On-X Aortic Prosthetic Heart Valve Low Dose Warfarin Post Approval Clinical Registry Study

The study is a prospective, multicenter, observational single arm study of newly-enrolled patients treated with the On-X Aortic Prosthetic Heart Valve. Assuming 20% attrition over 5 years and 40% of enrolled patients will be high risk home monitoring patients, 510 patients will be enrolled in the study. Patients will be under standard anticoagulation (INR 2.5 (2.0-3.0 range)) for at least 3 months before initiation of low dose (INR 1.8 (1.5-2.0 range)) anticoagulation treatment.

All centers will follow a common protocol in which eligible patients will be entered into the registry within 1 year after receiving the On-X Aortic Prosthetic Heart Valve. No special diagnostic or therapeutic procedures will be done for the purposes of the protocol and data will be collected prospectively on each patient for 5 years.

Data from all consenting patients at participating sites receiving the On-X Aortic Prosthetic Heart Valve will be entered into an online registry database. Patients will be recruited postoperatively within 1 year, most typically at the first post-discharge visit at between 2 and 6 weeks but also by special visit to their surgeon if agreed. Low dose therapy will begin no earlier than 3 months postop. Data entry is non-randomized, and continues until the sample size requirements are met and the enrollment into the registry is closed. To minimize bias all patients meeting the inclusion/exclusion criteria will be recruited and those agreeing to participate in the follow-up will be entered into the database. A screen failure log of those patients not agreeing to participate will be kept and entered identifying reasons for declined enrollment. Once entered patients will be analyzed as a whole and separated into groups by TE risk factors and warfarin monitoring method. All patients that have no contraindication for aspirin will be prescribed a daily 'baby' aspirin (75-100mg) for consistency to prior research and to current society guidelines.

The primary hypothesis for the registry is to confirm the results seen in the IDE trial, showing that using home monitoring or anticoagulation clinic with an INR target of 1.8 (range 1.5 to 2.0) does not significantly increase patient risk relative to current standard of practice.

The primary analysis will compare the overall composite event rate for the following events:

• Thromboembolism (TE)
• Valve thrombosis (VT)
• Major bleeding

Each of the component events and the combination of TE plus VT will be examined separately as secondary endpoints.

The sample size for the registry was calculated with the following assumptions:

• Incidence of composite outcome estimated via Poisson regression
• 1-sided test comparing the composite outcome to the reference value
• 5% significance level
• 90% power
• Expected overall composite proportion from the IDE high risk treatment group (pT) = 0.0457/patient-year (ptyr)
• Reference value (p95, upper 95% confidence bound from PROACT study) = 0.0693/ptyr
• 800 patient-years of follow-up
• Anticipated 5 years of follow-up per subject
• Loss to follow-up over 5 years of 20%

The resulting enrollment target is 510 subjects, which would result in approximately 816 patient-years in the high-risk home-monitoring group. These subjects will be recruited from at least 15 clinical centers and no more than 35 centers.

The sponsor will provide a secure Part 11 compliant online database for entry of required preoperative, operative, follow-up and adverse event data. All appropriate sections of the registry must be filled out accurately and completely.

To protect patient confidentiality, the sponsor will use the information from the registry for statistical purposes related to the hypothesis of the trial only and will not routinely collect all source document medical records, and when such records may be collected they will be de-identified, i.e. for adjudication of adverse events.

The statistical plan provided in this section follows generally the AATS/STS guidance which is referenced by both FDA and international standards as containing the preferred methods of analyzing heart valve study data. As the primary and secondary endpoints of this study are exclusively adverse events, the methods used to analyze these endpoints will be those specifically used with adverse events. The objective performance criteria (OPC) generally used for premarket studies will not be used in this post-approval study in preference of comparisons to rates calculated within AVR control arms of the prior IDE trial (G050208).

The cohort will be analyzed as a whole and in subgroups based on INR monitoring method and TE risk category where TE risk is defined by the clinical and laboratory criteria of G050208, except that history of known hypercoagulability will be acceptable for documentation and no new blood testing for such hypercoagulability will be required. Confounding factors to be examined include age at implant (under 50, 50-65 inclusive, over 65), gender, preoperative NYHA classification, occurrence of concomitant or prior cardiac surgery (including stent placement) and valve size (21 or less, 23 or more).

Early events will be presented as simple percentages. In this study early events will be presented in three categories: first the standard 30-day postop or date of discharge (whichever is longer) category, second the 3-months prior to eligibility for home monitoring category and third the period prior to enrollment (up to 1 year depending upon the patient).

Late events will be determined from the time of shift to low INR and the possible time for initiation of home monitoring when done. Late event analyses will be conducted by two methods: linearized rates (to cover total independent events) and Kaplan-Meier life tables (to cover time to first event).

A steering committee for the registry shall include at least 4 members with at least 2 cardiac surgeons, 1 cardiologist and 1 statistician. The committee shall provide input into the study conduct and any changes to study design, data elements required or statistical procedures to be used in analyses. They shall also review any publication that arises from the registry. The members will be identified prior to commencing enrollment in the registry. The committee will also either act as or appoint a separate clinical event committee (CEC) consisting of at least 2 experienced clinical investigators not participating in the registry. CEC members may be members of the steering committee but may also be in addition to the steering committee. Both steering committee and CEC will operate from an established charter for quality assurance.

The anticipated uses of the data collected will be FDA and other regulatory review as part of post approval study requirements and the publication of the aggregated data in peer reviewed meetings or journals. It is expected that the registry will be fully enrolled within 1 year after the last site is enrolled and that follow-up will be limited to 5 years per patient. Once fully enrolled the registry will be closed to further entry of patients.