Recovery is possible

The goal of treatment must always be clarified when considering palliative care. In terminal illness, the decision to withdraw active treatments and provide a supportive approach to symptom control is often appropriate. We question whether this is applicable in anorexia. Recovery is possible even in patients with longstanding severe anorexia. In a 10 year follow up of 76 severely ill women with anorexia, Eckert et al found that 18 (24%) had fully recovered, about half had a benign outcome, and only five (7%) had died.2 Ratnasuriya et al found a fairly constant rate of recovery during the first 12 years after onset of illness, with reduced likelihood of recovery after this.3 In the light of these findings, we believe that the hospice admission and treatment with opiates raises important issues about the difficulties some medical and nursing staff have in dealing with chronic mental illness.

Judgments may be clouded

Symptoms of depression, common in very underweight patients, may occur also in close family members. They may cause the sufferer, and his or her family, to underestimate previous positive clinical interventions and times of relative improvement. In this case it was reported that, “She had not committed herself to any therapeutic program, and had failed to gain any significant weight despite numerous episodes of intensive behaviour therapy and psychotherapy.”1 There is no intimation that clinical information had been sought from other hospitals where she had been treated. When the consequences of abandoning active treatment are so important the clinical assessment should include not only the patient's subjective report but also a review of all previous case notes and discussion with doctors who had been involved in the case.

In other medical settings the presence of depression is associated with the rejection of treatment even in situations with a good medical prognosis.4 Most general psychiatrists see few patients with anorexia each year and rarely have to manage the most severe and chronic cases. The emotional demands of working with a young person who is dying are arduous, and the resultant pessimism of patient and relatives may also affect the emotional response of the psychiatric team. In studies of the use of euthanasia and assisted suicide in Holland, a patient's feeling of hopelessness is one of the main factors in affecting the perceived appropriateness of euthanasia.5 When low weight levels are reached, staff may accept at face value statements that previous treatments have been ineffective, or assume that no other treatments will be effective in future, and therefore cease active treatment for a distressed patient who is not improving. Treatment with opiates can be effective in reducing pain and distress in physical illness, but this is not a recognised treatment of anorexia nervosa.

Anorectic patients, unlike those with physical terminal illnesses, fulfil the criteria for mental illness. A further complication is the impact of low body weight on cognitive function. In the Minnesota Studies normal volunteers were systematically starved over several months.6 Although none of these subjects initially suffered from anorexia nervosa, as they began to lose weight they developed anorectic patterns of eating, with preoccupation with food, bingeing, poor concentration, reduced libido, reduction in outside interests, social withdrawal, and apathy and they inaccurately perceived themselves to be overweight. Such perceptual abnormalities may lead to patients overestimating the width and size of their own faces by over 50% and reflect the impact of starvation on the brain.7 In most such patients, these features disappear with weight restoration. This raises important issues about the ability of patients who reach a very low body weight to give or withhold consent from treatment as a result of their mental disorder. Thus, treatment with nasogastric tube feeding on a medical ward when necessary (and imminent death could be argued to be such a case) can be justified under Section 3 of the Mental Health Act.8

Conclusions

When dealing with chronic illness, doctors should be able to tolerate distress and negativism and still offer support, control of symptoms, and effective treatment. This may require consistent care for years and necessitates a positive therapeutic stance. If possible, patients should be prevented from reaching such a low physical and emotional state that death seems the only acceptable option to them, their families, and doctors. Severe physical complications are best approached with medical care shared between physician and psychiatrist. Severe chronic anorexia is best treated by an experienced multidisciplinary team, with treatments of proved efficacy offered by experts in the specialty. When such skills are not available, a second opinion should be sought. Specialist centres make a valuable contribution to the assessment and treatment of such patients. The pessimism of patients and relatives at a time of exacerbation of illness should not prevent active treatment. Without this, decisions made benevolently may fail to offer patients adequate care.

Appropriateness of palliative care for anorectic patients

Specialist palliative care services were developed to address the needs of patients dying from cancer and their families. Such care is now considered appropriate for patients with any diagnosis causing active, progressive, and advanced disease and a limited prognosis.2 Currently, lack of resources restricts provision of specialist palliative care.

Anorexia nervosa is a complex illness that combines psychological and physical morbidity. Detailed information and effective communication are essential to establish appropriate goals of treatment. The staff of most hospice units discuss and agree the aims of inpatient admission with the patient, his or her family, and referring teams. Frequent review of care plans and discussion of changes in clinical care are routine.

Williams et al comment on the effects of starvation on cognition and the impact on the ability of such patients to give informed consent. This clarifies the need for specialist psychiatric advice in patients with anorexia nervosa and highlights the benefit of clear guidelines in their management. Interestingly, although some patients with cancer suffer marked cachexia, the potential effect on their cognitive function is not widely considered with respect to their making decisions about treatment. This may warrant further study.

Palliative care is not just terminal care

The possibility of recovery from anorexia nervosa, even for patients with a poor prognosis, is cited as a reason for withholding specialist palliative care. This argument no longer holds for cancer patients. The need for pain and symptom control is just as clear for patients with a 90% chance of long term survival as for those with a 10% chance. Close cooperation between specialists in palliative care and other medical disciplines can achieve optimal care without sacrificing survival chances. There is confusion over use of the terms palliative care and terminal care. Terminal care makes up a part of the spectrum of palliative care, when it is recognised and accepted that a patient will die within hours or days. At this stage all care is focused on providing comfort even if there is a risk that measures to do so may hasten death.

Involvement of specialist palliative care services for patients with anorexia nervosa would not necessarily involve admission to a hospice. Shared care in a hospital setting can allow a patient to remain in a psychiatric unit nursed by familiar ward staff who have psychiatric expertise. This environment may not always provide sufficient generic nursing skills for care of patients with severe debilitating physical problems. By contrast, acute medical wards may have difficulty in providing all the elements of satisfactory care for patients with complex psychiatric problems and difficult family dynamics. Perhaps the best care for severely ill anorectic patients requires the development of specialist units where staff are competent and confident in dealing with complex physical and psychological care. If cure or remission is being sought then a hospice inpatient unit is unlikely to be acceptable to most anorectic patients. However, if death is felt to be imminent the needs of a terminally ill anorectic patient could probably be best met in a hospice unit.

The association between receiving opiates and dying is a common misconception among those unfamiliar with opiate use in effective pain control. For severely malnourished patients with hypoproteinaemia and disturbances of renal and liver function, opiates are one of the safer drug groups to use. Appropriately prescribed and titrated, they act as a totally reversible means of relieving pain without necessarily hastening death. The patient to whom Williams et al refer suffered from several different sources of pain (pressure sores, osteoporotic fractures, and sciatic leg pain) and seemed to benefit from a small dose of diamorphine without being sedated.3 If a patient's pain is eased he or she may be more able to comply with the intensive behaviour therapy and psychotherapy necessary for recovery.

Conclusions

We argue that for patients with anorexia nervosa good palliative care is not a last resort. It should not exclude all other specific treatments and could work alongside these to provide optimal care potentially leading to remission or cure. Patients' eligibility for palliative care should be determined by their level of need and not purely by their diagnosis.