Article Text

Download PDFPDF

Original research
Thrombectomy for acute ischemic stroke in nonagenarians compared with octogenarians
  1. Eric S Sussman1,
  2. Blake Martin2,
  3. Michael Mlynash3,
  4. Michael P Marks1,
  5. David Marcellus2,
  6. Gregory Albers3,
  7. Maarten Lansberg3,
  8. Robert Dodd1,
  9. Huy M Do1,
  10. Jeremy J Heit1
  1. 1 Neurosurgery and Radiology, Stanford University Medical Center, Stanford, California, USA
  2. 2 Department of Radiology, Stanford University, Stanford, California, USA
  3. 3 Department of Neurology, Stanford Stroke Center, Stanford University, Stanford, California, USA
  1. Correspondence to Eric S Sussman, Neurosurgery and Radiology, Stanford University Medical Center, Stanford, California, USA; esuss11{at}


Introduction Multiple randomized trials have shown that endovascular thrombectomy (EVT) leads to improved outcomes in acute ischemic stroke (AIS) due to large vessel occlusion (LVO). Elderly patients were poorly represented in these trials, and the efficacy of EVT in nonagenarian patients remains uncertain.

Methods We performed a retrospective cohort study at a single center. Inclusion criteria were: age 80–99, LVO, core infarct <70 mL, and salvageable penumbra. Patients were stratified into octogenarian (80–89) and nonagenarian (90–99) cohorts. The primary outcome was the ordinal score on the modified Rankin Scale (mRS) at 90 days. Secondary outcomes included dichotomized functional outcome (mRS ≤2 vs mRS ≥3), successful revascularization, symptomatic intracranial hemorrhage (ICH), and mortality.

Results 108 patients met the inclusion criteria, including 79 octogenarians (73%) and 29 nonagenarians (27%). Nonagenarians were more likely to be female (86% vs 58%; p<0.01); there were no other differences between groups in terms of demographics, medical comorbidities, or treatment characteristics. Successful revascularization (TICI 2b–3) was achieved in 79% in both cohorts. Median mRS at 90 days was 5 in octogenarians and 6 in nonagenarians (p=0.09). Functional independence (mRS ≤2) at 90 days was achieved in 12.5% and 19.7% of nonagenarians and octogenarians, respectively (p=0.54). Symptomatic ICH occurred in 21.4% and 6.4% (p=0.03), and 90-day mortality rate was 63% and 40.9% (p=0.07) in nonagenarians and octogenarians, respectively.

Conclusions Nonagenarians may be at higher risk of symptomatic ICH than octogenarians, despite similar stroke- and treatment-related factors. While there was a trend towards higher mortality and worse functional outcomes in nonagenarians, the difference was not statistically significant in this relatively small retrospective study.

  • ischemic stroke
  • endovascular thrombectomy
  • nonagenarian
  • perfusion imaging

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


The management of acute ischemic stroke (AIS) due to large vessel occlusion (LVO) has changed dramatically over the past 5 years, as several randomized controlled trials have demonstrated significantly better functional outcomes in patients treated with endovascular thrombectomy (EVT).1–5 As a result of these landmark clinical trials, EVT has become the standard of care for AIS due to LVO.6

While the eligibility criteria used in the early stroke trials were relatively stringent, subsequent clinical trials have demonstrated the efficacy of EVT in expanded time windows,7 8 and several additional studies are underway to evaluate the efficacy of EVT in specific patient populations that were excluded from or under-represented in the previous trials.9–11 A randomized clinical trial evaluating the efficacy of EVT specifically in elderly patients with AIS due to LVO has not yet been completed, so the rationale for performing EVT in elderly patients is largely based on anecdotal experience and on generalization of clinical trial results from a younger patient population.

The available data suggest a positive treatment effect of EVT in the octogenarian stroke patient population,12–14 despite overall worse outcomes compared with younger patient populations.15–18 In contrast, there are insufficient data to determine the utility of EVT in the nonagenarian patient population. We sought to evaluate the efficacy of EVT in nonagenarians.

Methods and materials

The data that support the findings of this study are available from the corresponding author on reasonable request.

Patient cohort

Our Institutional Review Board approved this retrospective cohort study, which complied with the Health Insurance Portability and Accountability Act. Patient consent was waived by our review board. We performed a retrospective cohort study of all patients triaged for AIS due to LVO at our comprehensive stroke center between August 2010 and May 2018. Patients aged 80–89 (octogenarians) and 90–99 (nonagenarians) were selected for analysis.

Imaging evaluation

Patients underwent imaging evaluation with either MRI or CT, including angiography and perfusion imaging. Patients selected for EVT had AIS due to LVO of the internal carotid artery or first or second segment of the middle cerebral artery (MCA); core infarction <70 mL based on MRI DWI or CT perfusion (CTP) imaging with a relative cerebral blood volume <30% compared with the contralateral normal hemisphere; and a target mismatch between core and penumbra (mismatch ratio 1.8 and mismatch volume >15 mL).7 19

Endovascular thrombectomy (EVT) treatment details

EVT was performed on a Siemens Artis Zee biplane system in a dedicated neuroangiography suite. Patients were preferentially treated with conscious sedation monitored by an anesthesiologist, with conversion to general anesthesia in select cases based on physician discretion. All procedures were performed after obtaining common femoral artery access with an 8 Fr vascular sheath. A 6 Fr guiding catheter was positioned in the cervical internal carotid artery or V1 segment of the vertebral artery proximal to the LVO. Thrombectomy was performed with a combination of aspiration and stent retriever-mediated thrombectomy using FDA-approved devices.

Post-thrombectomy angiographic revascularization was quantified using the Thrombolysis in Cerebral Infarction (TICI) score, and successful revascularization was defined as TICI 2b–3.


Patient clinical and outcome data were determined from a prospectively maintained database of patients with AIS and from the electronic medical record. The primary outcome was the ordinal score on the modified Rankin Scale (mRS) at 90 days. Secondary outcomes included (1) binary clinical outcome that included good outcome (mRS ≤2) and poor outcome (mRS ≥3); (2) successful revascularization; (3) symptomatic intracranial hemorrhage, which was defined as a subarachnoid hemorrhage or parenchymal hematoma (PH1 or PH2)20 with a corresponding increase of ≥4 from the pre-treatment National Institutes of Health Stroke Scale (NIHSS) score; and (4) patient mortality at 90 days.

Statistical analysis

Patient demographics, clinical variables, and imaging data were compared between octogenarians and nonagenarians using the Fisher’s exact test for categorical variables and with a two-tailed t-test (two-sample assuming unequal variance) for continuous variables. The association between age and clinical outcome was assessed using an ordinal logistic regression model that was fitted to the primary outcome, and binary logistic regression models for each of the secondary outcomes. For these ordinal logistic regression models, the proportional odds assumption had to be met (p>0.05) before further analysis. Significance was set at alpha <0.05, and all reported results are two-sided. Statistical analysis was done using SAS 9.4.


One-hundred and eight patients met the inclusion criteria for this study, including 79 octogenarians (73.1%) and 29 nonagenarians (26.9%). There was a higher proportion of women in the nonagenarian cohort than in the octogenarian cohort (86.2% vs 58.2%; p<0.01). There were no other differences between the groups in terms of baseline demographics, functional status, or medical history (table 1). Left-sided LVOs accounted for 37.9% of nonagenarian cases and 64.6% of octogenarian cases (p=0.02); otherwise, there was no difference between the groups in terms of stroke characteristics (table 2). There was no difference in EVT metrics—including nearly equivalent rates of successful reperfusion defined as TICI 2b–3—between the two cohorts (79.3% in nonagenarians vs 78.5% in octogenarians, p=1.0; table 3).

Table 1

Baseline demographic and clinical data

Table 2

Stroke characteristics

Table 3

Endovascular thrombectomy metrics

Complete clinical outcome data are shown in table 4 and figure 1. With regard to the primary outcome (ordinal score on mRS at 90 days), there was a non-significant trend towards worse functional outcomes in the nonagenarian cohort compared with the octogenarian cohort (median mRS 6 vs 5, respectively; p=0.09). Good functional outcome, defined as mRS ≤2, was achieved in 12.5% of nonagenarians and in 19.7% of octogenarians (p=0.54). There was also a trend towards increased mortality in nonagenarians compared with octagenarians (63.0% vs 40.9%; p=0.07). There was a statistically significantly higher rate of symptomatic intracranial hemorrhage in the nonagenarian cohort than in the octogenarian cohort (21.4% vs 6.4%; p=0.03).

Figure 1

Modified Rankin Scale scores 90 days after treatment.

Table 4

Clinical outcomes

During the study time period, 12 additional nonagenarian patients were evaluated but were ultimately not taken for endovascular therapy, either due to prohibitively large core infarction, unfavorable target mismatch profile, or family preference to forego invasive therapy (see online supplementary table 1).


We found significantly higher rates of symptomatic reperfusion hemorrhage following EVT in nonagenarians compared with octogenarians, and a trend towards worse overall clinical outcomes and a higher mortality rate in the nonagenarian cohort, despite similar baseline functional status, stroke severity, and equivalent rates of revascularization. In fact, the rate of significant post-EVT hemorrhagic complications (21.4%) was nearly twice as high as the rate of good functional outcome (12.5%) in the nonagenarian cohort. This is particularly notable in the context of the stringent radiographic triage algorithm used at our institution,21 which selects optimal candidates for endovascular therapy with relatively small core infarctions and significant territory at risk, and hypothetically excludes patients that are predisposed to procedure-related complications and poor clinical outcomes.

Despite the relatively poor outcomes in very elderly patients in this series, these results should be interpreted in the context of the dismal natural history of AIS in this patient population. As shown in online supplementary table 1, nonagenarian patients who were triaged but ultimately not taken for EVT during the study period had similarly dismal outcomes, with numerically worse 3-month mRS scores. However, it is important to note that these patients were evaluated by a multidisciplinary stroke team and deemed to be poor candidates for EVT; a focused randomized controlled trial would be required to determine whether or not very elderly patients derive significant benefit from endovascular therapy.

While elderly patients are excluded from, or under-represented in, the majority of stroke clinical trials to date, this patient population accounts for a substantial proportion of all stroke patients in the USA.22 In the relevant literature, elderly patients are consistently reported to have worse clinical outcomes following AIS compared with their younger counterparts.14 22–24 However, these data are often confounded by other variables, including baseline stroke severity, functional status, and type of treatment administered. Furthermore, there is significant variability in age cut-off for classification as ‘elderly’, as well as inconsistency in the comparison group against which clinical outcomes in elderly stroke patients are measured.

The HERMES Collaboration, which pooled individual patient data from the five pivotal 2015 stroke clinical trials, included 198 patients aged ≥80 and identified a correlation between older age and functional dependence in both the intervention and control groups, despite a relatively increased effect size of intervention in the elderly subgroup.14 In contrast, Alawieh and colleagues17 18 reported the non-clinical trial real-world experience at a single center as well as at a multicenter cohort of comprehensive stroke centers, and demonstrated significantly higher rates of functional dependence, mortality, and hemorrhage in patients aged ≥80 compared with younger patients. In the single-center series, these authors did not identify a positive treatment effect of EVT over medical management in elderly patients.18 Similarly, Duffis and colleagues16 performed a meta-analysis of 2729 patients across eight studies and also found significantly higher rates of functional dependence, mortality, and symptomatic hemorrhage in those patients aged ≥80 compared with their younger counterparts.

Comparable thrombectomy data in the nonagenarian stroke population are more sparse. Mateen et al compared the clinical outcomes following intravenous thrombolysis in nonagenarians versus octogenarians and found no significant difference in symptomatic reperfusion hemorrhage rate, 30-day functional outcomes, or 90-day mortality between the two cohorts.23 In a similar comparison of clinical outcomes following IV tissue plasminogen activator (tPA) in nonagenarians and octogenarians, Sarikaya et al found lower rates of functional independence and higher rates of symptomatic ICH and mortality among nonagenarians.24 Notably, however, both of these studies included patients treated with IV tPA alone and are therefore not generalizable to the interventional treatment paradigm that has become the standard of care for AIS patients with LVO. Khan et al compared clinical outcomes of nonagenarian patients treated with EVT to those of all younger patients (age 18–89 years) and found a higher rate of poor outcome, defined as mRS >2, in the nonagenarian population, which was at least partially attributed to higher pre-stroke mRS scores in the nonagenarian cohort. There was no significant difference in mRS shift between nonagenarians and younger patients; however, when age was analyzed as a continuous variable, the authors identified a 5.1% increased risk of a poor outcome with every 1 year increase in age.25

Importantly, we found similar rates of functional independence and mortality to prior studies of EVT in elderly patients with AIS due to LVO.17 18 25 26 More specifically, we reported a rate of good functional outcome in 17.8% of the combined octogenarian and nonagenarian cohorts compared with 20.5% and 21% in the single-center and multicenter studies by Alawieh et al, respectively.17 18 In the nonagenarian cohort alone we found good functional outcomes in 12.5% compared with 11.1% in the study by Khan et al.25 Mortality and hemorrhage occurred in 47.3% and 38.5%, respectively, of all patients in our study compared with 38% and 42% in the multicenter study by Alawieh et al.17

There are several limitations to this study. Due to its retrospective design, there is intrinsic selection bias as individual patients were triaged for EVT at the discretion of the clinical team. Furthermore, the relatively small sample size, particularly of the nonagenarian cohort, may have limited our ability to detect statistically significant differences between the octogenarian and nonagenarian cohorts. Notably, however, this is among the largest reported series of nonagenarians undergoing EVT for AIS due to LVO.


Nonagenarian patients undergoing EVT for AIS due to LVO are at significantly higher risk of symptomatic reperfusion hemorrhage compared with octogenarians, despite similar stroke- and treatment-related factors. While there was a strong trend towards higher mortality rates and worse long-term functional outcomes in nonagenarians, the difference was not statistically significant in this relatively small retrospective study. Additional prospective and randomized studies are necessary to evaluate the efficacy of EVT in elderly patients, including nonagenarians.



  • ESS and BM contributed equally.

  • Contributors Each of the authors listed on this manuscript contributed substantially to the study design, data collection, data analysis, and manuscript review. Each of the authors has reviewed the final manuscript and has agreed with the submission of this manuscript to JNIS.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Patient consent for publication Not required.