Acute stroke in older people

Older people typically experience a poorer prognosis after stroke, with increased morbidity and mortality and poorer functional outcome.Reference Nakayama, Jørgensen, Raaschou and Olsen⁷ This is likely to be due to a number of factors including impaired regenerative capabilities, increased levels of co-morbid disease and organ dysfunction. Atrial fibrillation, cardiac failure, ischaemic heart disease, diabetes mellitus and previous stroke are all associated with a high risk of poor functional outcome and death following stroke, and all are more common in older people.Reference Nakayama, Jørgensen, Raaschou and Olsen^7, Reference Hier and Edelstein⁸ It is therefore arguable that it is more critical in this population that patients receive early effective comprehensive assessment and intervention.

Despite the poorer prognosis, there is good evidence that age should not be a deciding factor in stroke management.Reference Derex and Nighoghossian⁹ Whilst some of the associated poorer outcome is understandable given co-morbidities, there is concern that it may also be due, in part, to under-investigation and under-treatment. Evidence from the second UK National Sentinel Audit of StrokeReference Rudd, Hoffman, Down, Pearson and Lowe¹⁰ suggested that older people received less good quality care and were less liable to be treated within national guidelines than were younger people. They underwent brain scans less urgentlyReference Rudd, Hoffman, Down, Pearson and Lowe¹⁰ and older people have also been found to be less likely to have carotid imaging and (if indicated) carotid endarterectomy performed,Reference Fairhead and Rothwell¹¹ despite good evidence that they may actually benefit more from such interventions.Reference Rothwell, Eliasziw, Gutnikov, Warlow and Barnett¹² Older people are also less likely to be commenced on secondary prevention therapies or to be admitted to a stroke unit for rehabilitation,Reference Rudd, Hoffman, Down, Pearson and Lowe¹⁰ despite evidence that they obtain significant benefit from such interventions.

Preventing stroke

Whilst the purpose of this review is to consider acute management of stroke, it is important to emphasize that the best way of stopping stroke-related disability is to prevent the stroke from happening in the first place. Common modifiable risk factors for stroke in older people include hypertension, atrial fibrillation, arterial disease including ischaemic heart disease and carotid artery stenosis, diabetes, sedentary lifestyle, hyperlipidaemia, excess alcohol consumption and previous history of stroke or transient ischaemic attack (TIA). As with stroke in the younger population, elderly people should receive the primary and secondary preventative measures, as they are more susceptible to both stroke and its recurrence.⁵

Whilst many trials of prevention have excluded older people, there is robust evidence from other trials that primary and secondary prevention in older subjects should not be disregarded. The benefit of anti-platelet agents in the secondary prevention of stroke and TIA does not substantially reduce with age,Reference Gubitz and Sandercock¹³ and the BAFTA trial showed that anticoagulation of selected patients over 80 years with atrial fibrillation reduced stroke risk by >50% compared with aspirin.Reference Mant, Hobbs, Fletcher, Roalfe, Fitzmaurice, Lip and Murray¹⁴ The HYVET studyReference Beckett, Peters, Fletcher, Staessen, Liu, Dumitrascu, Stoyanovsky, Antikainen, Nikitin, Anderson, Belhani, Forette, Rajkumar, Thijs, Banya and Bulpitt¹⁵ showed the effectiveness of modest (average 15.0/6.1 mmHg) blood pressure lowering in reducing stroke risk by 30% in those aged over 80 years. The SPARCL trialReference Amarenco, Bogousslavsky, Callahan, Goldstein, Hennerici, Rudolph, Sillesen, Simunovic, Szarek, Welch and Zivin¹⁶ of the use of atorvastatin in secondary stroke prevention has been criticised for recruiting insufficient older people to allow sub-group analysis, but it did recruit people up to the age of 92 years, and the results showed no heterogeneity of effect of reduction in risk of recurrent stroke or TIA associated with age. A review of the results of the large NASCET and ESCT symptomatic carotid endarterectomy trials by Rothwell and colleaguesReference Rothwell, Eliasziw, Gutnikov, Warlow and Barnett¹² found an increased benefit of such intervention in those >75 years and indeed suggested that there was little benefit from operating on moderate (50–69%) internal carotid artery stenosis in those under 75 years.

Pre-hospital stroke

In a developing cerebral infarction, an area of intermediate blood flow (‘penumbra’) surrounding the infarct core is typically found. This penumbra, supplied by collateral vessels, represents potentially recoverable tissue that may transiently be exhibiting reduced or absent function through impaired metabolic function, but may be recoverable by acute interventions in some people.Reference Donnan, Baron, Ma and Davis¹⁷ It is this objective of trying to protect and recover penumbratous tissue that is the basis of modern acute stroke care. There is evidence that age-related changes in oxygen metabolism, vascular disease and in brain tissue reduce the volume of potentially recoverable penumbratous tissue and may also influence the duration over which the tissue may be recovered. Therefore, in order to reduce the morbidity and mortality associated with stroke in older people, it is essential that it is diagnosed and acute management initiated in a timely manner.

Outcome following stroke is better in patients who recognize their symptoms and then present rapidly to emergency services.Reference Feldmann, Gordon, Brooks, Brass, Fayad, Sawaya, Nazareno and Levine^19, Reference Kwan, Hand and Sandercock²⁰ Subjects who present within the first 4.5 hours experience improved outcomes,Reference Davalos, Castillo and Martinez-Vila^21, Reference Silvestrelli, Parnetti, Paciaroni, Caso, Corea, Vitali, Capocchi and Agnelli²² and may be eligible to receive thrombolytic therapy.Reference Lichtman, Watanabe, Allen, Jones, Dostal and Goldstein²³ Thus patient outcomes following acute strokes are better if care is organized and early access is facilitated. This ‘Stroke Chain of Survival’ emphasizes the need for coherent and timely identification and intervention in the stroke patient.Reference Adams, del Zoppo, Alberts, Bhatt, Brass, Furlan, Grubb, Higashida, Jauch, Kidwell, Lyden, Morgenstern, Qureshi, Rosenwasser, Scott and Wijdicks²⁴ This chain begins with the patient or patient's relatives identifying the stroke and urgently calling the emergency services. Hospital arrival is quicker for those who telephone for emergency services than for those who first contact their primary care physician or the hospital directly.Reference Harbison, Hossain, Jenkinson, Davis, Louw and Ford²⁵ Therefore, people should be educated on stroke recognition and made aware that they should call 911 (999) in the event of stroke. A recent analysis of data from the OXVASC studyReference Chandratheva, Lasserson, Geraghty and Rothwell²⁶ showed an increasing tendency for people to contact emergency services with symptoms of stroke or TIA sooner as they aged. There is no clear evidence that older people present later to the emergency department to seek stroke care.Reference Kwan, Hand and Sandercock²⁰ Although a study by Muangpaisan et al. showed that patients aged over 80 were more likely to present to the emergency department late,Reference Muangpaisan, Hinkle, Westwood, Kennedy and Buchan²⁷ other studies have not supported this finding.Reference Azzimondi, Bassein, Fiorani, Nonino, Montaguti, Celin, Re and D'Alessandro²⁸ Lacy and colleaguesReference Lacy, Suh, Bueno and Kostis²⁹ showed that subjects >55 years were more likely to present within 3 h of stroke onset, although this tendency was less marked in those over 80 years.

What is more consistently found to be a risk for delayed presentation is living alone.Reference Derex and Nighoghossian^9, Reference Wester, Rådberg, Lundgren and Peltonen³⁰ This is relevant to older people in that they are, perhaps, the population at risk of stroke most likely to be living alone. The Office of National Statistics in the UK, for example, has reported that, at the time of the last UK census, 60% of women and 29% of men over the age of 75 live alone. Therefore it is likely that the incidence of unwitnessed stroke symptom onset is higher in this group, thus increasing the time to presentation to the emergency department. It is probable that pre-existing disabilities, morbidities or new stroke-related disability may result in an older person not being able to call on emergency services, although these factors have not been evaluated to date.

Emergency department care

At the emergency department, the patient should be triaged as high-priority and the relevant laboratory tests and neuroimaging performed without delay. In all patients, so-called stroke mimics (conditions presenting like stroke, such as seizures, sepsis and hypoglycaemia) should be identified and managed appropriately.Reference Nor and Ford³¹ Hand and colleaguesReference Hand, Kwan, Lindley, Dennis and Wardlaw³² found that the presence of cognitive impairment or confusion in a patient presenting to the emergency department were associated with a significantly increased risk of the patient having a stroke mimic compared with a non-confused subject, which may suggest an increased risk of such presentation in older people. However, Vroomen and colleaguesReference Vroomen, Buddingh, Luijckx and Keyser³³ found that older people were substantially less likely to present with a ‘mimic’ than those <50 years (3% vs. 21%). They found that mimics such as complex migraines and conversion disorders were significantly more common in the young, whereas seizures were a little more common in older people. With increasing tendency towards intensive treatment of stroke patients there have been reports of patients with stroke mimics undergoing therapies such as thrombolysis. Winkler and colleagues found no effect of increasing age on likelihood of being thrombolysed inappropriately.Reference Winkler, Fluri, Fuhr, Wetzel, Lyrer, Ruegg and Engelter³⁴ Differentiation between strokes and mimics requires careful attention to history and physical examination, especially in older people where concomitant medical and neurological conditions are more likely to be present. Assessment tools, such as the Recognition of Stroke in the Emergency Room (ROSIER)Reference Nor, Davis, Sen, Shipsey, Louw, Dyker, Davis and Ford³⁵ scale, can be of use to rapidly screen for more obvious mimics, but no tool currently available can fully replace careful clinical assessment.

Brain imaging can be of help in supporting a diagnosis of acute stroke, but the diagnosis itself remains clinical. CT scanning remains the most common form of initial brain imaging in most parts of the world, and careful examination of CT head scans by experienced clinicians can permit identification of early signs of stroke within the first hour of the event in some subjects, although such identification can be very inconsistent.Reference Vu and Lev³⁶ Interpretation of CT scans of the head in older people can be more challenging than in younger subjects. Prevalence of undiagnosed background cerebrovascular disease increases in older people, especially those with vascular risks. Silent lacunar infarcts may be found in about 50% of those >80 yearsReference Vermeer, Koudstaal, Oudkerk, Hofman and Breteler³⁷ and background white matter disease or leukoaraiosis is seen to some extent in the majority of older subjects presenting with stroke.Reference Koton, Schwammenthal, Merzeliak, Philips, Tsabari, Orion, Dichtiar and Tanne³⁸ The presence of such concomitant ischaemic disease can make identification of early ischaemic changes more difficult. Tools such as the Alberta Stroke Program early CT score (ASPECTS) can be of assistance in evaluating scans,Reference Pexman, Barber, Hill, Sevick, Demchuk, Hudon, Hu and Buchan³⁹ with good inter-observer agreement found in identifying infarct size in the middle cerebral artery territory.Reference Demaerschalk, Silver, Wong, Merino, Tamayo and Hachinski^40, Reference Puetz, Dzialowski, Hill and Demchuk⁴¹

Hospitals are increasingly using more sophisticated CT imaging techniques such as CT angiography (CTA) and CT Perfusion (CTP) scanningReference Schaefer, Barak, Kamalian, Gharai, Schwamm, Gonzalez and Lev^42, Reference Köhrmann, Jüttler, Huttner, Nowe and Schellinger⁴³ to identify vascular occlusions and presence of recoverable penumbratous brain, particularly in circumstances where specialist neuroradiology or stroke services may not be as plentiful. Whilst these techniques may clearly be of advantage to some people, they do require administration of intravenous contrast, and in the case of CT perfusion this contrast is concentrated. Thus the techniques need to be used with caution where a subject has, or is suspected to have, renal impairment. A recent retrospective study by Hopyan and colleagues studying contrast induced nephropathy in subjects undergoing CTA or CTP for suspected stroke showed an incidence of 2.9%.Reference Hopyan, Gladstone, Mallia, Schiff, Fox, Symons, Buck, Black and Aviv⁴⁴ This is highly consistent with the 3% incidence recorded in another recent retrospective study by Mehdiratta and colleagues.Reference Mehdiratta, Schlaug, Kumar, Caplan and Selim⁴⁵ It is of interest that 3 of 5 subjects who suffered renal damage in Hopyan's population were subjects >75 years undergoing both CTA and CTP.

There is some evidence that older subjects may struggle to be scanned as quickly following admission as younger people. The 2004 National Stroke Audit in the UK reported that neuroimaging was performed within 24 h in 51% of those aged over 85 compared with 71% of those under 65 years of age.Reference Rudd, Hoffman, Down, Pearson and Lowe¹⁰ This is despite the study by Wardlaw and colleagues, which demonstrated in a cost benefit analysis of three different scanning strategies using a 70-year-old patient model, that immediate scanning following admission was most effective both in terms of cost and quality of life gains.

Thrombolysis in older people

Approximately 80% of strokes are thrombo-embolic. Reperfusing an area of brain with a thrombolytic agent following acute occlusion of a cerebral vessel may salvage the penumbra, thereby reducing the volume of dead brain tissue and resulting cerebral ischaemia, leading to a potentially better clinical outcome in the patient.Reference Wahlgren, Ahmed, Dávalos, Ford, Grond, Hacke, Hennerici, Kaste, Kuelkens, Larrue, Lees, Roine, Soinne, Toni and Vanhooren⁴⁷ This process of recovery of the penumbra may be important in older subjects where brain plasticityReference Todd, Kimber, Ridding and Semmler⁴⁸ and rehabilitation outcomes are possibly not as good,Reference Ergeletzis, Kevorkian and Rintala⁴⁹ perhaps less because of advancing age and more because of pre-morbid disability.Reference Luk, Cheung, Ho and Li⁵⁰

Recombinant tissue plasminogen activator (rtPA) is the only licensed thrombolytic agent in Europe for use in acute ischaemic stroke. The first large trial proving its efficacy was the NINDS thrombolysis trial published in 1995.⁵¹ Based predominantly on this and the results of the two ECASS studies,Reference Steiner, Bluhmki, Kaste, Toni, Trouillas, von Kummer and Hacke^52, Reference Hacke, Kaste, Fieschi, von Kummer, Davalos, Meier, Larrue, Bluhmki, Davis, Donnan, Schneider, Diez-Tejedor and Trouillas⁵³ the European Medicines Evaluations Agency licensed rtPA in 2002 to be used within 3 hours of stroke symptom onset at a dose of 0.9 mg/kg. Unfortunately the licence granted at that time was limited to subjects less than 80 years of age as the published data set for patients older than that was only 42 patients. In 2008 the ECASS 3 studyReference Hacke, Kaste, Bluhmki, Brozman, Dávalos, Guidetti, Larrue, Lees, Medeghri, Machnig, Schneider, von Kummer, Wahlgren and Toni⁵⁴ showed benefit for patients treated up to 4.5 h from stroke onset. However, this trial did not include subjects >80 years either, perversely because the extant licence was only for patients up to 80 years.

Evidence for thrombolysis in subjects >80 years is copious, but little of it is from randomized controlled trials. Observational studies consistently show a less favourable outcome in older subjects than is found in younger subjects.Reference Meretoja, Putaala, Tatlisumak, Atula, Artto, Curtze, Häppölä, Lindsberg, Mustanoja, Piironen, Pitkäniemi, Rantanen, Sairanen, Salonen, Silvennoinen, Soinne, Strbian, Tiainen and Kaste⁵⁵ However, this poorer outcome appears not to be due to an increased risk of intracerebral haemorrhage, the most common complication of thrombolytic therapy.Reference Sylaja, Cote, Buchan and Hill⁵⁶ At least in part it may be due to the high prevalence of co-morbid conditions in older people that affect stroke outcome, such as the presence of heart failure or atrial fibrillation. Patients suffering stroke over the age of 80 years typically have poorer outcomes without thrombolysisReference Hier and Edelstein⁸ and a recent study comparing outcomes of thrombolysis between subjects in the SITS and VISTA databases showed similar degrees of benefit in subjects below and above 80 years compared with age-matched controls.Reference Mishra, Ahmed and Andersen⁵⁷ This study used data from the SITS international stroke register on nearly 3500 subjects over 80 years who had received intravenous thrombolysis since the register's creation in 2002, so it is clear that many centres internationally will administer this therapy to selected older subjects outside the licence. The multinational, Third International Stroke TrialReference Whiteley, Lindley, Wardlaw and Sandercock⁵⁸ is continuing to randomize subjects >80 years to thrombolytic or conventional treatment, and its results will hopefully inform further thrombolytic treatment of older subjects.

Intra-arterial thrombolysis is typically performed in patients with extensive internal carotid, basilar artery and large middle cerebral artery thrombosis, where potential benefit of intravenous therapy is thought to be limited.Reference Meyers, Schumacher, Higashida, Barnwell, Creager, Gupta, McDougall, Pandey, Sacks and Wechsler⁵⁹ Evidence for the use of intra-arterial thrombolysis in older subjects is sparse. One retrospective study of 33 subjects showed equivalent safety but worse outcomes in those over 80 years,Reference Kim, Ford, Kidwell, Starkman, Vinuela, Duckwiler, Jahan and Saver⁶⁰ but there is little or no published trial-based evidence to guide practice in the elderly population.

Mechanical thrombectomy can be performed in subjects as an alternative or supplemental therapy to thrombolysis. Whilst a number of papers have been published showing that older subjects were entered into trials, there have been no reports to date specifically commenting on the efficacy of this approach in these patients.

Hemicraniectomy is performed following cerebral infarction in the distribution of the middle cerebral or internal carotid arteries where the subject suffers symptoms of raised intracranial pressure secondary to the developing oedema of the infarcted tissue, which is associated with a mortality rate in excess of 80%.Reference Hacke, Schwab, Horn, Spranger, DeGeorgia and von Kummer⁶¹ Surgical decompression has been found to benefit some patients in terms of reducing mortality and increasing the number independent after the stroke. Unfortunately, a systematic review of observational studies showed that the intervention had little benefit in subjects >50 yearsReference Gupta, Connolly, Mayer and Elkind⁶² and a number of subsequent clinical trials have been performed which excluded subjects >60 years.Reference Vahedi, Hofmeijer, Juettler, Vicaut, George, Algra, Amelink, Schmiedeck, Schwab, Rothwell, Bousser, Van Der Worp and Hacke⁶³ Thus hemicraniectomy for cerebral infarction is not typically performed in subjects over 60 years of age.

Intracerebral haemorrhage

Non-traumatic intracerebral haemorrhage causes between 10 and 30% of strokes admitted to acute hospitals, depending on populations studied.Reference Sacco, Mayer and Feldmann⁶⁴ As in the case of cerebral infarction, the incidence of intracerebral haemorrhage doubles with every decade over 50 years. The common causes of intracerebral haemorrhage in older subjects are hypertension, which typically results in deep ganglionic bleeds and cerebral amyloid angiopathy, which typically produces more superficial lobular haemorrhages that can recur in 25–40% of cases. There is currently no proven effective acute treatment for intracranial haemorrhage at any age, but American Heart Association guidelinesReference Broderick, Connolly, Feldmann, Hanley, Kase, Krieger, Mayberg, Morgenstern, Ogilvy, Vespa and Zuccarello⁶⁵ suggest that cautious lowering of systolic blood pressure below 180 mmHg in hypertensive subjects should be considered to limit haematoma expansion.Reference Anderson, Huang, Wang, Arima, Neal, Peng, Heeley, Skulina, Parsons, Kim, Tao, Li, Jiang, Tai, Zhang, Xu, Cheng, Heritier, Morgenstern and Chalmers⁶⁶ The Surgical Treatment of Intracerebral Haemorrhage (STICH) trialReference Mendelow, Gregson, Fernandes, Murray, Teasdale, Hope, Karimi, Shaw and Barer⁶⁷ showed no benefit in the surgical evacuation of intracerebral haematomas, although there may have been some benefit from treating superficial lobular haemorrhages (which may be more frequent in older subjects with amyloid angiopathy). The STICH 2 trialReference Mendelow and Unterberg⁶⁸ is further investigating this sub-group. It is generally agreed that patients with substantial cerebellar haemorrhages should be considered for decompressive surgeryReference Qureshi, Mendelow and Hanley⁶⁹ as a reasonably small volume of blood in the posterior cerebral fossa can result in significant pressure and damage to the brainstem.

Subarachnoid haemorrhage

Subarachnoid haemorrhage (SAH) makes up 3% of all strokes.Reference Sudlow and Warlow⁷⁰ In the elderly population it leads to a more unfavourable outcome than in the young.Reference Lanzino, Kassell, Germanson, Kongable, Truskowski, Torner and Jane⁷¹ This may be due to poorer management,Reference O'Sullivan, Dorward, Whittle, Steers and Miller⁷² worse neurological deficits at presentationReference Sugita, Nukui, Kobayashi, Horikoshi and Yagishita⁷³ and a higher incidence of co-morbidities. The risk rises with age and is 60% higher in those aged over 85 compared with those aged between 45 and 55 years.Reference de Rooij, Linn, Van Der Plas, Algra and Rinkel⁷⁴

Causes of SAH in the elderly differ from those in the young. Elderly patients are much less likely to have vascular malformations as a cause.Reference Stapf, Khaw, Sciacca, Hofmeister, Schumacher, Pile-Spellman, Mast, Mohr and Hartmann⁷⁵ The older patient is more likely to have head trauma leading to SAH due to an increased tendency to fall.Reference Masud and Morris⁷⁶ The older patient is also more likely to be on anticoagulant therapy, with a tendency towards increased bleeding risk from this treatment.Reference Torn, Bollen, Van Der Meer, Van Der Wall and Rosendaal⁷⁷ With age, there is also an increased tendency to develop amyloid angiopathy and these patients can often have multiple recurrent bleeds of both intracerebral and subarachnoid varieties.Reference Miller, Wardlaw and Lammie⁷⁸

The presentation of SAH in the elderly patient is typically similar to that in the young. There is often no focal neurological deficit. Fluctuating level of consciousness is often an important clue to diagnosis, as are headache, nausea, vomiting and impaired ability to concentrate.

There is evidence that intervention, including surgery, remains a beneficial treatment option in those >80 years.Reference Fridriksson, Hillman, Saveland and Brandt⁷⁹ In order to manage secondary insults including raised intracranial pressure, hypertension, hypoxaemia, pyrexia, hyperglycaemia, ischaemic deterioration, tentorial herniation, seizures, rebleeding and infection, all of which are associated with a poorer outcome, patients should ideally be admitted to a neurointensive care unit.Reference Enblad and Persson⁸⁰ Admission of the elderly patient to the neurointensive care unit is associated with an overall improved outcome.Reference Johansson, Cesarini, Contant, Persson and Enblad⁸¹

Early stroke unit care

Although much attention is given to more dynamic interventions such as thrombolysis, these are currently available to <10% of people suffering stroke in the UK, and the potential number of people eligible to receive them may be <20%. They cannot be used at all in patients with intracerebral haemorrhages. Other less dramatic treatments are of potential benefit to most or all stroke patients admitted acutely to hospital, perhaps the foremost of which is stroke unit care.

Effective stroke units have three essential components. The first, and possibly most important, is a specialist multidisciplinary team, the second a discreet, defined ward area for caring for the patients, and finally a robust governance structure with evidence-based pathways and protocols to define care.⁸² The Stroke Unit Trialists Collaboration meta-analyses⁸³ have shown that care in stroke units substantially reduces risk of both death and permanent disability by about one-third. This is supported by recently published evidence from the Canadian Stroke Network register.Reference Smith, Hassan, Fang, Selchen, Kapral and Saposnik⁸⁴ This intervention also results in a benefit to the whole stroke population with, for example, increased availability of stroke unit care to a large majority of the Scottish population being associated with a reduction in 6-month stroke mortality of 36% and an increase in discharge home of 28%.Reference Langhorne, Lewsey, Jhund, Gillies, Chalmers, Redpath, Briggs, Walters, Capewell, McMurray and Macintyre⁸⁵

Several studies have now confirmed that all age groups benefit from stroke unit care.Reference Terént, Asplund, Farahmand, Henriksson, Norrving, Stegmayr, Wester, Asberg and Asberg⁸⁶ The degree of early benefit may be smaller in older people. For example, a recent analysis of the Canadian Stroke Registry showed an absolute risk reduction of 2.3% (10.3 to 8.0%) in 7-day stroke mortality associated with stroke unit care, compared with an overall reduction of 3.7% (8.6 to 4.9%) for all age groups,Reference Saposnik, Kapral, Coutts, Fang, Demchuk and Hill⁸⁷ but the risk reduction in the over-80s was still significant. The effect on 30-day mortality was more pronounced, with stroke unit care in the over-80s being associated with an absolute risk reduction of death greater than in any other age group (over-80s, 5.5% reduction; overall, 4.6% reduction). What is perhaps more important is the effect of organization of care, with the highest level of organization being found to be associated with the lowest mortality and with the effect being most pronounced in the over-80 years age group.

Conflicts of interest

Neither author has a conflict of interest in respect of this paper.