Skip to main content
Online Exclusives

Examining Obligations When An Implant Procedure Fails: Reflections From A Surgeon

What should a surgeon do when an implant’s procedure outcomes are not as expected? This author discusses his experience with the Cartiva implant and reflects on how to approach new procedures. 

The Cartiva implant teaches us a lot about how to analyze a product for patient care. Being involved with the Cartiva company as a pre-launch consultant for the implant and seeing the outstanding results from the initial, randomized multicenter trial, I was sure this was an excellent option for patient care.1 When the results in my practice did not match the results of the clinical trial, it made me consider what happened and why. How is the community data so different from that of the initial trials? What is our responsibility to patients, the company and the FDA when a product is not working as described and approved?

The Cartiva implant has been on the market for approximately three years in the United States. For most orthopedic products, companies go through a process to obtain a 510k clearance from the Food and Drug Administration (FDA). The Cartiva implant went through a full “new” product approval process, which requires an extensive amount of clinical trial work, and is far more rigorous and costly than the process for a 510k clearance.

The results of the initial study commissioned for Cartiva by a consortium of surgeons suggested exceptional results.1 The data showed overwhelming performance and outcomes superior to anything on the market. The implant allowed for rapid recovery, weightbearing after surgery, expedited return to shoe gear and a simple, reproducible surgical procedure. 

I was among the first foot and ankle surgeons to be approached regarding bringing the Cartiva implant to market, and was introduced to the product about a year prior to its official release. It was very exciting and study details showed over 90 percent success, which was very impressive.1 Even more impressive was the fact that the procedure took about 20 minutes to perform and left the architecture of the joint intact, which is critical considering future procedures and the ease of revisional surgery should it be required. A group of us worked diligently on the implant release and guidelines, helping educate surgeons while also sharing our results with each other.

Over the course of approximately one year, our group placed about 50 Cartiva implants in patients. The procedures went very smoothly and were without complication. The patients had minimal pain and returned to shoe gear over three to four weeks with little difficulty. However, the longer-term results did not match anything close to the outcomes of the initial clinical trial.  

I began to notice a few months after regular performance of the procedure, that approximately 50 percent of our Cartiva post-operative patients had a reduction in joint space and what looked like a collapse of the implant. Discussions with other surgeons in the start-up group suggested similar findings. 

Over time, I began to discuss with both Cartiva and later Wright Medical the idea of subsidence, or collapse, of the implant into the bone, but this was not seen as a common problem or finding in the aforementioned multicenter trial commissioned by Cartiva. Wright Medical also felt its independent studies were very positive. I began to consider placing the implant two mms and then three mms proud instead of the one mm manufacturer guideline after group discussions and company suggestions. The implant would still push back into the bone in many of these cases, leaving patients with pain and the need for revision.  

This issue with this procedure’s complications as a whole has made me think about the proper considerations and actions with respect to the FDA, the company, the research group, myself and my patients. I hope this article will also help you as well with your thought process on new procedures and implants.  

Pertinent Questions When Evaluating Long-Term Results

How could the results of the initial study be so positive while the results in the community are not nearly as successful? In this case, there may have been some fault with the study itself, the questions asked and the time frame to assess the implant’s success or failure. I do not believe that radiographic studies were followed long-term, and the incidence of subsidence may not have been reported appropriately.   

Considering the cost and amount of data necessary to approve the Cartiva implant, this scenario would be a bit of a shock. Companies looking to bring a new product to the market follow the data trail. They may have an orthopedic or podiatric consultant, but they are driven by business and data. The clinical trial is set up by the company. While the design of the clinical trial has a great deal of input from the physicians involved, the company is primarily looking to the data to see if its product makes sense or not for release. The last thing the company wants is to spend $100 million bringing a product to market only to find out it does not work. 

It is difficult to believe anyone doing such a research project would risk providing skewed data as it carries liability and could severely ostracize a doctor or group of doctors. Therefore, my thought is that the data may not been have collected properly. Those involved with the study may not have asked the right questions, obtained the right radiographic studies or posed the right post-surgical questions regarding patient outcomes. While I do not know this for certain, there may have been faulty data in the initial trial that affected the results. 

Reassessing The Implant Procedure 

As a surgeon, I look at the data, consider the initial study results, consider the depth of the study and then look at the risk versus reward of the procedure. In this case, the data looked exceptional, there was reportedly a rigorous collection and study process for the initial study and the FDA approved the product. The procedure appeared to have very little risk, meaning a failure could be modified without much difficulty, which is always a plus.  

Again, as a surgeon, I discuss the potential risks and rewards with my patient. The group at my practice did about 50 cases with some modifications of implant position from one to three mm of prominence. Over a two-year period, we also had 50 patients with failures of the Cartiva procedure referred to us from other specialists. Fifty percent failure of our implant and an additional large number of failures referred to us from the community is far too much in comparison to the study results. This data must be and was reported to the Cartiva and Wright Medical, mainly verbally through the sales representatives. 

Was there an implant failure that is reportable to the FDA, was it surgeon error or was it just a bad result? That is the toughest part to analyze. Who reports this data to the FDA? What is done with the data? Are we as surgeons responsible to report a bad outcome, even if the implant looks pristine upon removal, but it subsided or sunk back into the bone? Why are some patients ecstatic with their results and others having such problems? There are so many questions to ponder. 

While I have my own considerations and questions, it is also important to address the patient and his or her considerations. I believe the manufacturer, the FDA, the research group and the surgeons who use this implant all have the best of intentions. No one is attempting to fool or pull the wool over anyone else’s eyes. However, once an implant procedure fails at a high rate, the company, the FDA and surgeons are all responsible to warn patients and each other. I have done this via communications with Cartiva and Wright Medical sales representatives, through patient education online and in the clinical literature.

Our group has also begun to look at our own data and found no correlative factors for which cases failed and which did not. It seems to be somewhat a roll of the dice. 

We have stopped using the implant as a primary treatment option unless the patient is fully informed as to the risk of failure. Some patients want this procedure and understand that if it fails, they can have a revision with little anticipated difficulty. Even so, should we perform this procedure if we have experienced a 50 percent failure rate?  

Discussing Surgical Alternatives

In our offices, we have returned to utilizing the HemiCAP DF® (Arthrosurface)implant. In our experience, it is stronger and has very limited complications over the course of time we have used it. It also seems to last for years and hold up very well. The implant does not rely on cortical bone for support but rather is stabilized mainly by the Morse taper screw system. This system is very solid and allows the implant to work well even if the bone is not of outstanding quality. I do caution against using this implant in patients with poor bone stock.  

We also have gone back to the tried and true option of first MPJ fusionin moderate to severe cases of osteoarthritis. A fusion procedure can be truly outstanding, endures the stresses of life very well and allows for complete or almost complete resolution pain in the great toe joint. There is some misconception about fusions not being good procedures. However, in my experience, fusions work well as long as the patient does not want to wear a high heel over two inches.  

Finally, we have also begun to perform more decompression osteotomy-type procedures in moderate arthritis cases in combination with a cheilectomy. This has also worked very well in the right cases. 

How does one choose the next step after failure of a Cartiva implant procedure? They have been fairly easy to convert to revisional procedures, which was one of the great selling points of the surgery. If it does not work, it is easy to convert. Conversion options have mainly included a fusion of the great toe joint. Fusion with a bone graft fill of the implant hole is fairly easy to perform. I usually choose plate fixation for added stability. In patients who are adamant to avoid a fusion, we have placed the HemiCAP DF implants when swapping for the eight mm Cartiva implant. We have also used silicone implants, with warnings of failure over time, when removing the 10 mm Cartiva implant.  

Final Thoughts

When examining my experiences with this implant procedure, I feel it is important to educate surgeons and patients as to its risks versus the benefits. In turn, surgeons should carefully consider use of the implant in more detail prior to placement as we are now finding the procedure has not worked as well in many surgeons’ hands in contrast to what was reported in the previous literature.1 Additionally, it is important to consider the future of this procedure. Is it safe for the market as presented? Are there other surgeons experiencing far superior results? If so, what are they doing differently? 

Under many circumstances, I do not choose to be an “early adopter” of a procedure. I prefer to wait and watch for a while. However, the initial results for the Cartiva implant were so promising, the procedure was so straightforward, and revision appeared to be so as well. I then adopted the procedure earlier than I usually would. 

That being said, I believe we all have an obligation to monitor our results, consider our failures and share positive and negative data with the surgical community in order to serve all patients in the most comprehensive and efficacious manner possible. 

Dr. Baravarian is an Assistant Clinical Professor at the UCLA School of Medicine. He is the Director and Fellowship Director at the University Foot and Ankle Institute in Los Angeles. (

Online Exclusives
By Bob Baravarian, DPM, FACFAS

1. Baumhauer JF, Singh D, Glazebrook M, et al. Prospective, randomized, multi-centered clinical trial assessing safety and efficacy of a synthetic cartilage implant versus first metatarsophalangeal arthrodesis in advanced hallux rigidus. Foot Ankle Int. 2016;37(5):457-69.

Back to Top