We retrospectively evaluated 74 consecutive patients who underwent 87 TTOs during RTKA in a single-surgeon, single-institution setting. All the procedures were performed between November of 1997 and December of 2006. There were 35 men and 39 women with an average age of 60 (29–89) years. 62 patients underwent 1 TTO, 10 patients had repeat TTO, and 1 patient had 3 TTOs in the same knee (). 1 patient also had bilateral TTO.
Figure 1. 3 TTOs on the same knee. A. Healed primary TTO in a 57-year-old man who underwent RTKA of the left knee. A bicortical screw proximally and 2 wires distally were used for osteotomy fixation. B. 4 years later, implant loosening and instability developed, (more ...)
Osteoarthritis (54 knees) was the most common indication for the primary total knee arthroplasty (TKA), followed by posttraumatic arthritis (10 knees), rheumatoid arthritis (6 knees), hemophilic arthropathy (3 knees), septic arthritis (1 knee), and osteosarcoma of the distal femur (1 knee).
The reason for RTKA with TTO was staged treatment of infection in 38 knees, stiffness and arthrofibrosis in 21 knees, aseptic loosening in 15 knees, instability in 6 knees, patellar maltracking in 2 knees, polyethylene wear-through in 2 knees, femoral component malrotation in 1 knee, heterotopic ossification in 1 knee, and tibiofemoral dislocation in 1 knee.
Preoperative lag of extension and flexion contracture were present in 23 and 30 knees, respectively. In 8 knees, a rectus snip (6 knees) or a V-Y turndown procedure (2 knees) had been performed earlier during another RTKA. Overall, the average number of RTKAs per patient was 1.5 (1–7).
Patients were scheduled to be evaluated clinically and radiographically preoperatively and postoperatively at 6 weeks, 3 months, 6 months, and annually thereafter—or at additional time intervals if residual symptoms and radiographic findings necessitated further examination. The average follow-up was 49 (6–108) months. None of the patients were lost to follow-up before healing of the osteotomy had occurred. Anteroposterior, lateral, and patellar radiographs were obtained from all the patients for the evaluation of implant position, patellar tracking, and status of the TTO. The measurement of length and width of osteotomy was done electronically with digitized lateral radiographs using imaging software. The osteotomy was classified as extramedullary when the bone cut was through the metaphyseal cancellous bone of the anterior tibia, or intramedullary when it extended more deeply from the inner surface of the tibial tubercle (). The osteotomized bone fragment was considered to be healed to the host tibia when radiographic evidence of bridging callus formation was observed on the lateral radiograph.
Figure 2. Intramedullary and extramedullary TTO A. Diagram illustrating a TTO elevated in a medial to lateral fashion. The anterior compartment muscles as well as the patellar tendon and patella are attached to the osteotomized bone fragment. The osteotomy is performed (more ...)
The procedure was performed through a medial parapatellar arthrotomy and the medial proximal tibia was dissected subperiostally to facilitate tibial external rotation and lateral patellar subluxation. Retropatellar adhesions were released and scar tissue dissected from the medial and lateral gutters. If patellar subluxation was associated with excess tension in the extensor mechanism and there was risk of patellar tendon avulsion, a TTO was carried out. The tibial tubercle together with a segment of anterior tibial crest was elevated in a medial to lateral direction. The bone cut was initiated with a thin oscillating saw and completed with 2 broad osteotomes, leaving a long tibial bone fragment for late repair. In comparison with the previously described technique of Whiteside and Ohl (1990) in which the distal end of the bone cut was made in a transverse manner, the osteotomy was distally tapered to avoid a stress riser in the anterior tibial cortex. Similarly, no step-cut was done proximally and the osteotomy was extended to the knee articular surface. The osteotomized bone segment, along with the attached anterior compartment muscles and patellar tendon, was then hinged laterally to expose the knee.
At the completion of the RTKA, the tibial tubercle was reduced in its anatomic position or alternatively displaced up to 1 cm medially and/or proximally to achieve optimal quadriceps tension, knee flexion, and alignment of the extensor mechanism. Lateral release was performed routinely to improve or correct patellar alignment. Osteotomy repair was performed with bicortical screws (9 knees), Luque wires (16 knees), 1 screw and wires (52 knees) (Figure ), or 2 screws and wires (7 knees).
In infected knee arthroplasty, a 2-stage procedure was followed with a time interval of 6–8 weeks. The TTO was made either during reimplantation of the components (second stage, 33 knees) or at the time of implant removal and introduction of cement-spacer (first stage, 5 knees). In the latter group, the bone fragment with the muscular attachments remained unfixed (Figure ) until the second stage of RTKA (Figure ).
Postoperatively, no weight bearing or range of motion restrictions were applied and knee flexion exercises were started on the day after surgery.
Statistical evaluation was carried out with the the SPSS software package vesrion 16.0. Since data histograms showed skewed distribution of the variables, nonparametric methods of analysis were chosen. Data are presented as median and range. Any differences in union time between extramedullary and intramedullary or first and repeat osteotomy groups were examined using the Mann-Whitney rank-sum test. The Kruskal-Wallis statistic was calculated to test for a dependency between fixation technique and union time. The changes in knee range of motion, extensor lag, and flexion contracture before and after surgery were evaluated with the Wilcoxon signed-rank test. Statistical significance was assumed for a p-value of < 0.05 (or determined with use of a 95% confidence interval).