|
|
| General |
| Study Status |
Completed |
| Application Number |
P050050 / PAS001 |
| Date Current Protocol Accepted |
05/09/2014
|
| Study Name |
Long Term Study
|
| General Study Protocol Parameters |
| Study Design |
Prospective Cohort Study
|
| Data Source |
New Data Collection
|
| Comparison Group |
Historical Control
|
| Analysis Type |
Analytical
|
| Study Population |
Adult: >21
|
| Detailed Study Protocol Parameters |
| Study Design Description |
The specific objective of this long-term study is to obtain information on the revision and removal rate for the STAR Ankle over time in order to establish the device¿s survivorship.
The post approval study is a prospective, multi-center, single-arm study at up to the 10 primary sites that participated in the pivotal and continued access studies. The study is continued follow-up of the premarket cohort and continued access cohort.The long-term results of Ankle Arthrodesis summarized from literature review by meta-analysis will be used as historical fusion control.
|
| Study Population Description |
The study population consists of all living subjects who participated in the continued access study.
Device Group: subjects implanted with the STAR Ankle
Comparison Group: Historical Fusion Control (i.e., the long-term results of Ankle Arthrodesis summarized from literature review by meta-analysis) Fusion results reported by Haddad, et al. (2007)7 will serve as the control group for the STAR Continued Access subjects. [Haddad SL, Coetzee JC, Estok R, Fahrbach K, Banel D, Nalysnyk L. Intermediate and long-term outcomes of total ankle arthroplasty and ankle arthrodesis. A systematic review of the literature. J Bone Joint Surg Am. 2007 Sep;89(9):1899-905.]
|
| Sample Size |
minimum 125 patients (STAR Ankle recipients)
The comparison of the STAR ankle is to a literature control, in Haddad et al. [2007], with a survivorship rate of 86.3%. An informal calculation indicates how many successes are needed in the STAR ankle group to claim non-inferiority. Sponsor assumed that the observed success rate for the fusion control was 94/109 = 86.3%, where the 109 is the effective sample size. Then, from StatXact-8 software, 40/47 = 85.1% is the smallest observed success rate so that a lower 90% confidence bound for the difference of success probabilities is no less than ¿15%. Note: This calculation uses two observed proportions, and is not a power-based calculation of detectable effect size.
Using PASS2005 sample size software, the sponsor performed a two-sample, exact non-inferiority calculation to compare the STAR ankle to a historical control with 109 subjects, a true success rate of 86.3%, and using a one-sided test with 5% Type 1 error rate, 80% power, and a 15% non-inferiority margin. The necessary sample size to show non-inferiority is 47 subjects in the STAR ankle study arm. However, SBi¿s intention is to enroll at minimum 125 patients and follow all who continue to participate. The meta-analysis performed by Haddad et al. [2007] had a sample size of 1262 subjects who received an arthrodesis. However, for purposes of the sample size calculation, the effective sample size in Haddad et al. [2007] is only 109, far less than the 1262 subjects in the articles that were included in Haddad¿s meta-analysis. A technical justification for this reduced number of control subjects appears in protocol Section 6.1.3.1
|
| Data Collection |
Primary Effectiveness Endpoint
Device Survivorship (Revision and Removal) at 8 years after implantation
Note: The information collected at these post-operative intervals will be added to the previously collected follow-up data (e.g., demographic, baseline, post-operative through 24-months).
Other Effectiveness Endpoints
Total Buechel-Pappas Scale score
(100-point scale consisting of subscales for pain (40 points), function (40 points), range of motion (15 points), and deformity (5 points))
Pain Visual Analog Scale (VAS) (100mm scale)
Quality of Life (SF-36)
AOFAS
Radiographic endpoints (assessed by Anterior-posterior and lateral x-ray images of the treated ankle)
Radiolucency
Migration
Safety Endpoints
All adverse events, including information concerning the nature, onset, duration, severity, and relationship to the device and operative procedure
|
| Follow-up Visits and Length of Follow-up |
96 months
48 months (± 8 weeks) , 72 months (± 8 weeks), and 96 months
(-8 weeks and any time after 96 Months) post-operatively. Baseline, 6 week, 12 month, and 24 month data will be used as collected for the study in which they originally participated.
|
| Interim or Final Data Summary |
| Actual Number of Patients Enrolled |
95
|
| Actual Number of Sites Enrolled |
5
|
| Patient Follow-up Rate |
91% (87/95)
|
| Final Safety Findings |
There were 82 device-related AEs in 37 subjects. Based on 87 subjects with 96-month follow-up, at least 12 (13.8%) were for polyethylene fracture, all requiring revision. 18 (20.7%) of the device-related adverse events were for formation of cysts, 16 of which required surgical treatment. 21 of 95 (22%) subjects had to undergo one or more secondary procedures related to revision or removal of the device.
|
| Final Effect Findings |
The overall Kaplan-Meier survivorship estimate for the cumulative surviving proportion based on the first removal/revision at 96 Months is 75.5% (std. error: ±3.9%). This was found to be not worse than the predefined arthrodesis control.
|
| Study Strengths & Weaknesses |
Long-term (8 year) follow-up are available in 91% of the enrolled study subjects, a study strength. Study weaknesses include: the study enrolled only 76% of the subjects from the continued access study from the IDE study and only 5 of 10 of the IDE study sites. Though some similarities among study participants and non-participants were demonstrated, selection bias remains a study limitation. Also, the study population may not be representative of real-world use.
|
| Recommendations for Labeling Changes |
A labeling change is recommended to include the long-term follow-up data generated in this Post-Approval study, including revision rates and device-related adverse event rates. The risk related to polyethylene fracture should be identified.
|
