Article Text

This article has a correction. Please see:

Download PDFPDF

Original research
The POST trial: initial post-market experience of the Penumbra system: revascularization of large vessel occlusion in acute ischemic stroke in the United States and Europe
  1. Robert Tarr1,
  2. Dan Hsu1,
  3. Zsolt Kulcsar2,
  4. Christophe Bonvin2,
  5. Daniel Rufenacht2,
  6. Karsten Alfke3,
  7. Robert Stingele3,
  8. Olav Jansen3,
  9. Donald Frei4,
  10. Richard Bellon4,
  11. Michael Madison5,
  12. Tobias Struffert6,
  13. Arnd Dorfler6,
  14. Iris Q Grunwald7,
  15. Wolfgang Reith7,
  16. Anton Haass7
  1. 1University Hospital, Cleveland, Ohio, USA
  2. 2Hôpitaux Universitaires de Genève, Geneva, Switzerland
  3. 3University of Kiel Medical Center, Kiel, Germany
  4. 4Swedish Medical Center, Denver, Colorado, USA
  5. 5St Paul's United Hospital, St Paul, Minnesota, USA
  6. 6Universitatsklinikum Erlangen, Germany
  7. 7Universitätsklinikum des Saarlandes, Germany
  1. Correspondence to Robert Tarr, University Hospital Case Medical Center, 11100 Euclid Ave., Cleveland, OH 44106, USA; robert.tarr{at}


Background and purpose The purpose of this study was to assess the initial post-market experience of the device and how it is compared with the Penumbra Pivotal trial used to support the 510k application.

Methods A retrospective case review of 157 consecutive patients treated with the Penumbra system at seven international centers was performed. Primary endpoints were revascularization of the target vessel (TIMI score of 2 or 3), good functional outcome as defined by a modified Rankin scale (mRS) score of ≤2 and incidence of procedural serious adverse events. Results were compared with those of the Penumbra pivotal trial.

Results A total of 157 vessels were treated. Mean baseline values at enrollment were: age 65 years, NIHSS score 16. After use of the Penumbra system, 87% of the treated vessels were revascularized to TIMI 2 (54%) or 3 (33%) as compared with 82% reported in the Pivotal trial. Nine procedural serious adverse events were reported in 157 patients (5.7%). All-cause mortality was 20% (32/157), and 41% had a mRS of ≤2 at 90-day follow-up as compared with only 25% in the Pivotal trial. Patients who were successfully revascularized by the Penumbra system had significantly better outcomes than those who were not.

Conclusion Initial post-market experience of the Penumbra system revealed that the revascularization rate and safety profile of the device are comparable to those reported in the Pivotal trial. However, the proportion of patients who had good functional outcome was higher than expected.

  • Ischemic stroke
  • mechanical thrombectomy
  • Penumbra
  • recanalization
  • functional outcome
  • artery
  • brain
  • stroke
  • device
  • thrombectomy

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.


Stroke is the third leading cause of death in the USA.1 Limited treatment options exist for the estimated 800 000 people annually who experience a new or recurrent stroke. Of these, ∼610 000 are first attacks while 185 000 are recurrences.2 At current trends, this number is projected to increase to one million per annum by the year 2050. Among available treatment options for acute ischemic stroke are the intravenous thrombolytic recombinant tissue plasminogen activator (rt-PA), the intra-arterial mechanical thrombectomy devices such as the MERCI retriever (Concentric Medical, Mountain View, California, USA) and the recently approved Penumbra system (Penumbra Inc., Alameda, California, USA). Annually, only ∼1.8%–3% of all ischemic stroke patients are treated with intravenous rt-PA therapy, due to the narrow time window and other reasons. Furthermore, <5% of the ∼800 000 annual stroke victims actually receive either intravenous or intra-arterial treatments, resulting in a significant unmet need in addressing such a devastating disease.

In 2008, the Penumbra system was launched for commercial use as a new generation of neuro-thrombectomy devices designed to reduce the clot burden in large intracranial vessels in acute stroke. The Pivotal trial results for the Penumbra system in support of US Food and Drug Administration (FDA) 510(k) clearance using the MERCI clot retriever as the predicate device was recently published.3 The results indicated that while the Penumbra system has a high rate of revascularization (80%), it has a lower than expected rate of good functional outcome (25%).

The purpose of this study was to assess the post-market experience of the Penumbra system at several academic and community hospitals in the USA and Europe, with particular focus on the rate of revascularization and the proportion of patients who achieved functional independence at 90-day follow-up.


The POST trial was a retrospective case review of 157 consecutive patients at seven international centers specifically designed to assess the safety and effectiveness of a stroke cohort treated by the Penumbra system since approval. The results were compared with those from the Penumbra Pivotal trial reported earlier.3

The Penumbra system is a mechanical device designed to reduce clot burden in acute stroke. The system comprises two components: first, the reperfusion catheter that has an optimized design for efficient navigation and aspiration, and second, the separator, which is an operator-controlled component that cleans and clears the reperfusion catheter to enable continuous aspiration.

Data collection for the POST study was performed by a retrospective case review of stroke patients who had been treated by the Penumbra system in accordance with the instructions for use. To avoid bias, the first 157 patients treated by the Penumbra system at seven international centers since approval were enrolled into the study. Patients who were treated off label were not included. Data from hospital admission included the patient history and thrombolysis in myocardial infarction (TIMI) score by digital subtraction angiography of the occluded target vessel (table 1).4–6 In addition, investigators also assessed the neurological and functional status of the patients at admission, 7 days after the procedure or at the time of discharge and 90 days after the procedure. The National Institutes of Health Stroke Scale (NIHSS) and the modified Rankin Scale (mRS) scores were utilized, respectively, for patients' neurological status and functional independence (table 2).7 All serious adverse events that occurred between the time of the procedure and 90 days after were reported.

Table 1

Definition of the thrombolysis in myocardial infarction (TIMI) scores

Table 2

Definition of the modified Rankin Scale

Baseline demographics, vessel location, procedural information and study outcome variables were summarized using descriptive summary statistics. The continuous variable summary statistics presented include the sample size, mean, median, SD, minimum and maximum. Categorical variables were described using counts and percentages. The two-sided Fisher's exact test was employed to analyze the association between the binary outcome variables and the TIMI subgroups for the POST study. The POST and Pivotal outcome differences were assessed using the two-sided Fisher's exact test.

Primary endpoints were revascularization of the target vessel (TIMI score of 2 or 3), good functional outcome as defined by a mRS score of ≤2 (table 2) and incidence of procedural serious adverse events (SAEs) inclusive of intracerebral hemorrhage (ICH) at 24 h after the procedure. Symptomatic ICH is defined as any evidence of hemorrhage from head CT scans within 24 h post-procedure and a deterioration of the NIHSS score by ≥4 points at any time during the 90-day follow-up.


The results of the current study are presented in tables 3, 4 and 5. The baseline characteristics of the 157 subjects enrolled in the POST study were comparable to those of the Pivotal trial. The only exception is the location of the target vessels, where the POST trial tended to have a lower rate of middle cerebral artery occlusions (58%) and a higher rate of vertebral basilar arteries occlusions (20%) (table 3). However, neither of these differences was statistically significant (p<0.05).

Table 3

Baseline characteristics of the two groups

Table 4

Comparison of endpoints between the two groups

Table 5

Outcome measures: closed versus open vessels

Consistent with the baseline characteristics, the rates of revascularization were similar between the two studies (table 4). Total procedural SAEs tended to be higher in the POST study but that was not statistically significant. The rate of symptomatic intracranial hemorrhage at 24 h tended to be lower in the POST study (6.4%) than in the Pivotal (11%), but did not achieve statistical significance. However, all-cause deaths at 90 days was significantly lower in the POST study (20%) than the Pivotal study (33%) (p<0.05). Consistent with this finding was a significantly higher proportion of patients who were functionally independent after treatment in the POST study (41% vs 25%) (table 4).

Similar results were observed when patients with successful revascularization were compared with those who had failed. As shown in table 5, those who were successfully revascularized by the Penumbra system had better outcomes, leading to significantly lower mortality and a higher rate of good functional outcome (p≤0.01).

With regard to adjunctive therapies, 32% (51/157) of the patients had intra-arterial lytic treatment during the procedure, 18.5% (29/157) received intravenous lytic treatment prior to the procedure (refractory to rtPA) and 23% (36/157) had both. A review of the data in table 6 revealed that although the intra-arterial rtPA group tended to have a higher rate of symptomatic ICH, adjunctive lytic therapy did not materially affect the revascularization rate or clinical outcomes of the patients treated by the Penumbra system.

Table 6

Effects of rtPA on revascularization, clinical outcome and ICH

Among the 157 patients treated by the Penumbra system, there were nine procedural events reported in nine patients (5.7%). These were dissection (two), perforation (two), ICH (one), access site hematoma (one), peripheral hemorrhage (one) and cardiac arrest (one).

A total of three device malfunctions were reported: two were related to fractures of the 032 reperfusion catheter and one was a breakage of a 041 separator. Both fractures occurred ∼40 cm from the tip of the 032 reperfusion catheters. In the case of the 041 separator breakage, the investigator encountered a hard clot and had to use considerable force to pull the separator after penetration into the thrombus, which caused a breakage ∼6 mm from the distal end of the device. The small tip of the 041 separator stayed within the thrombus and was not retrievable. None of these malfunctions led to an SAE or patient death.


The Penumbra system is a second generation of mechanical thrombectomy devices approved by the US FDA specifically for the revascularization of large vessel occlusion in acute ischemic stroke. However, published results from the Pivotal trial in support of this claim have been controversial. For example, despite having one of the highest revascularization rates among all intra-arterial therapies, the proportion of treated patients achieving functional independence was lower than expected.3 8–19 This has been cited as a ‘major unanswered question’ of the study.20 Thus, it remains unclear whether the high recanalization rate of the Penumbra system can improve functional outcome.

In this study, we compared the safety and effectiveness of the Penumbra system in the post-market setting against that reported in Penumbra Pivotal trial. The results indicate that the rates of revascularization were similar between the two studies. This concordant finding confirms that the TIMI-defined high rate of revascularization found in the Pivotal study can be replicated. It should be noted that scoring of the blood flow in the target vessel by TIMI was done by the investigators at each participating center, without the benefit of a core laboratory. Yet, the overall distribution of the TIMI scores was quite similar across the seven centers. These results suggest that this system is effective in reducing clot burden in a ‘real world’ setting, even without the rigorous constraints of a prospective trial for regulatory approval.

With regard to procedural safety, the event rate observed in this study of 5.8%, although higher than that reported in the Pivotal study (3%), was within the range of other intra-arterial therapies.3 8–19 As described in the results section, these SAEs were similar to those expected of other endovascular therapies for acute ischemic stroke. The fact that the SAE rate is higher in this study as compared with the Pivotal is not surprising and is consistent with most post-market studies where patients eligible for treatment are not selected as stringently as in a trial for the purpose of regulatory submission.

Consistent with the procedural safety profile is the rate of ICH in patients treated with the Penumbra system. The 24 h ICH rate of 21% is in the low range of historical rates, such as the Second Prolyse in Acute Cerebral Thromboembolism trial (PROACT II) (35.9%),9 Interventional Management of Stroke (IMS; 48.8%),13 MERCI Phase II (35.5%)18 and Multi-MERCI (38.7%).19 Further, the symptomatic ICH rate of 6.4% was more favorable than those reported for PROACT II (10.9%)9 and Multi-MERCI (9.8%).18 These results, coupled with the lower than expected all-cause mortality rate, confirmed that the Penumbra system's high rate of revascularization is associated with an acceptable safety profile.

Perhaps the most intriguing finding of this present study was the higher than expected number of patients who were functional independent at 90-day follow-up. Of the 157 patients enrolled, 41% had good functional outcome as defined by a mRS score of ≤2, which was significantly higher than that reported in the Pivotal or MERCI trials.3 18 19 As stated previously, the extent to which mechanical thrombectomy devices in general, and the Penumbra system in particular, improve patient functional outcome is controversial.20 This stems, in the most part, from the incongruent findings between the high rate of revascularization and the low rate of good functional outcome in patients treated with these systems. But in POST, we were able to demonstrate that the favorable rate of revascularization by the Penumbra system can lead to a high rate of functional independence and patients with open vessels do have significantly better functional outcomes than those with closed vessels, including a lower rate of all-cause deaths (tables 4 and 5). In a review of the baseline characteristics between the POST and Pivotal patients, we failed to find any major differences between the two patient populations that could explain the difference in patient outcomes. Nor were we able to find any procedural data that could shed a light on the discordant observations between the two studies.

It has been suggested that patients with acute stroke are a diverse population, not easily differentiated by the current clinical or angiographic criteria used to define treatment eligibility, and that their pathophysiology could be the key to resolving this problem. For example, there is increasing evidence that an imaging-defined size of the ischemic penumbra or core infarct volume is a better predictor of patient functional outcomes, perhaps superior to the presenting neurological status of the patients as reflected by the NIHSS scores or simple time from presentation, which is not always reliable.21–26 Unfortunately, these types of imaging were not part of the standard of care at the seven participating hospitals and therefore not available for analysis. Other limitations of this study were the retrospective nature of the study design so the study cohort was not a prospective, intent-to-treat population; the TIMI scores were not adjudicated by a core laboratory and the revascularization procedure was conducted in accordance with the standard of care, without the rigor and stringent control of a trial designed for regulatory submissions. Taken together, they speak to the need that the results must still be confirmed by future prospective trials.

In summary, the results from this current study confirmed that the Penumbra system is an effective mechanical thrombectomy device for the reduction of clot burden in acute ischemic stroke, which resulted in a TIMI-defined high rate of revascularization. And unlike the Pivotal trial, a higher than expected proportion of the treated patients were able to achieve functional independence. While a review of our database did not allow an explanation for the observed differences in outcome, it did raise the intriguing possibility that the variability in functional outcome observed between different clinical trials is caused by the inherent variability in patient stroke pathophysiology at presentation. Our study and other recent reports23 24 26 support the hypothesis that the stroke cohort that is eligible for mechanical thrombectomy is indeed a diverse population, and specific imaging analysis during triage may help to select patients who are most likely to benefit from this therapy. Future studies must be done to define and validate the constellation of specific clinical and imaging criteria at presentation that will identify this elusive patient cohort.


We want to express our sincere thanks to Penumbra Inc. for providing the database from the Penumbra Pivotal trial for our analysis. Special thanks are also extended to Drs Siu Po Sit and Arani Bose of Penumbra Inc. as well as to Ms Dawn Ueda, for their assistance in the preparation of this manuscript.



  • Competing interests None.

  • Ethics approval The study conformed to the national and institutional ethical requirements for the conduct of such a study at each participating center. Where appropriate, it was the responsibility of each Principal Investigator to obtain approval of the study protocol from the Institutional Review Board (IRB)/Ethics Committee (EC) and to keep the IRB/EC informed of trial progress.

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