It is well known that high-dose corticosteroid administration can cause osteonecrosis1,2. Compared with spontaneous osteonecrosis of the knee3, corticosteroid-induced osteonecrosis of the knee tends to be more extensive and often occurs in younger patients.
Recently, more attention has been given to regenerative tissue-engineering techniques that make use of autologous cells combined with implants to restore lost or dysfunctional tissues or organs4,5. We have had successful short-term results with use of tissue-engineered cartilage in the treatment of osteochondral defects of the knee6,7.
Tissue-engineering techniques, such as the implantation of calcium hydroxyapatite ceramic together with mesenchymal stem cells8-12, have also been applied to bone regeneration. Recently, a new type of hydroxyapatite with interconnected pores (IP-CHA) was introduced for use as a scaffold for bone regeneration13.
We present the case of a patient in whom steroid-induced osteonecrosis of the distal part of the femur was treated by implantation of tissue-engineered cartilage together with IP-CHA hybridized with cultured bone-marrow cells. The patient was informed that data concerning the case would be submitted for publication, and she consented.
A thirty-six-year-old female office worker had received corticosteroid therapy for the treatment of a connective-tissue disease since 1988. Oral prednisolone had been administered at an initial dosage of 60 mg per day and then gradually reduced. In 1991, the patient began to feel piercing pain in her left knee while walking. In 1993, osteonecrosis of the medial and lateral femoral condyles of the left knee was diagnosed. The knee pain gradually increased despite several nonoperative treatments, and in January 2004 she presented with severe pain in the left knee. In addition, she reported having a clicking sensation in the same knee, which was especially noticeable while walking.
Plain radiographs, made at the time of the first consultation, revealed radiolucent areas surrounded by sclerotic areas in the medial and lateral femoral condyles (Fig. 1, A). Magnetic resonance imaging also revealed large osteonecrotic areas in both the medial and the lateral femoral condyle (Fig. 1, B).
After approval was obtained from the ethical committee of our university and the patient was provided with detailed information about the various conventional treatment options, an arthroscopy was performed in June 2004. Large areas of osteochondral damage were seen on the medial and lateral femoral condyles. The lesion on the medial femoral condyle was located in the weight-bearing area and was 25 mm in length and 20 mm in width. The cartilage covering this lesion was almost detached from the subchondral bone, which was hard and sclerotic. The lesion on the lateral femoral condyle was also located in the weight-bearing area and was 20 mm in length and 15 mm in width, and the subchondral bone was exposed.
We harvested approximately 300 mg of cartilage from unloaded areas of the medial and lateral femoral condyles of the ipsilateral knee joint during the arthroscopy. Through enzymatic digestion and in accordance with previously described procedures6,7, we obtained 2.0 × 106 chondrocytes. The isolated chondrocytes were embedded in atelocollagen solution (3% type-I collagen; Koken, Tokyo, Japan) and cultured. After twenty-five days of culture, the atelocollagen gels containing chondrocytes became opaque and acquired a jelly-like consistency. Since the number of chondrocytes in the tissue-engineered cartilage could have increased during cultivation, we could not determine the exact cell numbers at the end of the culture period.
Bone-marrow aspirates (20 mL) were obtained from the left iliac crest of our patient, and bone-marrow cells were harvested and expanded in a monolayer culture system. At the end of the primary culture, adherent colonies were detached by treatment with 0.25% trypsin and 0.02% ethylenediaminetetraacetic acid. Although we refer to these cells as mesenchymal stem cells, it is important to note that we did not evaluate their stem-cell potential and thus they may have been comprised of a heterogeneous cell population. After twenty-five days of culture, the total number of mesenchymal stem cells was 1.2 × 105 (Fig. 2).
After arthrotomy, the lesion of the medial femoral condyle was exposed (Fig. 3, A). To remove the necrotic bone, three osseous sockets were made in the center of the lesion with use of the mosaicplasty system14. Two of the sockets were 8.5 mm in diameter and 12 mm in depth, and one socket was 6.5 mm in diameter and 12 mm in depth. Cylindrical blocks of IP-CHA of the same diameter were placed into the sockets (Fig. 3, B). The cultured mesenchymal stem cells, suspended in 1.5 mL of saline solution, were then infiltrated into the IP-CHA with use of a syringe, and fibrin glue (Bolheal; Kaketsuken, Kumamoto, Japan) that was mixed according to the manufacturer's instructions was sprayed onto the surface to prevent leakage of the mesenchymal stem cells. Next, the tissue-engineered cartilage was placed in the defect, and a synovial flap that was harvested from the suprapatellar pouch was sutured to cover the lesion (Figs. 3, C and D).
To treat the osteonecrosis of the lateral femoral condyle, two IP-CHA implants (6.5 mm in diameter and 12 mm in length) without tissue-engineered cartilage or mesenchymal stem cells were placed in the lesion with use of the same technique as described above.
After the knee joint had been immobilized for two weeks with a soft brace, range-of-motion exercises were started with use of a continuous passive motion device. Partial weight-bearing was allowed at five weeks after surgery, and full weight-bearing was permitted at eight weeks after surgery. The patient reported that the pain and clicking in the left knee disappeared soon after the operation.
We performed an arthroscopy at one year after the operation to assess the quality of cartilage repair. The patient provided informed consent for this procedure. At the time of arthroscopy, the medial femoral lesion was covered with smooth cartilage-like tissue and no osseous defects were apparent (Fig. 4, A). The grafted IP-CHA implants in the lateral femoral condyle were exposed intra-articularly and were noted to be covered with sparse fibrous tissue (Fig. 4, B). A 2.0-mm needle biopsy of the medial femoral condyle was performed. The biopsy tissue was fixed with 10% buffered formalin for one day and then stained with safranin O-fast green. The biopsy specimen from the medial femoral condyle consisted of a mixture of fibrous and cartilage-like tissue, the extracellular matrix of which showed staining with safranin O (Fig. 5). However, the intensity of staining with safranin O was much weaker than that which would be seen in normal cartilage, indicating that the regenerated tissue seemed to be deficient in proteoglycans. Although bone regeneration was observed in the IP-CHA implant, the cartilage-like tissue and IP-CHA separated during the staining procedures, indicating very fragile integration between them. Magnetic resonance imaging scans acquired two years after implantation showed that the medial femoral condyle was covered with a smooth and congruous tissue (Figs. 6, A, B, and C). When last seen at two years after the surgery, the patient was very satisfied with the results, especially since she had been able to return to her previous work.
In this case report, we describe a patient in whom steroid-induced osteonecrosis of the distal part of the femur was treated with implantation of tissue-engineered cartilage and hydroxyapatite hybridized with cultured bone-marrow cells.
Because the patient had severe tenderness, pain, and clicking in the medial compartment of the knee joint, we treated the lesions of the medial and lateral femoral condyles differently, suspecting that the most problematic lesion of the joint would be the osteonecrotic lesion of the medial femoral condyle. Because this was the first case to our knowledge in which this treatment was performed, we did not know the appropriate numbers of cells to use for both the tissue-engineered cartilage and the bone lesion. We transplanted all of the cultured cartilage cells to the lesion of the medial femoral condyle to focus on the most problematic lesion, and we did not transplant cells to the lesion in the lateral femoral condyle.
There are several treatment methods for large osteochondral defects. Although autologous chondrocyte implantation can repair cartilage lesions associated with small osseous defects, it cannot repair an underlying large osseous defect6,7. Autologous osteochondral transplantation or an osteochondral allograft may be indicated for large full-thickness articular defects14. Because this patient had a mixed connective-tissue disease, her own cartilage and subchondral bone might not have been normal or suitable for transplantation. We therefore decided to perform this new procedure, which is a possible treatment option for patients with steroid-induced osteonecrosis of the distal part of the femur. It is important to note that cartilage-like tissue regeneration and bone regeneration were achieved simultaneously in the medial femoral condyle of our patient even though we harvested only a small amount of cartilage and bone marrow. Further studies will be necessary to determine the size range of lesions that may be treated with this new procedure.
Although our result was good clinically and was confirmed by second-look arthroscopy, there are some limitations in this case report. First, the follow-up period for this patient was too short to evaluate the long-term survival of the regenerated tissue. Long-term follow-up of this patient will be particularly important because histological examination of the biopsy specimen showed that the regenerated tissue was not true hyaline cartilage and that the cartilage-like tissue and the IP-CHA implant were not well integrated as a unit. A second limitation is that we did not evaluate the stem-cell potential of the adherent cells from the bone marrow, which we referred to as mesenchymal stem cells. These cells probably consisted of a heterogeneous cell population that had varying potential to differentiate along several cell lines. It will be necessary to purify the cells to enable better tissue regeneration in the future, although the isolation procedure used in this study was very straightforward and clinically feasible15. Another potential limitation was that the lesion of the medial femoral condyle was covered with a synovial flap while the lesion of the lateral femoral condyle was not covered. Therefore, there was a possibility that the synovial tissue contributed to the tissue regeneration over the medial femoral condyle. Further studies that clarify the role of the transplanted synovial flap will be necessary to address this issue.
In conclusion, although further studies on a greater number of patients will be necessary to investigate the long-term efficacy of this procedure, we believe that the implantation of tissue-engineered cartilage and a hydroxyapatite implant hybridized with cultured bone-marrow derived cells has potential for future use in the treatment of large complex osteochondral defects. 