Article Text

Download PDFPDF

Original research
Utilization of CT perfusion patient selection for mechanical thrombectomy irrespective of time: a comparison of functional outcomes and complications
  1. Aquilla S Turk1,
  2. Eric MacKenzie Nyberg1,
  3. M Imran Chaudry1,
  4. Raymond D Turner2,
  5. Jordan Asher Magarik2,
  6. Joyce S Nicholas2,
  7. Christine A Holmstedt2,
  8. Julio Alejandro Chalela2,
  9. Angela Hays2,
  10. Christos Lazaridis2,
  11. Marc I Chimowitz2,
  12. Tanya N Turan2,
  13. Robert J Adams2,
  14. Edward C Jauch2
  1. 1Department of Radiology, Medical University of South Carolina, Charleston, South Carolina, USA
  2. 2Department of Neurosciences, Medical University of South Carolina, Charleston, South Carolina, USA
  1. Correspondence to Dr Aquilla S Turk, Department of Radiology, Medical University of South Carolina, 169 Ashley Ave, Charleston, SC 29425, USA; turk{at}


Background Patient selection for acute ischemic stroke has been largely driven by time-based criteria, although emerging data suggest that image-based criteria may be useful. The purpose of this study was to directly compare outcomes of patients treated within a traditional time window with those treated beyond this benchmark when CT perfusion (CTP) imaging was used as the primary selection tool.

Methods A prospectively collected database of all patients with acute ischemic stroke who received intra-arterial therapy at the Medical University of South Carolina was retrospectively analyzed, regardless of time from symptom onset. At presentation, CTP maps were qualitatively assessed. Selected patients underwent intra-arterial therapy. Functional outcome according to the modified Rankin scale (mRS) score at about 90 days was documented.

Results 140 patients were included in the study. The median time from symptom onset to groin access was 7.0 h. Overall, 28 patients (20%) had bleeding complications, but only 10 (7.1%) were symptomatic. The average National Institute of Health Stroke Scale (NIHSS) score for patients treated ≤7 h from symptom onset was 17.3 and 30.2% had a mRS score of 0–2 at 90 days. Patients treated >7 h from symptom onset had an average NIHSS score of 15.1 and 45.5% achieved a mRS score of 0–2 at 90 days (p=0.104). Patients in the two groups had similar rates of symptomatic intracerebral hemorrhage (8.5% and 5.8%, respectively; p=0.745).

Conclusions No difference was found in the rates of good functional outcome between patients treated ≤7 h and those treated >7 h from symptom onset. These data suggest that imaging-based patient selection is a safe and viable methodology.

  • Stroke
  • CT perfusion

Statistics from


Acute ischemic stroke remains a devastating condition and is a leading cause of morbidity and mortality affecting an estimated 800 000 people per year in the USA at an estimated cost of $41 billion in 2007.1 The most devastating strokes are generally those caused by proximal occlusions in the cervical and cerebral vasculature. Recanalization of these vessels with reperfusion of ischemic brain parenchyma is consistently associated with improved outcomes.2–4 Intravenous administration of tissue plasminogen activator (tPA) is approved for use within 3 h of symptom onset, with newer evidence suggesting a potential benefit to 4.5 h.5–8 However, intravenous tPA is poorly effective in recanalizing large vessel occlusions. Moreover, a recent meta-analysis of 502 patients with a perfusion mismatch treated with intravenous tPA beyond the approved time window failed to show evidence of benefit.9 Intra-arterial thrombolysis within 6 h and mechanical thrombectomy within 8 h have been shown to be effective in vessel recanalization.2 ,3 ,10–13

Recommendations for intra-arterial therapy inclusion criteria aimed at recanalization and reperfusion remain controversial and guidelines have generally been based on time from symptom onset. However, these guidelines were established prior to the widespread use of CT perfusion (CTP) imaging, which provides physiologic information regarding the area of brain infarcted and the surrounding potentially viable brain as well as the location of vascular occlusion. Thus, CTP has the potential to differentiate between patients in whom ischemia has progressed to infarct where recanalization would be futile and those with significant penumbra who may benefit from recanalization and reperfusion.14–18 Several recent reports have suggested that, with image-based selection, patients may be effectively treated beyond these strict time boundaries with good outcomes.19–23

Current trials such as the Mechanical Retrieval and Recanalization of Stroke Clots Using Embolectomy Trial (MR RESCUE), the Penumbra THERAPY Trial and the Efficacy and Safety Study of Desmoteplace to treat Acute Ischemic Stroke Trial (DIAS-3) are evaluating the usefulness of perfusion imaging to select acute stroke patients for treatment beyond conventional time limitations. In this study we evaluate the usefulness of patient selection for intra-arterial therapy based on CTP imaging rather than time-guided criteria by directly comparing functional outcomes and complications of patients treated within and outside conventional treatment time windows.


An Institutional Review Board approved retrospective review of a prospectively managed clinical database of all acute ischemic stroke intervention cases performed from 1 January 2008 to 13 October 2011 was undertaken. Chart review was further performed through an electronic medical record to assess clinical course and outcome.

Patient characteristics documented included age, gender, National Institute of Health Stroke Scale (NIHSS) score at presentation, time to presentation from last normal and modified Rankin Scale (mRS) score at 90 days or closest follow-up period to 90 days. mRS data were obtained from the Neurology Clinic record. If the medical record did not contain outcome information, an independent neurologist contacted the patient or family in order to determine the mRS score. Radiologic and angiographic imaging were reviewed to document the location of the vascular occlusion, Thrombolysis In Cerebral Infarction (TICI) flow after the procedure, procedural complications and the presence of post- or periprocedural intracerebral hemorrhage.

Patient selection

All stroke patients presenting with acute ischemic stroke and NIHSS score ≥ 8 or with eloquent territory at risk, irrespective of symptom duration, were considered potential candidates for endovascular therapy. All patients underwent non-contrast CT (NCCT), CT angiography (CTA) and CTP on admission to the emergency department. NCCT imaging determined the presence and extent of gross infarction and excluded underlying pathology such as mass or intracranial hemorrhage (ICH). CTA was used to identify the location of the vascular occlusion. The operating neurointerventionalist analyzed the CTP images to identify the presence and extent of penumbra relative to the core area of infarction. The primary endpoint was the percentage of patients achieving good functional outcome as measured by 90-day mRS score ≤ 2. Symptomatic hemorrhage was defined as a change in NIHSS score of ≥ 4 from the patient's baseline. The primary safety endpoints were symptomatic intracranial hemorrhage (sICH) within 36 h and mortality at 90 days.

CT perfusion imaging techniques

All CT imaging was performed according to a standardized institutional protocol. All CTP scans were performed using a Siemens 64- or 16-slice scanner with a 50 ml bolus of Omnipaque 350 contrast (GE Healthcare, Milwaukee, Wisconsin, USA). Perfusion datasets were post-processed on a GE Advantage Windows Workstation (General Electric, Milwaukee, Wisconsin, USA), Siemens Leonardo Workstation (Siemens Medical, Germany), or both. The CTA acquisition was subsequently performed with an 80 ml bolus injection of contrast followed by a saline chase. Two-dimensional 1 cm thick slab reformations were created in the axial, sagittal and coronal projections. The data were used to localize the location of the major vessel occlusion.

Data processing

CTP post-processing on the GE Workstation was performed manually by a senior radiology resident or by an experienced CT technologist yielding mean transit time (MTT), cerebral blood volume (CBV) and cerebral blood flow (CBF) maps using a deconvolution methodology on a commercially available perfusion package (GE AW; GE Healthcare). An experienced CT technologist performed semi-automated CTP post-processing on the Siemens Workstation. MTT, CBV and CBF maps were reconstructed using a maximum slope methodology with delay correction on Siemens commercially available perfusion software (VPCT Neuro; Siemens Healthcare, Forchheim, Germany). The arterial input and venous outflow curves were analyzed to ensure complete data sets. All CTP interpretations for treatment were made by the attending neurointerventionalist based on qualitative interpretation of the perfusion maps. The MTT maps were qualitatively assessed to define the area of brain at risk as delineated by at least two color band differences on the six-spectrum rainbow scale from the surrounding unaffected brain. CBV was evaluated to delineate the core region of infarction as the area with depressed CBV at least two color band differences on the six-spectrum rainbow scale from the surrounding unaffected brain. CBF maps were used to confirm these findings and to further define the area at risk. Patients in whom one-third or more of the middle cerebral artery (MCA) territory was infarcted or with ≤50% penumbra were not considered candidates for endovascular treatment unless an eloquent area was at risk.

Endovascular treatment

Patients eligible for fibrinolytic therapy who presented within the 0–4.5 h time window were treated with intravenous tPA before going to the angiography suite. Consent for endovascular therapy was obtained in all cases. All cases were performed under general anesthesia. All patients received heparin, consisting of a 1500 Unit bolus followed by intermittent boluses to maintain a baseline to 1.5 times baseline activated clotting time. The primary method of treatment in most cases was mechanical aspiration with the Penumbra aspiration system. Intra-arterial thrombolysis with tPA was used adjunctively at low doses ranging from 10 to 20 mg (administered in 3–5 mg aliquots) during the procedure if aspiration was not immediately effective. In cases refractory to the Penumbra aspiration system, other modalities including balloon angioplasty, stent placement or the Merci retrieval system were used. Intra-arterial abciximab was administered if acute intraprocedural thrombus formation was present. If a permanent stent was used either to recanalize the thrombosed vessel or to treat a proximal stenosis, the patient was given a weight-based loading dose of abciximab during stent placement and then administered 300 mg aspirin and 600 mg clopidogrel through a nasogastric tube immediately following the procedure. All patients were managed postoperatively in the neurosurgical intensive care unit after the endovascular procedure to ensure close neurologic monitoring and strict blood pressure control. Control head CT scans were performed 24 h after stroke intervention.

Data analysis

Baseline demographic, radiographic and outcomes data were analyzed using descriptive statistics including means, SDs, frequencies, percentages and medians. The 7 h time point was chosen as the time to segregate the patients into two groups as it was the median time to treat which allows for symmetric group comparison. Also, it is in the middle of the 6–8 h time period used by intra-arterial trials to provide appropriate comparable patient groups to reported literature. Comparisons between times to treatment groups were made using two-sided t tests for continuous measures, χ2 tests for categorical measures and Mann–Whitney U tests for non-parametric measures (medians). No adjustments were made for multiple comparisons and p values of <0.05 were considered significant. An unadjusted logistic regression was performed to determine the impact of recanalization on clinical outcomes.


One hundred and forty patients underwent endovascular treatment, with 54 (38.6%) receiving intravenous tPA prior to intra-arterial therapy. The overall mean age was 66.5 years (median 68 years) and 46.8% of the patients were men. The mean NIHSS score at presentation to our institution was 16 (median 16), and the locations of vascular occlusion were 20.0% internal carotid artery, 65.7% MCA and 14.3% posterior circulation (basilar artery, posterior cerebral artery or posterior inferior cerebellar artery (table 1).

Table 1

Baseline characteristics for all patients

The average time from the last point at which the patient was seen normal to groin vascular access was 11.3 h (median 7 h, range 1.75–72 h). Revascularization of TICI 2–3 was achieved in 87.2% of patients. The devices used to achieve revascularization were Penumbra 054 (n=27), Penumbra 041 (n=32), Penumbra 032 (n=22), Penumbra 026 (n=14), balloon angioplasty (n=19), stent (n=6), Merci L4 (n=2).

Overall, 28 patients (20%) had post-procedural intracerebral hemorrhage of which 10 (7.1%) were sICH. There were 12 (8.6%) procedure-related complications, which included eight dissections and four perforations. Six of the dissections were small and non-flow-limiting while two required placement of a stent. Three of the perforations were small, typically controlled with protamine and balloon inflation for 3 min. One patient with progressive basilar artery thrombosis who suffered multiple brain stem strokes had a fatal perforation of the basilar artery. There was no difference in hemorrhage rates in the 54 patients who received intravenous tPA before intra-arterial therapy (10/54=18.5%) and those who did not receive tPA (18/86=20.9%).

All patients were selected for endovascular procedures based on clinical and CTA/CTP criteria. The median time of symptom duration before groin puncture was 7.0 h. Patients were divided into groups on either side of the median for subgroup analysis. There were no significant differences in age or gender between the groups (table 2). The mean NIHSS scores were 17.3 and 15.1 for the two groups. For the two time periods, 30.2% and 45.5% achieved a 90-day mRS score of ≤ 2 (p=0.104; table 2). Mortality rates were 30.2% vs 21.2% (p=0.420), and the rate of sICH was 8.7% vs 5.6% (p=0.745) for the two groups. There were no statistically significant differences in the locations of the occlusion, however patients with carotid occlusions were more likely to present prior to 7 h and those with posterior circulation occlusions were more likely to present after 7 h. Recanalization of TICI 2B-3 was slightly more frequent in patients treated after 7 h than those treated prior to 7 h (82.6% vs 71.8%, p=0.16; table 2). The odds of a good neurologic outcome among those with TICI 2B-3 revascularization were 2.57 times the odds of a good outcome in those with TICI 1-2A revascularization (p=0.047, 95% CI 1.012 to 6.531; table 2).

Table 2

Differences between groups


Patient selection for mechanical thrombectomy in the setting of acute ischemic stroke beyond current recommended time intervals remains a challenge. Our study demonstrated rates of good functional outcome that are comparable to those reported in previous stroke studies (table 3).

Table 3

Comparison with historical studies

The low rates of ICH in our study may have been related to the judicious use of intra-arterial tPA following initial revascularization with mechanical methods as well as close periprocedural clinical monitoring in a dedicated neurocritical care unit. Moreover, we did not detect a difference in the rate of poor outcome between the stratified treatment groups. Although the NIHSS score at presentation was slightly higher in the early treatment group, this difference is unlikely to be reflected in outcomes as the NIHSS score in both groups indicated large vessel occlusions with a very poor natural history.24 Our findings suggest that the combination of the clinical assessment and CTP imaging, rather than strict time parameters, appropriately aids in the selection of patients suitable for endovascular revascularization in acute ischemic stroke.

Our data suggest that appropriately selected patients can be treated with intra-arterial mechanical thrombectomy well beyond conventional time windows. Patients in the delayed treatment window in our study had a viable ischemic penumbra at a mean time to procedure start of 11.3 h, indicating that they probably had robust collateral networks. The trend towards a higher rate of good functional outcome in the delayed group suggests that these robust networks may have conferred a selection bias. This observation should not, however, diminish the importance of mechanical thrombectomy and definitive revascularization. Collateral leptomeningeal pathways are pressure-dependent and patients often present with unsustainable hypertension to attempt to maintain perfusion via collateral pathways. The natural history of patients with a high NIHSS score is known to be very poor and the importance of recanalization has been consistently demonstrated.9 ,25–27

Conventional time-based treatment eligibility benchmarks are based on the estimated time of progression from cerebral oligemia and ischemia to infarction. However, in a given individual, the probability of progression from ischemia to infarction is stochastic rather than deterministic, and time since last seen normal is a poor proxy for perfusion status. Individual factors including anatomic variations in the circle of Willis, extent of leptomeningeal collateral circulation and cerebral perfusion pressure accord considerable variability. Progression to infarction may be completed within several minutes in some patients while others may have large territories of salvageable tissue after several days.9 Thus, patients should be given the benefit of physiologic imaging before the possibility of mechanical thrombectomy is ruled out as this enables the selection of patients with viable ischemic penumbra whose prognosis should greatly improve if they are revascularized.

The PROACT II trial first documented that stroke patients with higher rates of recanalization (66% with intra-arterial thrombolysis vs 18% with placebo; p=0.001) were more likely to have a 90-day mRS score of ≤ 2 than non-recanalized patients (40% vs 25%; p=0.04).12 This was further supported by subgroup analysis in the multi-Merci trial in which 49% of revascularized patients achieved an mRS score of ≤ 2 compared with 10% of non-revascularized patients (p<0.001).8 This study also showed a significant increase in 90-day mortality in patients who were not revascularized compared with those who were able to be revascularized (52% vs 25%; p<0.001). Similarly, the Penumbra POST trial found that 45% of revascularized patients achieved an mRS score of ≤ 2 compared with 13% of non-recanalized patients.

The use of CTP has been advocated over the last decade as a means of describing the perfusion status of the brain. The core infarct territory can be delineated with CBV maps, while the tissue at risk can be delineated with other perfusion metrics including MTT, time to peak and CBF maps.17 ,28 The use of this physiologic approach to select patients within a specified time window is currently being studied in several randomized trials.29–31 It is also being applied to everyday practice to help select patients who present outside conventional time windows32 and has been used at our institution over the last several years to triage all stroke patients for potential intra-arterial therapy (table 3).

Intravenous tPA remains the first line of treatment for patients with acute ischemic stroke presenting within the 4.5 h window. While it has been shown to be safe and relatively effective, many limitations exist. Only 4% of acute stroke patients receive intravenous tPA, mostly related to patient presentation beyond the approved time window.33 Among patients presenting within the approved time window, nearly half are ineligible to receive intravenous tPA because of exclusionary criteria. Furthermore, intravenous tPA does not perform well in recanalizing and sustaining recanalization in large vessel occlusions and is therefore less likely to be effective in treating large strokes (eg, NIHSS>8), including the vast majority of patients treated in this study.34 Administration of intravenous tPA was not a strong predictor of good or poor outcome in our study.

Our study is limited by the single-center retrospective nature of analysis of a prospectively maintained patient database. Patients excluded by unfavorable perfusion imaging are not tracked by our database, precluding this group as a potential comparison arm. Additionally, questions have been raised regarding the quantitative rigor of CTP imaging,35 and there is necessarily some component of subjectivity in the interpretation of CTP maps as well as their integration into the overall clinical picture.


The use of CTP imaging to select patients for acute stroke interventions irrespective of time was safe and effective in our single-center experience, with patient outcomes similar to those seen in other acute stroke trials. Patients with favorable CTP imaging presenting beyond conventional treatment windows should be considered for mechanical thrombectomy. The application of this approach could have a considerable impact on the number of patients eligible for interventional therapy. Future multicenter studies are needed for additional verification of CTP image-guided selection, rather than time-guided criteria, for endovascular revascularization of patients with acute ischemic stroke.



  • Contributors All authors made a material contribution to this article and its revision and gave their final approval for submission of the article to this journal.

  • Funding This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors

  • Competing interests None.

  • Ethics approval This study was a retrospective chart review so ethical approval was not required.

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

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.