We describe a method of reconstruction using tumour-bearing autograft treated by liquid nitrogen in 28 patients. The operative technique consisted of en bloc excision of the tumour, removal of soft tissue, curettage of the tumour, drilling and preparation for internal fixation or prosthetic replacement before incubation for 20 minutes in liquid nitrogen, thawing at room temperature for 15 minutes, thawing in distilled water for ten minutes, and internal fixation with an intramedullary nail, plate or composite use of prosthetic replacement. Bone graft or cement was used to augment bone strength when necessary.
The limb function was rated as excellent in 20 patients (71.4%), good in three (10.7%), fair in three (10.7%), and poor in two (7.1%). At the final follow-up six patients had died at a mean of 19.8 months after the operation, while 21 remained free from disease with a mean follow-up of 28.1 months (10 to 54). One patient is alive with disease. Bony union was seen at a mean of 6.7 months after the operation in 26 patients. Complications were encountered in seven patients, including three deep infections, two fractures, and two local recurrences. All were managed successfully. Our results suggest that this is a simple and effective method of biological reconstruction.
Advances in diagnostic imaging, neoadjuvant chemotherapy, and operative technique have made it possible to treat malignant tumours of bone and soft tissue by limb salvage. With multidisciplinary treatment, such techniques produce a functional and durable limb without reducing the long-term results. All the established methods of reconstruction such as the use of massive prostheses, allografts, combinations of allografts and prostheses or with bone cement have made limb salvage possible. Endoprosthetic replacement after excision of the tumour can provide excellent results more quickly than with other methods. In a large series1 the probability of a patient avoiding aseptic loosening for five years was 93.8% for a proximal femoral replacement, 67.4% for a distal femoral prosthesis and 58% for a proximal tibial implant. The survival rates for reconstructions around the knee now exceed 85% at five years.2 Survival at ten years after massive prosthetic replacement of the distal femur is approximately 50%3 while that of prostheses in the proximal humerus with mechanical failure as the end-point is 86.5% at 20 years.4 Development of extendible prostheses now allows their use in growing children.5–,7
Biological reconstruction may employ either living or dead bone. Recently, epiphyseal preservation and reconstruction with distraction osteogenesis have provided excellent function of the limb in selected cases.8–,10 Anatomical remodelling of the hip reconstructed with a massive allograft combined with a vascularised fibular transplant has been achieved in a child.11 Allografts are an example of biological reconstruction utilising dead bone. Mankin12 found that 77% of the allografts were still functional and competent. The best results were obtained with intercalary grafts while the poorest were with allograft arthrodesis. Fracture and nonunion reduce the rate of success. The addition of intramedullary cement to large-segment allografts improves their survival by decreasing the risk of fracture.13 Allograft prosthetic composite arthroplasty has also been used to solve the problem of degenerative changes occurring in osteoarticular allografts.14
Allograft is difficult to obtain in some Asian countries, especially in Japan, for socio-religious reasons. Therefore, recycling of bone has been widely used. Several methods have been developed to re-use the resected bone for reconstruction, including irradiation,15,16 autoclaving17,18 and pasteurisation.19,20 These methods require special equipment and strict thermal control. Heat treatment causes weakness of the bone and loss of the capacity for bone induction.21 We have developed a new method of treating autografts, based on in vitro and in vivo experiments,22 which uses the hypothermic effect of liquid nitrogen and which is a more useful recycling system.
Patients and Methods
We treated 28 patients, 17 men and 11 women with a mean age of 31.1 years (10 to 68) (Table I⇓). Informed consent and a letter of acceptance had been obtained from all patients. The standard pre-operative examinations included a history, clinical examination, radiography including the chest, CT, technetium-99m bone scintigraphy, thallium-201 scintigraphy, MRI angiography and routine laboratory tests. The pre-operative K2 chemotherapy protocol which consists of five courses of intra-arterial cis-platin, caffeine, and doxorubicin at intervals of three weeks23 had been used in 24 patients with high-grade sarcoma. At operation the tumour was excised en bloc, the soft tissue was removed, the tumour curetted and the excised bone prepared for internal fixation or prosthetic replacement before freezing. The excised portion was frozen in liquid nitrogen for 20 minutes, thawed at room temperature for 15 minutes, thawed in distilled water for ten minutes and then replaced with reconstruction by an intramedullary nail, plate or composite use of a prosthetic replacement. Bone graft or cement was used for mechanical support when necessary. Intravenous chemotherapy was continued during the post-operative period using cis-platin, caffeine and doxorubicin and/or high-dose methotrexate combined with citrovorum factor and vincristine (three courses each). The affected limb was assessed according to the functional evaluation system of Enneking.24
Details of the results are given in Table I⇑. At the final follow-up, six patients had died at a mean of 19.8 months after the operation at 5, 14, 32, 38, 14, and 16 months, respectively, while 21 remained free from disease at a mean follow-up of 28.1 months (10 to 54). One patient is alive with disease. Bony union was defined as complete cortical bridging in a long bone, or complete filling of the gap in the pelvis and sacrum. Bony union was seen at a mean of 6.7 months after the operation in 26 patients (92.8%). Non-union occurred in two. One of these died before bony union and in the other, the treated autograft was removed because it became infected before union. Function of the limb was rated as excellent in 20 patients (71.4%), good in three (10.7%), fair in three (10.7%) and poor in two (7.1%). Complications were encountered in seven patients, including three deep infections (10.7%), two fractures (7.1%), and two local recurrences (7.1%), all of which were managed successfully. One case of infection was controlled by partial resection of the autograft, one was treated by removal of the autograft, and the third by debridement and irrigation. Local recurrence arising from soft tissue in two patients was treated by additional wide excision. Two fractures were managed by internal fixation. We classified the reconstruction method into three types; Type 1-A (n = 14) was defined as intra-articular excision and reconstruction (Figs 1⇓ and 2⇓) and type 1-B (n = 9) as intercalary excision and reconstruction. Type 2 (n = 2) constituted excision and reconstruction of the whole joint (Figs 3⇓⇓ to 5⇓) and type 3 (n = 3) comprised in situ freezing of the osteoarticular or intercalary lesion with or without osteotomy (Figs 6⇓ and 7⇓).
Cryosurgery was first used in the management of bone tumours at the Memorial Sloan-Kettering Cancer Center in the United States in 1964 as a palliative procedure on a patient with a metastasis to the humerus from the lung.25,26 The use of liquid nitrogen for management of the primary lesion in osteosarcoma, was first described in 1984 by Marcove et al27 Repetitive freezing and thawing destroyed any tumour cells present at the margin of the curettage. The lesion was curetted, the cavity was frozen with liquid nitrogen, and then filled with cement. Immediate histological studies and those carried out at a second-look procedure showed no evidence of residual tumour. An en bloc excision of the tumour was not performed at that time.
Cryosurgery destroys tissue selectively by the controlled use of alternating freezing and thawing. Another possible cause of cell death during cryosurgery is ischaemic infarction due to thrombosis of the microcirculation.28 Cryosurgery is usually used with adjunctive treatment such as chemotherapy, immunotherapy and conventional surgery. In our basic research, human osteosarcoma tissue, cultivated in athymic mice, was placed in a cavity created in the cortex of a metatarsal bone of a Holstein cow, the shape of which is similar to that of the human tibia. The metatarsal bone was then soaked in liquid nitrogen for 20 minutes. Afterwards, the tumour tissue was implanted into the back of athymic mice. No regrowth of tumour was seen.22 Cell structures are destroyed twice during freezing and thawing. Usually, tumour tissues and cell cultures need to be frozen with an anti-freezing agent such as dimethyl sulphoxide to grow again after being thawed. Our freezing procedure using one freezing cycle of 20 minutes was safe in this clinical study, although there were two patients with local recurrence. This occurred in the soft-tissue and probably represented satellite lesions from insufficient surgical resection.
The advantages of reconstruction using tumour-bearing massive frozen autograft treated by liquid nitrogen are simplicity, osteoinduction, osteoconduction, a short treatment time, preservation of the cartilage matrix, a perfect fit, sufficient biomechanical strength, no infection, no need for a bone bank, easy attachment of tendons and ligaments and desirable bone stock. The disadvantages are degeneration of the cartilage over time, the impossibility of histological analysis of the whole specimen and related complications similar to allograft implantation. Biomechanical testing showed no significant difference in compression strength between intact bone and the bone treated by liquid nitrogen, whereas in autoclaved bone the strength was decreased.22 The functional results are comparable with other methods of reconstruction, and once incorporated by the host, frozen autografts offer the advantage of incorporation as bone stock and with soft-tissue attachments, unlike metal implants. For osteo-induction the activities of proteins and enzymes are preserved in bones treated in liquid nitrogen.29,30 Freezing with liquid nitrogen is widely used to measure and analyse tumour and healthy tissues in biomedical research. Garusi et al31 found that bone graft frozen by liquid nitrogen acts as a normal graft in rats. Our results suggest that tumour-bearing autograft treated by liquid nitrogen is a simple and effective option for biological reconstructions. It is best used for osteoblastic tumours while prosthetic or allograft reconstruction should be used for osteolytic lesions.
The biomechanical strength of the frozen autograft may be weaker than that of an allograft depending on the extent of destruction by the tumour but it is an excellent option for pelvic and sacral lesions since allograft matching and reconstruction are very difficult in these patients. Type-3 reconstruction keeps the continuity of the joint intact, giving excellent stability and function since no important ligaments are sacrificed. We are developing equipment for freezing in situ to shorten the operating time, to maintain stable hypothermia and to reduce unnecessary damage to surrounding normal tissues. Type-1 reconstruction requires the cutting of ligaments and the resutured tendon and ligament may not function well after freezing. Type-2 reconstruction does not require resuture of joint structures, but does need excision of the joint. Therefore, type-3 reconstruction is recommended if feasible.
Allografts used to reconstruct the bony defects after resection of a tumour offer many advantages, including reconstruction of the joint and incorporation of the graft to the host bone. However, the high incidence of complications makes the outcome unpredictable.32–,34 Fracture is one of the common complications and is thought to result from revascularisation of the allograft cortex. Chemotherapy increases the rate of fracture at the allograft junction.35 This phenomenon may be due to an allogenic immune response.36 Other complications include graft resorption, recurrence, and nonunion. Frozen autografts contain autogenous proteins, growth factors and cytokines, and do not elicit an immune reaction. They have the advantages of early bony union and low risk of bone resorption although they may also have some complications similar to allografts, such as infection, fracture, nonunion, or failure of the graft resulting from the use of dead bone.
Another important aspect of treatment by liquid nitrogen is cryoimmunology. It is possible that tissue proteins released from the frozen lesions have antigenic properties which initiate an immune response directed against the tumour. There have been reports that metastatic tumours have regressed after freezing of the primary tumour.37,38 Cryoablation of tumour tissue induces inhibition of secondary growth of the tumour and causes release of cytokines.39,40 Therefore, tumour-bearing massive frozen autograft may play a role in reducing local recurrence and lung metastasis by its cryoimmunological function.
Cartilage frozen by liquid nitrogen will progress to osteo-arthritic change in time as is seen in osteochondral allografts.41 Resurfacing total knee arthroplasty may be necessary for some patients in the future. Nevertheless, as bioengineering evolves, the ability to restore or repair cartilage may become a practical proposition. Recovery of chondrocytes has been observed in cryopreserved porcine articular cartilage by drilling a hole through the subchondral bone to the base of the cartilage.42
Reconstruction with tumour-bearing massive frozen autograft treated by liquid nitrogen in malignant bone and soft-tissue tumours is a simple and effective method of biological reconstruction. Long-term follow-up studies will provide more useful information and clarify the position.
No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.
- Received January 20, 2004.
- Accepted May 27, 2004.
- © 2005 British Editorial Society of Bone and Joint Surgery