We report our experience of using a computer
navigation system to aid resection of malignant musculoskeletal tumours
of the pelvis and limbs and, where appropriate, their subsequent
reconstruction. We also highlight circumstances in which navigation
should be used with caution. We resected a musculoskeletal tumour from 18 patients (15 male,
three female, mean age of 30 years (13 to 75) using commercially
available computer navigation software (Orthomap 3D) and assessed
its impact on the accuracy of our surgery. Of nine pelvic tumours,
three had a biological reconstruction with extracorporeal irradiation,
four underwent endoprosthetic replacement (EPR) and two required
no bony reconstruction. There were eight tumours of the bones of
the limbs. Four diaphyseal tumours underwent biological reconstruction.
Two patients with a sarcoma of the proximal femur and two with a
sarcoma of the proximal humerus underwent extra-articular resection
and, where appropriate, EPR. One soft-tissue sarcoma of the adductor
compartment which involved the femur was resected and reconstructed
using an EPR. Computer navigation was used to aid reconstruction
in eight patients. Histological examination of the resected specimens revealed tumour-free
margins in all patients. Post-operative radiographs and CT showed
that the resection and reconstruction had been carried out as planned
in all patients where navigation was used. In two patients, computer
navigation had to be abandoned and the operation was completed under
CT and radiological control. The use of computer navigation in musculoskeletal oncology allows
accurate identification of the local anatomy and can define the
extent of the tumour and proposed resection margins. Furthermore,
it helps in reconstruction of limb length, rotation and overall
alignment after resection of an appendicular tumour. Cite this article:
The use of a navigation system in musculoskeletal tumour surgery enables the integration of pre-operative CT and MRI images to generate a precise three-dimensional anatomical model of the site and the extent of the tumour. We carried out six consecutive resections of musculoskeletal tumour in five patients using an existing commercial computer navigation system. There were three women and two men with a mean age of 41 years (24 to 47). Reconstruction was performed using a tumour prosthesis in three lesions and a vascularised fibular graft in one. No reconstruction was needed in two cases. The mean follow-up was 6.9 months (3.5 to 10). The mean duration of surgery was 28 minutes (13 to 50). Examination of the resected specimens showed clear margins in all the tumour lesions and a resection that was exactly as planned.
