Segmental metallic prostheses have several advantages over biologic reconstruction in limb salvage surgeries, which include initial reliability with lower complication rates, ready availability and immediate stability allowing rapid return to full weight bearing with predictable function. However, long term risks for infection, local tumor recurrence, aseptic loosening, and implant failure are significant concerns for tumor surgeons [15].
Implant-related complications are still common in Orthopedic surgery and pose a serious problem. Several innovations were developed to counter or minimize these complications. One of these is the development of an endoprosthesis better matched to Asian knees by the Japanese Musculoskeletal oncology group. Nakamura et. al. [11] reported that the average MSTS score for the KLS system was 21.8 (72.5%). The complication rate was 34% (28 out of 80 patients), infection rate 8.5% (7 out of 82) and aseptic loosening rate 6% (5 out of 82). In another study, Tome et. al. [16] investigated the clinical outcome of the KLS system for malignant bone tumors of the femur. The 3 year prosthesis survival rate was 87.5% and the 5 year survival rate was 58.3%. The mean MSTS score was 81% and no prosthesis-related complications were noted.
This implant was further improved by coating it with iodine to prevent and/or treat periprosthetic infection. Tsuchiya et.al. [17] performed a clinical trial on the efficacy and safety of the iodine coated implants. A total of 222 patients with postoperative infection or compromised states were treated using iodine coated titanium implants. The mean age was 49 and the mean follow-up was 18 months. Iodine coated implants were used in 158 patients to prevent infection and 64 patients to treat active infections. In the prevention group, 3 (1.9%) patients developed infection, however, all 3 recovered without implant removal. In the treatment group, infection was eliminated in all patients. Despite using iodine coated implants, no abnormalities related to the thyroid gland were detected and good bony ingrowth was noted.
In our series, the most common complications for both groups were infection and aseptic loosening, which are consistent with published reports [1, 5-10].
It is reported that the infection rate after primary total joint replacement ranges from 0.2% to 4% [5, 7, 18]. For revision total joint arthroplasty, the risk of infection was 10-fold higher than for patients undergoing primary total knee arthroplasty [7]. In one study, the incidence of infection after revision joint surgery was 46% [6].
In our series, the infection rate after revision total knee arthroplasty was lower, 17% (2 of 12) and only one patient required removal of the implant for re-revision. All of the single-staged reconstruction patients were infection free at latest follow-up.
In limb salvage procedures, the infection rates are higher because patients are often immunocompromised. These patients undergo long and complex reconstructive procedures, often with large areas of dead space created after removal of bone and soft tissue, and have large wounds that often lack soft tissue coverage [19]. The incidence of infection after limb salvage surgery and endoprosthesis reconstruction ranges from 1% to 36% [20].
In a multicenter study reviewing the modes of failure after using an endoprosthesis for tumor reconstruction, Henderson [1] et al. reported that infection is the most common cause of failure after surgical reconstruction with an incidence of 34%. In our series, the infection rate was 10% (2 of 20 cases). Both patients required removal of the implant and revision using an iodine-coated endoprosthesis.
The incidence of aseptic loosening after primary joint arthroplasty ranges from 19% to 31%, and for revision arthroplasty, it was reported to be 19% [6, 10]. Contributing factors for loosening include patient specific factors (e.g. activity level, body mass index and body weight, gait mechanics, etc.), surgeon specific factors (e.g. surgical approach, component positioning, soft-tissue balancing, etc.) and implant specific factors (e.g. implant design, material, bearing couples, etc.) [21]. For limb salvage surgery and reconstruction using a tumor endoprosthesis, the incidence of aseptic loosening ranges from 5% to 27% [20]. Factors contributing to aseptic loosening of tumor endoprostheses include the extended length of the prosthesis and constrained joint design, both of which create high bending stresses at the prosthesis-bone interface [1].
The incidence of aseptic loosening was 20% for both our study groups and was within the reported incidence [6, 10, 20]. All cases of aseptic loosening were noted in the cementless KLS prosthesis subgroup. Most of the causes of aseptic loosening in our series were attributed to the side plates/screw fixation of this implant; 1 of 4 in the tumor group and 2 of 3 in the arthroplasty group. The femur is not tubular over its distal aspect and the endoprosthesis side plate follows this tapered end of the distal femur. The cortical fixation by the screws, however, is not rigid enough to maintain the straight side plate compressed along the cortical taper of the distal femur and combined with a high stress area; micromotion occurs at this interface postoperatively. This usually happens before any bony in-growth occurs thus, loosening results (Figure 1).
For the tumor group, the mean MSTS scores were 63 for the proximal femur, 67 for the distal femur and 92 for the proximal tibia and it is comparable to average reported MSTS from the literature (60%-90%) [11,20]. The best results were noted in patients who underwent proximal tibia reconstruction with the worst being in the proximal femur. These results are consisted with the study done by Gosheger et.al [22]. The 2 year prosthesis survival in our study was 79%. In the literature, reported 5 year survival of the lower extremity prostheses ranges from 69% to 87% [16, 20]. Currently, it is difficult to compare our results to published reports because of our short follow-up.
In the Arthroplasty group, the post-operative mean JOA score was 61 and the 2 year prosthesis survival was 84%. At present, there are no available reports on functional outcomes of the KLS system used for non-tumor cases.
No knee instability was noted in either group on the latest follow-up. This may be attributed to the constrained design of the KLS implant. Constrained knee prostheses provide great inherent stability to the knee by means of a link between the femoral and tibial components which prevents varus-valgus and translational stresses [2].
Limitations of our study include the small number of patients studied and the limited follow-up. To more accurately determine the functional outcomes and prosthesis survival, an increased population size and longer follow-up are needed.