The usefulness of wide excision assisted by a computer navigation system and reconstruction using a frozen bone autograft for malignant acetabular bone tumors: a report of two cases

We used OrthoMap software (Stryker, Kalamazoo, Michigan) as a computer navigation system. In the preoperative plan, the plot was based on computed tomography (CT) (case 1) or both CT and gadolinium-enhanced magnetic resonance imaging (MRI) (case 2). The CT and MRI slice thicknesses were 1.5 and 4 mm, respectively. We used only CT for case 1 because chemotherapy was extremely effective and the soft tissue mass disappeared; we used both CT and MRI for case 2 because chemotherapy was less effective and the soft tissue mass remained. We confirmed that the tumor did not grow in size 1 week before surgery, and then, we used the image acquired before chemotherapy. If the tumor grows, then, it is considered that the image should be the latest one. This process took 2–3 h. During the operation, the navigation station was located on the opposite side of the operator to eliminate any obstacles between the camera and the patient’s/instrument’s tracker. The patient’s tracker was placed on the iliac crest, and registration points were registered until the error was < 1 mm using the anterior superior iliac spine, sacroiliac joint, and exposed iliac crest as the registration points with comparisons with 3D imaging. Once we determined the osteotomy site, bones were cut using a chisel linked with navigation, a bone saw, and a threadwire saw (T-saw; Depuy AcroMed, Inc., Cleveland, OH) [39‐42]. Although determining navigation settings requires 30–60 min, doubts about determining osteotomy sites were dramatically reduced. The resection margin was pathologically confirmed using a small sample collected from a preserved host tissue.

The surgical procedure has previously been described [32]. To kill the tumor cells, rapid freezing and slow thawing were performed. In particular, the tumor-bearing bone was frozen in liquid nitrogen for 20 min, thawed at room temperature, and distilled in water for 15 min. When the tumor-bearing bone was frozen, moisture was removed, the bone marrow cavity was curettaged, and the bone was punctured in multiple places using a Kirschner wire to prevent fracture. The frozen autograft was replaced with the reconstruction. Bone graft or cement was used for mechanical support when necessary.

A 12-year-old girl presented with Ewing sarcoma of the left pelvis (PI-II). Neoadjuvant chemotherapy (4 courses of ifosfamide + etoposide and 4 courses of vincristine + doxorubicine + cyclophosphamide) was administered, and the extra-skeletal mass disappeared on MRI. Wide excision and reconstruction were performed. We planned osteotomy including the load surface of acetabulum assisted by a computer navigation system and reconstruction using a frozen bone autograft. The patient’s tracker was placed on the iliac crest; the error after registration was within 1 mm. Subsequently, by pulling the lower limbs, a gap of approximately 2 cm was made in the joint space and osteotomy was performed without dislocation of the hip joint (Fig. 2). In this case, navigation was particularly helpful in osteotomy of the acetabulum. Although osteotomy under direct view is possible via dislocation of the hip joint, it involves the risk of femoral head necrosis, and this cannot be confirmed from the cartilage surface. In addition, X-ray image alone cannot provide the orientation. Reconstruction was performed using plates and an artificial bone graft and autograft (normal iliac bone) to fill the cavity for the defect of the load surface of acetabulum after tumor curettage. The resection margins were free of tumor, and this was pathologically confirmed in a small sample collected from preserved host tissue. Pathological evaluation of curettage of cancellous bone indicated almost total necrosis of the tumor, which was classified as grade III/IV according to the Rosen and Huvos evaluation system [43]. After the completion of postoperative chemotherapy (three courses of ifosfamide + etoposide, 1 course of vincristine + doxorubicine + cyclophosphamide, and two courses of vincristine + cyclophosphamide), the patient was free of the disease. Her X-ray did not show osteoarthritis of the hip joint; she could walk normally with some claudication, and there were no limitations in the sitting posture (no limitation of the range of motion of the hip and knee joint) at her 23-month follow-up (Fig. 3). Moreover, we evaluated the outcome at the 23-month follow-up using the Musculoskeletal Tumor Society (MSTS) score [44], Toronto Extremity Salvage Score (TESS) [45], and 36-item Short-form Health Survey (SF-36) [46, 47]. Her MSTS score was 86.7 and TESS was 86.5. Additionally, her physical component summary, mental component summary, and role-social component summary scores in the SF-36 were 39.4, 59.9, and 44.1, respectively (each summary had a mean of 50 and a standard deviation of 10, according to data from 2007 for healthy Japanese individuals [47]).

A 71-year-old woman presented with dedifferentiated chondrosarcoma of the right pelvis (PII-III). The initial pathological diagnosis was osteosarcoma, and thus, neoadjuvant chemotherapy (3 courses of cisplatin + doxorubicine) was administered. The chemotherapy caused marked shrinkage and ossification of the tumor. Wide excision and reconstruction were performed. We planned osteotomy avoiding load surface of the acetabulum, in which there was no tumor, assisted by a computer navigation system. The patient’s tracker was placed on the iliac crest, and the error after registration was within 1 mm. In this case, navigation was also particularly useful for osteotomy of the acetabulum. The location and orientation of osteotomy were extremely important for preserving the load surface, but similar to other methods, it is currently impossible to precisely achieve this. Reconstruction was performed using plates and a frozen bone autograft (Fig. 4). The resection margins were free of tumor, and this was pathologically confirmed in a small sample collected from preserved host tissue. Pathological evaluation of curettage of the tumor, which was classified as grade II/IV according to the Rosen and Huvos evaluation system. The patient did not undergo postoperative chemotherapy as the final diagnosis was dedifferentiated chondrosarcoma (resistant to chemotherapy), and her physical status was limited. However, she was free of the disease her X-ray did not show osteoarthritis of the hip joint. She could walk almost normally with a cane at her 33-month follow-up. At that follow-up, her MSTS score was 63.3 and TESS was 68.8. Additionally, her physical component summary, mental component summary, and role-social component summary scores in the SF-36 were 26.0, 58.8, and 33.1, respectively.