MACI Sandwich Technique With Autologous Bone Graft

Abstract

Background:

Osteochondral lesions of the knee are difficult to manage. Such lesions can be effectively treated with the matrix-induced autologous chondrocyte implantation (MACI) “sandwich” technique using autologous chondrocytes on porcine collagen membrane in conjuction with autologous cancellous bone grafting (ABG). Few studies have examined this technique to restore the osteochondral unit.

Indications:

The MACI “Sandwich” procedure is indicated when osteochondral lesions are larger than 2 cm² in size and deeper than 8 to 10 mm, with symptoms consistent with the location of the lesion.

Technique Description:

The chondral defect is radically debrided back to healthy bone and cartilage tissue. The base of the bony cavity is drilled to enhance the vascular supply and promote healing. Next, the bony cavity is prepared to slightly undermine the subchondral bone surface around the articular margins, with the depth of the cavity wider than the opening, similar to a dental amalgam. The autologous cancellous bone is subsequently morselized and impacted up to the level of the native subchondral bone plate. The first MACI membrane with the cells facing up is placed directly onto the bone graft site and compressed with a neural patty. The neural patty is removed, and the second MACI membrane is then placed with the cells facing down. The edges are then micro-sutured to ensure stability.

Results:

The “sandwich” technique has superior survival rates compared with autologous bone grafting alone, with patients reporting decreased pain, improved function, and high satisfaction scores over a mid-term to long-term follow-up.

Discussion/Conclusion:

The MACI “sandwich” technique is an effective surgical intervention to restore the osteochondral unit and preserve the patient's native knee joint.

Patient Consent Disclosure Statement:

The author(s) attests that consent has been obtained from any patient(s) appearing in this publication. If the individual may be identifiable, the author(s) has included a statement of release or other written form of approval from the patient(s) with this submission for publication.

Graphical Abstract

This is a visual representation of the abstract.

Video transcript

This is Dr Tom Minas. Today, I would like to talk with you about my surgical technique for the matrix-induced autologous chondrocyte implantation (MACI) sandwich technique with autologous bone grafting. This technique was developed by Drs. Lars Peterson and Deryk Jones and published in the early 2000s.³ Prior to that time, I had been staging my bone grafting followed by autologous chondrocyte implantation (ACI), but this, I think, is an ingenious technique for osteochondral defects. These are my disclosures.

So, the “Sandwich” concept is to restore the osteochondral unit—“bone graft and chondrocytes”—all with autologous tissue. I'm going to go over a case study today to demonstrate this, specifically the surgical approach, and talk briefly at the end about postoperative management and good clinical outcomes.

So, here is the problem we face, an osteochondral defect, osseous and bony; chondrocytes will not fill this on its own. And what we're going to do is restore the osseous bed with autologous bone graft and the surface, and separate the surface cells from the marrow-derived cells in the bone. The goal here is very much like doing a dental procedure, a dental amalgam. You have a shallow cavity on the margins and deep in the center, and what we need to do is make the mouth of the defect smaller than the base of it so it holds the bone graft in. So, we vascularize the cavity by drilling into healthy vascularized bone, we remove all the abnormal bone with a high-speed bur, we undermine the edges so that we can place a membrane that will separate the surface chondrocytes from the deep-derived bone marrow cells, and then we put the second membrane on the top, and that is the way we talked about a sandwich, we sandwich the cells between 2 membranes.

The indications for me for the MACI sandwich technique include osteochondral bone loss in younger patients. Osteochondritis dissecans is frequently central or minimally uncontained, and it is deep. And if somebody has had previous drilling to try to repair it or microfracture, the subchondral bone is sclerotic. If the defect is deep and the walls are relatively vertical, this is a perfect case for a sandwich technique, contained osteochondral defects. Otherwise, indications include idiopathic avascular necrosis in the middle-aged man, bone cysts, failed osteochondral allografts, and failed synthetic plugs, which, of course, are not available presently. The treatment rationale—it’s just another tool in the toolbox, autologous bone and autologous chondrocytes.

Osteochondral allografts—I do use these in my practice, but these are usually for uncontained or contained large lesions in older patients with substantial bone loss. If they are very deep, I still find that they can undergo a creeping substitution and collapse an osteochondral allograft. So, I still would prefer to use a sandwich technique in a deeper lesion.

Smaller lesions, 1 to 2.5 cm², is the right place for an osteochondral autograft or a mosaicplasty technique. But the problem here is donor site symptoms. So, anything that is too large, and you have to remember the size of the knee that in a very small patient 2.5 cm may represent a very large lesion, here again, I would probably consider either allograft or MACI.

The young man I'm going to talk about is a 22-year-old athlete, a pitcher in college. He came to us with a catching mechanical sensation in his knee, very typical of osteochondritis dissecans. He was in the middle of his career, so we discussed the possibility of repairing the defect in situ versus doing a staged reconstruction. So, the way it is presented on radiographs is typical; in the intercondylar portion of the medial femoral condyle, you could see an osseous fragment to the base. And the question is: is this fragment big enough to repair, or is it fragmented? And when you look at a magnetic resonance imaging (MRI) scan, the problem with an MRI scan is that a calcified layer of cartilage typically looks like bone, and you may open it up and find there's no bony fragment large enough to secure this with compression screws. But a computed tomography (CT) arthrogram is a wonderful test for determining fragmentation of the fragment and the surface of the cartilage. And here, we can see there's a good bony fragment, and the surface of the cartilage does not appear to be cracked or broken.

Arthroscopically, we can see that there is some fracturing of the cartilage at the intercondylar and posterior margin, but that the condyle itself is actually quite firm. And here we're getting into the fractured area, and here into the osteochondral defect itself, and it's quite firm and well attached. So, to keep him going through his baseball season, we decided to proceed with just screw fixation, which was successful, and we took a backup cartilage biopsy in the event of a reconstruction if this osteochondral defect was to break away. And thankfully, he was able to get through his college career before it did fragment, and this is what we found. There was no damage to the opposing articular surface of his tibia after his fragment broke off and locked up his knee.

So here, this is his right knee, and you can see I'm doing an incision that is somewhat lateral to expose the defect and make a relatively large flap. And the reason I do this is to preserve the anterior sensation of the saphenous nerve to the knee so that he's able to kneel and function at a high level. And here we are now opening up his medial joint space, and I'm going to come down to the anterior aspect of the tibia, as well, so that I can expose the proximal tibia to obtain a bone graft. Here you can see we're coming and doing a medial arthrotomy.

When everything is exposed, you can see his medial femoral condyle surrounding damaged cartilage, and he also has an osteophyte developing already, and his condyle is becoming flat, which is typical of a chronic osteochondritis dissecans. So, we do a radical debridement with a ring curette back to healthy, full-thickness cartilage. And then we move forward, and we get a high-speed bur, and as I'd mentioned, we try to make a dental amalgam where the inlet of the defect is smaller than the deep surface so that it will maintain and keep the bone. We bur out all of the screw holes as well so that we have vascularized bone, get a fine drill, and drill into the subchondral bone to make sure we have a good vascular supply.

Here we have a fine-tipped pencil bur, and what we are doing is undermining the articular surface so that we can place the first membrane after the bone graft is placed to the level of the cartilage, but not overlapping. And here we are now, taking our autologous bone graft from the upper end of the tibia, and this is just medial to the attachment of the patellar tendon, this is a 1-cm-wide small saw, and I'm going to just pry this open, leaving it attached distally, and that gives us excellent access to the proximal cancellous bone of the tibia through the same incision. We can then harvest that, and it's actually pretty easy to do. It's just a small osteotome, a pituitary rongeur, and a curette. And we want to have nice, healthy, spongiosum cancellous bone. And the way to make sure that it stays in the defect is to morselize it very fine and then compact it, and it's surprising how stable and secure the bone graft will become.

Here we are first filling in the screw holes and packing them after the fibrous membrane within them has been removed, and we have also penetrated the subchondral bone. The next thing will be to compact that down so that we have a really tight bony backfill of the defect and using the freer (elevator) to reposition it underneath the adjacent cartilage so that when we put our first membrane down, it will stay in place mechanically. And here we have outpouring of our membrane into the operative sterile field, and you see the bottom left corner means cells are facing upward. I put this then on top of a sterile aluminum foil with a paper template, and I'm cutting out exactly to the shape we need to position the first membrane.

So, this is Tisseel fibrin glue on top of the cancellous packed bone. And now, we are putting the membrane directly onto the bone-grafted site and using our neural patty to compress this once our freer gets the membrane circumferentially underneath the cartilage, so it doesn't blow off when the tourniquet is let down. The neural patty is applied, which is thumb pressure, and we let our tourniquet down and then remove our neural patty so that we can then put on our second membrane.

The second membrane is face down, toward the first, and here we have another paper template, and we're going to cut this precisely. Here's our second membrane, which we then take our neural patty off. We check that everything is completely dry because we don't want to mix our marrow-derived cells with our end-differentiated chondrocytes. We place layer number 2 to the surface, place the edges up against the sidewall of the cartilage, and then micro-suture it, so it is absolutely stable and secure, as we did with generations 1 and 2. We're then going to go back to our bone graft site and backfill that deficiency with cancellous allograft bone chips and then close the trapdoor with a suture through the periosteum.

Finally, a pericapsular infiltration, this is done with the tourniquet down, and it's a mixture of saline mixed with ropivacaine, Toradol, clonidine, and epinephrine.

So, the rehabilitation after MACI sandwich technique is very similar to the isolated weightbearing condyle MACI. But sometimes, you may need to protect it a bit longer or with an unloader brace. So initially, the principles are range of motion (ROM), so we don't get adhesions to the transplants, and muscle tone. And what we do here is a continuous passive motion for the first 3 weeks. Then get them on a stationary bike, a hinged knee brace until their motor control to their quadriceps comes back, and an unloader brace occasionally for enlarged defects. When we get people back to weightbearing at 10 to 12 weeks, if it's a large lesion and I want to protect it, I will put them in an unloader brace.

Heal-toe walking is important early on. So, they restore their kinematics, progress to full body weight by 10 to 12 weeks, and ROM is 90° by 3 weeks, 110° by 6 weeks, and full ROM by 12 weeks. We use e-stim early on to prevent atrophy of the quadriceps as well as help with swelling, and patellar mobilizations are really critical to preventing Hoffa fat pad adhesions and stiffness of the knee. So, blood flow restriction is a new adjunct. And we use this once postoperative swelling has resolved; it helps to maintain the muscle mass and prevent atrophy afterward.

And I tell my patients the first year is just a functional recovery so that they can walk comfortably, bicycle, swim, hike, but no impact loading. We do an MRI scan at 1 year and want to ensure that the cartilage is starting to look isotonic on the repair site compared with the native adjacent cartilage, and that the edema in the bone graft is resolving. By 18 months, we usually start hard pivoting activities if the patient has a normal-looking MRI scan and is asymptomatic with no swelling, tenderness, or pain.

Here we can see our radiograph assessment postoperatively at his first visit, you can see the area is filled up quite nicely, and going onto 6 months postoperative, where we have a nice bone graft and cartilaginous cover by MRI scan, also an excellent clinical examination and a relatively pain-free knee.

You can see the patient that I have demonstrated in this video. His 1-year MRI scan, coronal and sagittal fat-suppressed imaging, shows minimal edema in the subchondral bone with nice restoration of the convexity of the condyle and near isotonic full-thickness repair of his cartilage surface. This is what we look for before we start increasing activity and to reassure patients everything is going the way it should.

The outcomes that we've looked at, looking at repair techniques of bone graft alone and then the ACI sandwich technique, demonstrate that superiority is very obvious and visible right from the start, with excellent clinical outcomes and much-improved survivorship over time. ⁴ So, the cells do make a difference, and bone graft alone is not enough. This was also noted by Bentley and his group when they did autologous bone graft with an MACI technique, and even with second-look biopsies had a very nice cartilaginous repair tissue and only 1 failure in their group of 14 patients, with 84% good and excellent outcomes. ⁶ The second-look arthroscopies also look excellent, and here you can see this large lateral femoral condyle osteochondritis dissecans 2-year second look, with a very good repair tissue fill.

When an osteochondral lesion is first identified, such as osteochondritis dissecans, the question I have is: is this a fixable lesion? Especially in a young patient, you will often see very thick, articular cartilage on an MRI scan, and you don't know whether this is all cartilage or whether the black layer that is deep is the calcified layer of cartilage versus bone. So, will it be amenable to open reduction internal fixation or arthroscopic fixation, or do I need to do a sandwich technique? The easiest way to determine the difference between the two is to do a CT arthrogram, which will show very clearly what is bone and what is cartilage. And that allows me to determine whether this is going to be a biologic reconstruction with bone and autologous chondrocytes or a fixation with internal metal screws. Usually, I use headless screws, and I don't go back to take them out.

Bony defects deeper than 6 to 8 mm should be considered for bone grafting and cartilage repair, especially if it's a chronic defect, as mentioned before, or the walls are very steep and the bone is sclerotic, or it's undergone drilling or microfracture. So, how do we do this? We do a radical debridement of the cartilage and the subchondral bone. Take a high-speed bur, remove all the abnormal bone, and get back to healthy bleeding bone. Use autologous bone to graft the defect site. Impact it like a dental amalgam so that it is stable. Undermine the margins so that we can put fibrin glue and a layer of membrane down. Let the tourniquet down, get a dry subchondral bone bed, and then move on to a thin second layer.

People have asked me if collagen can be used in isolation for the first layer. I've never tried it. It's not being published, but collagen can possibly be used on the first layer and just one layer of cells facing down toward that. But the technique was originally devised with periosteum with the cambium layer of cells up, the second layer down, and cells in between. That was the original concept of MACI or ACI sandwich technique. So, we have tried to keep it similar here, where the deep membrane is the impervious layer to the subchondral bone or the bone graft so that we separate marrow-derived cells from the surface and keep chondrocytes at the surface. In my series, I only suture the top layer because these are large defects, on average greater than 6 to 10 cm² on the surface, so I want to make sure I can move them immediately. The only way I feel safe to do so is to use micro-sutures with 6.0 Vicryl.

I’d like to thank my research assistant Lia, video production, which I thought was outstanding, by Sue Rothberg and her company, and of course, my mentor and very good friend, Lars Peterson, and sponsorship from the Vericel Medical Team.^1-6