When Should Neuroendovascular Care for Patients with Acute Stroke Be Palliative?

Abstract

Noncurative surgeries intended to relieve suffering during serious illness or near end of life have been analyzed across palliative settings. Yet, sparse guidance is available to inform clinical management decisions about whether, when, and which interventions should be offered when ischemic stroke and other neurological complications occur in patients whose survival is extended by other novel disease-modifying interventions. This case and commentary examine key ethical and clinical considerations in palliative neuroendovascular care of patients with acute stroke.

Case

Mr J is a 64-year-old man with metastatic non-small-cell lung cancer (NSCLC), who, while eating, abruptly developed right hemiplegia and aphasia. He had been diagnosed 10 months earlier with NSCLC; his estimated median life expectancy was approximately one year. After a course of chemotherapy with pemetrexed and carboplatin, Mr J started pembrolizumab, an anti-programmed death 1 immune checkpoint inhibitor offered to possibly extend his life but not as a cure for his cancer.^1,2 Since diagnosis, Mr J suffered multiple hematologic complications including thrombosis and hemorrhage. When brought to an emergency department, he was confirmed as full code and intubated on arrival, due to poor mental status and aspiration risk. Computed tomography (CT) of his head and neck revealed normal brain parenchyma and occlusion of the proximal left middle cerebral artery (MCA), which supplies blood to most of the brain’s left hemisphere, including critical areas for language and right-side sensorimotor function.³ The mechanism of Mr J’s left MCA occlusion was presumed to be thromboembolism, to which he was predisposed by hypercoagulability of malignancy, pembrolizumab,^4,5,6 and intracardiac hemostasis, given his known low left ventricular ejection fraction.

After discussion with Mr J’s health care proxy, GG, about his acute stroke as a likely a complication of his cancer, GG consented to Mr J undergoing an emergent thrombectomy. This neuroendovascular procedure is a minimally invasive alternative to more invasive interventions and is the standard method for thrombectomy for acute ischemic stroke with large-vessel occlusion via an endovascular approach. Emergent thrombectomy utilizes femoral artery access to position an intracranial catheter system that permits intracerebral thrombus removal, with goals of restoring blood flow to vascular territory downstream from an occlusion and enabling salvage of the ischemic penumbra—but not yet of irreversibly infarcted brain tissue—to restore neurological function and prevent further impairment.⁷ If thrombectomy is not performed, a large proximal-vessel stroke typically occurs, potentially leading to extended brain tissue infarction, cerebral edema and other symptoms of elevated intracranial pressure (ie, nausea, vomiting, headache, visual changes, and cranial neuropathies) that can exacerbate a patient’s impairment and suffering.^8,9,10,11

Mr J’s thrombectomy was uncomplicated and resulted in rapid and successful left MCA territory reperfusion. Subsequent brain magnetic resonance image (MRI), however, revealed multifocal infarcts affecting the left and the right hemisphere of Mr J’s brain and bilateral cerebellar hemispheres, consistent with his presumed cardioembolic etiology. Mr J was unable to communicate or meaningfully interact. After neurological examination, Dr N informed GG about key findings, including bilateral infarcts expected to produce long-term bilateral weakness, disordered speech, and cognitive impairment. Dr N also explained to GG that Mr J would likely need life support, including tracheostomy, gastrostomy, and rehabilitation if he survived much longer. GG expressed understanding and asked the team to prioritize Mr J’s comfort.

Commentary

More than 1 in 10 patients who present with acute ischemic stroke are estimated to have comorbid cancer.^12,13 As the median survival of patients with cancer improves with novel targeted therapies, frequency of acute stroke and other neurologic complications in this expanding population is expected to rise.^{14,15,16,17,18} Malignancy can predispose patients to ischemic stroke through hypercoagulability, nonbacterial thrombotic endocarditis, systemic treatment effects, or, rarely, tumor embolism or angioinvasion.^14,19 Since many patients seeking emergency evaluation of acute stroke symptoms may have comorbid cancer, clarifying ethical questions in these patients’ stroke care, especially those near the end of life, is key: Should clinicians try to preempt or reverse neurological dysfunction when the end of a patient’s life is near? When, to what extent, and according to whom should thrombectomy for patients with terminal illness be considered palliative? How should palliative or comfort care goals be set to guide appropriate neuroendovascular management decisions in the context of end-of-life care?

Palliative Thrombectomy Goals

Palliative care is defined by the World Health Organization (WHO) as “active total care of patients whose disease is not responsive to curative treatment” and aims to achieve “the best quality of life for patients and their families.”²⁰ The Center to Advance Palliative Care (CAPC) conceives the field as “specialized medical care for people with serious illnesses [that is] focused on providing patients with relief from the symptoms, pain, and stress of a serious illness.… The goal is to improve quality of life … and [palliative care] can be provided along with curative treatment.”²¹ Ethically, noncurative surgery intended to relieve symptoms in patients with serious illness or near the end of life have been analyzed in a range of contexts, particularly in surgical oncology.^{22,23,24,25,26,27,28,29,30,31} However, invasive neurological procedures intended to address indirect complications of terminal illnesses (ie, thrombosis due to hypercoagulability of malignancy) have received little clinical or ethical attention. Especially in clinical neuroscience, little is available to guide neuroendovascular intervention decisions with patients who develop ischemic stroke or other neurological complications near the end of life.

In Mr J’s case, thrombectomy did not appreciably reduce disability near the end of his life, but we argue that Dr N’s team’s decision to perform thrombectomy was ethically justifiable based on its concordance with Mr J’s goals gleaned through conversations with GG. As highlighted by the WHO and CAPC definitions, appropriate palliative care consists not merely in pain control but in the active total care of a patient who strives for the best quality of life. To the extent that neurological symptoms, including sensorimotor dysfunction (eg, weakness and numbness), headache, delirium, aphasia, dysarthria, imbalance, gait disturbance, and cranial neuropathies can detract from quality of life, it is incumbent upon clinicians to diligently address symptoms throughout a patient’s illness.^32,33,34,35 Among patients who experience acute ischemic stroke, more severe neurological impairment has been linked with significantly lower quality of life.^35,36

Neuroendovascular approaches, such as thrombectomy, are specifically intended to attenuate or prevent accumulated neurological disability^37,38,39 and are backed by robust randomized data.^40,41 The location and type of stroke and the extent of salvageable ischemic penumbra must be considered case-by-case in terms of whether foreseeable benefits of thrombectomy outweigh its risks. Eligibility criteria for late endovascular treatment trials for patient-subjects with acute ischemic stroke have included occlusion of proximal MCA-M1 or internal carotid artery (ICA) on CT or MR angiography, a score greater than 6 on the National Institutes of Health Stroke Scale up to 24 hours from time the patient was last seen well, significant ischemic penumbra, and factors such as age, baseline modified Rankin Scale score, and life expectancy. Exclusion criteria have varied by trial and have been a source of practice variation across centers.⁴² Another source of complexity is that criteria assessment and treatment decisions are typically made emergently.

Minimally invasive neuroendovascular interventions can aptly be considered palliative for a patient with limited life expectancy and should not be withheld based solely on a patient’s terminal comorbidity. Even if expected to live only weeks or months, it should be considered that the remainder of a patient’s life with additional stroke-related neurological impairment could exacerbate their end-of-life suffering. Though Mr J did not, many patients who undergo thrombectomy experience reversal of neurologic impairment or return to functional independence, due to their reduced risk of ischemic penumbra.⁴⁰ Successful reperfusion of ischemic penumbra may forestall the development of malignant cerebral edema, cerebral herniation, or other symptoms of increased intracranial pressure, as evidenced by lower rates of decompressive hemicraniectomy since the advent of mechanical thrombectomy.^43,44,45,46 Ethically, doing good and avoiding harm^47,48,49 likely require clinicians to offer palliative thrombectomy, even when a patient has incurable comorbidity.

Decision Sharing

Patient-centered care⁵⁰ requires assessing a patient’s functional status at baseline; their preferences, values, and goals,^51,52,53 perhaps as expressed by surrogates; and their prospects for recovery. Interpreting each stroke not as a discrete new disease, but in the context of a patient’s broader health state, relevant comorbidities, and illness narrative can foster holistic, goal-concordant intervention, and might help avoid unnecessary discontinuity or fragmentation in a patient’s care.^54,55,56,57 Clinicians’ conversations with patients and surrogates should emphasize that estimates of an intervention’s effects are extrapolated from studies in which subjects were drawn not from the unique population of patients with terminal illnesses but from a general population of subjects with minimal preexisting disability. Decision sharing and informed consent requires conveying uncertainty about how well an available evidence base applies to a particular patient.

Equity and Evolving Therapies

Further research is essential and key practical and ethical dimensions of clinical research in patient-subjects with advanced illnesses deserve careful consideration.^58,59,60,61 In one study of persons with metastatic NSCLC, newly diagnosed patients whose care plans integrated early palliative care experienced improved quality of life and mood,⁶² so goal-concordant palliative care should likely be part of studies about interventions that have evolved since 2010.^63,64 Equity as an organizational ethical value requires inclusion of all key stakeholders’ perspectives and goals—curative and palliative—when crafting policy and evaluating downstream implications of decisions to administer or withhold neuroendovascular interventions in individual cases.

Mr J had an acceptable health-related quality of life and—assuming his comparability to otherwise healthy patients with acute stroke—a higher chance of making a functional recovery with treatment than without, at very low procedural risk.⁶⁵ While caution must be exercised in generalizing from studies of acute stroke patients who did not have cancer, data indicate that thrombectomy for acute proximal MCA occlusion stroke may help stroke patients at the end-of-life experience fewer neurologic impairments and with reduced need for aggressive care and institutionalization following a sentinel cerebrovascular event.^64,66,67,68 While not a factor in this Mr J’s case, do-not-intubate (DNI) orders are common and worthy of mention here; DNI orders should not independently influence stroke care decisions “unless otherwise explicitly indicated,” as emphasized by an American Heart Association/American Stroke Association statement (Class IIa recommendation).⁶⁹ Generally, clinicians should express respect for patients’ rights to decline interventions⁷⁰ and should recognize that such interventions can have palliative roles by preventing debilitating neurological impairment and concomitant end of life suffering.^71,72 Palliative radiotherapy, including stereotactic radiosurgery for patients with advanced cancer with brain metastases, has been pursued,^73,74,75,76 as have deep brain stimulation for Parkinson’s disease management near the end of life⁷⁷ and palliative decompressive spinal surgery for patients with metastatic spinal cord compression.^78,79,80,81

Care Planning

Unlike decisions about these specific palliative interventions, however, decisions about stroke care are typically made quickly, given the urgency of acute stroke, its impact on patients’ capacities to participate in decision making, and exquisite time-sensitivity of implementing acute stroke interventions. Although specific neuroendovascular palliative intervention outcomes data are limited, advance care planning should include surrogate designation and discussion about minimally invasive intervention preferences, which could help safeguard value-concordant goal setting and decision sharing later.^71,82

Ethically appropriate palliative neuroendovascular care for patients with acute stroke includes more than pain control and extends to management of distressing physical, spiritual, emotional, and psychosocial symptoms.^83,84 Recognizing the relatively high frequency of neurological complications among patients with terminal illnesses,^{12,85,86,87,88,89} patient-centered palliative neuroendovascular care can be implemented with guidance from the following key ideas in Table 1.

Table 1.

Keys to ethically appropriate palliative neuroendovascular care

Elements of Ethically Appropriate Palliative Neuroendovascular Care
Recognize palliative care as more than pain control, extending to management of potentially disabling, distressing neurologic symptoms83,84,90
Clarify the patient’s (or surrogate’s) values and goals of care91
Avoid assumptions about a patient’s values, preferences or goals92
Discuss intended aims, prospective benefits, possible risks of a neuroendovascular intervention with a patient or surrogate
Explain the range of possible postprocedural outcomes to motivate transparency
Discuss likely outcomes of no neuroendovascular intervention or alternative interventions
Ensure decision making is sensitive to patient preferences, values and goals
Clearly document and communicate decisions to colleagues and care team members

AUTHOR INFORMATION

Dr. Young is a fellow in Neurology at Massachusetts General Hospital. His research examines ethical dimensions and philosophical frameworks underlying standards of care in clinical neuroscience and medicine, and he is devoted to improving translation of novel neurotechnologies for patients with neurological disorders.

Dr. Regenhardt is a neuroendovascular fellow and stroke scientist. His research interests span the spectrum of translation, from understanding white matter infarction to improving stroke systems of care.

Dr. Sokol is a resident-physician in Neurology who plans to pursue a hospice and palliative medicine fellowship after his residency training. His research interests are within neuropalliative care, specifically the adaptation and development of meaning-centered and other novel palliative care models for Huntington and other related neurodegenerative diseases to alleviate spiritual and existential suffering and improve other forms of health-related quality of life.

Dr. Leslie-Mazwi serves as Director of Endovascular Stroke Services at Massachusetts General Hospital. He has worked extensively with stroke systems of care, at state, regional and national level. He is passionate about improving care access to thrombectomy for patients with large vessel occlusions.