Special Therapeutic Issues: Hydration, Nutrition, and Antibiotics in End-of-Life Care

I have no pain, dear mother, now;

But oh! I am so dry:

Just moisten poor Jim’s lips once more;

And, mother, do not cry!

Edward Farmer, 1809–1876, The Collier’s Dying Child

As in the case of Mr. A. in the previous chapter, palliative care clinicians are frequently asked what they will or will not provide in terms of specific therapies: “Does your hospice allow IVs?” “You can’t let Mom starve to death!” “He has pneumonia. You’re giving him antibiotics, aren’t you?” Chapter 8 addresses how to respond to such inquiries in terms of communication skills. Here, I focus on the specific issues of hydration, nutrition, and antibiotics, issues that commonly arise at the end of life.

Hydration and Nutrition

The discussion of cachexia, anorexia, and asthenia in Chapter 5 leads naturally to a broader discussion of the role of hydration and nutrition. Eating and drinking are the means to two overt goals: sustaining life through the provision of calories, nutrients, and water and the alleviation of suffering associated with hunger, thirst, and decreased functional status. Both eating and drinking also represent basic pleasures in life. Beyond this, eating and drinking are also social activities that offer opportunities for nurturing and for being nurtured. In everyday life we are largely unaware that separate goals exist; eating and drinking are simply good things to do. In palliative care distinctions between life-prolonging and life-enhancing goals are often required. Confusion about goals in this area causes a great deal of distress in palliative and end-of-life care.¹

If the goals of care are directed solely to the provision of comfort (and enhancement of associated pleasures) and not life prolongation, then a life-prolonging goal is no longer relevant. In such cases decisions regarding all forms of feeding should be assessed relative to the goals of alleviating suffering and enhancing quality of life. Natural but assisted feeding, such as spoon-feeding, may alleviate hunger or thirst and usually provides simple pleasure. As such, it may be welcomed. However, force-feeding, especially of patients who have lost their appetites and thirst, may, in fact, cause distress, usually at the hands of well-intentioned family members or caregivers who feel compelled to get some arbitrary amount of food into a patient. This compulsion usually arises from a lack of clarity in goals and from values that equate volume of intake with nurturing and quality of care.

Looking more closely at the issue of artificial hydration, like any therapy it has potential benefits and burdens.²^,³^,⁴^,⁵ Most hospice practitioners believe that the burdens of artificial hydration at the very end of life usually outweigh the potential benefits. Dry mouth, as previously discussed, is a common problem, and true thirst may also be present, although studies suggest that thirst is less prevalent and troublesome than is dry mouth at the end of life and tends to fade near death.⁶^,⁷ However, there is little, if any, correlation between hydration status and both thirst and dry mouth at the end of life. For example, Ellershaw found in a prospective study no difference in the incidence of dry mouth or thirst between patients with normal hydration statuses as measured by serum osmolality and dehydrated patients.⁸ Musgrave found no evidence of relief of dry mouth or thirst, if present, in patients in advanced stages of dying who were treated with intravenous therapy.⁹ Ripamonti et al., in a study that compared treatments of malignant bowel obstruction, found no statistical difference in daily thirst or dry mouth scores for patients who were receiving IV hydration (>2000 cc/d) compared to those not parenterally hydrated (<500 cc/d). Interestingly, in this study hydrated patients experienced less nausea than did non-hydrated patients, perhaps related to better renal function and the ability to excrete toxins in hydrated patients.¹⁰

Many hospice workers believe hydration may worsen retained respiratory secretions at the very end of life. However, the effect of hydration on respiratory secretions at the end of life is unclear. Ellershaw found no correlation between hydration status and secretions in her study. None of the patients in that study were artificially hydrated.

It is clear that in some circumstances hydration can alleviate symptoms of delirium in dying patients.¹¹^,¹²^,¹³^,¹⁴ However, it may be technically difficult to maintain IV or SC hydration in patients who have poor access or are agitated. Fussing over IVs or, worse, restraining patients to enable IV hydration can contribute significantly to suffering for patients and families. Although decisions must be individualized, in my experience most patients do quite well without hydration in advanced stages of dying.

My impression is that most hospice workers, having treated many patients with and without hydration at the end of life, would generally prefer not to receive hydration when their turns come. One study supports this view. Andrews and Levine surveyed 96 hospice nurses about their experiences and opinions on this matter. “Overall, the study showed that hospice nurses who were experienced in the matter of terminal dehydration viewed it as beneficial in the terminal stages of life.”¹⁵ That is, surveyed nurses favored not using artificial hydration.

In Chapter 5 the limited effectiveness of artificial feeding was discussed relative to cachexia, anorexia, and asthenia. I now consider in more detail tube feeding as a means of providing hydration and nutrition. The following “Fast Fact” on tube feeding (adapted from reference 16) was generated to provide clinicians with some basic guidelines for their discussions and decision making with patients and families. It is remarkable how often clinicians suggest tube feeding or place tubes with no data to support the purported goals of therapy.

Tube Feed or Not Tube Feed?

Tube feeding is frequently used in chronically ill and dying patients. The evidence base for much of this use is weak at best. In the list below some of what is known (and not known) is summarized about tube feeding for specific indications.

For prevention of aspiration pneumonia:

• Reduction in the chance of pneumonia has been suggested for non-bedridden post-stroke patients in one prospective, nonrandomized study. For bedridden post-stroke patients, no reduction was observed.¹⁷ Three retrospective cohort studies that compared patients with and without tube feeding demonstrated no advantage to tube feeding for this purpose.

• Swallowing studies, such as videofluoroscopy, lack both sensitivity and specificity in predicting who will develop aspiration pneumonia. Croghan’s study of 22 patients who underwent videofluoroscopy demonstrated a sensitivity of 65% and specificity of 67% in predicting who would develop aspiration pneumonia within one year.¹⁸ In this study no reduction in the incidence of pneumonia was demonstrated in those who were tube fed.

• Swallowing studies may be helpful in providing guidance regarding swallowing techniques for populations amenable to instruction.

• Numerous observational studies have shown a high incidence of aspiration pneumonia in those who have been tube fed.

For life prolongation via caloric support: