FIRST EXPERIENCES USING A SURGICAL ROBOT FOR TOTAL KNEE REPLACEMENT

Abstract

Objectives: A surgical robot (ROBODOC®) is used for total knee replacement. The same system has been in clinical use for total hip replacement at BGU Frankfurt since 1994 and since March 2000 TKR is another clinical application. The presentation intends to give an overview of the system and of the first experiences in clinical use.

Background: The outcome of conventional total knee replacement has always been very dependent on the surgeon’s individual skills and routine. The most common mistakes have been malpositionings and malrotations of the prosthesis, which postoperatively caused varus and valgus malalignments of the lower limb resulting in an incorrect mechanical axis.

The system permits a three-dimensional pre-operative planning of the correct axis and rotation as well as the correct implant size. The introperative cutting is entirely executed by the surgical robot according to the preoperative planning.

Design / Methods: The ROBODOC® Surgical Assistant System consists of three major components: The pre-operative planning workstation called ORTHODOC®, the surgical robot and the robot control unit that receives the preoperative planning data and controls ROBODOC®. Presently, four titanium pins have to be implanted at the beginning of the procedure, one in the proximal femur, one in the distal femur, one in the proximal and one in the distal tibia. These pins are the landmarks for the following procedures.

A CT scan is made of the femoral head, the distal femur and the proximal tibia including all the pins and the ankle. A rod is laid on the patient’s leg to detect the motion during CT scan. The CT data is being transferred to the ORTHODOC® workstation on an optical disk. The ORTHODOC® displays three orthogonal cross-sections of the bone on a high-resolution screen. A manipulation on one of the cross-sections is shown in nearly real-time on the other two cross-sections.

The first step is to find all the four pins on the CT scan and to check their position. The next step is to create a femoral and tibial axis using four markers (i.e. proximal and distal femur and proximal an distal tibia). The bone is then aligned along the axis. Once those steps have been performed and implant can be selected from an implant library. The femoral component is the first part of the planning. Once the correct size, alignment and rotation have been found the tibial component is added and adjusted. The final step is to select the tibial liner.

Once the planning is finished a synthetic x-ray can be created which shows the postoperative result and helps to determine if the correct axis was planned. After finishing the planning a transfer tape that can be loaded into the ROBODOC® is created. The patient’s leg is positioned using a special leg holder and thigh support plate.

The patient’s knee should be flexed to an angle of approximately 70 to 80 degrees, a gap of 1 to 2 mm should be achieved. The patient is prepared and draped in the normal manner. Surgery proceeds normally and the regular approach for TKA is used. Once the exposure is finished and the four pins are clearly accessible, two Steinmann pins are inserted, one in the femur and one in the tibia. The Hoffmann II Orthopedic Fixation System is used to connect the two Steinmann pins and to distract the knee joint.

Now the robot is moved to the OR table. The femur and the tibia must be rigidly fixated to the robot base. Following this two bone motion monitors are attached to the bone, one to the femur and one to the tibia.

The registration program is started. Using ROBODOC®’s ball probe the four pins have to be located, including the use of pin extenders so that the robot can find the patient’s position on the OR table by comparing the data to the preoperative CT data of that particular patient. First the femoral pins are found, then the tibial pins. If the registration is correct the cutter can be installed, the irrigation system is connected and the robot starts cutting the surface for the planned implant.

First the femur is prepared, then the tibia. The final cut is the cruciform in the tibia for which a special cutter is required. Should bone motion occur during any part of the cutting procedure the robot will stop and the pins have to be re-registered. Once the cutting is finished, the robot is moved away from the OR-table, the pins and fixators are removed an the surgeon inserts the planned implant manually – normally using the cementless tecnique. The surgery is finished the traditional way.

Fifty patients who had received total knee replacement using the ROBODOC® System were followed up and examined using a pre-defined protocol: 23 patients were male, 27 were female. All cases showed severe signs of osteoarthrosis. In 35 cases we saw a varus deviation, in 13 cases a valgus deviation, in two cases there was no deviation from the axis. four patients were post-traumatic cases, in one case a complex osteotomy had been performed.

In 38 patients the cementless technique was used, in eight cases the tibial component was cemented and in four cases both components were cemented due to poor bone quality.

We had an obvious learning curve, OR time went down from 130 minutes for the first procedure to 90 minutes average OR time. Due to the three-dimensional pre-operative planning of the correct axis and rotation we saw a good alignment of the femoral and tibial component in all cases. Besides the optimal size of the components could be selected for the patients in this group.

Results/Conclusions: The system permits a three-dimensional pre-operative planning of the correct axis and rotation as well as the correct implant size. Due to the exact cut surfaces the cementless technique can be used in the makority of cases. The patients are permitted full weight bearing immediately postoperatively. None of the templates or tools that are needed for manual TKR are necessary when the ROBODOC® system is used which means an immense reduction of surgical tools for that procedure. The OR-time is not significantly longer compared to traditional TKR.

The surgeon has total control of the procedure at all times and the procedure can be finished manually if necessary. Correct aligment and rotation are the known preconditions for durability in TKR. The present disadvantages of the system are soft tissue management including ligament balancing, the rigid fixation and the use of pins (markers).

By June 2001 about 280 surgeries have been performed using the system. The development of a pinless system is already on its way and clinical testings are abour to start at BGU Frankfurt.