AHA Scientific Sessions 2013

Two new studies may cast some doubt upon the now widely held belief that core cooling improves outcomes in unconscious cardiac arrest patients.

The studies, presented at the American Heart Association scientific sessions, found that neither cooling to hypothermic levels, compared with normothermic, nor prehospital, compared with in-hospital, significantly improved mortality or neurologic outcomes in more 2,000 patients. The papers were simultaneously published – one in the New England Journal of Medicine, and the other in JAMA.

Induced hypothermia is now standard of care for unconscious survivors of out-of-hospital cardiac arrests. But some animal models suggest that the earlier cooling begins, the better outcomes result. Dr. Francis Kim and his colleagues, however, found almost identical outcomes in a group of 1,359 patients, whatever the timing of hypothermia induction (JAMA 2013 Nov. 17 [doi:10.1001/jama.2013.282173]).

Dr. Kim of the University of Washington, Seattle, and his coauthors randomized patients to induction in the field and during transport, or upon hospital arrival. The intervention group received an infusion of up to 2L of ice-cold normal saline, 7-10 mg pancuronium, and 1-2 mg diazepam, with a target temperature goal of 34°C.

In the control group, hypothermia induction occurred in the hospital, according to each site’s protocol, with either surface or intravascular regimens.

Patients were divided into two groups: those with ventricular fibrillation and those without. Patients without VF were older (68 vs. 62 years). Other baseline characteristics – including time from the call to the return of spontaneous circulation, heart rate, and blood pressure – were similar.

The in-field intervention decreased mean core temperature by about 1.2°C in patients with VF and 1.3°C in those without VF. These patients achieved the target core temperature of 34°C more than 1 hour sooner than patients cooled in the hospital.

Survival to discharge was not significantly different between the intervention and control groups (63% vs. 64% of those with VF; 19% vs. 16% of those without VF).

Nor were there significant differences in the neurologic status of full recovery or mild impairment at discharge, the investigators said. Among those with VF, good neurologic status occurred in 57% of the intervention group and 62% of the control group. For those patients without VF, good neurologic outcomes occurred in 14% of the intervention group and 13% of the control group.

There were no significant differences in the proportion of patients who awakened from coma or died without awakening in any of the groups. The median length of stay was within the VF group (9 days for both treatment groups), and among those without VF (about 11 days for each treatment group).

The intervention carried some evidence of increased harm, the authors noted. Significantly more of the intervention patients rearrested during transport (26% vs. 21% of the controls). They also had significantly lower oxygenation, increased pulmonary edema on the first chest x-ray, and greater use of diuretics in the first 12 hours of hospitalization.

"Importantly, the method of pre-hospital hypothermia may have been associated with early harm that could have masked subsequent improvement," the investigators noted. "Rearrest possibly worsened brain ischemia that did not affect early mortality, but manifested as increased risk of death later during the hospitalization."

The second study released at the AHA meeting questioned whether induced hypothermia confers significant benefit over maintaining a near-normothermic temperature. Patients cooled to 33°C had no better outcomes than those whose core temperatures were held at 36°C, Dr. Niklas Nielsen and colleagues reported Nov. 18 (N. Engl. J. Med. 2013 Nov. 17 [doi:10.1056/NEJMoa1310519]).

"Our trial does not provide evidence that targeting a body temperature of 33° C confers any benefit for unconscious patients admitted to the hospital after out-of-hospital cardiac arrest, as compared with targeting a body temperature of 36°C," wrote Dr. Nielsen of Helsingborg Hospital, Sweden, and his colleagues.

The team analyzed outcomes for 939 patients who had been randomized to the two cooling regimens. The study’s primary outcome was all-cause mortality by 180 days after enrollment. The secondary outcomes included a composite of death or poor neurologic outcome at 180 days as evaluated with the Cerebral Performance Category (CPC) and modified Rankin scales.

The intervention consisted of 36 hours of cooling and gradual rewarming. Treatment teams were allowed to use ice packs, ice-cold fluid, and intravascular or surface temperature management. Gradual rewarming to 37°C commenced after 28 hours; this was discontinued or tapered at 36 hours. After this, patients in both groups were treated with fever-control measures to prevent a rebound temperature of above 37.5°C.

The patients were a mean of 64 years old; most of them (about 75%) had VF as the first shocked rhythm. Spontaneous circulation returned a median of 25 minutes after the arrest. All were unconscious when they arrived at the hospital. The mean follow-up was 256 days.

Most (76%) had temperature management with a surface cooling system; the rest were cooled by an intravascular catheter. Three patients in the 33°C group and four in the 36°C group didn’t get the assigned intervention; 16 in the 33°C group were rewarmed sooner than the protocol, at the discretion of their physician.

During the first week of hospitalization, 247 patients (132 in the 33°C group and 115 in the 36°C group) had life support withdrawn. Reasons for withdrawal included brain death, multiorgan failure, and ethical concerns.

At final follow-up, 50% of patients in the 33°C group and 48% in the 36°C group had died – a nonsignificant difference. There were no significant differences in the composite outcome of death or poor neurologic outcome whether measured by the CPC or the modified Rankin scales.

These results were similar in all adjusted analyses, in the intent-to-treat population, and in the per-protocol population.

Adverse events were common, occurring in 93% of the 33°C group and 90% of the 36°C group. The most common was hypokalemia, which was more significantly more common in the 33°C group (19% vs. 13%).

The study differed in one important way from others that have supported the use of induced hypothermia – and this might have had a key impact on the overall finding.

"We did not allow the natural trajectory of temperature evolution in either group," the authors said. "We actively controlled the temperature during the intervention period and aimed to prevent fever during the first 3 days after cardiac arrest."

It’s difficult to compare these results with those of the Hypothermia After Cardiac Arrest Study – the 2006 trial which first found in favor of the practice, the authors said.

Mortality in both groups of the current study was lower than that seen in 2006 – probably because of the recent advances in prehospital and in-hospital critical care management for cardiac arrest patients.

"We did not find any harm with a targeted temperature of 33°C as compared with 36°C," the investigators wrote. "However, it is worth recognizing that for all outcomes, none of the point estimates were in the direction of a benefit for the 33°C group. On the basis of these results, decisions about which temperature to target after out-of-hospital cardiac arrest require careful consideration."

Dr. Kim had no financial disclosures. Several of his coauthors disclosed financial relationships with medical device companies. Dr. Nielsen had nothing to disclose. Five of his coauthors disclosed financial relationships with outside entities; however, none appeared to be related to the study.

[email protected]

Two new studies may cast some doubt upon the now widely held belief that core cooling improves outcomes in unconscious cardiac arrest patients.

The studies, presented at the American Heart Association scientific sessions, found that neither cooling to hypothermic levels, compared with normothermic, nor prehospital, compared with in-hospital, significantly improved mortality or neurologic outcomes in more 2,000 patients. The papers were simultaneously published – one in the New England Journal of Medicine, and the other in JAMA.

Induced hypothermia is now standard of care for unconscious survivors of out-of-hospital cardiac arrests. But some animal models suggest that the earlier cooling begins, the better outcomes result. Dr. Francis Kim and his colleagues, however, found almost identical outcomes in a group of 1,359 patients, whatever the timing of hypothermia induction (JAMA 2013 Nov. 17 [doi:10.1001/jama.2013.282173]).

Dr. Kim of the University of Washington, Seattle, and his coauthors randomized patients to induction in the field and during transport, or upon hospital arrival. The intervention group received an infusion of up to 2L of ice-cold normal saline, 7-10 mg pancuronium, and 1-2 mg diazepam, with a target temperature goal of 34°C.

In the control group, hypothermia induction occurred in the hospital, according to each site’s protocol, with either surface or intravascular regimens.

Patients were divided into two groups: those with ventricular fibrillation and those without. Patients without VF were older (68 vs. 62 years). Other baseline characteristics – including time from the call to the return of spontaneous circulation, heart rate, and blood pressure – were similar.

The in-field intervention decreased mean core temperature by about 1.2°C in patients with VF and 1.3°C in those without VF. These patients achieved the target core temperature of 34°C more than 1 hour sooner than patients cooled in the hospital.

Survival to discharge was not significantly different between the intervention and control groups (63% vs. 64% of those with VF; 19% vs. 16% of those without VF).

Nor were there significant differences in the neurologic status of full recovery or mild impairment at discharge, the investigators said. Among those with VF, good neurologic status occurred in 57% of the intervention group and 62% of the control group. For those patients without VF, good neurologic outcomes occurred in 14% of the intervention group and 13% of the control group.

There were no significant differences in the proportion of patients who awakened from coma or died without awakening in any of the groups. The median length of stay was within the VF group (9 days for both treatment groups), and among those without VF (about 11 days for each treatment group).

The intervention carried some evidence of increased harm, the authors noted. Significantly more of the intervention patients rearrested during transport (26% vs. 21% of the controls). They also had significantly lower oxygenation, increased pulmonary edema on the first chest x-ray, and greater use of diuretics in the first 12 hours of hospitalization.

"Importantly, the method of pre-hospital hypothermia may have been associated with early harm that could have masked subsequent improvement," the investigators noted. "Rearrest possibly worsened brain ischemia that did not affect early mortality, but manifested as increased risk of death later during the hospitalization."

The second study released at the AHA meeting questioned whether induced hypothermia confers significant benefit over maintaining a near-normothermic temperature. Patients cooled to 33°C had no better outcomes than those whose core temperatures were held at 36°C, Dr. Niklas Nielsen and colleagues reported Nov. 18 (N. Engl. J. Med. 2013 Nov. 17 [doi:10.1056/NEJMoa1310519]).

"Our trial does not provide evidence that targeting a body temperature of 33° C confers any benefit for unconscious patients admitted to the hospital after out-of-hospital cardiac arrest, as compared with targeting a body temperature of 36°C," wrote Dr. Nielsen of Helsingborg Hospital, Sweden, and his colleagues.

The team analyzed outcomes for 939 patients who had been randomized to the two cooling regimens. The study’s primary outcome was all-cause mortality by 180 days after enrollment. The secondary outcomes included a composite of death or poor neurologic outcome at 180 days as evaluated with the Cerebral Performance Category (CPC) and modified Rankin scales.

The intervention consisted of 36 hours of cooling and gradual rewarming. Treatment teams were allowed to use ice packs, ice-cold fluid, and intravascular or surface temperature management. Gradual rewarming to 37°C commenced after 28 hours; this was discontinued or tapered at 36 hours. After this, patients in both groups were treated with fever-control measures to prevent a rebound temperature of above 37.5°C.

The patients were a mean of 64 years old; most of them (about 75%) had VF as the first shocked rhythm. Spontaneous circulation returned a median of 25 minutes after the arrest. All were unconscious when they arrived at the hospital. The mean follow-up was 256 days.

Most (76%) had temperature management with a surface cooling system; the rest were cooled by an intravascular catheter. Three patients in the 33°C group and four in the 36°C group didn’t get the assigned intervention; 16 in the 33°C group were rewarmed sooner than the protocol, at the discretion of their physician.

During the first week of hospitalization, 247 patients (132 in the 33°C group and 115 in the 36°C group) had life support withdrawn. Reasons for withdrawal included brain death, multiorgan failure, and ethical concerns.

At final follow-up, 50% of patients in the 33°C group and 48% in the 36°C group had died – a nonsignificant difference. There were no significant differences in the composite outcome of death or poor neurologic outcome whether measured by the CPC or the modified Rankin scales.

These results were similar in all adjusted analyses, in the intent-to-treat population, and in the per-protocol population.

Adverse events were common, occurring in 93% of the 33°C group and 90% of the 36°C group. The most common was hypokalemia, which was more significantly more common in the 33°C group (19% vs. 13%).

The study differed in one important way from others that have supported the use of induced hypothermia – and this might have had a key impact on the overall finding.

"We did not allow the natural trajectory of temperature evolution in either group," the authors said. "We actively controlled the temperature during the intervention period and aimed to prevent fever during the first 3 days after cardiac arrest."

It’s difficult to compare these results with those of the Hypothermia After Cardiac Arrest Study – the 2006 trial which first found in favor of the practice, the authors said.

Mortality in both groups of the current study was lower than that seen in 2006 – probably because of the recent advances in prehospital and in-hospital critical care management for cardiac arrest patients.

"We did not find any harm with a targeted temperature of 33°C as compared with 36°C," the investigators wrote. "However, it is worth recognizing that for all outcomes, none of the point estimates were in the direction of a benefit for the 33°C group. On the basis of these results, decisions about which temperature to target after out-of-hospital cardiac arrest require careful consideration."

Dr. Kim had no financial disclosures. Several of his coauthors disclosed financial relationships with medical device companies. Dr. Nielsen had nothing to disclose. Five of his coauthors disclosed financial relationships with outside entities; however, none appeared to be related to the study.

[email protected]

Two new studies may cast some doubt upon the now widely held belief that core cooling improves outcomes in unconscious cardiac arrest patients.

The studies, presented at the American Heart Association scientific sessions, found that neither cooling to hypothermic levels, compared with normothermic, nor prehospital, compared with in-hospital, significantly improved mortality or neurologic outcomes in more 2,000 patients. The papers were simultaneously published – one in the New England Journal of Medicine, and the other in JAMA.

Induced hypothermia is now standard of care for unconscious survivors of out-of-hospital cardiac arrests. But some animal models suggest that the earlier cooling begins, the better outcomes result. Dr. Francis Kim and his colleagues, however, found almost identical outcomes in a group of 1,359 patients, whatever the timing of hypothermia induction (JAMA 2013 Nov. 17 [doi:10.1001/jama.2013.282173]).

Dr. Kim of the University of Washington, Seattle, and his coauthors randomized patients to induction in the field and during transport, or upon hospital arrival. The intervention group received an infusion of up to 2L of ice-cold normal saline, 7-10 mg pancuronium, and 1-2 mg diazepam, with a target temperature goal of 34°C.

In the control group, hypothermia induction occurred in the hospital, according to each site’s protocol, with either surface or intravascular regimens.

Patients were divided into two groups: those with ventricular fibrillation and those without. Patients without VF were older (68 vs. 62 years). Other baseline characteristics – including time from the call to the return of spontaneous circulation, heart rate, and blood pressure – were similar.

The in-field intervention decreased mean core temperature by about 1.2°C in patients with VF and 1.3°C in those without VF. These patients achieved the target core temperature of 34°C more than 1 hour sooner than patients cooled in the hospital.

Survival to discharge was not significantly different between the intervention and control groups (63% vs. 64% of those with VF; 19% vs. 16% of those without VF).

Nor were there significant differences in the neurologic status of full recovery or mild impairment at discharge, the investigators said. Among those with VF, good neurologic status occurred in 57% of the intervention group and 62% of the control group. For those patients without VF, good neurologic outcomes occurred in 14% of the intervention group and 13% of the control group.

There were no significant differences in the proportion of patients who awakened from coma or died without awakening in any of the groups. The median length of stay was within the VF group (9 days for both treatment groups), and among those without VF (about 11 days for each treatment group).

The intervention carried some evidence of increased harm, the authors noted. Significantly more of the intervention patients rearrested during transport (26% vs. 21% of the controls). They also had significantly lower oxygenation, increased pulmonary edema on the first chest x-ray, and greater use of diuretics in the first 12 hours of hospitalization.

"Importantly, the method of pre-hospital hypothermia may have been associated with early harm that could have masked subsequent improvement," the investigators noted. "Rearrest possibly worsened brain ischemia that did not affect early mortality, but manifested as increased risk of death later during the hospitalization."

The second study released at the AHA meeting questioned whether induced hypothermia confers significant benefit over maintaining a near-normothermic temperature. Patients cooled to 33°C had no better outcomes than those whose core temperatures were held at 36°C, Dr. Niklas Nielsen and colleagues reported Nov. 18 (N. Engl. J. Med. 2013 Nov. 17 [doi:10.1056/NEJMoa1310519]).

"Our trial does not provide evidence that targeting a body temperature of 33° C confers any benefit for unconscious patients admitted to the hospital after out-of-hospital cardiac arrest, as compared with targeting a body temperature of 36°C," wrote Dr. Nielsen of Helsingborg Hospital, Sweden, and his colleagues.

The team analyzed outcomes for 939 patients who had been randomized to the two cooling regimens. The study’s primary outcome was all-cause mortality by 180 days after enrollment. The secondary outcomes included a composite of death or poor neurologic outcome at 180 days as evaluated with the Cerebral Performance Category (CPC) and modified Rankin scales.

The intervention consisted of 36 hours of cooling and gradual rewarming. Treatment teams were allowed to use ice packs, ice-cold fluid, and intravascular or surface temperature management. Gradual rewarming to 37°C commenced after 28 hours; this was discontinued or tapered at 36 hours. After this, patients in both groups were treated with fever-control measures to prevent a rebound temperature of above 37.5°C.

The patients were a mean of 64 years old; most of them (about 75%) had VF as the first shocked rhythm. Spontaneous circulation returned a median of 25 minutes after the arrest. All were unconscious when they arrived at the hospital. The mean follow-up was 256 days.

Most (76%) had temperature management with a surface cooling system; the rest were cooled by an intravascular catheter. Three patients in the 33°C group and four in the 36°C group didn’t get the assigned intervention; 16 in the 33°C group were rewarmed sooner than the protocol, at the discretion of their physician.

During the first week of hospitalization, 247 patients (132 in the 33°C group and 115 in the 36°C group) had life support withdrawn. Reasons for withdrawal included brain death, multiorgan failure, and ethical concerns.

At final follow-up, 50% of patients in the 33°C group and 48% in the 36°C group had died – a nonsignificant difference. There were no significant differences in the composite outcome of death or poor neurologic outcome whether measured by the CPC or the modified Rankin scales.

These results were similar in all adjusted analyses, in the intent-to-treat population, and in the per-protocol population.

Adverse events were common, occurring in 93% of the 33°C group and 90% of the 36°C group. The most common was hypokalemia, which was more significantly more common in the 33°C group (19% vs. 13%).

The study differed in one important way from others that have supported the use of induced hypothermia – and this might have had a key impact on the overall finding.

"We did not allow the natural trajectory of temperature evolution in either group," the authors said. "We actively controlled the temperature during the intervention period and aimed to prevent fever during the first 3 days after cardiac arrest."

It’s difficult to compare these results with those of the Hypothermia After Cardiac Arrest Study – the 2006 trial which first found in favor of the practice, the authors said.

Mortality in both groups of the current study was lower than that seen in 2006 – probably because of the recent advances in prehospital and in-hospital critical care management for cardiac arrest patients.

"We did not find any harm with a targeted temperature of 33°C as compared with 36°C," the investigators wrote. "However, it is worth recognizing that for all outcomes, none of the point estimates were in the direction of a benefit for the 33°C group. On the basis of these results, decisions about which temperature to target after out-of-hospital cardiac arrest require careful consideration."

Dr. Kim had no financial disclosures. Several of his coauthors disclosed financial relationships with medical device companies. Dr. Nielsen had nothing to disclose. Five of his coauthors disclosed financial relationships with outside entities; however, none appeared to be related to the study.

[email protected]

Postmarket outcomes with the first commercially available U.S. transcatheter aortic valve replacement device are comparable with randomized trial and international registry experience, national registry results suggest.

The Edwards LifeSciences Sapien XT valve was successfully implanted in 92% of 7,710 TAVR patients, two-thirds with a transfemoral approach.

The primary outcomes of in-hospital all-cause mortality and stroke were 5.5% and 2%, respectively, Dr. Michael J. Mack reported at the American Heart Association scientific sessions.

Conversion to open-heart surgery was needed in only 1% of patients, but associated with a 49% incidence of in-hospital mortality.

Major bleeding and vascular complications were relatively low, at 3.5% and 6.4%, respectively, despite the overall lack of experience and use of larger sheath delivery systems. Acute renal insufficiency occurred in 5.5% of patients.

The 30-day death and stroke rates were 7.6% and 2.8% among 3,113 patients with sufficient follow-up. Most deaths were due to noncardiovascular causes, he noted.

This is the first public report from the national Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy (TVT) Registry, and comes amid calls for increased postmarket surveillance of high-risk medical devices.

The results also help address whether outcomes after a controlled commercial rollout of the first-generation valve hold up to those from clinical trials and the global TAVR experience, now based on second- and third-generation devices, according to Dr. Mack, past STS president and director of cardiovascular research and medicine at the Heart Hospital Baylor Plano in Plano, Texas.

The Food and Drug Administration initially approved the Edwards valve for severe, symptomatic aortic stenosis in inoperable patients in 2011 and expanded its label in September 2012 to include operable, high-risk patients using either a transfemoral or transapical approach.

The TVT analysis comprised 1,559 inoperable and 6,151 high-risk, operable patients who underwent TAVR at 224 participating registry hospitals from November 2011 to May 2013. Their median STS predicted risk of operative mortality (PROM) was 7%, but varied widely from 1.2% to 17.4%. Their median age was 84 years.

The risk profile of patients was generally lower in the registry than in randomized trials and European reports, and may have contributed to the good outcomes, Dr. Mack acknowledged. The mean PROM was 11.2% among inoperable patients and 11.8% among high-risk, operable patients in the pivotal PARTNER trial, which was used for the device’s approval.

Although incomplete data may have been entered into the risk calculator, the other plausible explanation is "risk creep," or that patients with lower surgical risk are being treated with the less-invasive procedure, leading to better outcomes, he noted.

Other possible contributors to the registry outcomes include a shorter U.S. learning curve due to international collaboration, inclusion of at least 35 highly experienced PARTNER study centers in the registry, and mandated training by the manufacturer including a company employed clinical specialist during all procedures.

Still, the results stack up well. Germany, which has led the TAVR revolution, had an in-hospital mortality of 5.1% with a transfemoral approach and 7.7% with a transapical approach in 2011 in the German Aortic Valve Registry.

In the PARTNER trial, 30-day mortality was 5% among inoperable patients, 3.7% among high-risk, operable patients by a transfemoral approach, and 8.7% among high-risk, operable patients by a transapical approach.

This compares with 30-day mortality rates of 6.1%, 4.6%, and 9.8%, respectively, in the TVT registry, Dr. Mack reported.

The TVT registry captured an estimated 88% of TAVR procedures performed between May 2012, when the Centers for Medicare and Medicaid Services issued its national coverage determination, and the end of the study. Notably, 30-day outcomes were reported from only 114 hospitals or 51% of the 244 hospitals in the registry. The results were published simultaneously with Dr. Mack’s presentation (JAMA 2013;310:2069-77. doi:10.1001/jama/2013.282043).

More complete long-term outcomes at 30 days and 1 year are needed to have a more comprehensive assessment of TAVR outcomes in the United States, Dr. Mack cautioned. Patient health status and quality of life outcomes will also be addressed in subsequent reports.

So far, it’s known that most (63%) patients were discharged home, 6% experienced new-onset atrial fibrillation, and 6.6% needed a new permanent pacemaker or implantable cardioverter-defibrillator.

The study was funded by the American College of Cardiology Foundation and the Society of Thoracic Surgeons. Dr. Mack reported no other disclosures; a co-author reported travel support from the STS and a second reported consulting for Janssen, Boehringer Ingelheim, and Sanofi.

[email protected]

Postmarket outcomes with the first commercially available U.S. transcatheter aortic valve replacement device are comparable with randomized trial and international registry experience, national registry results suggest.

The Edwards LifeSciences Sapien XT valve was successfully implanted in 92% of 7,710 TAVR patients, two-thirds with a transfemoral approach.

The primary outcomes of in-hospital all-cause mortality and stroke were 5.5% and 2%, respectively, Dr. Michael J. Mack reported at the American Heart Association scientific sessions.

Conversion to open-heart surgery was needed in only 1% of patients, but associated with a 49% incidence of in-hospital mortality.

Major bleeding and vascular complications were relatively low, at 3.5% and 6.4%, respectively, despite the overall lack of experience and use of larger sheath delivery systems. Acute renal insufficiency occurred in 5.5% of patients.

The 30-day death and stroke rates were 7.6% and 2.8% among 3,113 patients with sufficient follow-up. Most deaths were due to noncardiovascular causes, he noted.

This is the first public report from the national Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy (TVT) Registry, and comes amid calls for increased postmarket surveillance of high-risk medical devices.

The results also help address whether outcomes after a controlled commercial rollout of the first-generation valve hold up to those from clinical trials and the global TAVR experience, now based on second- and third-generation devices, according to Dr. Mack, past STS president and director of cardiovascular research and medicine at the Heart Hospital Baylor Plano in Plano, Texas.

The Food and Drug Administration initially approved the Edwards valve for severe, symptomatic aortic stenosis in inoperable patients in 2011 and expanded its label in September 2012 to include operable, high-risk patients using either a transfemoral or transapical approach.

The TVT analysis comprised 1,559 inoperable and 6,151 high-risk, operable patients who underwent TAVR at 224 participating registry hospitals from November 2011 to May 2013. Their median STS predicted risk of operative mortality (PROM) was 7%, but varied widely from 1.2% to 17.4%. Their median age was 84 years.

The risk profile of patients was generally lower in the registry than in randomized trials and European reports, and may have contributed to the good outcomes, Dr. Mack acknowledged. The mean PROM was 11.2% among inoperable patients and 11.8% among high-risk, operable patients in the pivotal PARTNER trial, which was used for the device’s approval.

Although incomplete data may have been entered into the risk calculator, the other plausible explanation is "risk creep," or that patients with lower surgical risk are being treated with the less-invasive procedure, leading to better outcomes, he noted.

Other possible contributors to the registry outcomes include a shorter U.S. learning curve due to international collaboration, inclusion of at least 35 highly experienced PARTNER study centers in the registry, and mandated training by the manufacturer including a company employed clinical specialist during all procedures.

Still, the results stack up well. Germany, which has led the TAVR revolution, had an in-hospital mortality of 5.1% with a transfemoral approach and 7.7% with a transapical approach in 2011 in the German Aortic Valve Registry.

In the PARTNER trial, 30-day mortality was 5% among inoperable patients, 3.7% among high-risk, operable patients by a transfemoral approach, and 8.7% among high-risk, operable patients by a transapical approach.

This compares with 30-day mortality rates of 6.1%, 4.6%, and 9.8%, respectively, in the TVT registry, Dr. Mack reported.

The TVT registry captured an estimated 88% of TAVR procedures performed between May 2012, when the Centers for Medicare and Medicaid Services issued its national coverage determination, and the end of the study. Notably, 30-day outcomes were reported from only 114 hospitals or 51% of the 244 hospitals in the registry. The results were published simultaneously with Dr. Mack’s presentation (JAMA 2013;310:2069-77. doi:10.1001/jama/2013.282043).

More complete long-term outcomes at 30 days and 1 year are needed to have a more comprehensive assessment of TAVR outcomes in the United States, Dr. Mack cautioned. Patient health status and quality of life outcomes will also be addressed in subsequent reports.

So far, it’s known that most (63%) patients were discharged home, 6% experienced new-onset atrial fibrillation, and 6.6% needed a new permanent pacemaker or implantable cardioverter-defibrillator.

The study was funded by the American College of Cardiology Foundation and the Society of Thoracic Surgeons. Dr. Mack reported no other disclosures; a co-author reported travel support from the STS and a second reported consulting for Janssen, Boehringer Ingelheim, and Sanofi.

[email protected]

Postmarket outcomes with the first commercially available U.S. transcatheter aortic valve replacement device are comparable with randomized trial and international registry experience, national registry results suggest.

The Edwards LifeSciences Sapien XT valve was successfully implanted in 92% of 7,710 TAVR patients, two-thirds with a transfemoral approach.

The primary outcomes of in-hospital all-cause mortality and stroke were 5.5% and 2%, respectively, Dr. Michael J. Mack reported at the American Heart Association scientific sessions.

Conversion to open-heart surgery was needed in only 1% of patients, but associated with a 49% incidence of in-hospital mortality.

Major bleeding and vascular complications were relatively low, at 3.5% and 6.4%, respectively, despite the overall lack of experience and use of larger sheath delivery systems. Acute renal insufficiency occurred in 5.5% of patients.

The 30-day death and stroke rates were 7.6% and 2.8% among 3,113 patients with sufficient follow-up. Most deaths were due to noncardiovascular causes, he noted.

This is the first public report from the national Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy (TVT) Registry, and comes amid calls for increased postmarket surveillance of high-risk medical devices.

The results also help address whether outcomes after a controlled commercial rollout of the first-generation valve hold up to those from clinical trials and the global TAVR experience, now based on second- and third-generation devices, according to Dr. Mack, past STS president and director of cardiovascular research and medicine at the Heart Hospital Baylor Plano in Plano, Texas.

The Food and Drug Administration initially approved the Edwards valve for severe, symptomatic aortic stenosis in inoperable patients in 2011 and expanded its label in September 2012 to include operable, high-risk patients using either a transfemoral or transapical approach.

The TVT analysis comprised 1,559 inoperable and 6,151 high-risk, operable patients who underwent TAVR at 224 participating registry hospitals from November 2011 to May 2013. Their median STS predicted risk of operative mortality (PROM) was 7%, but varied widely from 1.2% to 17.4%. Their median age was 84 years.

The risk profile of patients was generally lower in the registry than in randomized trials and European reports, and may have contributed to the good outcomes, Dr. Mack acknowledged. The mean PROM was 11.2% among inoperable patients and 11.8% among high-risk, operable patients in the pivotal PARTNER trial, which was used for the device’s approval.

Although incomplete data may have been entered into the risk calculator, the other plausible explanation is "risk creep," or that patients with lower surgical risk are being treated with the less-invasive procedure, leading to better outcomes, he noted.

Other possible contributors to the registry outcomes include a shorter U.S. learning curve due to international collaboration, inclusion of at least 35 highly experienced PARTNER study centers in the registry, and mandated training by the manufacturer including a company employed clinical specialist during all procedures.

Still, the results stack up well. Germany, which has led the TAVR revolution, had an in-hospital mortality of 5.1% with a transfemoral approach and 7.7% with a transapical approach in 2011 in the German Aortic Valve Registry.

In the PARTNER trial, 30-day mortality was 5% among inoperable patients, 3.7% among high-risk, operable patients by a transfemoral approach, and 8.7% among high-risk, operable patients by a transapical approach.

This compares with 30-day mortality rates of 6.1%, 4.6%, and 9.8%, respectively, in the TVT registry, Dr. Mack reported.

The TVT registry captured an estimated 88% of TAVR procedures performed between May 2012, when the Centers for Medicare and Medicaid Services issued its national coverage determination, and the end of the study. Notably, 30-day outcomes were reported from only 114 hospitals or 51% of the 244 hospitals in the registry. The results were published simultaneously with Dr. Mack’s presentation (JAMA 2013;310:2069-77. doi:10.1001/jama/2013.282043).

More complete long-term outcomes at 30 days and 1 year are needed to have a more comprehensive assessment of TAVR outcomes in the United States, Dr. Mack cautioned. Patient health status and quality of life outcomes will also be addressed in subsequent reports.

So far, it’s known that most (63%) patients were discharged home, 6% experienced new-onset atrial fibrillation, and 6.6% needed a new permanent pacemaker or implantable cardioverter-defibrillator.

The study was funded by the American College of Cardiology Foundation and the Society of Thoracic Surgeons. Dr. Mack reported no other disclosures; a co-author reported travel support from the STS and a second reported consulting for Janssen, Boehringer Ingelheim, and Sanofi.

[email protected]