SynCardia Systems - TAH-t Companion 2 Driver System (C2) and Risk of Mortality and Stroke - Letter to Health Care Providers
August 17, 2018
Dear Transplant Surgeons and Cardiologists,
We are writing to inform you that the FDA has reviewed the final results from the post-approval study conducted by SynCardia Systems, LLC. for their Temporary Total Artificial Heart (TAH-t) Companion 2 Driver System (C2 Driver System). These final results indicate a higher mortality rate and higher stroke rate for patients initially supported with the C2 Driver System compared to patients initially supported with the previous generation driver, the Circulatory Support System (CSS) Console. We recommend that you carefully consider these mortality and stroke results from the TAH-t post-approval study when making treatment decisions, and discuss the risks and benefits of the C2 Driver System with patients.
Syncardia’s TAH-t functions as a bridge to a heart transplant in a small population of heart failure patients with severe bi-ventricular failure. The TAH-t replaces a patient's native ventricles and valves to completely take over pumping of blood to both the pulmonary and systemic circulation. The TAH-t is used with an external pneumatic driver which activates the implanted device. The C2 Driver System is one of the available external pneumatic drivers.
The TAH-t was initially approved for use by the FDA in 2004, with the CSS Console as its initial driver system. The FDA approved the smaller C2 Driver System in 2012. Both driver systems were approved for use only in the hospital setting. In addition to the in-hospital driver systems, the FDA approved the portable Freedom Driver System in 2014. The Freedom Driver System can be used outside of the hospital, allowing some TAH-t patients to return home while on the device.
ANALYSIS OF PROBLEM
Background
As a condition of approval for the C2 Driver System, the FDA required SynCardia to conduct a post-approval study assessing postmarket performance. The study used data from the Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) and compared outcomes for patients who were initially supported with the C2 Driver System to patients initially supported with the CSS Console during the same time period (implanted on June 20, 2012 or after). The FDA's Post-Approval Study Database provides the study protocol parameters for the SynCardia post-approval study.
The FDA has previously apprised the health care community about interim results comparing C2 Driver System patients to CSS Console patients. On June 15, 2015, the FDA posted a letter indicating that interim data suggested a higher mortality rate for a subgroup of patients requiring pre-implant circulatory rescue interventions when using the C2 Driver System. On October 26, 2016, the FDA posted an update about the continued increased mortality rate for a subgroup of patients requiring pre-implant circulatory rescue interventions when using the C2 Driver System as well as additional interim study results that indicated a higher risk of neurological adverse events for C2 Driver System patients. On September 25, 2017, an update informed the health care community of a higher rate of cerebrovascular accidents (stroke) for C2 Driver System patients.
The FDA is issuing this current letter to inform health care providers about the final mortality and stroke results from the post-approval study, presented below. The final data summary for other post-approval study results can be found on the SynCardia post-approval study webpage.
Mortality Results
The primary endpoint of the post-approval study was survival, with “survival” defined as the patient having received a transplant, having transferred to the Freedom Driver, or being alive and continuing on the initial driver system. Survival was assessed at both three and six months post-implant. The survival rate for TAH-t patients initially supported with the C2 Driver System was compared to that for patients initially supported with the CSS Console, with the hypothesis being that the C2 Driver was non-inferior in survival rate to the CSS Console by a margin of no more than 10 percentage points.
Survival for patients initially supported with the C2 Driver System was significantly lower than survival for patients initially supported with the CSS Console at three months and six months post-implant (Table 1), and the non-inferiority hypothesis was not met.
All “No Survival” occurrences as listed in Table 1 were due to patient death.
Table 1. Survival Outcome, All patients (December 22, 2017 Final Report)
Through 3 Months post-implant | Through 6 Months post-implant | |||
Companion 2 | CSS Console | Companion 2 | CSS Console | |
No Survival◊ | 69 (34.5%) | 20 (22.5%) | 79 (39.5%) | 23 (25.8%) |
Survival | 131 (65.5%)† | 69 (77.5%) | 120 (60.0%)‡ | 66 (74.2%) |
Total | 200 | 89 | 200* | 89 |
◊ All “No Survival” occurrences were due to patient death.
†Lower bound of the two-sided exact 95% confidence interval is 58.5%, which is more than 10 percentage points below the observed rate for CSS Console. Non-inferiority hypothesis was not met at three months post-implant.
‡Lower bound of the two-sided exact 95% confidence interval is 52.9%, which is more than 10 percentage points below the observed rate for CSS Console. Non-inferiority hypothesis was not met at six months post-implant.
*Including one person who was lost to follow-up (0.5%).
In patients who did not require pre-implant circulatory rescue interventions (such as intra-aortic balloon pump or extracorporeal membrane oxygenation), there was a higher mortality rate at six months post-implant for C2 Driver System patients compared to CSS Console patients, even though at three months post-implant the mortality rates were similar (Table 2). For the subgroup of patients who did require pre-implant circulatory rescue interventions, the final results confirm the previously communicated finding of a higher mortality rate with the C2 Driver System at three months and six months post-implant (Table 3).
Table 2. Survival Outcome, Patients who did not receive pre-implant circulatory rescue interventions (December 22, 2017 Final Report)
Through 3 Months post-implant | Through 6 Months post-implant | |||
Companion 2 | CSS Console | Companion 2 | CSS Console | |
No Survival◊ | 37 (30.1%) | 10 (24.4%) | 45 (36.6%) | 10 (24.4%) |
Survival | 86 (69.9%) | 31 (75.6%) | 77 (62.6%) | 31 (75.6%) |
Total | 123 | 41 | 123* | 41 |
◊ All “No Survival” occurrences were due to patient death.
*Including one person who was lost to follow-up (0.8%).
Table 3. Survival Outcome, Patients who did receive pre-implant circulatory rescue interventions (December 22, 2017 Final Report)
Through 3 Months post-implant | Through 6 Months post-implant | |||
Companion 2 | CSS Console | Companion 2 | CSS Console | |
No Survival◊ | 32 (41.6%) | 10 (20.8%) | 34 (44.2%) | 13 (27.1%) |
Survival | 45 (58.4%) | 38 (79.2%) | 43 (55.8%) | 35 (72.9%) |
Total | 77 | 48 | 77 | 48 |
◊ All “No Survival” occurrences were due to patient death.
Stroke
A secondary objective of the post-approval study was to provide descriptive statistics on select adverse events. The INTERMACS neurological adverse event rate is a composite outcome of several separate adverse events including: transient ischemic attacks, cerebrovascular accidents (ischemic/embolic stroke, hemorrhagic stroke, and strokes from other causes), confusion, seizures, encephalopathy, and other neurologic events.
The stroke rate for patients initially supported with the C2 Driver System was statistically significantly higher than the stroke rate for patients initially supported with the CSS Console at three months and six months post-implant (Table 4). Other neurological adverse event rates were similar between the two groups.
Table 4. Stroke Outcome, All patients (December 22, 2017 Final Report)
Through 3-months post-implant | Through 6-months post-implant | |||
Companion 2 | CSS Console | Companion 2 | CSS Console | |
Stroke | 53 (26.5%) | 7 (7.9%) | 54 (27.0%) | 7 (7.9%) |
No Stroke | 147 (73.5%) | 82 (92.1%) | 145 (72.5%) | 82 (92.1%) |
Total | 200 | 89 | 200* | 89 |
*Including one person who was lost to follow-up (0.5%).
Stratification by pre-implant circulatory rescue intervention status demonstrated that the stroke rate was higher for patients using the C2 Driver System at three months and six months post-implant irrespective of the need for pre-implant rescue intervention (see SynCardia post-approval study webpage).
Data Limitations
Although the differences in mortality and stroke outcomes in this study are notable, we do not know the root cause for these observed differences, and the results are not adjusted for potential confounders. Cohort designation was ‘as-treated,’ based upon a subject’s initial driver system at time of TAH-t implantation; any effects due to driver system exchanges within 6 months are therefore not captured by these results.
The information provided in this letter is meant to assist physicians in understanding the performance of the TAH-t using the C2 Driver System as seen in the post-approval study. We recognize that the CSS Console is no longer available as a driver option for TAH-t patients, and that consequently, physicians may determine that the C2 Driver System is the most appropriate option for patients with severe biventricular failure in need of mechanical circulatory support.
RECOMMENDATIONS
The FDA has the following recommendations for health care providers:
Carefully consider these mortality and stroke results from the TAH-t post-approval study when making treatment decisions, and discuss the risks and benefits of the C2 Driver System with patients.
Report any adverse events or suspected adverse events experienced with the SynCardia TAH-t and Driver Systems. Voluntary reports can be submitted through MedWatch, the FDA Safety Information and Adverse Event Reporting program. Device manufacturers and user facilities must comply with the applicable Medical Device Reporting (MDR) regulations. Health care personnel employed by facilities that are subject to the FDA's user facility reporting requirements should follow the reporting procedures established by their facilities. Prompt reporting of adverse events can help the FDA identify and better understand the risks associated with medical devices.
Return all devices associated with, or suspected to be associated with, any adverse events to the manufacturer for evaluation to help them and the FDA better understand the issue.
FDA ACTIONS:
The FDA will continue to work with SynCardia to ensure that the product labeling addresses the post-approval study findings. We will continue to keep the public informed if new or additional information becomes available.
CONTACT US
If you have questions about this communication, please contact the Division of Industry and Consumer Education (DICE) at [email protected], 800-638-2041 or 301-796-7100.
Sincerely,
/s/
William Maisel, MD, MPH
CDRH Chief Medical Officer
Center for Devices and Radiological Health
U.S. Food and Drug Administration