Case Report: Are Clinicians Obligated To Medically Treat a Suicide Attempt in a Patient with a Prognosis of Weeks?

Introduction

A 69-year-old man, Mr. H., was brought to the emergency department via ambulance after family members found him unresponsive in his apartment near open bottles of hydrocodone/acetaminophen and potassium chloride. Naloxone was administered in the emergency department leading to a temporary improvement in the patient's mental status, during which the patient was able to confirm that he ingested an unknown amount of hydrocodone/acetaminophen. He was unable to elaborate a reason for the ingestion. The patient had a known history of a brain mass and a computed tomography (CT) scan of the head performed in the emergency department showed a left frontotemporal mass with surrounding edema and mass effect. He was started on intravenous dexamethasone and admitted to the intensive care unit (ICU) for monitoring after his acetaminophen level was found to be 133 mg/L (a potentially hepatotoxic level given the unknown duration of time after the acetaminophen ingestion). He was also given N-acetylcysteine as treatment of his acetaminophen overdose.

Background

The patient's brother, who was his primary caregiver and health care power of attorney, provided most of the history due to the patient's altered mental status. Mr. H. lived alone. He had a 2-year history of depressive symptoms. The brother attributed Mr. H.'s depressive symptoms to the death of Mr. H.'s partner 2 years ago. While the patient visited with a therapist briefly for a “grief reaction” related to the loss, he was never known to have received any formal treatment for depression. Three months prior to the current admission, Mr. H. suffered a right-sided pontine stroke. During evaluation for this stroke, he was found to have an incidental left frontotemporal mass. Imaging at the time was suggestive of malignancy. Mr. H. declined surgical resection or biopsy to confirm the diagnosis, despite strong recommendations from his physicians.

After this admission, Mr. H. was discharged to a skilled nursing facility to undergo rehabilitation. While there, he often spoke of his desire to return home to spend the rest of his life in more familiar surroundings. Mr. H. also voiced thoughts of wanting “all of this to end” and a desire “to be reunited” with his deceased partner. Despite these concerning statements, mental health consultation was not requested. Per the patient's wishes, he was discharged home from the facility. Over the next 2 months, Mr. H. became more socially withdrawn. His brother later observed that Mr. H. began to experience progressive difficulty with ambulation, speech, and activities of daily living. Despite pleas from his family and caregivers, Mr. H. steadfastly refused to seek medical attention for these deficits.

Difficult Decisions

Upon arrival to the ICU, Mr. H.'s mental status deteriorated and he became obtunded. The ICU team, concerned about ability to maintain his airway, approached his family about the possibility of intubation for airway protection. Mr. H.'s family, who had been supportive of his decision to forgo medical treatment for his brain mass, requested a do-not-resuscitate/do-not-intubate (DNR/DNI) order be written and declined the offer to intubate the patient. They cited Mr. H.'s longstanding wish to avoid medical treatment for his brain mass along with their belief that he would not have wanted any “aggressive” therapies. The ICU team expressed concerns that writing a DNR/DNI order would allow the patient to die prematurely as a direct result of his suicide attempt.

To assist in these complex medical decisions, the ICU team consulted the inpatient palliative care team.

Mr. H.'s progressive neurologic deterioration and functional decline in the weeks prior to admission suggested that he had a prognosis of weeks even without the complication of the suicide attempt. If one accepts that it is a core obligation of medical professionals to prevent patients from harming themselves, then the clinicians caring for Mr. H. faced a clinical and ethical dilemma. How aggressively should they treat a patient with a profoundly poor long-term premorbid prognosis following a presumed suicide attempt? Mr. H.'s had made it clear historically that he did not want any further medical evaluation or care for his probable malignancy. It must be assumed, based on his providers' previous respect of care wishes, that the patient had been judged competent to refuse further evaluation and treatment of his brain tumor.

Consequently, Mr. H.'s physicians were faced with defining and implementing a compassionate and ethical care plan to address the medical sequelae of his suicide attempt in the context of his probable malignancy, poor clinical prognosis, and previous stated wish not to receive medical care. Which medical interventions are the clinicians ethically obligated to perform in the context of his assumed suicide attempt? Does a clinician's duty to prevent suicide necessitate the use of therapeutic antidotes (N-acetylcysteine) or aggressive medical interventions (endotracheal intubation or cardiopulmonary resuscitation) if such interventions cannot restore a patient to a “meaningful” existence and quality of life as defined by the patient? These were some of the profound questions which the palliative care and intensivist services were confronting in the care of this patient.

Discussion

A number of important issues regarding depression, despair, and suicidality in terminally ill patients were raised in this case. We will briefly review the epidemiology and risk factors for suicidality in patients with life-limiting illnesses and discuss the ethical challenges raised by this case.

Completed suicides by the terminally ill appear to be rare, making up 2%–4% of completed suicides. ¹ However, the rates of completed suicide, suicide attempts and suicidal ideation in those with advanced illnesses are increased over the general population. ² Rates of suicidality rise with increasing co-morbidity and burden of disease.^3,4 Advanced age, male gender, having acquired immune deficiency syndrome (AIDS), a family history of suicide, and comorbid psychiatric illness are all associated with increased suicide risk. ⁵

Patients with malignancy are at risk for suicide. Data suggest that patients with cancer patients may commit suicide at rates twice that of the general population. ⁶ In addition, up to 8.5% of terminally ill cancer patients express a sustained and pervasive wish for an early death, ⁷ and 7.8% of cancer patients report suicidal thoughts. ⁸ Increased physical symptom distress, poor performance status, feelings of hopelessness or being a burden to others, diminished spiritual well-being, and concerns about loss of independence have all been associated with increased rates of suicidal ideation or a desire for hastened death in cancer patients.^7–11

An important role of the physician is to identify suicidal patients and to intervene clinically to prevent self-harm. At times this intervention may require that patients be detained, hospitalized, and treated against their will. ¹² A suicidal patient's decision-making capacity may be impaired by major depression or other mental illnesses. As a consequence, actions taken to harm one's self are not considered to be autonomous but rather an unfortunate and dangerous outcome of mental illness. ¹³ Because it is thought that a suicidal patient's mental illness is negatively affecting their decision-making capacity, society has deemed that the physician has the clinical responsibility to prevent suicide and protect the patient. This duty may supersede the patient's right to autonomy.

Interventions to prevent self-harm, whether initiated before or after a suicide attempt, offer further time and opportunities to evaluate the level of the threat, further ascertain the patient's decision-making capacity, and treat the patient's underlying mental illness or other source(s) of suffering. The fact that the patient may have an underlying premorbid life-limiting illness, such as a malignancy, has not been viewed historically as reason to alter a physician's responsibility to prevent self-harm. A notable exception to this is where legally sanctioned, medically assisted suicides occur after a deliberative and evaluative process such as under the Oregon Death with Dignity Act. ¹⁴ However, even under the Oregon Death with Dignity Act, physicians are instructed to refer patients who are suffering depression for counseling to ensure the patient's depression or other psychological illness is not causing impaired judgment before proceeding with the process of physician-assisted suicide. ¹⁴

Interventions to prevent suicide vary in degree and should attempt to appropriately balance the seriousness of the threat with the restriction on the patient's liberty. Some examples include maximizing social support from family, friends, and the community. It is also important to remove potential means in which patients may harm themselves. If there is a comorbid psychiatric illness, arranging for voluntary psychiatric treatment in willing patients experiencing suicidal thoughts is an important treatment option. If a patient who represents a potential danger to himself is unwilling to undergo psychiatric treatment; one must consider the appropriateness of seeking involuntary treatment. All of the aforementioned interventions have as their goal the successful treatment of the precipitant(s) that have culminated in the patient's desire to seek death.

In this case there was sufficient evidence to suggest that Mr. H.'s presentation represented a suicide attempt occurring in the context of depressive symptomatology and perhaps an emerging encephalopathy from his brain tumor. Therefore, based on the previously established ethical and professional standards of suicide prevention, attempts to reverse the acute toxicities resulting from the overdose would be reasonable. However, in the case of Mr. H., these treatments would be provided in direct contrast to his previously voiced sentiments that he would not want to be admitted to the hospital again or receive any “aggressive” forms of medical care. This conflict became evident early in Mr. H.'s presentation as there was a concern that he may require intubation for airway protection and require antidotal treatment and supportive care for acetaminophen toxicity as direct consequences of a suicide attempt. The ICU clinicians were understandably anxious with the family's request to limit treatment.

In our clinical analysis, Mr. H.'s poor prognosis should be viewed as confounding the ethical and societal precept that all cases of attempted suicide require treatment in an effort to correct the self-harm. Given his history of steady neurologic decline and previously stated clear wish not to have his tumor treated, which all involved in his care respected, he probably only had weeks to live even in the absence of the suicide attempt. It is also likely that those weeks would involve worsening debility, encephalopathy, and communication impairment. None of these negative consequences of his underlying medical illness would be mitigated by treating his suicide attempt. With such a prognosis, it may be argued that the utility of medical interventions to reverse a suicide attempt is much less compelling than in a similar medical scenario in which the patient's underlying medical prognosis was felt to be months or years. Furthermore, Mr. H. historically had declined aggressive medical therapy offered in the hopes of briefly extending his life. Rescinding his expressed DNR/DNI order and implementing advanced forms of life support such as mechanical ventilation were unlikely to restore the patient to health in a way that would have been acceptable to him. In fact, medical care to reverse the suicide attempt could have been viewed as potentially harmful by prolonging dying in a way incommensurate with the patient's previously expressed sense of a dignified death.

Due to these considerations, the palliative consultation service concluded that Mr. H.'s medical circumstances created a need to balance the physicians' obligation to prevent suicide and their desire to respect a patient's previously expressed wish in the context of a devastating underlying illness, as opposed to focusing on just one of these obligations. Even if such therapies would be indicated to support a patient through the initial period after a suicide attempt to prevent its success, they may be viewed as not obligatory or even appropriate in this specific situation. Had we known of the patient's suicidal ideation and subsequent plans to attempt suicide, a multitude of protective and potentially therapeutic interventions would have been appropriate. However, intervening with medical treatments which were nonrestorative and objectionable to the patient (and his family) was felt to be inappropriate by the treating physicians, health care power of attorney, and the patient's family even if this meant his lifespan was foreshortened by a small amount time by the suicide.

Battin ¹⁵ and other ethicists have alluded to an ethical principal that allowed physicians to consider what would be the “least worst death” in the calculus of their medical decision-making. Accordingly, it may be morally wrong for clinicians to treat or reverse potentially fatal medical conditions if doing so only puts the patient on a path to a more painful or difficult death. ¹⁴ By this analysis, it could be argued that in the case of Mr. H. it would have been most appropriate not to treat his acetaminophen toxicity, acknowledging the possibility the patient would directly die from an acetaminophen overdose.

Encountering terminally ill patients for whom it may not be appropriate to intervene to prevent the success of a suicide attempt is fortunately rare. It has been our experience that most serious suicide attempts in terminally ill patients are either immediately successful or that patients receive aggressive life-prolonging interventions in the initial aftermath of an attempt without question. Withdrawal of treatments to prevent the success of a suicide attempt may be appropriate if certain criteria are met: the life-prolonging treatment is unlikely to restore a patient to a sufficient level of health due to a poor prognosis from the underlying terminal illness, there exists a sufficient level of certainty of a short prognosis, and the withdrawal of life sustaining interventions is consistent with a competent patient's previously stated wishes.

As events unfolded in the case of Mr. H., a “middle-ground” plan was implemented with the support of the family, ICU and palliative care teams. A DNR/DNI order was placed. Mr. H received supportive treatments and monitoring which were acceptable to the family including intravenous fluids, N-acetylcysteine, and dexamethasone. In retrospect, providing nonburdensome interventions targeted toward the treatment of the suicide attempt (i.e., N-acetylcysteine) while forgoing medical interventions targeted to prolong Mr. H.'s life from his malignancy may have been the most appropriate care plan.

Mr. H.'s family hoped his ability to communicate would improve, at least transiently, so they could say goodbye. Unfortunately, his mental status never improved sufficiently for meaningful conversations. He remained bed-bound with minimal oral intake. Psychiatry was eventually consulted and dismissed his psychiatric emergency detention as it was felt that the patient would not benefit from psychiatric hospitalization given his medical status and poor prognosis. Hospice care was arranged and Mr. H. was discharged to a residential hospice facility on hospital day 7. He died 10 days after discharge secondary to complications from his brain mass.

History revealed that Mr. H. was showing signs and symptoms of a depressive disorder ever since the death of his partner 2 years prior. A clinical assessment of his depressive symptoms could have helped identify an underlying depressive disorder and led to appropriate treatment. Further assessment exploring spiritual, religious, and existential concerns may have identified a sense of loss, abandonment, and difficulty finding meaning in his life as he adjusted to his own terminal illness and his partner's death. There appears to have been missed opportunities, both at the time of his partner's death, the diagnosis of his brain mass, and prior to his self-discharge from his rehabilitation program to intervene psychiatrically and potentially improve his quality of life and avoid his subsequent suicide attempt.

Mr. H. had repeatedly expressed goals to avoid further medical evaluation and to die at home. These goals would have been well served by enrollment in a hospice program. Hospice services were not offered to Mr. H. until the final days of his life after his suicide attempt. Unfortunately, by this time it was too late to create a safe home care plan. Acknowledging that Mr. H.'s depression may not have resolved with therapy and the uncertainty that Mr. H. would have accepted hospice services earlier in his disease trajectory, it is probable that a multidisciplinary hospice team with its attention to safety, emotional well-being, and spiritual needs could have prevented his suicide attempt and helped him meet his stated goal of dying at home. This appears to be another missed opportunity to potentially prevent this sad and unfortunate outcome.