Since August 2021, Exactech has been working closely with FDA to understand and evaluate potential risks created by the packaging nonconformance (discussed below) that affected multiple product lines. To eliminate confusion and concern in the market, Exactech is now proceeding with a recall and removal of Equinoxe UHMWPE devices packaged in nonconforming packaging, as recommended by FDA. This general advisory notice summarizes important aspects of the clinical experience of Exactech’s Equinoxe anatomic total shoulder (aTSA) and reverse total shoulder (rTSA) implants since 2004. In light of past recalls of other Exactech implants packaged in nonconforming vacuum bags, this general advisory notice also describes the results of mechanical testing conducted by an independent lab to characterize the strength and oxidation associated with polyethylene devices in nonconforming packaging up to the 5-year shelf life of the Equinoxe, along with a comparison describing the biomechanical differences in the shoulder vs. the Knee, Ankle, and Hip.
Beginning in August 2021, Exactech initiated recalls of its Knee, Ankle, and Hip conventional Ultra High Molecular Weight Polyethylene (UHMWPE) devices in non-conforming packaging because the multi-layer vacuum bags used to package these implants lacked a secondary oxygen barrier layer of ethylene vinyl alcohol (EVOH). (Attachment 1) As described in the recall letters, the use of these non-conforming bags may enable increased oxygen diffusion to the UHMWPE insert, which may increase oxidation relative to inserts packaged with the specified additional oxygen barrier layer. Over time, oxidation can potentially degrade the mechanical properties of conventional UHMWPE, which, in conjunction with other surgical factors, can lead to both accelerated wear debris production and bone loss, and/or component fatigue cracking/fracture, all leading to corrective revision surgery. We recalled the Knee, Ankle, and Hip devices because the clinical effects of polyethylene oxidation were observed in some of these patients.
The majority of our UHMWPE devices packaged from 2004 to August 2021, including approximately 88% of the Equinoxe UHMWPE devices, were packaged in these nonconforming bags. As such, the entire clinical history of the Equinoxe aTSA (released November 2004) and the Equinoxe rTSA (released March 2007) has utilized this non-conforming packaging. In this notice, we are providing a summary of the clinical experience, mechanical test results, and biomechanical information so that you can understand this real-world evidence to inform your clinical practice and monitor your Equinoxe shoulder patients.
Clinical Data
The clinical evidence analyzed includes long-term registry data from multiple countries, long-term PMCF data (since 2004), and a 2005-2024 literature review. The results of each are summarized below:
Registry Data – Exactech analyzed revision rates and the indications for revision surgery for aTSA and rTSA across the UK and Australian national joint registries to identify if the packaging nonconformance resulted in any increased rates of polyethylene related failures.
- UK and Australian National Joint Registries – Exactech compared the all-cause revision rates of the 2,321 Equinoxe aTSA patients and the 10,673 Equinoxe rTSA patients to all other aTSA and rTSA patients in the UK and Australian registries from 2011 to 2023. (Attachment 2)
-
- The UK registry enrolled 5,391 Equinoxe rTSA patients and reported 138 revisions (2.5%); the cumulative all-cause rTSA revision rate in the UK at 10 years was 4.5%. The Australian registry enrolled 5,282 Equinoxe rTSA patients and reported 168 revisions (3.2%); the cumulative all-cause rTSA revision rate in Australia at 8-years was 5.9%.
- No difference in all-cause revision rate was observed in the Australian registry (n=5,282, HR=0.95, p=0.514) or in the UK registry (n=5,391, HR = 1.10, p=0.308) relative to all other rTSA shoulders.
- The UK registry enrolled 1,606 Equinoxe aTSA patients and reported 65 revisions (4.0%); the cumulative all-cause aTSA revision rate in the UK at 10 years was 6.4%. The Australian registry enrolled 715 Equinoxe aTSA patients and reported 61 revisions (8.5%); the cumulative all-cause aTSA revision rate in Australia at 9-years was 14.3%.
- No difference in all-cause revision rate was observed in the UK registry (n=1,606, HR=1.19, p=0.182).
- A difference after 1.5 years was observed in the Australian registry (n=715, HR = 1.65, p=0.003) relative to all other aTSA shoulders. (The devices sold in the UK are the same as devices sold in Australia.) Exactech observed differences in patient cohorts; these differences were not identified by the Australian registry as being causative, but we provide them for your information and evaluation: Equinoxe aTSA patients were generally younger (14.2% more patients <65 years at the time of surgery), more active/healthy (8.8% more Equinoxe patients had an ASA ≤2 and 12.5% less Equinoxe patients were deceased), had more glenoid deformity (>30% of Equinoxe aTSA patients received a posterior augment glenoid), and had a greater percentage of male patients (7.0% more males) as compared to all other aTSA shoulders in Australia.
- An analysis of the reasons for revisions in each registry was conducted to quantify the potential polyethylene failure modes (dissociation, wear/lysis, fracture, and loosening). (Attachment 2) The cause of any revision surgery is often multi-factorial. It should be noted that the UK registry reports all reasons for a given revision procedure, such that in the UK, 163 reasons for revision were reported for the 138 rTSA revision cases and 86 reasons for revision were reported for the 65 aTSA revision cases. On the other hand, the Australian registry only reports 1 reason for any given revision procedure. Therefore, some polyethylene related failure modes in Australia may not be captured if a polyethylene failure mode was not reported as the primary cause for the revision.
- For rTSA from 2011 to 2023, revision rates for potential polyethylene failure modes are the same as for other rTSA shoulders.
- Australian and UK registries combined reported 0.15% of Equinoxe rTSA patients revised for dissociation (n=16), 0.06% revised for polyethylene wear/lysis (n=6), 0.02% revised for implant fracture (n=2), and 0.36% revised for loosening (n=38).
- For aTSA from 2011 to 2023, revision rates for potential polyethylene failure modes are the same as for other aTSA shoulders.
- Australian and UK registries combined reported 0.3% of Equinoxe aTSA patients revised for polyethylene wear (n=7), 0.22% revised for lysis (n=5), 0.13% revised for dissociation (n=3), 0.13% revised for implant fracture (n=3), and 1.38% revised for loosening (n=32).
- For rTSA from 2011 to 2023, revision rates for potential polyethylene failure modes are the same as for other rTSA shoulders.
- The UK registry enrolled 5,391 Equinoxe rTSA patients and reported 138 revisions (2.5%); the cumulative all-cause rTSA revision rate in the UK at 10 years was 4.5%. The Australian registry enrolled 5,282 Equinoxe rTSA patients and reported 168 revisions (3.2%); the cumulative all-cause rTSA revision rate in Australia at 8-years was 5.9%.
Exactech’s multi-center post-market clinical follow-up (PMCF) outcomes study – Exactech maintains a prospective PMCF multi-center clinical outcomes study across more than 40 sites in the US and Europe. From November 2004 to July 2023, 4,431 Equinoxe aTSA and 11,206 Equinoxe rTSA patients were enrolled. (Attachment 3) Each clinical site utilizes the same data entry forms and the reasons for complications and revision are captured with checkboxes on the standardized adverse event form.
- 18 Equinoxe aTSA patients had some notation of polyethylene wear, osteolysis, or massive bone loss in the adverse event form for an overall occurrence rate of 0.41%.
- 8 Equinoxe rTSA patients had some notation of polyethylene wear, osteolysis, or massive bone loss in the adverse event form for an overall occurrence rate of 0.07%.
Peer reviewed clinical publications – A comprehensive review of 304 peer reviewed publications of various study designs (213 clinical studies, 39 computer studies, 23 bench tests, 15 case studies, 8 registry studies, 5 surgical techniques, and 1 retrieval analysis) from 2005 to 2024 by 658 different authors reported no cases of UHMWPE oxidation or polyethylene delamination with Equinoxe. (Attachment 4) Conflicts of interest are reported in each peer-reviewed study and 141 of these 304 peer-reviewed publications listed an Exactech employee as an author. Additionally, the clinical data source used for nearly all of the Equinoxe clinical publications with an Exactech author is the Equinoxe multicenter clinical database (so there is some data overlap with the PMCF section above). Finally, it should be noted that these publications had multiple different study designs and not all of the clinical studies, case studies and registry analyses reported complications and revisions.
Shoulder Biomechanics
The shoulder is biomechanically different than the knee, hip, or ankle. Peak loading in the shoulder is unrelated to gait, and because we do not walk on our shoulders, peak loads in the shoulder are lower, generally 1 to 1.5 times body weight, and occur less frequently than peak loads in gait-loaded joints, which can be 3 to 6 times body weight and may occur multiple times in each gait cycle. Because of the lower implant stress in the shoulder, polyethylene related failure modes are less common with shoulder arthroplasty than with total knee arthroplasty, total hip arthroplasty, and total ankle arthroplasty. While shoulder motion may occur on the same frequency of knee/hip/ankle motion, a review of the aforementioned national joint registries indicate that revisions for polyethylene wear, dissociation, osteolysis, and fracture are rare with shoulder arthroplasty across all manufacturers.
Mechanical Test Results
Exactech hired an independent test lab to perform real-time oxidation testing and small punch strength testing on UHMWPE inserts packaged in nonconforming bags: (Attachment 5)
- The oxidation index (OI) test results up to the 5-year shelf life of the Equinoxe measured a maximum OI of 0.70. As described in the UHMWPE Biomaterials Handbook, an OI less than 1.0 is considered low and unlikely to have a substantially negative impact on mechanical behavior.
- Strength test results of UHMWPE measured at OI=0.74 and demonstrated similar work-to-failure strength as UHMWPE with no oxidation (OI=0).
Exactech will continue to review the clinical performance of the Equinoxe Shoulder system and investigate all reported adverse events.
ATTACHMENTS
All referenced attachments are available upon request. Please email advisory-notice@exac.com
- Vacuum bag construction details
- UK and Australian Registry analysis
- PMCF results
- Peer reviewed publication summary
- OI Test Results and Mechanical strength test results
- Device UDI listing