There have been conflicting results reported in the

Transcription

1 Bulletin Hospital for Joint Diseases Volume 61, Numbers 1 & Total Knee Replacement Following High Tibial Osteotomy Sanjeev Madan, M.Ch.Orth., F.R.C.S. (Orth), M.Sc.(Orth), M.B.A., R. K. Ranjith, F.R.C.S., and Nicholas J. Fiddian, F.R.C.S. Abstract There have been conflicting reports in the literature regarding the results of post high tibial osteotomy knee arthroplasty. This study aims to assess the clinical and radiological results as well as patient satisfaction following post high tibial osteotomy knee arthroplasty and to identify the variables affecting the results. A total of 29 cases of post osteotomy knee arthroplasties were retrospectively analyzed. Preoperative and postoperative range of movement and deformity were recorded with knee scores at latest follow-up. Weightbearing radiographs were taken preoperatively and at latest follow-up. Patients were asked if they were satisfied with their results. Average follow-up was 7.5 years with an average interval between osteotomy and knee arthroplasty of 4.7 years. Average age at osteotomy was 62.5 years. Seventeen patients had Maquet s dome osteotomy and 12 had closing wedge osteotomy. The average knee score was Overall 5 (17.2%) patients had a poor result. There was a significantly greater pre-osteotomy subluxation (9 mm) in the failed knees compared to the successful arthroplasties (3.64 mm) with p = using the Mann-Whitney U test. This has to be tempered by the fact that numbers were small. Failed arthroplasties had a valgus cut of the tibia (1.5 ) compared to either a neutral or slight varus cut in successful arthroplasties (average 2.42 varus). Lateral translation of tibial tray was greater in the failed knees whereas posterior translation was greater in the successful knees. However, differences in tibial tray angle and translation were not statistically significant. Four (21.05%) cases went on to revision of their arthroplasty in Sanjeev Madan, M.Ch.Orth., F.R.C.S.(Orth), M.Sc.(Orth), M.B.A., R. K. Ranjith, F.R.C.S., and Nicholas J. Fiddian, F.R.C.S., are in the Department of Orthopaedic Surgery, The Royal Bournemouth Hospital, Castle Lane East, Bournemouth, England. Reprint requests: Sanjeev Madan, Department of Orthopaedics, Sheffield Children s Hospital, Sheffield S10 2TH, England. less than 8 years, 2 for painful valgus instability. This study concludes that post high tibial osteotomy knee arthroplasty is a technically demanding operation with altered anatomy and significant balancing problems. Results of knee arthroplasty after high tibial osteotomy are inferior to that of primary arthroplasties and the amount of lateral subluxation prior to surgery may adversely affect results. There have been conflicting results reported in the literature concerning the results of total knee replacement (TKR) done after previous high tibial osteotomy (HTO). Cameron and colleagues 1 reported 17.5% poor results in 131 knee arthroplasties done after failed high tibial osteotomies at 2 to 15 years follow-up and an overall complication rate of 11.2%. These findings were echoed by Insall and associates 2 who reported poor results in 14% of cases in a 3.6 year follow-up of 22 knees. However, Takai and coworkers, 3 in a 6 year follow-up of 12 knees, found results comparable to a control group of 12 primary knee arthroplasties. Toksvig-Larsen and colleagues, 4 in a comparison of 40 matched knees, found no difference between primary and post tibial osteotomy knee arthroplasties after a 3-month, 6-month, 1-year, 2-year, 5-year and 10-year follow-up. However these authors were only able to provide data for only 5 of their patients at the 10-year interval. This study aims to assess the clinical and radiological results as well as patient satisfaction following knee arthroplasties for failed tibial osteotomy. We have tried to identify the various variables affecting our results. Materials and Methods A total of 72 high tibial osteotomies were carried out at the Royal Bournemouth Hospital between July 1976 and February 1990, of which 29 were converted to total knee arthroplasties and these were retrospectively analyzed. We

2 6 Bulletin Hospital for Joint Diseases Volume 61, Numbers 1 & Table 1 Clinical and Radiographic Measurements Failed Successful All Dome Wedge TKRs TKRs TKRs Age at HTO Time to TKR ROM Post HTO ROM Post TKR PRE HTO Alignment POST HTO Alignment POST TKR Alignment Current Knee Score FFD Post HTO FFD Post TKR Subluxation Pre HTO Subluxation Post HTO Tibial Tray Angle Lateral Translation Posterior Translation Percentage of 11.8% 25% 17.2% Unhappy Patients Negative values in alignment refer to varus and positive values refer to valgus. recorded the preoperative and postoperative deformity and range of movement. Weightbearing radiographs were taken preoperatively and at latest follow-up. Alignment was read off the radiographs, preoperative stage of osteoarthritis was recorded according to Ahlback s classification, 5 and lateral subluxation of the tibia was recorded in millimeters. The lateral subluxation was a radiographic finding measuring the uncovered part of the tibia laterally on standing x-rays. Postoperative x-rays were examined for implant alignment, angle of tibial cut, patellar height, tibial tray translation, and any lucency or sign of failure. American Knee Society scores were recorded at the latest follow-up. In addition, patients were asked whether they were satisfied with their operation and if it had relieved their symptoms. The correlation between the outcome with the method of osteotomy (i.e., Maquet s dome or closing wedge procedure) was assessed. The operating surgeons were not involved in the clinical assessment. The outcome of TKR following dome HTO was compared with that after wedge TKR using the Fisher s exact test. The continuous data was compared using the Mann- Whitney U test for these two groups of patients. The average follow-up after total knee arthroplasty was 7.5 years (range: 5.2 to 10.3 years) and the average interval to knee arthroplasty after tibial osteotomy was 4.7 years. The average age at tibial osteotomy was 62.5 years (range: 42 to 70 years). Seventeen patients had Maquet s dome osteotomy and 12 had closing wedge osteotomy. During the period under review all dome osteotomies were performed by one surgeon and the wedge osteotomies by another surgeon in the same institution. One patient, who had a Maquet s dome osteotomy, underwent a subsequent revision to a closing wedge osteotomy for inadequate correction before being further revised to a knee arthroplasty. The average number of previous operations patients had before having their osteotomies was 0.8. Average range of movement was (range: 85 to 110 ) and the knee score was (range: 86 to 146) at the latest follow-up (Table 1). Twenty-nine patients had total knee arthroplasty because of persistent pain even after high tibial osteotomy. Results Of the 29 patients in our study that had TKR, five patients (17.2%) had an unsatisfactory result from their arthroplasties. Two TKR patients (11.8%) following dome osteotomy and three TKR patients (25%) following closing wedge osteotomy had unsatisfactory outcome. Four knee arthroplasties have since been revised (Fig. 1) and a fifth patient had lateral laxity and instability of 12 degrees on varus strain with no pain at the 6-year follow-up (Fig. 2). This patient is satisfied with his current level of function and does not wish to consider further surgery at this point. Complications There were a total of 7 complications; 4 patients had valgus instability; 1 patient had multidirectional instability; 1 patient had elevation of the joint line by 8 mm; and 1 patient had cellulitis that responded to antibiotics. Four patients in our series underwent revision surgery. The problems encountered at knee arthroplasty after a high tibial osteotomy included significant soft tissue imbalance, tibial bone loss requiring tibial augmentation wedges in 2 cases, altered slope of the tibial plateau as well as translational deformity of proximal tibia (Table 2). These difficulties led to a compromised result with the arthro-

3 Bulletin Hospital for Joint Diseases Volume 61, Numbers 1 & Figure 1 Constrained total knee replacement for soft tissue laxity after revision of post high tibial osteotomy total knee replacement. plasty. The patellar height as measured by the ratio LP/LT was, however, not significantly compromised in our series and was on average 0.98 (range: 0.76 to 1.23) post tibial osteotomy and 1.0 post knee arthroplasty. Although the results were apparently better with post dome knee arthroplasty with only 11.8% of patients of dome osteotomy unhappy compared to 25% of patients who underwent wedge osteotomy, this difference was not statistically significant. However because of the small sample size there is the possibility of a type 2 error. The posterior translation of the tibial tray appeared to be significantly increased in post dome cases while lateral translation of tibial tray appeared to be increased in post wedge cases (Table 2). The tibial tray also tended to be in a greater degree of varus post dome osteotomy (Fig. 2). There was no significant difference in knee scores between the two groups. Overall a significant number (17.2%) were dissatisfied with their knee arthroplasties with 4 (13.8%) going on to a revision arthroplasty in less than 8 years. Two revision arthroplasties were for painful valgus instability, one had a 7 valgus cut on the tibia and the other had a 6 valgus cut on the tibia. One of these had patellar resurfacing with lateral release at 3 years 9 months postoperatively followed by full revision 6 years and 10 months later. One patient underwent revision within 7 months for painful multidirectional instability. A fourth patient had persistent pain with 6 varus alignment. This patient had a 4 varus cut on the tibia. A fifth patient with valgus instability but no pain has not yet been revised. Of the 29 patients who went on to a knee arthroplasty, 14 were dissatisfied with their tibial osteotomy from the outset and 15 had a good result initially. After knee arthroplasty the 4 that were revised had a less than satisfactory result right from the outset. There was a significantly greater pre-osteotomy lateral tibial subluxation at 9 mm in the failed knees compared to the successful arthroplasties at 3.64 mm with p = using the Mann-Whitney U test (Fig. 3). This has to be tempered by the fact that the number of patients in this series is small. Failed arthroplasties had a valgus cut of the tibia (average: 1.5 ) compared to either a neutral or slight varus cut in successful arthroplasties (average: 2.42 varus) as depicted in Figure 4. Lateral translation of tibial tray was greater in the failed Figure 2 Opening of the lateral joint space after total knee replacement.

4 8 Bulletin Hospital for Joint Diseases Volume 61, Numbers 1 & Figure 3 Lateral subluxation of tibial plateau in millimeters. knees, whereas posterior translation was greater in the successful knees. However, differences in tibial tray angle and translation were not statistically significant. Table 2 Problems Faced at Arthroplasty Number of cases Patellar tracking problem with lateral release 2 Mediolateral instability 3 Severe medial bone loss requiring wedges 2 Depressed lateral plateau 1 Deficient lateral femoral condyle 1 Internal torsion of proximal tibia 2 External torsion of proximal tibia 1 Lateral translation proximal tibia 18 Recurvatum of proximal tibia 3 Medial slope of plateau 3 Lateral slope of plateau 3 Increased posterior slope of plateau 1 Posterior translation of proximal tibia 1 Anterior translation of proximal tibia 1 Hard sclerotic bone 1 Figure 4 Negative values indicate varus angulation and positive values indicate valgus angulation. Discussion The problems encountered at knee arthroplasty after a high tibial osteotomy included significant soft tissue imbalance, tibial bone loss requiring tibial augmentation wedges in 2 cases, altered slope of the tibial plateau as well as translational deformity of proximal tibia (Table 2). Soft tissue imbalance at the long-term follow-up following high tibial osteotomy was also observed by Bettin and colleagues. 6 They found that after valgisation osteotomy, a stretching of medial collateral occurred. They found that the lateral collateral ligament remained lax in the long-term, due perhaps to a functional adaptation process to the postoperative loading in valgus angulation on the medial collateral ligament. The patellar height as measured by the ratio LP/LT was, however, not significantly compromised in this series and was on average 0.98 (range: 0.76 to 1.23) after tibial osteotomy and 1.0 after knee arthroplasty. This is in contrast to the experience of Insall and associates. 2 Conversion of a failed HTO to a TKR presents technical challenges not generally encountered in primary arthroplasty. Operative time for revision of an osteotomy to a TKR is generally more than the time it takes to perform a primary arthroplasty. Medial scarring from osteotomy can be more challenging while exposing and soft tissue balancing at the time of TKR. A closing wedge HTO shortens the distance of the tibial tubercle from the joint line, making exposure of the proximal part of the tibia and lateral dislocation of the patella more difficult, which increases the risk of avulsion of the tibial tubercle. 2,7 Generally, 8 mm to 14 mm of bone is removed from the lateral side of the proximal tibia at the time of HTO. Thus resecting this amount of bone during TKR results in excessive bony resection medially, and even more resection laterally. This may result in difficulties with excessive laxity and soft tissue imbalance and also necessitates use of a very thick prosthesis on the tibial side to restore the joint line. Osteotomy of the tibial tubercle may be required to expose the proximal part of the tibia for insertion of the tibial component in patients who have marked shortening of the proximal part of the tibia. The truncated metaphyseal portion of the proximal part of tibia can present potential difficulties in seating the central stem or lateral-fixation lug of a tibial component that may abut the lateral aspect of the tibial cortex. Malrotation of the

5 Bulletin Hospital for Joint Diseases Volume 61, Numbers 1 & proximal tibia may occur at the time of HTO, which could cause patellofemoral subluxation or rotatory instability if not recognized. 2 Alignment of the tibial baseplate is critical to success in TKR. Rotation of the tibia should be judged at the tubercle, with great care taken to avoid internal rotation of the tibial component. When preparing the tibia for insertion of the component, minimal bony resection is required to prevent the above-mentioned problems. However, because of the deficient bone stock laterally, this may not be possible and, therefore, lateral structural bone grafts or metal spacers should be used to attain normal soft tissue balance. Two of our patients had valgus cuts of the tibia when they had TKR. This could be due to the misjudgment of the proximal tibial bony deficiency. These patients had soft tissue imbalance and instability after TKR and required a revision to a constrained TKR. The patients could have had better soft tissue balance if metal spacers or bone grafts were used. The posterior cruciate ligament (PCL) is often scarred and shortened after HTO. Therefore, adequate balancing may require resecting the PCL and using a PCL substituting component. In evaluating the patient who has a failed proximal tibial osteotomy for potential TKR the intra-articular deformity of the tibia with respect to the extra-articular alignment of the limb must be carefully observed. A knee that is in varus angulation and has significant loss of bone from the medial tibial compartment and then has an osteotomy that results in an overcorrected valgus position presents the dilemma, at the time of TKR, of requiring a proximal tibial cut in an attempt to correct for dissimilar intra-articular and extra-articular deformities. This may result in inadvertent sacrifice of the collateral ligaments, with radical removal of bone. 7 Therefore bone grafting or spacers should be used more readily in cases of revision of HTO to TKR when indicated. The significant complication rate and technical difficulties associated with conversion of high tibial osteotomy to total knee arthroplasty suggests that tibial osteotomy should be reserved for a very narrow group of patients in whom primary arthroplasty is unsuitable. After knee arthroplasty the 4 which were revised had a less than satisfactory result right from the outset. This is in contrast to the findings of Staheli and associates 8 who found no difficulties in converting tibial osteotomies to knee arthroplasty, whereas Katz and colleagues 7 reported results similar to our series. However, Gill and coworkers 9 found greater difficulty in converting unicompartmental knee replacements compared to tibial osteotomy. Other authors have found the reverse to be true. Chakrabarty and colleagues, 10 Schai and colleagues, 11 and Thornhill and colleagues 12 reported no difficulties in converting unicompartmental knee replacements to total knees, while, Jackson and coworkers, 13 in a comparison of matched patients who had undergone either high tibial osteotomy or unicompartmental knee replacements, found a higher complication rate while revising the osteotomy group. Conclusions Post high tibial osteotomy knee arthroplasty is a technically demanding operation with altered anatomy and significant balancing problems. Results of knee arthroplasty after high tibial osteotomy are inferior to that of primary knee arthroplasty. Preosteotomy subluxation of tibia may have an adverse effect on knee arthroplasty results. Valgus cut of tibia appeared to fare badly as compared to a slight varus or neutral cut. Post dome knee arthroplasty appeared to do better than post wedge knee arthroplasty, although this was not statistically significant. Acknowledgments We would like to acknowledge and thank the following individuals: Dr. Peter Thomas of Bournemouth University for helping us with the statistical analysis of our data and Mr. Blakeway, Mr. Dinley, Mr. Jowett and Mr. Kernohan of the Royal Bournemouth Hospital for allowing us to review their patients. References 1. Cameron HU: Knee Arthroplasty: Limits and other problems: Total knee replacement following high tibial osteotomy and unicompartmental knee. Orthopaedics 19(9): , Insall JN, Joseph DM, Msika C: High tibial osteotomy for varus gonarthrosis: A long-term follow up study. J Bone Joint Surg 66A: , Takai S, Yoshino N, Hirasawa Y: Revision total knee arthroplasty after failed high tibial osteotomy. Bull Hosp Jt Dis 56(4): , Toksvig-Larsen S, Magyar G, Onsten I, Ryd L, Lindstrand A: Fixation of the tibial component of total knee arthroplasty after high tibial osteotomy. J Bone Joint Surg 80B: , Ivarsson I, Myrnerts R, Gillquist J: High tibial osteotomy for medial osteoarthritis of the knee. J Bone Joint Surg 72B: , Bettin D, Karbowski A, Schwering L, Matthiab HH: Time dependent clinical and roentgenographical results of Coventry high tibial valgisation osteotomy. Arch Orthop Trauma Surg 117:53-57, Katz MM, Hungerford DS, Krackow KA, Lennox DW: Results of total knee arthroplasty after failed proximal tibial osteotomy for osteoarthritis. J Bone Joint Surg 69A: , Staeheli JW, Cass JR, Morrey BF: Condylar total knee arthroplasty after failed proximal tibial osteotomy. J Bone Joint Surg 69A:28-31, Gill T, Schemitsch EH, Brick GW, Thornhill TS: Revision total knee arthroplasty after failed unicompartmental knee

6 10 Bulletin Hospital for Joint Diseases Volume 61, Numbers 1 & arthroplasty or high tibial osteotomy. Clin Orthop 321:10-18, Chakrabarty G, Newman JH, Ackroyd CE: Revision of unicompartmental arthroplasty of the knee. J Arthroplasty 13(2): , Schai PA, Suh JT, Thornhill TS, Scott RD: Unicompartmental knee arthroplasty in middle aged patients. J Arthroplasty 13(4): , Thornhill TS, Scott RD: Unicompartmental total knee arthroplasty. Orthop Clin North Am 20(2): , Jackson M, Sarangi PP, Newman JH: Revision knee arthroplasty: Comparison of outcome following primary proximal tibial osteotomy or unicompartmental arthroplasty. J Arthroplasty 9(5): , 1994.