By Jeffrey Nichols, M.D.
NYCLTCEN Co-Director, Medical Director, Cabrini Center for Nursing and Rehabilitation, and Chief of Geriatrics, Cabrini
Recent issues of this newsletter have contained articles by several knowledgeable and thoughtful members of our Board presenting varying opinions and perspectives regarding what has proven to be the most frequently raised and disputed issues in long-term care ethics, i.e., decisions regarding artificial nutrition and hydration for dementia patients. Clearly none of these articulate position papers represents an officially endorsed policy for the New York City Long-Term Care Ethics Network or its Board. Indeed, the role of the Network is to stimulate thought and discussion and to assist the formation of Ethics Committees, rather than to declare what is ethical and what is not. Still, how can there be such diversity of opinion on an issue so central to what we all do? I believe that all participants in the discussion are actually in agreement on the basic ethical issues. Where they disagree is on the underlying medical issues which frame the discussion.
We are all agreed that decisions regarding artificial hydration and nutrition should be made by a capacitated patient based on the burdens and benefits of the proposed procedures as experienced by the patient. Similarly, we all agree that a resident who has advance directives which specify a desire to accept or to refuse a proposed procedure should have those directives honored in the event that the resident loses capacity. But for dementia patients the decision for tube feeding rarely presents itself while the patient still possesses the capacity to make it. Even when patients have executed a living will, the standard living will form does not specifically address the issue, while the New York State Health Care Proxy form merely allows the proxy to decide on the issue without instructing the proxy how to decide. Using the principle of substituted judgement, families are asked to search their memories for statements the patient might have made in the distant past regarding complex medical issues faced today--attempting to reach the elusive goal of "clear and convincing" evidence. And decisions based on the ethical principle of the patient's best interest must, necessarily, be based on our current understanding of the patient's medical condition and prognosis.
How these issues are presented to the family should be based on solid biomedical information, not on the personal values and beliefs of the care-givers or of the decision-makers. Whether a patient is terminally ill should not be a matter of opinion or values, but a statement based on reproducible, measurable patient parameters and well-documented patient outcomes. While the full burden of a feeding tube may be difficult to assess, the medical risks and benefits which it might offer should be known based on scientific data. Only then could residents, families, and proxies make true informed decisions.
How close do we come to this ideal? Closer, actually, than you might think. Dementia is actually a category of diseases of which Alzheimer's disease represents between 60% and 90% of the cases. Thanks to the work of Dr. Reisberg at New York University, among others, advanced Alzheimer's disease can be staged and life expectancies, regardless of treatment, can be associated with those stages. The FAST scale can identify with acceptable accuracy a group of patients with a life expectancy of six months or less, the generally accepted definition of terminal. This scale has been accepted by the National Hospice Organization as appropriate criteria for the admission of AD patients to hospice programs, and is supported by the Alzheimer's Association as defining a group of patients with terminal Alzheimer's disease. It has been validated by clinical observations on literally thousands of end-stage patients. These criteria are probably equally accurate in the identification of the final stages of dementia associated with multi-infarct dementia, Parkinson's disease or with Pick's disease, although insufficient evidence exists at present to confirm this.
That medical science should be able to define "terminal" phases of these dementias is entirely consistent with out understanding of these illnesses. They are all progressive neurogenerative illnesses. Since the brain is a vital organ, the accumulation of sufficient degenerative damage might reasonably be expected to ultimately produce a fatal outcome. Indeed the National Bureau of Vital Statistics accepts Alzheimer's disease as a cause of death--the ninth most common cause of death for adult Americans in 1990. Yet even these statistics are deeply flawed, because a small number of State Health Departments do no accept Alzheimer's disease as a fatal illness and, unfortunately, New York's is one of them. It is argued that Alzheimer's patients do not die of their disease but rather of infections, pulmonary emboli, or other complications of immobility and impaired physical functioning. But this is no less true for cancer patients, whose terminal events are almost never the actual erosion of the tumor into a vital structure but almost always the total physiological burden of the disease, often with these same complications which are the terminal events for Alzheimer's disease. Since at this point there is no treatment which has been demonstrated to arrest or to slow the progression of end-stage dementing disease, it should properly be regarded as a terminal state. Families and other surrogates should be made aware that the patient is dying at the time that medical decisions are being made.
Where do Dr. Blandford's recommendations on the assessment and management of abnormal feeding behaviors fit into this picture? Ten years ago, at the annual John Hopkins Dysphagia Conference, I asserted that approximately two thirds of all residents met standard criteria for possible dysphagia and that little to nothing was then know about appropriate evaluation or management of those residents. Since that time (but by no means because of my remarks) some literature has developed around these issues, yet the best available research remains contradictory and confusing. It is not clear that feeding tubes of any type (nasogastric, gastrostomy, or jejunostomy) prevent aspiration, decrease decubitus ulcers, promote comfort in any way, or prolong life. Feeding tubes have been shown to shorten life-expectancy in many cancers. 1991 National Medicare data showed that 20% of Medicare patients for whom Medicare paid for feeding gastrostomy and who were subsequently transferred to a skilled nursing facility were dead within 30 days. Median survival was 29.4 weeks. Several studies have shown shortened survival times for tube-fed as opposed to spoon-fed residents. Tube feeding may lead to more frequent and more severe pneumonias by encouraging bacterial overgrowth of upper airways. Part of the difficulty in assessing this and other data about feeding tubes, is the wide variety of indications and conditions for which they have been used. One of the few studies done in long-term care which showed prolonged survival of tube-fed patients included only 175 for whom the indication was dysphagia, while 67% of tubes had been placed for weight loss or "refusal to eat." A review of a Wisconsin MDS data sample of 434 tube- fed long-term residents showed that 10% ambulated independently and 35% reportedly were able to make themselves understood most of the time. These are clearly a very different population from end-stage demonstration patients. What Dr. Blandford's research, supported by research from Dr. Ladislov Volicer at Boston University and other researchers around the country, achieves, is to identify a large group of late-stage (not end-stage) dementia patients who should not receive feeding tubes, not because they are futile, not because it is ethically wrong per se, but medically because it is wrong and unnecessary. These patients can be fed orally if we assess them appropriately and take the time and preparation necessary for oral feeding. And I am sure that we would all agree that it is unethical to ask a family to consent to a feeding tube because a facility will not assign staff to do necessary feedings.
Nevertheless, we are all familiar (as Rabbi Kronick points out) with severely neurologically impaired residents who have had very prolonged survivals-- assisted it would seem, by enteral feedings. Indeed, there would have been no Quintan or Cruzan cases if tube-feeding was not potentially compatible with prolonged survival. These long-surviving residents are generally victims of severe brain trauma or large strokes. Although neurologically impaired, they do not have progressive illnesses, and their prognosis may be highly variable. At this time there is no good information to offer families as to the likelihood that a feeding tube will prolong life. For these sadly incapacitated patients, Rabbi Kronick's concerns to honor the personhood of the individual through a careful assessment of their values and the values and beliefs of the family and close associates is entirely appropriate and necessary. We can not escape our obligation to the resident to offer an intervention which may convey significantly enhanced survival time to at least some subgroups.
What is certainly not clear is the precise relationship between the FAST scale and Dr. Blandford's Feeding Behavior Inventory. Is there a phase of Alzheimer's disease when severe oropharyngeal and pharyngoesophageal dysphagia occur but when the patient has not otherwise met neurological criteria for terminal status? I suspect the answer to this question is no, that these are simply additional markers for terminal status. As yet unpublished data seem to confirm that this is clearly a fruitful area fur further research. If, indeed, my suspicion is correct, feeding tubes for Alzheimer's disease would become obsolete, since in the middle and late stages they would be unnecessary, while in the final stages they would be ineffective or possibly harmful.
Next Article
Previous Article
Back to News Home Page
