Endovascular treatment of acute ischaemic stroke in octogenarians and nonagenarians compared with younger patients

Abstract

Purpose: Endovascular therapy for emergent large vessel occlusion has been established as the standard approach for acute ischaemic stroke. However, the effectiveness and safety of endovascular therapy in the very elderly population has not been proved.

Objective: To determine the safety and effectiveness of endovascular therapy in octogenarians and nonagenarians.

Methods: We retrospectively reviewed all patients who underwent endovascular therapy at two stroke centres between April 2012 and July 2018. Functional outcome was assessed using the modified Rankin scale at 90 days after stroke or at discharge. A favourable outcome was defined as a modified Rankin scale score of 0–2 or not worsening of the modified Rankin scale score before stroke. Outcome was compared between younger patients (aged 46–79 years, n = 40) and octogenarians and nonagenarians (aged 80–97 years, n = 19).

Results: Octogenarian and nonagenarian patients had pre-stroke functional deficit (modified Rankin scale score >1) more frequently than younger patients (57.9% vs. 20.0%, respectively, P = 0.0059). No difference was observed between very elderly and younger patients in the rate of successful reperfusion (89.5% vs. 67.5%, respectively, P = 0.11), favourable functional outcome (47.4% vs. 45.0%, respectively, P = 1.00) and mortality (21.1% vs. 27.5%, respectively, P = 1.00). On multiple regression analysis, successful reperfusion, concomitant use of intravenous thrombolysis, and out-of-hospital onset were independent predictors of favourable outcome (P = 0.0003, 0.015 and 0.028, respectively).

Conclusions: Successful reperfusion, concomitant use of intravenous thrombolysis, and out-of-hospital onset were clinical predictors of favourable outcome. However, we did not observe an age-dependent effect of clinical outcome after endovascular therapy.

Introduction

Since the publication of six randomised trials,^1–6 endovascular therapy (EVT) mainly using a stent retriever has been established as the standard approach for acute ischaemic stroke patients with emergent large vessel occlusion (ELVO).⁷ However, in four of these six positive trials, an age limit of 80 or 85 years was applied, and the other two trials enrolled only 81 and 85 patients aged 80 years or older, respectively. In addition, all six trials required independent premorbid function as an inclusion criterion. Subgroup analysis in these trials showed the efficacy of EVT in the elderly.³ Similarly, the SWIFT PRIME, ESCAPE, EXTEND-IA and REVASCAT (SEER) collaboration pooled a total of 129 patients from the four trials and showed a significant reduction in mortality in the subgroup of patients aged 80 years or older compared to younger patients.⁸ The HERMES trial performed pooled analysis on 198 patients aged 80 years and older and also showed the efficacy of EVT.⁹ However, because some authors reported age as a predictive factor of poor clinical outcome and increased mortality,^10,11 the efficacy and safety of EVT in octogenarians and nonagenarians remains controversial.

Conversely, the number of Japanese patients treated by EVT in the clinical setting has been growing due to the ageing of society. Therefore, the aim of this study was to analyse retrospectively the safety and efficacy of EVT in elderly patients aged 80 year and older, including patients with pre-stroke morbidity. We compared the outcomes between elderly patients (≥80 years old) and younger patients (<80 years old) following EVT. As more than half of the elderly patients recruited had pre-stroke morbidity, we set the primary endpoint as a modified Rankin scale (mRS) score of 0–2 or not worsening of the mRS score as a favourable outcome.

Methods

Study sample

We retrospectively analysed consecutive, prospectively collected patients with ELVO who were treated with EVT at two stroke centres between April 2012 and August 2018. ELVO was diagnosed using computed tomography (CT)/CT angiography, magnetic resonance imaging (MRI)/magnetic resonance angiography (MRA), and both of CT and MRI/MRA in four, 41 and14 patients, respectively. One patient with a pre-stroke mRS score greater than 4 and two patients who had an embolic stroke during endovascular surgery for cerebral aneurysm were excluded from the analysis. This study was reviewed and approved by institutional review boards at Kyoto Katsura Hospital, Kyoto, Japan.

Digital subtraction angiography

All endovascular procedures were performed by experienced neurointerventional surgeons using a biplane angiography unit (Allura Xper FD20/20; Philips Medical Systems, Best, The Netherlands) with patients under conscious sedation. In one patient, general anaesthesia was introduced because of a deterioration of consciousness during the procedure. Endovascular access was achieved by way of a standard transfemoral approach, except for two patients in whom a transbrachial approach was used due to difficulty in catheterisation of the right common carotid artery by the transfemoral approach. A MERCI retriever (Concentric Medical, Mountain View, CA, USA), aspiration device, stent retriever and combined use of aspiration and a stent retriever were used as a first line treatment in two, 13, 18 and 24 procedures, respectively. Final reperfusion was assessed using a modified thrombolysis in cerebral infarction (TICI) scale including TICI 2c, i.e. near complete perfusion except for slow or distal emboli in a few distal cortical vessels.¹²

Clinical outcome

Functional outcome was assessed using the mRS at 90 days after stroke or at discharge. A favourable outcome was defined as an mRS score of 0–2 or not worsening of the mRS before stroke, because it is not uncommon that elderly patients have pre-stroke morbidity. Futile reperfusion was considered when a favourable outcome was not obtained in spite of successful reperfusion.

Statistical analysis

Demographic, clinical and angiographic variables were compared between the elderly (80 years old) and younger patients. Continuous variables are reported as mean ± standard deviation or as median. All statistical analyses were performed using JMP software for Mac, version 11 (SAS Software, Cary, NC, USA). Predictors of interests included age, National Institutes of Health Stroke scale (NIHSS) on admission, successful reperfusion, concomitant use of intravenous thrombolysis, location of occluded vessel (anterior circulation vs. posterior circulation), site of stroke presentation (inhospital vs. out of hospital), use of stent retriever, procedure time and pre-stroke morbidity. Age was dichotomised to 46–79 years versus 80–97 years. Variable selection in the model was performed by backward selection. Significance was established at the 0.05 level (two-sided).

Results

A total of 59 patients underwent EVT for ELVO at the two stroke centres during the study period. Nineteen patients were older than 79 years and 40 were aged 46–79 years. The baseline clinical and radiological characteristics of the patients are summarised in Table 1. The mean pretreatment NIHSS was 23.4 for the elderly patients and 20.4 for the younger patients, which were not significantly different (P = 0.13). Pre-stroke morbidity was observed more often in the elderly patients than in the younger patients (57.9% vs. 20.0%, respectively, P = 0.0059). Elderly patients had a tendency to have cardiogenic stroke, but it was not significantly different between the groups (84.2% vs. 62.5%, respectively, P = 0.13). There were no differences between both groups in sex distribution, the use of intravenous thrombolysis and location of the occluded vessels (anterior circulation vs. posterior circulation). Elderly patients were treated more frequently in the later period of this study (from 2016 onwards) when our workflow to manage acute ischaemic stroke patients was improved drastically.

Table 1.

Baseline characteristics of the cohort.

Variable	Age <80 (n = 40)	Age ≥80 (n = 19)	P value
Age, means (SD)	68.5 (8.69)	87.0(4.99)
Women	15 (37.5)	11 (57.9)	0.17
Since 2016	19 (47.5)	16 (84.2)	0.01
Inhospital onset	4	11	0.75
Cardiogenic emboli	25 (62.5)	16 (84.2)	0.13
tPA iv	21 (52.5)	13 (68.4)	0.28
ASPECTs, mean (SD)	7.85 (0.36)	8.64 (0.51)	0.19
NIHSS, mean (SD)	20.40 (9.43)	23.42 (9.50)	0.13
Prestroke morbidity (mRS >1)	8 (20.0)	11 (57.9)	0.0059
Location of arterial occlusion			0.41
ICA	15	10
MCA	19	8
BA	6	1
Stent retriever use	25 (62.5)	16 (84.2)	0.13
Combined use of aspiration and stent	13 (32.5)	11 (57.9)	0.09
Procedure time, mean (SD)	81.78 (5.54)	65.58 (8.04)	0.10

tPA: tissue plasminogen activator; iv: intravenous; SD: standard deviation; NIHSS: National Institutes of Health Stroke scale; mRS: modified Rankin scale; ICA: internal carotid artery; MCA: middle cerebral artery; BA: basilar artery.

The safety and efficacy outcomes between the groups are shown in Table 2. The frequency of successful reperfusion (TICI 2b, 2c or 3) was not significantly different between the groups (89.5% vs. 67.5%, respectively, P = 0.11). Interestingly, near or complete reperfusion (TICI 2c or 3) was observed more frequently in the elderly group than in the younger group (57.9% vs. 27.5%, respectively, P = 0.042). Symptomatic intracerebral haemorrhage was observed in one patient (2.5%) in the younger group. The distribution of 90-day mRS of both groups is shown in Figure 1. There was no difference in futile revascularisation, procedure time, favourable outcomes and mortality between both groups.

Table 2.

Angiographic and clinical outcomes of the cohorts: values in parenthesis indicate percentage.

Variable	Age <80 years (n = 40)	Age ≥80 years(n = 19)	P value
TICI 2b/2c/3	29 (67.5)	17 (89.5)	0.11
TICI 2c/3	11 (27.5)	11 (57.9)	0.042
Futile revascularisation	12/27 (44.4)	9/17 (53.0)	0.75
Symptomatic ICH	1 (2.5)	0	1
mRS 0–2	16 (40)	6 (31.6)	0.58
mRS 0–2 or no worsening of mRS	18 (45)	9 (47.4)	1
Mortality	11 (27.5)	5 (21.1)	1
Procedure time, minutes	81.78 (36.53)	65.58 (31.63)	0.96

TICI: thrombolysis in cerebral infarction; ICH: intracerebral haemorrhage; mRS: modified Rankin scale.

Figure 1.

Distribution of the 90-day modified Rankin scale in elderly and younger patients is shown. There was no difference between both groups (P = 0.38).

The associations between potential clinical factors and favourable functional outcome are summarized in Table 3. Multivariate analysis revealed that successful reperfusion, the concomitant use of intravenous thrombolysis, and out-of-hospital outset were independent predictors of favourable outcome (P = 0.0003, 0.015 and 0.028, respectively). Age, location of occluded vessel (anterior circulation vs. posterior circulation) and baseline NIHSS did not affect functional outcome (P > 0.05).

Table 3.

Multivariate analysis showing prognostic factors of favourable outcome after EVT.

	P value
Successful recanalisation	0.0003
Inhospital vs. out-of-hospital onset	0.028
NIHSS	0.24
Age (<80 vs. ≥80 years)	0.38
tPA iv	0.015

EVT: endovascular therapy; NIHSS: National Institutes of Health Stroke scale; tPA: tissue plasminogen activator; iv: intravenous.

Discussion

Age

Since the publication of several randomised trials,^1–6 an increasing number of patients with acute ischaemic stroke is being treated by EVT. In Japan, as the elderly population is increasing rapidly, progressively more octogenarians or nonagenarians have been treated by EVT, in spite of the fact that there is little evidence to show the benefit of EVT in this subgroup. Although pooled data analysis of randomised controlled trials also showed the benefit of EVT in this subgroup,⁹ there was a limited number of patients older than 79 years. Thus the age-dependency of clinical outcome after EVT remains controversial.^8,11,13–23 In the present study, age did not affect the clinical outcome irrespective of whether pre-stroke morbidity was considered. Similarly, a higher age was not associated with a longer procedure time, adverse events and poorer revascularisation rate. Interestingly, near or complete occlusion of TICI 2c or 3 was observed more frequently in the elderly patients for unknown reasons and, as a result, we had more favourable outcomes than reported previously^20,24 (Table 4). One possible explanation for this is that after publication of the positive randomised controlled trials, we instituted a drastic improvement of our workflow for the management of stroke patients with large vessel occlusion, and the combined use of a stent retriever and aspiration catheter²⁵ was used as a first-line procedure afterwards. Another explanation is that elderly patients were treated more often with intravenous thrombolysis, although it was not statistically significant between both groups.

Table 4.

Comparison with historical data.

	Systemic review20	Multicentre study24	Present series
Number	860	346	19
TICI 2b/3	78	88	90
TICI 2c/3	n.a.	38	58
Procedure time, minutes	99	45	66
Symptomatic ICH	8	7	0
mRS 0–2	27	21	32
Mortality	34	38	21

TICI: thrombolysis in cerebral infarction; ICH: intracerebral haemorrhage; mRS: modified Rankin scale.

National Institutes of Health Stroke scale

The NIHSS has been accepted as an established determinant of clinical outcome after stroke.²⁶ However, the significance of baseline NIHSS has become less important in the neurothrombectomy era,²⁷ because a remarkable improvement in the NIHSS can be expected after successful reperfusion of large vessel occlusion even when baseline NIHSS is high.^3,4

Successful reperfusion

There was no difference in the rate of successful reperfusion between both groups. Moreover, near or complete reperfusion of modified TICI 2c/3, which is highly associated with a favourable outcome,²⁸ was achieved more often in the elderly patients. Thus we should not be concerned about technical difficulties when considering the indication of EVT for large vessel occlusion in the elderly. Successful reperfusion was the most important predictor of favourable clinical outcome after EVT in our analysis, as reported in previous studies.^29,30 Our data suggest that a favourable outcome can be expected after successful reperfusion even in the elderly, and EVT should not be withheld in this population simply because of age.

Concomitant use of intravenous thrombolysis

In our data, the concomitant use of intravenous thrombolysis was significantly associated with a favourable outcome. One of theoretical advantages of the concomitant use of intravenous thrombolysis is that tissue plasminogen activator-induced fibrin degradation may lead to easier clot detachment from an occluded vessel wall.³¹ Another is its potential contribution to the recanalisation of microemboli migrating distally from endovascularly treated vessels. A recent subgroup analysis of the ESCAPE trial (Small Core and Anterior Circulation Proximal Occlusion with Emphasis on Minimizing CT to Recanalization Times Trial) showed that the concomitant use of thrombolysis reduced the likelihood of an infarct in a new previously unaffected territory complicating EVT.³² Conversely, the disadvantages of the concomitant use of intravenous thrombolysis include the possible exacerbation of haemorrhagic transformation, delay of workflow and cost. Thus the benefits and risks of the concomitant use of intravenous thrombolysis remain controversial.^33–38 The results of ongoing randomised controlled studies comparing direct EVT and EVT with concomitant intravenous thrombolysis, such as SWIFT DIRECT (NCT03192332), will be of great interest to resolve this issue.

Inhospital presentation

Inhospital stroke constitutes between 6.5% and 15.0% of all strokes.³⁹ It is well known that it is associated with a poor outcome,^39–42 because many inhospital stroke patients are admitted to hospital for surgical procedures, concomitant malignancy or cardiac disease. A delay in the recognition, assessment and contraindication to intravenous thrombolysis are suggested to be other causative factors.^39,40,42 In our analysis, 15 patients had inhospital stroke, constituting 25.4% of the cohort, and were associated with a poorer outcome compared to the out-of-hospital stroke patients (26.7% vs. 52.3%, respectively). Five of these 15 patients (33.3%) had a concomitant malignancy and 10 of them (66.7%) had congestive heart failure. In addition, six of them (40.0%) had stroke during or after the surgical procedure. The mortality of inhospital stroke associated with malignancy and that of perioperative stroke was 60% and 33.3%, respectively. Four of the six patients (66.7%) who were originally admitted to hospital due to congestive heart failure and subsequently had a stroke eventually died in hospital.

Limitations

The limitations of our study are the small number of patients and its retrospective design. In addition, the pre-stroke mRS was determined after discussions with the patients’ families or was taken from the medical records. This study, as far as we know, is the first report of the clinical outcomes of octogenarians and nonagenarians treated by EVT considering pre-stroke mRS, and the outcomes were better than those reported previously in this group of patients.

Conclusions

In our analysis, more than half of the octogenarians and nonagenarians treated by EVT had pre-stroke morbidity. Successful reperfusion, the concomitant use of intravenous thrombolysis and out-of-hospital onset were clinical predictors of favourable outcome. However, we did not observe an age-dependent effect of clinical outcome after EVT. Thus EVT should not be withheld for large vessel occlusion in octogenarians and nonagenarians with acute ischaemic stroke.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Conflict of interest

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.