August 13, 2014
5 min read

Autologous chondrocyte implantation in the patella is only part of the story

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

Recently Gomoll and colleagues published an excellent study entitled “Autologous chondrocyte implantation in the patella: A multicenter experience.” The study describes the use of autologous chondrocyte implantation in patients with patellofemoral disorders. It is worth a careful read. However, the title of the project only tells part of the story.

This is a large case study with 110 patients who underwent a variety of patellofemoral surgery (always including autologous chondrocyte implantation [ACI]) at four different centers with minimum 4-year follow-up. In addition to patellar chondral damage, there were bipolar lesions (i.e., trochlear chondral lesions) in 27% of the patients. Patellar defects averaged about 5 cm2 and were central patellar lesions in 72% of patients. Other concomitant procedures figured prominently in the treatment of these patients and included anteromedialization of the tibial tuberosity (AMZ) in 69%, lateral release or lengthening in 41% and vastus medialis advancement in 20%. I could not tell from reading the paper and the online supplement if any patients had a truly isolated ACI treatment in the patella and/or trochlea.

Pain and instability

William Post

William R. Post

The stated indication for treatment was unremitting pain resistant to nonoperative treatment, which was said to include 3 months to 6 months of supervised physical therapy. Indications were not clearly defined for all the operations. Patellofemoral joint space narrowing had to be less than 50% for ACI and the indication for tuberosity transfer was history of patellar instability (dislocation or subluxation), maltracking on physical examination and tibial tuberosity trochlear groove measurement of more than 15 mm, which is a threshold lower than the 20 mm often quoted in the patellofemoral literature. Lateral release/lengthening were performed when there was decreased patellar mobility.

Regarding the question of indications for AMZ, it seems likely that the patients varied somewhat between centers as the center with the most patients reported a 97% rate of tibial tubercle osteotomy compared with rates between 53% and 60% at the other three centers. If the rate of AMZ was this high and the indication for AMZ required instability, it seems most of these patients indeed had a combination of pain and instability.

As such, perhaps the study might be considered to be a study about the value of adding ACI to patellofemoral realignment or perhaps the value of adding patellofemoral realignment to ACI — depending of course on one’s perspective on which might be considered the primary treatment. Of course, in reality it is difficult to separate the effects of these two components of the procedure. So perhaps we should consider this retrospective case report study for what it really is — a study of patellofemoral realignment/unloading and ACI for patients with patellofemoral pain and instability.

Importance of ACI

These are difficult patients to treat and the authors should be congratulated for their efforts and success. Patients were satisfied overall in the majority of cases. The question, however, remains how important ACI was to the results. Early studies of ACI (without realignment/unloading) for patellofemoral articular damage had little success and resulted in the current off-label status of ACI in the patellofemoral joint. As surgeons incorporated principles of unloading and realignment, it is sometimes said that results of ACI improved. What really improved was the outcome for the patient when unloading/realigning was included with ACI.

Sorting out whether ACI is necessary for these patients is an important question and this study is another arrow in our quiver as we pursue this understanding. The best results in this study were in patients who had more limited articular cartilage damage, as are the results in studies of realignment/unloading without ACI. One wonders how much better these patients are than those treated by AMZ without ACI, if indeed they are any better. For any discussion of this, our literature serves as historical control.

A complete review of the results of AMZ is way beyond the scope of this blog and the reader is referred to an excellent review by Sherman and colleagues on this subject. However, a few key points deserve analysis. The authors focused on the study by Pidoriano and colleagues, which included 37 patients who underwent AMZ and reported results based on location of the chondral lesions. Treatment of chondral lesions included debridement and for grade 4 lesions drilling. They reported that overall for the grade 4 lesions, 70% had good to excellent subjective results. The emphasis of their conclusion was that the results correlated more with the location of the lesion than with the depth. Medial facet (nine cases) and central diffuse (five cases) had less satisfactory results. Perhaps most notably only one of the five central patients had a good result. In this area of the patella, it is notable that Gomoll and colleagues have achieved more satisfactory results, however, we must remember that the historical control to which we are asked to compare included only five patients.

Further eroding the definitive value of this historical control, Pidoriano and colleagues noted that if worker’s compensation patients are removed from their study results, the overall good to excellent results improved to 50% for the medial facet and diffuse central groups. So when the worker’s compensation patients are removed from consideration we now have less than five knees in which there were 50% good or excellent results. I think we need to be careful drawing conclusions on this limited data.

Basic science studies

Basic science studies of tibial tuberosity elevation have shown consistent unloading of pressures in the entire patellofemoral joint, including the trochlea. Relatively speaking, the distal portion of the patella is unloaded most. When combined with medialization of the tuberosity, loads have been shown to relatively increase some on the medial patella. However, recall that such studies are not done on knees with pre-existing malalignment in which the medial patella may have been physiologically under-loaded as a result of the chronic tilt and lateral patellar translation that is commonly present. Such unloading makes sense for treatment of the pre-existing overload and also to create a less stressful environment for the transplanted chondrocytes.

Another factor I personally think is probably one key to the success of these patients is a rational rehabilitation program that allows for osseous healing and restoration of homeostasis. That is not to say I think the rest alone was all these patients needed. Certainly rest and gentle progressive rehab can produce sometimes surprisingly excellent results even without surgery, but meticulous attention to avoiding overload is imperative in the postoperative care of this type of patient. The rehab program in this study was less aggressive than prior reports of AMZ or other realignment operations. Patients in this study were touchdown weight bearing for 6 weeks, gradually progressed weightbearing until 3 months and were discouraged from open chain knee extension and stair climbing for 6 months postoperatively.

This brings us to the hypothesis that restoring homeostasis of load acceptance in the patellofemoral joint may well be the underlying principle unifying the treatment outcomes we observe. Certainly it is reasonable to try and improve the malalignment of forces surgically when it can be clearly demonstrated clinically. In my own experience, this is best done with SPECT scans which simultaneously illustrate metabolic overload and alignment. Unloading/realigning with a goal of homeostasis would ideally include biological resurfacing to facilitate load transfer.

Not without costs

Autologous chondrocyte implantation is not without costs, both economical costs and morbidity. The need for two operations (harvesting cells then implanting/reconstructing) produces two sets of medical costs as well as economic costs when patients’ lives are disrupted twice for the purpose of surgery and rehabilitation. If the study design had been cartilage harvest and realignment initially, and if patients had the option of declining further surgery, I wonder how many would have undergone another open procedure for ACI? Remember the reoperation rate for “minor” problems related to ACI, such as periosteal graft hypertrophy and lysis of adhesions (which Gomoll and colleagues did not consider as treatment failure), may not be minor to patients undergoing another surgery.

Having said all that somewhat critically, in the worst cases where pan-patellar chondrosis was present and results of unloading procedures alone have been inconsistent or at least uncertain in the literature, this study contains some promising outcomes.


Gomoll AH. Am J Sports Med. 2014;doi:10.1177/0363546514523927.

Pidoriano AJ. Am J Sports Med. 1997;25(4):533-537.

Sherman SL. Am J Sports Med. 2013;42(8):2006-2017.

William R. Post, MD, is on the board of directors of the The Patellofemoral Foundation. He can be reached at Mountaineer Orthopedics Specialists, 2195 Cheat Rd., Suite 2, Morgantown WV 26508; email:

Disclosure: Post has no relevant financial disclosures.