Refusal of termination of life support in the intensive care unit of Mulago Hospital (Kampala, Uganda).

INTRODUCTION

We report a case of refusal of termination of life support in a brain dead patient in the Intensive Care Unit (ICU) in Mulago Hospital (Kampala, Uganda), which is a unit with resource limited settings. Ethics approval was obtained from the Ethics Committee of Mulago hospital for the publication of the present case study. The aim of this case study is to pose ethical concerns among developing countries, for the degree of contemplating these challenges that may vary from one place to another, according to resourses availability, educational level and national cultural differences.

CASE PRESENTATION

A 72 year old female Ugandan was admitted at 6 pm to a district hospital more than 150 miles from Kampala with sudden loss of consciousness. Four years earlier, she had been diagnosed with a posterior cerebral aneurysm. Physical examination revealed very high blood pressure and a Glasgow coma scale of 6/15 and unequal pupil sizes. Management for an unconscious patient with hypertensive stroke was initiated. The following day she was transferred to the national referral hospital in Kampala for further care. Upon arrival at 4.45 pm a CT scan of the brain showed massive intracerebral and intra-ventricular haemorrhage with severe cerebral edema. She was received at the ICU at 5 pm. At this time the Glasgow coma scale was 7/15 with systolic blood pressures swinging between 130 and 190 mmHg and diastolic between 70 and 110 mmHg without intervention. She had irregular breathing and was intubated, and started on synchronized intermittent mandatory mechanical ventilation. The following day at 11.00 am the blood pressure crashed to 70/40 and the patient was started on vasopressor support with dopamine. At this time she had declining respiratory effort and was started on continuous mandatory ventilation. One hour later she had evident clinical features of brainstem death: dilated non-reacting pupils, absent corneal, cough, gag reflexes and dolls eye movements. Brain death was further confirmed by EEG at 2 pm which showed flat tracings. The entire medical team which comprised of a neurosurgeon, a neuro-physician, a general surgeon, an anaesthesiologist, an internal medicine physician and the nurse in-charge of the ICU proposed to terminate the life support system at a single meeting. The hospital ethics committee was not consulted since it only handles research issues. The relatives were informed about the findings of brain death by the medical team at 2.30 pm and were allowed to see the patient who was still on the life support machines. They included the husband, two daughters, and two in-law relatives. The information was delivered in English well over 30 minutes and adequate time was allowed for questions. The relatives, who had strong Christian beliefs, understood the explanation of brain death but maintained that since the heart and other organs were still functional, the life support systems should not be switched off. They argued that since it is God who gives and takes life they had no mandate to terminate the life of another. The medical team did not press them to change their mind. The patient was allowed to continue on life support as well as other components of medical care including vasopressor support and feeding. At 5 pm the patient had a cardiac arrest with ECG tracings showing asystole. At this time the medical team decided not to do any cardiac resuscitation using in house operational guidelines which are still in the process of being developed. The relatives were told that the heart had stopped and therefore the patient was dead. They agreed with this new position and were in agreement with switching off the machines. The life support machines were switched off at 5.30 pm.

The patient was a devout Christian housewife and progressive well to do farmer with university training. She was nondrinker of alcohol and non-smoker. She was HIV seronegative and had been a relatively well controlled hypertensive on nifedipine and enalapril. She had history of focal convulsions that were currently controlled by carbamazepine. There was no evidence of psychiatric illness. Following the diagnosis of posterior cerebral aneurysm four years earlier, she had understood the possibility of instant death in the future but had made no advance directives. There were many other relatives, close friends and religious leaders who were consulted by the family during the discussions. All these were in agreement with the husband that the life support systems should not be terminated.

DISCUSSION

The establishment of intensive care units in the United States in the 1960's and 1970's led to the famous cases of Karen Quinlan and Nancy Cruzan. They were in persistent vegetative states following ingestion of drugs or exposure to cold after a road traffic accident respectively. (1) The two cases led to intense legal and ethical debates that largely shaped the approach to unresponsive patients in the US. Similarly, the ability to prolong life of critically ill patients in developing nations has over the years been enhanced by the establishment of intensive care units albeit at a lower pace compared to that of developed nations. In both developed and developing nations, clinicians and proxy decision makers face ethical dilemmas daily. The discussion of this case illustrates the differences between the developed and developing nations.

Lack of advance directives in developing nations

Advance Directives were introduced in the US following the Karen Quinlan case in 1991.1 This was prompted by the fact that Karen's choices for what could be done to her should she become unconscious were not known by her proxy decision makers. Advance directives help to objectively express ones choices for future care when incapacitated. They take several forms including: Living wills, appointment of a surrogate and Durable Power of Attorney for Health Care Matters.

At this hospital and in the country at large, advance directives have not been used. A culture of non-documentation is the norm in many developing nations. Conveying of information by oral tradition is widely practiced and preferred to the written format in all aspects of life. This is the most likely method employed by proxy decision makers in conveying the wishes made by patients before their incapacitation. Because it is not documented, it is not readily verifiable. Before disregarding this method of information conveyance, we should remember that for millennia, African societies have strongly maintained and propagated information through oral tradition in all the major aspects of life.

In many African settings, it is not customary to discuss death and dying issues during ones life openly as it is in the West. Because of this cultural setting, it is unusual for people while they are not sick to discuss and make choices in advance concerning future actions should they become incapacitated to articulate their choices. It is also not usual in these settings, to formally discuss prognostic issues at the bedside. Clinicians working in these conditions are therefore unlikely to be as open as their compatriots in the West in discussing death and dying at the bedside with their patients or proxy decision makers.

No hierarchy of proxy decision makers in developing nations

A proxy decision maker is a person who makes decisions for an incapacitated patient by default when the incapacitated patient has not executed an advance directive. Such a person would not have been delegated by the patient to act in that capacity as in the case of the surrogate. The proxy decision maker is also distinct from a legal guardian who is appointed by Court. The absence of advance directives and rarity of appointment of legal guardians in most African contexts means that proxy decision makers are a vital resource. Since proxies can be determined in various ways, it is helpful to generate a hierarchical order to guide their functions. As an example, the Florida Statutes prefer the following order; first is the patients spouse followed respectively by adult children, patients' parent, adult siblings of the patient, adult relative and lastly close friend. (2) Despite the lack of such statutes in Uganda, it is noteworthy that in the case of our patient, the main proxy decision maker was the spouse assisted by the adult children. However in our case we see the in-laws appearing earlier in the hierarchy before other closer relatives and friends. We also see religious leaders featuring prominently as players in the decision making process especially in regard to interpretations of death and dying. In Africa, there are large families with closely knit networks in the context of polygamy and complex relationships that span beyond the immediate family. They include clan, community and tribal heads, religious heads, friends, peers and workmates and in some cases employers. In most African settings, the role of these individuals is well accepted and their influences on the decision making processes are significant. (3) It would obviously be unrealistic to draw a hierarchical structure for all these individuals. It would however be unreasonable to follow the individualistic approach of Western societies and relegate the communitarian approach of African societies in such processes. Any considerations for drawing a hierarchy of proxy decision makers in the African contexts should find means of accommodating this wider context.

Furthermore there are no legal or institutional definitions for a hierarchical structure of proxy decision makers in clinical matters in Uganda. This contrasts deeply with the US where various states have legal definitions for proxy decision makers. (4)

Conflict among proxy decision makers

It is not unusual to have conflicts among proxy decision makers in the developed world despite the advances made in regulating their roles. (5) Surprisingly, in the unregulated climate as illustrated by this case from Kampala, we didn't have any conflict. This may be partly explained by the highly paternalistic and male dominated society in which the word of the husband was likely to be taken as final and any dissent of the daughters could be culturally ignored. The male dominance could also explain the involvement of the male in-laws, the male religious heads and largely male component among the friends. However gender transitions in many developing nations including Uganda are steadily negating the male dominance by empowering females in many aspects. (6,7) When these gender disparities narrow we may possibly witness similar conflicts as in the West among proxy decision makers. When this occurs in a situation where there are no clear guidelines on the hierarchical structure among the proxy decision makers, greater confusion is likely to ensue.

The definition and conceptualization of death

For most of mans known history, death was taken as cessation of the heartbeat and breathing. This is sometimes called the whole-body criterion or the cardio-respiratory criterion of death. With introduction of artificial respiration and open heart transplantation in the 20th century enabling continuation of life beyond the cessation of ones heart or lung function, newer definitions of death were required. This lead to the concept of brain death. The concept of brain death has been developed in the US since the 1968 Harvard Criteria were operationalized. (8) Later distinctions were made between brain death and persistent vegetative states. (9) The Uniform Determination of Death Act (UDDA) which is operational in most US states (10) provides that: An individual who has sustained either irreversible cessation of circulatory and respiratory functions, or irreversible cessation of all functions of the entire brain, including the brain stem, is dead. (11) Clinicians in developing nations use mostly the cardio-respiratory criteria for determining death. In a few cases, particularly in large referral hospitals, they may use the brain death criteria, as in our case. However, most people in developing nations still understand death to mean cessation of heartbeat and respiration. Most have probably not heard of concepts like brain death. They are also unlikely to know or understand the differences between brain death and persistent vegetative states. This inevitably brings clinicians working in developing nations in immediate conflict with the proxy decision makers as in our case. One of the forces behind refining death definitions in the US was organ donation. (11) In Uganda, and in many African nations, there is no practice of harvesting organs from brain dead patients. The absence of this practice may explain delays in formulating policies based on brain death. Therefore there are immense differences in the legal, moral, and ethical dimensions of death definitions between the US and African nations.

Rationing issues

There remains the danger of setting lower thresholds for terminating life support systems in the context of limited resources. Our hospital has only 12 beds in the general adult ICU that literally serves not only a city of three million people but an entire region that includes neighboring countries. It is a public health care unit and the government meets the costs of health care. The average length of stay is five days with a range of 1-33 days. (12) There was a significant decline in ICU residence among patients with sepsis in US hospitals from an average of 17 days in 1979-1984 period to that of 11.8 days in 1995-2000 period. (13) This observation may mean that clinicians discharge patients from ICU's faster in the Ugandan context than the US.

Medical futility in context of scarce resources

In many African settings, drugs, bed space and human personnel are critical scarce resources whose use may need to be rationed. This becomes particularly so when clinicians are faced with futile medical care. Contemporary ethical thought is that it is rational to withhold scarce resources from those to whom benefit is not expected. (14) In our case, despite the diagnosis of brain death, the clinicians proceeded to give feeding and vasopressor support. There are no written policies in this hospital on medical futility. Even in the US where numerous policies and statutes have been written on medical futility, it is not always clear on what may be done. (15)

This lack of consensus on medical futility should however not deter African nations from dialoguing on best practices applicable to resource limited settings. In all situations, the dignity of the dying patient should be maintained especially in publicly funded health care facilities where every one is entitled to equal access to service. (16)

Lack of institutional guidance documents and bedside ethics committees in developing nations

There is often a gap in developing nations in the aspect of capacity to have documented guidance or policies as in this hospital. It is very common to hear comments like the documents are still in the process of development. There is usually a lack of hospital ethics committees in developing nations yet their presence is standard practice in health units in developed nations. (15) In recent years and as a requirement for international collaborative research, research ethics committees have flourished, not only at institutional but also at national level. (17) However in most clinical settings, ethics committees that address bedside ethical issues are lacking. A hospital ethics committee would be vital in offering expert advisory services to clinicians, patients and proxy decision makers especially in times of conflicts. (14,15)

CONCLUSIONS

The refusal of termination of life support systems by the proxy decision makers of an elderly patient in Kampala illustrates the significant differences in the knowledge, attitudes and practices of the dying processes between African and Western societies. Further studies addressing these differences are warranted to define better approaches to the dying processes in African settings taking into consideration advances that have been made in the West.

REFERENCES

(1.) Comas. Nancy Cruzan and Karen Quinlan in: Classic Cases in Medical Ethics. Second Edition. Gregory E. Pence page 31.

(2.) Florida Statutes 754.401--The proxy.

(3.) Sirkku K. Hellsten. Human Rights in Africa: From Communitarian Values to Utilitarian Practice. Vol 5 No-January 2, 2004 Pages 61-85.

(4.) Legal issues at http://endoflife.northwestern.edu/legal_issues/legal.pdf (Accessed on March 8,2009).

(5.) Council on Ethical and Judicial Affairs 4-a-01. Surrogate Decision Making. Reference Committee on Amendments to Constitution and Bylaws at http://www.ama-assn.org/ama1/pub/upload/mm/369/ceja_4a01.pdf as on 08 March 2009.

(6.) Gender and Women empowerment at www.undp.or.ug (Accessed on March 1, 2009).

(7.) Women's Empowerment and Demographic and Health Outcomes in Uganda Demographic and Health Survey Report by Uganda Bureau of Statistics 2006 (Ed). Pages 238-261.

(8.) Ad Hoc Committee of the Harvard Medical School to Examine the Definition of Brain Death. "A Definition of Irreversible Coma". JAMA 1968;205:337.

(9.) Medical Aspects of the Persistent Vegetative State; The Multi-Society Task Force on PVS.N Engl J Med 1994;330:1499-1508.

(10.) Available at http://www.law.upenn.edu/bll/archives/ulc/fnact99/1980s /udda80.htm.

(11.) Controversies in the Determination of Death in the Presidential Commission on Determination of Death Report: A White paper by the Presidents Commission on Bioethics, December 2008: Page 2-14. Available at www.bioethics.gov.

(12.) Mulago Hospital Monthly Statistics July 2008. Unpublished data.

(13.) Martin GS, Mannino DM, Eaton S, Moss M. The epidemiology of sepsis in the United States from 1979 through 2000. N Engl J Med. 2003; 348 (16): 1546-54.

(14.) Bruce McIntosh. Medical Futility. Northeast Florida Medicine Supplement. Published on January 26, 2008.

(15.) Truog RD. Tackling Medical Futility in Texas. N Engl J Med 2007;357:1-3.

(16.) Universal Declaration of Human rights (1948) at www.who.int./hhr/readings/ declarations/en/index.html (Accessed on March 1, 2009).

(17.) The Uganda National Council for Science and Technology. Established 1990 by Act of Parliament of the Laws of Uganda (209). Accessed at http://www.uncst.go.ug.

Frederick Nelson Nakwagala and Jane Nakibuuka

Intensive Care Unit, Mulago Hospital, Kampala, Uganda

Corresponding author: Frederick Nelson Nakwagala, MD Intensive Care Unit, Mulago Hospital, Box 7051 Kampala, Uganda

Email: [e-mail]