Thrombectomy fails to improve PCI outcomes, ups stroke risk

SAN DIEGO – Routine manual thrombectomy before percutaneous coronary intervention did not improve 180-day outcomes and was linked with an increased risk of stroke in patients with acute ST-segment elevation MI in the TOTAL trial.

Routine thrombectomy had no effect on the primary outcome of cardiovascular death, MI, cardiogenic shock, or New York Heart Association class IV heart failure, occurring in 6.9% of thrombectomy patients and 7% of PCI-only patients.

However, the study’s primary safety endpoint of stroke at 30 days doubled in patients undergoing routine thrombectomy before PCI to 33 events (0.7%), compared with those who had PCI with only bailout thrombectomy (16 events [0.3%]; P = .015).

The same pattern was observed with stroke or transient ischemic attack within 30 days (42 vs. 19 events; hazard ratio, 2.21; P = .003) and continued for stroke within 180 days (52 vs. 25 events; HR, 2.08; P = .002).

“The stroke findings are unexpected and we believe require confirmation in other datasets. A detailed case-by-case review is underway to help us understand the etiology and the relationship with the procedure,” lead author Dr. Sanjit S. Jolly said at the annual meeting of the American College of Cardiology.

Enthusiasm for manual thrombus aspiration was sparked by a survival benefit observed in the single-center, prospective TAPAS trial in ST-segment elevation MI patients (STEMI), and the procedure was widely adopted.

The more recent, multicenter TASTE trial, however, reported that routine thrombectomy before PCI failed to significantly reduce 30-day mortality in 7,244 STEMI patients, though there were trends toward reductions in stent thrombosis and hospitalization for recurrent MI.

TOTAL (Manual Aspiration Thrombectomy Plus PCI vs. PCI Alone in STEMI) randomly assigned 10,063 patients within 12 hours of STEMI symptoms to primary PCI either with upfront manual thrombectomy or only bailout thrombectomy if the PCI strategy failed.

The lack of significant differences between groups in the primary outcome was also true in all the components of the primary outcome. Furthermore, there was no effect on the primary outcome based on thrombotic burden, a question that remained unanswered after TASTE, Dr. Jolly reported. The TOTAL results were published online simultaneously with his report (N. Engl. J. Med. 2015 March 16 [doi:10.1056/NEJMoa1415098]).

“TOTAL and TASTE emphasize the need to conduct large randomized trials of common interventions, even when small trials appear positive,” Dr. Jolly said.

Discussant Dr. Steven Nissen, chair of cardiovascular medicine at Cleveland Clinic, described the routine use of thrombectomy as “a sad story about device regulation in the United States” in that the evidence level needed to get a medical device on the market is so far below that required for drug approval that patients undergo procedures without good randomized trial evidence to show they even work.

“We dodged a bullet recently with renal denervation when everyone thought it would work, and when you finally tested it, it didn’t,” Dr. Nissen said. “Let this be a lesson to us: We need to have more rigorous studies of medical devices before they get to market and get used in very large numbers of people.”

Currently, aspiration thrombectomy carries a IIa recommendation for use with PCI in the most recent ACC/American Heart Association guidelines for the management of patients with STEMI (J. Am. Coll. Cardiol. 2009;54:2205-41).

When asked whether the guidelines should change based on the TOTAL and TASTE results, Dr. Jolly said there should be a clear recommendation that routine thrombus aspiration should not be the appropriate approach, while the issue of bailout aspiration may be left to clinician judgment.

The finding of late strokes is difficult to understand and should be interpreted with caution because of the small number of strokes occurring between 30 and 180 days, he said. Detailed analysis of all strokes will be presented at a later meeting, but Rankin Scale scores show several strokes were “very debilitating.” There is a consistency in the data, as a meta-analysis of smaller trials also identified an increased stroke risk with adjunctive thrombectomy.

Discussant Dr. Gregg W. Stone, director of cardiovascular research and education at Columbia University Medical Center in New York, said a mechanism for periprocedural stroke with aspiration can be envisioned, but that understanding the risk of ongoing, late stroke is more difficult.

As for why thrombectomy didn’t work, “aspiration is incredibly inefficient, thromboemboli still occur before, during, and after aspiration, the timing of aspiration is often too late to benefit most patients,” and other mechanisms of myonecrosis may predominate, such as reperfusion injury, he observed.

Dr. Stone said the TOTAL results should change practice and that the guideline recommendation should be downgraded to IIb.

“There are some patients who have a very large thrombus burden who have trouble dealing with all that thrombus in the cath lab who might benefit, and it is impossible to design randomized trials for small sections and groups of patients,” he said. “I wouldn’t make it class III by any means, but I think it’ll take a long time for that reduction in use to actually transmit through clinical practice, because I must say interventional cardiologists love the idea of simply removing thrombus with a relatively easy-to-use device.”

Dr. David Kandzari, director of interventional cardiology at the Piedmont Heart Center in Atlanta, said in an interview that TOTAL will make operators much more selective and cautious in their performance of thrombectomy until further insights into the stroke issue are available. Thrombectomy should be reserved for bailout instances and not as a front-line therapy, he said.

On the other hand, the stroke rate in the early phase was not significantly different between groups in an as-treated analysis, and an opportunity exists to investigate potential differences between stroke and nonstroke patients to determine whether other comorbidities rather than thrombectomy per se may account for the stroke signal, Dr. Kandzari observed.

TOTAL was funded by the Canadian Institutes of Health Research, Canadian Network and Centre for Trials Internationally, and Medtronic. Dr. Jolly disclosed receiving consulting fees and honoraria from AstraZeneca, speaking fees for St. Jude, and research grants from Medtronic. Dr. Nissen has received research support from and is a consultant/adviser to numerous pharmaceutical companies; all honoraria or consulting fees go directly to charity so that he receives neither income nor a tax deduction. Dr. Stone reported consulting honoraria from Guided Delivery Systems, Miracor, and Reva, and ownership interest or partnership in Arstasis, Caliber, VNT, Micardia, Biostar family funds, and Medfocus family funds. Dr. Kandzari reported research and grant support from Medtronic, Biotronic, Abbott Vascular, and Boston Scientific.

pwendling@frontlinemedcom.com

SAN DIEGO – Routine manual thrombectomy before percutaneous coronary intervention did not improve 180-day outcomes and was linked with an increased risk of stroke in patients with acute ST-segment elevation MI in the TOTAL trial.

Routine thrombectomy had no effect on the primary outcome of cardiovascular death, MI, cardiogenic shock, or New York Heart Association class IV heart failure, occurring in 6.9% of thrombectomy patients and 7% of PCI-only patients.

However, the study’s primary safety endpoint of stroke at 30 days doubled in patients undergoing routine thrombectomy before PCI to 33 events (0.7%), compared with those who had PCI with only bailout thrombectomy (16 events [0.3%]; P = .015).

The same pattern was observed with stroke or transient ischemic attack within 30 days (42 vs. 19 events; hazard ratio, 2.21; P = .003) and continued for stroke within 180 days (52 vs. 25 events; HR, 2.08; P = .002).

“The stroke findings are unexpected and we believe require confirmation in other datasets. A detailed case-by-case review is underway to help us understand the etiology and the relationship with the procedure,” lead author Dr. Sanjit S. Jolly said at the annual meeting of the American College of Cardiology.

Enthusiasm for manual thrombus aspiration was sparked by a survival benefit observed in the single-center, prospective TAPAS trial in ST-segment elevation MI patients (STEMI), and the procedure was widely adopted.

The more recent, multicenter TASTE trial, however, reported that routine thrombectomy before PCI failed to significantly reduce 30-day mortality in 7,244 STEMI patients, though there were trends toward reductions in stent thrombosis and hospitalization for recurrent MI.

TOTAL (Manual Aspiration Thrombectomy Plus PCI vs. PCI Alone in STEMI) randomly assigned 10,063 patients within 12 hours of STEMI symptoms to primary PCI either with upfront manual thrombectomy or only bailout thrombectomy if the PCI strategy failed.

The lack of significant differences between groups in the primary outcome was also true in all the components of the primary outcome. Furthermore, there was no effect on the primary outcome based on thrombotic burden, a question that remained unanswered after TASTE, Dr. Jolly reported. The TOTAL results were published online simultaneously with his report (N. Engl. J. Med. 2015 March 16 [doi:10.1056/NEJMoa1415098]).

“TOTAL and TASTE emphasize the need to conduct large randomized trials of common interventions, even when small trials appear positive,” Dr. Jolly said.

Discussant Dr. Steven Nissen, chair of cardiovascular medicine at Cleveland Clinic, described the routine use of thrombectomy as “a sad story about device regulation in the United States” in that the evidence level needed to get a medical device on the market is so far below that required for drug approval that patients undergo procedures without good randomized trial evidence to show they even work.

“We dodged a bullet recently with renal denervation when everyone thought it would work, and when you finally tested it, it didn’t,” Dr. Nissen said. “Let this be a lesson to us: We need to have more rigorous studies of medical devices before they get to market and get used in very large numbers of people.”

Currently, aspiration thrombectomy carries a IIa recommendation for use with PCI in the most recent ACC/American Heart Association guidelines for the management of patients with STEMI (J. Am. Coll. Cardiol. 2009;54:2205-41).

When asked whether the guidelines should change based on the TOTAL and TASTE results, Dr. Jolly said there should be a clear recommendation that routine thrombus aspiration should not be the appropriate approach, while the issue of bailout aspiration may be left to clinician judgment.

The finding of late strokes is difficult to understand and should be interpreted with caution because of the small number of strokes occurring between 30 and 180 days, he said. Detailed analysis of all strokes will be presented at a later meeting, but Rankin Scale scores show several strokes were “very debilitating.” There is a consistency in the data, as a meta-analysis of smaller trials also identified an increased stroke risk with adjunctive thrombectomy.

Discussant Dr. Gregg W. Stone, director of cardiovascular research and education at Columbia University Medical Center in New York, said a mechanism for periprocedural stroke with aspiration can be envisioned, but that understanding the risk of ongoing, late stroke is more difficult.

As for why thrombectomy didn’t work, “aspiration is incredibly inefficient, thromboemboli still occur before, during, and after aspiration, the timing of aspiration is often too late to benefit most patients,” and other mechanisms of myonecrosis may predominate, such as reperfusion injury, he observed.

Dr. Stone said the TOTAL results should change practice and that the guideline recommendation should be downgraded to IIb.

“There are some patients who have a very large thrombus burden who have trouble dealing with all that thrombus in the cath lab who might benefit, and it is impossible to design randomized trials for small sections and groups of patients,” he said. “I wouldn’t make it class III by any means, but I think it’ll take a long time for that reduction in use to actually transmit through clinical practice, because I must say interventional cardiologists love the idea of simply removing thrombus with a relatively easy-to-use device.”

Dr. David Kandzari, director of interventional cardiology at the Piedmont Heart Center in Atlanta, said in an interview that TOTAL will make operators much more selective and cautious in their performance of thrombectomy until further insights into the stroke issue are available. Thrombectomy should be reserved for bailout instances and not as a front-line therapy, he said.

On the other hand, the stroke rate in the early phase was not significantly different between groups in an as-treated analysis, and an opportunity exists to investigate potential differences between stroke and nonstroke patients to determine whether other comorbidities rather than thrombectomy per se may account for the stroke signal, Dr. Kandzari observed.

TOTAL was funded by the Canadian Institutes of Health Research, Canadian Network and Centre for Trials Internationally, and Medtronic. Dr. Jolly disclosed receiving consulting fees and honoraria from AstraZeneca, speaking fees for St. Jude, and research grants from Medtronic. Dr. Nissen has received research support from and is a consultant/adviser to numerous pharmaceutical companies; all honoraria or consulting fees go directly to charity so that he receives neither income nor a tax deduction. Dr. Stone reported consulting honoraria from Guided Delivery Systems, Miracor, and Reva, and ownership interest or partnership in Arstasis, Caliber, VNT, Micardia, Biostar family funds, and Medfocus family funds. Dr. Kandzari reported research and grant support from Medtronic, Biotronic, Abbott Vascular, and Boston Scientific.

pwendling@frontlinemedcom.com

SAN DIEGO – Routine manual thrombectomy before percutaneous coronary intervention did not improve 180-day outcomes and was linked with an increased risk of stroke in patients with acute ST-segment elevation MI in the TOTAL trial.

Routine thrombectomy had no effect on the primary outcome of cardiovascular death, MI, cardiogenic shock, or New York Heart Association class IV heart failure, occurring in 6.9% of thrombectomy patients and 7% of PCI-only patients.

However, the study’s primary safety endpoint of stroke at 30 days doubled in patients undergoing routine thrombectomy before PCI to 33 events (0.7%), compared with those who had PCI with only bailout thrombectomy (16 events [0.3%]; P = .015).

The same pattern was observed with stroke or transient ischemic attack within 30 days (42 vs. 19 events; hazard ratio, 2.21; P = .003) and continued for stroke within 180 days (52 vs. 25 events; HR, 2.08; P = .002).

“The stroke findings are unexpected and we believe require confirmation in other datasets. A detailed case-by-case review is underway to help us understand the etiology and the relationship with the procedure,” lead author Dr. Sanjit S. Jolly said at the annual meeting of the American College of Cardiology.

Enthusiasm for manual thrombus aspiration was sparked by a survival benefit observed in the single-center, prospective TAPAS trial in ST-segment elevation MI patients (STEMI), and the procedure was widely adopted.

The more recent, multicenter TASTE trial, however, reported that routine thrombectomy before PCI failed to significantly reduce 30-day mortality in 7,244 STEMI patients, though there were trends toward reductions in stent thrombosis and hospitalization for recurrent MI.

TOTAL (Manual Aspiration Thrombectomy Plus PCI vs. PCI Alone in STEMI) randomly assigned 10,063 patients within 12 hours of STEMI symptoms to primary PCI either with upfront manual thrombectomy or only bailout thrombectomy if the PCI strategy failed.

The lack of significant differences between groups in the primary outcome was also true in all the components of the primary outcome. Furthermore, there was no effect on the primary outcome based on thrombotic burden, a question that remained unanswered after TASTE, Dr. Jolly reported. The TOTAL results were published online simultaneously with his report (N. Engl. J. Med. 2015 March 16 [doi:10.1056/NEJMoa1415098]).

“TOTAL and TASTE emphasize the need to conduct large randomized trials of common interventions, even when small trials appear positive,” Dr. Jolly said.

Discussant Dr. Steven Nissen, chair of cardiovascular medicine at Cleveland Clinic, described the routine use of thrombectomy as “a sad story about device regulation in the United States” in that the evidence level needed to get a medical device on the market is so far below that required for drug approval that patients undergo procedures without good randomized trial evidence to show they even work.

“We dodged a bullet recently with renal denervation when everyone thought it would work, and when you finally tested it, it didn’t,” Dr. Nissen said. “Let this be a lesson to us: We need to have more rigorous studies of medical devices before they get to market and get used in very large numbers of people.”

Currently, aspiration thrombectomy carries a IIa recommendation for use with PCI in the most recent ACC/American Heart Association guidelines for the management of patients with STEMI (J. Am. Coll. Cardiol. 2009;54:2205-41).

When asked whether the guidelines should change based on the TOTAL and TASTE results, Dr. Jolly said there should be a clear recommendation that routine thrombus aspiration should not be the appropriate approach, while the issue of bailout aspiration may be left to clinician judgment.

The finding of late strokes is difficult to understand and should be interpreted with caution because of the small number of strokes occurring between 30 and 180 days, he said. Detailed analysis of all strokes will be presented at a later meeting, but Rankin Scale scores show several strokes were “very debilitating.” There is a consistency in the data, as a meta-analysis of smaller trials also identified an increased stroke risk with adjunctive thrombectomy.

Discussant Dr. Gregg W. Stone, director of cardiovascular research and education at Columbia University Medical Center in New York, said a mechanism for periprocedural stroke with aspiration can be envisioned, but that understanding the risk of ongoing, late stroke is more difficult.

As for why thrombectomy didn’t work, “aspiration is incredibly inefficient, thromboemboli still occur before, during, and after aspiration, the timing of aspiration is often too late to benefit most patients,” and other mechanisms of myonecrosis may predominate, such as reperfusion injury, he observed.

Dr. Stone said the TOTAL results should change practice and that the guideline recommendation should be downgraded to IIb.

“There are some patients who have a very large thrombus burden who have trouble dealing with all that thrombus in the cath lab who might benefit, and it is impossible to design randomized trials for small sections and groups of patients,” he said. “I wouldn’t make it class III by any means, but I think it’ll take a long time for that reduction in use to actually transmit through clinical practice, because I must say interventional cardiologists love the idea of simply removing thrombus with a relatively easy-to-use device.”

Dr. David Kandzari, director of interventional cardiology at the Piedmont Heart Center in Atlanta, said in an interview that TOTAL will make operators much more selective and cautious in their performance of thrombectomy until further insights into the stroke issue are available. Thrombectomy should be reserved for bailout instances and not as a front-line therapy, he said.

On the other hand, the stroke rate in the early phase was not significantly different between groups in an as-treated analysis, and an opportunity exists to investigate potential differences between stroke and nonstroke patients to determine whether other comorbidities rather than thrombectomy per se may account for the stroke signal, Dr. Kandzari observed.

TOTAL was funded by the Canadian Institutes of Health Research, Canadian Network and Centre for Trials Internationally, and Medtronic. Dr. Jolly disclosed receiving consulting fees and honoraria from AstraZeneca, speaking fees for St. Jude, and research grants from Medtronic. Dr. Nissen has received research support from and is a consultant/adviser to numerous pharmaceutical companies; all honoraria or consulting fees go directly to charity so that he receives neither income nor a tax deduction. Dr. Stone reported consulting honoraria from Guided Delivery Systems, Miracor, and Reva, and ownership interest or partnership in Arstasis, Caliber, VNT, Micardia, Biostar family funds, and Medfocus family funds. Dr. Kandzari reported research and grant support from Medtronic, Biotronic, Abbott Vascular, and Boston Scientific.

pwendling@frontlinemedcom.com