Palliative Care and Strokestroke is recognized and treatment commenced,...

Palliative Care and Stroke

Peter Eastman and Brian Le

Contents1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

2 Palliative Care and Acute Stroke . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32.1 Palliative Care Recognition Within International Stroke Guidelines . . . . . . . . . . . . . . 32.2 Integration of Stroke and Palliative Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52.3 Specialist Palliative Care Utilization in Acute Stroke: A Review of the Current

Literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62.4 Shared Stroke and Palliative Care Issues in the Acute Setting . . . . . . . . . . . . . . . . . . . . 7

3 Palliative Care in the Post-Stroke Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143.1 Pain Syndromes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143.2 Post-Stroke Depression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163.3 Quality of Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173.4 Caregiver Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

4 Conclusions/Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

AbstractThere have been significant advances in theprevention and management of stroke overthe last few decades. Despite these important

developments, stroke, both in the acute andchronic phases, remains a major cause of mor-bidity and mortality. The value of integratingpalliative care principles and practices intostroke care management is being increasinglyrecognized across a range of domains includ-ing symptom management, assistance withcomplex decision-making, discharge planning,and end-of-life care. This chapter will explorethe logistics, benefits, complexities, and chal-lenges associated with the evolving relation-ship between stroke and palliative careservices.

P. Eastman (*)Department of Palliative Care, Barwon Health, NorthGeelong, VIC, Australia

Department of Palliative and Supportive Care, RoyalMelbourne Hospital/Melbourne Health, Parkville, VIC,Australiae-mail: [email protected]

B. LeDepartment of Palliative and Supportive Care, RoyalMelbourne Hospital/Melbourne Health, Parkville, VIC,Australiae-mail: [email protected]

# Springer International Publishing AG 2018R. D. MacLeod, L. Block (eds.), Textbook of Palliative Care,https://doi.org/10.1007/978-3-319-31738-0_59-1

1

mailto:[email protected]

mailto:[email protected]

https://doi.org/10.1007/978-3-319-31738-0_59-1

1 Introduction

There have been significant advances in the acutetreatment and rehabilitation of stroke over the lastfew decades. Contributing factors includeimproved early detection/recognition, timely hos-pital presentation, the use of thrombolytic andendovascular therapies, early initiation of rehabil-itation, and the introduction of organized strokecare. Ongoing improvements in stroke preven-tion, treatment, and rehabilitation are crucial asstroke prevalence is expected to increase with theaging population. The lifetime risk of having astroke has been reported to be approximately onein five for females and one in six for males(Mozaffarian et al. 2015) although this figurewas calculated from a predominantly Caucasiancohort within the United States and may thereforenot be easily generalizable. Stroke incidencevaries between countries and is influenced by arange of health, cultural, geopolitical, and socio-economic factors. Globally the burden of stroke isincreasing as measured by the absolute number ofpeople affected, stroke survivors, and disability-adjusted life years lost (Feigin et al. 2014). Muchof this burden is seen in low- to middle-incomecountries.

Strokes constitute a heterogeneous group ofconditions which can be described by mechanismand/or location of injury. Approximately 80–85%of strokes are ischemic in nature, with the remain-der being either intracerebral hemorrhage (ICH)or subarachnoid hemorrhage (SAH). ICH gener-ally occurs following arterial wall rupture in ves-sels weakened by chronic hypertension whileSAH typically follows aneurysmal rupture withinthe circle of Willis or trauma (Simmons and Parks2008). Ischemic subtypes include thromboticstrokes due to occlusive arterial atheroscleroticdisease, embolic strokes which are typically car-diac in origin, and lacunar infarcts which occurfollowing the occlusion of the small arteries thatsupply deep brain structures. Broadly speakingthe extent of injury or damage associated with astroke will depend upon the blood vesselsinvolved and the areas and extent of brainaffected.

The most important factor in the initial man-agement of acute stroke is time. The sooner astroke is recognized and treatment commenced,the better the outcome is likely to be. Another keyfactor in optimizing chance of recovery is venueof care. There is clear evidence that patients whoare managed in a designated stroke center or unitwith a specialist multidisciplinary team do betterthan patients managed elsewhere. In a Cochranereview of 28 trials and 5855 participants whichcompared stroke care unit (SCU) care with alter-native care, SCU was associated with significantreductions in the odds of death, dependency, and/or institutionalized care (Stroke Unit Trialists’Collaboration 2013). The benefits of managementwithin a SCU have been demonstrated for all sub-types of ischemic strokes (Smith et al. 2010a) aswell as for ICH (Langhorne et al. 2013).

After initial cardiorespiratory stabilization,immediate stroke management is individualizedaccording to a range of factors including type andseverity of stroke, presence of contraindications,medical comorbidities, and patient and familywishes. Following an ischemic stroke in appropri-ate patients who do not have contraindications,the initial goal is to provide intravenous (IV)thrombolytic therapy as quickly as possible. Theprovision of a single dose of IV recombinanttissue plasminogen activator (IV rt-PA) within4.5 h of the commencement of symptoms hasbeen associated with a substantially improvedchance of independent function at 3 months poststroke (Maldonado et al. 2014). The use of endo-vascular therapies following ischemic stroke is anarea of ongoing research and development. Ther-apies under this umbrella include endovascularthrombolysis, thrombectomy, and stent retrievertechnology. It has been suggested that theseapproaches may provide superior recanalizationin situations such as proximal vessel occlusionwhen systemic thrombolysis may be less effica-cious although until recently results from random-ized trials were mixed. Two recently publishedtrials however, both of which were stopped earlydue to efficacy, have shown significant improve-ments in functional outcome (Campbell et al.2015; Goyal et al. 2015) as well as mortality(Goyal et al. 2015) for the combination of

2 P. Eastman and B. Le

endovascular thrombectomy and IV rt-PA com-pared to IV rt-PA alone, in ischemic stroke withproximal cerebral arterial occlusion.

The management of ICH involves acute stabi-lization including management of blood glucoseand temperature, careful blood pressure control,quality nursing care, prevention of complications,early rehabilitation, and prevention of recurrenthemorrhage (Hemphill et al. 2015). Time isagain crucial as it is common for deterioration tooccur in the period soon after an ICH. While thebenefits of surgical removal of hemorrhage formost supratentorial ICH have not been proven,prompt surgical management is recommendedfor patients with a cerebellar ICH who are deteri-orating neurologically or have brain stem com-pression (Simmons and Parks 2008; Hemphill etal. 2015). Emergent therapies under investigationin ICH include minimally invasive surgical tech-niques and the use of biological neuroprotectiveagents.

Despite advances in prevention, acute treat-ment, and rehabilitation, stroke remains a promi-nent cause of morbidity and mortality worldwide.It is a major cause of disability in the United Stateswhere up to a third of patients require admission toa long-term care facility following stroke(Mozaffarian et al. 2015). A range of psycholog-ical and physical sequelae have been reported bystroke survivors including anxiety and depres-sion, pain syndromes, and communication diffi-culties. Not surprisingly high levels of distress andcaregiver burden are also common among rela-tives and loved ones. The ideal model for manag-ing the important physical and psychosocial needsof stroke survivors and their families is yet to beestablished, and it is not clear where palliativecare services sit in this paradigm.

While mortality rates vary internationally andhave decreased over the last 20 years (Feigin et al.2014), stroke remains the second commonestcause of death worldwide (World Health Organi-zation 2014). A range of important and diverseneeds have been reported by dying patients andtheir families including symptom management,psychosocial support and assistance with prog-nostication, treatment decision-making, andfuture planning. The utilization of palliative care

principles and practices would seem appropriatein meeting these needs, and there is increasingrecognition, including from international guide-lines, of the importance of integrating palliativecare into stroke care (National Stroke Foundation2010; Holloway et al. 2014; Casaubon et al. 2016;Intercollegiate Stroke Working Group 2016).However a range of challenges to effective inte-gration have also been identified from both astroke and palliative care perspective. The aim ofthis chapter is to explore the relationship betweenstroke services and palliative care in both theacute and longer-term setting. The current levelof involvement and proposed models of integra-tion will be discussed as well as some of thecommon palliative care needs and issues identi-fied by patients, families, and health profes-sionals. It is hoped this will provide acontemporaneous overview of the current situa-tion and highlight some important areas for futurethought, research, and investment.

2 Palliative Care and AcuteStroke

There is growing recognition that some patientsand families following an acute stroke have needsthat will be best met by integrating palliative carephilosophies and practices into their strokemanagement.

2.1 Palliative Care RecognitionWithin International StrokeGuidelines

There is increasing recognition of the relevance ofpalliative care principles for some patients follow-ing a stroke within published international strokeguidelines. While the depth and breadth of infor-mation varies across guidelines, there is generalagreement concerning the importance of high-quality end-of-life care within stroke servicesand some support for palliative care throughoutthe stroke care trajectory.

In a comprehensive scientific statement fromthe American Heart Association/American Stroke

Palliative Care and Stroke 3

Association, Holloway and colleagues exploredthe dynamics, logistics, and complexities associ-ated with palliative and end-of-life care in stroke(Holloway et al. 2014). The authors advocate forthe availability of palliative care for all patientsand their families following a severe or life-threat-ening stroke, throughout the disease course. Intheir proposed model, the management of mostpalliative care problems would be carried out bythe stroke service multidisciplinary team withspecialist palliative care services available forsecondary consultation in situations of heightenedcomplexity. Importantly, the authors provide evi-dence-based recommendations and guidance toaid implementation, including practical educationregarding communication skills, goal setting, andsymptom assessment and management.

The Canadian Stroke Best Practice Recom-mendations emphasize the importance of foster-ing palliative care expertise within stroke centers(Casaubon et al. 2016). The guidelines make adistinction between palliative care and end-of-life care noting that palliative care can occur incombination with life-prolonging therapies and isnot reserved only for those who are imminentlydying. A palliative approach is suggested in thesetting of catastrophic stroke or multiplecomorbidities in order to support both patientand family. Replicating the recommendations ofHolloway and colleagues (2014), the guidelinesmake clear that the stroke service multi-disciplinary team should have the appropriate pal-liative care skills to support dying stroke patientsbut also that there be access to specialist expertise.Specific indications for specialist palliative carereferral include the management of unstablesymptoms and assistance with decision-makingand with managing complex family and psycho-social dynamics. Recommendations are also madein relation to Advance Care Planning (ACP), spe-cifically that stroke survivors, their families, andcaregivers should be supported to participate in it.

In Australia, the 2010 Clinical Guidelines forStroke Management recommend that all patientsdying following a stroke have access to care that isaligned with palliative care principles and prac-tices (National Stroke Foundation 2010). Thisincludes consideration of physical, psychosocial,

spiritual, and cultural needs as well as guidancewith prognostication and the diagnosis of dying indeteriorating patients or those following a severestroke. The guidelines note that in many less com-plex situations, this care will be provided bystroke service staff, and these staff will requireongoing education and support in relation toend-of-life care. The importance of clear andtimely communication between stroke services,patients, and families is stressed, and family meet-ings are suggested as an appropriate forum forthese discussions.

The 2016 fifth edition of the UK National Clin-ical Guideline for Stroke include, among their keyrecommendations, the proposition that multi-disciplinary stroke teams should consider high-qual-ity end-of-life care as a core component of theirwork (Intercollegiate Stroke Working Group2016). The need for enhanced education and sup-port for stroke service staff in palliative care princi-ples is emphasized, as is access to specialistpalliative care and the opportunity for timely trans-fer to home if this is the desirable outcome. Amongchanges from the previous iteration, the 2016 guide-lines caution against imposing burdensome restric-tions upon patients dying from strokewith particularmention made about the pragmatic management ofimpaired swallowing and oral intake at end of life.The guidelines acknowledge the range of physicaland psychological stressors that can occur followinga large stroke and propose that the appropriate man-agement of these problems can help ameliorate dis-tress for patients and their families at end of life.

When considered together, a number of com-mon themes are evident across the guidelines.These include:

• General support for palliative care principles askey components of stroke care.

• Recognition of the presence of multi-dimensional symptom issues for many patientsand their families following a large stroke.

• The importance of clear and ongoing communi-cation between stroke services, patients, andfamilies particularly in relation to issues such asprognostication, goals of care, and management.

• Support for a service model in which the multi-disciplinary stroke team provides the majority


of palliative and end-of-life care with second-ary support available from specialist palliativecare services.

2.2 Integration of Stroke andPalliative Care

Although integrating palliative care and acutestroke services is considered important forpatients, families, and health professionals alike,how this integration might actually take place isunder-researched and not without its challenges.Establishing the palliative care needs of strokepatients and their families is a crucial first step inan integrated approach; however data addressingthis question are limited and little is known abouthow these needs might change over time (Stevenset al. 2007).

Potential methods to assess palliative careneeds might include triggers built into existingstroke care pathways – simple bedside promptsor more formal targeted needs assessment tools.Creutzfeldt and colleagues (2015) showed that theuse of a simple four-question “palliative careneeds checklist” during neurology ward roundswas an effective prompt for recognizing and meet-ing patient and family needs. Questions in thechecklist addressed the presence of distressingsymptoms, how the patient and family were cop-ing, and whether the goals of care or treatmentapproach needed to be modified. Burton and col-leagues (2010) used the Sheffield Profile forAssessment and Referral for Care (SPARC) toolto identify high levels of palliative care needs in aprospective study of 191 consecutive strokeadmissions to two UK hospitals. The SPARCwas developed to screen patients with a range ofadvanced diseases for specialist palliative care(SPC) referral and incorporates five domains cov-ering physical, psychosocial, spiritual, and func-tional issues. The authors concluded that the useof the SPARC tool particularly in more disabledstroke patients provided a valuable trigger for staffto consider a range of palliative care issues.

In some ways acute stroke and palliative caremight not be considered particularly compatible.

The acute stroke environment is one of high-intensity neuro-restorative care with an emphasison survival, while palliative care is commonlyconsidered more meditative in approach withless concern placed upon survival and more onquality of life. Challenges in introducing pallia-tive care into general stroke management arehighlighted by qualitative studies in which strokeunit staff have questioned the juxtaposition of asimultaneous curative focused and palliativeapproach (Gardiner et al. 2013) and have viewedpalliative care as being predominantly about end-of-life care and representative of failure of strokemanagement (Burton and Payne 2012). Beyondthe acute stroke phase, categorizing a patient as“palliative” has been identified as a barrier tocombining rehabilitation and palliative care (Bur-ton and Payne 2012).

Traditional palliative care models developedpredominantly for patients with malignant dis-eases may not be easily transferrable to stroke.With its typically sudden onset and decidedlyunpredictable clinical course, stroke may behavedifferently to many cancers or other chronic non-malignant conditions in which palliative care iscommonly involved. Such differences werehighlighted by a study of 544 patients admittedto a tertiary SCU that found more than 50% of the87 patients who died had been completely inde-pendent and well immediately prior to their stroke(Eastman et al. 2013). Caregivers of stroke survi-vors have also been noted to have different needswhen compared to those of patients with cancer.They are typically older and frailer and as a resultmay lack important support and social networks(Stevens et al. 2007).

Despite a current lack of clarity concerning thecharacteristics and function of palliative carewithin acute stroke services (Burton and Payne2012), there does appear to be general acceptanceof a pragmatic service model in which the multi-disciplinary stroke team provides the majority ofpalliative and end-of-life care with secondary sup-port available from SPC (Holloway et al. 2014;Creutzfeldt et al. 2015; Casaubon et al. 2016).SPC services are also ideally placed to providethe necessary education, training, and support toenable the provision of high-quality nonspecialist


palliative care. Importantly, for any model to beapplicable to patients throughout the diseasecourse as has been suggested (Holloway et al.2014; Creutzfeldt et al. 2015), recognition willbe required from stroke services that palliativecare principles can be positive adjuncts to dis-ease-modifying therapies rather than simplyapplying to end-of-life care. Additionally amodel combining stroke and palliative care willneed to consider important system issues such asavoiding overburdening already-stretched pallia-tive care services and not deskilling stroke unitstaff (Stevens et al. 2007).

Using qualitative data obtained from patients,families, and stroke service staff, Burton andPayne (2012) constructed a theoretical frameworkon how the integration of stroke services andpalliative care might occur. They proposed thatintegration was underpinned by six key mecha-nisms, namely, clinical legitimacy, capacity, fam-ily engagement, early integration, recognition ofcomplexity, and recognition of dying. Thesemech-anisms could themselves be influenced by a rangeof clinical and organizational factors which varyaccording to the existing structures and processeswithin individual healthcare settings. Burton andPayne (2012) propose that analysis of the relation-ships and interactions influencing these six keymechanisms will facilitate service developmentand improve the care of patients and familiesfollowing stroke.

Overall the development and optimization ofintegrative models will depend not only on thewillingness of the stroke and palliative care sec-tors to embrace the process but also organizationaland governmental support to ensure adequatefunding and support. Underpinning the wholeprocess is the requirement for ongoing collabora-tive quantitative and qualitative research focusingnot just on outcomes but also factors such as needsassessment, appropriate patient stratification,timing and indication for SPC referral, cost/bene-fit analysis, and patient, family, and health profes-sional experiences.

2.3 Specialist Palliative CareUtilization in Acute Stroke: AReview of the Current Literature

Data quantifying the level of SPC involvement instroke are limited and predominantly focused oninpatient consultation services. While variabilityis seen between health organizations internation-ally, there is evidence to suggest that the overalluptake of palliative care services for patients fol-lowing an acute stroke is increasing. In a largecross-sectional study investigating palliative careutilization in nontraumatic intracerebral hemor-rhage (ICH), Murthy and colleagues (2016) ana-lyzed admissions to over 1000 American hospitalsusing the National Inpatient Sample, the largestinpatient health database in the United States. Ofthe 311,217 included admissions, 32,159 (10.3%)received palliative care, with a substantial annualincrease in palliative care involvement seen (4.3%in 2007 to 16.2% in 2011). In an analysis ofadministrative data of 4894 patients who diedwithin 30 days of an ischemic stroke, 23% of allpatients were enrolled in hospice for end-of-lifecare, a threefold increase from previously reportedrates (duPreez et al. 2008).

A small amount of research has explored pre-dictors for SPC involvement in patients followingstroke as well as differences between referred andnon-referred patients. The influence of both socio-demographic and clinical factors on referral pat-terns has been reported with data typicallyobtained from either large administrative datasetanalyses or retrospective cohort studies.

In an analysis of palliative care involvementfor patients following nontraumatic ICH (Murthyet al. 2016), significant predictors for palliativecare involvement included ICH severity, associ-ated comorbidities, female gender, advanced age,hospital location, and ethnicity. Several of thesefactors were also predictors for hospice utilizationin 4894 patients who died within 30 days of anischemic stroke (duPreez et al. 2008). Predictorswhich increased the likelihood of hospice utiliza-tion included older age, female gender, dementia,


Caucasian descent, and hospital length of staygreater than 3 days. Mechanical ventilation,gastrostomy, and uncomplicated diabetes mellitusas a comorbidity correlated with decreased hos-pice enrolment.

Holloway and colleagues (2010) reviewed allpatients referred to an inpatient palliative careconsultation service within a large US teachinghospital over a 3-year period. Compared topatients with cancer or common nonmalignantdiagnoses including chronic obstructive pulmo-nary disease (COPD), stroke patients tended tobe more functionally impaired, less likely tohave decision-making capacity, and were morelikely to die in the hospital. For the 1551 strokepatients analyzed during the study period, 6.5%received palliative care consultation. Strokepatients seen by the palliative care service wereolder, stayed longer in the hospital, and were morelikely to be discharged to hospice. However, themajority of stroke patients who died or weredischarged to hospice were not seen by SPC.

In an Australian review of 544 admissions overa single year to a metropolitan, tertiary SCU, justover 11% of patients were referred to SPC (East-man et al. 2013). The predominant reason forreferral was end-of-life care, and a number ofpredictors for referral were identified. Theseincluded female gender, older age, increased dis-ability pre stroke, ICH, and living alone or in aresidential aged care facility prior to stroke. Six-teen per cent of all SCU admissions died duringthe review period with just over half seen by thepalliative care team. This is approximately doublethe rate reported from both Swiss and Irish strokeservices where only 26% and 24%, respectively,of decreased stroke patients received palliativecare consultation (Mazzocato et al. 2010;Ntlholang et al. 2016). Stroke patients referred tothe Australian SPC who died during the admis-sion were older, more disabled, and had a signif-icantly longer length of stay than those notreferred. In contrast to the total SCU population,the proportion of patients with an ICH was higherin deceased stroke patients who did not receivepalliative care consultation (50% vs. 37%). Thisin combination with the significantly shortermedian survival of non-referred patients

(2.5 days vs. 6 days) supports previous observa-tions that stroke service staff are generally com-fortable managing clearly terminal patients(Rogers and Addington-Hall 2005).

In a retrospective review of 54 consecutivedeaths on an Irish teaching hospital stroke serviceover a 2-year period, several differences werefound between patients referred and not referredto SPC (Ntlholang et al. 2016). Those patientswhose cause of death was judged to be unrelatedto their stroke were more likely to be seen by theSPC team. These patients also had a longermedian time between stroke and death, againsupporting the notion that stroke unit staff aremore comfortable managing the palliative careneeds of patients who die soon after a severestroke, and less so in those who survive longer.

Taken together these data suggest that referralsto SPC, particularly to inpatient consultation ser-vices, following acute stroke are increasing. Thismay reflect increased recognition of the benefitsassociated with palliative care in areas such assymptom management, complex end-of-life deci-sion-making, and communication. Referral toSPC is influenced by a range of clinical, socio-demographic, and health service factors, but notsurprisingly stroke patients referred to palliativecare tended to be older, frailer, and sicker. Impor-tantly the majority of stroke patients, includingthose who were severely disabled or died, werenot referred to SPC. While epidemiological dataregarding referral rates and predictors are usefulfor service provision and planning, they do notanswer important questions such as whether allpatients with complex needs who would benefitfrom SPC input are currently being referred orwhether non-referred “less complex” strokepatients are receiving timely and appropriate non-specialist palliative care (Stevens et al. 2007).This represents an area for ongoing research.

2.4 Shared Stroke and PalliativeCare Issues in the Acute Setting

2.4.1 PrognosticationThe prediction and communication of prognosis isan important but inherently difficult part of stroke


care. Given the majority of patients will survive astroke, prognostication is relevant to all stages ofstroke care from initial acute presentation to post-acute recovery and finally to rehabilitation and/ordischarge. Accurate prognostication provides aroad map to guide treatment decision-makingand allows patients and families to plan for thefuture. Inaccurate prognostication runs the risk ofdepriving some patients of the best chance ofrecovery while exposing others to burdensometreatments that may only prolong suffering. Inessence prognostication attempts to providesome degree of certainty while being at the sametime inherently uncertain. In a qualitative studythat examined patient and family members’ per-spectives of acute stroke care, honest discussionsaround prognosis despite uncertainty were valuedby relatives even when it was predicted that prog-nosis might be poor (Payne et al. 2010).

Studies investigating prognosis followingstroke have tended to focus on mortality morethan functional outcome or recovery. Numerousclinical, sociodemographic, and institutional pre-dictors for mortality have been identified, withstroke severity and older age generally reportedto have the strongest associations (Smith et al.2010b). The National Institutes of Health StrokeScale (NIHSS) provides a quantification of strokeseverity and is a highly reliable predictor of mor-tality after stroke both on its own and in combi-nation with other variables (Smith et al. 2010b;Frontera et al. 2015). It is a 15-item validatedbedside assessment tool which incorporates test-ing of consciousness, language, sensation, andmotor function.

Stroke type itself is also a strong predictor ofmortality. This is illustrated by a Danish study of39,484 hospitalized stroke patients in which intra-cerebral hemorrhage (ICH) was associated with afourfold increased risk of dying compared toischemic events in the immediate post-strokeperiod (Andersen et al. 2009). Interestingly thedifference in mortality between the two stroketypes progressively decreased over time until at3 months, stroke type no longer correlated withmortality. Temporal differences in the factorsassociated with mortality have also been reportedafter ischemic stroke. In a Canadian study of 3631

patients following ischemic strokes, stroke sever-ity was found to be a significant predictor ofmortality at all time points, while clinician expe-rience was only significant at seven and 30 daysand age and medical comorbidities only signifi-cantly associated with 30-day and 1-year mortal-ity (Saposnik et al. 2008).

Numerous multivariate prediction models foroutcome and mortality following stroke are avail-able although the quality and clinical utility ofthese models varies (Holloway et al. 2014).Examples include the Hunt-Hess scale for sub-arachnoid hemorrhage and the ICH score whichincorporates Glasgow Coma Scale, ICH volume,intraventricular hemorrhage, age, and site of ICHorigin. ICH scores of three or greater have beenreported to have a sensitivity and specificityapproaching 80% and 90%, respectively, for mor-tality following ICH (Simmons and Parks 2008).

The applicability of prognostic models to thereal world is related in part to the commonality ofthe variables included and also by the sample fromwhich the data are drawn. Prediction models, forexample, generated using data from cohorts withhigh numbers of clinical trial participants or onlypatients managed in specialized stroke centersmay not be easily applicable to a broader, com-munity-based stroke population. Similarly modelsthat include clinical information from imagingmodalities that are not widely available may alsolack generalizability.

One particular concern raised by a number ofstroke researchers when considering mortalityprediction models is the concept of “withdrawalbias.” It is argued that because current modelshave incorporated patients who either never com-menced or had life-sustaining therapies with-drawn, there is a potential bias toward mortality(or treatment withdrawal) as well as difficulty inestablishing the true effect (positive or otherwise)of these therapies following severe stroke. This islikely to be particularly relevant for stroke typeswith poor prognosis such as ICH. Given this theimportance for clinicians to understand thestrengths and limitations of these models, whenusing them to establish and communicate progno-sis to patients and families, has been highlighted(Holloway et al. 2014; Frontera et al. 2015). This


may be particularly the case when consideringinstigating or transitioning to end-of-life care.

Prognostication based upon clinician experi-ence and expertise is an alternative to a model-based approach. Unlike fixed mathematicallygenerated models, clinicians have the advantageof being able to consider an individualized rangeof factors and adjust their estimations in real time.Clinician-based estimates however may vary con-siderably between individuals and be influencedby both optimistic and pessimistic judgements(Holloway et al. 2014; Frontera et al. 2015). Theinvolvement of the multidisciplinary team andutilization of a second opinion (including fromSPC) have been suggested as ways to overcomesome of the issues with clinician-based estimates.

Despite the inherent difficulties associatedwith prognostication, the importance of providingpatients and their families with a sense of thefuture cannot be underestimated. This is likely tobe best achieved using a combination of clinicianacumen and evidence-based estimation modelsindividualized to each patient and their uniquecharacteristics.

2.4.2 Limitations of TreatmentThe majority of deaths after stroke follow eitherthe limitation, cessation, or foregoing of poten-tially life-prolonging therapies (Kelly et al. 2012;Holloway et al. 2014; Creutzfeldt et al. 2015;Alonso et al. 2016). In the United States, up to60% of all stroke deaths follow the removal ofmechanical ventilation (Holloway et al. 2014)although this rate is likely to be different in othercountries. Decision-making regarding limitationsof treatment in acute stroke is complex as it needsto incorporate a range of patient, sociocultural,and clinical factors including autonomy, patientand family wishes, and prognostication. Addition-ally in the acute phase when uncertainty is high,management decisions need to be made quicklyso as to maximize the chances of injury reversaland potential recovery. The acute stroke settinghas been described as a “fast-paced, chaotic envi-ronment wrought with hope and disappointment,relief and anxiety” (Creutzfeldt et al. 2015), andtherefore it is hardly surprising that decision-mak-ing in this setting is challenging. Complicating the

situation further, many patients have impairedcommunication as a consequence of their stroke,meaning the responsibility for these complex andoften life-and-death decisions will involve familyand loved ones.

Treatment limitations can take numerous formsand may alter over time. In the acute setting, theymay include decisions regarding thrombolyticand/or endovascular therapy or aggressive resus-citative approaches such as cardiorespiratoryresuscitation or mechanical ventilation. In somepatients, particularly the more elderly, frail, orthose following a severe stroke, the decision totake a palliative approach with an emphasis oncomfort might be taken at the outset. This mightinvolve, among other things, the foregoing orwithdrawal of antibiotic therapy, hydration, orsupplementary feeding. In the majority of circum-stances, these discussions and decisions will notinvolve SPC but be undertaken by members of thestroke team or emergency department. Relativesof stroke patients have reported experiencing dis-comfort when they felt excluded from the deci-sion-making process, overly responsible for thedecisions made (Cowey et al. 2015), or whenparticipating in discussions if they knew do-not-resuscitate (DNR) orders had already been madeby the medical team (Payne et al. 2010). Thishighlights the importance of mutual, shared deci-sion-making between clinicians, patients, andfamilies. The importance of good communicationskills underpinned by an awareness of palliativecare practice and philosophy has been advocatedas a valuable component of both neurology prac-tice and training (Holloway et al. 2014;Creutzfeldt et al. 2015).

The impact of treatment limitations on acutestroke outcomes has been fairly extensively stud-ied with particular interest in the effect of earlyDNR orders (and other limitations) on stroke mor-tality. A number of papers have shown an associ-ation between DNR orders and increasedmortality in both ICH and ischemic stroke inde-pendent of other established mortality predictors(Zahuranec et al. 2007; Holloway et al. 2014;Parry-Jones et al. 2016). In a study of 270 non-traumatic ICH, the presence of early treatmentlimitations (defined as DNR orders alone or in


combination with early withdrawal and/or earlydeferral of other life-sustaining therapies) wasassociated with a twofold increase in mortalityindependent of a range of established predictorsincluding age, Glasgow Coma Scale score, andICH volume (Zahuranec et al. 2007). It has beensuggested that the association between DNRorders and early mortality following stroke (andICH in particular) represents a “self-fulfillingprophecy” whereby the prediction of poor prog-nosis leads to limitations of care which ultimatelyproduce a poor prognosis. Accordingly, and in theabsence of clear advance care directives, cautionis advised about making early treatment limita-tions decisions in the immediate post-strokeperiod.

Following the acute stroke phase, ongoingconsideration of treatment goals is influenced bythe evolving clinical picture, patient and familywishes, and expectations in relation to recovery.This can be a time of significant uncertainty forpatients, families, and clinicians and one in whichdecision-making can be become more complexwith emphasis on factors such as prognosis,potential withdrawal of life-prolonging therapies,and discharge planning (Rogers et al. 2005; East-man et al. 2013). Decisions around the withdrawalof hydration and nutrition can be particularly chal-lenging and were recorded as a source of conflictin nearly half of all interactions between familymembers and staff in a study of 104 patients whodied in a Canadian SCU (Blacquiere et al. 2009).Assistance with this complex decision-makingand communication is a common reason for SPCinvolvement following stroke and is valued byother health professionals. In one study, strokestaff noted that the benefits of access to SPC inthese complex scenarios included “reassurance”and “support for decision-making” (Burton andPayne 2012).

For relatives of patients who have had a severestroke, acting as a surrogate decision-maker canbe a two-edged sword. On the one hand, feelingsufficiently involved in decision-making is a pre-dictor of high satisfaction with end-of-life care forbereaved relatives of stroke patients (Young et al.2009). On the other, when relatives have reflectedupon their experiences as surrogates for patients

who had suffered severe strokes, a number ofconflicts and struggles were described (de Boeret al. 2015). These include the strain of makingdecisions under time pressure, feeling unpreparedor underqualified to speak for their relative, anddealing with uncertainty and change. Despite dif-ferences in the experiences reported betweeninterviewees, there was an overall tendency forsurrogates to follow medical advice, highlightingagain the importance of patient- and family-cen-tered communication in this setting.

Longer-term positive and negative conse-quences have been reported for surrogatesinvolved in making treatment decisions for others(not specific to stroke). In a systematic review(Wendler and Rid 2011) which included 2854surrogates across 40 quantitative and qualitativestudies, nine papers found that being involved indecision-making had beneficial effects for somesurrogates. Importantly however up to a third ofsurrogates reported negative emotional conse-quences associated with making decisions fortheir loved one including stress, guilt, and doubtabout whether they had done the right thing.These negative emotions were typically sustainedfor months and sometimes years which has impor-tant implications for bereavement services. Feel-ing confident they were following a treatment planconsistent with patient preferences seemed to mit-igate some of the emotional burden on surrogates,perhaps highlighting the value of Advance CareDirectives as a way of planning for future healthcare.

2.4.3 Symptom Burden FollowingAcute Stroke

The recognition and appropriate management ofboth physical and psychological symptoms hasbeen highlighted as an unmet need for patientsand their families following acute stroke(Addington-Hall et al. 1995; Stevens et al. 2007;Burton and Payne 2012). While published litera-ture is limited, symptom burden has been reportedin between 65% and 98% of dying stroke patients.A broad range of physical symptoms have beenreported with varying frequency in patients fol-lowing an acute stroke including fatigue, nausea,restlessness, and issues with urination and


defecation. Pain has been variably described withreported prevalence rates of between 30% and70% (Addington-Hall et al. 1995; Mazzocato etal. 2010; Ntlholang et al. 2016; Eriksson et al.2016).

There is evidence that psychological distress iscommon in patients following an acute stroke,although differences in prevalence are noted inthe limited available published literature. Psycho-logical distress, including anxiety, dysthymia, andloneliness, was found to be present in almost halfof 191 consecutive stroke admissions in one UKstudy (Burton et al. 2010) and in 25% of dyingstroke patients referred to a Swiss palliative careconsultation service (Mazzocato et al. 2010). Bycontrast, in a separate review of 54 consecutivedeaths in an Irish specialist stroke service, psy-chological distress was reported in only onepatient (Ntlholang et al. 2016). Investigating thepalliative care needs of patients following an acutestroke, Burton and colleagues (2010) found thatwhile spiritual or religious concerns were low, upto a 25% of patients were worried about death anddying. Additionally, many were concerned aboutongoing dependence and disability and the impactthis might have on their loved ones. About aquarter felt that their care needs would exceedthe capabilities of their families creating thepotential for additional distress and even dishar-mony in an already emotionally fraught time.

The importance of considering the psycholog-ical impact of stroke upon patients and familieswas highlighted in a study investigating bereavedfamily members’ satisfaction with the care pro-vided to patients palliated after ischemic stroke(Blacquiere et al. 2013).While overall satisfactionwith palliative care was high, lower satisfactionrates were reported for treatment of anxiety anddepression and for the level of emotional supportprovided to families. The impact of stroke uponloved ones and families is frequently profound, inlarge part due to the abrupt change in function andcognition often associated with stroke. In the lon-ger term, a number of adverse outcomes havebeen identified in stroke caregivers includingmental health issues, worsening physical health,and financial burden (van Heugten et al. 2006;Carod-Artal and Egido 2009). This potential for

longer-term adverse sequelae adds further weightto the importance of considering the psychosocialneeds of stroke patients and their families in theacute phase.

While variability in symptom burden follow-ing acute stroke has been reported, respiratorysymptoms including dyspnea and secretionsappear to be particularly prevalent. This is likelyto be due, at least in part, to their demonstrabilitywhen compared to less visibly obvious symptomssuch as pain and anxiety. In one prospectivecohort study of 22 patients admitted to a SCUand felt likely to die within 3 months of admis-sion, all of the 20 patients who subsequently diedexperienced respiratory symptoms during theirlast hours of life (Rogers and Addington-Hall2005). Respiratory secretions or “death rattle”were recorded in just over 60% of 1626 dyingstroke patients in a Swedish database review(Eriksson et al. 2016), and dyspnea was thecommonest recorded symptom in 54 consecutivepatients who died in an Irish hospital stroke ser-vice (Ntlholang et al. 2016) and in 81% of strokepatients referred to a Swiss palliative care consul-tation team (Mazzocato et al. 2010). The potentialdifficulty of managing dyspnea in dying patientswas highlighted in this study as only 48% ofpatients were felt to be free from dyspnea duringtheir last 48 h of life (compared to 81% who wereassessed as being free from pain).

Using data from the Swedish Register of Pal-liative Care, Eriksson and colleagues (2016) com-pared the prevalence of six symptoms (pain,“death rattle,” dyspnea, anxiety, confusion, andnausea) between 1626 patients who died follow-ing a stroke and 1626 patients dying from cancer.Interesting differences in symptom prevalence,awareness, and management were noted betweenthe groups. While all symptoms were present inthe stroke group (nausea 7.6%, confusion 7.9%,dyspnea 16.3%, anxiety 18.9%, pain 42.7%, and“death rattles” 60.7%) when compared with theirmatched counterparts with cancer, stroke patientswere significantly more likely to experience“death rattles” but less likely to experience anyof the other five symptoms. Importantly however,staff caring for stroke patients were significantlyless likely to know whether a patient suffered


from any of the target symptoms compared to staffcaring for patients with malignant disease. Itmight be anticipated that this difference wasrelated to higher rates of reduced consciousnessin dying stroke patients; however this does notseem to have been the case as the ability to self-determine until the last days of life was equivalentbetween groups (73.3% of stroke patients, 74.3%in cancer). In keeping with the differences seen insymptom prevalence and staff awareness of symp-toms, stroke patients were significantly morelikely to have as-required medications chartedfor “death rattle” but less likely to have them forpain, nausea, or anxiety. Overall this study high-lights the potential differences in palliative careneeds between dying stroke and cancer patientsand lends support to individualized approachesrather than assuming that one model of care willfit all. Crucially it also reinforces the importanceof stroke service staff being appropriately edu-cated and supported in the provision of generalpalliative care.

The considerable variability in symptom prev-alence seen in the published literature is likely dueto a range of factors including communication andconsciousness impairment and inconsistent symp-tom assessment. Additionally accurate quantifica-tion of symptom burden for patients following anacute stroke is hampered at least in part by the factthat most studies addressing the question havebeen retrospective reviews of precollected data.Further prospective longitudinal research isrequired to expand the currently limited datasetaddressing this question and to guide the devel-opment of appropriate interventions andapproaches to meet the important needs ofpatients and their families.

2.4.4 End-of-Life Care and DyingFollowing Acute Stroke

The recognition or diagnosis of dying is compli-cated and largely arbitrary. Difficulties identifyingthe time point at which end-of-life care (EOLC)might be initiated are commonly reported bystroke care staff (Burton and Payne 2012; Gardi-ner et al. 2013; Cowey et al. 2015) leading topotential under- or overtreatment and delay inproviding palliative and end-of-life care. A large

contributor to this difficulty is the commonalitybetween many of the features associated withdying and those seen in patients following asevere stroke who may subsequently recover. Arange of factors have been used to identify dyingstroke patients including stroke characteristics(including subtype and severity), clinical course(in particular the lack of meaningful recovery orongoing deterioration), and physiological param-eters (such as altered breathing patterns) (Coweyet al. 2015). In a study of patients who died in astroke unit, disturbed consciousness, early dys-phagia, and large supratentorial strokes were indi-cations for initiation of EOLC (Alonso et al.2016). In a separate mixed-methods study involv-ing 23 Scottish stroke unit health professionals,over a quarter reported using intuition (at least inpart) to recognize dying (Cowey et al. 2015).

The responsibility for decision-making regard-ing commencement of EOLC for hospitalizedstroke patients has generally rested with the SCUmedical team, although the importance of inputand insight from other members of the multi-disciplinary team cannot be underestimated. Aswe have seen due to the nature of stroke presen-tation and management, SPC services are infre-quently and reactively involved in this processtypically at times of increased complexity or sig-nificant uncertainty. Involvement of family mem-bers in decision-making around end-of-life care isimportant but not without its problems particu-larly if there is discordance between relativesand healthcare professionals (Rogers andAddington-Hall 2005). Feeding patients after asevere stroke has been highlighted as an areawhere motivations and opinions may differbetween family members and staff. Rogers andAddington-Hall (2005) noted that while relativesand stroke unit staff shared the common motiva-tion of wanting to avoid prolongation of suffering,stroke unit staff were also concerned about thepatient starving or having their chance of recoveryimpinged due to lack of nutrition. Despite thesedifficulties family involvement allows a uniqueperspective on the patient and their place in theworld and has been shown to be a predictor ofhigh satisfaction with EOLC. As for all other areasof stroke care, the importance of clear,


unambiguous, empathic, and effective communi-cation cannot be overstated.

Two distinct patterns of dying following anacute stroke have been described, namely, arapid, sudden death or prolonged dying. A pro-longed dying phase can be particularly difficultfor family members, and this is likely to be exac-erbated if there was expectation of a quick death,they had not been informed of the possibility ofprolonged dying, and in the setting of severedysphagia (Cowey et al. 2015). Not surprisinglythe transition from recovery-focused care toEOLC can be challenging for family memberswith feelings of isolation and abandonmentreported (Payne et al. 2010). However whendeath is considered likely, the most importantthings identified by relatives and stroke unit staffalike are the avoidance of distress and the main-tenance of comfort and dignity (Rogers andAddington-Hall 2005; Payne et al. 2010).

End-of-life care (EOLC) pathways have beenused to guide and optimize multidisciplinary,holistic care for dying patients and their familiesincluding following stroke. In this setting EOLCpathways aim to cover not only physical symp-toms but also psychological, spiritual, and culturalconsiderations, desired place of care, and after-death management. The Liverpool Care Pathway(LCP) is an example of an EOLC pathway whichhas been used worldwide in both malignant andnonmalignant conditions. Recently the LCP hasattracted considerable media attention and contro-versy in part due to concerns about its perceivedoverapplication at times, the inappropriate denialof nutrition and hydration in some cases, anddeficits in communication with patients and fam-ilies about its use. A subsequent UK government-commissioned independent review, whileacknowledging the principles underpinning theLCP, expressed among a range of concerns thatit was too frequently used as a “tick-box” exerciseand recommended that it be progressively phasedout and replaced by individualized care plans(Neuberger et al. 2013).

There is limited evidence addressing the use ofEOLC pathways following severe stroke withavailable data seeming to be generally supportiveof their use. The Australian Clinical Guidelines

for Stroke Management in a small section devotedto palliative care include a recommendation thatpathways for stroke palliative care can be used toimprove the care of people dying following astroke but acknowledge that evidence to supportthis recommendation is weak (National StrokeFoundation 2010). Very little is known about thequantitative impact of EOLC pathways on careparameters following severe stroke althoughimprovements in both documentation and clinicalpractice were observed in a small retrospectiveaudit pre- and post the implementation of theLCP on a 12-bed stroke unit (Jack et al. 2004).Examples of the changes in clinical practice seenincluded increases in the discontinuation of inap-propriate medications (from 40% to 100%), in thecharting of subcutaneous medications (from 20%to 85%), and in the assessment of religious needs.While these results appear promising, they need tobe interpreted cautiously given the retrospectivestudy design and small sample size.

Two qualitative studies have addressed theperceptions of healthcare professionals regardingthe use of the LCP in English and Scottish strokeunits (Gardiner et al. 2013; Cowey et al. 2015). Ingeneral satisfaction has been reported with its use,with staff in one study suggesting that the LCPwas a core element of high-quality palliative care(Gardiner et al. 2013). Surveyed family memberswere also generally satisfied with LCP-based carealthough tended to be more concerned with ade-quate control of problems rather than whether anEOLC pathway was used (Cowey et al. 2015).Importantly family members were able to influ-ence EOLC including when the LCP was used.This meant that management plans, includingcomponents such as the ongoing provision ofparenteral hydration, were adapted to incorporatetheir wishes. This negotiated mutual pragmatismis perhaps what Neuberger and colleagues (2013)were envisaging when they recommended thereplacement of the LCP with individualized careplans. Interestingly while both of these studiesundertook their qualitative data collection priorto the 2013 independent LCP review, the paperby Cowey and colleagues (2015) was publishedsubsequent to it. In their conclusions, the authorsacknowledged the withdrawal of the LCP in the


UK but noted that it continued to be used world-wide. They reiterated that in their study, familymembers were more concerned with distressingstroke-related problems than the LCP. In a suc-cinct summary encapsulating the complexity ofEOLC following stroke, they concluded that“such problems are enduring in nature and remainas clinical challenges whether end-of-life carepathways are used or not.”

3 Palliative Care in the Post-Stroke Phase

The last couple of decades have seen emergingrecognition of the role palliative care can playbeyond end of life. Earlier involvement inadvanced malignant disease, for example, hasbeen demonstrated to lead to improvements insymptom burden, quality of life, and psychologi-cal well-being. Increased awareness of the sub-stantive symptom burden and distress associatedwith a range of noncancer conditions, includingchronic respiratory, renal, and neurological disor-ders, has highlighted the benefits of multi-disciplinary palliative care in symptommanagement, complex decision-making, andcare facilitation. In stroke while there is a devel-oping evidence base for palliative care in the acutephase, the picture is much less clear for the post-stroke period. This is worthy of further consider-ation because as previously discussed the majorityof patients will ultimately survive an acute stroke.

For those people who survive their initialstroke, the focus of most clinical programs andthe literature has understandably been on second-ary prevention and stroke rehabilitation. Accord-ingly there is comparatively little published dataevaluating the involvement or potential role ofpalliative care services in post-stroke care. Whiledata examining palliative care in this setting arelimited, there is evidence of a range of oftenunder-recognized symptoms and morbidity inpatients, families, and caregivers followingstroke.

The symptoms and issues experienced bypatients and their caregivers in the chronic post-stroke phase will be familiar to many palliative

care clinicians. They can include pain, depression,functional disability, seizures, bladder and boweldysfunction, and caregiver stress/fatigue. Whilethese symptoms are likely to share similaritieswith those seen in patients with cancer, end-organ failure, and/or neurological degenerativedisorders, they may also display features uniqueto stroke. For example, the immediate onset ofprofound change in physical and cognitive func-tion that often accompanies acute stroke contrastswith the subacute development of illness and dis-ability in other malignant and nonmalignant con-ditions. Additionally the management ofsymptom burden in an environment where patientsurvival might be measured in years rather thanweeks or months is outside the traditional skill setof many palliative care clinicians.

In some ways the chronic post-stroke periodmight be considered analogous to cancer survi-vorship as both may share features of ongoingsymptoms in the context of likely long-term sur-vival. The role of palliative care in cancer survi-vorship, while variable worldwide, remainsrelatively undefined, debatably appropriate, andinfluenced by a broad range of clinical and struc-tural factors including limitations in clinicianexperience, workforce issues, and fundingmodels. It seems likely that involvement in thepost-stroke phase will provide palliative care, atboth a clinician and organizational level, withsimilar challenges regarding clinical appropriate-ness and service provision.

3.1 Pain Syndromes

The most common types of pain reported in strokesurvivors are central post-stroke pain, hemiplegicshoulder pain, painful spasticity, musculoskeletalpain, and tension-type headache (Creutzfeldt et al.2012). Accurate pain assessment following strokecan be challenging due to the consequent commu-nication deficits that may include dysphasia anddysarthria and/or changes in consciousness thatoccur in some stroke survivors.


3.1.1 Central Post-Stroke PainCentral post-stroke pain (CPSP) is a neuropathicpain syndrome that is both highly distressing andfrequently refractory to treatment. Sometimesreferred to as Dejerine-Roussy syndrome, it wasfirst described in 1906 (Vartiainen et al. 2016).There is variability in the reported rates of CPSPdevelopment among stroke survivors as well asconcerns that it may be under-recognized. In anItalian population-based study which included1494 post-stroke patients, symptoms and sensorychanges consistent with CPSP were present in11% of stroke survivors (Raffaeli et al. 2013).Other authors have reported prevalence rates ofbetween 3% and 8% when all stroke survivors areconsidered (Vartiainen et al. 2016) and up to 35%in those with specific thalamic lesions. While thepathogenesis of CPSP is not well established,postulated mechanisms for development includehyperexcitability of injured sensory networks,changes in central inhibitory mechanisms, or cen-tral nervous system neurotransmitter imbalances.The varying incidence and prevalence of CPSP isexplained by the unclear etiology of this painsyndrome as well as the lack of universallyaccepted diagnostic criteria. The diagnosis ofCPSP requires the exclusion of pain caused byjoint contracture, peripheral nerve disorders, andspasticity.

The pain of CPSP may be spontaneous orevoked. Spontaneous pain can occur either con-tinuously or intermittently, while evoked pain istypically precipitated by stimuli such as touch,movement, stress, or temperature change(Creutzfeldt et al. 2012; Vartiainen et al. 2016).Sensory abnormalities such as allodynia,dysesthesia, paresthesia, and hyperalgesia arecommon features (Creutzfeldt et al. 2012; Raffaeliet al. 2013). The pain is typically experiencedwithin the area of sensory impairment and usuallyat the point of maximal deficit. Spinothalamicabnormalities such as temperature-sensory abnor-mality are frequently described. CPSP maydevelop at any time from immediately to yearsafter a stroke; however onset most commonlyoccurs at 3–6 months (Raffaeli et al. 2013).

Management of CPSP is challenging as there isa paucity of evidence to guide practice. Optimal

management is likely to be best achieved by uti-lizing a combination of pharmacological and non-pharmacological treatment approaches. A rangeof pharmacological agents have been trialed inthe management of CPSP with a particular focuson antidepressants and anticonvulsants. The effi-cacy and tolerability of these agents has variedacross trials. Amitriptyline at a final dose of 75 mgdaily was associated with significant improve-ments in pain when compared to placebo in asmall double-blind, placebo-controlled crossoverstudy conducted in 15 nondepressed patients(Leigon and Boivie 1989). Importantly there waslittle in the way of adverse effect associated withamitriptyline. No statistically significant benefitwas seen for carbamazepine (at doses of up to800 mg/day) in the same study.

In terms of other agents, there is some evidenceof benefit from both case series and a controlledtrial for lamotrigine; however, there is limitedevidence for gabapentin and none for opioids(Frese et al. 2006). The efficacy and safety ofpregabalin in the management of CPSP wasassessed in a double-blind, placebo-controlledrandomized trial in 219 patients. The mean doseof pregabalin received was 356.8 mg, and after13 weeks, pain had improved in both treatmentarms (pregabalin baseline pain score 6.5, end-point pain score 4.9; placebo baseline pain score6.3, end-point pain score 5.0); however there wasno significant difference between the groups. Pre-gabalin was associated with improvements insleep and anxiety in the study; however, morethan 50% of patients in the pregabalin groupreported some adverse effects (compared to 23%in the placebo group), with dizziness and somno-lence being most common (Kim et al. 2011).

Non-pharmacological therapies including cog-nitive behavioral therapy and stimulation thera-pies including transcutaneous electrical nervestimulation and acupuncture have been describedin small case series; however, controlled trial evi-dence is lacking. Surgical pain therapies such asrhizotomy, sympathectomy, cordotomy, and deepbrain stimulation have not been formally evalu-ated for the treatment of CPSP and should beconsidered only in the context of approved clini-cal trials.


3.1.2 Hemiplegic Shoulder PainHemiplegic shoulder pain (HSP) is the most com-mon post-stroke pain, occurring in 11–83% ofpatients (Creutzfeldt et al. 2012). The prevalenceof HSP increases with worsening motor impair-ment. HSP occurs as a consequence of joint sub-luxation and sensory and motor deficits. Carefulattention to joint positioning in hemiplegicpatients and physical therapies are the mainstayof treatment, with some emerging data for intra-articular steroid injections and intramuscularinjections of botulinum toxin-A. Analgesia withanti-inflammatory medications, paracetamol, andtopical heat or ice and soft-tissue massage can alsoassist with initial pain management.

3.1.3 Spasticity and MusculoskeletalPain

Pain due to contractures, pressure areas, and spas-ticity are best prevented through the use of phys-ical and rehabilitative therapies, body positioning,and range of motion exercises. Baclofen anddantrolene are used for post-stroke spasticity;however, their side effects including sedation,confusion, and dizziness can limit their use. Bot-ulinum toxin can be used, particularly in upper-limb spasticity, to improve functional outcome(Creutzfeldt et al. 2012).

3.2 Post-Stroke Depression

Depression is part of a constellation of neuropsy-chiatric disorders that are recognized to be asso-ciated with stroke. Estimates of the frequency ofpost-stroke depression (PSD) vary due to the het-erogeneity of assessment and reportingapproaches; however, a recent meta-analysiswhich included 61 studies and over 25,000 peoplefound a frequency for PSD of 31% out to 5 yearspost stroke (Hackett and Pickles 2014). PSD onsetcan occur at any time following a stroke, with aprospective, longitudinal study of over 200patients, observing rates of PSD to be relativelystable across a range of time points from 1 monthto 18 months post stroke (De Ryck et al. 2014).Stroke severity and the resultant degree of physi-cal and functional disability are the strongest

predictors of post-stroke depression, with otherrecognized predictors including cognitive impair-ment, dysphasia and aphasia, apraxia, and pre-morbid history of depression or anxiety (DeRyck et al. 2014; Robinson and Jorge 2016).Evidence for the impact of social support andstroke lesion location upon PSD is conflicting;however, there are reports of association betweenleft frontal and left basal ganglia lesions and PSD(Robinson and Jorge 2016).

Post-stroke depression is associated withincreased mortality, poorer engagement in reha-bilitation, decreased quality of life, and socialisolation (Creutzfeldt et al. 2012; De Ryck et al.2014; Robinson and Jorge 2016). An independentand direct association exists between depressionseverity and functional impairment although thisrelationship is likely to be reciprocal in manycases. The increased mortality associated withPSD has been reported to be due to greater car-diovascular mortality, with PSD associated dis-ruptions in autonomic nervous system functionpostulated as a potential explanatory mechanism(Robinson and Jorge 2016).

A number of double-blind, placebo-controlledtreatment trials for PSD have been undertakenusing both tricyclic antidepressants and selec-tive-serotonin uptake inhibitors (SSRI) since1984 (Robinson and Jorge 2016). Although sam-ple numbers were generally small across the trials,most reported improvements in depression scoreswhen compared to placebo. A meta-analysis thatincluded 17 trials (13 using pharmacologicalagents and 4 psychotherapy) and 1655 patients(Hackett et al. 2008) found that pharmacotherapyconferred a small but significant benefit in treatingdepression and reducing depressive symptoms;however, this was coupled with an increase inadverse effects. No evidence of benefit was dem-onstrated for psychotherapy; however, other stud-ies have reported positive outcomes for briefpsychosocial interventions (includingpsychoeducation and family support) when com-bined with antidepressants (Robinson and Jorge2016). The impacts of antidepressant medicationson physical and functional outcomes are less welldescribed (Creutzfeldt et al. 2012); however,improvements in motor, cognitive, and functional


capacity have been demonstrated (Robinson andJorge 2016). There is emerging evidence thatpharmacotherapy (in particular SSRIs) may havea preventative role in PSD as well as interestingdata, suggesting antidepressants might improvestroke survival independent of either successfuldepression treatment or the presence of depressionin the first place (Robinson and Jorge 2016).

3.3 Quality of Life

Of patients who survive 30 days following acutestroke, half will die within 5 years, and of thesurvivors, approximately 30% will remain dis-abled, with 14% requiring institutional care(Hankey et al. 2002). Quality of life (QOL)while difficult to define and inherently subjectivein nature is generally considered to be a multi-dimensional construct incorporating multiplebroad domains including physical, social, andmental. Recognition of the potential impact ofstroke upon QOL has seen the development of anumber of stroke-specific health-related QOLscales in recent years. These scales include spe-cific factors relevant to stroke patients includingvision and language impairments, with examplesbeing the Stroke Impact Scale and the Burden ofStroke Scale (Carod-Artal and Egido 2009).

A wide range of factors have been shown toimpair QOL following stroke including depen-dency in activities of daily living (ADL), motordysfunction, aphasia, presence of depression,CPSP, sexual dysfunction, and limited social sup-ports (Carod-Artal et al. 2000; Choi-Kwon et al.2006; Carod-Artal and Egido 2009). Reductionsin QOL have been described by stroke survivorswith differing levels of functional impairmentfrom the profoundly disabled to those whoregained independence in ADL function but didnot get back to premorbid functional levels and inothers who were not able to return to work. AKorean study that examined influences on QOLin 151 first-time stroke patients found that ADLdependency, CPSP, depression, and lower socio-economic status were all important explanatoryfactors for lower QOL 3 years post stroke (Choi-Kwon et al. 2006). Although variation exists in

the literature stroke type, lesion location, age, andgender have not consistently been associated withlower QOL (Choi-Kwon et al. 2006) althoughthere is some evidence to suggest females havelower QOL following stroke than men (Carod-Artal et al. 2000; Carod-Artal and Egido 2009).

Importantly recognition of the negative influ-ence of a broad range of factors upon QOL high-lights the need for close attention to be paid tothese things in the post-stroke period. In particu-lar, vigilance when it comes to the assessment andtreatment of depression and pain and rehabilita-tive efforts focused on maximizing functional out-comes and independence are crucial componentsin the preservation of QOL for patients followinga stroke.

3.4 Caregiver Issues

While patients after stroke confront a myriad ofissues, caregivers are also confronted with a newand often devastatingly different world. This isoften exacerbated by the sudden onset of changewhich necessitates the rapid acquisition of newknowledge, skills, and acceptance of changed cir-cumstances. This contrasts other more chronicillnesses where disease progression is typicallygradual and predictable, allowing caregiversgreater opportunity for adjustment and adaptation.Cognitive deficits including perceptual and lan-guage change along with motor deficits and func-tional dependency add additional complexity tothe caregiver role. Additionally many caregiversare themselves elderly with their own health prob-lems and often contracting social and supportnetworks.

Worsening physical health, lower QOL, highlevels of stress and mood disorder, social isola-tion, and financial burden have all been reportedby stroke caregivers (van Heugten et al. 2006;Carod-Artal and Egido 2009). Anxiety anddepression are common with rates of depressionestimated to be upward of 25%. Given the recog-nized high levels of burden among stroke care-givers and the crucial role they play in supportingpatients at home, increasing attention is being paidto assessing and managing stroke caregiver needs.


Importantly despite the clear challenges that thecaregiver role can bring, evidence suggests thatmany stroke caregivers still experience increasedappreciation of life and role fulfillment.

A lack of attention to caregivers in establishedstroke guidelines prompted the development ofevidence-based clinical practice guidelines spe-cific to stroke caregivers in the Netherlands inthe mid-2000s (van Heugten et al. 2006). Theseguidelines cover a range of topics including riskfactors for caregiver burden, approaches toassessing burden, and possible caregiver interven-tions. Factors identified as predictive of increasedcaregiver burden include both patient characteris-tics (particularly limited functional capacity andsignificant cognitive impairment) and caregiverfactors including preexisting psychological ill-health.

The impact of interventions upon stroke care-givers has been evaluated across a number ofstudies with varying degrees of benefit reported.Interventions identified to provide at least someimprovement in caregiver wellbeing include theprovision of information, social and practical sup-ports, training in the caregiver role, attention tomental health issues, and counseling (Kalra et al.2004; van Heugten et al. 2006). Clinicians have akey role to play in preparing caregivers for theirrole. In retrospective surveys of bereaved care-givers, satisfaction with care following a strokeis positively correlated with involvement of doc-tors and nurses who are knowledgeable aboutstroke, clinicians being open to discussing fearsand concerns, and caregivers being involving indecision-making processes (Young et al. 2009).

While after-stroke care largely focuses on sec-ondary prevention and rehabilitation, it is clearthat symptoms and complex care needs includingpain syndromes, depression, impaired quality oflife, functional impairment, and caregiver issuesare highly prevalent following stroke. These areall things familiar to multidisciplinary palliativecare clinicians, who accordingly are conceivablywell positioned to address them utilizing expertiseand experience drawn from other chronic dis-eases, as well as collaboration with colleaguesfrom stroke and rehabilitation teams.

4 Conclusions/Summary

Despite significant advances in prevention, recog-nition, and treatment, stroke remains a majorcause of morbidity and mortality worldwide. Thevalue of integrating palliative care principles andpractices into stroke management is being increas-ingly recognized particularly in relation to issuessuch as symptom management, complex deci-sion-making, and establishment of goals of care.Ongoing research is required addressing not onlythese important patient and family needs but alsothe theoretical basis, clinical requirements, anddepartmental structures required for successfulintegration. In the acute phase, the majority ofpalliative care is currently and likely to continueto be undertaken by members of the stroke teamwith secondary support available from specialistpalliative care services on an as-needed basis.Additionally specialist palliative care is ideallyplaced to provide the necessary education andguidance to ensure the provision of high-qualitycare.

The role of palliative care in the more chronicpost-stroke phase is less clear with little publisheddata addressing the question. There is clear evi-dence of significant symptom burden, reducedquality of life, and caregiver distress for patientsand their families in this period, and these are allthings familiar to multidisciplinary palliative careteams. In many ways palliative care clinicians arewell placed to take a lead role in the managementof many of these problems although their chronic-ity is likely to impose clinical, organizational, andworkforce challenges upon the profession.

The ongoing success of the integration of pal-liative care and stroke services will be contingenton a number of factors, not least of which beingbuy-in from both specialties. For palliative carethis will perhaps require a conscious shift awayfrom the cancer-related models and ideologies ofthe past and an acknowledgement that adaption,conciliation, and collaboration will be crucial inmoving forward. This is relevant not just forstroke but also other nonmalignant diseases andthe ever-changing cancer landscape. In essencethis is the challenge for modern palliative careand the next generation of specialists.


References

Addington-Hall J, Lay M, Altmann D, McCarthy M.Symptom control, communication with health profes-sionals, and hospital care of stroke patients in the lastyear of life as reported by surviving family, friends, andofficials. Stroke. 1995;26(12):2242–8.

Alonso A, Ebert AD, Dörr D, Buchheidt D, Hennerici MG,Szabo K. End-of-life decisions in acute stroke patients:an observational cohort study. BMC Palliat Care. 2016;15(1):1.

Andersen KK, Olsen TS, Dehlendorff C, KammersgaardLP. Hemorrhagic and ischemic strokes compared strokeseverity, mortality, and risk factors. Stroke.2009;40(6):2068–72.

Blacquiere DP, Gubitz GJ, Dupere D, McLeod D, PhillipsS. Evaluating an organized palliative care approach inpatients with severe stroke. Can J Neurol Sci. 2009;36(06):731–4.

Blacquiere D, Bhimji K, Meggison H, Sinclair J, SharmaM. Satisfaction with palliative care after stroke. Aprospective cohort study. Stroke. 2013;44:2617–9.

Burton CR, Payne S. Integrating palliative care withinacute stroke services: developing a programme theoryof patient and family needs, preferences and staff per-spectives. BMC Palliat Care. 2012;11(1):1.

Burton CR, Payne S, Addington-Hall J, Jones A. Thepalliative care needs of acute stroke patients: a prospec-tive study of hospital admissions. Age Ageing. 2010;39(5):554–9.

Campbell BC, Mitchell PJ, Kleinig TJ, Dewey HM,Churilov L, Yassi N, et al. Endovascular therapy forischemic stroke with perfusion-imaging selection.N Engl J Med. 2015;372(11):1009–18.

Carod-Artal FJ, Egido JA. Quality of life after stroke: theimportance of a good recovery. Cerebrovasc Dis.2009;27(Suppl 1):204–14.

Carod-Artal J, Egido JA, Gonzalez JL, de Seijas EV. Qual-ity of life among stroke survivors evaluated 1 year afterstroke. Stroke. 2000;12:2995–3000.

Casaubon LK, Boulanger J-M, Glasser E, Blacquiere D,Boucher S, Brown K, et al. Canadian stroke best prac-tice recommendations: acute inpatient stroke careguidelines, update 2015. Int J Stroke. 2016;11(2):239–52.

Choi-Kwon S, Choi JM, Kwon SU, Kang DW, Kim JS.Factors that affect the quality of life at 3 years post-stroke. J Clin Neurol. 2006;2:34–41.

Cowey E, Smith LN, Stott DJ, McAlpine CH, Mead GE,Barber M, et al. Impact of a clinical pathway on end-of-life care following stroke: a mixed methods study.Palliat Med. 2015;29(3):249–59.

Creutzfeldt CJ, Holloway RG, Walker M. Symptomaticand palliative care for stroke survivors. J Gen InternMed. 2012;27:853–60.

Creutzfeldt CJ, Holloway RG, Curtis JR, Palliative Care A.Core competency for stroke neurologists. Stroke.2015;46(9):2714–9.

de Boer ME, Depla M, Wojtkowiak J, Visser MC,Widdershoven GA, Francke AL, et al. Life-and-deathdecision-making in the acute phase after a severestroke: interviews with relatives. Palliat Med. 2015;29(5):451–7.

De Ryck A, Fransen E, Brouns R, Geurden M, Peij D,Marien P, et al. Post-stroke depression and its multifac-torial nature: results from a prospective longitudinalstudy. J Neurol Sci. 2014;347:159–66.

duPreez AE, Smith MA, Liou JI, Frytak JR, Finch MD,Cleary JF, et al. Predictors of hospice utilization amongacute stroke patients who died within thirty days.J Palliat Med. 2008;11(9):1249–57.

Eastman P, McCarthy G, Brand CA, Weir L, Gorelik A, LeB. Who, why and when: stroke care unit patients seenby a palliative care service within a large metropolitanteaching hospital. BMJ Support Palliat Care. 2013;3(1):77–83.

Eriksson H, Milberg A, Hjelm K, Friedrichsen M. End oflife care for patients dying of stroke: a comparativeregistry study of stroke and cancer. PLoS One.2016;11(2):e0147694.

Feigin VL, Forouzanfar MH, Krishnamurthi R, MensahGA, Connor M, Bennett DA, et al. Global and regionalburden of stroke during 1990–2010: findings from theglobal burden of disease study 2010. Lancet. 2014;383(9913):245–55.

Frese A, Husstedt IW, Ringelstein EB, Evers S. Pharma-cologic treatment of central post-stroke pain. Clin JPain. 2006;22:252–60.

Frontera JA, Curtis JR, Nelson JE, Campbell M, GabrielM, Mosenthal AC, et al. Integrating palliative care intothe care of neurocritically ill patients: a report from theimproving palliative care in the ICU project advisoryboard and the center to advance palliative care. CritCare Med. 2015;43(9):1964–77.

Gardiner C, Harrison M, Ryan T, Jones A. Provision ofpalliative and end-of-life care in stroke units: a qualita-tive study. Palliat Med. 2013;27(9):855–60.

Goyal M, Demchuk AM, Menon BK, Eesa M, Rempel JL,Thornton J, et al. Randomized assessment of rapidendovascular treatment of ischemic stroke. NEJM.2015;372(11):1019–30.

Hackett ML, Pickles K. Part I: frequency of depressionafter stroke: an updated systematic review and meta-analysis of observational studies. Int J Stroke. 2014;9(8):1017–25.

Hackett ML, Anderson CS, House A, Xia J. Interventionsfor treating depression after stroke. Cochrane DatabaseSyst Rev. 2008;(4):CD003437. https://doi.org/10.1002/14651858.CD003437.pub3.

Hankey GJ, Jamrozik K, Broadhurst RJ, Forbes S, Ander-son CS. Long term disability after first-ever stroke andrelated prognostic factors in the Perth communitystroke study 1989-1990. Stroke. 2002;33:1034–40.

Hemphill JC, Greenberg SM, Anderson CS, Becker K,Bendok BR, Cushman M, et al. Guidelines for themanagement of spontaneous intracerebral hemorrhagea guideline for healthcare professionals from the


https://doi.org/10.1002/14651858.CD003437.pub3

https://doi.org/10.1002/14651858.CD003437.pub3

American Heart Association/American Stroke Associ-ation. Stroke. 2015;46(7):2032–60.

Holloway RG, Ladwig S, Robb J, Kelly A, Nielsen E, QuillTE. Palliative care consultations in hospitalized strokepatients. J Palliat Med. 2010;13(4):407–12.

Holloway RG, Arnold RM, Creutzfeldt CJ, Lewis EF, LutzBJ, McCann RM, et al. Palliative and end-of-life care instroke a statement for healthcare professionals from theAmerican Heart Association/American Stroke Associ-ation. Stroke. 2014;45(6):1887–916.

Intercollegiate Stroke Working Group. National clinicalguideline for stroke 5th edition [Online]. London:Royal College of Physicians; 2016. Available: www.strokeaudit.org/SupportFiles/Documents/Guidelines/2016-National-Clinical-Guideline-for-Stroke-5t-(1).aspx.

Jack C, Jones L, Jack BA, Gambles M, Murphy D,Ellershaw JE. Towards a good death: the impact ofthe care of the dying pathway in an acute stroke unit.Age Ageing. 2004;33(6):625–6.

Kalra L, Evans A, Perez I, Melbourn A, Pateal A, KnappM, et al. Training carers of stroke patients: randomizedcontrolled trial. BMJ. 2004;328:1099.

Kelly AG, Hoskins KD, Holloway RG. Early stroke mor-tality, patient preferences, and the withdrawal of carebias. Neurology. 2012;79(9):941–4.

Kim JS, Bashford G, Murphy TK, Martin A, Dror V,Cheung R. Safety and efficacy of pregabalin in patientswith central post-stroke pain. Pain. 2011;152(5):1018–23.

Langhorne P, Fearon P, Ronning OM, Kaste M, PalomakiH, Vemmos K, et al. Stroke unit care benefits patientswith intracerebral hemorrhage systematic review andmeta-analysis. Stroke. 2013;44(11):3044–9.

Leigon G, Boivie J. Central post-stroke pain: a controlledtrial of amitriptyline and carbamazepine. Pain.1989;36:27–36.

Maldonado NJ, Kazmi SO, Suarez JI. Update in the man-agement of acute ischemic stroke. Crit Care Clin.2014;30(4):673–97.

Mazzocato C, Michel-Nemitz J, Anwar D, Michel P. Thelast days of dying stroke patients referred to a palliativecare consult team in an acute hospital. Eur J Neurol.2010;17(1):73–7.

Mozaffarian D, Benjamin SJ, Go AS, Arnett DK, BlahaMJ, Cushman M, et al. Heart disease and stroke Statis-tics-2015 update: a report from the American HeartAssociation. Circulation. 2015;131:e29–e322.

Murthy SB, Moradiya Y, Hanley DF, Ziai WC. Palliativecare utilization in nontraumatic intracerebral hemor-rhage in the United States. Crit Care Med.2016;44:575–82.

National Stroke Foundation. Clinical guidelines for strokemanagement. Melbourne; 2010.

Neuberger J, Guthrie C, Aaronovitch D. More care, lesspathway: a review of the Liverpool care pathway. Lon-don: Department of Health; 2013.

Ntlholang O, Walsh S, Bradley D, Harbison J. Identifyingpalliative care issues in inpatients dying followingstroke. Ir J Med Sci. 2016;185:741–4.

Parry-Jones AR, Paley L, Bray BD, Hoffman AM, JamesM, Cloud GC, et al. Care-limiting decisions in acutestroke and association with survival: analyses of UKnational quality register data. Int J Stroke. 2016;11(3):321–31.

Payne S, Burton C, Addington-Hall J, Jones A. End-of-lifeissues in acute stroke care: a qualitative study of theexperiences and preferences of patients and families.Palliat Med. 2010;24(2):146–53.

Raffaeli W, Minella CE, Magnani F, Sarti D. Population-based study of central post-stroke pain in Rimini Dis-trict, Italy. J Pain Res. 2013;6:705–11.

Robinson RG, Jorge RE. Post-stroke depression: a review.Am J Psychiatry. 2016;173(3):221–31.

Rogers A, Addington-Hall J. Care of the dying strokepatients in the acute setting. J Res Nurs. 2005;10(2):153–67.

Saposnik G, Hill MD, O’Donnell M, Fang J, Hachinski V,Kapral MK. Variables associated with 7-day, 30-day,and 1-year fatality after ischemic stroke. Stroke.2008;39(8):2318–24.

Simmons BB, Parks SM. Intracerebral hemorrhage for thepalliative care provider: what you need to know.J Palliat Med. 2008;11(10):1336–9.

Smith EE, Hassan KA, Fang J, Selchen D, Kapral MK,Saposnik G, Investigators of the Registry of the Cana-dian Stroke Network, Stroke Outcome Research Can-ada (SORCan) Working Group. Do all ischemic strokesubtypes benefit from organized inpatient stroke care?Neurology. 2010a;75(5):456–62.

Smith EE, Shobha N, Dai D, Olson DM, Reeves MJ, SaverJL, et al. Risk score for in-hospital ischemic strokemortality derived and validated within the get with theguidelines–stroke program. Circulation. 2010b;122(15):1496–504.

Stevens T, Payne SA, Burton C, Addington-Hall J, JonesA. Palliative care in stroke: a critical review of theliterature. Palliat Med. 2007;21(4):323–31.

van Heugten C, Visser-Meily A, Post M, Lindeman E. Carefor carers of stroke patients: evidence-based clinicalpractice guidelines. J Rehabil Med. 2006;38(3):153–8.

Vartiainen N, Perchet C, Magnin M, Creac’h C, Convers P,Nighoghossian N, et al. Thalamic pain: anatomical andphysiological indices of prediction. Brain. 2016;139(3):708–22.


http://www.strokeaudit.org/SupportFiles/Documents/Guidelines/2016-National-Clinical-Guideline-for-Stroke-5t-(1).aspx.



Wendler D, Rid A. Systematic review: the effect on surro-gates of making treatment decisions for others. AnnIntern Med. 2011;154(5):336–46.

World Health Organisation. The top 10 causes of death.2014. http://www.who.int/mediacentre/factsheets/fs310/en/. Accessed 18 Nov 2016.

Young AJ, Rogers A, Dent L, Addington-Hall JM. Expe-riences of hospital care reported by bereaved relatives

of patients after a stroke: a retrospective survey usingthe VOICES questionnaire. J Adv Nurs. 2009;65(10):2161–74.

Zahuranec DB, Brown DL, Lisabeth LD, Gonzales NR,Longwell PJ, Smith MA, et al. Early care limitationsindependently predict mortality after intracerebral hem-orrhage. Neurology. 2007;68(20):1651–7.


http://www.who.int/mediacentre/factsheets/fs310/en

http://www.who.int/mediacentre/factsheets/fs310/en

Palliative Care and Strokestroke is recognized and treatment commenced,...

Documents

Transcript of Palliative Care and Strokestroke is recognized and treatment commenced,...