are relatively unchanged.4, 6 The major laboratory finding of pure water loss is hypernatremia. Once significant pure water loss develops, mental status changes ensue, beginning with mild confusion and progressing to obtundation and coma. The clinical presentation in this condition has been likened to the symptom complex in hyperglycemia and other hyperosmolar states that cause brain cell dehydration. Profound neurologic damage has been reported in children. The neuropsychiatric symptoms may affect the sense of thirst, and may further hinder the ability to replace fluids. Symptoms seem to be more severe when hypernatremia develops rapidly. Fatigue, muscular weakness, and low-grade fever also have been reported. Dehydrated terminally ill patients usually present with mixed disorders of salt and water depletion, resulting typically from abnormal gastrointestinal or renal losses; normal water losses from the skin, lungs, or kidneys; and from failure to replace these losses. Isotonic dehydration, a condition that regularly is overlooked in discussions of the symptoms of fluid and electrolyte disorders, may occur. Clinicians with a special interest in the care of the terminally ill7, 8 describe thirst and dry mouth in dehydrated patients, but do not report encountering the other worrisome symptoms, such as headache, nausea and vomiting, or cramps, which are cited in some reviews of water deprivation. Personal observation of hospice patients suggests that nausea and vomiting are not regular sequelae of hyponatremia dehydration (nor of other forms of dehydration), although these symptoms frequently cause fluid and electrolyte deficiency. In the author's experience, thirst and dry mouth are the only seriously troubling and commonly encountered symptoms that can be attributed to dehydration in terminally ill patients.9 These symptoms may be satisfactorily relieved by amounts of oral fluid too small to significantly reverse metabolic abnormalities3, 10 or by maintaining moisture in the mouth with water, ice chips, or various forms of artificial saliva.7-9 Postural hypotension may be bothersome in dehydrated patients who are ambulatory. Bedbound patients report primarily lethargy, drowsiness, and fatigue, but these symptoms are rarely a source of much distress. Disorders of thirst and the mental status changes that foster and perpetuate salt and water deficiencies may protect against discomfort or obliterate the awareness of suffering. Indeed, patients who become dehydrated may be too lethargic to be troubled by symptoms potentially produced by fluid deprivation. The decision to administer fluids to a dehydrated patient is often determined by the symbolic or emotional meaning of such measures to the patient, family, and caretakers. Even irreversibly comatose, terminally ill persons are commonly given intravenous fluids, sometimes at a “keep open” rate, which is physiologically almost useless. Maintenance of fluid and electrolyte balance may be considered ordinary (ie, standard or necessary, rather than extraordinary or heroic), whereas the withholding of such treatment may be viewed by the family and others as causing suffering or as evidence of not caring about the patient.11 However, insofar as the physical comfort of the patient guides management12-15 (or until more systematic studies are available on the subjective effects of dehydration and its treatment), fluid depletion in dying patients should be regarded as a disorder with relatively benign symptoms. Successful treatment of the discomfort of thirst and a dry mouth generally does not require rehydration. Administration of salt and water by enteral (oral or, sometimes, rectal), subcutaneous, or intravenous routes should be considered for the purpose of restoring health and prolonging life, and may be justified readily in selected patients (for instance, those who still enjoy meaningful well-being but either temporarily or permanently cannot ingest liquids). Correction of hypernatremia or hyponatremia occasionally may be indicated to provide symptomatic relief. However, when the goal of management is to promote comfort or relieve suffering, treatment generally can be confined to simple measures in the form of mouth care.